Handbook of Optical Biomedical Diagnostics, Vol.2 Methods, 2nd Edition

HANDBOOK OF
OPTICAL
BIOMEDICAL
DIAGNOSTICS
SECOND EDITION
Volume 2: Methods
Valery V. Tuchin
EDITOR
Library of Congress Cataloging-in-Publication Data
Names: Tuchin, V. V. (Valerii Viktorovich), editor.

Title: Handbook of optical biomedical diagnostics / Valery V. Tuchin, editor.
Other titles: Optical biomedical diagnostics
Description: Second edition. | Bellingham, Washington : SPIE Press, [2016] |
Includes bibliographical references and index.
Identifiers: LCCN 2015038341| ISBN 9781628419092 (v. 1 : alk. paper) | ISBN
9781628419139 (v. 2 : alk. paper)
Subjects: | MESH: Diagnostic Imaging–methods. | Microscopy, Confocal. |
Spectrum Analysis.
Classification: LCC R857.O6 | NLM WN 180 | DDC 616.07/54–dc23
LC record available at http://lccn.loc.gov/2015038341
Published by
SPIE
P.O. Box 10
Bellingham, Washington 98227-0010 USA
Phone: +1 360.676.3290
Fax: +1 360.647.1445
Email: [email protected]
Web: http://spie.org
Copyright © 2016 Society of Photo-Optical Instrumentation Engineers (SPIE)
All rights reserved. No part of this publication may be reproduced or distributed in

any form or by any means without written permission of the publisher.
The content of this book reflects the work and thought of the authors and editors.
Every effort has been made to publish reliable and accurate information herein, but the
publisher is not responsible for the validity of the information or for any outcomes
resulting from reliance thereon. All known errata will be posted on the book’s page on
our website.
Printed in the United States of America.

First printing.
Images on the cover were reprinted with permission from the following publications:
• A. Doronin, C. Macdonald, and I. Meglinski, “Propagation of coherent polarized
light in highly scattering turbid media,” J. Biomed. Opt. 19(2), 025005 (2014).
• S. J. Kirkpatrick and D. D. Duncan, “Noncontact microstrain measurements in
orthodontic wires,” J. Biomed. Mater. Res. 29, 1437–1442 (1995).
Table of Contents
Preface xv
List of Contributors xix
III Scattering, Fluorescence, Infrared, and Raman Spectroscopy

of Tissues 1
Alexander V. Priezzhev and Juergen Lademann
1 Optical Study of RBC Aggregation in Whole Blood Samples and
Single Cells 5
Alexander V. Priezzhev, Kisung Lee, Nikolai N. Firsov, and Juergen Lademann
1.1 Introduction. Microrheological Structure of Blood: Biophysical and
Clinical Aspects 5
1.2 Importance of Quantitative Measurement of Red Blood Cell
Aggregation and Deformability Parameters 9
1.3 Arrangement of a Couette-Chamber-Based Laser Backscattering
Aggregometer 12
1.3.1 Measurement procedure 13
1.4 Kinetics of the Aggregation and Disaggregation Process
in Whole Blood Samples 15
1.4.1 Determination of the characteristic parameters of the
aggregation and disaggregation process in whole
blood samples 15
1.5 Parameters Influencing the Aggregation and Disaggregation
Measurements 16
1.5.1 Effect of blood sample temperature 16
1.5.2 Effect of blood sample oxygenation 18
1.5.3 Effect of sedimentation 19
1.5.4 Effect of hematocrit 19
1.6 Comparison of Aggregation and Disaggregation Measurements
with Sedimentation Measurements 20
1.7 Laser Tweezers as a New Tool for Studying RBC
Aggregation at the Single-Cell Level 21
1.7.1 Laser tweezers operation principle and experimental
arrangement 22
1.7.2 Sample preparation and measurement procedure 24
v
vi Table of Contents
1.7.2.1 Measurement of the disaggregation force 24

1.7.2.2 Measurement of the aggregation force 25
1.8 Hemorheological Characterization of Various Diseases by
Aggregation and Disaggregation Measurements of
Blood Samples 26
References 29
2 Light Scattering Spectroscopy of Epithelial Tissues: Principles and

Applications 37
Lev T. Perelman and Vadim Backman
2.1 Introduction 37
2.2 Microscopic Architecture of Mucosal Tissues 39
2.2.1 Morphology of the cell 40
2.2.2 Histology of mucosae 44
2.2.3 Introduction to histopathology of early cancer and dysplasia 47
2.3 Principles of Light Scattering 50
2.3.1 Rigorous solution of the direct scattering problem 51
2.3.2 Approximate solutions of the scattering problem 53
2.3.3 Numerical solutions of the scattering problem 58
2.4 Light Scattering by Cells and Subcellular Structures 59
2.5 Light Transport in Superficial Tissues 66
2.6 Detection of Cancer with Light Scattering Spectroscopy 70
2.6.1 Diagnosis of early cancer and precancerous lesions
with diffusely scattered light 71
2.6.2 Diagnosis of early cancer and precancerous
lesions with single-scattered light 77
2.6.3 Imaging of early cancer and precancerous lesions with an
endoscopic polarized scanning spectroscopy instrument 83
2.7 Confocal Light Absorption and Scattering Spectroscopic
Microscopy 88
Acknowledgments 92
References 92
3 Reflectance and Fluorescence Spectroscopy of Human Skin in vivo 99

Yuri P. Sinichkin, Nikiforos Kollias, George I. Zonios, Sergei R. Utz, and
Valery V. Tuchin
3.1 Introduction 99
3.2 Human-Skin Back Reflectance and Autofluorescence Spectra
Formation 100
3.2.1 Diffuse reflectance spectrum 100
3.2.2 Autofluorescence spectra 105
3.3 Simple Optical Models of Human Skin 112
3.3.1 Simple skin model for reflectance analysis 112
3.3.2 Simple skin model for autofluorescence analysis 115
Table of Contents vii
3.4 Combined Reflectance and Fluorescence Spectroscopy

Method for in vivo Skin Examination 118
3.4.1 Correction of the internal absorption effect in
fluorescence emission 118
3.4.2 Determination of melanin and erythema indices 119
3.4.3 Monitoring of hemoglobin oxygenation 122
3.5 Color Perception of Human-Skin Back Reflectance and
Fluorescence Emission 127
3.5.1 Color analysis of reflectance and fluorescence spectra 128
3.5.2 Color imaging 133
3.6 Polarization Reflectance Spectroscopy 136
3.7 Polarization Imaging 139
3.8 Sunscreen Evaluation using Reflectance and Fluorescence
Spectroscopy 143
3.9 Control of Skin Optical Properties 147
3.9.1 Introduction 147
3.9.2 Skin compression and stretching 148
3.9.3 Immersion optical clearing 151
3.9.3.1 In vitro spectrophotometry 155
3.9.3.2 In vivo spectral reflectance measurement 159
3.9.3.3 Frequency-domain measurements 161
3.9.4 Skin blood flow imaging 163
3.9.5 OCT imaging 163
3.9.6 Confocal microscopy 164
3.9.7 Fluorescence and Raman signal detection 165
3.9.8 Second harmonic generation 166
3.9.9 Skin heating 167
3.9.10 UV radiation 168
3.9.11 Applications 168
3.9.12 Conclusion 170
Conclusion 170
Acknowledgments 170
References 171
4 Infrared and Raman Spectroscopy of Human Skin in vivo 191
Gerald W. Lucassen, Peter J. Caspers, Gerwin J. Puppels,
Maxim E. Darvin, and Juergen Lademann
4.1 Introduction: Basic Principles of IR and Raman Spectrosopy 191
4.2 Fourier Transform Infrared Spectroscopy of Human-Skin Stratum
Corneum in vivo 193
4.2.1 Experimental ATR-FTIR setup 195
4.2.2 Human-skin stratum corneum spectra and band
assignments 196
4.2.3 ATR-FTIR spectrum of water 198
viii Table of Contents
4.2.3.1 Water bending mode and low-wave-number region 199

4.2.4 Stratum corneum hydration measurements 200
4.2.4.1 OH stretch region 200
4.2.4.2 Fit on water spectrum 201
4.2.5 Band analysis of hydrated and normal skin 202
4.2.5.1 Penetration depth of the IR beam 205
4.2.5.2 Fits of the hydrated-skin stratum corneum spectra 206
4.2.5.3 Comparison with MF and IR absorbance ratio 207
4.3 Confocal Raman Microspectroscopy of Human Skin in vivo 209
4.3.1 Setup for in vivo confocal Raman microspectroscopy 211
4.3.2 Water and natural moisturizing factor in human
skin epidermis 215
4.3.3 Raman spectra of human skin constituents in vitro 215
4.3.4 Profiling the water content and NMF content in
human skin in vivo 219
4.3.4.1 Water 219
4.3.4.2 NMF 221
4.4 Resonance Raman Spectroscopy of Cutaneous Carotenoids in vivo 223
4.4.1 Properties and role of cutaneous carotenoids 223
4.4.2 Setup for in vivo resonance Raman spectroscopy of
carotenoids 225
4.4.3 Selective detection of carotenoids in the human skin 226
4.4.4 In vivo measurements of the influence of UV irradiation
on human skin 228
4.4.5 In vivo measurements of the influence of IR irradiation
on human skin 229
4.4.6 In vivo measurements of the influence of VIS irradiation
on human skin 229
4.4.7 Factors influencing the concentration of carotenoids
in human skin 230
4.4.8 Distribution of carotenoids in human skin 230
4.4.9 Conclusions 231
Acknowledgments 231
References 231
5 Fluorescence Technologies in Biomedical Diagnostics 241
Herbert Schneckenburger, Wolfgang S. L. Strauss, Karl Stock, and
Rudolf Steiner
5.1 Introduction 241
5.1.1 Fundamentals 241
5.1.2 Potential diagram 242
5.1.3 Jablonski diagram and kinetic rates 243
5.1.4 Fluorescence anisotropy 244
5.2 Intrinsic and Extrinsic Fluorescence 245
Table of Contents ix
5.2.1 Intrinsic fluorophores 245

5.2.2 Fluorescent markers 246
5.3 Spectroscopic, Microscopic, and Imaging Techniques 248
5.3.1 Fluorescence spectroscopy 248
5.3.2 Fluorescence microscopy 250
5.3.3 Imaging techniques 252
5.4 Time-Resolved Fluorescence Spectrosopy and Imaging 254
5.4.1 Time-correlated single photon counting 254
5.4.2 Phase fluorometry 256
5.4.3 Time-gated fluorescence spectroscopy 258
5.4.4 Time-resolved fluorescence imaging 259
5.5 Total Internal Reflection Fluorescence Spectroscopy
and Microscopy (TIRFS/TIRFM) 262
5.5.1 Theory of TIRFS/TIRFM 263
5.5.2 Technical set-up 264
5.5.3 Combination of TIRFS/TIRFM with innovative
fluorescence microscopic techniques 266
5.5.4 Application of TIRFS/TIRFM in cell biology 267
5.6 Energy Transfer Spectroscopy 268
5.6.1 Basic mechanisms 268
5.6.2 FRET applications 271
5.7 Wide-Field 3D Microscopy 273
5.7.1 Structured illumination 273
5.7.2 Light sheet fluorescence microscopy (LSFM) 274
5.8 Laser Scanning and Multiphoton Microscopy 275
5.8.1 Introduction 275
5.8.2 Performance of confocal laser scanning microscopes 276
5.8.3 Applications of CLSM 280
5.8.4 Multiphoton microscopy 281
5.8.5 Super-resolution and single-molecule detection 284
5.9 Concluding Remarks 287
References 287
IV Coherent-Domain Methods for Biological Flows and Tissue

Structure Monitoring 305
J. David Briers and Sean J. Kirkpatrick
6 Laser Speckles, Doppler, and Imaging Techniques for Blood and

Lymph Flow Monitoring 309
Ivan V. Fedosov, Yoshihisa Aizu, Valery V. Tuchin, Naomichi Yokoi,
Izumi Nishidate, Vladimir P. Zharov, and Ekaterina I. Galanzha
6.2 Doppler and Speckle Techniques 314
x Table of Contents
6.2.1 Laser Doppler technique 314

6.2.2 Laser speckle technique 315
6.2.3 Interrelation 316
6.3 Two-Wavelength Near-Infrared Speckle Imaging 317
6.3.1 Optical system 317
6.3.2 Frame-rate analysis of blood flow 318
6.3.3 Blood flow measurements in humans 319
6.3.4 Blood flow measurements in rats 320
6.3.5 Simultaneous monitoring of blood flow and concentration 320
6.3.6 Measurements for humans 322
6.3.7 Experiments on rats 323
6.4 Low-Coherence Speckle Interferometry 325
6.5 Quantitative Characterization of Blood Flow Rate 327
6.5.1 The use of laser Doppler anemometry for measurements
of absolute blood flow velocity 327
6.5.2 Intravital particle image velocimetry of capillary
blood flow 333
6.6 Intravital Microscopy (IM) for Monitoring Blood and Lymph Flows 337
6.7 Intravital Transmission Digital Microscopy (ITDM) 338
6.8 Intravital Fluorescent Digital Microscopy (IFDM) 339
6.9 Optical Clearing 340
6.10 In vivo Flow Cytometry 340
6.11 In vivo Lymph Flow Cytometry (LFC) 343
6.12 Animal Models 343
6.13 Biomedical Applications 346
6.13.1 Optical lymphography 346
6.13.1.1 Indocyanine Green (ICG) lymphography 347
6.13.1.2 Integrated fluorescent angio- and lymphography 347
6.13.1.3 Monitoring lymph flow profile 347
6.13.2 In vivo label-free imaging of lymphatic function 348
6.13.2.1 Lymph flow 349
6.13.2.2 Experimental lymphedema 350
6.13.2.3 Nicotine intoxication 351
6.13.2.4 Nitric oxide 352
6.13.2.5 High-power laser–induced thermal effects on
lymph vessels 352
6.13.3 In vivo flow cytometry 353
6.13.3.1 Label-free image flow cytometry 353
6.13.3.2 In vivo lymph and blood fluorescent flow
cytometry 357
6.14 Summary 362
Acknowledgments 362
References 363
Table of Contents xi
7 Real-Time Imaging of Microstructure and Function Using Optical

Coherence Tomography 385
Christine P. Hendon and Andrew M. Rollins
7.2 Optical Coherence Tomography Principles 386
7.2.1 Time-domain OCT 388
7.2.2 Frequency-domain OCT 389
7.2.2.1 Spectrometers 393
7.2.2.2 Light sources 394
7.3 Functional Imaging 396
7.3.1 Doppler OCT 396
7.3.2 Polarization-sensitive OCT 397
7.4 Applications of OCT 398
7.4.1 Ophthalmology 398
7.4.2 Cardiology 400
7.4.3 Oncology 406
7.5 Conclusions 411
References 412
8 Speckle Technologies for Monitoring and Imaging Tissues and

Tissue-Like Phantoms 429
Dmitry A. Zimnyakov, Olga V. Ushakova, David J. Briers, and
Valery V. Tuchin
8.2 Diffusing-Wave Spectroscopy (DWS) as a Tool for Tissue
Structure and Cell Flow Monitoring 430
8.3 Laser Speckle Contrast Analysis (LASCA) for Measuring
Blood Flow 442
8.3.1 Statistical properties of laser speckle 442
8.3.2 Time-varying speckle 442
8.3.3 Full-field methods 443
8.3.4 Single-exposure speckle photography 444
8.3.5 Laser speckle contrast analysis (LASCA) 444
8.3.6 The question of speckle size 445
8.3.7 Theory 446
8.3.8 Practical considerations 448
8.3.9 Early applications of the LASCA technique 449
8.3.10 Important developments of the basic LASCA technique 450
8.3.11 Conclusions 452
8.4 Modification of Speckle Contrast Analysis to Improve Depth
Resolution and to Characterize Scattering Properties
of a Probed Medium 453
8.5 Various Modifications of Laser Speckle Contrast Imaging 463
xii Table of Contents
8.6 Imaging Using Contrast Measurements of Partially Developed

Speckles 467
8.7 Monitoring Tissue Thermal Modification with a Bundle-Based
Full-Field Speckle Analyzer 470
8.8 Summary 486
Acknowledgments 487
References 487
9 Optical Assessment of Tissue Mechanics 497

Sean J. Kirkpatrick, Donald D. Duncan, Brendan F. Kennedy, and
David D. Sampson
9.2 Introduction to Prior Edition 499
9.3 Tissue Mechanics and Medicine 500
9.3.1 Dermatology 501
9.3.2 Oncology 501
9.3.3 Ophthalmology 502
9.3.4 Cardiology 504
9.3.5 Other application areas 504
9.4 Constitutive Relations in Biological Tissues 505
9.5 Laser Speckle Patterns Arising from Biological Tissues 511
9.5.1 First-order statistics 512
9.5.2 Second-order statistics 514
9.6 Elastography Measurements by Tracking and Translating
Laser Speckle: The Transform Method 515
9.6.1 Potential error sources 521
9.6.2 Applications of laser speckle elastography to hard
and soft tissues 522
9.7 Alternative Processing Algorithms for Calculating
Speckle Shift 526
9.7.1 Non parametric speckle shift estimators 526
9.7.2 Parametric speckle shift estimators 527
9.7.2.1 A minimum mean square error estimator 528
9.8 Expanding to Higher Dimensions 531
9.9 What is Really Measured in Laser Speckle-Tracking Elastography? 534
9.9.1 Lagrangian description of motion of particles in
object space 534
9.9.2 Relationship between elastograms and SEDFs 536
9.10 In vivo Laser-Speckle-Tracking Optical Elastography 538
9.11 Performance Comparisons 538
9.12 Generalizations 541
9.13 Elastography of Tissues with Optical Coherence Tomography 544
Table of Contents xiii
9.13.1 Variants of OCE 547

9.13.1.1 Compression OCE 548
9.13.1.2 Surface wave/shear wave OCE 549
9.13.2 OCE probes 550
9.14 Acoustically Modulated Speckle Imaging 550
9.15 Conclusions 553
References 553
10 Optical Clearing of Tissues: Benefits for Biology,
Medical Diagnostics, and Phototherapy 565
E. A. Genina, A. N. Bashkatov, Yuri P. Sinichkin, I. Yu. Yanina, and
V. V. Tuchin
10.1 Fundamentals of Optical Clearing (OC) of Tissues
and Cells 565
10.2 Immersion OC 568
10.3 Compression OC 581
10.4 Photochemical, Thermal, and Photothermal OC 585
10.5 Applications of Optical Clearing 587
10.5.1 Optical coherence tomography 587
10.5.2 Optical projection tomography 593
10.5.3 Fluorescence imaging 594
10.5.4 Photoacoustic imaging 597
10.5.5 Nonlinear and Raman microscopy 600
10.5.6 Terahertz spectroscopy 603
10.6 Determination of OCA and Drug Diffusion Coefficients in Tissues 604
10.7 Conclusion 610
Acknowledgments 610
References 610
Index 639
Preface
This Handbook is the second edition of the monograph initially published in
2002. The first edition described some aspects of laser–cell and laser–tissue
interactions that are basic for biomedical diagnostics and presented many
optical and laser diagnostic technologies prospective for clinical applications.
The main reason for publishing such a book was the achievements of the last
millennium in light scattering and coherent light effects in tissues, and in the
design of novel laser and photonics techniques for the examination of the
human body. Since 2002, biomedical optics and biophotonics have had rapid
and extensive development, leading to technical advances that increase the
utility and market growth of optical technologies. Recent developments in the
field of biophotonics are wide-ranging and include novel light sources,
delivery and detection techniques that can extend the imaging range and
spectroscopic probe quality, and the combination of optical techniques with
other imaging modalities.
The innovative character of photonics and biophotonics is underlined by
two Nobel prizes in 2014 awarded to Eric Betzig, Stefan W. Hell, and William
E. Moerner “for the development of super-resolved fluorescence microscopy”
and to Isamu Akasaki, Hiroshi Amano, and Shuji Nakamura “for the
invention of efficient blue light-emitting diodes which has enabled bright and
energy-saving white light sources.” The authors of this Handbook have a
strong input in the development of new solutions in biomedical optics and
biophotonics and have conducted cutting-edge research and developments
over the last 10–15 years, the results of which were used to modify and update
early written chapters. Many new, world-recognized experts in the field have
joined the team of authors who introduce fresh blood in the book and provide
a new perspective on many aspects of optical biomedical diagnostics.
The optical medical diagnostic field covers many spectroscopic and laser
technologies based on near-infrared (NIR) spectrophotometry, fluorescence
and Raman spectroscopy, optical coherent tomography (OCT), confocal
microscopy, optoacoustic (photoacoustic) tomography, photon-correlation
spectroscopy and imaging, and Doppler and speckle monitoring of biological
flows.1–45 These topics—as well as the main trends of the modern laser
diagnostic techniques, their fundamentals and corresponding basic research
xv
xvi Preface
on laser–tissue interactions, and the most interesting clinical applications—are

discussed in the framework of this Handbook. The main unique features of
the book are as follows:
1. Several chapters of basic research that discuss the updated results on light
scattering, speckle formation, and other nondestructive interactions of
laser light with tissue; they also provide a basis for the optical and laser
medical diagnostic techniques presented in the other chapters.
2. A detailed discussion of blood optics, blood and lymph flow, and blood-
aggregation measurement techniques, such as the well-recognized laser
Doppler method, speckle technique, and OCT method.
3. A discussion of the most-recent prospective methods of laser (coherent)
tomography and spectroscopy, including OCT, optoacoustic (photoa-
coustic) imaging, diffusive wave spectroscopy (DWS), and diffusion
frequency-domain techniques.
The intended audience of this book consists of researchers, postgraduate
and undergraduate students, biomedical engineers, and physicians who are
interested in the design and applications of optical and laser methods and
instruments for medical science and practice. Due to the large number of
fundamental concepts and basic research on laser–tissue interactions presented
here, it should prove useful for a much broader audience that includes students
and physicians, as well. Investigators who are deeply involved in the field will
find up-to-date results for the topics discussed. Each chapter is written by
representatives of the leading research groups who have presented their classic
and most recent results. Physicians and biomedical engineers may be
interested in the clinical applications of designed techniques and instruments,
which are described in a few chapters. Indeed, laser and photonics engineers
may also be interested in the book because their acquaintance with a new field
of laser and photonics applications can stimulate new ideas for lasers and
photonic devices design. The two volumes of this Handbook contain 21
chapters, divided into four parts (two per volume):
• Part I describes the fundamentals and basic research of the extinction of
light in dispersive media; the structure and models of tissues, cells, and
cell ensembles; blood optics; coherence phenomena and statistical
properties of scattered light; and the propagation of optical pulses and
photon-density waves in turbid media. Tissue phantoms as tools for
tissue study and calibration of measurements are also discussed.
• Part II presents time-resolved (pulse and frequency-domain) imaging
and spectroscopy methods and techniques applied to tissues, including
optoacoustic (photoacoustic) methods. The absolute quantification of
the main absorbers in tissue by a NIR spectroscopy method is discussed.
An example biomedical application—the possibility of monitoring brain
activity with NIR spectroscopy—is analyzed.
Preface xvii
• Part III presents various spectroscopic techniques of tissues based on

elastic and Raman light scattering, Fourier transform infrared (FTIR),
and fluorescence spectroscopies. In particular, the principles and
applications of backscattering diagnostics of red blood cell (RBC)
aggregation in whole blood samples and epithelial tissues are discussed.
Other topics include combined back reflectance and fluorescence, FTIR
and Raman spectroscopies of the human skin in vivo, and fluorescence
technologies for biomedical diagnostics.
• The final section, Part IV, begins with a chapter on laser Doppler
microscopy, one of the representative coherent-domain methods applied
to monitoring blood in motion. Methods and techniques of real-time
imaging of tissue ultrastructure and blood flows using OCT is also
discussed. The section also describes various speckle techniques for
monitoring and imaging tissue, in particular, for studying tissue
mechanics and blood and lymph flow.
Financial support from a FiDiPro grant of TEKES, Finland (40111/11)
and Academic D.I. Mendeleev Fund Program of Tomsk National Research
State University have helped me complete this book project. I greatly
appreciate the cooperation and contribution of all of the authors and co-
editors, who have done a great work on preparation of this book. I would like
to express my gratitude to Eric Pepper and Tim Lamkins for their suggestion
to prepare the second edition of the Handbook and to Scott McNeill for
assistance in editing the manuscript. I am very thankful to all of my colleagues
from the Chair and Research Education Institute of Optics and Biophotonics
at Saratov National Research State University and the Institute of Precision
Mechanics and Control of RAS for their collaboration, fruitful discussions,
and valuable comments. I am very grateful to my wife and entire family for
their exceptional patience and understanding.
Valery V. Tuchin
April 2016
References
1. F. A. Duck, Physical Properties of Tissue: A Comprehensive Reference
Book, Academic, London (1990).
2. A. P. Shepherd and P. A. Oberg, Laser Doppler Blood Flowmetry,
Kluwer, Boston (1990).
3. J. B. Pawley (Ed.), Handbook of Biological Confocal Microscopy, Plenum
Press, New York (1990).
4. T. Wilson (Ed.), Confocal Microscopy, Academic Press, London (1990).
5. K. Frank and M. Kessler (Eds.), Quantitative Spectroscopy in Tissue, pmi
Verlag, Frankfurt am Main (1992).
xviii Preface
6. G. Müller, B. Chance, R. Alfano, et al. (Eds.), Medical Optical

Tomography: Functional Imaging and Monitoring, IS 11, SPIE Press,
Bellingham (1993).
7. V. V. Tuchin (Ed.), Selected Papers on Tissue Optics Applications in
Medical Diagnostics and Therapy, Milestones Series MS 102, SPIE Press,
Bellingham (1994).
8. B. R. Masters (Ed.), Confocal Microscopy, MS 131, SPIE Press,
Bellingham (1996).
9. O. Minet, G. Mueller, and J. Beuthan (Eds.), Selected Papers on Optical
Tomography, Fundamentals and Applications in Medicine, MS 147, SPIE
Press, Bellingham (1998).
10. V. V. Tuchin, Tissue Optics: Light Scattering Methods and Instruments for
Medical Diagnosis, SPIE Tutorial Texts in Optical Engineering, Tutorial
Text Series, 38 SPIE Press, Bellingham (2000).
11. B. R. Masters (Ed.), Selected Papers on Optical Low-Coherence
Reflectometry and Tomography, MS 165, SPIE Press, Bellingham (2001).
12. B.E. Bouma and G.J. Tearney (Eds.), Handbook of Optical Coherence
Tomography, Marcel-Dekker, New York (2002).
13. T. Vo-Dinh (Ed.), Biomedical Photonics Handbook, Boca Raton, CRC
Press (2003); 2nd ed. (2014).
14. H.-P. Berlien and G.J. Müller (Eds.), Applied Laser Medicine, Springer-
Verlag, Berlin (2003).
15. P. Prasad, Introduction to Biophotonics, Wiley-Interscience, Hoboken,
New Jersey (2003).
16. J.R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed., Springer
Science þ Business, New York (2006).
17. V.V. Tuchin, Tissue Optics: Light Scattering Methods and Instruments for
Medical Diagnosis, 2nd ed., PM 166 (2007); 3rd ed., PM254, SPIE Press,
Bellingham, WA (2015).
18. L.V. Wang and H.-I. Wu, Biomedical Optics: Principles and Imaging,
Wiley-Interscience, Hoboken, New Jersey (2007).
19. Q. Luo, L. Wang, and V.V. Tuchin (Eds.), Advances in Biomedical
Photonics and Imaging, World Scientific, New Jersey, London, Singapore
et al. (2008).
20. G. Ahluwalia (Ed.), Light Based Systems for Cosmetic Application,
William Andrew, Inc., Norwich, New York (2008).
21. W. Bock, I. Gannot, and S. Tanev (Eds.), Optical Waveguide Sensing and
Imaging, NATO SPS Series B: Physics and Biophysics, Springer,
Dordrecht (2008).
22. W. Drexler and J.G. Fujimoto (Eds.), Optical Coherence Tomography:
Technology and Applications, Springer, Berlin (2008); 2nd ed. Springer,
Berlin (2015).
Preface xix
23. E. Baron (Ed.), Light-Based Therapies for Skin of Color, Springer, New
York (2009).
24. K.-E. Peiponen, R. Myllylä, and A. V. Priezzhev, Optical Measurement
Techniques, Innovations for Industry and the Life Science, Springer-Verlag,
Berlin, Heidelberg (2009).
25. L. Wang, Ed., Photoacoustic Imaging and Spectroscopy, CRC Press,
Taylor & Francis Group, London (2009).
26. V.V. Tuchin (Ed.), Handbook of Optical Sensing of Glucose in Biological
Fluids and Tissues, CRC Press, Taylor & Francis Group, London (2009).
27. A. Wax and V. Backman (Eds.), Biomedical Applications of Light
Scattering, McGraw-Hill, New York (2010).
28. V. V. Tuchin, Lasers and Fiber Optics in Biomedical Science, 2nd ed.,
Fizmatlit, Moscow (2010).
29. X.-C. Zhang and J. Xu, Introduction to THz Wave Photonics, Springer,
New York (2010).
30. V.V. Tuchin (Ed.), Handbook of Photonics for Medical Science, CRC
Press, Taylor & Francis Group, London (2010).
31. F. S. Pavone (Ed.), Laser Imaging and Manipulation in Cell Biology,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim (2010).
32. V.V. Tuchin (Ed.), Advanced Optical Flow Cytometry: Methods and
Disease Diagnoses, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
(2011).
33. D. A. Boas, C. Pitris, and N. Ramanujam (Eds.), Handbook of Biomedical
Optics, CRC Press, Taylor & Francis Group, London (2011).
34. J. Popp, V.V. Tuchin, A. Chiou, and S.H. Heinemann (Eds.), Handbook
of Biophotonics, vol. 1: Basics and Techniques, WILEY-VCH Verlag
GmbH & Co. KGaA, Weinheim (2011).
of Biophotonics, vol. 2: Photonics for Health Care, WILEY-VCH Verlag
GmbH & Co. KGaA, Weinheim (2012).
of Biophotonics, vol. 3: Photonics in Pharmaceutics, Bioanalysis and
Environmental Research, WILEY-VCH Verlag GmbH & Co. KGaA,
Weinheim (2012).
37. V.V. Tuchin, Dictionary of Biomedical Optics and Biophotonics, SPIE
Press, Bellingham, WA (2012).
38. M. J. Leahy (ed.), Microcirculation Imaging, Wiley-VCH Verlag GmbH
& Co. KGaA, Weinheim (2012).
39. R.K. Wang and V.V. Tuchin (Eds.), Advanced Biophotonics: Tissue
Optical Sectioning, CRC Press, Taylor & Francis Group, London (2013).
40. H. Jelinkova (Ed.), Lasers for Medical Applications: Diagnostics, Therapy
and Surgery, Woodhead Publishing, Ltd., Cambridge (2013).
xx Preface
41. F. S. Pavone and P. J. Campagnola (Eds.), Second Harmonic Generation

Imaging, CRC Press, Taylor & Francis Group, Boca Raton, London,
New York (2014).
42. F.S. Pavone, P.T.C. So, and P.M.W. French (Eds.), Proc. of the
International School of Physics ‘Enrico Fermi,’ Course 181 – Microscopy
Applied to Biophotonics, Societa Italiana di Fisica, Bologna (2014).
43. B. Querleux (Ed.), Computational Biophysics of the Skin, CRC Press,
Taylor & Francis Group, London (2015).
44. F.D. Dip, T. Ishizawa, N. Kokudo, and R. Rosenthal (Eds.), Fluores-
cence Imaging for Surgeons: Concepts and Applications, Springer Science
þ Business Media, New York (2015).
45. I. J. Bigio and S. Fantini, Quantitative Biomedical Optics: Theory,
Methods, and Applications, Cambridge University Press, Cambridge
(2016).
List of Contributors
Yoshihisa Aizu Ekaterina I. Galanzha

Muroran Institute of Technology, University of Arkansas for Medical
Japan Sciences, USA
Vadim Backman E. A. Genina

Northwestern University, USA Saratov National Research State
University and Tomsk National
A. N. Bashkatov Research State University,
Saratov National Research State Russia
University and Tomsk National
Research State University, Christine P. Hendon
Russia Columbia University, USA
David J. Briers Brendan F. Kennedy

Kingston University, UK University of Western Australia,
Australia
Peter J. Caspers
Erasmus University Rotterdam, Sean J. Kirkpatrick
The Netherlands Michigan Technological University,
USA
Maxim E. Darvin
Nikiforos Kollias
University Clinic Charité, Germany
University of British Columbia,
Canada
Donald D. Duncan
Portland State University, USA Juergen Lademann
University Clinic Charité, Germany
Ivan V. Fedosov
Saratov National Research State Kisung Lee
University, Russia Lomonosov Moscow State
University, Russia
Nikolai N. Firsov
Russian State Medical University, Gerald W. Lucassen
Russia Philips Research, The Netherlands
xxi
xxii List of Contributors
Izumi Nishidate Wolfgang S. L. Strauss

Tokyo University of Agriculture University of Ulm, Germany
and Technology, Japan
Valery V. Tuchin
Lev T. Perelman Saratov National Research State
Harvard University, USA University, Tomsk National
Research State University, and the
Alexander V. Priezzhev Institute of Precision Mechanics and
Lomonosov Moscow State Control, Russia
University, Russia
Olga V. Ushakova
Gerwin J. Puppels Saratov Technical University,
Erasmus University Rotterdam, Russia
The Netherlands
Sergei R. Utz
Andrew M. Rollins Saratov State Medical University,
Case Western Reserve University, Russia
USA
I. Yu. Yanina
David D. Sampson
Saratov National Research State
University of Western Australia,
Medical University, Russia
Australia
Herbert Schneckenburger Naomichi Yokoi

University of Ulm and Aalen Asahikawa National College of
University, Germany Technology, Japan
Yuri P. Sinichkin Vladimir P. Zharov

Saratov National Research State University of Arkansas for Medical
University and Tomsk National Sciences, USA
Research State University, Russia
Dmitry A. Zimnyakov
Rudolf Steiner Saratov Technical University,
University of Ulm, Germany Russia
Karl Stock George I. Zonios

University of Ulm, Germany University of Ioannina, Greece
Part III: Scattering,
Fluorescence, Infrared, and
Raman Spectroscopy of
Tissues
This part of the Handbook describes the basic principles and diagnostic
applications of optical techniques based on detecting and processing the
scattering, fluorescence, FT IR, and Raman spectroscopic signals from
various tissues, with an emphasis on blood, epithelial tissues, and human skin.
Chapter 1 covers the approaches to quantitative measurement of the
spontaneous aggregation kinetics of red blood cells in whole blood samples
and the biophysical and clinical importance of these measurements. It is
shown that real-time measurement of the backscattered light intensity
provides information on a number of important characteristics of blood
related to the hemorhological status of the donor. Although there are a
number of parameters influencing the aggregation and disaggregation
measurements, statistically significant correlations with different diseases
can be obtained that have high diagnostic value for clinicians. The relation of
blood aggregation and sedimentation measurements is also discussed. It is
shown that the new emerging modality of laser manipulation and trapping
(laser tweezers) is very helpful when studying the individual features of
interaction between cells, measuring the corresponding forces and the kinetics
of cells aggregation and disaggregation.
Chapter 2 overviews the principles and applications of light scattering
spectroscopy of epithelial tissues. It describes novel techniques capable of
identifying and characterizing pathological changes in these tissues at the
cellular and sub-cellular levels and providing structural and functional
information about the tissue. The discussion is focused on studying epithelial
morphology in living tissues without tissue removal aiming at noninvasive or
minimally invasive detection of precancerous and early cancerous changes in a
variety of organs such as esophagus, colon, uterine cervix, oral cavity, lungs,
1
2 Part III
and urinary bladder. The main goal of this chapter is to provide the readers
with basic tools necessary to understand the potentials of biomedical light
scattering spectroscopy, including sufficient medical and biological back-
ground and principles of light scattering by cells and sub-cellular structures.
The relation of single and multiple scattering in tissue is particularly
considered. Finally, the applications of various types of light scattering in
detection of early cancer and precancerous conditions are reviewed. In
addition, several recently developed clinical tools are described including the
endoscopic polarized scanning spectroscopy (EPSS) instrument, which is
compatible with existing endoscopes. It scans large areas of the esophagus
chosen by the physician and has the software and algorithms necessary to
obtain quantitative, objective data about tissue structure and composition,
which can be translated into diagnostic information in real time. This process
enables the physician to take confirming biopsies at suspicious sites and
minimize the number of biopsies taken at nondysplastic sites. Another newly
developed technique, called confocal light absorption and scattering
spectroscopic (CLASS) microscopy, combines light-scattering spectroscopy
(LSS) with confocal microscopy. In CLASS microscopy, light-scattering
spectra are the source of the contrast. Another important aspect of LSS is its
ability to detect and characterize particles well beyond the diffraction limit.
Chapter 3 discusses the applications of reflectance and fluorescence
spectroscopies for the assessment of the optical properties of human skin in
relation to different diseases, environmental factors, and the effectiveness of
various treatments. Applied to the skin in vivo, these techniques provide
information on the structure of epidermis and dermis, on the quantity and
density of blood vessels, on the concentration and spatial distribution of
chromophores and fluorophores in skin, and on the nature of skin metabolic
processes. The authors discuss the potential advantages and possible
applications of the combined use of reflectance and fluorescence spectroscopy
of skin for the evaluation of erythema and pigmentation indices, the
determination of hemoglobin oxygenation and concentration, and the
investigation of the efficacy of topical sunscreens. Simple models are used
to analyze changes in skin reflectance and fluorescence spectra as a result of
morphological and functional alterations in skin, or as a result of treatment
effects. Such changes can be monitored by imaging techniques, in particular,
in polarized light and analyzing the color characteristics of the reflected light.
Ways to improve the accuracy of skin diagnostics and the efficiency of skin
therapy by analyzing and controlling the skin optical parameters are also
discussed in this chapter. In particular, the authors demonstrate how to
control the sensitivity of skin reflectance spectra by compression and
stretching. A special emphasis is made on the potentialities of immersion
optical clearing and corresponding decrease in the scattering coefficient in
tissue studies. Ways to raise the efficiency of optical clearing, e.g., by
Scattering, Fluorescence, Infrared, and Raman Spectroscopy of Tissues 3
accelerating the penetration of the index-matching compounds by enhancing

skin permeability through creating a lattice of microzones (islets) of limited
thermal damage in the stratum corneum, are also discussed.
Chapter 4 discusses the basic principles and potentialities of in vivo
diagnostics of human skin by vibrational spectroscopic techniques, namely,
Fourier transform infrared spectroscopy and confocal Raman microspectro-
scopy. The detailed information on the molecular composition, structure, and
organization of the skin and, in particular, the content of water and natural
moisturizing factor in human skin epidermis that can be obtained with these
techniques is highlighted. The results of the research, reviewed in this chapter,
provide the means for various applications of these techniques in cosmetics,
pharmacology, clinical diagnosis, treatment monitoring, and surgery. A large
part of the chapter is devoted to the resonance Raman spectroscopy of
cutaneous carotenoids. These substances form an antioxidant network of
living skin and quick in vivo measurement of their amount in skin is very
important when estimating the status of a human organism. Distribution of
carotenoids in the human skin and the factors influencing their concentration
are discussed.
Finally, Chapter 5 overviews different fluorescence technologies used in
biomedical diagnostics. It provides information on the basic principles of
fluorescence spectroscopy, microscopy, and imaging, including the continuous-
wave, time-gated, and time-resolved variants. Theory and applications to cell
biology of total internal reflection fluorescence spectroscopy and microscopy,
energy transfer spectroscopy and wide-field 3D microscopy (including
structured illumination and light sheet microscopies) are described in detail.
This is followed by a discussion of the principles as well as current and
possible future applications of laser scanning and multiphoton microscopy. In
the last part of the chapter, the super-resolution and single-molecule detection
possibilities are briefly discussed.
Overall, the chapters provide readers with knowledge of a very important
and quickly developing field of optical biomedical diagnostics.
Alexander V. Priezzhev
Juergen Lademann
Co-editors
Chapter 1
Optical Study of RBC
Aggregation in Whole Blood
Samples and Single Cells
Alexander V. Priezzhev and Kisung Lee
Lomonosov Moscow State University, Moscow, Russia
Nikolai N. Firsov
Russian State Medical University, Moscow, Russia
Juergen Lademann
University Clinic Charité, Berlin, Germany
1.1 Introduction. Microrheological Structure of Blood:

Biophysical and Clinical Aspects
Historically, the first observations of red blood cell (RBC) association into
larger structures in human blood were performed in the 18th century. Since
that time extensive research has been performed to explain this phenomenon,
which determines the rheological properties of blood, affects the micro-
vascular blood flow, and often leads to the development of blood flow
abnormalities in large vessels.
The RBC aggregation process takes place in both in vivo and in vitro
conditions. Normal human RBCs suspended in a solution of high molecular
weight proteins or synthetic colloids of various chemical contents tend to
associate into linear structures resembling rouleaux, while in an isotonic salt
solution the cells stay monodisperse. A manifold of different experiments with
blood samples showed that the extent of RBC aggregation and the
aggregation rate depend not only on the concentration of cells, but on
physico-chemical properties and the concentration of blood plasma proteins.
This brought the researchers to a conclusion that this phenomenon is a result
of the interaction of RBCs and macromolecules.
5
6 Chapter 1
Microscopic observations of RBC motion in in vitro conditions show that

the association of RBCs into doublets and further into larger aggregates is an
active process that can be implemented in either of three ways: by cells
sticking “edge to surface,” the crawling of one cell over another, and covering
one cell by another by turning.1,2 This process starts after two initially
separated RBCs approach each other at a distance of around 25 nm. The
approach can happen differently depending on the flow conditions. Hence, the
kinetics of aggregation is different in a regular shear flow, residual motion in a
cuvette, at RBC sedimentation in a vertical tube, in a field of ultrasound
oscillations, etc.3 Based on experimental measurements, note that thermal
Brownian motion does not suffice for bringing a large number of cells into
contact during a characteristic time of several seconds due to the very low
value of their translation diffusion coefficient (D , 10–12 cm2/s). The diffusion
of RBCs in shear flow (shear diffusion) is several orders of magnitude higher.
Thus, shear flow is one of the major mechanisms of cell collisions, which are
necessary for aggregation. Sedimentation in a gravity environment may be
another such mechanism. Experiments performed with whole blood samples
in a microgravity environment showed that the resulting aggregates differ
from those formed under normal gravity.4
The general thermodynamic cause of RBC aggregation is related to the
existence of a large phase separation interface in a suspension of discocytes and,
correspondingly, the existence of redundant free energy. By aggregating, the cells
reduce their surface energy.1 The mechanisms of cohesion interaction of RBCs in
aggregates have been extensively studied during the last 25 years. One of the
hypotheses about the biophysical mechanism of aggregation is based on a
“bridging” model of membrane interaction.5–9 According to this model the large
macromolecules constituting blood plasma, such as fibrinogen, IgG, IgM, and
lipoproteids, are adsorbed on the RBC membrane surface. When two cells
approach each other at a distance of 30 to 40 nm, the free side of a
macromolecule, earlier adsorbed on the membrane of one cell, is also adsorbed
on the membrane of the second cell. This is how a “bridge” is formed. The
movement of the cells around this “bridge” results in the further formation of
other “bridges.” The resulting cohesion force sticking individual RBCs together
corresponds to the number of “bridges” formed between them. The “bridge”
forces are nonspecific and appear only when individual cells (singly or as parts of
aggregates) casually approach each other at a critical distance.
According to another hypothesis,10 the major factors of RBC aggregation
in the presence of biopolymers are flexible structures formed in the solution by
these molecules. These structures act as osmotic cells and molecular sieves,
pushing RBCs out of the space occupied by them. This is supported by the
electric charge similarity of the RBC membranes and major polymer
molecules in blood plasma. Because they are concentrated in very limited
Optical Study of RBC Aggregation in Whole Blood Samples and Single Cells 7
volumes, the RBCs have a higher probability of collisions and, consequently,

of aggregation.
To avoid an aggregation collapse, the RBC suspension must have an
efficient system of stabilization preventing the formation of strong aggregates
in physiological conditions. This system is based on electrostatic repulsion
forces. The RBC membrane surface is covered with negatively charged
carboxy groups of cyalic acid. They supply the membranes with a negative
electrostatic charge that prevents cells from very close rapprochement. The
repulsion force per unit surface between two normally charged cells separated
at a distance of 20 nm is $ 0.01N/m2. The energy of electrostatic interaction
between RBCs is about 1.9 10–14 J per cell. Electrokinetic repulsion is one of
the mechanisms that maintains RBC suspension stability. Neutralization of
the electric charge on RBC membranes by treating them with neuraminidase
strongly enhances the aggregation process in blood.11
Another mechanism of stabilization is based on the concurrence of
albumin molecules in blood plasma for the macromolecule binding sites on
RBC membranes. The extent of RBC aggregation is inversely proportional
to the concentration ratios of albumin/fibrinogen and albumin/globulin in
plasma.12 Albumin molecules themselves do not cause aggregation due to
their small size in blood (molecular mass around 5 104 dalton), but can
reduce the number of “bridges” by occupying the binding sites. Direct
experiments with exogenous dextrans of different molecular mass show that
those with M . 4 104 kDa induce the growth of aggregates while those with
M , 4 104 kDa induce their destruction.13
RBC aggregation is a reversible process. In shear flow conditions, the
aggregates lose single cells or groups of cells and acquire the others.
The recombination of aggregates can be described by kinetic equations.14 The
aggregates permanently change their composition, and their mean aggregate
size is defined by the flow shear rate and the cohesion forces between the cells.
The rise of the shear rate in an experimental flow cuvette induces cell
detachment from the aggregates. The reduction of the shear rate leads to the
opposite process.
Figure 1.1 shows a typical distribution of the number of cells in the
aggregates versus the flow shear rate. It follows from such experiments that in
in vivo conditions, the aggregation structure of blood changes along the
vascular network, in which the shear rate may change from zero to several
hundred sec–1 corresponding to the values of shear stress.
At a certain value of the shear rate, the rouleaux become totally destroyed
and the blood behaves as a suspension of single cells. A further rise of the
shear rate induces a rotation (tank tread motion) of the cell membranes
around their inner contents Ref. 15 as well as the orientation and deformation
(elongation) of the cells.
8 Chapter 1
Figure 1.1 Number distributions of RBCs in aggregates of different sizes at initial

stationary flow and after a fixed time (from upside down) after its cessation. Along the
horizontal axis—the number of cells in one aggregate; along the vertical axis—the part of the
cells comprising the aggregates of a given size. From upside down: (1) stationary flow,
aggregation state; (2) after 20 (A, C) and 10 s (B); (3) after 50 (A, C) and 60 s (B); (4) after
120 (A, B) and 75 s (C).
The morphological studies of RBC aggregates in stasis or under shear

stress were performed in different flow conditions.15–21 A normal aggregation
process, started after stopping the shear flow that is destroying the aggregates,
finally results in the appearance of a three-dimensional network of rouleaux
and/or large aggregates containing up to more than 500 cells.
According to H. Schmid-Schoenbein et al.,22 the pathological aggregation
morphologically differs from the normal physiological one. While normal
aggregation usually terminates at a level of dispersed or loosely connected
rouleaux, in the cases of strong pathologies, aggregate clumps ranging in size
up to 500 mm can be seen. Typical shapes of RBC aggregates at normal and
pathological aggregation are compared in Fig. 1.2.
The elevated strength of cohesion between individual cells is the main
feature of pathological aggregation. The aggregates that are not destroyed by
the flow shear forces and block the blood microcirculation in in vivo
conditions are called sludges. The classification of aggregates in terms of
increasing strength (rouleaux ! rouleaux-clumps ! clumps ! rouleaux-
sludges ! sludges) as proposed in Ref. 23 reflects the possibilities of transition
from one- to three-dimensional structures.
Aggregation of RBCs is a feature typical of the blood of humans and
many other mammals, birds, amphibia, reptiles, fishes, etc., but it practically
does not happen in goat and bovine blood because of the normally spherical
shape of their RBCs.17 The shape and deformability of RBCs are related to
their aggregation properties and also determine the microrheological structure
of blood. An abrupt increase of osmotic pressure of blood plasma leads to
Figure 1.2 Microphotographs showing typical shapes of RBC aggregates at normal

(a) and pathological (b) aggregation (published with kind permission of H. Kiesewetter,
Institut für Transfusionsmedizin, Charité, Berlin).
creasing the cell membranes, which reduces the aggregation of RBCs. The
studies of RBC deformability showed24–26 that as the shear rate is increased
from zero to 200 s–1, the relative elongation p ¼ (ab)/(a þ b), where a and b
are the correspondingly long and short half-axes of the ellipsoid modeling the
elongated cell, increases quite rapidly, whereas at shear rates changing from
200 to 600 s–1 the cell relative elongation changes more slowly and does not
exceed the value of p ¼ 0.7. Long-lasting mechanical stress reduces the RBC
deformability. However, the deformability is gradually restored after some
time depending on the duration and intensity of the applied stress. Computer
analysis of the experimental data obtained by means of laser diffractometry27
allows one to conclude that at high shear rates the major factor of RBC
deformability is the viscosity of the intercellular medium (solution of
hemoglobin), while at low shear rates the major factors are the viscoelastic
properties of the membrane and the shape of the cells.
1.2 Importance of Quantitative Measurement of Red Blood Cell

Aggregation and Deformability Parameters
The altered aggregation and disaggregation behavior of erythrocytes reflects
the pathological states of the human organism. The elevated aggregation of
RBCs leads to generalized disorders of microcirculation. For example, the
experimental perfusion of mesentery blood vessels with RBC suspension
enriched with high-molecular-weight dextran leads to layering the flow and
blocking the vessels with sludges.28 In another experiment, the administration
into the blood flow of RBCs treated with neuraminidase neutralizing the
negative charge of the membranes induced drastic disturbances in the regional
blood flow in the organs.29 The blood flow in the spleen decreased by 60%, in
the liver 34%, in the lungs 22%, and in the feet 22%.
10 Chapter 1
In acute experiments carried out on dogs, the effect of RBC aggregation

on heart function was tested.12 The coronary system of the heart was
sequentially perfused with normal blood and with blood aggregated by high-
molecular weight dextran (M ¼ 5 105 kDa). The blood with altered
rheology showed a negative effect in all experiments with an acute (2 to 5
times) decrease of the coronary blood flow and a three to four times decrease
of the force and power of myocardium isometric contractions.
Many types of pathologically altered RBC aggregation, e.g., the
hyperaggregation syndrome complicated with reduced RBC deformability,
are not compatible with the life of an organism. Hence, the quantitative
determination of the macro- and microrheological properties of blood is an
important problem for medical laboratory diagnostics. This can partly be
performed by means of viscometry, a well-established technique widely used
in research and in clinics. However, this technique does not yield data on the
aggregation rate and the strength of aggregates.
The RBC aggregation behavior directly influences the results of well-
established routine analyses of blood by sedimentation measurements that
are usually performed in clinical rheological laboratories.30 These
measurements are influenced by a manifold of different parameters, such
as, e.g., the hematocrit, the plasma viscosity, the temperature, the form and
the size of erythrocytes, the composition of the blood plasma, and last but
not least, the aggregation properties.31 More specific and selective
information about blood rheology and, hence, the physical condition of
the human body is contained in the kinetics of RBC aggregation and
disaggregation and in the data on the forms and properties of erythrocyte
aggregates. Therefore, an analytical method of their measurement should
be clinically available.
Intensive development of optical quantitative blood aggregometry started
in the late 1960s, especially due to the works of H. Schmid-Schoenbein and co-
workers,21,22 who developed a technique for in vitro monitoring RBC
aggregation at different shear rates in a transparent viscometer and registering
the aggregation kinetics by means of photometry of light transmitted through
relatively thin blood layers. Later, other researchers extensively used this
approach (see, e.g., Ref. 30). Other experimental arrangements for transmis-
sion measurements include a thin glass capillary flow system (Hahn et al.31)
and a tapered rotation disc system (Jung et al.32). Tukhvatulin et al.17
developed a technique of photometric transmission measurements through
thin layers of blood affected by mechanical vibrations of variable intensity,
allowing for studying reversible aggregation of RBCs in microvolumes. Later
studies also introduced microfluidic slit rheometry for the measurement of the
aggregation parameters in whole blood samples.33,34
The three basic approaches currently used to assess RBC aggregation
in vitro as discussed above are schematically shown in Fig. 1.3.
Figure 1.3 Basic approaches for the determination of the erythrocyte aggregation and
disaggregation by optical measurements.
It is worth noting here that optical transmission measurements with blood

samples are possible only with samples of thicknesses not exceeding 100 mm.
This measuring distance is too small to assess a complete three-dimensional
aggregation process taking into account the large sizes of aggregates in
pathological cases as mentioned above. The assessment of RBC aggregation
and disaggregation processes in whole blood not perturbed by limiting factors
can be made only in samples of thicknesses exceeding 500 mm. This condition
can be met only by performing measurements in a backscattering mode.
A light backscattering technique relating the kinetics of the signal remitted
from thick layers of blood to the kinetics of spontaneous RBC aggregation
was initially introduced to hemorheological research by Usami and Chien35
12 Chapter 1
and Stolz et al.36 and Donner et al.37 Later Firsov et al.,38–41 Priezzhev
et al.,42,43 Hardeman et al.,44 and Potron et al.45 developed this approach
enabling researchers to perform quantitative measurements of kinetic indices
of RBC aggregation and estimate the hydrodynamic strength of the
aggregates. A cylindrical Couette cell is typically used for such measurements.
Numerical simulations of the backscattering signal from model blood
samples in different geometries performed using the Monte Carlo algorithm
were implemented for optimizing the operation of the device and signal
processing and interpretation.43,46
1.3 Arrangement of a Couette-Chamber-Based Laser

Backscattering Aggregometer
The arrangement of an aggregometer based on measuring the remitted signal
from a blood sample placed into the gap of a cylindrical Couette chamber is
more advantageous than other arrangements due to several reasons. In
particular, in this case, the blood-filled gap between the cylinders is oriented
vertically, so that slow sedimentation of RBCs does not manifest itself during
the aggregation measurements that typically last several minutes, and does not
perturb the results. A schematic layout of such an arrangement is shown in
Fig. 1.4.
The experimental set-up consists of a mechanical part, a receiving and
analyzing system, and a light source (typically a compact GaAs diode laser)
with an output power of about 1 mW. The mechanical part includes a Couette
chamber, an electric motor, and a reduction gear.
The Couette chamber device consists of two coaxial cylinders placed
one inside the other. A blood sample fills the gap between the cylindrical
surfaces. Typically the gap width is around 1 mm and the volume of the
Figure 1.4 Schematic layout of a rotational Couette cell-based erythro-aggregometer.

1. fiber coupled diode laser, 2. optical emitting and detecting head, 3. Couette chamber
comprising two co-axial cylindrical cups, 4. computer-controlled stepping motor with
reduction gear, 5. computer.
sample ranges from 2.3 to 2.5 mL. The outer (hollow) cylinder wall is
transparent to the illuminating light. The surface of the inner cylinder is
specially coated to eliminate reflection. The inner or outer cylinder can be
rotated with varying step rates, so that the shear rate in the fluid flow
inside the gap changes from the minimum value ranging from 1.5 to 2.5 s–1
to the maximum value ranging from 840 to 1500 s–1, depending upon the
design. A specific feature of Couette flows is that the magnitude of the shear
rate is the same everywhere inside. It is determined only by the rotation rate
of the cylinder and the radii of the cylinders. This means that all RBCs in
the gap are in similar hydrodynamic stress conditions that are nondepen-
dent on their exact location, which is very important for aggregation
measurements.
To perform the measurements, a collimated probe beam of an intensity-
modulated laser (lp ¼ 670 or 780 nm) illuminates a small volume of blood
(typically V ¼ 1–2 mm3, depending on the focusing) at the half-height level of
the gap. The incident light peak intensity in the probe beam is I0 , 0.3 mW/
mm2. Multiple tests performed with blood samples showed that laser
irradiation of blood at these wavelengths and of such an intensity affects
neither the rheological nor the optical properties of a sample throughout the
measurement procedure, which typically lasts about 10 minutes.
After passing through the transparent wall of the outer cylinder, the
probing beam is scattered multiple times by RBCs all throughout the depth of
the blood sample. A fiber-optic unit consisting of several detecting apertures
placed outside the chamber in the vicinity of the probing beam detects a
portion of the incident light scattered in the backward direction. The process
of diffusion and the multiple scattering of the photons in the blood volume
can be considered as follows. Part of the probe light is scattered by the first
layer of RBCs. As discussed in Chapter 2, due to the high anisotropy of blood,
most of the photons are scattered at low angles. They are sequentially
scattered in the deeper and deeper layers. Only a small portion of the photons
that are multiply scattered at large angles reach the detector apertures at the
outer surface of the sample and contribute to the output signal. Experiments
show that the apertures located at different positions relative to the incident
probe beam and flow direction detect different numbers of photons
backscattered from flowing RBCs. This phenomenon of scattering asymmetry
is considered in more detail in Ref. 47.
1.3.1 Measurement procedure

The measurements are performed with freshly drawn blood, usually stabilized
with EDTA. After the blood sample is poured into the measuring gap, the
maximum rotation rate of the cylinder is applied so that all aggregates are
destroyed and individual RBCs become completely separated and deformed
(elongated) by the shear forces. Note that this procedure does not destroy the
14 Chapter 1
membranes of individual cells. At this phase, the detectors yield an output

signal of a constant level.
After this level is fixed, the rotation of the cylinder is instantaneously
halted, and the shear stress in the sample ceases. During a fraction of a
second, the directed motion of the blood stops and RBCs attain their
normal discoid shape and chaotic orientation. This gives rise to the
remitted signal. Indeed, our model calculations show that a thick layer of
closely packed elongated and oriented spheroid particles scatters less light
in backward direction than that of stochastically oriented discoids of
similar volume. The difference in signal levels corresponding to the
maximum flow rate and stopped flow conditions is proportional to the
deformability of RBCs. Hence, this difference can be calibrated in terms of
a deformability index. The absolute value of this difference depends on the
geometry of the detecting unit, which can be optimized for RBC
deformability measurements.
Having re-attained their initial shape, the RBCs start to spontaneously
aggregate. Along with cell aggregation, the remission signal monotonously
decreases. The process of spontaneous aggregation consists of several
stages, with each stage corresponding to the formation of aggregates of
different sizes and structures. As soon as the growth of three-dimensional
aggregates or the rouleaux network stops, the remission signal reaches
its lowest level. In the experiments, the full aggregation occurs in ca.
2 minutes after cessation of the flow. The measured curve of the remitted
signal intensity versus time is strongly nonlinear. So far there is no exact
theory of light scattering from such large irregularly shaped particles as
aggregating RBCs. Hence, we cannot apply a best fitting procedure to
compare the experimental and theoretical kinetic curves and to straight-
forwardly obtain the time-dependent size and shape parameters of the
growing aggregates. However, digital simulations and model calculations
based on well defined approximations yield the monotonously decaying
remission signal kinetics.48 This proves the correctness of experimental
signal interpretation.
After obtaining a constant signal corresponding to full aggregation of the
RBCs, the shear rate is gradually increased until the full disaggregation takes
place again. With the increasing rotation velocity, the remission signal
increases. It achieves a maximum when all erythrocytes are again separated. A
further increase in the shear rate leads to the deformation of the cells and to a
reduction of the remission signal. If the movement is suddenly stopped at this
phase, the RBCs again attain their original form and the remission signal
immediately increases. After that the signal decreases further because of the
beginning aggregation process. This procedure of aggregation and disaggre-
gation measurements could be repeated several times with the same blood
sample.
Figure 1.5 Typical time course of remission signal measured with a blood sample drawn
from a healthy individual (Hct ¼ 40).
1.4 Kinetics of the Aggregation and Disaggregation Process

in Whole Blood Samples
The left part of Fig. 1.5 shows the time course of the remission signal
attributed to the aggregation kinetics typical of the blood of a healthy donor.
On the right side of this figure, a typical course of the remission signal during
the step-wise induced disaggregation process is shown.
1.4.1 Determination of the characteristic parameters of the

aggregation and disaggregation process in whole blood
samples
Extensive investigations were performed to determine the measurement
parameters that characterize the aggregation and disaggregation kinetics. The
selection of these parameters can be made according to the time course of the
detected remission signal during the aggregation and disaggregation
processes. If the remission signal presented in Fig. 1.6 is transferred into a
semilogarithmic plot, the best fitting approximation of the measuring curve
yields two straight lines with different slopes, characterized by the angles a1
and a2 (Fig. 1.7). The parameters T1 ¼ ctg a1 and T2 ¼ ctg a2 are attributed
to the formation time of the small and large aggregates. These parameters are
used in the following for the characterization of the aggregation process of the
erythrocytes.
In correspondence with the best fitting approximation of the measuring
curve, the kinetics of the RBC aggregation process in our experimental
conditions is best fitted by the following function:
Tt Tt
I a¼ C1 e 1 þ C2 e 2 þ C3 (1.1)
where Ia is the full amplitude of aggregation; T1, T2 are the characteristic

times of the linear and three-dimensional aggregates’ formation; t is the
current time; C1, C2, C3 are constants.
16 Chapter 1
Figure 1.6 Kinetics of the remission signal during the aggregation process.
In analogy with the aggregation process, the kinetics of disaggregation can

be analyzed. For every value of the step rate introduced in the share rate, the
corresponding remission signal can be determined. The obtained relation
between the remission signal and the shear rate has a similar course as in the case
of the aggregation kinetics. The characteristic parameters, b1 and b2, are related
to the hydrodynamic stability (strength) of the large and small aggregates,
respectively.
bg bg
I d I g ¼ C4 e 1 þ C5 e 2 (1.2)
Here, g is the shear rate; Id is the amplitude of disaggregation; Ig¼0 is the
amplitude of the remission signal at g ¼ 0; b1, b2 are the hydrodynamic
strengths of large and small aggregates; C4, C5 are the constants.
1.5 Parameters Influencing the Aggregation and Disaggregation

Measurements
1.5.1 Effect of blood sample temperature
Hahn et al.31 measured the influence of the temperature of blood samples on
the remission signal using a double Ulbricht sphere arrangement. It was
Figure 1.7 Approximation of the remission signal during the aggregation process.
shown that the temperature change ranging from 28 to 37°C has no effect on
the remission signal at flow conditions. As for the temperature dependence of
the aggregation properties of RBCs in stasis or at low shear stress, the
experimental results obtained by different authors are rather controversial.
Our experiments, performed with a Couette system and blood samples of
healthy individuals,49 show that though the scatter of the measured
parameters is rather high, the characteristic times T1 and T2 do not regularly
depend on temperature from 2 to 37°C. The main conclusion is that the
aggregation parameters of RBCs can be reproducibly measured at room
temperatures (20–25°C) without any corrections. Temperatures higher than
37°C induce a quick increase in the aggregation rate (by about three times at
45°C). In the temperature range of 45 to 50°C, the adequate measurements are
hardly possible due to the thermal denaturation of blood. In some
pathological cases (e.g., cryoglobulinemia, Sjogren disease) a sharp decrease
of T1 and an increase of the strength of aggregates, b, were recorded.
However, other authors using different techniques and conducting
experiments in different conditions report somewhat different and variable
18 Chapter 1
results (see, e.g., Refs. 50–52). In one such experiment performed by means of
photometric aggregometry,50 20 blood samples of normal donors and 20
blood samples of patients with venous ulcers of the leg were examined at 3, 10,
20, 30, and 37°C. It was shown that with decreasing temperature, red cell
aggregates become more resistant to hydrodynamic dispersion and more
prone to growing under low shear stress. It is concluded that a decrease in
temperature causes an increase in the adsorptive energy of red cell
aggregation, which is most likely due to an increase in molecular adsorption
stress. RBC aggregate formation as an overall process is retarded by a
decrease in temperature, which is primarily due to an increase in plasma
viscosity that causes an increased damping of aggregate formation.
Accordingly, the rate constant of aggregate formation corrected for plasma
viscosity increases with decreasing temperature. The authors explain the
obtained temperature dependence of the kinetic parameters by a theoretical
model that suggests an increase in the contact area between aggregating red
blood cells as the rate-limiting step of red cell aggregation. They conclude that
as a whole, red cell aggregation is favored by a lowering of temperature.
1.5.2 Effect of blood sample oxygenation

The dependence of the remission signal on the measuring wavelength and the
degree of blood sample oxygenation is presented in Fig. 1.8.
The measurements were performed using the double Ulbricht sphere
arrangement.53 The absolute level of the remission signal depends on the degree
of oxygenation of the blood sample and on the measuring wavelength. However,
the time courses of the signals measured along the aggregation and disaggreg-
ation processes with samples of different oxygenation are absolutely similar.
Figure 1.8 Effect of blood sample oxygenation on remission intensity.53

The effect of oxygenation on the level of the remission signal can be reduced if
the measuring wavelength is chosen near the isobestic point of blood at 805 nm.
Whether or not the determined dependence of the remission signal on the
degree of oxygenation influences the kinetic measurements of whole blood
samples in the rotating cylinder system was checked using two different diode
lasers. One of them emitted at a wavelength of 670 nm, at which the
oxygenation effect has a significant influence on the remission signal, whereas
the other one was at 780 nm, at which the influence can be neglected. Long-
term measurements showed no difference in the time course of the remission
signal during aggregation and disaggregation at these wavelengths. Taking
into consideration these results, the effect of oxygenation on the remission
signals could be neglected under the conditions of the standard procedure of
blood sample preparation. This result is comprehensible, because blood
samples are kept in very limited contact with air during sampling and
preparation. The open surface of the sample in the gap, i.e., in contact with
the air, is very small in comparison to the whole sample surface.
1.5.3 Effect of sedimentation

Investigations were carried out to check whether or not there is any influence of
the sedimentation processes on the remission signal during the aggregation
kinetics measurements. A blood sample was placed into the gap of the Couette
chamber, and the remission signal measurement was started immediately without
cylinder rotation. The detecting optical fibers were located near the surface of the
blood sample at different heights along the gap. No sooner than two or three
minutes after the intensity of the remission signal stabilized at a level of full
aggregation, as shown in Fig. 6, were small deviations in the signal intensity
observed. These fluctuating deviations developed, on average, as the decrease of
the remission slowly developed over time. They are similar to the fluctuating
kinetics of sedimentation reported by Voeikov54 and Kondakov et al.55
The time dependence of the maximum and minimum levels of signal
intensity, between which the fluctuation takes place, is schematically presented
in Fig. 1.9 for blood samples from five different donors.53 This figure shows
that during the aggregation measurements inside the time interval of 2 minutes,
the sedimentation process does not influence the remission signal.
1.5.4 Effect of hematocrit

The hematocrit values have a strong influence on the absolute level of the
remission signal and on the time course of the measured curves. This
dependence is quite obvious as the probability of RBC contacts needed for
aggregation is proportional to the hematocrit. For the analysis of different
blood samples, it is necessary to perform measurements with the same
hematocrit value, e.g., Hct ¼ 40%. Therefore, the blood samples have to be
dissected before the measurements. This time-consuming procedure will be
20 Chapter 1
Figure 1.9 Effect of RBC sedimentation process on the remission signal intensity
(variation limits).
unnecessary if the device-specific dependence of the measuring characteristic

parameters on the hematocrit values is determined. By using this relation, the
characteristic parameters determined for any hematocrit value can be
calculated for the fixed hematocrit value used for comparison of the
measuring results.
1.6 Comparison of Aggregation and Disaggregation

Measurements with Sedimentation Measurements
Sedimentation measurements are the standard method for blood analysis in
rheological laboratories. It is based on the phase separation of the cell
components from blood plasma. The sedimentation velocity is a value that is
influenced directly or indirectly by different factors, such as aggregation of
erythrocytes, hematocrit, plasma viscosity, and temperature. Lademann
et al.56 compared the information obtained by aggregation and disaggregation
measurements and by sedimentation measurements of the same blood
samples. Three different types of blood samples were investigated:
1. blood samples of patients with symptoms of an illness and an increased
blood sedimentation velocity;
2. blood samples of patients with symptoms of an illness and a normal blood
sedimentation velocity; and
3. blood samples of healthy volunteers with a normal blood sedimentation
velocity.
Clear differences of the aggregation and disaggregation parameters were
found in the first case in comparison to the parameters determined for healthy
volunteers. Aggregation and disaggregation measurements allowed the
determination of pathological states such as in the case of sedimentation

measurements.
In the second case, it was not possible to determine pathological states by
sedimentation measurements, regardless of the illnesses of the patients. Using
aggregation measurements, it was also not possible to detect pathological
states. Only the disaggregation parameters showed characteristic differences
in relation to the parameters of healthy volunteers. The reason for the higher
sensibility of the disaggregation measurements in comparison to the
aggregation measurements can be the integral determination of the
disaggregation parameters during the interval when the cylinder rotation
rate is constant. The aggregates did not change their shapes during this time in
contrast to the rapid changes during the spontaneous aggregation.
The investigation demonstrated that the aggregation and particularly the
disaggregation measurements are better suited for the detection of pathologi-
cal stages by blood analysis than the sedimentation measurements. Further-
more, it is possible to determine the classes of diseases that are the reasons for
the pathological states by aggregation and disaggregation measurements.
1.7 Laser Tweezers as a New Tool for Studying RBC

Aggregation at the Single-Cell Level
In order to fully assess the mechanism of RBC aggregation at the single cell
level, a number of methods that allow for studying the cell-cell interactions
were developed in the 1970s. The micropipette aspiration technique, widely
used to study the interactions between pairs of cells, allowed to make the first
quantitative measurements of the energy of interactions between RBCs. Later
on, the methods of atomic force microscopy (AFM) and laser tweezers (LT)
were applied to study the interaction of RBCs. The advantage of AFM is that
it allows one to measure forces from 10 to 1000 pN,57–59 which is hardly
achievable with LT. A series of studies performed with the AFM tip
functionalized by fibrinogen macromolecules showed the strong interaction of
the tip with the RBC membrane.60 Measurements in dextran solutions using
AFM showed good correlation with the existing interaction mechanism of
RBCs for neutral macromolecules.58
Compared to AFM, LT allow one to measure forces more precisely starting
from ranges of 0.1 pN to 100 pN, and is available to manipulate cells without
any mechanical contact.61 The interaction forces between RBCs measured with
LT in autologous plasma were found to be in the range of a tenth of a pN.62–64
The LT also allowed for measuring the interaction between RBCs in carefully
controlled interaction options (e.g., area of contact and time of interaction).63,64
These techniques enabled the researchers to study and elucidate the
mechanics of RBC aggregation by measuring the RBC interactions on cell
doublets. The latest works, especially with laser tweezers, allowed the
22 Chapter 1
obtaining of detailed kinetics of RBC interaction and provided the grounds

for new ideas about the mechanics of aggregation and the role of aggregation-
inducing factors.9,62,64 Therefore, the single cell study techniques are crucial
for understanding the aggregation mechanics, factors, and parameters that
have yet to be discovered.
1.7.1 Laser tweezers operation principle and experimental

arrangement
Laser tweezers allow for measuring the cell-to-cell interactions without
mechanical contact with the cells. The tightly focused laser beam can exert
forces from femtonewtons to hundreds of piconewtons. Therefore, LT can be
used to trap and manipulate single cells without damaging them. Once the
appropriate calibration is done, the device can be used to measure very low
forces. This method was proposed by Arthur Ashkin in 1970, and during the
last few decades it had been used widely in the field of biophotonics and has
been nominated as one of the most promising findings of the 21st century.65,66
To study RBC aggregation, double channel LT were typically used.
In the studies of RBCs’ aggregation, the LT were first used by
Bronkhrost.9 Even though force measurements were not performed, the
interaction time and surface area dependence of the RBCs’ aggregation force
were observed. Later on, these parameters were quantitatively measured by
other authors.62–64,67 It was shown that in plasma, the interaction forces
between RBCs are around 10 pN.
A typical schematic layout of an LT setup is given in Fig. 1.10. The
experimental setup consists of the following main parts: laser, objective, and
camera used to obtain the image. The experimental setup uses a tightly focused
laser beam to trap the cells. The tightly focused laser beam forms an optical
Figure 1.10 Schematic layout of a double channel laser tweezers’ setup.

trap—a light field gradient that traps the nearby microparticles. Typically a
microparticle in the optical trap is considered as a particle attached to a spring
(optical trap) with stiffness k. The farther away the particle is from the beam
focus, the higher is the returning force applied by the optical trap. In a typical
study of RBCs’ interactions, the forces are matched to the maximum returning
force of the optical trap. When a trapped cell escapes from the optical trap, the
interaction force has exceeded the trapping force.
The problem of cells’ heating is solved by matching the laser wavelength
to the minimum of the RBC absorption. The calculations made in Ref. 62
showed that the effect of heating is a few K per every 100 mW of laser power
(depending on the chamber size and geometry). The laser power used in the
experiments typically does not exceed 50 mW. Therefore, the laser heating
effect on cells is negligible. Experimental observations proved that trapping an
RBC for several minutes causes no visible changes in the cell. The typical
measurement time for one cell is 1–2 minutes. Using a laser beam power of
50 mW is sufficient for measuring interaction forces ranging from 0.5 to
30 pN with 10–20% deviation.
The setup records the images from forward-scattered light to a video
camera. The dynamic video fragments are analyzed using adequate software
to obtain detailed information about the measurement procedure and the
results. Typical images of the RBCs before and after trapping are presented in
Fig. 1.11. It is seen that the cells tend to rotate to a side-oriented state because
the optical trap exerts more longitudinal forces than lateral forces.
The measurements of RBCs’ interactions in vitro with LT are typically
performed with cells suspended in the autologous plasma or some other
solution. Therefore, the parameters of interaction can be precisely measured
Figure 1.11 Sequences of cell trapping by laser tweezers in a diluted solution of RBCs:
(a) single cells on the glass surface, (b) two individual RBCs trapped in double-channel laser
tweezers.
24 Chapter 1
depending on both the cell membrane properties and the solution content.
LTs allow for measuring the pure aggregational interaction between the two
RBCs during the formation of a doublet as the first step in the formation of
larger aggregates, independent of the other cells.
1.7.2 Sample preparation and measurement procedure

The chambers used for LT measurements typically consist of two glass plates
separated by a small gap of 0.1 mm and filled with a highly diluted suspension of
blood cells (0.5%). The measurements are performed with freshly drawn blood
stabilized with EDTA. In the experiments with autologous plasma, it is crucial
that the plasma is free of platelets as they tend to be trapped together with RBCs
and interfere with the measurement. A portion of platelet-poor plasma is
obtained from whole blood by centrifugation and a drop of blood is added into
it. The suspension is put into the chamber and RBCs are allowed to settle down.
The individual non-interacting RBCs are lifted from the surface and are used for
experiments. The RBCs on the bottom of the chamber form a monolayer, as
shown in figure 2, and their shape remains discoid for 4–5 hours.
Most of the experiments using laser tweezers to study RBC aggregation are
performed operating with two cells. Therefore, one experiment typically contains
the results obtained from several dozens of RBC doublets. Even though the
measurements conducted on one sample cannot comprise a large ensemble of
cells (millions) as in whole blood aggregometry, the measured parameters do not
have significant variations and are reproducible. Typical deviations of measured
parameters are within 20–30%. Deviations are mostly caused by individual
differences between cells. The measurements can distinguish the aggregation
parameters of the pathological and healthy donor cells fairly well.64
The measurements with laser tweezers of the parameters characterizing
RBC aggregation and disaggregation are somewhat different.
1.7.2.1 Measurement of the disaggregation force

Figure 1.12 presents the sequences of measurement of the disaggregation forces.
We refer to the disaggregation force as the force required to separate the
individual cells in an RBC doublet. The measurement sequences are the
following: (1) two independent non-interacting cells are trapped in separate laser
tweezers; (2) the cells are lifted from the bottom of the chamber; (3) the cells are
Figure 1.12 Sequences of disaggregation process measurement with laser tweezers.

1) Two individual RBCs trapped in different LTs; 2) RBCs attached manually and held
together; 3-4) sequential separation of RBCs with laser tweezers.
Figure 1.13 The different end-points of disaggregation (Ref. 62).
attached to each other with the desired initial interaction surface and interaction
times; (4) laser tweezers are moved away from each other to separate the RBCs
at a constant trapping power; (5) the minimum force required to separate the
RBCs is found by repeating step 4 while slowly increasing the trapping power.
The typical value of the force, required to separate the cells overlapped in the
RBC doublet in autologous plasma was found to be around 10 pN.62–64
In a number of works, it was shown that the disaggregation process could
proceed in three different ways as shown in Fig. 1.13.62–64 In the first case, an
unbreakable point contact between cells is observed at the final stage of
disaggregation; in the second case, the disaggregation process is characterized
by strong interaction forces within a certain area of the cells’ surfaces’ overlap;
in the third case, the RBC doublet is easily fully disaggregated.
The force required to disaggregate RBCs in autologous plasma is
dependent on the interaction time. During the first 2 seconds, the cells
disaggregate easily, and after that the interaction becomes stronger.9 The
quantitative measurements proved that the disaggregation force increases
during the first few seconds.58,64
1.7.2.2 Measurement of the aggregation force
In the case of the measurement of the aggregation process, the velocity of
spontaneous aggregation is measured as shown in Fig. 1.14. The measurement
sequences are the following: (1) two independent non-interacting cells are
trapped in separate laser tweezers; (2) the cells are lifted from the surface of
the chamber; (3) the cells are attached to each other with a local interaction
surface; (4) laser tweezers are turned off; (5) RBCs start to spontaneously
overlap; (6) the velocity of the process is measured. The typical value of the
Figure 1.14 Typical measurement steps: 1) two individual RBCs are trapped in different
laser tweezers (a); 2) the cells are attached with laser tweezers after which the latter are
switched off (b); 3-5) the velocity of RBCs spontaneous overlapping is measured (c-e).
26 Chapter 1
aggregation velocity is found to be 0.3 mm/sec. It was shown that this

parameter is significantly changed in pathological states.64
The aggregation force was defined as the minimum trapping force
required to stop the RBCs’ spontaneous aggregation (overlapping). The forces
of the RBCs’ aggregation in autologous plasma were found to be significantly
lower than the forces required to disaggregate the cells.63
To conclude, the initial studies of RBC aggregation conducted with laser
tweezers have proved a good perspective of their future use in this field for
ensuring a better assessment of the parameters, factors, and mechanics of this
complicated phenomenon.
1.8 Hemorheological Characterization of Various Diseases

by Aggregation and Disaggregation Measurements
of Blood Samples
The kind and concentration of biomolecules that influence the aggregation
and disaggregation properties of the erythrocytes are altered in the case of the
pathological states of the human body. These alterations are reflected by the
changes in the aggregation and disaggregation parameters that are different
for healthy people and patients.
Different attempts were carried out to use the aggregation and
disaggregation measurements not only for the detection but also for the
classification of pathological diseases.
Lademann et al.68 showed differences in the aggregation and disaggre-
gation parameters of blood samples obtained from healthy volunteers and
patients suffering from diabetes mellitus type I. The blood glucose
concentrations of the patients determined in the rheological laboratory varied
from 120 mg/dl to 180 mg/dl in the investigated blood samples. Those of
healthy donors were ≤75 mg/dl. A hematocrit of 40% at 37°C was established
by diluting the cell-mass in the autologous plasma.
A comparison of the values of the measured characteristic aggregation
and disaggregation parameters obtained for healthy donors and patients is
presented in Fig. 1.15. The values of all parameters show clear differences
when comparing the blood samples from healthy donors and patients
suffering from diabetes. The strongest difference could be found for the
disaggregation parameters.
These findings are supported by the results of Babu et al.69 who showed
that with the increase of glucose concentration, the erythrocytes’ aggregation
is enhanced in diabetic patients compared to that of healthy subjects.
RBC aggregation affects low shear blood viscosity and microvascular flow
dynamics, and was markedly enhanced in several clinical states. Until recently,
most research efforts were focused on the relations between the suspending
medium composition (i.e., protein levels, polymer type, and concentration) and
aggregate formation. However, there is now an increasing amount of
Aggregation Disaggregation
100
80
values of the parameters
60
40
20
0
T1 T1 T2 T2 β1 β1 β2 β2
healthy donors patients suffering diabetes
Figure 1.15 Aggregation and disaggregation parameters of blood samples from healthy
donors and patients suffering from diabetes.
experimental evidence indicating that RBC cellular properties can markedly

affect aggregation, with the term “RBC aggregability” used to describe the
cell’s intrinsic tendency to aggregate. Variations of aggregability can be large,
with some changes of aggregation substantially greater than those resulting
from pathologic states. The work by Rampling et al.70 provides an overview of
this topic, and includes such areas as donor-to-donor variations, polymer-
plasma correlations, effects of RBC age, effects of enzymatic treatment, and
current developments related to the mechanisms involved in RBC aggregation.
Hemorheological disorders are involved in the development of ischemic
heart disease and represent a factor that plays a significant role in the
pathogenesis of its development. The authors of Ref. 71 found that the
erythrocyte aggregability increased almost two times in the blood of investigated
patients as compared to the healthy control group. The aggregability was
positively correlated with the severity of the disease. The most pronounced
hemorheological disorders were found in patients with heart failure. Studying the
alterations of the ability of erythrocytes to aggregate in the conditions of brain
ischemia, the authors of Ref. 72 have shown that endothelialy derived NO
correlates with erythrocyte aggregability and has a positive impact on the
restoration of cerebral blood flow in the initial stage of acute brain ischemia.
A comprehensive analysis of the influence of different classes of diseases
on the aggregation and disaggregation parameters was performed by Firsov
et al.73,74 They found a clear correlation of the measured aggregation and
disaggregation parameters with different diseases. The results are summarized
in Table 1.1.
The morbidity of several diseases often causes heavy pathological states. The
rheology and, consequently, the aggregation and disaggregation kinetics in these
28
Table 1.1 Correlation of the aggregation and disaggregation parameters measured in whole blood samples from healthy donors and patients
suffering from different diseases.73
Disease N T1, s ± RMS T2, s ± RMS I02.5, % ± RMS b1, s–1 ± RMS b2, s–1 ± RMS
Chronic glomerulonephritis 31 9.4 ± 0.6*** 62.2 ± 1.6*** 17.5 ± 2.7** 32.0 ± 3.8 was not distinguished
The same with nephrotic syndrome 17 6.4 ± 0.9*** 61.5 ± 2.8*** 3.3 ± 4.5*** 46.0 ± 7.3** was not distinguished
Systemic lupus erythematosus 8 8.6 ± 0.5*** 55.6 ± 8.3** 10.4 ± 5.2** 35.9 ± 4.0* was not distinguished
Hereditary hypercholesteremia 48 11.4a ± 0.8* 35.0a ± 1.1*** 15.8b ± 1.5*** 45.7 ± 2.8*** 84.4c ± 6.3
Pulmonary hypertension 19 18.0 ± 1.2*** 45.8 ± 1.7* 23.5 ± 2.8 26.0 ± 3.3 70.0 ± 5.0
Intestinal tumors preoperatively (.60 years) 10 14.1 ± 1.9* 46.2 ± 4.0 14.3 ± 3.3** 56.0 ± 4.0*** was not distinguished
Psoriasis 57 12.5 ± 0.5* 40.0 ± 1.2 22.2 ± 1.2* 28.0 ± 1.4 61.6d ± 6.6
Psoriatic arthritis 45 8.2e ± 0.6*** 51.6 ± 1.4*** 1.7 ± 3.2*** 57.3e ± 5.6*** was not distinguished
Coronary disease 22 6.9 ± 0.2*** 37.9 ± 1.0** was not distinguished 33.0 ± 1.0** 55.2 ± 2.3
Coronary disease with diabetes 17 6.2 ± 0.1*** 34.1 ± 0.5*** was not distinguished 30.0 ± 0.9* 57.3 ± 2.0
Healthy donors 40 12.5 ± 0.6 41.4 ± 0.8 25.7 ± 0.8 26.3 ± 1.1 was not distinguished
Significance of difference between the group of patients and the healthy donors is designated with * for p , 0.05, ** for p , 0.01 and *** for p , 0.001. a n ¼ 40; b n ¼ 33; c n ¼ 14; d n ¼ 39;
e
n ¼ 11.
Chapter 1
Figure 1.16 Aggregation and disaggregation kinetics in a blood sample obtained from a
patient in a strong pathological state.68
cases are usually completely destroyed, as is shown in Fig. 1.16 for a blood
sample obtained from a patient suffering from cancer.68 The aggregation and
disaggregation parameters are strongly different for those of healthy persons.
Significant variations of the results obtained by different authors using
different equipment, sample preparation procedures, and experimental
protocols when measuring the RBC aggregation and disaggregation
parameters raised the issue of standardization of this work, which turned
out to be very difficult issue to solve. As a partial solution of this problem, an
international group of experts has set up new guidelines for hemorheological
laboratory techniques, including the RBC aggregometry in Ref. 75.
Summarizing the outlined results, we can state that the described optical
measurement techniques designed for studying blood aggregation and
disaggregation are well suited for the determination of the pathological status
of patients suffering from different diseases. The class of diseases can be
estimated. This method will produce broad practical applications in medical
healthcare, once it becomes possible to standardize both the sample
preparation and measurement protocols and the algorithms of the presenta-
tion and analyses of the results.
Acknowledgments
Dr. Priezzhev acknowledges the support of this work by the Russian Science
Foundation grant #14-15-00602.
References
1. J. S. K. Fung and P. B. Canham, “The mode and kinetics of the human
red cell doublet formation,” Biorheology, 11, 241–251 (1974).
2. M. J. Dunlop, M. Martin, P. B. Canham, and C. P. S. Taylor, “Kinetics
of adhesive interaction in vitro of human erythrocytes in plasma,”
Microvasc. Res., 28, 62–74 (1983).
30 Chapter 1
3. V. A. Levtov, S. A. Regirer, and N. H. Shadrina, a) «Aggregation and

diffusion of erythrocytes,» Modern Problems of Biomechanics, 9: Blood
Rheology and Microcirculation, Moscow, 5–33 (1994), b) Blood
Rheology, Medicine Publ., Moscow (1982), c) “On red blood cell
aggregation,” Contemporary Problems of Biomechanics, G. G. Chernyi
and S. A. Regirer-eds, CRC Press, Boca Raton-Ann Arbor-Boston,
55–74 (1990); and S. A. Regirer, “Diffusion of blood cells,” Contempo-
rary Problems of Biomechanics, G. G. Chernyi and S. A. Regirer-eds,
CRC Press, Boca Raton-Ann Arbor-Boston, 75–98 (1990).
4. L. Dintenfass, a) “Execution of “ARC” experiment on space shuttle
“Discovery” STS 51-C: some results on aggregation of red blood
cells under zero gravity,” Biorheology, 23(4), 331–347 (1986); b)
“Experiment on STS 51-C: effect of weightlessness on the morphology
of aggregation of human red cells in disease,” Report at the 38th
Congress of the International Astronautical Federation, Brighton,
October 10–17 (1987).
5. A. Katchalsky, D. Danon, A. Nevo, and A. de Vries, “Interaction of
basic polyelectrolytes with the red blood cell,” Biochem. Biophys. Acta.,
33, 120–138 (1959).
6. S. Chien, “Electrochemical and ultrastructural aspects of red cell
aggregation,” Bibl. Anat., 11, 244–250 (1973).
7. M. W. Rampling, “Plasma-protein induced aggregation of erythrocytes:
its causes, estimation, and effects on blood flow,” Stud. Biophys., 134,
91–94 (1989).
8. N. Maeda, Y. Izumida, A. Seiyama, N. Tateishi, and T. Shiga,
“Interaction between plasma high molecular weight proteins and
erythrocytes on the erythrocyte aggregation,” First World Congress of
Biomechanics, La Jolla, CA, USA, 2, 247, 1990.
9. P. J. H. Bronkhorst, J. Grimbergen, G. J. Brakenhoff, R. M. Heethaar,
and J. J. Sixma, “The mechanism of red cell (dis)aggregation
investigated by means of direct cell manipulation using multiple optical
trapping,” Brit. J. Haematol., 96, 256–258 (1997).
10. S. M. Bychkov and S. A. Kuz’mina, “The role of glicozaminoglicans and
proteoglicans in erythrocyte aggregation and adhesion,” Bull. Eksper.
Biol. Med. (Moscow), 3, 284, 1977.
11. S. Chien, K.-M. Jan, and S. Usami, “Roles of electrostatic repulsion on
mechanical shearing in red cell aggregation,” Bibl. Anat., 13, 93–94
(1975).
12. L. Dintenfass, Blood Rheology. Viscosity Factors in Blood Flow, Ischemia
and Trombosis, Butterworth, London (1976)
13. M. J. Gregersen, B. Peric, S. Usami, and S. Chien, “Relation of
molecular weight of dextran to its effects on viscosity and sedimentation
rate of blood,” Bibl. Anat., 4, 58–61 (1964).
14. T. Murata and T.W. Secomb, “Effects of shear rate on rouleaux

formation in simple shear flow,” Biorheology, 25, 113–122 (1988).
15. T. Fischer and H. Schmid-Schoenbein, “Tank-tread motion of red cell
membranes in viscometric flow: behavior of intracellular and extracellu-
lar markers,” Blood Cells, 3, 347 (1977).
16. L. Dintenfass, H. Jedrzejczyk, and A. Willand, “Photographic,
stereological and statistical methods in evaluation of aggregation of
red cells in disease. Part I,” Biorheology, 19, 567–577 (1982).
17. R. T. Tukhvatulin, V. A. Levtov, V. N. Shuvaeva, and N. H. Shadrina,
“Aggregation of erythrocytes in blood placed in macro- and microcuv-
ettes,” Fiziol. Zhurnal SSSR (Moscow), 72 (6), 775–784 (1986).
18. T. L. Fabry, “Mechanisms of erythrocyte aggregation and sedimenta-
tion,” Blood, 70(5), 1572–1576 (1987).
19. N. H. Shadrina, L. A. Strel’nikova, Yu. I. Levkovich, and V. A. Levtov,
“Study of erhythrocyte aggregation in flowing blood by a microphotog-
raphy technique,” Fiziol. Zhurnal SSSR (Moscow), 62, 214–217 (1974).
20. V. A. Levtov, N. I. Nikiforov, A. S. Poppel, and N. H. Shadrina, “On
the study of aggregation properties of blood,” Regional and Systemic
Blood Circulation, Leningrad, 49–59, 1978.
21. H. Schmid-Schoenbein, P. Gaehtgens, and H. Hirsch, “On the shear rate
dependence of red cell aggregation in vitro,” J. Clin. Invest., 47(6), 1447–
1454 (1982).
22. H. Schmid-Schoenbein, H. Reiger, G. Gallash, and H. Schartner,
“Pathological red cell aggregation (clump aggregation),” Recent Adv.
Clin. Microcirc. Res. Part 2, Basel, etc., 484–489 (1977).
23. L. Dintenfass, Blood Viscosity, MTR Press, Ltd. (1985)
24. M. Besis and N. Mohandas, “Deformability of normal, shape-altered
and pathological red cells,” Blood Cells, 1, 315 (1975).
25. C. Allard, N. Mohandas, and M. Besis, “Red cell deformability changes
in hemolytic anemias estimated by diffractometric methods (ektacyto-
metry),” Blood Cells, 3, 209 (1977).
26. W. Groner, N. Mohandas, and M. Besis, “New technique for measuring
erythrocyte deformability with ektacytometry,” Clin. Chem., 26(10),
1435 (1980).
31. G. Wolf, R. Bayer, and D. Ostuni, “Stress-induced rigidification of
erythrocytes as determined by laser diffraction and image analysis,” Opt.
Eng., 31 (7), 1475 (1992).
28. Y. Suzuki, M. Soutani, N. Tateishi, and N. Maeda, “Changes of flow
behavior of erythrocytes in microvessels and flow resistance accompa-
nied by erythrocyte aggregation,” Biorheology, 33 (1), 85 (1996).
29. H. Schmid-Schoenbein, Fluid dynamics and hemorheology in vivo: the
interaction of hemodynamic parameters and hemorheological “proper-
ties” in determining the flow behavior of blood in microvascular
32 Chapter 1
network, Clinical Blood Rheology, D.O. Gordon and M.D. Lowe–

Editors, CRC Press, Inc., Boca Raton, Florida (1988).
30. H. Kiesewetter and H. Radtke, “Blood sedimentation: an old clinical
procedure with new aspects,” Klin. Wochenschr., 61 (13), 621–624
(1983).
31. A. Hahn, A. Roggan, and D. Schädel, Minimal Invas. Med., 7, 79–90.
(1996).
32. F. Jung, A. Seegert, H. G. Roggenkamp, C. Mrowietz, H. P. Nüttgens,
H. Kiesewetter, H. Zeller, and G. Müller, “Simultaneous recording of
hematocrit, erythrocyte aggregation and disaggregation: methodology,
quality control and reference ranges,” Biomed. Tech., 32(5), 117–125 (1987).
33. S. Shin, Y. Yang, and J.S. Suh, “Measurement of erythrocyte
aggregation in a microchip stirring system by light transmission,” Clin.
Hemorheol. Microcirc., 41, 197–207 (2009).
34. S. Shin, J. X. Hou, and J.S. Suh, “Measurement of cell aggregation
characteristics by analysis of laser-backscattering in a microfluidic
rheometry,” Korea-Aust. Rheol. J., 19(2), 61–66 (2007).
35. S. Usami and S. Chien, “Optical reflectometry of red cell aggregation
under shear flow,” Bibl. Anat., 11, 91–97 (1973).
36. J.-F. Stolz, F. Paulus, and M. Donner, “Experimental approaches to
erythrocyte aggregation,” Clin. Hemorheol., 7, 109 (1987).
37. M. Donner, M. Siadat, and J.-F. Stolz, “Erythrocyte aggregation:
approach by light scattering determination,” Biorheology, 25, 367 (1988).
38. N. N. Firsov, A. V. Priezzhev, and O. M. Ryaboshapka, “Study of
erythrocyte aggregation kinetics in shear flow in vitro by light scattering
technique,” Proc. SPIE, 1991, 17–25 (1992).
39. N. N. Firsov, A. V. Priezzhev, O. M. Ryaboshapka, and I. V. Sirko,
“Aggregation properties of erythrocytes of whole blood under shear stress
by backscattering nephelometry,” Proc. SPIE, 1884, 283–290 (1993).
40. N. N. Firsov, N. B. Lapteva, B. A. Levenko, A. V. Priezzhev, S. G.
Proskurin, and O. M. Ryaboshapka, “Laser scattering studies of
structural and dynamic colloidal properties of protoplasm and blood,”
Progress in Colloid and Polymer Science, 93, 81–84 (1993).
“Diagnostic potentials of laser nephelometry of aggregating erythrocytes
suspensions,” Proc. SPIE, 1922, 139–144 (1993).
42. A. V. Priezzhev, N. N. Firsov, and O. M. Ryaboshapka, “Experience of
application of nephelometry for the analysis of aggregational state of
blood in a clinic of internal,” Proc. SPIE, 2136, 114–118 (1994).
43. A. V. Priezzhev, O. M. Ryaboshapka, I. V. Sirko, and N. N. Firsov,
“Optimization of the design of erythronephelometer,” Bulletin of the
Russian Academy of Sciences, (English transl. “Izvestia Rossiiskoi
Akamedii Nauk, Seriya Fizicheskaya”), 59, 168–173 (1995).
44. M. R. Hardeman, P. T. Goedhard, J. G. G. Dobbe, and K. P. Lettinga,

“Laser-assisted optical rotational cell analyser (L.O.R.C.A.); I. A new
instrument for measurements of various structural hemorheological
parameters,” Clin. Hemorheol., 14(4), 605–618 (1994).
45. G. Potron, D. Jolly, P. Nguyén, J. L. Mailliot, and B. Pignon,
“Approach to erythrocyte aggregation through erythrocyte sedimenta-
tion rate: application of a statistical model in pathology,” Rev. Fr.
Hematol., 36, 241–247 (1994).
46. M. Yu. Kirillin and A. V. Priezzhev, “Monte Carlo simulation of laser
beam propagation in a plane layer of the erythrocyte suspension:
comparison of contributions from different scattering orders to the angular
distribution of light intensity,” Quant. Electron., 32(10), 883–887 (2002).
47. A. V. Priezzhev, S. G. Khatsevich, and V. V. Lopatin, “Asymmetry of
light scattering from Couette flow of RBC suspensions: application for
biomonitoring of blood samples,” Proc. SPIE, 3567, 213–232 (1999).
48. V. V. Lopatin and A. V. Priezzhev, “Multiple light scattering by
ensembles of aggregating spheroids,” Moscow University Physics
Bulletin, 5, 19–22 (1999).
49. I. V. Sirko, N. N. Firsov, O. M. Ryaboshapka, and A. V. Priezzhev,
“Temperature dependence of erythrocyte aggregation in vitro by
backscattering nephelometry,” Proc. SPIE, 2982, 314–318 (1997).
50. F. J. Neumann, H. Schmid-Schonbein, and H. T. Ohlenbusch,
“Temperature dependence of red cell aggregation,” Pflugers-Arch.,
408(5), 524–530 (1987).
51. N. Maeda, M. Seike, and T. Shiga, “Effect of temperature on the
velocity of erythrocyte aggregation,” Biochem. Biophys. Acta: Biomem-
branes, 904(2), 319–329 (1987).
52. P. Snabre, H. Baumler, and P. Mills, “Aggregation of human RBC after
moderate heat treatment,” Biorheology, 22(3), 185–195 (1986).
53. J. Lademann, H.-J. Weigmann, W. Sterry, A. Roggan, G. Muller, A. V.
Priezzhev, and N. N. Firsov, “Investigation of the aggregation and
disaggregation properties of erythrocytes by light scattering measure-
ments,” Laser Physics, 9(1), 357–362 (1999).
54. V. L. Voeikov, “Physico-chemical and physiological aspects of erythro-
cyte sedimentation reaction,” Uspekhi Fiziologicheskikh Nauk (Moscow),
29(4), 55–73 (1998).
55. S. E. Kondakov, V. L. Voeikov, Yu. I. Gurfinkel, and A. Yu. Dmitriev,
«Dynamics of erythrocyte sedimentation rate as a new diagnostic tool, »
Proc. SPIE, 3252, 54–61 (1998).
56. J. Lademann, H.-J. Weigmann, W. Sterry, A. V. Priezzhev, and N. N.
Firsov, “Investigation of the aggregation and disaggregation properties
of erythrocytes in blood flow by optical techniques,” Autometria
(Russia), 5, 67–73 (2000).
34 Chapter 1
57. P. Steffen, A. Jung, D. B. Nguyen, T. Muller, I. Bernhardt, L. Kaestner,

and C. Wagner, “Stimulation of human red blood cells leads to Ca2 þ
-mediated intercellular adhesion,” Cell Calcium, 50, 54–62 (2011).
58. P. Steffen, C. Verdier, and C. Wagner, “Quantification of depletion-
induced adhesion of red blood cells,” Phys. Rev. Let., 110, 018102-1–
018102-5 (2013).
59. F. A. Carvalho and N. C. Santos, “AFM-based force spectroscopy–
biological and biomedical applications,” Life, 64(6), 465–472 (2012).
60. F. A. Carvalho, S. Connel, G. Miltenberger-Miltenyi, S. V. Pereira,
R. A. S. Ariens, and N. C. Santos, “Atomic Force Microscopy-Based
Molecular Recognition of a Fibrinogen Receptor on Human Erythro-
cytes,” ACS NANO, 4(8), 4609–4620 (2010).
61. K. C. Neuman and M. Block, “Optical trapping,” Rev. Sci. Instrum.,
75(9), 2787–2809 (2004).
62. A. Yu. Maklygin, A. V. Priezzhev, A. V. Karmenyan, S. Yu. Nikitin,
I. S. Obolesnkii, A. E. Lugovtsov, and K. Lee, “Measurement of
interaction forces between red blood cells in the aggregate by optical
tweezers,” Quant. Electron., 42(6), 500–504 (2012).
63. L. Lee, A. V. Danilina, M. Kinnunen, A. V. Priezzhev, and I. Meglinski,
“Probing the red blood cells aggregating force with optical tweezers,”
IEEE J. Sel. Top. Quantum Electron., 22(3), 7000106 (2016).
64. M. D. Khokhlova, E. V. Lyubin, A. G. Zhdanov, S. Yu. Rykova, I. A.
Sokolova, and A. A. Fedyanin, “Normal and system lupus erythema-
tosus red blood cell interactions studied by double trap optical tweezers
direct measurements of aggregation forces,” J. Biomed. Opt., 17(2),
025001-1–025001-6 (2012).
65. D. J. Stevenson, F. Gunn-Moore, and K. Dholakia, “Light forces the
pace: optical manipulation for biophotonics,” J. Biomed. Opt., 15(4),
041503-1–041503-21 (2010).
66. A. Ashkin, “Acceleration and trapping of particles by radiation
pressure,” Phys. Rev. Let., 24(4), 156–159 (1970).
67. A. Fontes, H. P. Fernandes, M. L. Barjas-Castro, A. A. de Thomaz,
L. Y. Pozzo, L. C. Barbosa, and C. L. Cesar, “Red blood cell
membrane viscoelasticity, agglutination and zeta potential measurement
with double optical tweezers,” Proc. SPIE, 6088, 608811-1–608811-10
(2006).
68. J. Lademann, H.-J. Weigmann, W. Sterry, H. Kiesewetter, A. Roggan,
G. Muller, A. V. Priezzhev, and N. N. Firsov, “Investigation of the
aggregation and disaggregation properties of erythrocytes by light
scattering measurements,” Proc. SPIE, 3923, 26–31 (2000).
69. N. Babu and M. Singh, “Influence of hyperglycemia on aggregation,
deformability and shape parameters of erythrocytes,” Clin. Hemorheol.
Microcirc., 31(4), 273–280 (2004).
70. M. W. Rampling, H. J. Meiselman, B. Neu, and O. K. Başkurt,

“Influence of cell-specific factors on red blood cell aggregation,”
Biorheology, 41, 91–112 (2004).
71. T. Urdulashvili, N. Momtselidze, M. Mantskava, N. Narsia, and
G. Mchedlishvili, “Hemorheological disorders and arteriolar resistance
during ischemic heart disease,” Clin. Hemorheol. and Microcirc., 30(3-4),
399–401 (2004).
72. M. Beridze, N. Momtselidze, R. Shakarishvili, and G. McHedlishvili,
“Effect of nitric oxide initial blood levels on erythrocyte aggregability
during 12 hours from ischemic stroke onset,” Clin. Hemorheol. and
Microcirc., 30(3-4), 403–406 (2004).
73. N. N. Firsov, A. Bjelle, T. V. Korotaeva, A. V. Priezzhev, and O. M.
Ryaboshapka, “Clinical application of the measurements of spontaneous
erythrocyte aggregation and disaggregation. A pilot study,” Clin.
Hemorheol. Microcirc., 18(2-3), 87–97 (1998).
“Aggregation and disaggregation of erythrocytes in whole blood: study
by backscattering technique,” J. Biomed. Opt., 4(1), 76–84 (1999).
75. O. K. Baskurt, M. Boynard, and G. C. Cokelet et al. “New guidelines for
hemorheological laboratory techniques,” Clin. Hemorheol. Microcirc.,
42, 75–97 (2009).
Alexander V. Priezzhev received his PhD from the Faculty of

Physics, Lomonosov Moscow State University (MSU),
Moscow, Russia in 1971 and 1975, respectively. He has led
and participated in various national and international
research projects and scientific conferences on biophotonics,
medical physics, and biomedical optics. He is head of the
Laboratory of Laser Biomedical Photonics, Faculty of
Physics and International Laser Center, MSU. His areas of
expertise include nanobiophotonics, biomedical optics, light scattering
diagnostics, physics, and rheology of biological fluids. He is a member of
SPIE and editorial board member of several scientific journals.
Kisung Lee received his M.S. degree in physics from the

Irkutsk State University, Irkutsk, Russia in 2011. He is
currently pursuing his PhD in biophysics in the Lomonosov
Moscow State University, Moscow, Russia and D.Sc. (Tech)
degree in the University of Oulu, Oulu, Finland. His research
interests include biophotonics, biomedical optics, hemor-
heology, and biophysics of cell-cell interaction. He is a
member of SPIE.
36 Chapter 1
Nikolai N. Firsov is a professor of physics in Russian

National Research Medical University named after N.I.
Pirogov. He holds the degrees of MD and Doctor of Sci. in
biology. He authored and co-authored many fundamental
and applied research results in physics of biological fluids, in
bio- and hemorheology. He is an inventor of and holds
patents on red blood cell aggregometry techniques. His areas
of expertise include medical physics, biophysics, physics and
rheology of biological fluids, and light scattering diagnostics of blood cells.
Juergen Lademann graduated as a physicist from the

Lomonosov Moscow State University in 1980 and received
his PhD from that university in 1984. In 1991 he received his
PhD in spectroscopy from the Friedrich-Schiller University
of Jena. He obtained his venia legendi in electrical
engineering from the University of the Armed Forces of
Munich in 1995, and in biophysics from the Humboldt
University of Berlin in 1996. Since 1996, he has been
Director of the Center of Experimental and Applied Cutaneous Physiology
at the Department of Dermatology, Venerology and Allergology of the
Charité – Universitätsmedizin Berlin. In 2001, he was appointed Professor of
Dermatology by the Charité. In October 2014, he was elected to the
Praesidium of the International Federation of Societies of Cosmetic Chemists.
He also serves as an editorial board member of the Journal of Biomedical
Optics.
Chapter 2
Light Scattering Spectroscopy
of Epithelial Tissues: Principles
and Applications
Lev T. Perelman
Harvard University, Cambridge, USA
Vadim Backman
Northwestern University, Evanston, USA
2.1 Introduction
Over the past decades, substantial progress has been made in medical
diagnostic technologies that target anatomic changes at the organ level.
Techniques such as magnetic resonance imaging (MRI) and spectroscopy
(MRS), x-ray computed tomography (X-ray CT), and ultrasound make it
possible to “see through the human body.” At the same time, there is clearly a
need for the development of diagnostic techniques that use our current
knowledge of the cellular and subcellular bases of disease. The diagnostic
techniques applicable in situ (inside the human body) that can provide
structural and functional information about the tissue at the cellular and
subcellular levels, the kind of information that is currently obtainable using
only in vitro methods requiring tissue removal, will have great implications for
the detection and prevention of diseases as well as targeted therapy.
Recently various optical techniques, such as elastic1–3 and Raman
scattering,4 absorption,5,6 and fluorescence7–9 have been used to study human
tissue in situ noninvasively. The goal of these studies is not only to provide
accurate diagnosis of various diseases, but also to better understand the
genesis of diseases. In this chapter we describe novel techniques capable of
identifying and characterizing pathological changes in human tissues at the
cellular and subcellular levels that are based on light scattering. Light
scattering spectroscopy (LSS) has long been used to study a great variety of
37
38 Chapter 2
materials ranging from isolated atoms to complex condensed matter systems.

Biological tissue is yet another example of a complex system that can be
studied with LSS.10 Light scattering can provide structural and functional
information about the tissue. This information, in turn, can be used to
diagnose and detect disease. One exciting application of biomedical optics is
the noninvasive or minimally invasive detection of precancerous or early
cancerous changes in human epithelium, the cellular layer lining the inner
surfaces of the body. Detection and diagnosis of such conditions is
particularly important because most tumors are readily treatable if diagnosed
at an early stage. Unfortunately, many forms of precancerous lesions are
difficult to detect using conventional diagnostic techniques.
Cancers of epithelial origin, or carcinomas, which represent more than
85% of all cancers, are generally preceded by a precursor condition known as
dysplasia. Dysplasia is confined to the epithelial layer and is characterized by
cellular proliferation, abnormal appearance of the cell nuclei, and changes in
tissue architecture. In many cases, dysplastic tissue is flat and indistinguish-
able from the surrounding nondysplastic tissue. Since it cannot be
distinguished, detection of such dysplastic tissue is based on random biopsy.
The biopsied tissue is then fixed, stained, and examined by a surgical
pathologist. Microscopic examination of biopsy specimens of dysplastic tissue
shows characteristic morphologic changes in cell nuclei, which are the
hallmarks of dysplasia and early cancer. The nuclei become enlarged,
crowded, and hyperchromatic (abnormally darkly stained). Although the
gross and microscopic appearance of dysplasia in different organs and
different types of epithelium can vary significantly, these changes are common
to all types of precancerous and early cancerous conditions.
Despite these seemingly well-defined differences between normal and
dysplastic tissues, there can be significant disagreement among even
experienced pathologists on the diagnosis. In some cases, the interobserver
agreement is as low as 50%. One possible explanation for such disagreement is
the lack of objective quantitative information—in some sense pathology is
both science and art. The diagnosis, the choice of treatment and,
consequently, the patient’s well-being depend on a pathologist’s ability to
recognize a morphological pattern known to be associated with a disease.
Diagnostic methods based on objective measurement techniques can
supplement or even surpass the conventional approaches to clinical diagnosis.
In this chapter we discuss LSS-based methods to measure epithelial morphology
in living tissues that do not require tissue removal. Such techniques can be used
for noninvasive or minimally invasive detection of precancerous and early
cancerous changes and other diseases in a variety of organs such as the
esophagus, colon, uterine cervix, oral cavity, lungs, and urinary bladder.11
This chapter is not intended to be a complete survey of light scattering in
tissue. Its main goal is to provide the reader with the basic tools necessary to
Light Scattering Spectroscopy of Epithelial Tissues: Principles and Applications 39
understand the principles of biomedical light scattering spectroscopy as well

as the underlying medical and physical background. The structure of this
chapter is as follows: Sec. 2.2 provides the medical and biological background
for the rest of the chapter. It discusses the microscopic architecture of
superficial tissues as well as changes associated with cancerous and
precancerous transformations in these tissues. In other words, it addresses
the question “what are the potential sources of light scattering in superficial
tissues?” Section 2.3 provides the reader with the basic principles of light
scattering needed to better understand Sec. 2.4, where these principles are
applied to light scattering by cells and subcellular structures. Section 2.5
discusses the relation of single and multiple scattering in tissue. Section 2.6
reviews applications of various types of light scattering in the detection of
early cancer and precancerous conditions. Finally, Sec. 2.7 discusses the new
technique, called confocal light absorption and scattering spectroscopic
(CLASS) microscopy, which combines LSS with confocal microscopy.
2.2 Microscopic Architecture of Mucosal Tissues

The properties of light scattering in biological tissues depend inherently on
tissue architecture. Various tissue inhomogeneities such as cellular organelles,
extracellular matrix, etc., may affect light propagation in tissue. This
translates into unique spectroscopic, polarization, or angular features of
scattered light emerging from tissue. Therefore, qualitative or quantitative
information about tissue macroscopic and microscopic structures can be
obtained with proper interpretation of these features. However, solving such
an inverse problem without any a priori knowledge of tissue structure is
virtually impossible due to the exceedingly complex organization of the
biological tissues. Most tissues are highly inhomogeneous and are character-
ized by a high degree of complexity, ranging from a few angstroms to
centimeters. Thus, solving the inverse problem would require precise
knowledge of the electromagnetic field at any point inside the tissue, which,
of course, is never a possibility. Despite this seemingly discouraging fact, if
certain assumptions about the tissue structure can be made, some important
properties of the tissue can be measured easily with scattered light.
In this section we discuss the basics of the morphology of human tissues
with special emphasis on the structure of the mucosal tissues lining the inner
and outer surfaces of the human body. Our particular interest in the mucosal
tissues is substantiated by the fact that, as mentioned in the introduction, it is
the mucosal tissues that are most readily accessible with visible light. Also,
detection of pathological changes such as precancerous conditions known as
dysplasia and carcinoma in situ (CIS) in the mucosal epithelium represents one
of the biggest challenges in modern medicine. Malignant and premalignant
transformations in the epithelium will be discussed in the following section in
40 Chapter 2
great detail. The following discussion is by no means exhaustive. Rather, it

aims to provide a brief introduction to the histology of normal tissues and
selected pathological conditions at its irreducible minimum necessary to
comprehend the principles and applications of light scattering spectroscopy.
For a reader who searches for more detailed and complete information about
the histology and pathology, we would highly recommend Refs. 12–14.
Any biological tissue consists of variable quantities of cells and
extracellular matrix. Four types of tissues are being identified: epithelium,
connective tissue, muscle, and nervous tissue. Some anatomists distinguish a
fifth tissue type—the blood. Here, however, we follow a more standard
classification and identify the blood as a specialized subtype of connective
tissue. A number of tissues comprise functional units called organs. For
example, the esophagus consists of epithelium covering its inner (lumenal)
surface situated on top of the layers of connective tissue and muscle
containing variable amounts of nerves and blood vessels. Several organs
compose organ systems such as skeletal, circulatory, respiratory, gastrointes-
tinal, and genitourinary systems. For example, the gastrointestinal system
consists of the organs of the oral cavity (tongue, lips, etc.), esophagus,
stomach, small and large intestines, rectum, and glands of the digestive system
located outside the walls of the alimentary canal (pancreas, liver, gallbladder,
etc.). We start our discussion with the description of the organization of the
cell, considering an epithelial cell as a model. Then we discuss the major
histological properties of the mucosa and the tissues of which it is composed.
2.2.1 Morphology of the cell

So far, more than 200 different cell types have been identified. Despite this
diversity, most cells possess many common features. We will consider a cell of
the membranous epithelium lining surfaces of most human organs as a model.
The cell is bounded by a membrane, the plasmalemma. It is a phospholipid
bilayer 8.5 to 10 nm in thickness with integral and peripheral proteins
embedded in it. These proteins provide functional properties of membranes
and may extend out of the inner or outer surface of the plasmalemma by as
much as 10 nm. Thus, the overall thickness of the plasmalemma might range
from 8.5 to about 30 nm.
Two major cell compartments are the nucleus and the surrounding
cytoplasm. The cytoplasm contains organelles, which are metabolically active
subcellular organs, and inclusions, which are metabolically inactive. The
following is a list of major cytoplasmic organelles and inclusions and their
properties (Fig. 2.1):
1. Mitochondria typically have a shape of a prolate spheroid. Their size varies
greatly even within a single cell. The large dimension of a mitochondrion
may range from 1 mm to 5 mm. The larger diameter typically varies between
Figure 2.1 0DMRU RUJDQHOOHV DQG LQFOXVLRQV RI WKH FHOO6HHFRORUSODWHV
0.2 mm to 0.8 mm. The mitochondria are quite flexible and may easily change
their shape. Their major function is to generate adenosine triphosphate
(ATP) via oxidative phosphorilation, thus providing the energy required for
the cell. Since metabolic requirements of different cells vary, the numbers of
mitochondria differ depending on the cell size and its energy needs. For
example, nonsecreting cells of the epithelial lining of the internal surface of
the colon have few mitochondria. In contrast, the hepatocytes—liver cells—
contain a few thousand mitochondria. A mitochondrion is composed of a
7-nm-thick outer membrane and an inner membrane that is about 6 nm in
thickness. The inner membrane is folded to form cristae. It is studded with
numerous mushroom-like 15-nm particles that possess a 10-nm head and 5-
nm tail connecting them to the membranous wall. The 10- to 20-nm-wide
space between the outer and the inner membranes is called the membrane
space and appears relatively empty. The space between the cristae of the
inner membrane is wider, ranging from 40 to 200 nm, and is filled with the
mitochondrial matrix. This matrix is not uniform, but rather is a meshwork
of nearly spherical matrix granules composed of densely packed phospho-
lipoproteins. These granules are 30 to 50 nm in diameter and have a complex
internal structure. They contain multiple 1- to 3-nm compartments separated
by dense septa.
2. Endoplasmic reticulum (ER) is composed of tubules and flat sheets of
membranes distributed over the intracellular space. The outer diameter of
these tubules ranges from 30 to 100 nm. Their wall thickness is about
10 nm. There are two types of endoplasmic reticulum: rough endoplasmic
reticulum (RER) that functions in the synthesis of cholesterol and lipids,
and smooth endoplasmic reticulum (SER) that functions in the synthesis
of proteins. The RER differs from the SER in that it bears 20- to 25-nm
42 Chapter 2
spherical or sometimes spheroidal particles called ribosomes. Although

small, the ribosomes are complex particles consisting of 1-nm units that
contain ribonucleic acids (RNA) and proteins.
3. Golgi apparatus is composed of a group of 4 to 10 flattened parallel
membrane-bounded cisternae and functions in the modification and
packaging of the macromolecules. The overall thickness of this organelle
can range from 100 to 400 nm.
4. Lysosomes are 0.25 to 0.8 mm bodies of various shapes, ranging from
highly irregular to almost perfectly spherical. They differ in their internal
structure as well: some can be homogeneous and some may consist of a
collection of dense granules of 20 to 100 nm in diameter embedded in a
surrounding matrix of lesser density. The numbers of lysosomes are highly
variant for different cells as well: the cells of the membranous epithelial
lining of the cervix, for example, contain just a few lysosomes, while
hepatocytes may posses a few hundred. They are filled with hydrolytic
enzymes and other molecules and participate in intracellular digestion.
5. Peroxisomes are 0.2 to 1.0 mm spheroidal bodies of lower densities than
lysosomes that are more sparse in the metabolically active cells such as
hepatocytes where they are counted in hundreds.
6. Cytokeleton is composed of filamentous arrays of proteins. Its three major
components are microtubules, which are about 25 nm in diameter with a
wall 9 nm thick and a 15 nm lumen, 10 nm in diameter intermediate
filaments, and 7 nm in diameter microfilaments.
7. Various cytoplasmic inclusions, such as lipids, glycogen, secretory granules,
and pigments, come in all different sizes ranging from 20 to 500 nm. They
might be of various shapes but usually appear to be near spherical. The
surface roughness of an inclusion can range from 2 to 40 nm.
Although these cytoplasmic components appear to be greatly different in
their structure and organization, a few generalizations can be made: 1. most
of the cytoplasmic organelles and inclusions are smaller than 1 mm in size;
2. they are not homogenous bodies but rather complex structures that possess
an exceedingly intricate internal structure.
At this point we turn our attention to the largest cell organelle, the nucleus,
which deserves special attention. The nucleus is usually spherical or spheroidal,
although in some cases it may appear infolded or lobulated. The importance of
the nucleus is apparent from its function. The nucleus houses the chromosomes,
where most of the genetic material essential for the cell activity and replication
is contained. It is the site of messenger RNA (mRNA), transfer (tRNA), and
ribosomal RNA (rRNA) transcription, and of ribosome assembling from
proteins and rRNA. The overall sizes of the nuclei vary depending on the cell
type, its metabolic state, and phase of the cycle—it is usually enlarged when
active transcription of the genes occurs. Typical nuclei are from 5 to 10 mm in
diameter. One of the striking differences between the nucleus and the other
organelles is the fact that most of the nuclei of the same cell line, for example,
columnar epithelial cells of the intestine, are all of similar size and shape. In
some cases, the standard deviation of the nuclear diameters is less than 5% of
the average diameter. On the contrary, the sizes of other organelles,
mitochondria for instance, vary widely even within a single cell. (Note that
the statistical properties of the population of mitochondria and other small
organelles are roughly preserved over a cell line.)
The major components of the cell nucleus are the nuclear envelope,
chromatin, nucleolus, and nuclear matrix, all embedded in the nucleoplasm.
1. The nucleus is enclosed by the nuclear envelope, a system of two
membranes separated by a 10- to 30-nm space, the prerinuclear cisterna.
The outer membrane may have ribosomes attached to it. Some parts of
the membrane contain fenestration called nuclear pores that are complex
structures with a circular shape about 100 nm in diameter covered by 15-
to 20-nm spherical particles.
2. Chromatin is one of the major components of the nucleus. It contains
the strands of deoxyribonucleic acid (DNA) that encode the genetic
information of the cell. Although the DNA is organized in the
chromosomes, these are not appreciable in the interphase, or nondividing
nucleus, and become apparent only with cell division. Two types of
chromatin are distinguished. The portion of the DNA that is not being
transcribed forms condensed or heterochromatin. The transcribed part of
the DNA comprises an extended or euchromarin. This distinction was
brought into use by the histologists who microscopically examined stained
tissue sections. When stained with basophilic contrast dye, the nucleus
appears bluish due to the high affinity of the heterochromatin to such dyes.
On the other hand, the euchromatin is poorly stained and is not visible with
the microscope. The fundamental etiology of this difference between two
types of chromatin has yet to be explained. Although it is known that the
euchromatin is less dense than the heterochromatin, this fact itself is not
sufficient to explain the difference in the staining pattern. The clumps of the
heterochromatin are irregular and may vary from 0.5 to 2 mm in size. The
heterochromatin is made up of closely tangled 30 nm fibrils. The fibrils are
composed of beaded strands with regularly spaced smaller subunits, the
nucleosomes. These have a cylindrical shape and are connected by the
filaments, which are 4 nm in diameter. In the euchromatin, the strands of
the nucleosomes are uncoiled and do not form the fibrillar structure.
3. The nucleolus is responsible for the transcription of ribosomal RNA
(rRNA), the assembly of subunits of ribosomes. The size of the nucleolus
may range from 0.5 mm up to about 1 mm. It appears to be a more or less
round network of anastomosing strands, called pars granulose. These
strands are made from 15-nm ribonucleoprotein particles surrounded by
even smaller filaments. There are a few rounded structures within the
nucleolus filled with fibrillar material of a lower density, called fibrillar
centers, which have a dimension of about 80 nm.
44 Chapter 2
The cell nucleus, like most other cell organelles, is not a uniform object
and possesses a complex internal structure. The larger inhomogeneities are
formed by smaller structures, which in turn are composed of even smaller
building blocks. We have to notice, however, that studies have established
that the density variations within the cell nucleus are smaller than the one
between the nucleus and the cytoplasm. The implications of this fact for light
scattering by the cells will be discussed in the following sections.
2.2.2 Histology of mucosae

The walls of most internal cavities of the body such as the gastrointestinal
tract, respiratory tract, cervix, etc. generally consist of four layers (not all
layers are present in all organs’ walls): mucosa, submucosa, tunica muscularis,
and serosa or adventitia. Generally speaking, the walls of hollow human
organs are formed by alternating layers of connective and muscular tissues
and the epithelium coating the inner and outer surfaces of the wall (Fig. 2.2).
Mucosa, the innermost layer, may consist of three layers. Its surface is
always lined by some type of epithelium. Epithelia are separated from
underlying tissues by a thin noncellular epithelium-derived 100 nm thick layer
called the basement membrane. Thickness of the epithelium may vary from 10
to about 300 mm depending on its type. We will describe the various types of
epithelia in greater detail later. The epithelium is usually supported by a layer
of connective tissue whose thickness can range from 50 to a few hundred
microns (Fig. 2.3). Depending on the density of this connective tissue, it is
Figure 2.2 Low-power microphotograph of section of the wall of the esophagus illustrating
the major structures of the gastrointestinal tube. L–lumen, E–epithelium, CT–connective
tissue, MM–muscularis mucosa, BV–submucosal blood vessel, SM–submucosa, MP–
PXVFXODULV SURSULD6HHFRORUSODWHV
Figure 2.3 Microphotograph of the H&E-stained section of the bronchial wall illustrating
major components of the mucosa: epithelium and connective tissue6HHFRORUSODWHV
classified as a part of mucosa or submucosa. For example, the loose

connective tissue underlying the epithelial lining of the colon is considered to
be a part of mucosa and is called lamina propria. Denser connective tissue
supporting the epithelium of the cervix is referred to as a part of submucosa.
Some mucosae—distal esophagus is an example—include a layer of scattered
or continuous smooth muscle cells just below the lamina propria. It is usually
not thicker than 100 mm but in some cases may reach up to 400 mm.
The submucosa is a 400- to 1000-mm layer of moderately dense connective
tissue that supports and provides vascular, nervous, and lymphatic supply to
the mucosa. The tunica muscularis consists of a few layers of smooth muscle
cells, blood vessels of various sizes, and nerves. It is from 0.5 to 3 mm thick.
The outermost layer of a tissue wall, a 0.5- to 3-mm thick serosa, consists of
connective tissue covered by a single layer of epithelial cells called
mesothelium (the outermost layer in some organs is referred to as adventitia).
Epithelia form either glands or membranes. It is the membranous
epithelium that covers the mucosal surfaces and it is our major interest.
Hereafter, we will discuss the membranous epithelium only. Epithelia are
composed of very closely packed, contiguous cells, with very little or no
intracellular material in the extracellular space. They are avascular and do not
possess a nerve supply. Epithelia are classified according to the number of cell
layers, shape of the cells, and the free surface specializations.
Based on the number of cell layers, an epithelium can be classified as
simple, stratified, pseudostratified, or transitional (Fig. 2.4).
1. The simple epithelium consists of a single cell layer. The examples are
endothelium, the epithelium of the blood vessels (2 mm thick), and
intestinal epithelium (15 to 20 mm thick). The thicknesses of such epithelia
do not usually exceed 30 mm.
46 Chapter 2
Figure 2.4 Examples of epithelia: (a) simple columnar epithelium, (b) stratified squamous
epithelium, (c) pseudostratified epithelium, (d) transitional epithelium6HHFRORUSODWHV
2. The stratified epithelium is formed by a number of cell layers. It can be as

thick as 0.5 mm. In a common subtype of this epithelium, squamous
stratified epithelium (see below), the cells flatten out while they move from
the basal (bottom) to the apical (lumenal) surfaces during the process of
maturation. Stratified squamous epithelia are found in skin, oral cavity,
esophagus, and exocervix.
3. The pseudostratified epithelium consists of a single layer of tall cells that
appear to be stratified. It is about 30 to 50 mm thick. Examples are
epithelia of male urethra and proximal respiratory tract (trachea and
primary bronchi).
4. The transitional epithelium is found primarily in the urinary bladder,
consists of a multiple cell layers, and is characterized by large round cells
on the surface and tall cells on the bottom. In the bladder, it is usually
composed of 7 cell layers and is about 150 to 200 mm thick.
Based on the shape of the cells, an epithelium is classified as squamous,
cuboidal, or columnar.
1. The uppermost cells of a squamous epithelium are more or less flat.
A squamous epithelium can be simple or stratified (see examples above).
The squamous cells are just a few microns thick and have a large surface
area.
2. The shape of the cuboidal cells is suggested by their name. They are
typically about 10–15 mm wide and thick. The nuclei are usually spherical.
While true stratified epithelium is rarely found, examples of simple
cuboidal epithelia are abundant: epithelium of the ovary, ductal

epithelium in the breast, etc.
3. A typical columnar cell has a cylindrical shape and is about 10 mm wide
and 20 to 30 mm tall. The nuclei are located parabasely and are usually
shaped as prolate spheroids with a longer axis oriented along the axis of
the cell. Most of the columnar epithelia are simple. Examples are
intestinal epithelium, epithelium of the uterus, and endocervix.
The free surface of an epithelial cell may form several specialized
structures such as microvilli, cilia, stereocilia, etc. Classification of epithelia
based on such properties is frequently used. However, it is of less importance
for our discussion.
Since the connective tissue is one of the major components of the mucosa
and/or submucosa, it is worthwhile to briefly consider its structure. Among
the eight types, there is one that is of immediate interest, namely loose
(areolar) connective tissue. (The other types are dense irregular, dense regular,
adipose, reticular, cartilage, bone, and blood. Dense irregular connective
tissue is found in some of the submucosae. We do not discuss its structure due
to its relative similarity to that of the areolar connective tissue.) It is
ubiquitously present in most mucosae and submucosae. The connective tissue
has both cellular and intercellular components. The intercellular materials are
fibers, amorphous ground substance, and tissue fluid.
Three types of fibers are recognized: collagen, reticular, and elastic fibers.
Collagen fibers are composed of specific proteins. They appear as
unbranched, randomly oriented strands 0.5 to 5 mm in diameter. They
consist of clearly distinct parallel fibrils, which are 50 to 90 nm in diameter,
separated by about 100 nm. Reticular fibers are thin, 10 nm in diameter, and
branched. Elastic fibers are almost as thin as reticular fibers. They branch and
form a loose three-dimensional network.
The cells of loose connective tissue are fibroblasts, macrophages, plasma
cells, mast cells, and some others. Despite such a variety and in contrast to the
epithelium, most connective tissues possess only limited quantities of the cells
and are mainly composed of noncellular materials.
2.2.3 Introduction to histopathology of early cancer and dysplasia

So far we have considered normal histology of cells and tissues composing
human mucosae. The major application of optical techniques such as LSS is
thought to be studying, detecting, and diagnosing various pathological
conditions. Detection of early cancer and precancerous conditions such as
dysplasia or carcinoma in situ (CIS) has been the Holy Grail of biomedical
optics for many years. It is the immense importance of the problem and
attractiveness of a potentially noninvasive diagnosis that stimulated
researchers from various fields to attack this problem. It is important to
48 Chapter 2
notice that more than 85% of all cancers originate from the epithelia; colon,
lung, and cervical cancers are examples. Such cancers are referred to as
carcinomas. If diagnosed at one of the preinvasive stages, most would-be
carcinomas are readily curable. Unfortunately, detection of such lesions is not
always possible with conventional diagnostic techniques. During the last few
years, LSS has appeared to be a powerful tool to considerably advance the
detection of early cancer. Its success is based on the fact that the earliest and
most ubiquitous changes occurring with precancer are alterations of the
histology of the affected epithelial cells, and their nuclei in particular. The
purpose of the following section is to outline the basic histopathology of early
cancerous and precancerous changes happening in the epithelia.
While carcinomas may originate from perfectly healthy epithelium, it has
been shown that the probability of a malignant transformation is much
greater for the epithelium that has already been damaged. For example, many
cancers grow on the basis of metaplasia. The term metaplasia describes a
pathological condition in which a normal mucosa is replaced by mucosa of
another type that is not normally present in the affected organ. One of the
examples of metaplasia is a replacement of normal stratified squamous
epithelium of the esophagus by intestinal mucosa covered by the simple
columnar epithelium, which is believed to be a consequence of the persistent
gastroesophageal reflux commonly known as “heartburn.” This condition is
referred to as Barrett’s esophagus. Patients with Barrett’s esophagus are
estimated to have a 40 times increased chance to develop esophageal
adenocarcinoma than the rest of the population. Squamous cell carcinoma of
the lungs is another example. It originates from stratified squamous
epithelium that has replaced respiratory (pseudostratified ciliated) epithelium
normally found in the large bronchi.
Carcinogenesis is a complex process. It starts from the exposure of a cell
to a carcinogenic agent, such as ionizing radiation, a chemical, a virus, etc.
This exposure results in DNA damage and mutation. It is said that the cell
becomes “initiated.” Many genes such as oncogenes and cancer suppressor
genes whose mutations may lead to a progression to malignancy have been
identified. By definition, the cells become malignant when they are able to
penetrate the basement membrane separating the epithelium from the
underlying connective tissue. The cells are almost never able to penetrate
the basement membrane immediately after their initiation. Thus, the
malignancy is usually preceded by a precursor stage that is usually referred
to as dysplasia. The dysplastic cells exhibit disorderly but not malignant
proliferation. Dysplasia almost always precedes but not necessarily results in
cancer. Thus, it would be proper to say that dysplastic cells have malignant
potential. Depending on the severity of the disease, the dysplasia is classified
as either low grade or high grade or as mild, moderate, or severe. Sometimes
pathologists use the term “indefinite for dysplasia” in cases when identifying
dysplasia is difficult. Finally, if the overall thickness of the epithelial layer is

affected, the dysplasia is called carcinoma in situ (CIS). In some organs such
as the cervix, dysplasia CIS might be called squamous interepithelial lesion
(SIL), or squamous intraepithelial neoplasia (SIN). The use of such different
terminology, though it may seem confusing, is widely accepted by pathologists
and clinicians.
Now we are ready to address the question of how the dysplastic
epithelium is different from the normal epithelium. Despite the fact that
dysplasia may vary greatly in its macroscopic appearance depending on the
organ and stage of the disease, there are a few major morphologic features
that are ubiquitously present in the affected cells and make them appear
distinctly different from their normal counterparts. Namely, dysplastic cells
and their nuclei exhibit considerable pleomorphism (variation in size and
shape), cell nuclei are hyperchromatic (appear darkly stained with nuclear
dyes due to excessive quantities of chromatin), and are abnormally enlarged.
Additionally, if the mitotic figures are present in a nonbasal part of the
epithelium, abnormal mitosis may be observed. At a higher level of
organization, architectural changes such as an overall disorganization of the
epithelium that is closely related to loss of the normal maturation of cells can
be seen. Figure 2.5 illustrates these differences between normal and dysplastic
stratified squamous epithelia of the uterine cervix. It is important to reiterate
that although the macroscopic appearance of dysplastic lesions may differ (for
example, dysplasia in the colon forms polyps called adenomas versus flat
dysplastic lesions found in Barrett’s esophagus; both mucosae share the same
type of epithelium, simple columnar), the morphologic features of dysplastic
cells and architectural alterations occurring in the epithelial layer listed above
are characteristic for essentially all types of dysplasia in all organs.
Figure 2.5 Dysplasia in the cervical stratified squamous epithelium. On the left,
nondysplastic epithelium is preserved6HHFRORUSODWHV
50 Chapter 2
As mentioned, if a dysplastic lesion is detected, it can almost always be

cured with surgery or other types of therapy. Therefore, it is not surprising
that dysplasia surveillance is considered to be a crucial step toward cancer
prevention. Despite significant attention, no universal means to detect
dysplastic lesions has been found. Polypoid dysplastic lesions such as colon
adenomas are apparently the easiest to detect. These can be detected and
removed with endoscopy. Many dysplastic lesions, however, do not form
grossly apparent formations and are indistinguishable from the surrounding
nondysplastic tissue. Moreover, the size of these “flat” dysplasias might be no
larger than 1 mm in diameter. Even such small lesions can bear lethal
metastatic potential. Detection of such lesions is possible only with random
biopsy. For example, patients with Barrett’s esophagus undergo annual
surveillance with random biopsy at endoscopy. Other examples of flat
dysplasia include bladder dysplasia, colon dysplasia associated with ulcerative
colitis, cervix, and many others. The difficulties associated with random
biopsy are apparent: the probability of detecting a 1-mm lesion in the
esophagus or any other organ with a large surface area is minute. Moreover,
the diagnosis and classification of dysplasia is purely qualitative and is based
on subjective interpretation of the biopsied materials by a pathologist. Studies
have shown that pathologists’ inter- and intra-observer agreement may be as
low as 50% in some cases.15 This is why the absolute majority of dysplastic
lesions are left undetected until they lead to metastatic cancer.
Cancer starts with epithelial cells penetrating the basement membrane into
the underlying connective tissue. Cancers are classified according to stage
(spread of malignant cells) and grade (degree of cytologic abnormalities such
as anaplasia). The higher the grade of the tumor, the more aggressive it is,
and, generally speaking, the poorer the patient’s prognosis is. The anaplasia is
marked by essentially the same morphologic features as dysplasia: cellular and
nuclear pleomorphism, nuclear hyperchomasia, and enlargement. The
chromatin is usually coarsely clumped and distributed along the nuclear
membrane. The malignant nuclei frequently possess large and dense nucleoli.
This succinct overview suffices to describe the most important histologic
features of dysplastic and malignant tissues. Further discussion of tumor-
ogenesis is out of the scope of our discussion.
2.3 Principles of Light Scattering

In this section we overview several techniques that are often used to describe
light scattering from biologically relevant objects such as cell organelles,
inclusions, etc., described in the previous section. First of all, we will define
the scattering particle. As discussed above, most subcellular organelles are not
uniform and have a complex shape and structure. Nevertheless, we will refer
to them as scattering “particles.” By a particle we will mean any bounded
region of refractive index variation. One can think of two types of problems
related to light scattering by a single particle: direct and inverse scattering
problems. The direct problem is formulated as follows: given a particle of
known structure (alternatively, given complete information about the
distribution of the refractive index inside the particle) illuminated by a plane
wave of particular polarization, find the electromagnetic field inside and
outside the particle. Usually, however, the simplified formulation of the
problem suffices: find the electromagnetic field at large distances from the
particle. The inverse problem is to find the distribution of the particle’s
refractive index based on the measurable electromagnetic wave scattered by
this particle. Although not solvable in a closed form in most cases, the first
problem is much easier than the second one. In fact, it is virtually impossible
to solve the inverse problem in the case when the solution of the direct
problem is not known. While the inverse problem is not solvable in most
relevant cases, one can get at least partial information about the particle from
the scattering data. In order to achieve this, understanding the direct problem
is crucial.
2.3.1 Rigorous solution of the direct scattering problem

Consider a particle illuminated by a plane electromagnetic wave
Ei ¼ E0 eiðkrvtÞ , (2.1)
where E0 is the wave amplitude, k is the wave vector, and v is the frequency.
In the wave zone, distances r are large compared to the wavelength l and the
particle’s characteristic dimension a; the radiation scattered by the particle is a
spherical wave Es. The amplitudes of both the incident and scattered waves
can be expressed as a combination of two components, parallel, Ei2 and Es2,
and perpendicular, Ei1 and Es1, to the scattering plane. The scattering
amplitude matrix S(u,f) relates the components of the scattered wave and
those of the incident wave, Es ¼ SEi or, more explicitly,16

E s2 eiðkrvtÞ S2 S3 E i2
¼ , (2.2)
E s1 ikr S4 S1 E i1
where r ¼ r(u,f) is a direction of propagation of the scattered light given by

the polar angles u and f in the spherical system of reference associated with
the particle and, generally speaking, complex functions Sk ¼ Sk(u,f),
k ¼ 1,. . . ,4. The scattering amplitude matrix is the fundamental property
that gives a complete description of the scattering process and solves the direct
scattering problem. Moreover, all other properties describing the scattering
event can be expressed through the matrix components. For example, the
scattering cross section, ss, which is defined as the geometrical cross section of
52 Chapter 2
a particle that would produce an amount of scattering equal to the total

observed scattered power at all directions, is given by
2p 0
ss ¼ k 2
∫ ∫ðjS1 þ S4 j2 þ jS1 þ S4 j2 Þd cos udf: (2.3)
0 p
The elements of the scattering matrix depend on the spatial distribution of

the refractive index within the particle. For example, if the particle is
homogenous and spherical, then there are only two parameters that the matrix
depends on: the diameter of the sphere and its refractive index. We note as
well that for a spherically symmetrical particle, or, more generally, any
particle that is cylindrically symmetrical in respect to the direction of
propagation of the incident light, elements S3 ¼ S4 ¼ 0 and the matrix is
diagonal. Moreover, S1 and S2 are functions of the scattering angle u only and
do not depend on f. In this case, the expression for the intensities of scattered
light polarized along and orthogonally to the scattering plane are proportional
to the respective components of the incident light
jS 2 ðuÞj2
I jjs ¼ I jji , (2.4)
k 2 r2
jS 1 ðuÞj2
I ⊥s ¼ I ⊥i . (2.5)
k 2 r2
Generally speaking, to find the matrix elements, one needs to solve
Maxwell’s wave equations with proper boundary conditions of the electric
and magnetic field continuities. The methods of solving the wave equations
are not discussed here, since they can be found elsewhere.17 Such a solution is
the most rigorous and the one that is usually the most difficult to obtain. In
fact, there are just a few cases where the analytical solutions to the wave
equation have been found. In 1907, Gustav Mie obtained the solution for the
scattering of a plane wave by a uniform sphere. The functions S1 and S2 are
expressed as an infinite series of Bessel functions of two parameters, kd and
kmd, with k the wave number, d the diameter of the sphere, and m the relative
refractive index of the sphere.18 We note that the parameters the Mie solution
depends on are the maximal and the minimal phase shifts of the light wave
passing through the system. If the sphere is optically denser than the
surrounding medium, m . 1, the maximal phase shift, kmd, is achieved when
a photon propagates through the center of the sphere. On the other hand, the
photon that is not scattered by the particle has undergone a minimal phase
shift, kd. Other examples of particles for which the scattering problem has
been analytically solved are cylinders, coated spheres, uniform and coated
spheroids, strips, and planes.19 In all these cases, the amplitudes are expressed
as an infinite series. The coefficients of some of these series are obtained as yet
another series. Some of the series are ill converging.
2.3.2 Approximate solutions of the scattering problem

Apparent difficulties obtaining a rigorous solution to the wave equations have
led to the development of other means of solving the scattering problem. Such
alternative solutions, which are indeed used in practice much more extensively
than rigorous solutions, have evolved in two directions: 1. approximate
solutions of the scattering theory and 2. methods of solving exact wave
equations numerically. First, we consider how the scattering problem can be
simplified under certain approximations.
A great variety of the structures that cell organelles are built of, such as
the tubules of endoplasmic reticulum, cisternae of Golgi apparatus, etc., are
small compared to the wavelength. Light scattering by such particles is known
as Rayleigh scattering and was described by Rayleigh in 1871.16 In this
approximation, the electric field is considered to be homogenous over the
volume of the particle. Therefore, the particle behaves like a dipole and
radiates in all directions. In a most relevant case of isotropic polarizability a
of the particle, the scattering amplitude matrix becomes

S2 S3 cos u 0
¼ ik a
3
. (2.6)
S4 S1 0 1
The scattered light has a well-known broad angular distribution. The

scattering cross section is simply
8
ss ¼ pk 4 a2 . (2.7)
3
We note that because a is proportional to the particle’s volume, the scattering

cross section scales with the particle’s linear dimension a as a6 and varies
inversely with l4.
When the particle is not small enough to allow the Rayleigh approxima-
tion, the discrete dipole approximation (DDA) could be employed.20 It is
particularly applicable to particles with a size comparable to their wavelength.
Thus, it may be useful to study light scattering by small organelles such as
mitochondria, lysosomes, etc. From a certain point of view, DDA is an
extension of the Rayleigh approximation. A scattering particle is considered
to be an array of N point dipoles with the spacing between these dipoles being
small compared to the wavelength. Thus, the particle need not be uniform. On
the contrary, this approximation might be a powerful tool to describe
scattering from composite particles, as most small organelles are. In this
approximation, the elements of the scattering matrix are expressed through
the dipole moments of each of the dipoles. To find the dipole moments, one
needs to solve a system of 3N linear equations with coefficients dependent on
the polarizabilities of the dipoles.
54 Chapter 2
Another approach to describe the scattering by particles comparable to

the wavelength involves the Rayleigh-Gans approximation.18 It is applicable
if the following two conditions are satisfied.
1. Relative refractive index of the particle is close to 1, |Dm 1|≪1, where
max0 ½nðrÞ
r∈V
Dm ¼ , (2.8)
min0 ½nðrÞ
r∈V
with n(r) the refractive index at a point r and V 0 the volume encompassing
the volume of the particle V and its immediate surroundings.
2. The phase shift is small, 2ka|Dm 1|≪1, with a, the linear dimension of
the particle. Such a “soft” particle can be considered an array of
noninteracting dipoles.
The refractive index of most cell organelles ranges from 1.38 to 1.42.21–23
The refractive index of the cytoplasm of most cells varies from 1.34 to
1.36. Therefore, both conditions of the Rayleigh-Gans approximation are
satisfied for most small organelles. This approximation is derived by applying
Rayleigh’s equations (2.6) to any volume element dV within the particle. It
can be easily shown that

S2 S3 ik 3 V cos u 0
¼ Rðu,fÞ , (2.9)
S4 S1 2p 0 1
with
Rðu,fÞ ¼
1
VV
∫ðmðrÞ 1Þeidðr,u,fÞdr, (2.10)
where m(r) is the relative refractive index at a point r, and d is the phase of the
wave scattered in direction (u,f) by the dipole positioned at a point r.
If a particle is sufficiently homogenous,
maxðjmðrÞ 1jÞ minðjmðrÞ 1jÞ ≪

r∈V r∈V
1
VV
∫jmðrÞ 1jdV ,
then

S2 S3 ik 3 ðm 1ÞV cos u 0
¼ Rðu,fÞ , (2.11)
S4 S1 2p 0 1
where m is the relative refractive index averaged over the volume of the
particle, and function Rðu,fÞ ¼ V1 ∫eid dV is the so-called form factor. One can
predict that the total intensity of light scattered by a small organelle increases
with the increase of its refractive index as (m 1)2 and with its size as a6. The
angular distribution of the scattered light differs from that of Rayleigh
scattering. For u ¼ 0, the form factor equals unity. In other directions, |R| , 1,
so the scattering in the forward direction prevails.
While most subcellular structures are smaller than or comparable to the
wavelength, there is one whose size is significantly larger than that of the other
organelles: the nucleus. None of the above-mentioned approximations could
be applied to the cell nucleus. Rather it falls into a different limiting case,
namely it is much larger compared to the wavelength. Such large particles
produce interesting but complicated scattering patterns. The approximate
theory of light scattering by large particles was first proposed by van de Hulst
in 1957.18 The van de Hulst approximation was originally formulated for
spherical particles only. However, it can be extended to large particles of an
arbitrary shape. Although the van de Hulst theory does not provide a
universal means for finding the scattering matrix for all scattering angles even
in the case of a homogenous sphere, it does enable scattering amplitudes to be
obtained in the near-forward direction as well as the scattering cross section.
Consider a particle that satisfies the following two conditions. The first is
the same as condition 1 of the Rayleigh-Gans approximation, |Dm 1|≪1.
The other is directly opposite to condition 2. We require that the phase shift be
large, 2ka|m 1|≫1. The phase shift of the ray emerging from the particle
depends on where this ray entered the particle and how much it was deflected.
The difference in the phase shift will create constructive or destructive
interference on a screened position far from the particle. Applying the
Huygens’ principle, one can obtain16
k2
SðuÞ ¼ ∫∫ð1 eijðrÞ Þeidðr,uÞd 2 r,
2p A
(2.12)
where r is a vector in the plane orthogonal to the direction of propagation of

the incident light, j is the phase shift gained by a light ray that enters the
particle at a position given by r and passes through the particle along a
straight trajectory relative to the phase shift gained by ray propagating outside
the particle, and d is the phase difference between the rays scattered by
different parts of the particle (Fig. 2.6). As well, we omitted the subscript after
S, since S1 ¼ S2. The integration is performed over the geometrical cross
section of the particle, A. The phase shifts depend on the particle shape and
refractive index. For example, for a spherical particle of radius a and relative
refractive index m,
j ¼ 2kaðm 1Þ cos g (2.13)
d ¼ ka sin u sin g cos w, (2.14)

where g is an angle between the radial direction and the direction of the initial
ray, and w is an azimuthal angle of a vector oriented toward an element of the
surface of the particle.
56 Chapter 2
Figure 2.6 d, the phase shift between the two light rays, is shown.
Equation (2.12) enables one to obtain the scattering amplitude for a soft
large particle of an arbitrary shape. In the case of a sphere, substituting the
expressions for the phase shifts Eqs. (2.13) and (2.14) into Eq. (2.12), and
performing the integration in the spherical coordinates, one can obtain14
pffiffiffiffiffiffiffiffiffi
a
SðuÞ ¼ k 2
∫ð1 e2ikðm1Þ a2 r2
ÞJ 0 ðkruÞrdr. (2.15)
0
This integral cannot be evaluated analytically. It must be either expanded18 or

approximated. The latter approach gives the following expression for the
intensity of small angle scattering:
rffiffiffiffi
a4 k 2 I i J 1 ðxuÞ p J 1∕2 ½yðuÞ 2
I s ðuÞ ¼ pffiffiffiffiffiffiffiffiffi
r2 xu 2 yðuÞ
2
2xðm 1Þ sin yðuÞ 2
þ ðcos yðuÞ , (2.16)
y2 ðuÞ yðuÞ
where x ¼ ka is called the size parameter, and

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
yðuÞ ¼ x u2 þ 4ðm 1Þ2 .
As one can see, the scattered intensity is highly forward directed. The width of
the first scattering lobe is about l/a. Generally speaking, the larger the
particle, the stronger and narrower the first lobe.
Figure 2.7 Total cross section s as a function of z ¼ x(m1).
The scattering cross section can be obtained using the optical theorem16
either from Eq. (2.16) or directly from Eq. (2.12) and equals

sin½2xðm 1Þ sin½xðm 1Þ 2
ss 2pa 1
2
þ . (2.17)
xðm 1Þ xðm 1Þ
This is a famous expression first obtained by van de Hulst. It shows that large
spheres give rise to a very different type of scattering than the small particles
considered above do. Both the intensity of the forward scattering and the
scattering cross section are not monotonous functions of wavelength. Rather,
they exhibit oscillations with the wavelength, called an interference structure
(Fig. 2.7). The origin of these oscillations is indeed interference between the
light ray passing through the center of the sphere and one not interacting with
it. The frequency of these oscillations is proportional to x(m 1), so it
increases with the sphere size and refractive index.
Before we leave the topic of light scattering by a sphere, let us briefly
discuss scattering in the near-backward direction. Unfortunately, obtaining
the intensity of light scattered at large angles is difficult, since one has to
consider several effects that influence the scattering pattern. For example,
multiple internal reflections and surface waves all contribute to the intensity in
the backward direction.18 It can be shown that the scattering in the near-
backward direction is given by
jSðu0 Þj2 ∝ J 20 ðxu0 þ hÞ, (2.18)
where angle u 0 is counted from the exact backward direction, and h is an

oscillating function of size parameter x and phase shift xm. One can see that
the intensity of the scattered light peaks not only in the exact forward
direction, but in the near-backward direction as well. However, the
58 Chapter 2
backscattering peak is significantly smaller than the forward scattering

peak. Another difference is that it is shifted from the exact backward
direction. This shift varies inversely with the size of the sphere and its
refractive index.
Concluding the discussion of light scattering by large particles, we would
like to take a final look at general expression Eq. (2.12). Substituting S(u)
from Eq. (2.12) into the optical theorem16
4p
ss Re Sð0Þ, (2.19)
k2
one can obtain the scattering cross section of an arbitrarily shaped and not
necessarily homogenous large particle:
ss ¼ 2∫∫f1 cos½jðrÞgd 2 r. (2.20)

A
It can easily be shown that the value of this integral, the scattering cross
section, and other integral properties of the scattering pattern depend mostly
on the maximal phase shift that a light ray can gain while traveling along a
straight trajectory inside the particle, rather than on the specific shape of the
particle. This fact was confirmed in a great number of experiments. For
example, it was found that light scattering by red blood cells can accurately be
described using the van de Hulst approximation, despite the fact that red
blood cells are highly nonspherical.24
2.3.3 Numerical solutions of the scattering problem

The analytical approaches discussed above are dependent on strict simplifying
assumptions about a tissue scatterer’s geometry and refractive index. It is not
at all clear how limiting these assumptions are and what to do when they are
not applicable in the first place. Recently, significant progress has been
achieved in the development of numerical approaches to solve the scattering
problem and to address these issues. Here we will discuss in detail two
numerical methods that have been found to be very useful in studying light
scattering from biological objects: the discrete particle model (DPM)25 and
the three-dimensional finite difference time domain (FDTD) method.26,27
The DPM is an extension of the analytical methods presented in
Sec. 2.3.2. It models a complex nonuniform scatterer as a combination of
discrete uniform objects of simple, usually spherical, shape. To conduct these
calculations, one needs to assume a particular distribution of the scatterers’
sizes and refractive indices. Normal or lognormal distribution functions are
most frequently used. Then the scattering fields are calculated numerically as
a superposition of the scattering fields produced by each of the particles using
either Mie theory or its approximations. As one can see, although this model
might give some insight into the impact of the internal structure of a particle
on the scattering pattern, it is by no means rigorous.
The FDTD presents a more general approach that does not suffer from
the limitation of the discrete particle model. It enables the computation of
scattering amplitudes for inhomogeneous objects of arbitrary shape. The
FDTD aims to find the numerical solution of Maxwell’s equations for an
electromagnetic wave propagating in a medium with given variations of the
refractive index. The equations are discretized in time and space on a four-
dimensional grid. The value of the refractive index is given for each of the
grid’s pixels. Apparently, the higher the spatial and temporal frequencies of
the grid, the more accurate the model is. However, there is a price to pay. The
complexity of the algorithm increases as l–6t–2 with a reduction of the grid
spatial spacing l and temporal interval t. It was established that l , l/10,
where l is the wavelength of the incident light guarantees reasonable
accuracy. Thus, the step is chosen as l min{l/10,dn} with dn the special
dimension of refractive index variations. Proper boundary conditions are to
be applied as well. For example, the Liao boundary conditions have been used
to simulate unbounded media.28 The incident wave is considered to be a
sinusoidal plane wave source. Thus, the system of first-order finite difference
equations is constructed. The values of six components of the electromagnetic
wave (Ex, Ey, Ez, Hx, Hy, Hz) are to be found at each of the grid elements
using conventional algorithms.
Despite its generality, the FDTD suffers from a few drawbacks. First, it is
computationally intensive. Second, although it does allow a solution for the
scattering problem, it does not necessarily help with the understanding of the
scattering characteristics. And, after all, the foremost important reason to
solve the direct scattering problem is to elucidate the general properties of the
scattering that enable at least a partial solution of the inversion problem. On
the other hand, the FDTD can be exceedingly useful in comparing various
approximate analytical models with otherwise difficult to perform experi-
ments and, therefore, in developing the most accurate analytical descriptions.
In the end, a word of caution: it would be tempting to think that the use of
numerical modeling avoids the necessity for making any simplifying
assumption regarding the origin of tissue scattering. While it is partially true
and numerical modeling can release some of the assumption, it must always
be realized that the model is only as good as its input parameters.
2.4 Light Scattering by Cells and Subcellular Structures

In the previous sections, we overviewed the structure of epithelial cells and
mucosal tissues and discussed some of the analytical and computational
methods widely used to calculate the amplitudes of the light scattered by small
particles. Now we shall see how histology and physics merge to provide the
60 Chapter 2
understanding of light scattering by living cells and tissues. Our goal will be
twofold. First, we will examine which cell structures are responsible for
various scattering patterns. Second, we will address the question of what kind
of simplified models can be used to predict the scattering patterns produced by
various subcellular structures.
As we have seen, the cells and tissues have complex structures with a
very broad range of the scatterers’ sizes: from a few nanometers, which is the
size of a macromolecule, to 7–10 mm, which is the size of a nucleus, and to
20 mm, which is the size of a cell itself. The picture is complicated even
further by the fact that only a few cell structures are homogenous with a
uniformly distributed refractive index. Most cell organelles and inclusions
are themselves complex objects with a spatially varying refractive index.
Thus, picturing a cell as being composed of a number of distinct uniform
objects of proper geometrical shape is, generally speaking, not correct. On
the other hand, several studies have confirmed that many organelles, such as
mitochondria, lysosomes, and nuclei, do posses an average refractive index
substantially different from that of their surroundings (Table 2.1); therefore,
viewing a cell as an object with a continuously or randomly varying
refractive index is not accurate either. A more accurate model acknowledges
subcellular compartments of various sizes with refractive indices, though not
constant over the compartment’s volume, but different from that of the
surroundings. A similar model would describe an epithelium or a mucosal
tissue as well.
Studies of light scattering by cells have a long history. The first works in
this area investigated the angular dependence of the scattered light. Most of
the experiments were performed on a single wavelength and the angular
distribution of the scattered light was measured either with an array of
photodetectors, fiber optics, or charge-coupled devices (CCD). Brunsting et al.
initiated a series of experiments aiming to relate the internal structure of
living cells with the scattering pattern by measuring forward and near-forward
scattering by cell suspensions.29 This became one of the first attempts to explain
light scattering by cells using rigorous quantitative approaches.
The researchers used cells of several types such as Chinese hamster’s
oocytes (CHO), HeLa cells, and nucleated blood cells. They compared the
resulting angular distribution of the scattered light with the one predicted by
the Mie theory and found that a very good agreement between the theory and
Table 2.1 Index of Refraction Values

Organelle Refractive Index References
Extracellular fluid 1.35–1.36 21
Cytoplasm 1.36–1.37 22
Nucleus 1.38–1.47 21, 23
Mitochondria 1.38–1.41 21
experiment was achieved when a cell was approximated as a denser sphere

imbedded into a larger and softer sphere. The sizes of these spheres
corresponded to the average sizes of the cell nuclei and cells, respectively.
Considering scatterers of smaller sizes in the model did not improve the fits.
This result agrees well with the scattering theory. Indeed, as discussed in
Sec. 2.3, the particles that were large compared to the wavelength produce the
scattered field that peaks in the forward and near backward directions in
contrast to smaller particles, which scatter light more uniformly. In fact, the
width of the forward peak, uf, is proportional to the ratio of the wavelength, l,
to the particle’s size, a:
l
uf ¼ .
a
Therefore, it is not surprising that the structures with the biggest dimensions,
namely cells and their nuclei, are the major scatterers in the forward direction.
Another interesting conclusion can be drawn from these experiments:
despite the nonhomogeneity and a not perfectly spherical shape of the cells
and their nuclei, the experimental results were explained using the Mie theory,
which deals with uniform spheres. This finding should not be surprising as
well. It can be anticipated in the light of the van de Hulst theory. Indeed, the
scattering pattern produced by a large particle (for simplicity, we refer to
particles that are large compared to the wavelength as “large” particles) in the
near-forward direction depends mostly on the maximal phase shift undergone
by the incident light wave rather than variations of the refractive index inside
the particle [see Eq. (2.15)]. This result was later supported by Sloot et al.23 in
the experiments with white blood cells (leukocytes) and Hammer et al. in the
experiments with red blood cells.24 The former group of researchers found
that light scattering by the leukocytes in the near forward direction could be
explained if each cell was approximated as being composed of two concentric
spheres, one being the cell itself and the other being the nucleus. The latter
group showed that the near-forward scattering of light by red blood cells can
accurately be described using the van de Hulst approximation, which was
derived for large particles of spherical shapes rather than the concave-convex
disks that the red blood cells really are. One important comment must be
made. There is an apparent distinction between the experiments with cell
suspensions and those involving living epithelia: when the scattering by a cell
suspension is measured, the cells might provide important contribution to the
scattering. On the contrary, the epithelia consist of one or several layers of
contiguous cells with minute (a few nanometers) or no extracellular space at
all. Thus, the contribution to the scattering from a cell as a whole is expected
to be much less and the subcellular structures are to be viewed as major
scatterers.
62 Chapter 2
Figure 2.8 Schematic of the goniometer system used for measuring angular scattering
distribution, P(u). A collimated light beam from the He-Ne laser is incident on a cylindrical
sample cell containing a suspension of biological cells. The scattered light is measured as a
function of angle by a photomultiplier tube, which is rotated around the sample cell. The thick
black line represents a thin tube that is black on the inside and is critical for angular
resolution and elimination of stray light. The intensity as a function of wavelength must be
multiplied by the geometrical factor, cosu, to account for the change in acceptance angle as
the detector is rotated around the sample.
Extensive studies of the angular dependence of light scattering by cells

using a goniometer (Fig. 2.8) were carried out by Mourant et al.30
Measurements of light scattering from cells and cell organelles were
performed from 2 to 171 deg and from 9 to 168 deg, respectively. In both
cases, the unpolarized light was delivered by a He-Ne laser at 632.8 nm. The
angular resolution was about 0.5 deg and most of the data was taken for every
2 deg. The concentration of the cells was 105 cells/mL. This concentration was
chosen so that multiple scattering events would be rare. Two types of cells
were used in their experiments: immortalized rat embryo fibroblast cells (M1)
and a ras-transfected clone (MR2), which is highly tumorigenic. The cells
were suspended in phosphate-buffered saline and kept on ice. Nuclei and
mitochondria were isolated from MR1 cells by standard methods and
resuspended mannitol sucrose buffer.
Examples of the angular scattering distributions, P(u), for M1 cells, and
their nuclei and mitochondria, are shown in Fig. 2.9. The researchers found
that MR1 cells produced similar scattering patterns. As can be seen, the
scattering from the nuclei most closely resemble that from the cells. In
particular, the scattering at near-forward angles is attributable to the nuclei.
At the same time, mitochondria were found to scatter more strongly at larger
angles, from 15 to 160 deg.
One can also notice an increase in the scattering by the nuclei and cells in
the near-backward directions for angles above 160 deg. As discussed in the
previous section, such an increase is characteristic for scatterers that are large
compared to the wavelength of the incident light. On the contrary, particles of
smaller or comparable sizes with the wavelength do not give rise to the
Figure 2.9 Measured angular scattering distribution, P(u), for cells, nuclei, and mitochon-
dria. Values below 9 deg and above 168 deg were extrapolated. All curves are normalized
so that the integral of the P(u) over the 4p solid angle equals unity.
backscattering peak. Their intensity profile is rather flat for these angles,
which is illustrated by the flat P(u) for angles u . 150 deg. For comparison,
the intensity of light scattered by the cell nuclei and cells themselves, which is
more than an order of magnitude larger than the wavelength, does exhibit a
prominent peak in a near-backward direction, as expected.
These studies showed that the cell structures responsible for light
scattering might be correlated with the angle of scattering. The following
generalizations, though by no means universal, can provide a convenient
paradigm to think about light scattering by biological cells:
1. When a cell is suspended in a buffer solution of lower refractive index, the
cell itself is responsible for small angle scattering. This result has been used
in flow cytometry to estimate the cell sizes.31 However, this is not the case
for contiguous layers of cells such as epithelia, where all cells are
surrounded by the other cells with similar optical properties rather than a
medium of a different refractive index.32 In the latter case, most of the
scattering is attributable to particular structures within the cells.
2. At slightly larger angles, the nucleus is primarily responsible for
scattering. Moreover, it is the major scatterer in all forward and
near-forward directions, in the case when the cell is a part of a contiguous
layer.
3. Smaller organelles, cell inclusions, and suborganellar and subnuclear
inhomogeneities are likely responsible for scattering at larger angles.
Unfortunately, it is difficult to be more specific due to the fact that there is
too great a variety of cell components with sizes smaller than or
comparable to the wavelength. To name a few: mitochondria, lysosomes,
Golgi apparatus, mostly all cell organelles other than the nucleus, plus
64 Chapter 2
subnuclear inclusions such as nucleolus and clumps of chromatin—all

have sizes of this scale. Scattering may originate from organelles
themselves or their internal components. Moreover, one should not forget
that light scattering might occur from the patterns formed by a number of
neighboring objects. The complexity of this picture does not imply that
scattering at large angles cannot provide any clues about the origin of the
scattering. The angular dependence might potentially elucidate whether
the scattering originates from the objects of regular or irregular shape,
spherical or elongated, inhomogenous or uniform. However, rigorous
approaches to accomplish this have yet to be developed. In some cases, the
large angle scattering can be attributed to a specific predominant
organelle. For example, the research conducted by Beavoit et al.21
provides strong evidence that mitochondria are primarily responsible for
light scattering from hepatocytes, the cells known to possess very
numerous mitochondria that can reach a count of thousands. However,
the hepatocytes are exceptions with their rich mitochondrial content. Most
other cells including most mucosal epithelia possess much fewer
mitochondria.
4. Finally, the increase of the light scattering intensity in the near-backward
directions is attributable to larger structures within the cell, such as the
nucleus.
Not only does light scattered by cell nuclei have a characteristic angular
distribution peaked in the near-backward directions, but it also exhibits
spectral variations typical for large particles, as discussed in Sec. 2.3.
Perelman et al. observed such a spectral behavior in the light backscattered
from the nuclei of human intestinal cells.10 The cells, approximately 15 mm
long, affixed to glass slides in a buffer solution, formed a monolayer of
contiguous cells similar to the epithelial lining of the colon mucosa. In the
experiments, an optical fiber probe (NA ¼ 0.22) was used to deliver white light
from a xenon arc lamp onto the sample and to collect the reflected signal.
After the measurement was performed, the cells were fixed and stained with
H&E, a dye that renders otherwise transparent cell nuclei visible under
microscope examination and that is widely used in biology and medicine to
examine tissue morphology. Microphotographs of the monolayer were
obtained, and the size distribution of the nuclei was measured. It centered
at about 6 mm and had a standard deviation of approximately 0.5 mm.
Figure 2.10 compares the wavelength varying component of light
backscattered by the cells, Rs(l), measured in the experiments with one
calculated with the Mie theory using the size distribution of the cell nuclei
determined via microscopy. As can be seen, both spectra exhibit similar
oscillatory behavior. The fact that light scattered by a cell nucleus exhibits
oscillatory behavior with the frequency depending on its size was used to
develop a method of obtaining the size distribution of the nuclei from the
1.06
1.04
1.02
Rs
1
0.98
0.96
0.94
350 450 550 650
λ (nm)
Figure 2.10 Oscillatory component of light backscattered by a monolayer of intestinal
cells. The dotted line shows the spectrum measured in the experiment. The solid line shows
the signal predicted by the Mie theory.
spectral variations of light backscattered by biological tissues. As discussed in

Sec. 2.6, this method was successfully applied to diagnose precancerous
epithelia in several human organs in vivo.11
Not only the organelles themselves, but also their components can scatter
light. This raises an important question: how does scattering depend on the
internal structure of a scattering object, meaning variations of the refractive
index inside the cell or an organelle? For example, what is the difference
between the scattering pattern produced by a relatively uniform nucleus
and one with a complex internal structure, and, therefore, highly varying
refractive index? Unfortunately, it is rather difficult to address these issues
experimentally. As mentioned in the previous section, finite-difference time-
domain (FDTD) simulations provide a means by which to study spectral and
angular features of light scattering by arbitrary particles of complex shape and
density. It might be considered as a numerical experiment substituting for a
real experiment in cases when the latter is not possible. Using FDTD and
choosing proper models, one can learn a great deal about the origins of light
scattering.
Drezek et al. investigated the influence of cell morphology on the
scattering pattern.32 Figure 2.11 demonstrates the effect of changing the
frequency of the spatial variation of the refractive index while keeping
the mean refractive index value constant for a cell with a randomly generated
dielectric structure. The figure plots the normalized scattering pattern for two
cells. Both cells have a mean refractive index of n ¼ 1.4 with uniformly
distributed variations between n ¼ 1.35 and n ¼ 1.45. In one cell, the spatial
frequency of the index variations ranges from 5 to 20 mm–1, while in the other
cell, the spatial frequency is lower and ranges from 2 to 10 mm–1. The curves
66 Chapter 2
Figure 2.11 Scattering pattern of angular dependence of the intensity of light scattered by
two cells with randomly assigned dielectric structure. The spatial frequency of the refractive
index fluctuations is higher in the top curve (labeled high frequency) than in the lower curve
(labeled low frequency). The mean refractive index is the same for both curves.
demonstrate that as the frequency of the refractive index variations increases,

the scattering intensity becomes higher at large angles. However, scattering
intensities are similar for small angles and in the near-backward directions.
Because in these simulations the cell was modeled as an object embedded into
a medium with a lower refractive index, the results can be generalized for any
large structure within the cell, such as the cell nucleus. Thus, the internal
structure of an organelle does affect the scattering at large angles, but not in
the forward or backward directions. In fact, this finding is not paradoxical
and should be expected: light scattered in the forward or backward directions
depends strongly on the larger structures within an organelle, for example, the
organelle itself. It samples the average properties of the organelle, which were
kept constant in the simulations. On the other hand, smaller structures within
the organelle scatter strongly in the intermediate angles. Thus, light scattering
at these angles is influenced by its internal structure.
2.5 Light Transport in Superficial Tissues

Light transport in biological tissue is dominated by elastic scattering. The
primary scattering centers are thought to be the collagen fiber network of the
extracellular matrix, the mitochondria, and other intracellular substructures,
all with dimensions smaller than the optical wavelength. However, larger
structures, such as cell nuclei, typically 5 to 15 mm in diameter, can also
scatter light. Because of the ubiquity of light scattering, its effects are
enmeshed with those of absorption, making interpretation of tissue spectra
difficult. On the other hand, this feature can actually enrich the information
provided by spectroscopic techniques.
Consider a beam of light incident on an epithelial layer of tissue.
A portion of this light is backscattered from the epithelial cells, while the
remainder is transmitted to deeper tissue layers where it undergoes multiple
scattering and becomes randomized. All of the diffusive light that is not
absorbed in the tissue eventually returns to the surface, passing once more
through the epithelium where it is again subject to scattering from the cells.
Thus, the emerging light will consist of a large component of multiply
scattered light and a smaller component of singly scattered light. For a thin
slab of epithelial tissue, an approximate solution of the transport equation for
the signal R(l) collected within an acceptable solid angle Vc is given by the
following expression10
RðlÞ 1 etðlÞ
¼ etðlÞ þ hhI ðl, s0 Þpðl,s, s0 ÞiVi þ hI d ðl,s0 Þpðl,s,s0 Þi2p iVc ,
RðlÞ hI d ðl,sÞiVc i
(2.21)
where t is the optical thickness, Ii (l,s) is the intensity of the incident light
delivered in solid angle Vi, Id(l,s) is the intensity of the light emerging from
the underlying tissue, and
hf ðs,s0 ÞiV ¼ ∫f ðs,s0 Þds0

V
for any function f and solid angle V, with s a unit vector pointing outward
from the tissue surface in an arbitrary direction. The quantity
RðlÞ ¼ hI d ðl,sÞiVc ∕hI i ðl,sÞiVc
is the reflectance of the diffusive background. In the case when the scatterers
are distributed uniformly throughout the thickness of the epithelium and
discrete particle approximation is applicable, the optical thickness is simply
X
tðlÞ ¼ ss ðl,jÞN j ,
j
with ss(l,l) the scattering cross section of a particle of type j (the type of a
particle may be defined as its shape, size, refractive index, etc.) and Nj the
distribution function of the particles depending on their type (number of
68 Chapter 2
scatterers of type j per unit area). In the same approximation, the effective
scattering phase function is
1X
pðl,s,s0 Þ ¼ pðl, j,s,s0 Þss ðl, jÞN j
t j
with p(l, j,s,s 0 ) the scattering phase function for a particle of type j. In the case
of spherical scatterers, this function is determined by the Mie theory. The first
term in Eq. (2.22) describes the attenuation of the diffusive component, and
the terms in brackets describe backscattering of the incident light and forward
scattering of the diffusive component by the scatterers within the epithelium,
respectively.
For a small Vc the forward scattering and backscattering terms in
Eq. (2.21) can be expanded in t(l). The forward scattering term varies with
the wavelength in phase with t(l), as required by the optical theorem,16
whereas the backscattering term is out of phase. Thus, the light scattered in
the epithelium introduces into the signal of light emerging from a tissue a fine
wavelength dependent component, whose spectral shape depends on the
optical properties of the scatterers and that can be accounted for using the
scattering theory. In turn, some of the properties of the scattering particles
within the epithelium might be revealed by analyzing the spectral variations of
the reflected signal R(l).
The fine structure component is typically just a few or less than a percent
of the total reflected signal and is ordinarily masked by the background of
diffusely scattered light from underlying tissue, which itself exhibits spectral
features due to absorption and scattering. The spectrum of the diffusely
scattered light can be described using a quantity called diffuse reflectance.
Diffuse reflectance is determined by absorption and reduced scattering
coefficients, ma and m0s , respectively. Absorption and scattering coefficients, in
turn, depend linearly on the concentrations of the various tissue components.
Several researchers have employed models, often based on the diffusion
approximation, to extract optical parameters from diffuse reflectance.
Quantitative results have been obtained in blood oximetry33,34 and in the
study of the optical properties of various animal tissues such as rat prostate,35
chicken breast,36 and canine gut.37 Zonios et al. developed a method for
modeling the reflectance of colon tissue and extracting such properties as
hemoglobin concentration, hemoglobin oxygen saturation, effective scatterer
density, and effective scatterer size.38 The method is based on the model
derived from the diffusion approximation and assumes colon tissue to be a
homogeneous semi-infinite turbid medium. Part of the incident light is
absorbed in the tissue, whereas the nonabsorbed part is subject to multiple
scattering and eventually emerges from the surface as diffuse reflectance.
A certain fraction of this emerging light is collected by the probe, whereas the
remaining part escapes undetected. The amount of the light collected depends
on the optical properties of tissue as well as on the probe radius rc . This radius
serves as a scale length, enabling ma and m0s to be determined.
Starting with an expression derived by Farrell et al.,39 Zonios et al. obtained
an analytical expression for the diffuse reflectance collected by the probe:
0 mr0
m0s mz0 ð1þ43AÞmz0 emr1 4 e 2
Rp ðlÞ ¼ 0 e þe z0 0 1 þ A z 0 0 ,
ms þ ma r1 3 r2
(2.22)
with
m ¼ ½3ma ðma þ m0s Þ1∕2 ,
1
z0 ¼ ,
m0s þ ma
r01 ¼ ðz20 þ r02

cÞ
1∕2
,
and
2 1∕2
4
r02 02
¼ z0 1 þ A þ rc
2
.
3
The parameter A depends in the known way on the refractive index n of
the medium. For a given probe geometry there is an optimal value of the
effective probe radius rc, which can be determined by calibrating Eq. (2.22)
using the reflectance measurement of a tissue phantom with known optical
properties.
In the visible spectral range in colon tissue, hemoglobin encountered in
both oxygenated and deoxygenated forms appears to be the only significant
absorber. Thus the total absorption coefficient, ma(l) is given by
ma ðlÞ ¼ ln 10 cHb ½a εHbO2 ðlÞ þ ð1 aÞ εHb ðlÞ, (2.23)
with a the hemoglobin oxygen saturation parameter and cHb the total
hemoglobin concentration. The molar extinction coefficients, εHbO2 (l) and
εHb (l), are well known.40
In general, the reduced scattering coefficient m0s ðlÞ is the sum of
contributions from the various tissue scatterers. Detailed information about
these individual scatterers is not well known. Therefore, one can express
m0s ðlÞ ¼ rs s0s ðlÞ, (2.24)

with rs the effective scattering density and s0s ðlÞ the effective reduced
scattering cross section. Equation (2.24) models the tissue scattering properties
70 Chapter 2
in an average way, as if tissue contained a single well-defined type of scatterer.

In general, s0s ðlÞ depends on the refractive index, shape, and size of the
scatterer, as well as on the refractive index of the surrounding medium.
In summary, such models predict that the spectra of the diffusely scattered
light depend on a few tissue parameters, such as density of tissue scatterers,
their scattering cross section (which is a function of the scatterer size and
refractive index), the concentration of hemoglobin, and oxygen saturation of
the tissue. As we will see in Sec. 2.6, spectral analysis of the reflected light
enables determination of these parameters, which, in turn, could be used for
tissue characterization and detection of precancerous changes.
2.6 Detection of Cancer with Light Scattering Spectroscopy

In the previous section we described an approach that distinguishes two
components of light returned from a biological tissue by means of scattering.
The photons returned after a single scattering in the backward or near-
backward directions produced a so-called single-scattering component. The
photons returned after multiple scattering events produced the diffuse
reflectance. Accordingly, the part of the tissue responsible for the formation
of the single scattering component is said to be the “upper” layer. Although
this approach is very simplistic, it does provide a convenient framework to
explain some of the important effects in the formation of the back-reflected
signal.
The spectra of both single-scattering and diffusive signals contain valuable
information about tissue properties. However, the type of information is
different. The single-scattering component is sensitive to the morphology of
the upper tissue layer, which in case of any mucosal tissue almost always
includes or is limited by the epithelium. Its spectroscopic features are related
to the microarchitecture of the epithelial cells, sizes, shapes, and refractive
indices of their organelles, inclusions, and suborganellar components and
inhomogeneities. Thus, analysis of this component might be useful in
diagnosing diseases limited to the epithelium, such as preinvasive stages of
epithelial cancers, dysplasias, and carcinomas in situ (CIS).
The diffusive component contains information about tissue scatterers and
absorbers as well. However, due to multiple scattering, information about
tissue scatterers is randomized as light propagates into the tissue, typically
over one effective scattering length (0.5 to 1.5 mm, depending on the
wavelength). Moreover, the diffusive light samples considerably deeper tissue
than the single-scattered component does. With the exception of a very few
cases, such as the thick skin on the sole of the foot, the tissue sampled by the
diffusive light is never limited to the epithelium. Thus, the diffusive
component is much less sensitive to individual structures of the epithelial
cells. Rather it carries important information about tissue architecture
and composition and is particularly useful in diagnosing diseases in the

following cases:
1. The disease is not limited to the epithelium. Invasive cancer is an
example.
2. The diseased epithelial cells themselves do not penetrate the basement
membrane, but the normal structure of the under-epithelial tissue is
altered in the process of the disease. Some developed dysplastic lesions are
able to induce the angiogenesis resulting in the increased blood supply to
the lesion. Since the hemoglobin contained in the red blood cells is one of
the major endogenous absorbers of most mucosal tissues, its concentration
increase affects the spectrum of the diffusive light in a manner predicted
by Eq. (2.22).
3. The disease is limited to the epithelium, but the epithelial optical thickness
is increased so much that a considerable portion of the diffusive light is
confined to this layer. In some cases, the later stages of CIS of the
stratified squamous epithelia of such organs as the cervix or oral cavity
can make the epithelium appear thick enough to trap a significant part of
the diffusive photons. Because the scattering and particularly absorbing
properties of the epithelia and underlying connective tissues are different,
such a change in the epithelial thickness might affect the spectral features
of the diffusive component.
Generally speaking, the single-scattering component is more important in
diagnosing the initial stages of the epithelial precancerous lesions, while the
diffusive component carries valuable information about more advanced
stages of the disease. In this section, we will discuss the application of
both components to the diagnosis and detection of epithelial cancers and
precancerous lesions.
2.6.1 Diagnosis of early cancer and precancerous lesions with

diffusely scattered light
Diffuse reflectance spectra are usually collected in a broad wavelength range
that might span from 320 to 1100 nm with the use of an optical fiber probe.
The probe may consist of as few as two optical fibers. In this case, one fiber is
used for delivery and the other is used for collection of the diffusely reflected
light. In order to increase the signal-to-noise ratio, or to sample larger areas
and depths, it might be necessary to incorporate a large number of optical
fibers into the probe. Some groups reported using as many as 30 fibers.41 The
diameters of the fibers are chosen to satisfy specific technical requirements of
the system. For example, achieving efficient coupling of the probe and a CCD
or spectrophotometer may restrict the overall diameter of the probe and,
therefore, the diameters of the fibers. The numerical aperture (NA) of the
fibers is chosen using similar considerations. In many experiments involving
72 Chapter 2
measurement of the diffuse reflectance, the fibers are chosen to be less than
200 mm in diameter with NA ¼ 0.22.
For protection of the probe surface, a few millimeters shield made of an
optically transparent material such as quartz is put on the tip of the probe.
A thicker shield would allow photons of shorter pathlengths to be collected.
On the other hand, a thinner shield selects for photons traveling longer paths
inside the tissue and, therefore, sampling deeper tissue. Thus, the thickness of
the shield can be chosen depending on how deep the tissue of greatest interest
is located. A probe based on a frequently used design consists of a 200-mm
central delivery fiber concentrically surrounded by six collection fibers of the
same diameter. The probe tip is covered by a 1-mm quartz shield beveled to
prevent specular reflection from the surface of the probe.
Figure 2.12 shows an example of a system designed to collect the diffuse
reflectance from a tissue in a broad spectral range. The light from a white light
source, such as a xenon or tungsten lamp, is delivered onto the tissue surface
via the delivery arm of the probe. The other arm collects the returned light and
directs it to the spectrophotometer. The data are then transferred to a
computer for display and analysis. The reflectance signal R(l) is defined as
follows
SðlÞ DðlÞ
RðlÞ ¼ , (2.25)
S ref ðlÞ DðlÞ
where S(l) is a signal of the reflected light measured with the spectrophotom-
eter, Sref(l) is a reference signal defined as 100% reflectance measured with a
white standard, and D(l) is a spectrophotometer’s dark current signal.
Figure 2.12 An example of equipment used to acquire reflectance spectra from skin
tumors in situ.
The spectrum of the diffuse reflectance R contains information about

tissue scatterers and absorbers, and therefore, about tissue structure and
biochemical composition. These properties, in turn, could be related to disease
diagnosis. Two major approaches to extract this information from the
reflectance spectra, which could be called the analytical approach and the
diagnostic approach, have been developed.
1. Analytical approach. This approach aims to answer the following
questions: why does a spectrum have a particular shape and amplitude?
Can it be explained with an analytical model? What are the principal
characteristics of tissue that affect the spectrum? How are these
characteristics parameterized? As long as these questions are answered,
an algorithm can be developed to obtain these parameters from the
measured spectrum. In the case where these parameters correlate with the
pathophysiology of the disease, their values are diagnostically significant
and a diagnostic algorithm can readily be developed. For example, the
diffusion model discussed in the previous section was applied to explain
the signals from normal colonic mucosa and adenomatous polyps. This
model was then used to obtain the effective scattering and absorption
coefficients m 0 s and ma that correlated with the disease state.38
2. Diagnostic approach. This approach does not aim to explain the spectral
features. It does not require development or assuming any analytical
approach describing the formation of the spectrum. Rather, the spectra
are examined to elucidate some spectral metrics that correlate with the
disease with no concern about the underlying cause of these differences. In
the simplest case, this can be achieved by visual examination of the
spectral shape and observing prominent differences between the spectra of
nondiseased and diseased tissues. For example, the hemoglobin absorp-
tion, which indicates higher blood supply to the tissue, can significantly
alter a spectrum by diminishing the signal at characteristic wavelengths.
However, in most cases a more rigorous approach is necessary. For
example, statistical techniques such as principal component analysis42 or
even neural network algorithms have been used. It might seem that the
diagnostic approach is less “proper” than the analytical approach. This is
not entirely correct. While it is true that all the diagnostic features singled
out using the diagnostic approach can be obtained using rigorous
analytical models, the development of such models is often difficult or
even impossible due to the exceedingly complex organization of the tissue.
The diagnostic capabilities of the diffuse reflectance spectroscopy were
tested in most organs accessible by visible light (with light either delivered and
collected directly or through optical fibers). The priority in application of
reflectance spectroscopy to diagnose lesions probably belongs to ancient
Egyptian or Chinese doctors who successfully used the color of a lesion as a
74 Chapter 2
diagnostic criterion. Both analytical and diagnostic approaches have been

used. Let us start with examples that could be considered an application of the
analytical approach. Skin is obviously the most accessible of all the organs
and was the first one to be studied with reflectance spectroscopy. Several
authors developed diagnostic metrics based on the spectral regions of
hemoglobin absorption bands (400 to 440 and 540 to 580 nm). Since the
intensity of the signal in these regions of the spectrum varies inversely with the
amount of hemoglobin present in the tissue, other conditions being equal,
this metric semiqualitatively characterizes the amount of blood supply to the
tissue.43,44 As for the diagnostic approach, Marchesini et al. developed
another metric to distinguish between malignant melanoma of skin and
benign naevi with high sensitivity.45 They normalized all lesion spectra by the
signal from the nearby skin to obtain so-called “effective reflectance spectra,”
RL ðlÞ
RE ðlÞ ¼ , (2.26)
RH ðlÞ
and found that the best discrimination was achieved if four spectral features
were used: 1) slope of the reflectance spectrum, RE(420) to RE(600), 2) the
area enclosed between the straight line that joins RE(600) to RE(780) and
RE(l) in the range from 600 to 700 nm, 3) the mean of RE(l) between 700 and
780 nm, and 4) the area enclosed between the straight line obtained by linear
fitting RE(l) and RE(l) in the 700 to 780-nm wavelength range. Mourant et al.
demonstrated that the reflectance spectroscopy was able to diagnose the
transitional carcinoma of the urinary bladder and adenoma and adenocarci-
noma of the colon and rectum with good accuracy.2,46 In these cases, the
hemoglobin absorption metric was used.
As we have already mentioned in the previous section, Zonios et al.
studied the capability of the diffuse reflectance spectroscopy to diagnose
colonic precancerous lesions, adenomatous polyps in vivo.38 Figure 2.13 shows
typical diffuse reflectance spectra from one adenomatous polyp site and one
normal mucosa site. Significant spectral differences are readily observed,
particularly in the short-wavelength region of the spectrum, where the
hemoglobin absorption valley around 420 nm stands out as the prominent
spectral feature. This valley is much more prominent in the spectrum of the
adenomatous polyp. This feature, as well as more prominent dips around 542
and 577 nm, which are characteristic of hemoglobin absorption as well, are all
indicative of the increased hemoglobin presence in the adenomatous tissue.
Apparently, the differences between these spectra are due to changes in
the scattering and absorption properties of the tissues. Both the absorption
dips and the slopes of the spectra are sensitive functions of the absorption and
scattering coefficients, providing a natural way to introduce an inverse
algorithm that is sensitive to such features. The authors quantified the
absorption and scattering properties using the diffusion-based model
Figure 2.13 Typical normal and adenomatous polyp spectra (thick curves) and modeled
spectra (thin curves).
Figure 2.14 Scattering spectra obtained from the data shown in Fig. 2.13 (thin curves) and
corresponding Mie theory spectra (thick curves). The effective scattering sizes are indicated.
discussed in Sec. 2.5. Equation (2.22) was fit to the data using the Levenberg-
Marquardt minimization method. Thus, the total hemoglobin concentration
cHb and hemoglobin oxygen saturation a were obtained. The optimal reduced
scattering coefficient ms 0 (l) was also found for each wavelength l, ranging
from 360 to 685 nm. It was found that ms 0 (l) has a spectral dependence that
resembles a straight line declining with wavelength l. The slope of ms 0 (l)
decreases with the increasing effective size of the scatterers, ds (Fig. 2.14).
76 Chapter 2
Effective Scatterer Size (μm)
0
0 50 100 150
Hemoglobin Concentration (mg/dL)
Figure 2.15 Binary plot of total hemoglobin concentration cHb vs. effective scatterer size
ds. Open circles–nondysplastic colon, solid circles–adenomatous polyps (dysplasia).
This allowed the effective scatterer size to be determined from the known
ms 0 (l). The model fits shown in Fig. 2.13 are in very good agreement with the
experimental data.
The authors applied this algorithm to analyze the spectra collected from
several adenomatous as well as nondysplastic tissue sites, obtaining values
of the four parameters for each site probed. Figure 2.15 shows a binary plot
of the effective scatter size versus hemoglobin concentration. Note that the
normal mucosa data tend to form a cluster, while the adenomatous polyp
data are separated and characterized by a wider spread and irregular
distribution. Adenomatous colon polyps were characterized by increased
hemoglobin concentration, in agreement with published results that
precancerous tissues such as adenomatous polyps exhibit increased
microvascular volume.47,48 The hemoglobin oxygen saturation was found
to be approximately 60% on average for both normal mucosa and
adenomatous polyps. This result is reasonable, as the measurements were
performed in the capillary network of the mucosa, where oxygen is
transferred from hemoglobin to tissue.
Characteristic differences in the scattering properties were also observed
between the two tissue types studied. For adenomatous polyps, the average
effective scattering size was larger, and the average effective scatterer density
was smaller as compared to normal mucosa. The range of effective scattering
sizes was in good agreement with that reported for average scatterer sizes of
biological cell suspensions.30 Although the reason for the increase of the
scatterers’ sizes in the adenomas is not clear, one can speculate that it is due
to the increased cellular content of the polyps. These studies have
demonstrated the potential of diffuse reflectance spectroscopy to obtain
quantitative information about tissue structural composition in vivo and in
real time.
2.6.2 Diagnosis of early cancer and precancerous lesions with single-

scattered light
The single scattering component of the returned light contains information
about the structure of the uppermost epithelial cells. It has been shown that
LSS enables quantitative characterization of some of the most important
changes in tissues associated with precancerous and early cancerous
transformations, namely, enlargement and crowding of epithelial cell
nuclei.10,11,49 As discussed above, typical nondysplastic epithelial cell nuclei
range in size from 4 to 10 mm. In contrast, dysplastic and malignant cell nuclei
can be as large as 20 mm. Single scattering events from such particles, which
are large compared to the wavelength of visible light (0.5 to 1 mm), can be
described by the Mie theory. This theory predicts that the scattered light
undergoes small but significant spectral variations. In particular, the spectrum
of scattered light contains a component that oscillates as a function of
wavelength (see Sec. 2.3.2). The frequency of these oscillations is proportional
to the particle size. Typically, normal nuclei undergo one such oscillation
cycle as the wavelength varies from blue to red, whereas dysplastic/malignant
nuclei exhibit up to two such oscillatory cycles. Such spectral features were
observed in the white light directly backscattered from the uppermost
epithelial cell nuclei in human mucosae.10
When the epithelial nuclei are distributed in size, the resulting signal is a
superposition of these single frequency oscillations, with amplitudes
proportional to the number of particles of each size. Thus, the nuclear size
distribution can be obtained from the amplitude of the inverse Fourier
transform of the oscillatory component of light scattered from the nuclei.10
Once the nuclear size distribution is known, quantitative measures of nuclear
enlargement (shift of the distribution toward larger sizes) and crowding
(increase in area under the distribution) can be obtained. This information
quantifies the key features used by pathologists in the histologic diagnosis of
dysplasia and CIS, and can be important in assessing premalignant and
noninvasive malignant changes in biological tissue in situ.
However, single scattering events cannot be directly observed in in vivo
tissues. Only a small portion of the light incident on the tissue is directly
backscattered. The rest enters the tissue and undergoes multiple scattering from
a variety of tissue constituents where it becomes randomized in direction,
producing a large background of diffusely scattered light. Light returned after a
single scattering event must be distinguished from this diffuse background. This
requires special techniques because the diffusive background itself exhibits
prominent spectral features dominated by the characteristic absorption bands
of hemoglobin and the scattering of collagen fibers, which are in abundance in
the connective tissue lying below the epithelium.
Several methods to distinguish single scattering have been proposed.
Field-based light scattering spectroscopy50 and spectroscopic optical
78 Chapter 2
coherence tomography51 were developed for performing cross-sectional

tomographic and spectroscopic imaging. In these extensions of conventional
optical coherence tomography (OCT),52 information on the spectral content
of backscattered light is obtained by detection and processing of the
interferometric OCT signal. These methods allow the spectrum of back-
scattered light to be measured either for several discreet wavelengths,50 or
simultaneously over the entire available optical bandwidth from 650 to
1000 nm51 in a single measurement.
Another method49 is based on the fact that initially polarized light loses its
polarization when traversing a turbid medium such as biological tissue.
Consider a mucosal tissue illuminated by linearly polarized light. A small
portion of the incident light will be backscattered by the epithelial cell nuclei.
The rest of the signal diffuses into the underlying tissue and is depolarized by
multiple scattering. In contrast, the polarization of the light scattered
backward after a single scattering event is preserved. Thus, by subtracting
the unpolarized component of the reflected light, the contribution due to
the backscattering from epithelial cell nuclei can be readily distinguished. The
residual spectrum can then be analyzed to extract the size distribution of the
nuclei, their population density, and their refractive index.
This method was implemented as follows: collimated polarized light from
a broadband source is delivered on a tissue sample. The returned light is split
into two orthogonally polarized signals, I|| with the polarization vector
parallel to that of the incident light and I⊥ with the perpendicular polarization
vector, by means of a broadband polarizing beamsplitter cube. The output
from this cube is delivered through optical fibers into two channels of a
multichannel spectroscope (Fig. 2.16). This enables the spectra of both
components, I|| and I⊥, to be simultaneously measured in the range of 400 to
900 nm. The experiments have shown that taking away the unpolarized
component of the reflected light by means of subtracting I⊥ from I|| allowed
the single scattering component to be accurately restored.
The applicability of the technique to biological tissues was tested in studies
with normal and cancerous human colon tissue samples obtained immediately
after surgical resection. Figure 2.17 shows the size and refractive index
distributions of the epithelial cell nuclei obtained for normal and cancerous
tissue samples. For the normal tissue sample, the average diameter was found
to be d ¼ 4.8 mm, the standard deviation of the sizes was s ¼ 0.4 mm, and the
relative refractive index of the nuclei was m ¼ 1.035. For the cancerous tissue
sample, the corresponding values were d ¼ 9.75 mm, s ¼ 1.5 mm, and m ¼ 1.045.
As can be seen in Fig. 2.17, the populations of normal and cancerous cell
nuclei are clearly distinguishable. Cancerous nuclei are noticeably enlarged
and have a higher refractive index. The increase in the nuclear refractive index
from normal to cancerous tissue is also characteristic. As mentioned earlier,
cancerous and dysplastic nuclei are known to stain darker than benign nuclei
Multichannel
spectroscope
Mirror
Beam splitting
Broad-band Lens Polarizer polarization
source analyzer
Optical fiber
Aperture Beam
splitter
Tissue
sample
Figure 2.16 Schematic diagram of polarization LSS system.
Figure 2.17 Size distributions and refractive index distributions, N(d,m), of epithelial cell
nuclei obtained with LSS for the (a) normal and (b) cancerous colon tissue samples.
(hyperchromaticity). This suggests a higher concentration of nuclear solids

such as DNA, RNA, and proteins within the cancerous nuclei. Studies have
shown that the refractive index of a cell organelle increases linearly with the
concentration of its solid components. Therefore, a higher value of the
refractive index obtained for the cancerous nuclei is indicative of nuclear
hyperchromaticity.
80 Chapter 2
The third method of diffusive background removal is based on the

observation that this background is typically responsible for more than
95–98% of the total reflectance signal. Therefore, the diffusive background is
responsible for the coarse features of the reflectance spectra. The diffusion
approximation-based model discussed above may account for this component
by fitting to its coarse features. After the model fit is subtracted, the single
backscattering component becomes apparent and can be further analyzed to
obtain the nuclear size distribution.10 This method is simpler to implement,
because it does not require the use of polarized light, but is computationally
more intensive.
The promise of LSS to diagnose dysplasia and CIS were tested in in vivo
human studies in four different organs and in three different types of
epithelium: columnar epithelia of the colon and Barrett’s esophagus,
transitional epithelium of the urinary bladder, and stratified squamous
epithelium of the oral cavity.11 All clinical studies were performed during
routine endoscopic screening or surveillance procedures. In all of the studies,
an optical fiber probe delivered white light from a xenon arc lamp to the tissue
surface and collected the returned light. The probe tip was brought into gentle
contact with the tissue to be studied. Immediately after the measurement, a
biopsy was taken from the same tissue site. The biopsied tissue was prepared
and examined histologically by an experienced pathologist in the conventional
manner. The spectrum of the reflected light was analyzed and the nuclear size
distribution determined. The majority of the distributions of dysplastic cell
nuclei extended to a larger size.
These size distributions were then used to obtain the percentage of nuclei
larger than 10 microns, and the total number of nuclei per unit area
(population density). As noted above, these parameters quantitatively
characterize the degree of nuclear enlargement and crowding, respectively.
Figure 2.18 displays these LSS parameters in binary plots to show the degree
of correlation with histological diagnoses. In all four organs, there is a clear
distinction between dysplastic and nondysplastic epithelium. Both dysplasia
and CIS have a higher percentage of enlarged nuclei and, on average, a higher
population density, which can be used as the basis for spectroscopic tissue
diagnosis.
In these clinical studies, LSS has been restricted to sampling of millimeter-
size regions of tissue using a contact probe. To render this technology more
practical for clinical applications, one needs to extend its capabilities to
analyze wide areas of epithelial linings of the body. Recently, several novel
technologies have been developed that bridge spectroscopy and imaging,
i.e., spectroscopic imaging. In these imaging modalities each pixel or voxel of
the imaged object, surface, or volume is represented not by a single number, as
in conventional imaging, but by a linear array that is a spectrum of light
scattered elastically or inelastically or transmitted through each pixel or voxel.
300
300 (a) (b)
-2
-2
Population density, 10 mm
2
2
200 200
100 100
0 0
0 20 40 60 0 20 40 60
Percentage of large nuclei Percentage of large nuclei
400
(c) (d)
-2
-2
600
2
2
300
400
200
200
100
0 0
0 10 20 30 0 20 40 60
Percentage of large nuclei Percentage of large nuclei
Figure 2.18 Dysplasia/CIS classifications for four types of tissue obtained clinically with
LSS, compared with histologic diagnosis. In each case the ordinate indicates the percentage
of enlarged nuclei and the abscissa indicates the population density of the nuclei, which
parametrizes nuclear crowding. (a) Barrett’s esophagus: nondysplastic Barrett’s mucosa
(○), indefinite for dysplasia (▪), low-grade dysplasia (•), high-grade dysplasia (▴); (b) colon:
normal colonic mucosa (○), adenomatous polyp (▴); (c) urinary bladder: benign bladder
mucosa (○), transitional cell carcinoma in situ (▴); (d) oral cavity: normal (○), low-grade
dysplasia (•), squamous cell carcinoma in situ (▴).
Kidder et al. and Cabib et al. developed Fourier transform infrared (FTIR)
spectroscopic imaging to study the biochemical composition of tissues.53,54
Kidder et al. applied FTIR spectroscopic imaging to observe biochemical
modifications in brain tissue.55 Sowa et al. showed that spectroscopic imaging
can be used to study tissue perfusion.56 Farkas et al. developed a new
modality of spectroscopic imaging by combining it with analytical cytol-
ogy57,58 and applied spectral imaging for cancer detection and diagnosis.59 In
spectroscopic optical coherence tomography (OCT), the capabilities of OCT
were enhanced by combining conventional OCT with Fourier transform
spectroscopy.60
LSS was extended to allow imaging applications as well.61 This LSS-based
imaging allows mapping variations in the size of epithelial cell nuclei of living
82 Chapter 2
tissues over wide surface areas. The resulting images provide direct
quantitative measurements of nuclear enlargement and chromatin content,
which can be translated into clinical diagnoses. The technique can be used for
noninvasive or minimally invasive detection of precancerous changes in a
variety of organs, such as the colon and oral cavity.
In LSS imaging, a light source with a broad illumination spectrum is used
to illuminate the imaged tissue. The light from this source is collimated,
polarized, and transmitted through one of several narrow-band filters to select
the desired wavelengths covering the visible spectral range. A pair of equifocal
achromatic lenses separated by twice their focal length collects the light
backscattered by the sample. This so-called 4-f system ensures that the special
distribution of light in the plane distanced one focal length between the
collecting lenses depends on the angular distribution of light emerging from
the tissue. Therefore, an aperture positioned at the center of the lens system
determines the angular distribution of light scattered by the sample and
collected by the CCD, which is placed one focal length away from the outer
lens. The single scattering component is distinguished from the multiple
scattering component by means of polarization discrimination using an
analyzing polarizer, as in the polarization LSS. The CCD collects images for
each of the illumination wavelengths. After all the filters are used, each pixel is
represented by an LSS spectrum, which is analyzed using the Mie theory as in
the other LSS modalities.
LSS imaging was applied to study ex vivo colon tissue samples that were
obtained immediately after resection from patients undergoing colectomy
for familial adenomatous polyposis. Colonic adenomas are precancerous
dysplastic lesions exhibiting all of the characteristics of dysplastic lesions,
including cell nuclear enlargement, pleomorphism, and hyperchromasia.
The adenomas are surrounded by normal tissue covered by a single layer of
epithelial cells. For each pixel (25 mm 25 mm) of the imaged field
(1.3 cm 1.3 cm), the analysis of the LSS spectra enabled the size and
refractive index of the nuclei in each pixel to be obtained. Then the imaged
field was divided into 125 mm 125 mm regions and the percentage of nuclei
larger than 10 mm was obtained for each of these regions. As discussed
above, this statistic, which characterizes the degree of nuclear enlargement,
is highly significant for the diagnosis of dysplastic lesions in the colon and
several other organs. The resulting color-coded plot is shown in Fig. 2.19. As
expected, the nuclei are enlarged in the central, adenomatous region, but not
in the surrounding nondysplastic tissue.
These results demonstrate that LSS has the potential to provide a means
for detecting epithelial precancerous lesions and preinvasive cancers
throughout the body. LSS is advantageous compared to conventional
diagnostic techniques in that it can provide objective, quantitative results in
real time without the need for tissue removal. The first clinical application
Percentage of large
nuclei
0.5
40 - 50
mm 30 - 40
1 20 - 30
10 - 20
1.5
0 0.5 1 1.5
mm
Non-dysplastic
Adenom a mucosa
Figure 2.19 LSS images of colon tissue sample showing the spatial distribution of the
percentage of enlarged nuclei. The adenoma observed histologically is marked by an
ellipse6HHFRORUSODWHV
may be to guide random biopsies of previously undetectable, endoscopically

invisible lesions. This could lead to new diagnostic and imaging technologies
that would significantly improve the efficacy of cancer screening and
surveillance procedures.
2.6.3 Imaging of early cancer and precancerous lesions with an

endoscopic polarized scanning spectroscopy instrument
LSS-based detection of dysplasia in Barrett’s esophagus (BE) has been
demonstrated successfully using a simple proof-of-principle single-point
instrument.10,11 This instrument was capable of collecting data at randomly
selected sites, which were then biopsied. The data was processed off-line, and a
comparison with biopsy results was made at a later time. The high correlation
between spectroscopic results and pathology was sufficiently promising to
justify the development of the clinical device, which is described herein.
A recently developed62 clinical endoscopic polarized scanning spectros-
copy (EPSS) instrument is compatible with existing endoscopes (Fig. 2.20). It
scans large areas of the esophagus chosen by the physician, and has the
software and algorithms necessary to obtain quantitative, objective data
about tissue structure and composition, which can be translated into
diagnostic information in real time. This enables the physician to take
confirming biopsies at suspicious sites and minimize the number of biopsies
taken at non-dysplastic sites.
The instrument detects polarized light coming primarily from the epithelial
layer. Although principally using the polarization technique to extract
diagnostic information about dysplasia, the EPSS instrument also sums the
two polarizations to permit the use of diffuse reflectance spectroscopy, which
can also provide information about the early stages of adenocarcinoma.63
84 Chapter 2
Figure 2.20 Clinical EPSS instrument. The EPSS instrument is shown in the endoscopy
suite before the clinical procedure, with the scanning probe inserted into the working channel
of an endoscope. The insets show details of the scanning probe tip and the control box.
The EPSS instrument is a significant advance over the single-point fiber-

optic instrument in that: (1) it scans the esophagus and has the software and
algorithms necessary to obtain quantitative, objective data about tissue
structure and composition, which can be translated into diagnostic informa-
tion and guide biopsies in real time; (2) it employs collimated illumination and
collection optics, which enables the instrument to generate maps of epithelial
tissue not affected by the distance between the probe tip and the mucosal
surface, making it dramatically less sensitive to peristaltic motion; (3) it
incorporates both the polarization technique for removing the unwanted
background in the LSS signal, and single backscattering in the diffuse
reflectance spectroscopy signal; (4) it integrates the data analysis software
with the instrument in order to provide the physician with real time diagnostic
information; (5) it combines LSS information with diffuse reflectance
spectroscopy information measured by the same instrument, thereby
improving the diagnostic assessment capability.
The instrument makes use of commercially available gastroscopes and
video processors. A standard PC is adapted to control the system. Commercially
available spectrometers are also employed.
For use during endoscopy, the polarized scanning fiber optic probe is
inserted into the working channel of a standard gastroendoscope
(e.g., Olympus GIF-H180 used in the procedures reported below) and the
gastroenterologist introduces the endoscope through the mouth. Spectroscopy
of the entire Barrett’s segment is performed by scanning adjacent sections, 2 cm
in length, with the polarized scanning probe as follows. The endoscope tip is
positioned and the probe is extended 2 cm beyond the endoscope tip, placing it
at the distal boundary of a BE region chosen for examination. One complete
rotary scan of the esophageal wall is completed. The probe is withdrawn
linearly 2 mm back into the endoscope tip and another rotary scan is
completed. This is repeated for 10 rotary scans, so that an entire 2 cm length of
BE is scanned; then, the endoscope tip is withdrawn 2 cm and the next length of
BE is examined. The instrument collects 30 data points for each rotary scan
and performs ten steps during a linear scan (2 mm per step), collecting 300 data
points in 2 min for each 2-cm segment of BE. The scanning time can be reduced
to as little as 20 sec by utilizing a more efficient scanning mechanism.
Qiu et al.62 performed in vivo measurements using EPSS during 10 routine
clinical endoscopic procedures for patients with suspected dysplasia at the
Interventional Endoscopy Center (IEC) at Beth Israel Deaconess Medical
Center (BIDMC). Patients reporting to the IEC at BIDMC had undergone
initial screening at other institutions and were referred with confirmed BE and
suspicion of dysplasia. Protocols were reviewed and approved by the BIDMC
Institutional Review Board.
Patients reporting for routine screening of Barrett’s esophagus who had
consented to participate in the study were examined. The EPSS polarized fiber
optic probe was inserted into the working channel of the gastroendoscope and
the gastroenterologist introduced the endoscope through the mouth. The
EPSS instrument performed optical scanning of each complete, continuous
region of the luminal esophageal wall chosen for examination by the
gastroenterologist. Data from the optical scans were recorded for each linear
and angular position of the probe tip as parallel and perpendicular
polarization reflectance spectra, corrected for light source intensity and
lineshape. The backscattering spectrum at each individual spatial location was
extracted by subtracting perpendicular from parallel polarized reflectance
spectra. The backscattering spectra were then normalized to remove
amplitude variations due to peristalsis. The mean of the normalized spectra
was calculated. The difference from the mean for each site was calculated,
squared, and summed over all spectral points. A site was considered likely to
be dysplastic if this parameter was greater than 10% of the summed mean
squared. No data points are needed for calibration of this simple diagnostic
rule. This analysis is straightforward and can be done in real time. By
extracting the nuclear size distributions from the backscattering spectra for
each individual spatial location the researchers found that this simple rule is
86 Chapter 2
Figure 2.21 Nuclear size distributions for one high-grade dysplasia site and one non-
dysplastic site in BE of one of the patients. Dark (red and pink online) regions of the map
indicate areas suspicious for dysplasia based on nuclear size distributions extracted from
the backscattering spectra for each individual spatial location. Non-dysplastic BE sites had
nuclear size distributions centered about 5-6 mm diameters while sites marked as suspicious
for dysplasia have nuclear size distributions with a main peak centered from 9 to 15 mm. The
arrows indicate the specific locations on the esophageal surface for which the size
distributions are extracted from the polarized LSS data.
approximately equivalent to a contribution of greater than 25% from enlarged

nuclei over 10 microns in diameter (Fig. 2.21).
Two observations support the clinical feasibility of this method. First,
spectroscopic data collected during clinical procedures confirm that the
polarization technique is very effective in removing unwanted background
signals. Second, the issue of peristaltic motion is addressed in the EPSS
instrument. During a procedure, it is difficult to maintain a fixed distance
between the optical probe head and the esophageal surface due to peristaltic
motion and other factors. Therefore, an important feature of the EPSS
instrument is its ability to collect spectra of epithelial tissue that are not
affected by the orientation or distance of the distal probe tip to the mucosal
surface. This is achieved with collimated illumination and collection optics.
Analysis of parallel polarization spectra collected at ten BE locations during a
standard clinical procedure showed that although amplitudes of the spectra
differ from point to point, the spectral shape is practically unchanged and,
more importantly, the oscillatory structure containing diagnostically signifi-
cant information is intact.
During the initial stage of the project the researchers collected a total of
22,800 EPSS spectra in 10 clinical procedures, covering the entire scanned
regions of the esophagus. The capabilities of the clinical method were validated
by comparing EPSS data with the subsequent pathology at each location where
biopsies were taken. For the first two patients, pathology was reported per
quadrant not per biopsy. For the other patients, 95 biopsies were collected at
EPSS locations given by their distances from the mouthpiece of the endoscope
and their angles relative to the start of the EPSS scan. Pathological examination
revealed a total of 13 dysplastic sites out of which 9 were high-grade dysplasia
(HGD). The rest of the sites were diagnosed as non-dysplastic BE.
The diagnostic parameters for each EPSS location were extracted from the
backscattering spectra, i.e., the residuals of the parallel and perpendicular
spectral components collected by the EPSS instrument. The results were
presented in the form of pseudo-color maps. Double blind comparison of the
EPSS maps with the biopsy reports revealed 11 true positive (TP) sites, 3 false
positive (FP) sites, 80 true negative (TN) sites, and 1 false negative (FN) site.
Thus EPSS measurements are characterized by a sensitivity of 92% and
specificity of 96%.
Several of the BE patients enrolled in the study who underwent routine
endoscopy and biopsy with EPSS presented with pathologies that revealed no
dysplasia, and the patients were dismissed. However, in some of these patients
the EPSS scan indicated probable sites of focal dysplasia, which were located
in regions where biopsies had not been taken. One of the patients was recalled
and biopsies were taken at the three sites indicated by EPSS in addition to the
standard-of-care protocol. Pathology confirmed HGD in all three EPSS
directed biopsies and one more HGD at a point located between two EPSS
indicated sites (Fig. 2.22). The latter site, considered to be a false negative, is
Figure 2.22 Biopsies taken during the initial and follow-up endoscopy procedures for patient
A, overlaid on the EPSS map acquired during the initial procedure (left panel). Three follow-up
biopsies were guided by the EPSS map and pathology confirmed HGD for each (indicated at
360°). High-resolution endoscopic (HRE) image of a location with invisible HGD (right panel)
with narrow band imaging (NBI) enabled. Video capture was acquired in subject A at one of
the locations where invisible dysplasia was missed by visual examination by HRE with NBI,
but were located by EPSS, and later confirmed by pathology. The site is marked by an arrow.
Note that the site is visually indistinguishable from the surrounding nondysplastic BE tissue
6HHFRORUSODWHV
88 Chapter 2
very close to the sites indicated by EPSS and may arise from imperfect
correspondence of the actual biopsy site with an EPSS mapped site (a problem
that will be addressed in future instrument and algorithm development). The
patient now has been given radio frequency ablation (RFA) treatment.
These focal dysplasias were missed by standard-of-care procedures that
blindly biopsy a tiny fraction of esophageal tissue according to a prescribed
protocol, but they were caught and confirmed by the capability of EPSS to
examine the entire esophageal epithelium millimeter-by-millimeter and detect
dysplastic cells—enabling early treatment and in all likelihood saved patients'
esophagi, and perhaps their lives.
The frequency of dysplasia in the patient sample was consistent with that of
the pre-screened patient population referred to the BIDMC IEC for
confirmation and treatment, but was higher than would be expected in the
general BE patient population. In fact, the rarity of HGD detection in the
general population of BE patients underscores the importance of having a more
comprehensive and effective method for gastroesophageal cancer screening.
2.7 Confocal Light Absorption and Scattering Spectroscopic

Microscopy
Recently, a new type of microscopy that employs intrinsic optical properties
of tissue as a source of contrast has been developed.64 This technique, called
confocal light absorption and scattering spectroscopic (CLASS) microscopy,
combines LSS with confocal microscopy. In CLASS microscopy, light
scattering spectra are the source of the contrast. Another important aspect of
LSS is its ability to detect and characterize particles well beyond the
diffraction limit.
A schematic of the CLASS microscope is shown in Fig. 2.23. System
design of the CLASS microscope provided for broadband illumination with
either a Xe arc lamp for the measurements performed on extracted
organelles in suspension, or a supercontinuum laser (Fianium SC-450-2)
for the measurements performed on organelles in living cells. Both sources
used an optical fiber to deliver light to the sample. To insure that CLASS
microscopy detects organelles inside living cells and correctly identifies
them, Itzkan et al.64 complemented the CLASS instrument with a wide field
fluorescence microscopy arm, which shares a major part of the CLASS
optical train.
Depth sectioning characteristics of a CLASS microscope can be
determined by translating a mirror located near the focal point and aligned
normal to the optical axis of the objective using five wavelengths spanning the
principal spectral range of the instrument (Fig. 2.24). The half-width of the
detected signal is approximately 2 mm, which is close to the theoretical value
for the 30 mm pinhole and 36X objective used.65 In addition, the shapes of all
Figure 2.23 Schematic of the prototype CLASS/Fluorescence microscope.
1.00 500 nm
550 nm
600 nm
0.75 650 nm
700 nm
S, a.u.
0.50
0.25
0.00
-6 -4 -2 0 2 4 6
z (μm)
Figure 2.24 Depth sectioning of CLASS microscope along vertical axis.
five spectra shown in Fig. 2.24 are almost identical (500 nm, 550 nm,
600 nm, 650 nm, and 700 nm), which demonstrates the excellent chromatic
characteristics of the instrument. Small maxima and minima on either side of
the main peak are due to diffraction from the pinhole. The asymmetry is due
to spherical aberration in the reflective objective.66
90 Chapter 2
Figure 2.25 Fluorescence image of the suspensions of carboxylate-modified 1.9 mm

diameter microspheres exhibiting red fluorescence (left side), the image reconstructed from
the CLASS data (middle) and the overlay of the images (right side) (a). Image of the mixture
of three sizes of fluorescent beads with sizes 0.5 mm, 1.1 mm, and 1.9 mm mixed in a ratio of
4:2:1 (left side), the image reconstructed from the CLASS data (middle) and the overlay of
the images (right side) (b). Image of live 16HBE14o- human bronchial epithelial cells with
lysosomes stained with lysosome-specific fluorescence dye (left side), the image
reconstructed from the CLASS data (middle), and the overlay of the images (right side) (c)
6HHFRORUSODWHV
Itzkan et al.64 tested the combined CLASS/Fluorescence instrument on
suspensions of carboxylate-modified Invitrogen microspheres, which exhibit
red fluorescence emission at a wavelength of 605 nm with excitation at
580 nm. The microspheres were effectively constrained to a single layer
geometry by two thin microscope slides coated with a refractive index
matching optical gel. Fig. 2.25(a) shows (from left to right) the fluorescence
image of the layer of 1.9-mm diameter microspheres, the image reconstructed
from the CLASS data, and the overlay of the images. Figure 2.25(b) shows a
mixture of three sizes of fluorescent beads with sizes of 0.5 mm, 1.1 mm, and
1.9 mm mixed in a ratio of 4:2:1. Note the misleading size information evident
in the conventional fluorescence images. A 0.5 mm microsphere that is either
close to the focal plane of the fluorescence microscope or carries a high load of
fluorescent label produces a spot that is significantly larger than the
microsphere’s actual size. The CLASS image [middle of Fig. 2.25(c)] on the
other hand does not make this error and correctly reconstructs the real size of
the microsphere. One also can see that prior fluorescence labeling does not
affect the determination of the objects with CLASS measurements.
To confirm the ability of CLASS to detect and identify specific organelles in
a live cell, Itzkan et al.64 performed simultaneous CLASS and fluorescence
imaging of live 16HBE14o- human bronchial epithelial cells, with the lysosomes
stained with a lysosome-specific fluorescent dye. The fluorescence image of the
bronchial epithelial cell, the CLASS reconstructed image of the lysosomes, and
the overlay of two images are provided in Fig. 2.25. The overall agreement is
very good. However, as expected, there is not always a precise, one-to-one
correspondence between organelles appearing in the CLASS image and the
fluorescence image. This is because the CLASS image comes from a single, well
defined confocal image plane within the cell, while the fluorescence image
comes from several focal “planes” within the cell throughout the thicker depth
of field produced by the conventional fluorescence microscope. Thus, in the
fluorescence image, Itzkan et al.64 observe the superposition of several focal
“planes,” and thus additional organelles above and below those in the single,
well-defined confocal image plane of the CLASS microscope.
Figure 2.26 shows the ability of CLASS microscopy to do time sequencing
on a single cell. The cell was incubated with DHA for 21 hours. The time
Figure 2.26 The time sequence of CLASS microscope reconstructed images of a single
cell. The cell was treated with DHA and incubated for 21 hours. The time indicated in each
image is the time elapsed after the cell was removed from the incubator6HHFRORUSODWHV
92 Chapter 2
indicated in each image is the time elapsed after the cell was removed from the
incubator. In this figure, the nucleus, which appears as the large blue organelle,
has its actual shape and density reconstructed from the CLASS spectra. The
remaining individual organelles reconstructed from the CLASS spectra are
represented simply as spheroids whose size, elongation, and color indicate
different organelles. The shape of the nucleus has changed dramatically by the
third hour and the nuclear density, indicated by color depth, has decreased with
time. The organelles almost completely vanish by 4 hours.
Since the CLASS microscopy requires no exogenous labels, thus avoiding
their potential interference with cell processes, it is applicable to viable cell
and tissue preparations, enabling the observation of cells and organelles
functioning at scales on the order of 100 nm. Applications for CLASS
microscopy in such diverse areas as prenatal diagnosis, in vitro fertilization
(IVF) or drug discovery are all linked by the potential of this technique to
nondestructively observe functional intracellular processes.
Human embryo development and quality, as well as response to
environmental factors, might be monitored progressively at all critical stages
using CLASS. Since the CLASS measurement is nondestructive and requires
no exogenous chemicals, a given embryo in vitro could be monitored over time
before implantation. These kinds of progression studies are not possible with
the techniques currently available.
An important part of the drug discovery process is to monitor changes in
organelle morphology in cells treated with compounds being screened for
therapeutic or toxic effects. Currently, imaging using numerous fluorescence
markers67 or electron microscopy using non-viable cell preparations68–70 are
being used to detect these changes. However, CLASS could be used to
monitor organelle responses in multiple cell lines in parallel, in real time, using
viable cells with no exogenous markers. Not only would CLASS enable more
rapid screening, but it would also provide results more likely to be predictive
of animal and ultimately human outcomes.
Acknowledgments
We would like to thank Irving Itzkan for encouragement and Olga Perelman
and Eugenia Fingerman for inspiration and patient help.
References
1. A.G. Yodh and B. Chance, “Spectroscopy and imaging with diffusing
light,” Physics Today, 48(3), 34–40 (1995).
2. J.R. Mourant, I. J. Bigio, and J. Boyer et al., “Spectroscopic diagnosis of
bladder cancer with elastic light scattering spectroscopy,” Lasers Surg.
Med., 17, 350–357 (1995).
3. S.L. Jacques, J.R. Roman, and K. Lee, “Imaging superficial tissues with
polarized light,” Lasers Surg. Med., 26, 119–129 (2000).
4. R.P. Rava, J.J. Baraga, and M.S. Feld, “Near-infrared Fourier-
transform Raman spectroscopy of human artery,” Spectrochemica Acta
A 47(3-4), 509–512 (1991).
5. M.S. Patterson, B.C. Wilson, and J.W. Feather et al., “The
measurement of dihematoporhyrin ether concentration in tissue by
reflectance spectrophotometry,” Photochem. Photobiol., 46(3), 337–343
(1987).
6. R.R. Anderson and J.A. Parish “The optics of human skin,” J. Invest.
Dermatol. 77, 13–19 (1981).
7. R.R. Alfano, A. Prahdan, G.C. Tang, and S.J. Wahl, “Optical
spectroscopic diagnosis of cancer and normal breast tissues,” J. Opt.
Soc. Amer. B 6(5), 1015–1023 (1989).
8. R.R. Richards-Kortum, R. Rava, and M. Fitzmaurice et al.,
“A one-layer model of laser-induced fluorescence for diagnosis of
disease in human-tissue— applications to atherosclerosis,” IEEE Trans.
Biomed. Eng. 36, 1222–1232 (1989).
9. K.T. Schomacker, J.K. Frisoli, and C.C. Compton et al., “Ultraviolet
laser-induced fluorescence of colonic tissue–basic biology and diagnostic
potential,” Lasers Surg. Med. 12(1), 63–78 (1992).
10. L.T. Perelman, V. Backman, and M. Wallace et al., “Observation of
periodic fine structure in reflectance from biological tissue: a new
technique for measuring nuclear size distribution,” Phys. Rev. Lett., 80,
627–630 (1998).
11. V. Backman, M. Wallace, and L.T. Perelman et al., “diagnosing cancers
using spectroscopy,” Nature, 405 (2000).
12. D.W. Fawcett, A Textbook of Histology, Charman & Hall, New York
(1994).
13. L.D. Hiatt, Color Atlas of Histology, Williams & Wilkins, Baltimore
(1994).
14. R.S. Cotran, S.L. Robbins, and V. Kumar, Robbins Pathological Basis of
Disease, W.B. Saunders Company, Philadelphia (1994).
15. B.J. Reid, R.C. Haggitt, and C.E. Rubin et al., “Observer variation in
the diagnosis of dysplasia in Barrett’s esophagus,” Hum. Pathol., 19,
166–178 (1988).
16. R.G. Newton, Scattering Theory of Waves and Particles, McGraw-Hill
Book Company, New York (1969).
17. J.D. Jackson, Classical Electrodynamics, John Wiley & Sons, New York
(1975).
18. H.C. van de Hulst, Light Scattering by Small Particles, Dover
Publications, New York (1957).
19. M. Kerker, The Scattering of Light, Academic Press, New York (1969).
94 Chapter 2
20. B.T. Draine and P.J. Flatau, “Discrete dipole approximation for
scattering calculations,” J Opt. Soc. America, A11, 1491–1499 (1994).
21. B. Beauvoit, T. Kitai, and B. Chance, “Contribution of the mitochon-
drial compartment to the optical properties of rat liver: a theoretical and
practical approach,” Biophys. J., 67, 2501–2510 (1994).
22. J. Beuthan, O. Milnet, and J. Helfmann et al., “The spatial variation of
the refractive index in biological cells,” Phys. Med. Biol., 41, 369–382
(1996).
23. P.M.A. Sloot, A.G. Hoekstra, and C.G. Figdor, “Osmotic response of
lymphocytes measured by means of forward light-scattering-theoretical
considerations,” Cytometry, 9, 636–641 (1988).
24. M. Hammer, D. Schweitzer, and B. Michel et al., “Single scattering by
red blood cells,” Appl. Opt., 37, 7410–7418 (1998).
25. J.M. Schmitt and G. Kumar, “Optical scattering properties of soft
tissue: a discrete particle model,” Appl. Opt. 37, 2788–2797 (1998).
26. A. Tafloe, Computational Electrodynamics: The Finite Difference Time
Domain Method, Artech, Boston (1995).
27. A. Dunn and R. Richards-Kortum, “Three-dimensional computation
of light scattering from cells,” IEEE J. Sel. Top. Quantum Electron., 2,
898–905 (1996).
28. Z. Liao, H. Wong, B. Yang, and Y. Yuan, “A transmitting boundary for
transient wave analysis,” Sci Sin Ser. A 27, 1063–1076 (1984).
29. A. Brunsting and F. Mullaney, “Differential light scattering from
spherical mammalian cells,” Biophys. J., 14, 439–453 (1974).
30. J.R. Mourant, J. P. Freyer, and A. H. Hielscher et al., “Mechanisms of
light scattering from biological cells relevant to noninvasive optical-
tissue diagnosis,” Appl. Opt., 37, 3586–3593 (1998).
31. J.V. Watson, Introduction to Flow Cytometry, Cambridge Univ. Press,
Cambridge (1991).
32. R. Drezek, A. Dunn, and R. Richards-Kortum, “Light scattering
from cells: finite-difference time-domain simulations and goniometric
measurements,” Appl. Opt., 38, 3651–3661 (1999).
33. J.M. Schmitt, “Simple photon diffusion analysis of the effects of multiple
scattering on pulse oximetry,” IEEE Trans. Biomed. Eng. 38, 1194–1203
(1991).
34. S. Takatani and J. Ling, “Optical oximetry sensors for whole blood and
tissue,” IEEE Eng. Med. Biol. 3, 347–357 (1994).
35. M.R. Arnfield, J. Tulip, and M.S. McPhee, “Optical propagation in tissue
with anisotropic scattering,” IEEE Trans Biomed Eng 35, 372–381 (1988).
36. A. Kienle, L. Lilge, and M.S. Patterson et al., “Spatially resolved
absolute diffuse reflectance measurements for noninvasive determination
of the optical scattering and absorption coefficients of biological tissue,”
Appl. Opt. 35, 2304–2314 (1996).
37. S. Takatani and M. Graham, “Theoretical analysis of diffuse reflectance

from a two-layer tissue model,” IEEE Trans. Biomed. Eng. 26, 656–664
(1979).
38. G. Zonios, L.T. Perelman, and V. Backman et al., “Diffuse reflectance
spectroscopy of human adenomatous colon polyps in vivo,” Appl. Opt.,
38, 6628–6637 (1999).
39. T.J. Farrell, M.S. Patterson, and B.C. Wilson, “A diffusion theory
model of spatially resolved, steady-state diffuse reflectance for the
non-invasive determination of tissue optical properties,” Med. Phys., 19,
879–888 (1992).
40. O.W. van Assendelft, “Spectrophotometry of Haemoglobin Derivatives,”
Springfield, Ill.: C. C. Thomas (1970).
41. V.P. Wallace, D. C. Crawford, and P.S. Mortimer et al., “Spectropho-
tometric assessment of pigmented skin lesions: methods and feature
selection for evaluation of diagnostic performance,” Phys. Med. Biol.,
45, 735–751 (2000).
42. M. Fitzmaurice, “Principles and pitfalls of diagnostic test development:
implications for spectroscopic tissue diagnosis,” J. Biomed. Opt. 5(2),
119–130 (2000).
43. J.B. Dawson et al., “A theoretical and experimental study of light
absorption and scattering by in vivo skin,” Phys. Med. Biol., 25, 696–709
(1980).
44. J.W. Feather, M. Hajizadeh-Saffar, G. Leslie, and J.B. Dawson,
“A portable scanning reflectance spectrophotometer using visible
wavelengths for the rapid measurement of skin pigments,” Phys. Med.
Biol., 34, 1301–1315 (1989).
45. R. Marchesini et al., “In vivo spectrophotometric evaluation of
neoplastic and non-neoplastic skin pigmented lesions. II: discriminant
analysis between nevus and melanoma,” Photochem. Photobiol., 55,
151–154 (1992).
46. J.R. Mourant, J. Boyer, and T. Johnson et al., “Detection of
gastrointestinal cancer by elastic scattering and absorption spectro-
scopies with the Los Alamos Optical Biopsy System,” Proc. SPIE
Vol. 2387, 210–217 (1995).
47. G.L. Tipoe and F.H. White, “Blood vessel morphometry in human
colorectal lesions,” Histol. Histopathol., 10, 589–596 (1995).
48. S.A. Skinner, G.M. Frydman, and P.E. O’Brien, “Microvascular
structure of benign and malignant tumors of the colon in humans,”
Digest. Dis. Sci., 40, 373–384 (1995).
49. V. Backman, R. Gurjar, and K. Badizadegan et al., “Polarized light
scattering spectroscopy for quantitative measurement of epithelial
cellular structures in situ,” IEEE J. Sel. Top. Quant. Elect., 5, 1019–
1026 (1999).
96 Chapter 2
50. C. Yang, L.T. Perelman, and A. Wax et al., “Feasibility of

field-based light scattering spectroscopy,” J. Biomed. Opt, 5, 138–143
(2000).
51. U. Morgner, W. Drexler, and F.X. Kartner et al., “Spectroscopic optical
coherence tomography,” Optics Lett., 25, 111–113 (2000).
52. D. Huang, E.A. Swanson, and C.P. Lin et al., “Optical coherence
tomography,” Science 254, 1178–1181 (1991).
53. L.H. Kidder, I.W. Levin, and E. Neil Lewis et al., “Mercury cadium
telluride focal-plane array detection for mid-infrared Fourier-transform
spectroscopic imaging,” Opt. Lett., 22, 742–744 (1997).
54. D. Cabib, R.A. Buckwald, Y. Garini, and D.G. Soenksen, “Spectrally
resolved Fourier transform spectroscopy (spectral imaging): a powerful
tool for quantitative analytical microscopy,” Proc. SPIE Vol. 2678 (1996).
55. L.H. Kidder, P. Colarusso, and S.A. Stewart et al., “Infrared
spectroscopic imaging of the biochemical modifications induced in the
cerebellum of the Niemann-Pick type C mouse,” J. Biomed. Opt., 4, 7–13
(1999).
56. M.G. Sowa, J.R. Payette, M.D. Hewko, and H.H. Mantsch, “Visible-
near infrared multispectral imaging of the rat dorsal skin flap,”
J. Biomed. Opt., 4, 474–481 (1999).
57. A.A. Pollice, C.A. Smith, and K. Brown et al., “Multiparameter analysis
of human epithelial tumor cell lines by laser scanning cytometry,”
Cytometry, 42, 347–356 (2000).
58. P. Yang, D.L. Farkas, and J.M. Kirkwood et al., “Macroscopic spectral
imaging and gene expression analysis of the early stages of melanoma,”
Mol. Med., 5, 785–794 (1999).
59. D.L. Farkas and D. Becker, “Applications of spectral imaging: detection
and analysis of human melanoma and its precursors,” Pigment Cell Res.,
14, 2–8 (2001).
60. U. Morgner, W. Drexler, and F.X. Kartner et al., “Spectroscopic optical
coherence tomography,” Opt. Lett., 25, 111–113 (2000).
61. R. Gurjar, V. Backman, J. Van Dam, and L.T. Perelman et al.,
“Significant breakthroughs in early cancer detection: Early cancer
detection with white light,” The American Institute of Physics Bulletin of
Physics News, 477, 1 (2000).
62. L. Qiu, D. Pleskow, R. Chuttani, E. Vitkin, J. Leyden, N. Ozden,
S. Itani, L. Guo, A. Sacks, J.D. Goldsmith, M.D. Modell, E.B. Hanlon,
I. Itzkan, and L.T. Perelman. “Multispectral Scanning during Endos-
copy Guides Biopsy of Dysplasia in Barrett’s Esophagus,” Nature
Medicine 16, 603–606 (2010).
63. I. Georgakoudi, B. C. Jacobson, J. Van Dam, V. Backman, M. B.
Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A.
Thomas, L. T. Perelman, and M. S. Feld, “Fluorescence, reflectance,
and light-scattering spectroscopy for evaluating dysplasia in patients

with Barrett's esophagus,” Gastroenterology, 120, 1620–1629 (2001).
64. I. Itzkan, L. Qiu, H. Fang, M.M. Zaman, E. Vitkin, L.C. Ghiran,
S. Salahuddin, M. Modell, C. Andersson, L.M. Kimerer, P.B. Cipolloni,
K.-H. Lim, S.D. Freedman, I. Bigio, B.P. Sachs, E.B. Hanlon, and L.T.
Perelman, “Confocal Light Absorption & Scattering Spectroscopic
(CLASS) Microscopy Monitors Organelles in Live Cells with no
Exogenous Labels,” Proc. Natl. Acad. Sci. USA, 104, 17255–17260
(2007).
65. T. Wilson and A.R. Carlini, “Size of the detector in confocal imaging
systems,” Opt. Lett., 12, 227–229 (1987).
66. B.A. Scalettar, J.R. Swedlow, J.W. Sedat, and D.A. Agard, “Dispersion,
aberration and deconvolution in multi-wavelength fluorescence images,”
Journal of Microscopy, 182, 50–60 (1996).
67. J.C. Yarrow, Y. Feng, Z.E. Perlman, T. Kirchhausen, and T.J.
Mitchison, “Phenotypic Screening of Small Molecule Libraries by
High Throughput Cell Imaging,” Combinatorial Chemistry & High
Throughput Screening, 6, 279–286 (2003).
68. M. El Mouedden, G. Laurent, M.P. Mingeot-Leclercq, and P.M.
Tulkens, “Gentamicin-induced apoptosis in renal cell lines and
embryonic rat fibroblasts,” Toxicol Sci, 56, 229–239 (2000).
69. F. Van Bambeke, C. Gerbaux, J.M. Michot, M.B. d’Yvoire, J. P.
Montenez, and P.M. Tulkens, “Lysosomal alterations induced in
cultured rat fibroblasts by long-term exposure to low concentrations of
azithromycin,” J Antimicrob Chemother, 42, 761–767 (1998).
70. S. Carryn, H. Chanteux, C. Seral, M.-P. Mingeot-Leclercq, F. Van
Bambeke, and P.M. Tulkens, “Intracellular pharmacodynamics of
antibiotics,” Infectious Disease Clinics of North America, 17, 615–634
(2003).
Lev T. Perelman joined the faculty of Harvard University in

2000 where he is currently a professor with appointments in
the departments of Medicine, Obgyn and Reproductive
Biology, and the Biological and Biomedical Sciences
program. He is also the Director of the Center for Advanced
Biomedical Imaging and Photonics at Beth Israel Deaconess
Medical Center. Prior to that he served for six years as a
Principal Scientist at MIT George R. Harrison Spectroscopy
Laboratory. Professor Perelman’s current research interests are primarily
focused on the application of optics to medicine and biology and include light
scattering spectroscopy, surface-enhanced Raman spectroscopy, nanopho-
tonics, optical imaging, and cancer detection with light.
98 Chapter 2
Vadim Backman, PhD, is the Walter Dill Scott Professor of

Biomedical Engineering at the McCormick School of
Engineering and Applied Sciences, Northwestern University
and Program Leader, Cancer and Physical Sciences, at the
Robert H. Lurie Comprehensive Cancer Center. He received
a PhD in Medical Engineering from Harvard University and
Massachusetts Institute of Technology. His research is
focused on bridging advances in biophotonics into biomedi-
cal research and clinical medicine. He develops novel optics technologies for
characterization and imaging of biological tissue with a focus on the
nanoscale and molecular levels. His research spans from cancer biophysics
to novel optical diagnostic and imaging techniques to multi-center clinical
trials. Dr. Backman has received numerous awards including being selected as
one of the top 100 young innovators in the world by the MIT Technology
Review Magazine and has served on multiple NIH and NSF review panels, as
well as chair of various scientific conferences such as OSA Biomed and SPIE
Biomedical Applications of Light Scattering. He has published more than 170
papers in peer-reviewed journals including Nature, Nature Medicine, PNAS,
and Physical Review Letters and holds over 20 patents. In the past three years,
he has served as the principal investigator on 20 grants from the National
Institutes of Health and National Science Foundation, including an NIH
Bioengineering Research Partnership. At Northwestern, he teaches advanced
classes in optics and human physiology. He is the co-founder of two biotech
companies.
Chapter 3
Reflectance and Fluorescence
Spectroscopy of Human
Skin in vivo
Yuri P. Sinichkin
Saratov National Research State University, Saratov, Russia
Tomsk National Research State University, Tomsk, Russia
Nikiforos Kollias
University of British Columbia, Vancouver, Canada
George I. Zonios
University of Ioannina, Ioannina, Greece
Sergei R. Utz
Saratov State Medical University, Saratov, Russia
Valery V. Tuchin
Institute of Precision Mechanics and Control, Russian Academy of Sciences,
Saratov, Russia
3.1 Introduction
Assessment of the optical properties of the skin is very helpful for the
quantification of the content and spatial distribution of the various biological
components in skin. It is also useful for the diagnosis of skin diseases,
investigation of the impact of different environmental factors (chemical agents,
drugs, UV-radiation, temperature, etc), and evaluation of the effectiveness of
skin treatments.
Currently, reflectance and fluorescence spectroscopy are probably the most
developed among the available optical methods for investigation of skin in vivo.
99
100 Chapter 3
Reflectance and fluorescence from skin carry information about the structures of
the epidermis and dermis, the quantity and density of blood vessels, the
concentration and spatial distribution of chromophores and fluorophores in
skin, and the nature of skin metabolic processes.
The latest improvements in fiber optics, electronics, and computer
technologies have made reflectance spectroscopy a common and popular
technique for skin analysis. Typical applications include in vivo quantitative
analysis of skin erythema and pigmentation, determination of cutaneous color
variation, monitoring of dermatological treatment effects, and study of skin
biophysics.1–28
Fluorescence spectroscopy has also benefited from recent technological ad-
vances that made available new light sources, supersensitive optical multichannel
analyzers, and charge-coupled device (CCD) or complementary metal-oxide
semiconductor (CMOS) detectors with high temporal and spatial resolution.
A number of particular applications of fluorescence spectroscopy has
already been identified as being very useful in the study of skin. These
include ultraviolet A (UVA) light excited fluorescence,29–35 the use of
fluorescence for diagnostic purposes, determination of skin photoaging,29,36,37
determination of the level of primary melanin deposits,38 assessment of skin
erythema and pigmentation,34,35 and diagnosis of skin tumors.30,31,39–41
Methods of skin imaging for histological purposes, based on skin
autofluorescence (AF), are also being intensively developed.30,42,43
The absorption and scattering properties of skin affect both the AF and the
reflectance spectra. Therefore, the combined use of fluorescence and reflectance
may provide additional information for the analysis of skin tissue biophysics.
In this chapter, we discuss the potential advantages of the combined use of
reflectance and fluorescence spectroscopy of skin for the evaluation of erythema
and pigmentation indices, determination of total hemoglobin and its oxygen-
ation, and investigation of the efficacy of topically applied sunscreens. Skin
reflectance and fluorescence spectra alterations, caused by morphological and
functional changes in skin with aging, disease development, or therapy, can be
adequately analyzed with the help of simple skin models. Such an approach often
leads to new diagnostic methods that utilize skin optical imaging techniques and
color measurements often combined with AF and polarization-sensitive
techniques for the examination of skin. Improvement of skin diagnostic accuracy
and therapy depends largely on the analysis of the skin optical properties.
3.2 Human-Skin Back Reflectance and Autofluorescence

Spectra Formation
3.2.1 Diffuse reflectance spectrum
Human skin is an inhomogeneous absorbing medium with strong scattering
properties. Light interaction with such a medium has a complicated
Reflectance and Fluorescence Spectroscopy of Human Skin in vivo 101
character.1,44 The horny skin layer (stratum corneum) has an average

refractive index higher than that of air. This layer is responsible for the
specular reflection of light on the skin/air interface (5–7% of the total incident
light). A significant part (93–95%) of the light is scattered and absorbed by the
remaining layers of the skin, i.e., the epidermis, dermis, basal lamina, blood
vessels, etc.2,44–47
Absorption is only one mode by which light can interact with skin.
Absorption of ultraviolet (UV) and visible light in skin is due to electronic
excitation of aromatic or conjugated unsaturated chromophores. There are
many kinds of chromophores in skin, but a few major chromophores
predominantly determine the optical absorption within each skin layer.
Absorption spectra of the main skin chromophores are shown in Fig. 3.1.
Proteins found in the epidermis contain the aromatic amino acids trypto-
phan and tyrosine, which have a characteristic absorption band near 270–
280 nm; urocanic acid and the nucleic acids also contribute to this absorption
band, with a maximum near 260 to 270 nm (Fig. 3.2). Epidermal melanin plays
an important role in limiting the penetration depth of light in the skin: it
effectively absorbs at all wavelengths from 300 to 1000 nm, but the strongest
absorption occurs at shorter wavelengths in the near-UV spectral range.48,49
Some of the major dermal chromophores are oxyhemoglobin, deoxyhe-
moglobin, bilirubin, carotenoids, and porphyrins. Both the oxygenated and
deoxygenated forms of hemoglobin absorb light. Oxyhemoglobin has its
strongest absorption band at 415 nm (the Soret band), and it has two
secondary absorption bands at 542 and 577 nm. Deoxyhemoglobin has its
Figure 3.1 Spectral absorption of the main skin chromophores. For chromophores marked
with (*), the range indicated is a half-width of the band. FAD: flavin adenine dinucleotide;
NADH: reduced form of nicotinamide adenine dinucleotide; NAD: nicotinamide adenine
dinucleotide; DNA: deoxyribonucleic acid.
102 Chapter 3
Figure 3.2 UV absorption spectra of major chromophores of human skin [dioxyphenyla-

lanine (DOPA)-melanin, 1.5 mg % in H2O; urocanic acid, 104 M in H2O; DNA, calf thymus,
10 mg % in H2O (pH 4.5); tryptophan, 2 104 M (pH 7); tyrosine, 2 104 M (pH 7).44
Figure 3.3 Molar attenuation spectra for solutions of major visible light-absorbing human
skin pigments. (1) DOPA-melanin (H2O); (2) oxyhemoglobin (H2O); (3) hemoglobin (H2O);
(4) bilirubin (CHCl3).44
primary absorption band at 430 nm, and it has a single secondary absorption
band at 555 nm. Both hemoglobins exhibit the lowest absorption at
wavelengths longer than 620 nm.2,6,44,46,50 Bilirubin has two relatively broad
absorption bands near 330 and 460 nm (Fig. 3.3).44
In the infrared (IR) spectral range, the skin absorption spectrum is
essentially determined by the absorption of water.46
In addition to absorption, skin is also characterized by strong light
scattering properties.2,44–47,51 The scattering results from inhomogeneities in
the refractive index that correspond to microstructural inhomogeneities, and

depends on the size and shape of the scatters and the wavelength of the light.
Typical scatterers found in skin have dimensions in the range of 0.05 to 10 mm;
i.e., they can be larger, comparable, or smaller than the wavelength of light.
As a result, different types of scattering (ranging from Rayleigh to Mie
scattering) take place within the skin. Structures with dimensions greatly
exceeding the light wavelength in epidermis (keratohyalin granules) and in
dermis (collagen fibers) are responsible for highly forward-directed scattering.
Because epidermis is not as thick as dermis, scattering in the epidermis is less
important than dermal scattering when determining the penetration of optical
radiation in skin. Dermal connective tissue is practically entirely responsible
for the majority of light scattering that takes place in the skin, and it also
determines the diffuse pattern of light distribution within the skin and the
formation of the backscattered diffuse reflectance. Scattering is generally
stronger in the UV spectral range, but strong epidermal absorption is the
important factor responsible for the reduction of the backscattered light and
generation of the skin reflectance spectrum.44 Thus, absorption and scattering
determine the amount of light emerging from the skin surface, which is closely
related to the diffuse reflectance coefficient Rd.
There are two main factors influencing the intensity and spectral features
of the skin reflectance Rd. The relative contribution of different skin layers is
one factor. Since the pigmented epidermis and the dermis that contains blood
vessels carry the main chromophores, absorption by these chromophores
limits the penetration depth and, therefore, their contribution to the diffuse
reflectance spectrum.
In the UV spectral range (,300 nm), Rd is generally very small due to
strong epidermal absorption that reduces the amount of backscattered light
to the same levels as Fresnel’s reflectance. The penetration depth of
optical radiation within the epidermis does not exceed a few cell layers, and
epidermal chromophores have a small effect on the diffuse reflectance
spectrum.
In the UVA spectral range (320 to 400 nm), skin reflectance exceeds
Fresnel’s reflectance, which indicates an increase in backscattered radiation.
The penetration depth of optical radiation increases up to hundreds of
micrometers, and epidermal chromophores (Fig. 3.1) affect the shape of the
reflectance spectrum.
In the visible spectral range (400 to 700 nm), the penetration depth is
between 0.5 to 2.5 mm.46 In this case, both absorption and scattering play a
dominant role in the formation of the diffuse reflectance spectrum. The
fraction of backscattered light increases due to multiple scattering within the
skin. The value of Rd is between 15 to 70%, and the reflectance spectrum has a
sharp minimum in the spectral range of 415 to 430 nm, due to hemoglobin
absorption in the dermis.
104 Chapter 3
In the spectral range of 600 to 1500 nm absorption is even lower,

scattering dominates absorption, and penetration depth can be up to 8 to
10 mm.45,51,52 Light within the skin is entirely diffuse, thus the diffuse
reflectance increases and the value of Rd rises to 35 to 70%. The reflectance
spectrum has characteristic dips in the 540 to 580 nm spectral range that are
due to the secondary absorption bands of hemoglobin. Additional weak
minima in the reflectance may be noted due to the absorption of carotene
(480 nm) and bilirubin (460 nm).
In the near-IR (NIR) spectral range, skin reflectance increases up until
800–900 nm and then decreases due to increasing water absorption.44,46
Representative diffuse reflectance spectra of human skin are shown in
Fig. 3.4.
The experimental setup used is another factor influencing the intensity and
spectral features of the diffuse reflectance Rd. In vivo, the most accurate
reflectance measurements may be made using an integrating sphere technique.
However, the size and localization of many dermal sites under investigation
limit its use. Usually, a fiber-optic-based system is used for in vivo skin
reflectance measurements. Experimental setups employing fiber optics are
usually of two types. Some instruments consist of Y-shaped flexible fiber optic
guides that have two legs. The first may consist of a single fiber with a large
core diameter that is centrally placed and delivers the excitation light to the skin
surface. The second may have several fibers with a smaller diameter
surrounding the central fiber, collecting the light scattered back by the skin.53–55
Measurements depend on the geometry of the optical fibers used, and
there are some issues that must be taken into consideration: (1) accurate
Figure 3.4 Typical diffuse reflectance spectrum of human skin.44

measurements require collection of backscattered light at all possible angles;

(2) the tips of the collection fibers must be arranged in such a way as to collect
the backscattered light from a surface area of the skin larger than the
excitation light spot. The latter is especially used in the NIR wavelength
range, because strong scattering and weak absorption in this range
significantly increase the escape area for backscattered light.56
Other experimental setups are capable of measuring the spatially resolved
diffuse reflectance, with the source and detector separated by a variable
distance r on the skin surface.57 In this type of measurement, the measured
reflectance is defined by the depth probed by the light. This depth is defined
by the photon path distribution function for the photon migration from a
source to a detector and depends on the distance rsd between the source and
detector. As a result, the shape of the reflectance spectrum is also defined by
the distance rsd.57
3.2.2 Autofluorescence spectra

Human skin contains various types of native fluorophores with unique
absorption and emission spectra, different fluorescence quantum efficiencies,
different fluorescence decay times, and different distributions within the skin.
Some fluorophores have similar absorption and fluorescence spectra, and
typically, fluorescence spectra measured on the skin surface are the result of
the overlapping bands of such fluorophores. Skin also contains non-
fluorescent chromophores such as hemoglobin and melanin. These chromo-
phores may absorb fluorescence emitted by other fluorophores, thus
introducing dips and peaks in the fluorescence spectra.
When the excitation wavelength is increased, new fluorophores are
involved in the formation of the fluorescence spectrum shape. The closer the
excitation wavelength to the center of the so-called therapeutic/diagnostic
window (600–900 nm), the larger the penetration depth of the excitation light
in tissue, and the larger the tissue volume probed by the excitation light. As a
result, new kinds of fluorophores located in deeper skin layers contribute to
the total tissue fluorescence measured.
The dependence of fluorescence on excitation and emission wavelength
can best be depicted with a 3D plot (Fig. 3.5). Measurements were performed
on skin samples of 20 20 mm2 size with subcutaneous fat obtained
interoperatively from five patients in the course of skin plastic surgery (three
samples from the abdomen and two from the lower extremities). A simple
inspection of the spectra presented leads to two basic observations: human
skin exhibits a rather characteristic autofluorescence (AF) pattern, and skin
AF intensity is subject to marked individual variations.
A 2D contour plot of a 3D skin AF pattern, usually referred to as a
fluorescence excitation-emission map (EEM) is shown in Fig. 3.6. One of the
goals of fluorescence spectroscopy is the identification of excitation
106 Chapter 3
Figure 3.5 3D plots of the human skin in vitro AF spectra at different excitation
wavelengths: (a) 40 year old man, (b) and (c) 60 and 87 year old women, accordingly.61
wavelengths suitable for differentiation of various pathological conditions.

This is closely related to the identification of the chromophores responsible for
this differentiation. Most of the biological components that are either related
to skin tissue structure or are involved in metabolic and functional processes
generate fluorescence emission in the UV-visible spectral region. As a result,
different morphofunctional conditions of the skin related to histological,
Figure 3.6 The excitation-emission maps (EEMs) of in vivo skin AF emission.59
Figure 3.7 Spectral ranges of fluorescence of the main skin chromophores. For
chromophores marked with (*), the range indicated is a half-width of the band.
biochemical, and physiochemical alterations can be characterized, in

principle, on the basis of information available in fluorescence EEMs.31,58–60
In Figs. 3.5 and 3.6, characteristic spectral features and patterns can be easily
identified.
The fluorescence maximum in the 320–370 nm range with a peak at
340 nm arises with excitation in the 250–290 nm range (peak at 280 nm).
Based on the information available in Fig. 3.7, where the spectral ranges of
fluorescence of the main skin chromophores are shown,29,44,62–64 it can be seen
that the skin AF in the UVA range is dominated by the fluorescence bands of
aromatic amino acids, namely tyrosine and tryptophan. There is only a slight
108 Chapter 3
variation in UVA fluorescence between different skin sites. This may be

attributed to the absence of AF attenuation by melanin, which is deposited
mainly within the epidermis. Tyrosine and tryptophan content in epidermis is
more than twice that of the whole skin, and this is why epidermis has a high
AF in the UVA range. This also explains why the fluorescence of psoriatic
stratum corneum is significantly higher than that of normal stratum
corneum.61
Investigation of the nature of skin autofluorescence in the 400–480 nm
range (with maximum near 460 nm) is more difficult. In this case, the
excitation peak (near 360 nm) as well as the emission peaks are constant.
Among the endogenous skin fluorophores investigated, two of the most
promising are the different forms of nicotinamide adenine dinucleotide
(NAD)39,65 and keratin40 located in epidermis and dermal collagen.29 The
reduced (NADH) and oxidized (NAD þ ) forms of NAD take part in cellular
metabolism, and the intensity of their specific fluorescence (fluorescence
maxima near 460 nm and 435 nm, respectively) is used not only for
differential diagnostics of metabolism disfunction,66 but also in quantitative
NADH detection.67 Sterenborg et al.40 reported a similarity between the AF
spectrum of human skin in vivo and the emission spectrum of keratin
(maximum near 450 nm) measured in vitro.
Collagen is one of the most important skin fluorophores. Approximately
75% of the dry weight of dermal tissue is composed of collagen fibers.68
Collagen is the main structural component of connective tissue and accounts
for about 90% of the protein in human dermis. There are at least five types of
collagen; types I (approximately 80%) and III (approximately 20%) are
found in dermal collagen, and type IV is found in the cellular basement
membrane.69 Collagen fibers exhibit a constant density throughout all
dermal layers.70
The observations regarding the central role of epidermal chromophores
(keratin, NADH) in the formation of the AF spectrum of human skin is based
on the fact that in vitro fluorescence spectra of keratin and NADH are very
similar to in vivo AF spectra of human skin.40
In the case of collagen and elastin, which are located predominantly
within the papillary and reticular layers of dermis, the situation is a bit
different. Here, both excitation and emission light is attenuated because of
absorption by melanin. In addition, fluorescence intensity in the 400 to
480 nm range is subject to attenuation by other skin chromophores:
hemoglobin, porphyrins, carotenoids, etc. (see Fig. 3.1). Both the total
intensity and the spectral features may be affected.33
Comparison of the fluorescence spectrum of collagen in vitro with the AF
spectrum of in vivo human skin, both measured using a fiber optic sensor
under identical conditions, revealed a difference in the peak emission
wavelengths.35 This may be due to the fact that the AF spectrum of dermal
Figure 3.8 Fluorescence spectrum of (1) collagen (in vitro), (2) AF spectrum of human skin
(in vivo), and the fluorescence spectrum of collagen after optical filtering by the dermal blood
plexus [thickness: (3) 50 mm and (4) 100 mm; blood content of 5%; oxygenation 50%;
modeling]. Excitation wavelength 337 nm.35
collagen is significantly affected by hemoglobin absorption. As shown in

Fig. 3.8, the AF spectra of human skin and collagen are essentially identical,
accounting for optical filtering by the dermal blood plexus (thickness of
50 mm or 100 mm; blood content of 5%).
The disturbance of collagen metabolism caused by aging leads to collagen
deposition within dermis.29,71 This provides an explanation for the observed
significant increase of the skin AF in elderly persons.
Experimental investigations of skin AF with different blood content35
confirm the assumption that collagen is the main fluorophore that contributes
to the AF of the skin. Figures 3.9 and 3.10 represent the temporal dynamics of
the AF skin spectra involved in the process of UVA-erythema formation; and
Figs. 3.11 and 3.12 present the AF spectra of skin with different degrees of
erythema formation at different pressures applied to the skin. The main
change observed is a significant decrease of AF intensity during erythema
formation and an increase of AF intensity that corresponds to an increase of
skin pressure.
Variations in blood content in the superficial dermal vascular plexus and
in melanin content in epidermis are the main reasons for the changes observed
in the skin AF shown in Figs. 3.9–3.12.35 In Fig. 3.12, the increase of AF
intensity as a function of pressure is due to a decrease in the blood content
within the dermis, and the difference in the fluorescence intensity between
normal skin and the skin with erythema at pressure values .105 Pa is due to
110 Chapter 3
Figure 3.9 3D plot of human skin AF after UVA irradiation with 4 minimal erythema
doses (MEDs).35
Figure 3.10 Temporal dependence of the AF intensity for in vivo skin with developing
erythema: (○) lFL = 460 nm; (•) lFL = 420 nm; and (□) lFL = 500 nm.35
melanin absorption. Desquamation of human skin results in a reduction of

melanin content, which explains the increase of AF intensity of compressed
skin before and after desquamation (Fig. 3.11).
NADH and keratin have certain contributions to skin AF. Moreover,
results of calculations72 show a 2 to 2.5-fold decrease in the fluorescence of
collagen on the surface of skin as compared to that of the dermal layers. Even
so, however, the fluorescence intensity of collagen is sufficiently strong so as
to dominate the skin AF spectra.
Skin AF excited in the visible region (450 to 500 nm) can be attributed to
fluorophores such as carotenoids and flavins.62,63 Skin AF in the red spectral
region is characterized by much lower intensity compared to the UVA and
Figure 3.11 In vivo human skin AF with UVA-erythema for different values of external
mechanical pressure: (a) 7 days after UV irradiation; (b) 14 days after irradiation; (□) AF of
normal skin; (□) 1.4 104 Pa; (○) 8.4 104 Pa; (•) 14 104 Pa; and (▴) erythema without
pressure.35
blue spectral ranges. In the red, excitation maximum is located near 400 nm,
and emission maximum is near 600 nm. In addition to this AF maximum,
there are several less pronounced peaks in the 580 to 640 nm range.
Fluorescence in this range is mostly due to porphyrins.30,31,41,73
112 Chapter 3
Figure 3.12 AF intensity of human skin (lFL = 460 nm) vs. external mechanical pressure:
(□) 7 days after UVA irradiation; (•) 14 days after irradiation. AF intensity of human skin
without erythema and compression (a dotted line) is marked as a reference value.35
3.3 Simple Optical Models of Human Skin

3.3.1 Simple skin model for reflectance analysis
A simple model for the analysis of skin reflectance is based on the
assumption that skin consists of three or four layers, where each layer
homogeneously transmits and scatters light.6,7,50,54 The relationship between
the simplified model (Fig. 3.13) and the anatomic structure of skin is
apparent. The stratum corneum (layer 1) is responsible for only 5 to 7% of
the reflectance (including the Fresnel’s reflectance) and allows most of the
Figure 3.13 Simplified model of the layered structure of the skin (Reproduced with
permission from Ref. 6).
light to scatter forward. In the epidermis (layer 2), light is strongly absorbed
by melanin and the non-absorbed part reaches the hemoglobin-reach
papillary dermis (layer 3). The remaining part of the light is then diffusely
reflected by dermal collagen (layer 4). The light that is diffusely scattered by
the dermis reaches the skin surface after passing twice through the
hemoglobin and melanin-rich layers.
Assuming that the reflectance due to the first three layers is significantly
lower than that of the fourth layer, i.e., Rd,1, Rd,2, Rd,3 ≪ Rd,4, the total skin
reflectance is given by the equation:
I
Rd ¼ ¼ T 21 T 22 T 23 Rd,4 , (3.1)
I0
which is based on the assumption that absorption by melanin and hemoglobin
are more prominent than scattering in the top three layers, therefore, light
transmitted through these layers Ti (i ¼ 1, 2, 3) can be described by the Beer-
Lambert law and T1 1.
For a turbid medium such as human skin, the reflectance Rd determines
the apparent optical density (OD) of the medium:2
OD ¼ log Rd : (3.2)
The optical density of the skin can be expressed as50,54
X
OD ¼ 2 log e εi d i logðRd,4 Þ, (3.3)
i
where εi is the extinction coefficient of the i’th layer with thickness di. Since
OD is defined as a sum, in vivo reflectance measurements allow for the in vivo
estimation of skin chromophore content (melanin and hemoglobin, located in
the second and third layers, respectively):
OD ¼ 2 log e½ðεext c1 d 2 Þmel þ ðεext c2 d 3 Þoxy þ ðεext c3 d 3 Þdeoxy logðR4 Þ, (3.4)
where εext is the extinction coefficient, ci the concentration, and di the optical
thickness of the i’th chromophore layer.
Figure 3.14 shows the temporal dependence of OD for in vivo skin with
developing erythema, and Fig. 3.15 shows the in vivo OD spectral distribution
for erythematous human skin for different values of external mechanical
pressure. Curves in Figs. 3.14 and 3.15 demonstrate the erythema and
pressure effect on skin absorption. In particular, Fig. 3.14 shows the increase
of blood content in the skin, which rises by 80% within the first day after UVA
irradiation.
The application of an external mechanical compression on the order
of 105 Pa on skin in vivo leads to a decrease of both of its scattering properties
and its absorption caused by a decrease of the hemoglobin of blood contained
114 Chapter 3
Figure 3.14 Temporal dependence of OD for in vivo skin with developing erythema;
(•) lFL = 460 nm; (▴) lFL = 575 nm; and (▪) lFL = 650 nm.54
Figure 3.15 The erythematous skin (three days after irradiation) OD spectral distribution
for different values of external mechanical pressure: (1) without pressure; (2) 5.6 104 Pa;
(3) 8.4 104 Pa; (4) 1.4 105 Pa.54
in the skin. This process is inertial and proceeds within a period of time of
about 3 to 4 min. After the removal of the compression, the recovery time of
the skin tissue ( 40 to 50 min) considerably exceeds the stabilization time of its
parameters after application of external mechanical compression (several
minutes). At the initial moment of time after the removal of the compression,
the fullness of blood vessels in the skin increase considerably (by a factor of
2.4 compared to the normal skin).74
When an external mechanical pressure is applied to human skin in vivo,
the influence of hemoglobin on the reflection spectra is effectively reduced and
after a short optical clearing time the carotenoid absorption becomes easily
discernable in a 460 to 500 nm spectral window and its optical density can be
calculated with high accuracy.75
3.3.2 Simple skin model for autofluorescence analysis

By analogy with the simple optical model for reflectance spectral analysis, a
corresponding simple model can be thought of for skin AF spectral analysis.
Skin tissue is again considered as consisting of distinct layers: epidermis,
papillary dermis, and the layers beneath the papillary dermis (Fig. 3.16). The
spatial distribution of the fluorophores contributing to human skin AF is
assumed to be uniform within the limits of the epidermal (NADH, keratin)
and dermal (collagen) layers. The excitation light and collagen fluorescence
are affected by melanin and hemoglobin content because of the optical
filtering effect.
The intensity IAF of skin AF is defined by the total fluorescence intensities
I1 and I4 arising from the epidermal (NADH, keratin) and dermal (collagen)
skin layers, respectively.
In the one-dimensional approximation, the intensity of escaping fluores-
cence I4 from the dermal layer can be presented as8,35,76,77
I 4 ðlF L Þ ¼ ∫z FðlEX ,zÞhðlF L ,lEX ÞTðlF L ,zÞdz, (3.5)
where lEX and lFL are the excitation and emission wavelengths, F(lEX, z)
is the fluence distribution of excitation light within the fourth layer, and
h(lFL, lEX) is the fluorescence quantum yield of dermal collagen.
Figure 3.16 Simplified skin model for AF analysis.35

116 Chapter 3
The transfer function T(lFL,z) describes the propagation of collagen

fluorescence light (optical filtering) and can be written as
X

TðlFL Þ ¼ exp εi ðlF L Þd i , (3.6)
i
where εi ðlF L Þ is the absorption of the i’th layer with thickness di at the
wavelength of fluorescence emission lFL.
The average value of the excitation of light intensity in the fourth layer
can be expressed as
X
I ðlEX Þ ¼ I 0 ðlEX Þ exp εi ðlEX Þd i , (3.7)
i
where I0(lEX) is the intensity of excitation light incident on the skin surface as
a collimated beam, and εi(lEX) is the extinction coefficient of the i’th layer at
the wavelength of the excitation light. Then the intensity of the escaping
fluorescence can be calculated as follows:
X

I 4 ðlF L Þ ¼ hðlF L ,lEX ÞI 0 ðlEX Þ exp ½εi ðlEX Þ þ εi ðlF L Þd i : (3.8)
i
Normalizing the fluorescence intensity I4 (lFL) to h(lFL,lEX) I0(lEX) gives
I 4 ðlFL Þ
R ¼ , (3.9)
hðlF L ,lEX ÞI 0 ðlEX Þ
and the quantity OD* can be introduced:
X
1
OD ¼ log ¼ log e f½εi ðlEX Þ þ εi ðlF L Þd i g : (3.10)
R i
The quantity OD* can be called the “apparent effective optical density”
because it is directly proportional to absorption. As compared to OD, changes
of OD* are affected by changes in the absorption properties of the skin at the
excitation and emission wavelengths, thus OD* provides information about
the absorption properties of skin (blood and melanin) at two different
wavelengths.
Since it is generally difficult to obtain absolute quantitative informa-
tion regarding the fluorophores from AF measurements, mainly because
of difficulties involved in the determination of the fluorescence quantum
yield h, fluorescence spectroscopy is usually employed in the analysis of
the relative fluorophore content. While reflectance spectroscopy, where
reflected light is normalized to a reflectance standard (such as BaSO4),
allows the absolute determination of OD, it only allows for the

measurement of relative changes in skin absorption. This is mainly due
to the unknown quantity h.
DOD* is defined by skin AF intensity changes only:
1
hI 0 hI 0 I
DOD ¼ OD2 OD1 ¼ log 2 log 1 ¼ log F2 L , (3.11)
I FL I FL I FL
where I 1F L and I 2F L are two different AF intensities.

Changes in OD* do not depend on hI0 , and hI0 has been chosen in such a
way that the R* and R values do not differ dramatically.
The epidermal fluorescence introduces a constant contribution to skin AF
and does not significantly alter the dependence of OD* on blood and melanin
content.
The OD* spectra for skin with erythema, obtained at different pressures
on the skin, are shown in Fig. 3.17 (AF spectra of skin with different degrees
of erythema formation at different pressures on the skin were shown in
Fig. 3.11). Figure 3.18 shows the pressure dependence of OD* for skin with
seven-day erythema. As pressure on the skin increases, blood content
decreases and OD* decreases as well. The difference between OD* values
for erythematous skin at 105 Pa and normal skin at the same pressure
(horizontal line in Fig. 3.18) is due to melanin content.
Figure 3.17 In vivo OD* spectral distribution for human skin with UVA-erythema (after
seven days of irradiation) for different values of external mechanical pressure: (▪) normal
skin; (▵) 14 104 Pa; (○) 2.8 104 Pa; and (•) erythema without pressure.35
118 Chapter 3
Figure 3.18 OD* (lFL = 460 nm) for erythematous human skin (after seven days of
UV irradiation) vs. external mechanical pressure; (dashed line) - normal skin under pressure
of 105 Pa.35
3.4 Combined Reflectance and Fluorescence Spectroscopy

Method for in vivo Skin Examination
Human skin back reflectance and AF spectra are affected by the same
tissue absorption and scattering properties. As a result, information
about the absorption properties of skin (in particular, hemoglobin and
melanin content) can be obtained from either reflectance or autofluor-
escence measurements.34,35,54 Diffuse reflectance and AF spectroscopy
complement each other and together provide additional information
about the tissue under study.78,79
3.4.1 Correction of the internal absorption effect in

fluorescence emission
Fluorescence diagnostic techniques are based on identification of the fluoro-
phores responsible for abnormal fluorescence signals, and on the detection of
abnormal distributions of endogenous fluorophores within the skin. For such
identification, AF spectra corrected for absorption and scattering tissue
properties with the use of reflectance spectra are very useful.80
In the case of human skin, the combined use of reflectance and
fluorescence is very useful and allows for an effective analysis of in vivo
pffiffiffiffiffiffi
autofluorescence.32,78,79,81,82 There are models suggesting that Rd or Rd may
be used as a first-order approximation, f, for correcting the fluorescence
spectrum for skin absorption and scattering effects.30,79,83 In this case, the
corrected (intrinsic) fluorescence spectrum is given by the ratio of the

measured spectrum and f.
3.4.2 Determination of melanin and erythema indices

By measuring the reflectance in a specific part of the spectrum, it is possible to
come up with parameters called “pigment indices.” These parameters are
useful in the estimation of cutaneous chromophore content, and in the
quantitative characterization of pathological tissue conditions. The skin
reflectance spectrum includes spectral regions in which the measured in vivo
skin reflectance is determined mainly by only a specific chromophore, thus
individual pigment indices can be calculated. Pigment indices have been
introduced for the description of melanin pigmentation, bilirubin pigmenta-
tion, and the degree of erythema.3,5,9,20,48,84–88
The apparent absorption spectrum of melanin is approximately
linear with a wavelength at wavelengths longer than 600 nm, and it has
a stronger dependence on wavelength at shorter wavelengths. It also
exhibits a maximum around 335 nm (as compared to amelanotic skin)
(Fig. 3.19).
Using diffuse reflectance spectra from skin, Andersen and Bjerring
proposed an algorithm for the calculation of melanin based on its absorption
in the UVA region (360 to 390 nm). Due to the strong absorption of melanin
in the UV range, it has been proposed that the slope of the in vivo reflectance
Figure 3.19 The apparent absorption spectrum of constitutive epidermal melanin

pigmentation (Used with permission from Melanin: Its Role in Human Photoprotection,
Vadenmar Publ. Co., Overlang Park, Kan.89)
120 Chapter 3
spectrum between 365 and 395 nm correlates with melanin content.7,90 The
melanin index M is defined as:
M ¼ 100ðOD365 OD395 Þ, (3.12)
where subscripts denote the wavelength in nanometers. However, hemoglobin
also absorbs in this region (rising slope of the Soret band) and should be taken
into account as well.
Similarly, Dwyer et al. estimated the fraction of the epidermis that
contains melanin using the reflectance values at 400 and 420 nm.91 Although
they suggested that this parameter when measured at the upper inner arm may
be used as a predictor of risk for skin cancer, it should be noted that skin
remittance at these wavelengths are determined to a great extent by the Soret
band of oxyhemoglobin.
Due to minimal influence of hemoglobin, the near-infrared spectral range
is an alternative region for determining the melanin index. Kollias and
Baqer4,48 reported that the slope of the skin OD in the spectral range of 620 to
720 nm is directly proportional to the melanin content in the epidermis.
Specifically, the melanin index is defined by the slope of OD(l) in the region
above 620 to 640 nm:4,6,48
M ¼ 100ðOD650 OD700 Þ, (3.13)
where the lines denote the mean value of OD measured near 650 and 700 nm.
Changes in hemoglobin content in the dermis (or changes in hemoglobin
oxygenation) lead to changes in absorption in the 535- to 585-nm spectral
range (oxygenated hemoglobin has a double-peak maximum at 542 nm and
577 nm, while deoxygenated hemoglobin has a single peak at 555 nm).
Therefore, skin reflectance in this spectral range can be used to obtain
information about blood content and hemoglobin oxygenation.
Methods for the determination of the erythema index, E, using in vivo
reflectance spectroscopy are also well developed. The most widely used
definition for the erythema index is given by the area under the spectral curve
OD(l) in the region of 510 to 610 nm:6
E ¼ 100½OD560 þ 1.5ðOD545 þ OD575 Þ 2.0ðOD510 þ OD610 Þ, (3.14)
where subscripts denote wavelengths in nanometers (Fig. 3.20).
Simplified methods for erythema index determination are based on
comparison of skin optical density in the green (560 nm) (high hemoglobin
absorption) and red (650 nm) (low hemoglobin absorption) spectral
ranges,87
E ¼ 100ðOD560 OD650 Þ: (3.15)
This simplified method has the disadvantage that it ignores the contribution
of melanin absorption in the green spectral range. The three-wavelength
Figure 3.20 (a) Reflectance spectrum R and (b) spectrum of optical density OD of
(1) erythematous and (2) normal human skin.88
method92 incorporates corrections of the erythema index for melanin absorp-

tion. Melanin and erythema indices are defined as follows:

OD2 OD3
M¼k , (3.16)
Dl23

M
E ¼ 100 OD1 OD2 Dl12 , (3.17)
k
where Dl12 ¼ l2 l1, Dl23 ¼ l3 l2, l1 ¼ 560 nm, l2 ¼ 650 nm, l3 ¼

710 nm, and k is a normalization factor.
122 Chapter 3
The major advantages of the skin diffuse reflectance spectra and “pigment
indices” are their versatility and specificity because of the high information
contained in the full spectrum. Furthermore, the skin diffuse reflectance
spectra properties can be presented as biologically relevant parameters
(melanin content, blood oxygenation, blood stasis, etc.).93
The specific features of diffuse reflectance from different areas of human
face skin for laser and non-laser sources of visible and NIR light have been
investigated94 to localize the closed-eye (eyelid) region. In the visible spectral
range, the eyelid skin surface can be discriminated by determining the melanin
index. Its values for the eyelid skin exceed those for other parts of the faces of
volunteers with skin types II and III.
The eyelid skin can also be differentiated from the other facial skin areas
by comparing the skin reflectance in the NIR at the wavelengths corre-
sponding to the presence and absence of water absorption bands. The skin
reflectance in the eyelid area measured at a wavelength corresponding to the
water absorption band and normalized to the reflectance at the wavelength at
which the water absorption is minimal, are much lower than those for the
other face regions. This can be explained by the light penetration into eyeball
tissue with a high water content.94
The skin AF spectra can also be used for determination of erythema and
melanin skin indices. Equations (3.9) and (3.10) express OD* of normal
(ODN*) and erythematous (ODE*) skin:

hI 0 hI
ODN ¼ log N ; ODE ¼ log E 0 , (3.18)
I FL I FL
where lFL ¼ 460 nm.
The erythema index, which is proportional to absorption of fluorescence
(i.e., blood content), can be calculated from the equation
N
I
E ≡ kðODE ODN Þ ¼ k log FEL , (3.19)
I FL
where the value of k is chosen in such a way that erythema index values
obtained with reflection and fluorescence methods do not differ dramatically.
The melanin index can be defined by a similar equation, but fluorescence
intensities I N E
F L and I F L are measured under an external mechanical pres-
sure on skin 10 Pa (bloodless skin).
5
3.4.3 Monitoring of hemoglobin oxygenation

Steady-state reflectance is the simplest among several available optical
techniques for the study of hemoglobin saturation in tissue. There are two
main ways of implementing the steady-state reflectance technique. There has
been significant progress by several research groups in using steady-state NIR
Figure 3.21 Near-infrared attenuation [log10] for (1) a 1-cm layer of oxyhemoglobin,
(2) deoxyhemoglobin, and (3) water; 210-mM hemoglobin concentration in water.98
light to monitor hemoglobin saturation and its changes in living organs57,95–98

(see also Chapter 9 in Vol. 1). In the “therapeutic/diagnostic window” (600 to
900 nm) of a tissue, reduced hemoglobin (Hb) and oxygenated hemoglobin
(HbO2) are the two major absorbing chromophores. These two forms of
hemoglobin have very well known absorption spectra (Fig. 3.21). In this
spectral range, the reduced scattering coefficient, m0s , is weakly dependent on
the wavelength. These facts allow for evaluation of hemoglobin oxygen
saturation in tissue by measuring either the absolute absorption coefficient,ma ,
or the ratio of the absorption coefficient at two different wavelengths99 (dual-
wavelength oximetry technique).
The hemoglobin oxygen saturation Y is given by the individual concentra-
tions of deoxyhemoglobin, cdeoxy, and oxyhemoglobin, coxy, as follows93
coxy
Y¼ : (3.20)
coxy þ cdeoxy
The absorption coefficients at the two wavelengths l1 and l2 are
ma ðl1 Þ ¼ εl1 l2
deoxy cdeoxy þ εoxy coxy , (3.21)
ma ðl2 Þ ¼ εl2 l2
deoxy cdeoxy þ εoxy coxy , (3.22)
where ε is the extinction coefficient (cm–1M–1). Therefore, (3.20) becomes:

ma ðl1 Þ l2
εl1
deoxy ma ðl2 Þ εdeoxy
Y ¼ m ðl Þ : (3.23)
l2
ma ðl2 Þ ðεoxy
a 1
εl2 l1 l1
deoxy Þ ðεoxy εdeoxy Þ
124 Chapter 3
According to Eq. (3.21), the oxygen saturation Y can be calculated if the

extinction coefficients for the oxygenated and deoxygenated hemoglobin at
the selected wavelengths are known a priori.
Other researchers have used changes in the measurement of optical
density (OD) to evaluate changes in tissue oxygenation.95,100 The measured
diffuse reflectance Rd from the skin surface by a fiber-optic reflectance
spectrometer can be thought of as the “apparent” optical density:
ma L
OD ¼ logðaÞ, (3.24)
2.3
where L represents the pathlength and a is an unknown factor depending on
the efficiency of the backscattered light from skin, the geometry of the incident
beam, and the wavelength of light. As long as a and L are independent of the
chromophore concentration, this equation provides a simple relation for
calculating absolute concentrations from measured OD variations at two
wavelengths. As a result, this method provides only relative changes in
hemoglobin saturation and so far has not been successfully used to accurately
quantify absolute blood oxygenation.
Another technique used is the introduction of a differential pathlength
factor, DPF, to compensate for pathlength increase due to multiple light
scattering in tissue.101 In this method, spatially resolved optical density
measurements are related to the skin absorption coefficient, ma, via a simple
linear equation:
OD ¼ ma srsd þ G, (3.25)
where rsd is the source-detector separation and s is a DPF. G is determined

purely by the scattering coefficient, ms, and other geometrical factors. The
absorption coefficient, ma , is given by the sum of the absorption coefficients
for each chromophore, which, in turn, are determined by the absolute
concentration c and extinction coefficient ε for each chromophore.
If N chromophores contribute to the optical density (mainly, hemoglo-
bins, melanin, and water, see Fig. 3.21), then Eq. (3.25) can be re-written as
follows:
X
N
ODðlÞ ¼ a þ bl þ ci εi ðlÞ: (3.26)
i¼1
Measurements of OD at a minimum of N þ 3 wavelengths enable a, b and ci to

be determined using standard methods of linear regression.
According to another approach,96,102 spatially resolved steady-state
reflectance data can be used to measure the optical properties and blood
oxygenation in tissue, based on the slope of the OD measurements.
According to diffusion theory, for a given source-detector separation rsd

that is much larger than lt, where lt is the photon transport mean-free path,
(rsd 20 40lt), a linear dependence of log½r2sd Rd ðrsd Þ on the source-detector
separation can be obtained in the following form:

mef f 0 1
log½rsd Rd ðrsd ,rsd Þ ¼
2 0
r logðamt Þ þ log mef f þ 0 , (3.27)
2.3 sd rsd
where r0sd is the middle point of the chosen pminimum and maximum source-
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
detector separation, m0t ¼ ma þ m0s , mef f ¼ 3ma ðma þ m0s Þ, and a is a constant
independent of rsd. Equation (3.27) indicates that the slope of
log½r2sd Rd ðrsd ,r0sd Þ versus the source-detector separation, rsd, gives meff, which
is a function of the product of ma and m0t . In the 600- to 900-nm wavelength
range, the reduced scattering coefficients, m0s , exhibit a weak dependence on
the wavelength. Thus, in a dual-wavelength method, it follows that96
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
slopeðl1 Þ mef f ðl1 Þ ma ðl1 Þ
¼ : (3.28)
slopeðl2 Þ mef f ðl2 Þ ma ðl2 Þ
Equation (3.28) suggests that determination of the slopes from two-

wavelength steady-state reflectance measurements can result directly in the
blood oxygenation regardless of the initial light intensity. In practice, the
values of meff are calculated from the slope of the OD measurements, and OD
can be calculated on the basis of reflectance measured for a calibration sample
with known optical properties similar to tissue and the tissue sample under
study,
cal 0 cal
Rd ðrsd ,r0sd Þ mef f mcal
ef f mt mef f þ ð1∕r0sd Þ
OD ¼ log ¼ rsd þ log 0cal þ log
Rd ðrsd ,r0sd Þ 2.3 mt mef f þ ð1∕r0sd Þ
(3.29)
As hemoglobin changes from the deoxygenated to the oxygenated form,
the extinction coefficient decreases at 760 nm, but increases at 850 nm (see
Fig. 3.21). Based on the relation ma ¼ extinction coefficient [hemoglobin
concentration], a decrease/increase in extinction coefficient leads to a
decrease/increase in ma when the total hemoglobin concentration remains
constant. Consequently, a decrease/increase in ma will result in a decrease/
increase in meff if the scattering properties remain unchanged [see Eq. (3.28)].
Figure 3.22(a) shows changes in meff at 760 nm and 850 nm of a sample
solution during the process of blood oxygenation and deoxygenation. The
corresponding oxygen saturation values are plotted in Fig. 3.22(b).
In the visible spectral range Eq. (3.3) can be used to calculate skin
chromophore content (cidj), when εext is known a priori. In order to perform
an estimation of the amount of optical chromophore, the absorption
126 Chapter 3
Figure 3.22 (a) Changes in meff of the blood liquid model measured at 760 nm and 850 nm
during a process of deoxy ! oxy ! deoxy circle. (b) Corresponding changes in hemoglobin
saturation of the liquid model. The calibration sample is the initial deoxygenated blood-yeast
model solution (Used with permission; see Ref. 96).
characteristics of in vivo melanin, oxygenated and deoxygenated hemoglobin,

and the reflectance of collagen (Rd,4) must be measured.50 For the estimation
of c2d3 and c3d3 and oxygen saturation, OD must be measured at several
wavelengths, and multiple linear regression can be subsequently employed to
calculate the concentrations.
A similar method has been applied for the in vivo investigation of
UV-induced changes in oxy- and deoxyhemoglobin103 in inflamed human
skin.104,105 The effects of melanin absorption are calculated by fitting a
standard melanin absorption spectrum to the measured apparent absorption
spectrum in the range 620-720 nm.4 After subtracting the melanin contri-
bution, the concentration of HbO2 and Hb are calculated by a 3-point fit (560,
577 and 630 nm).
Feather et al.20,100 have developed indices for hemoglobin concentration
(H) and oxygen saturation (Y) based on the gradient of the OD spectrum at
selected wavelengths:

OD544 OD527.5 OD573 OD544
H ¼ 100 ; (3.30)
16.5 29

5.1 103 OD573 OD558.5 OD558.5 OD544
Y¼ þ 42, (3.31)
H 14.5 14.5
where ODl is the skin optical density at wavelength l nm. The above indices
can be clinically valuable for monitoring skin response to various
dermatological treatments.
In vivo measured hemoglobin indices for a white skin normally do not
exceed a value of 0.5. The hemoglobin index of skin as calculated from
Eq. (3.30) is principally a function of the hemoglobin content in the sub-
papillary plexus. Assuming that this function is linear and that the in vitro
model represents the in vivo optical properties of skin, then a conversion factor
from the hemoglobin index H to the approximate amount of hemoglobin per
unit area may be derived as follows:106
Hemoglobin content ¼ H 3.8 104 g cm2

or
Hemoglobin content ¼ H 2.3 108 mol cm2 : (3.32)
Previous studies20 demonstrate that oxygen saturation of hemoglobin

in vitro can be determined accurately by reflectance spectrophotometry and
that the hemoglobin index is insensitive to changes in oxygenation.
3.5 Color Perception of Human-Skin Back Reflectance and

Fluorescence Emission
The skin is unique in the sense that it is an organ highly accessible to direct
visual inspection. Visual inspection of cutaneous morphology is the mainstay
of clinical dermatology. A change in the skin color is one of the most
important clues to the severity of a skin lesion and the interpretation of skin
test reactions. Perception of skin color is influenced by many factors,
128 Chapter 3
including its structure and its constituent elements, the light source used for
illumination, and the detector (eyes, camera, film, PC monitor, and printer) by
which it is observed.107,108
Skin is characterized by its surface structure, such as scale, wrinkles, and
hair (both color and texture); the concentration and distribution of absorbers
in the stratum corneum, epidermis (melanin), and dermis (hemoglobin,
bilirubin, etc.); and the concentration and distribution of scatterers in the
stratum corneum (melanin dust), epidermis (melanosoms and melanin
granules, cellular structures), and dermis (collagen, erythrocytes, etc.). Light
transport in skin (including propagation, absorption, scattering, and
remittance of light at different wavelengths) is the main factor determining
skin color.107,109 An ideal light source would be one with similar spectral
characteristics as sunlight, and the ideal spectral sensitivity of the detector
would be that which is similar to the human eye. Moreover, for a complete
description of color as perceived by a human observer, the exact physiology of
the eye must be taken into account.109,110
Skin color change resulting from different skin diseases or local/systemic
drug administration has been one of the major subjects of clinical and
experimental skin chromametry. Chromametry is based on the interpretation
of color perception due to the reflection from skin under white-light
illumination.11,13,18,19,111 It also deals with interpretation of skin color under
conditions other than white-light illumination, such as human skin
autofluorescence.112
Besides color, the exact shape of the lesion plays an important role
in performing a diagnosis. Visualization, documentation, monitoring,
measurement, and classification of morphologic manifestations of various
cutaneous processes have attracted the use of digital imaging tech-
niques that are based on computer processing of skin images.102 Recently,
digital color imaging techniques have found various applications in
dermatology, such as melanoma screening, psoriasis and erythema
detection, color analysis of nevi, monitoring of wheal and flare reactions,
etc.25,26,113,114
3.5.1 Color analysis of reflectance and fluorescence spectra

Standard spectrophotometric methods for determining the color of an object
are based on measurements of the reflected light intensities at three different
wavelength ranges that correspond to the three color-sensitive bands of the
retina in the human eye. Three color matching parameters xðlÞ, yðlÞ, and
zðlÞ, which represent the spectral sensitivity of a standard observer, are
defined in the CIE1931(Yxy) color system.115 Three tristimulus parameters
X, Y, Z, can then be calculated for any reflecting object on the basis of the
spectral distribution of the light emitted by the source and the reflectance
spectrum of the sample.
The tristimulus values are defined as the following sums:
X
780 nm
X¼ Rd ðlÞxðlÞDl,
380 nm
X
780 nm
Y¼ Rd ðlÞyðlÞDl,
380 nm
X
780 nm
Z¼ Rd ðlÞzðlÞDl, (3.33)
380 nm
where Rd is the total skin diffuse reflectance. Y is a brightness expressed

as a percentage based on a perfect reflectance of 100% (the light re-
flected by a standard white etalon made, for example, from BaSO4). The
chromaticity coordinates x, y of the CIE xy Chromaticity Diagram are
defined as
X Y
x¼ , y¼ : (3.34)
X þY þZ X þY þZ
The x, y parameters are sufficient for characterizing the skin color under fixed
illumination conditions and given a standard observer. In order to obtain an
orthogonal coordinate system for color description, the CIE1976 (L*a*b*)
system was introduced, in which color is represented as a vector in a three-
dimensional space defined by the coordinates L*, a*, and b* that form a
Cartesian coordinate system116 (Fig. 3.23):
1∕3 1∕3
Y 1∕3 X Y
L ¼ 116 16; a ¼ 500 ;
Y0 X0 Y0
1∕3 1∕3
Y Z
b ¼ 200 : (3.35)
Y0 Z0
X0, Y0, Z0 are the nominally white object-color stimuli given by the CIE
standard C, Y0 ¼ 100. For the above equations to yield valid results, X/X0,
Y/Y0, and Z/Z0 must be greater than 0.008856. The coordinate L*
correlates with “psychometric lightness” (relative brightness or “gray
value”), while a* and b* are chromatic coordinates (a* varies from green
to red, and b* varies from blue to yellow), equivalent to the two other
quantities commonly used in color description, the “hue” and the color
saturation (vividness) or “chroma:”
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
b
Hue ¼ arctan , Chroma ¼ ða Þ2 þ ðb Þ2 : (3.36)
a
130 Chapter 3
Figure 3.23 L*a*b* color space (modified from Ref. 116).
The total color difference DE* between two color stimuli [(L, a, b)i and
(L, a, b)j] is calculated from the following equation:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
DE ij ¼ ðDLij Þ2 þ ðDaij Þ2 þ ðDbij Þ2 , (3.37)
where DLij ¼ Li Lj , Daij ¼ ai aj , and Dbij ¼ bi bj .
Erythema is often evaluated using the a* parameter, whereas pigmenta-
tion is evaluated by the values of L*, b*, or combinations of them,
e.g., DE.11,13,26,117–120
Although usually a* correlates closely with the erythema index of the
narrow-band instruments, L* and b* show weak correlations with the melanin
index.117,121 In particular, an increase in hemoglobin concentration can
decrease both values of L* and b* in the absence of any change in melanin
pigmentation,93 for example, during application of sub-diastolic pressure with
a pressure cuff.122 Moreover, a* values are influenced by melanin concentra-
tion.11,123,124 In UVA-induced persistent pigment darkening, the b* value was
found to initially decrease and later increase, as the yellow component of
newly generated melanin becomes prominent.119
The L*a*b* parameters provide a measure of the perception of skin color
and can, therefore, emulate how the dermatologist or the average person
perceives skin. On the other hand, such a system of parameters cannot be
expected to simulate an analytical tool that could explain the physiological
reason for the skin appearance.
Chardon et al. have proposed the use of a vector representation for the
UV-induced tanning reaction in the L*a*b* space.119,125 The authors showed
that in the three-dimensional L*a*b* space, all skin colors of subjects with fair
complexion fall within a “banana” shaped volume termed skin color volume.
Increases in skin pigmentation can be graphed as a shift on the L*–b* plane,
whereas skin reddening (erythema) is represented as a shift on the L*–a*
plane. In an attempt to quantify skin pigmentation, the “individual typology
angle (ITA)” or “alpha characteristic angle” has been proposed, defined as the
vector direction in the L*–b* plane:

L 50 180
I TA ¼ arctan , (3.38)
b p
where ITA is given in degrees. This parameter has been validated as an
expression of skin pigmentation by the analysis of diffuse reflectance
measurements. However, it has also been found that an increase in the local
concentration of deoxyhemoglobin (e.g., by application of a pressure cuff) has
a similar effect on ITA as an increase in melanin pigmentation and can,
therefore, visually simulate pigmentation.122 Therefore, although ITA may be
a measure of perceived skin pigmentation, it still lacks the information about
the molecular origin of skin color (e.g., whether the perception of ‘pigment’ is
due to an increase of melanin or deoxyhemoglobin concentration).93
An approach that has attracted attention in the last decade involves the
determination of color as a psychophysical parameter, by taking into account
the way it is perceived by the human eye and brain.13 Instruments for color
measurement based on this approach are called either chromaticity meters or
chroma meters, and they employ illumination based on a mixture of the three
basic primary colors. They provide a good description of the color variability
of different skin types,13,126 the color changes associated with erythema and
tanning,11,13 the colorimetric assessment of drug effects on the skin,18,19 the
effects of age and sun exposure,16 etc.
Another aspect associated with the color of human skin involves skin
fluorescence. UVA-induced AF emission in skin occurs in the visible
spectral range and hence may contribute to skin color. Changes in AF may
lead to changes in color perception, which can be expressed in terms of the
three color parameters: L*, chroma, and hue. These parameters, along with
the color difference DE, can be calculated using Eqs. (3.33)–(3.37) where
the AF spectrum IFL(l) is substituted in place of the diffuse reflectance
spectrum Rd(l).
Figure 3.24 shows color changes in fluorescence and reflectance of skin
with UVA-erythema and skin with external application of mechanical
pressure. The changes in color perception of both AF and reflected white
light are related to blood content in the dermis. In the case of reflected white
light, the hue decreases during erythema development and increases with
applied pressure to the skin. These results are expected because erythematous
skin has a reddish tinge and skin with pressure applied has a less reddish tinge
compared to normal skin. In contrast, changes in skin blood content only
132 Chapter 3
Figure 3.24 Colors of human skin AF radiation (3,4) and white light reflected from the skin
(1,2) with developed UVA-erythema (1,3) and the skin with applied mechanical pressure
(2,4) presented in L*a*b* color space.112
Figure 3.25 Changes of color parameters for human skin with developed UVA-erythema:
(1) autofluorescence, (2) reflectance.112
slightly influence the chroma and lightness L* parameters. As a result, the

color difference DE is defined mostly by the changes in hue.
In the case of skin AF, the behavior of the color parameters is different.
Figures 3.25 and 3.26 show that hue and chroma changes are not very
Figure 3.26 Changes of color parameters for human skin with applied mechanical
pressure: (1) autofluorescence, (2) reflectance.112
significant (chroma decreases in skin with low blood content) as opposed to

changes in lightness L*. In the case of skin with variable blood content, the
spectral distribution of AF essentially does change, and only the fluorescence
intensity depends on blood content. The color parameters in the case of AF
depend on fluorescence intensity, thus IFL was normalized to get L* for
reflected light with AF being identical for normal skin.
Although chromametry provides a good description of skin color, it
cannot provide information on the chromophores responsible for color. Only
detailed analysis of the reflectance and/or fluorescence spectra may yield such
information.
3.5.2 Color imaging

The appearance of skin is of vital importance to the field of medicine. During
the diagnosis of skin diseases, careful observation and assessment of the
appearance of the diseased area is always the first and most important step.
Recently, photo-diagnosis and phototherapy have become popular methods
for treating skin diseases. In these techniques, light is used to detect and treat
lesions in the skin. Such techniques are non-invasive, hence patients are not
subjected to pain and scars during the treatment. In order to increase the
precision of such systems, we need more precise models of the interaction of
light with dermal tissues.127
Recent developments in CCD (charge coupled devices) and CMOS
(complementary metal-oxide semiconductor devices) that are used in digital
134 Chapter 3
cameras have made possible high-resolution and accurate color image

acquisition. In this way, reflectance and fluorescence images of human skin
may be recorded just as they would have been perceived by the human eye in real
time. The color images thus obtained can be used for analysis of color “texture,”
as they contain color information for each pattern element forming the image.
There are many types of digital imaging instrumentation and methods.
Digital images can be obtained by an analog video camera whose signal can
be converted to a series of numbers by the analog-to-digital convertor in a
frame grabber of a computer board, or by direct acquisition via a digital
camera that can transfer a series of numbers representing the image to digital
storage media or to a computer.108 The equipment used to capture an image
defines the spatial resolution of the resulting digital image. Typically, real-
time color digital cameras have a spatial resolution about 1400 pixels in
horizontal and 1000 pixels in vertical.
Images may have a color resolution of 256 levels at each color band of
red, green, and blue (RGB), indicating that the brightness of each band has
256 levels (0 ¼ darkest; 255 ¼ brightest). Regions of interest can be selected
from the total pattern for R, G, and B bands separately or by an additive
mixture of the three basic colors.
Using the mean brightness values of white standard (WR,G,B) and those of
the skin (SR,G,B), the integrated reflectance (RR,G,B) and skin apparent optical
density (ODR,G,B) in each spectral band (R, G, and B) can be determined: 25
S R,G,B
RR,G,B ¼ ; (3.39)
W R,G,B

1
ODR,G,B ¼ 100 log : (3.40)
RR,G,B
Then, according to the simple skin reflectance model [Eqs. (3.3)–(3.4)], when
selecting a green (G) and a red (R) band, the following equations can be
written: 25

1
ODR 2εext,R d 2 c þ log
mel mel
, (3.41)
R4,R

1 1
ODG 144ODR 2εhemo d
ext,G 3 chemo
þ log 1.44 log :
R4,G R4,R
(3.42)
Equations (3.41) and (3.42) have been used to evaluate spatial alterations in
the distribution of melanin and hemoglobin content in skin.25
The green color component of the images was used for wheal and flare
areas’ calculations.113 The light intensity threshold of the green color was used
for discriminating between normal unaffected skin and skin with wheal or
flare or skin with intracutaneous PAF-acether injection. For each pixel, light
intensities above the threshold were manually set to maximum white, and light
intensities below the threshold were set to zero (black). This resulted in the
generation of 1-bit resolution images (black and white images) calculated for
the wheal and flare areas.
The psoriatic skin lesions were assessed from photographs using a
computer image analysis system, with green color normalized to the original
color image.114
The estimation of the skin color parameters in the CIE1976 (Lab*) system
can be obtained using a color-camera-based technique.128 For this purpose,
the R, G, B color coordinates of the images are converted into quasi-L*, a*,
b* parameters using equations similar to Eqs. (3.35), where X/X0 ≡ RR,
Y/Y0 ≡ RG, and Z/Z0 ≡ RB.
Barel et al. have studied skin color changes after application of a
commercial tanning milk using an imaging system (Visi-Chroma VC-100;
Biophotonics, Lessines, Belgium) that records a typical RBG image and can
convert it to several CIE standard scales including L*a*b* and Yxy.129 Image
analysis software is available that can convert images from the RGB space to
the L*a*b* space provided that the acquired images have been properly
calibrated.
Erythema and melanin content in port wine stain lesions have been
evaluated by a similar method of converting an RGB image to the L*a*b*
space, assuming that the a* index image represents the erythema distribution
map and the L* index image the inverse of the melanin distribution map.130
A new method for quantitative analysis of 2D distributions of erythema
and pigmentation in the human skin using the skin computer image was
reported in Ref. 131. The recorded images of the skin using two interference
filters at wavelengths of 650 nm and 700 nm give the possibility to form,
according to Eq. (3.16), a new image of skin, where the pigmentation index is
the visualization parameter. Similarly, the recorded images using three
interference filters at 560 nm, 650 nm, and 700 nm give the possibility to form,
according to Eq. (3.17), a new image of skin, where the erythema index is the
visualization parameter.
The contrast of an erythematous skin image, when the erythema index is
used as the visualization parameter, as well as the contrast of the skin image
with melanin pigmentation, when the pigmentation index is used as
visualization parameter, was three times larger than in the case of the color
image (R-, G-, or B-images).131
Fluorescence photography appears to be a useful tool in the evaluation of
topical acne therapy.42 It also allows monitoring of patient compliance, and a
comparative analysis of images. UV-excited fluorescence is useful as an
ancillary technique in the evaluation of epidermal melanin. The maximum of
136 Chapter 3
UV-induced fluorescence emission of collagen occurs in the visible spectrum,

centered at 420 nm. The in vivo absorption of melanin at 420 nm is two times
greater and at 360 nm is approximately five times greater than at 540 nm.
Thus, epidermal melanin detection with fluorescence photography is 10 times
more sensitive than that with visible light reflectance photography.43
3.6 Polarization Reflectance Spectroscopy

Despite the fact that the development of the method of reflectance
spectroscopy of biological tissues allowed one to elaborate a variety of
diagnostic methods and devices that found wide application in experimental
and clinical medicine, the capabilities of this method are far from being fully
employed. Probing of biological tissue by polarized radiation, followed by
measurement of the spectral composition of the co- and cross-polarized
components of radiation backscattered by biological tissue, allows one not
only to quantitatively estimate the chromophore composition of the medium,
but also estimate the depth at which a particular chromophore is located.
Analysis of the polarization characteristics of light scattered by probed
biological tissues allows one, in some cases, to obtain radically new results in
the studies of the morphological and functional state of a biological tissue,
which are one of the most important trends of modern medical diagnostics.132
When linearly polarized light is normally incident on the skin, the specular
reflection by the stratum corneum (5%) is polarized in the same direction as
the incident light. The component of the incident light that is not reflected by
the stratum corneum/air interface enters the skin; one part of it is reflected
from the subsurface tissues and retains its polarization, and the remaining
non-absorbed part propagates in the dermis.133,134 After several successive
scattering events in the dermis, light loses all memory of its initial polarization
state.134,135 Thus, the light exiting the skin as diffuse reflectance is randomly
polarized, i.e., all possible polarization orientations are equally represented in
the reflected spectrum. In this way, light reflected by the skin consists of two
components: one that maintains the orientation of polarization of the incident
beam and one that is randomly polarized. The polarization discrimination of
light diffusely reflected by biological tissue using separate detection of two
orthogonally polarized components allows one to extract the scattered-
radiation components stipulated by the scattering from either surface or in-
depth layers of a tissue. The direction of polarization of one of these
components (the co-polarized component) corresponds to the direction of
polarization of a linearly polarized probing light, while another component of
the scattered light with a polarization vector orthogonal to the probing-light
polarization vector (the cross-polarization component) is due mainly to the
diffuse scattering of light in the probed volume and carries information on the
in-depth layers of the studied object. Such an approach is defined as
polarization-sensitive spectroscopy of elastic scattering, or polarization

reflectance spectroscopy.
In this section, we present the results demonstrating the prospects for
using the method of polarization reflectance spectroscopy for diagnostics of
the state of skin tissue, in particular, for estimation of the depth and blood
content in skin blood vessels.136
For skin tissue, the average scattering transport length in the optical
range of the spectrum is estimated to be about 100 mm. Structures with
dimensions greatly exceeding the light wavelength in epidermis (kerato-
hyalin granules) and in dermis (collagen fibers) play a dominant role in the
scattering, and, as a result, the depolarization length for linearly polarized
radiation, i.e., the e-folding distance for the decrease in the initial degree of
polarization of partial components in the scattering medium, turns out to
be comparable with the scattering transport length, which exceeds the
thickness of the epidermis. On the other hand, the presence of efficient
chromophores in skin (melanin in epidermis and hemoglobin in dermis)
must lead to an increase in the degree of residual polarization of the
backscattered radiation in spectral ranges corresponding to the absorption
bands of the chromophores.137 Moreover, these chromophores are
predominantly located at different depths that can be estimated from the
presence of typical absorption bands of the mentioned chromophores in the
difference polarization spectra.
The setup for realization of this method included two optical fiber
bundles: one with a wideband polarization filter mounted at the output of the
bundle to transport a linearly polarized light to the object under study, and the
other with a polarization filter that can be positioned in parallel or
perpendicular to the incident radiation and is used to collect radiation
reflected by skin. The setup allows one to measure the spectra RII(l) and
R⊥(l) of diffusely reflected radiation for parallel and perpendicular
orientation of the polarization filters, respectively. From the measured
spectra RII(l) and R⊥(l), difference polarization spectra DR(l) and spectra
PrL ðlÞ of the degree of residual polarization are calculated using the following
equations:
DRðlÞ ¼ RI I ðlÞ R⊥ ðlÞ, (3.43)
RI I ðlÞ R⊥ ðlÞ
PrL ðlÞ ¼ : (3.44)
RI I ðlÞ þ R⊥ ðlÞ
To demonstrate this method we present diffuse reflectance spectra taken

from the skin surface on the inner side of a forearm of a volunteer in the
course of layer-by-layer removal of surface layers of epidermis using glue-
stripping technology, as well as from the skin sites with erythema induced by
138 Chapter 3
Figure 3.27 Spectra of the degree of residual polarization of linearly polarized probing
radiation diffusely reflected by human skin (type II according to Fitzpatrik) with erythema of
different grades (a): 1 corresponds to the erythema index E ¼ 157, 2 – E ¼ 223, 3 – E ¼ 249,
4 – E ¼ 275, and 5 – E ¼ 290. Spectra of the apparent optical density of human skin with
erythema of different grades (b): 1 corresponds to the erythema index E ¼ 137, 2 – E ¼ 157,
3 – E ¼ 213, 4 – E ¼ 249, and 5 – E ¼ 288.136
UV radiation. The degree of erythema was quantitatively estimated with the

help of an erythema-melanin meter.92
Figure 3.27(a) shows the spectral distributions of the residual polarization
degree PrL ðlÞ of radiation backscattered from skin with erythema of different
grades. It should be mentioned that the spectral dependences of the residual
Figure 3.28 Variation in the difference polarization spectra of skin with the layer-by-layer
removal of surface layers of epidermis: 1 corresponds to normal skin, 2 – to skin with the
stripping layers of thickness 40 mm, 3 – 50 mm, 4 – 60 mm, and 5 – 70 mm.136
polarization degree are similar to the spectra of the effective optical thickness
OD of skin, presented in Fig. 3.27(b), and are indicative of the presence of
melanin (increased polarization degree in the short wavelength part of the
spectrum) and blood (increased polarization degree in the spectral absorption
bands of hemoglobin) in skin. An increase in the erythema grade, which reflects
the fact of increase in the absorption properties of skin stipulated by an increase
in the blood concentration in papillar dermis, leads to an increase in the residual
degree of polarization within the absorption bands of hemoglobin.
Blood of the papillar dermis is almost not manifested in the difference
polarization spectrum of normal skin. However, due to thickening of the
epidermis as a result of its layer-by-layer removal, the polarization state of
light interacting with the papillar-dermis region is retained to a remarkable
degree. This is seen as the appearance of absorption bands of hemoglobin in
the difference polarization spectrum (Fig. 3.28). The intensity of hemoglobin
absorption bands in the difference polarization spectrum makes it possible to
estimate the thickness of the epidermal layer of skin, or, more exactly, the
depth of blood vessels in skin tissue. The thickness of the peeled-off epidermal
layer was estimated by the number of sequential retrievals (a layer of thickness
about 4 mm was peeled off for each stripping).
3.7 Polarization Imaging

The features of the formation of polarization characteristics of the back-
scattered radiation with the initial linear polarization allows for visualization
140 Chapter 3
within the scattering media, including biological tissues, by analyzing spatial

distributions of polarization characteristics (intensities of the co- and
cross-polarization components and the residual polarization degree) of the
backscattered radiation.
The simplicity of technical arrangement is one of the advantages of this
method. Imaging techniques based on polarization employ two polarization
filters: one linear polarizer is mounted directly on the incident light source; a
second, identical polarizer is placed in front of the camera lens and is free to
rotate with respect to the first polarizer. The second polarizer provides light
passing through the camera lens to be selectively oriented either parallel or
perpendicular to the direction of polarization of the illuminating light.
Polarization imaging of the skin can be used in two ways. First, by using
polarized light photography at parallel polarizers, it is possible to selectively
probe surface or subsurface features of the skin.133,138 Surface features can be
viewed without interference by underlying pigmented lesions or erythema,
and may be examined without interference due to light coming from the
subsurface skin features.
Second, with the crossed polarizers, the surface features (such as scale,
wrinkles, and pore opening) are eliminated from the image, which allows the
evaluation of pigment distribution and erythema. It has been shown that
polarized light photography, with crossed polarizers, is more sensitive than
standard flash photography.139 The visualization of erythema distribution as
well as the extent of inflammatory lesions is much better realized in polarized
photography,140 such that retrospective evaluation of photographs yields
results that have comparable sensitivity to clinical assessment.141 Polarized
light photography has been employed in the evaluation of UV reactions,
response of psoriatic lesions to treatment, extent of inflammation in acne,
photodamaged skin, and irritancy.140–142 The major advantages in polarized
light photography are that images emphasizing either surface or subsurface
skin features may be obtained without altering either the camera angle, angle
of illumination, or source of illumination. The quality of the images is such
that retrospective evaluation is possible and yields results that compare well
with clinical assessment.
Skin imaging can also be combined with measurements of the degree of
polarization.131,134 With two images received on the basis of the co-polarized
component, I||, and cross-polarized component, I⊥, measurements can be
algebraically combined for isotropic media to yield a residual polarization
degree image (PrL ):
I jj I ⊥
PrL ¼ : (3.45)
I jj þ I ⊥
The advantage of using residual polarization degree as a visualization

parameter is due to the fact that the numerator in Eq. (3.45) is sensitive only
Figure 3.29 A freckle. A polarization image removes the melanin from a freckle. Freckle
melanosomes do not appear to influence the image.116
to polarized light, and rejects randomly polarized diffuse reflectance. For

example, a freckle is a superficial pigmentation that acts like a surface
absorption filter that attenuates light passing in/out of the skin. Such a filter
will affect both I|| and I⊥ images to the same degree. Therefore, when the
residual polarization degree is used as an imaging parameter, the attenuation
factor due to melanin is common to all terms in Eq. (3.45) and cancels out. As
a result, freckle melanosomes do not appear to influence the image (Fig. 3.29).
On the other hand, pigmented nevi may present melanin pigment at some
depth within the skin. Melanin absorbers imbedded deeper in the scattering tissue
may violate the assumption that an optical filter formed by the superficial tissue
layers is equally affecting I|| and I⊥ images. In such a case, the calculated residual
polarization degree image may not perfectly eliminate the melanin, however, it is
enough to see apparent scatter from the nevus structure (Fig. 3.30).
An image of object in the scale of values of the residual polarization
degree PrL in the spectral band corresponding to absorption of certain
chromophores in biological tissue makes it possible to localize areas of
increased content of chromophores. Such areas correspond to increased-
brightness fragments of an image.
Figure 3.30 A benign pigmented nevus. A polarization image removes the melanin and
shows apparent scatter from the nevus structure.134
142 Chapter 3
Figure 3.31 Polarization images of a burn-injured site of skin surface: (a) the co-polarized
component of the diffusely reflected radiation; (b) the cross-polarization component; (c) the
image obtained by using the degree of residual polarization of diffusely reflected light as the
visualization parameter.131
Figure 3.31 presents monochromatic images of a burn-injured skin surface

obtained in the spectral band of hemoglobin absorption (in this case, an
interference light filter with a wavelength of about 550 nm is installed in front
of the video-camera objective) for polarizers with parallel [Fig. 3.31(a)] and
orthogonal [Fig. 3.31(b)] orientations, and shows an image [Fig. 3.31(c)]
where the visualization parameter is the residual polarization degree. It should
be mentioned that the image contrast of the polarization degree (approxi-
mately 0.49) considerably exceeds the contrast of other monochromatic
images (0.08–0.13), which indicates the fairly high efficiency of this method.
The depth of polarization degree imaging is typically 100 to 150 mm. The
quantitative measurement of epithelial cellular structures in situ provided
by polarized light scattering spectroscopy is described by Backman et al.143
(see also Chapter 2 in this volume).
A polarization enhanced multispectral imaging device gives the possibility
of real-time visualization of skin structures with the resolution down to tens of
microns over a wide-field of approximately 4 cm2 (Fig. 3.32).144 In particular,
Figure 3.32 Cross-polarized imaging at different wavelengths emphasizes different

features of human skin. Skin photograph (A). In vivo cross-polarized image at
440 nm (B). In vivo cross-polarized image at 570 nm (C). In vivo cross-polarized image at
690 nm (D).144
cross-polarized 440 nm images were used for noninvasive quantitative

evaluation of dermal collagen.
3.8 Sunscreen Evaluation using Reflectance and Fluorescence

Spectroscopy
The use of spectroscopic methods for studying the effects of topically applied
drugs is of great interest to dermatologists and cosmetologists. The main
advantages of the methods are that they are noninvasive, measurements are
quick, and they provide high sensitivity and reproducibility. The study of
reflectance allows for real-time assessment of drug penetration by direct
measurement of the characteristic absorption bands of drugs under study, or
via indirect evaluation of drug effects (erythema, immediate pigment
darkening, delayed pigmentation, phototoxic reactions due to topical or
systematic photosensitization, production and migration of cells mediated by
sunburn, etc.).13,50,81,128,145–150
144 Chapter 3
The reflectance of in vivo skin is relatively low in the UV range; and the
ability to measure sunscreen optical characteristics can be significantly
impaired by strong light absorption and scattering within the epidermis. In
addition, for in vivo phototesting of sunscreen formulations with a high sun
protection factor (SPF) in the UVA SPF(A), many (seven or more) hours of
UVA exposure are required.
An alternative method is the fluorescence technique based on the assess-
ment of changes in fluorescence intensity of skin labeled with exogenous,
highly fluorescent dyes (dansyl chloride, acridine orange, etc), when excited by
UVA radiation.151
The following equation is used for the calculation of the SPF(A):152
X
400 nm
ðCIEl E l Þ
320 nm
SPFðAÞ ¼ , (3.46)
X
400 nm
ðT l CIEl E l Þ
320 nm
where Tl is the sunscreen transmission at wavelength l; CIEl is the CIE
action spectrum value at wavelength l; and El is the spectral irradiance
of terrestrial midday midsummer sunlight for southern Europe at wave-
length l.153–155
Changes in dye fluorescence intensity when sunscreen is applied to the
skin allow for determination of the sunscreen transmission in the UVA range,
and therefore, determination of SPF(A) as well. Skin AF excited by UVA
light may also be used for this purpose.83 Assuming that collagen is
responsible for most of the AF emission of skin, the fluorescence intensity of
normal skin and skin with sunscreen applied can be expressed as
N,S N,S N,S
F L ðlF L Þ ¼ I 0 ðlEX ÞT 1 ðlEX ÞT 2 ðlEX ÞT 3 ðlEX Þ
I NS
hT 3N,S ðlF L ÞT N,S N,S
2 ðlF L ÞT 1 ðlF L Þ (3.47)
where I0(lEX) is the intensity of the excitation radiation incident on the skin
surface; T1, T2, and T3 are the transmittances of corresponding skin layers at
the excitation and emission wavelengths (see Section 3.3.2.); h is the quantum
efficiency of collagen fluorescence; superscript N denotes normal skin with no
sunscreen applied, and superscript S denotes skin with sunscreen.
Assuming that the diffusion of the sunscreen molecules takes place only
within the epidermis (layer 1), then T N2 ¼ T 2 and T 3 ¼ T 3 . In this case, the
S N S
decrease of fluorescence intensity of the skin with the applied sunscreen

relative to normal skin AF is
I SF L T S1 ðlEX Þ T S1 ðlFL Þ
¼ : (3.48)
IN 1 ðlEX Þ T 1 ðlF L Þ
TN N
FL
T N,S
1 ðlFL Þ is the epidermal transmittance measured in the visible spectral
range. It can be determined from the in vivo reflectance spectra of normal skin
and that of skin with sunscreen applied.
The reflection factors RN (for normal skin) and RS (for skin with sun-
screen applied) can be defined as (see Section 3.3.1.)
RN,S ðlF L Þ ¼ ðT 1N,S ðlFL ÞÞ2 ðT N,S 2 N,S 2

2 ðlFL ÞÞ ðT 3 ðlF L ÞÞ R, (3.49)
where R is the collagen reflection factor. Assuming that sunscreen

predominantly affects the optical properties of the first layer, we get the
following:
S
RS ðlFL Þ T 1 ðlF L Þ 2
¼ : (3.50)
RN ðlF L Þ 1 ðlFL Þ
TN
Combining Eqs. (3.48) and (3.50), we have
T S1 ðlEX Þ I S ðlFL Þ∕I N FL ðlF L Þ

TðlEX Þ ¼ ¼ pFLffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi: (3.51)
T 1 ðlEX Þ
N
RS ðlFL Þ∕RN ðlF L Þ
T is the reduction of the epidermal transmittance after application of the

sunscreen, and 1/T is the protection factor at the wavelength lEX.
This algorithm, based on an estimation of AF intensity and reflectance of
human skin before and after application of sunscreen, allows for an
assessment of wavelength dependence of SPF(A) through evaluation of T
(lEX) from Eq. (3.51). T(lEX) may be termed “in vivo transmittance of
sunscreen applied onto skin.” The accuracy of the protection factor
determination increases substantially when lFL is chosen in the range where
sunscreen fluorescence is negligible.
The in vivo transmittance spectra of various tested sunscreens, calculated
as described above, are shown in Fig. 3.33. Commercially available
formulations have been designated as a, b, and c, and are listed in Table 3.1.
This technique also enables estimation of SPF(A) over extended periods of
time after sunscreen application. Changes in the integrated transmittance T
over time, obtained from the skin treated with sunscreen, are presented in
Fig. 3.34. Here T is determined as
X
400 nm
T¼ TðlEX Þ, (3.52)
320 nm
and it is normalized with reference to skin with no sunscreen applied. For all
studied sunscreens, the maximum reduction of skin AF, i.e., the photoprotec-
tion maximum, was observed 1 h after application.
146 Chapter 3
Figure 3.33 In vivo transmittance spectra of tested sunscreens 20 min after application.
(1) Contralum Ultra; (2) pH5-Eucerin; (3) Ilrido Plus.83
Table 3.1 Tested sunscreens and calculated SPF(A)

Manufacturers’ Calculated SPF(A),
Product Trade name of sunscreen SPF(A) mean SD
a pH5-Eucerin (lotion) SPF ¼ 15 5.4 0.54
b Ilrido Plus (milk) SPF(A) ¼ 9 7.3 0.82
c Contralum Ultra (cream) SPF(A) ¼ 7 6.2 0.73
d Anthelum “S” (cream) SPF(A) ¼ 10 8.9 1.2
e Ecran Total (lotion) SPF(A) ¼ 15 11.2 2.1

The data about SPF(A) are not available.
Figure 3.34 AF changes followed up the application of tested sunscreens, lFL = 500 nm.
(1) Contralum Ultra; (2) pH5-Eucerin; (3) Ilrido Plus.83
3.9 Control of Skin Optical Properties

3.9.1 Introduction
Reflectance, absorption, scattering, and fluorescence in living tissues,
and particularly in skin, can be effectively controlled by various
methods.25,35,44,47,54,156–229 Sensitization of the skin is extensively used in
optical diagnosis, photodynamic therapy (PDT) of malignant neo-
plasm,164–167,169 UVA photochemotherapy44 of psoriasis and other
proliferative disorders, and selective photodestruction of individual
components of tissue.170
The main limitations for skin and subcutaneous tissue optical imaging and
spectroscopy are the low penetration depth of light and/or low image contrast,
because of high tissue scattering in skin. The following optical modalities
suffer strongly: Doppler and speckle blood flow monitoring, optical coherence
tomography (OCT), confocal microscopy, second harmonic generation
(SHG) imaging, multi-photon spectroscopy, polarization imaging, and
Raman spectroscopy. Thus, optical clearing due to the reduction of the
scattering ability of tissue components is an attractive method for controlling
the optical properties of many tissues, which provides many benefits for
successful application of the mentioned optical modalities.197–200 For optical
diffusion imaging, diffusion wave blood flow monitoring and other principally
scattering-based techniques, some reduction of scattering also could be fruitful
to provide more flexibility in getting quantitative information about tissue
lesions and functioning. One of the optical clearing approaches is based on
impregnation of tissue by an immersion liquid whose refractive index is higher
than the refractive index of interstitial fluid (ISF). When applied to skin, the
diffusion resistance of the stratum corneum (SC) makes the transdermal
delivery of immersion agents and water lost by skin difficult. This happens
because SC possesses a protective function due to cell keratinization, tight
packing, and existence of lipid bridges between them. Thus, SC is a dense
medium with a poor penetration for foreign molecules. Dermis is the next
thicker layer of the skin, which is mostly fibrous tissue well supplied by blood,
and thus can be easily impregnated by exogenous or endogenous liquids
(immersion agents). Subcutaneous tissue contains a big portion of a fat
cellular layer, which is much less penetrative for diffusing molecules than
dermis. Such a specific structure of skin defines the methodology of its
effective optical clearing, which is related to the immersion of the refractive
indices of scatterers (keratinocytes components in epidermis, collagen and
elastin fibers in dermis) and ground matter.199
In this section, the optical clearing of SC, epidermis, and dermis will
be briefly analyzed in the framework of receiving more precise and valuable
information from reflectance spectroscopy, polarization measurements, Raman
spectroscopy, confocal microscopy, and optical coherence tomography (OCT),
148 Chapter 3
as well as from nonlinear spectroscopies, such as two-photon fluorescence and

second harmonic generation (SHG). In vitro, ex vivo, and in vivo spectroscopic,
polarization, and OCT studies of human skin and underlying tissues are
presented. Optical clearing agents (OCA), cosmetic preparations, and drug
delivery, as well as skin enhanced permeation, is under discussion. Some
important applications of the optical immersion technique in medicine and
cosmetology, including glucose noninvasive sensing and laser tattoo removal,
will be demonstrated.
3.9.2 Skin compression and stretching

It is possible to change optical transmittance through soft tissue by means of
compression or stretching.45,54,158–163,199 Reduction of transmittance can be
achieved due to removal of blood and interstitial liquid (water) from the
compressed site. Chan et al.161 observed that compression caused leaking
around the skin specimen. Some of the extracellular fluid along the edge of the
skin sample was forced out upon compression. This results in a higher
refractive index of the ground matter, whose value becomes close to that of
scatterers (cell membrane or collagen fibers) (see Chapter 1 of Vol. 1). Blood
removal from the compressed area also changes absorption and scattering. On
the other hand, compression reduces specimen thickness, which might increase
the effective concentration of scatterers and immovable chromophores within
the compressed area of tissue.159,161 Therefore, compression may also give rise
to an increase in scattering and absorption coefficients. For pressure
uniformly distributed over the sample surface of a few square centimeters,
the increase in scatterer and absorber concentration is likely to be stronger
than the reduction in index mis-match.164 Generally, at a compression
uniformly distributed over a sample surface, a decrease in optical reflectance
and increase in transmittance are observed in spite of the increase of
absorption and scattering coefficients. Corresponding data for human skin are
presented in Table 3.2.
It should be noted that the relative contributions of the mechanisms
mentioned affecting tissue optical properties is expected to change if a point-
wise compression is applied.156,185 The optical translucence of tissue is
significantly enhanced over a period of several minutes in the course of a
point-wise compression.156 This kinetic can be explained by tissue water
displacement caused by a local stress.
In vivo study of UV-induced erythematous human skin demonstrates a
high sensitivity of skin reflectance spectra to compression. See corresponding
data in Figs. 3.11, 3.12, 3.15, 3.17, and 3.18.
The in vivo application of an external mechanical compression on the
order of 105 Pa on skin leads to a decrease of the reflectance coefficient of the
skin in the spectral range of 400–1000 nm due to the decrease of skin
scattering properties. The magnitude of a dip in the spectrum in the range of
Table 3.2. Change of the optical properties at 500 and 810 nm of human skin samples of
approximately 2 2 cm2 at pressures of 0.0, 0.1 and 1.0 kg/cm2 uniformly distributed over
the sample surface.161 In vitro measurements were done using an integrating sphere
technique and the adding-doubling method for deconvolution of absorption and reduced
scattering coefficients. N is the type of the sample: 1–3: skin from a Caucasian male; 4–6:
skin from a Hispanic male; 7–9: skin from a Caucasian female. D is the thickness of the
sample.
No pressure Pressure 0.1kg/cm2 Pressure 1kg/cm2
N l, nm D, mm ma, cm–1 m0s , cm–1 D, mm ma, cm–1 m0s , cm–1 D, mm ma, cm–1 m0s , cm–1
1 500 0.40 4.7 34.0 0.15 11.7 86.4 0.12 10.0 81.9
810 0.2 10.2 0.5 22.9 0.1 24.2
2 500 0.23 4.8 69.1 0.05 22.6 320.9 0.05 19.3 309.5
810 0.2 19.6 0.3 108.1 0.3 105.7
3 500 0.25 5.9 47.0 0.13 11.8 94.9 0.13 11.7 78.6
810 0.4 17.6 1.1 27.0 1.3 30.2
4 500 0.70 3.4 23.9 0.35 6.2 44.1 0.28 7.5 47.7
810 0.6 7.5 0.9 15.1 1.5 12.7
5 500 0.78 3.6 21.8 0.62 4.6 37.1 0.48 4.7 35.5
810 0.9 6.8 1.6 12.0 0.8 9.1
6 500 0.63 4.4 27.0 0.48 4.4 31.6 0.33 6.5 37.9
810 1.1 8.4 0.3 7.2 0.3 8.7
7 500 0.42 5.3 21.3 0.30 6.7 26.1 0.27 7.7 28.7
810 0.5 9.7 1.5 13.2 1.5 14.8
8 500 0.50 5.6 24.2 0.30 9.7 36.8 0.20 13.1 49.4
810 1.3 8.1 1.9 12.2 2.3 14.0
9 500 0.50 4.8 26.3 0.34 5.8 31.8 0.23 9.2 42.4
810 1.1 6.9 0.8 8.6 1.2 10.6
500 to 600 nm also decreases (Fig. 3.35a) due to a decrease of skin absorption
caused by the hemoglobin of blood contained in the skin. This process is
inertial and proceeds within a period of time of about 3 to 4 min.74
After removal of the external mechanical compression, the recovery of the
skin occurs within 40 to 50 min, and, in this case, during the first several
seconds, a considerable increase in the content of blood (hemoglobin) (by a
factor of 2.4 compared to the norm) occurs [Figs. 3.35(b) and 3.36]. This can
also be seen from Fig. 3.37, which presents temporal changes in the
reflectance coefficient of the skin at two wavelengths (540 and 700 nm).
It may be mentioned that during the first several seconds after removal
of the external mechanical compression, a considerable increase of the
degree of oxygenation (by a factor of 2 to 4 compared to the norm)
(Fig. 3.38) exists, which can be caused by a sharp inflow of arterial blood to
the volume of the skin tissue that was subjected to the external mechanical
compression.
In the NIR, two processes control the in vivo skin reflectance:
alterations of the absorbing and the scattering properties of the skin. The
external compression decreases both absorption and scattering of the skin,
150 Chapter 3
44
42
Reflectance, % 40
38
36
34 1
32
2
30 3
4
28 5
500 600 700 800 900
Wavelength, nm
(a)
44
40
36
Reflectance,%
4
32
5
28 1
24 2
3
20
500 600 700 800 900
Wavelength, nm
(b)
Figure 3.35 Temporal changes in in vivo reflection spectra of skin from the human forearm
(a) upon application of an external mechanical compression (p ¼ 110 kPa) and (b) after its
removal. Curves in (a): (1) norm, t ¼ (2) 15, (3) 105, (4) 200, and (5) 290 s; curves in
(b): t ¼ (1) 0, (2) 10, (3) 150 s, (4) 40 min, and (5) norm.74
and the decrease of the scattering properties as a result of the displacement

of water out of the volume of skin subjected to the compression is
predominant.230 It has been found that, under the application of an external
compression of 110 kPa, the water content in the skin decreases by
about 10%.231
Figure 3.36 Dynamics of erythema index E of skin tissue under conditions of an external
mechanical compression (within the time interval of 0 to 330 s) and after its removal (within a
time interval longer than 330 s). p = (1) 13.9 and (2) 110 kPa.74
44
2
40
36
Reflectance, %
32
1
28
24
20
0 100 200 300 400 500 600

Time, s
Figure 3.37 Temporal changes in the reflection coefficients of skin from the human
forearm at two wavelengths l under conditions of the application of an external mechanical
compression (within the time interval of 0–290 s) and after its removal (within a time interval
longer than 290 s); l = (1) 540, (2) 700 nm.74
3.9.3 Immersion optical clearing

It is possible to achieve a definite reduction in scattering by matching the
refractive indices of the scattering centers and background material, by
means of intratissue administration of appropriate chemical agents–OCA.
152 Chapter 3
100
80
Oxygenation degree
60
40
20
-20
0 100 200 300 400 500 600
Time, s
Figure 3.38 Dynamics of the degree of oxygenation of blood hemoglobin of skin tissue
under conditions of an external mechanical compression (p ¼ 110 kPa) (within the time
interval of 0–290 s) and after its removal (within a time interval longer than 290 s). The
dashed line shows the time interval within which there is no blood in the skin tissue.74
Experimental studies on optical clearing of normal and pathological skin,

and the management of reflectance and transmittance spectra using
water, glycerol, glycerol-water solutions, glucose, sunscreen creams,
cosmetic lotions, gels, and pharmaceutical products are described in
Refs. 44, 47, 146, 157, 158, 171–180, 186–194, 197, 201–229 Controlling of
the skin optical properties was achieved by controlling the refractive indices
of scatterers (keratinocyte components in epidermis, collagen and elastic
fibers in dermis) and the background material.
The principles of the optical immersion technique is based on the
impregnation of a tissue by a biocompatible chemical agent, which also may
have some hyperosmotic properties. Any connective (fibrous) tissue can be
effectively impregnated by a liquid agent or its water solution. The
transmission of a collimated light beam of intensity I0 by a tissue layer of
thickness d . 1/meff is defined by the exponential law that accounts for
multiple scattering:158
T c ¼ I ðdÞ∕I 0 I 0 bs expðmeff dÞ, (3.53)
where bs accounts for additional irradiation of the upper layers of a tissue due
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
to backscattering (photon recycling effect), mef f ¼ 3ma ðma þ m0s Þ.
Owing to the fibrous structure of skin dermis – the main portion of skin, it
is quite reasonable to assume that the kinetics of fluid diffusion within the skin
could be approximated by free diffusion. Therefore, to describe the kinetics of
the refractive index change and corresponding decrease of the scattering

coefficient when a chemical agent freely diffuses within the interfiblillar
substance of a tissue, the following approximate solution of matter diffusion
equation can be used197,203
∂C a ðx,tÞ ∂2 C a ðx,tÞ
¼ Da , (3.54)
∂t ∂x2
where Ca(x, t) is the fluid concentration, Da is the agent coefficient of
diffusion, and x is the spatial coordinate.
The volume-averaged concentration of an agent Ca(t) in the first-order
approximation has a view197,203

C a ðtÞ ¼ ∫0 C a ðx,tÞdx ≅ C a0 1 exp
1 d t
, (3.55)
2 t
When an agent is administrated through only one sample surface (topical
agent application) the characteristic diffusion time is:
d2
t : (3.56)
Da
At tissue impregnation by a chemical agent, the refractive index of the
background (interfibrillar) media n0 is a time-dependent function of the agent
concentration that penetrates into a sample Ca(t) and is defined by Eq. (3.55).
The time-dependent volume fraction of the agent within the tissue sample fa is
proportional to its concentration Ca, thus using the law of Gladstone and
Dale, we can get:
n0 ðtÞ ¼ n0i ðtÞf 0 ðtÞ þ na f a ðtÞ, (3.57)

where f0(t) þ fa(t) ¼ 1, and n0i is the initial refractive index of the base
material without the agent.
The expression for the scattering coefficient, derived for a system of
noninteracting thin cylinders with a number of fibrils per unit area rs has a
form:204
5 4
p a ½n0 ðtÞ3 2
ms ðtÞ ≅ rs ð½mðtÞ 1Þ 1 þ
2 2
, (3.58)
l30 ½mðtÞ2 þ 12
where rs ¼ fcyl/pa2, fcyl is the surface fraction of the cylinders’ faces, a is
the cylinder radius, m(t) ¼ ns/n0(t) is the relative index of refraction of the
cylinders (scatterers) to the background (interfibrillar space), and l0 is the
wavelength in vacuum. Due to the square dependence, the sensitivity to
indices’ matching is very high, for instance, a change of m from 1.1 to 1.01
give about a 100-fold decrease of scattering coefficient, ms2 ≅ 0.01ms1.
154 Chapter 3
The described model of a tissue is applicable to any fibrous tissue

including skin dermis and muscle.
Human skin consists of the following three principal layers: SC, viable
epidermis, and dermis. SC is a lipid-protein biphasic structure, having a
thickness of only 10–20 mm on most surfaces of the human body. Due to cell
membrane keratinization and tight packing of cells and lipid bridges between
them, SC is a dense medium with a poor penetration for foreign molecules.202
The excellent diffusional resistance of the SC makes the transdermal delivery
of immersion agents and water lost by skin difficult to overcome.
To understand how to control the transport and barrier functions of the
skin work, it is important to have knowledge of the water distribution within
the different layers.202 The SC receives water from within the body, but water
may also be taken up from the environment. In vivo the diffusion of water
across the SC is a passive process that can be modified with application of
osmotic OCAs. The outside SC layer is certainly drier than the innermost
cornified layer or adjacent moist granular layer. Thus, there exists a
concentration gradient causing transepidermal water loss that can be
increased by osmotic OCA application. Low permeation of normal skin is
determined by the SC; however, a viable epidermis and dermis, in spite of
their much better permeability than the SC, may significantly delay the OCA
diffusion inside the body because of their higher thicknesses.
No significant difference was found for the diffusion across epidermis and
232
SC. The diffusion coefficient Dw of the flow of water through tissue
corresponds to a viscous flow through a very fine porous medium. As has been
determined in strongly hydrated SC, Dw is about four orders of magnitude less
than the self-diffusion coefficient in water.232 The diffusivity (Dw) of water in
SC increases from 3 10–10 to 10–9 cm2/sec as the surrounding relative
humidity increases from 46% to 81%. The average water content of the SC in
a normal state is in the range from 15% to 30% (by weight) as measured from
the outmost to the innermost layers. The normal hydration of the viable cell
layers of the epidermis is not significantly different from that of the dermis,
which is 70% by weight.202
Dermis is the next thicker layer of the skin, which is mostly fibrous tissue,
thus it can be easily impregnated by exogenous or endogenous liquids
(OCAs). Subcutaneous tissue contains a big portion of a fat cellular layer,
which is much less penetrative for diffusing molecules than dermis. Such a
specific structure of skin defines the methodology of its effective optical
clearing, which is related to the matching of refractive indices of scatterers
(keratinocyte components in epidermis, collagen, and elastin fibers in dermis)
and ground matter.197–200
To compare the diffusion time for different skin layers, we suppose that
for small molecules, such as glycerol and propylene glycol, the diffusion
coefficient is close to that of water, which is, Da ¼ 3 10–10 cm2/sec for SC.
As the SC thickness is in the range of d ¼ 10–20 mm, it follows from Eq. (3.55)
that ta can be ranged from 0.9 to 3.6 hrs. For a living epidermis of
thickness 100 mm and OCA diffusivity of Da ¼ 3 10–8 cm2/sec,
approximately, a diffusion time ta 0.9 hr can be provided. Two orders
higher diffusivity of the living epidermis in comparison with the SC is due to
the higher permeation ability of the epidermal cell membrane, which is similar
to the permeability of membranes of other epithelial cells. For a 1-mm dermis
thickness and typical diffusivity of fibrous tissue, Da ¼ 3 10–6 cm2/sec,212ta
can be estimated as 0.9 hr. In accordance with these estimations, 2.7–5.4 hrs
is needed for OCA diffusion through a skin layer. Evidently, in dependence on
tissue condition and site on the body, this time can be different.
Approximately equal contributions to the time delay of OCA permeation is
provided by all three major skin layers with 10-mm SC, 100-mm living
epidermis, and 1-mm dermis, of 2.7 hrs.
For a thicker, 20-mm SC, the resistance of SC dominates, thus various
vehicles and methods for reversible disruption of the SC protective function
should be used to provide a reasonable time for OCA diffusion. The SC is
functioning not only as a barrier against OCA penetration into skin, but also
as a reservoir for topically applied substances.202,233 Skin appendages, in
particular sebaceous glands, also serve as a reservoir and pathway for clearing
agents.234
3.9.3.1 In vitro spectrophotometry

Thin epidermal layers demonstrate high efficacy of immersion optical
clearing. Data of collimated transmittance measurements using a He-Ne
laser with 633 nm are presented in Table 3.3, which illustrates the influence of
different immersion agents on the average transmittance of thin epidermal
layers expressed as the ratio of the mean values of intensity before and after
agent application. At skin epidermal stripping using glue and glass substrates,
only the inner layers could be impregnated by a lotion. Their thickness is in
the range of 20–30 mm, thus these data mostly characterize the diffusivity of
the upper cell layers of living epidermis and its inclusion in the total scattering
of the sample.
An in vitro study of rat dorsal skin impregnated by anhydrous glycerol,
when the agent was applied to the dermal side of the skin sample, showed a
Table 3.3 Efficiency of the immersion agent application expressed as a ratio of mean
transmitted intensities before and after application of lotion to the skin stripping; n is the
index of refraction of the used lotion.173
Immersion 1 2 3 4 5 6 7
agent n = 1.449 n = 1.380 n = 1.356 n = 1.348 n = 1.354 n = 1.337 n = 1.396
〈IA〉/〈IB〉 12.8 3.7 4.9 4.1 5.9 5.3 7.9

1, 2, 3, 4, 5: glycerol-water-like compositions; 6: ultrasound gel; 7: DMSO 50%
156 Chapter 3
power wavelength dependence of the reduced scattering coefficient in the

wavelength range from 500 to 1200 nm, ms0 l–h, with a reduced scattering
coefficient at 500 nm ms0 50 cm–1 and h ¼ 1.12 for normal skin, and with a
subsequent decrease in ms0 (500 nm) and h with increased time in glycerol
(mostly due to the dehydration effect), ms0 30 cm–1 and h ¼ 1.09 for 5 min,
ms0 20 cm–1 and h ¼ 0.85 for 10 min, ms0 12 cm–1 and h ¼ 0.52 for 20 min,
and ms0 23 cm–1 and h ¼ 0.9 for the rehydrated sample kept in physiological
phosphate buffered saline solution for 20 min.177
More prolonged administration of glucose (up to 6 hrs) and glycerol (up
to 45 min) into the fresh rat skin samples at room temperature in the course
of tissue collimated transmittance measurements was also done.199,205 These
studies were performed to clarify the mechanisms of the skin optical clearing
and to optimize the technique. To avoid tissue damage and to provide less
viscosity of the chemical agent, a glycerol-water solution (88%) and 40%-
glucose, both of which are available in a drug store, were used as immersion
agents. Figure 3.39 illustrates the typical collimated transmittance spectra
and optical clearing kinetics. It is seen that the untreated rat skin is poorly
transparent for the visible light. Both glucose and glycerol administration
make this tissue highly transparent; a 15-fold increase of the collimated
transmittance for glucose [Fig. 3.39(a)] and 10-fold increase for glycerol at
700 nm for the samples with a fatty layer kept in solution for 45 min were
found. The efficiency is substantially greater with a removed fatty layer
[Fig. 3.39(b)], for which about a 50-fold transmittance increase is seen.
Optical clearing potential (OCP), defined as the ratio of values of tissue
reduced scattering coefficient before and after agent action, OCP ≡
ms0 (before)/ms0 (after), were measured in vitro for a variety of agents at their
application to the dermis side of human skin using a Franz diffusion
chamber.188 There was found no correlation between OCP and refractive
index for the used agents with indices in the range from 1.43 to 1.48, as well as
no correlation with osmolality in a wide range from 1,643 to 26,900 mOsm/kg,
but the highest values of OCP from 2.4 to 2.9 were provided by the agents
having both the highest refractive index and osmolality, such as glycerol, 1,4-
butanediol, and 1,3-butanediol.
Using near infrared spectroscopy (800-2200 nm), mass and water loss
measurements, and transdermal skin resistance measurements, such enhancers
of skin permeability as dimethyl sulfoxide (DMSO) and oleic acid were
compared with propylene glycol (PG) application onto the epidermal surface
of the samples of fresh porcine skin with a thickness of 1.52 0.18 mm.207
Because of clinical safety reasons, oleic acid could be an optimal choice as an
enhancer for the optical clearing of skin, because it is recognized as being safe
and free of side effects, and DMSO has some potential toxicity. After
application of the oleic acid solution (0.1 M of oleic acid and PG-40) the total
transmittance measured at the wavelength of 1278 nm of the skin sample
Figure 3.39 The time-dependent collimated transmittance of the rat skin samples
measured at different wavelengths at administration of immersion solution - 40%-glucose
in a bath. (a) Sample thickness 0.73 mm, with hypodermic fatty layer; (b) sample thickness
0.57 mm, with removed hypodermic fatty layer.205
increased by 41% and 58%, respectively, for 30- and 60-min treatment, while
the diffuse reflectance decreased by 39% and 47%, respectively.
A method of accelerating penetration of the index-matching com-
pounds by enhancing skin permeability through creating a lattice of micro-
zones (islets) of limited thermal damage in the SC was proposed.208 A
combination of a flashlamp system (EsteLux, Palomar Medical Technolo-
gies, Inc.) and a specially designed appliqué with a pattern of absorbing
158 Chapter 3
470 nm
650 nm
1,5
1,4
Relative transmittance
1,3
1,2
1,1
1,0
-500 0 500 1000 1500 2000 2500 3000 3500 4000

Time, sec
Figure 3.40 Transmittance spectra recording using Frantz diffusion chamber: left panel
demonstrates experimental arrangement (1 – physiological saline; 2 – OCA; 3 – tissue
sample; 4 – lens; 5 – water jacket; 6 – illuminating and collecting fibers); right panel shows
spectral kinetics of optical transmittance of pig skin specimen of 1-mm thickness at topical
application of 40%-glucose (see insertion), 2 pulses 27 J/cm2 (20 ms) with islet damage
mask þ 2 pulses 27 J/cm2 (20 ms) were applied to perforate stratum corneum.208
centers (center size 75–100 mm, lattice pitch 450–500 mm) has been
used to create the lattice of islets of damage (LID). Several index-matching
agents, including glucose and glycerol, have been tested. A high degree of
optical clearance of a full-thickness pig, rat, and human skin in vitro and
in vivo has been demonstrated with 40%-glucose and 88%-glycerol solution
after creating an LID with a few optical pulses (fluence 14-36 J/cm2, 20 ms
pulse duration) (Fig. 3.40).
One of the possible mechanisms of more effective optical clearing of skin
by an osmotic OCA at LID application is connected with more effective
dehydration of skin due to local disruption of the SC. A special experiment
with Yucatan pig skin samples at LID (two 20-millisecond pulses of 30 J/cm2
each with appliqué and two 20-msec pulses of 30 J/cm2 each after appliqué
removal) done for a sample non-treated by OCA and treated with 40%-
glucose showed that the sample area strictly corresponding to the LID area
was dehydrated more effectively than the surrounding area free of LID. The
dehydrated area was clearly seen as a more translucent area with less
thickness. Tissue shrinkage was of 20–25% in the thickness of the sample with
LID kept at room temperature for 2 hrs and up to 40% in the thickness of the
sample treated with 40%-glucose and kept in an oven at a temperature of 51°C
during 2 hrs. Thus, more effective skin dehydration is expected at LID due
to SC partial ablation. Besides, the local heating of the living epidermis
under the SC may enhance skin permeability due to induced phase transition
of epidermal intercellular lipids from the gel phase to liquid crystalline
phase.235,236
3.9.3.2 In vivo spectral reflectance measurement

In vivo topical application of glycerol, glucose, x-ray contrasts, propylene
glycol, cosmetic lotions, and gels also made human skin more translucent,
within a time period from a few minutes to a few hours.199 Water loss or
increase by means of moisturizing substances seriously influences skin optical
properties.
To enhance OCA permeation through SC, a number of specific
procedures, such as heating, electrophoresis, and sonophoresis, are usually
applied.202 To increase efficiency of the topical application of the OCAs,
gelatin gels containing clearing agents (verografin, glycerol, or glucose) were
designed.209 The diffusion rate of the agents within the gel layer can be rather
high, and this along with a comparatively large volume of the gel provided the
constant concentration of OCA, equal to agent content in the gel, at the skin
surface. For the intact skin of a volunteer, the best kinetics, i.e., the rate and
the degree of clearing (17%), was observed in the case of verografin-gel; after
40 min of observation, clearing still proceeds with a marked rate, while for
glycerol-gel, after 27 min the curve flattens out.
As the barrier function of the skin is associated mainly with SC, the
measurement was carried out on the skin after 30–50-mm-epidermal glue
stripping. Application of glucose-gel to the skin without the upper epidermis
layer gave a rapid 10% drop of reflected light intensity. Glycerol-gel gave
better results; over the time of observation, the decrease of the reflected signal
ranged up to 20%, and was twice that which was attained for intact skin.
The electophoretic applicator and gel with twice the content of gelatin were
also applied for human skin optical clearing.209 The results for glycerol-gel
showed that when the active electrode was connected as an anode, the
reduction of scattering by 20% was attained.
In vivo topical application of glycerol, glucose, trazograph (x-ray contrasting
substance), cosmetic lotions, and gels also made human skin more translucent
within a period of a few tenths of minutes.153,158,161,169,170,179–182 Physiological
glucose concentrations in human skin can be measured by NIR optical
methods due to refractive index matching.171–173 NIR glucose absorption
bands in a 2.0–2.5-mm spectral range also allows for the use of back
reflectance measurements from the skin in this range for noninvasive glucose
sensing.173 In addition to reflectance, several fluorescence-based techniques
for glucose sensing are available.173,174,215
The administration of glucose or glycerol by intradermal injection into rat
or hamster skin causes the decrease of reflectance and the corresponding
increase of tissue transmittance.176,197,199,205,210 This effect was observed at all
wavelengths during 15–18 min after glucose injection.210,211 The greatest
degree of tissue reflectance changes is found at the wavelengths from 580 to
750 nm, where scattering dominates. At the 16th min, the reflectance of the
skin was minimal (maximal transmittance); it decreased about 3.5 fold at
160 Chapter 3
700 nm. It was shown that a glycerol injection causes a more prolonged effect
of tissue optical clearing, but reflectance decreased a little bit less than for the
glucose injection. This can be explained by a higher viscosity of glycerol and
by its mostly indirect action via tissue dehydration.
Within one hour after a 40%-glucose intradermal injection applied to a
human healthy volunteer, the skin reflection coefficient decreases in average
by a factor of 3.8 and then exhibits a slow increase, which indicates that
glucose is eliminated from the observation area and the skin reflectance tends
to restore itself to the initial level.212 Based on these results and the proposed
skin clearing model, the main contribution to clearing was the refractive index
matching between collagen fibrils of the dermis (n ¼ 1.46) and the interstitial
space (initially n ¼ 1.36) to which glucose (n ¼ 1.39) diffuses. The diffusion
coefficient of glucose in dermis estimated from these experimental data is
DG ¼ (2.56 0.13) · 10–6 cm2/s; this value is 3.6 fold less than that for glucose
diffusion in water at 37°C, DG 9.2 · 10–6 cm2/s, and reflects the character of
dermis permeability for glucose. This and other data on diffusion coefficients
are presented in Table 3.4.
Water loss or increase by means of moisturizing substances seriously
influences skin optical properties.190–194 NIR reflectance spectroscopy is used
as a method to directly determine changes in free, bulk, and protein-bound
water and to assess scattering effects in skin for the evaluation of skin care
products.191 The following spectral bands are associated with water: free
water, 1879 nm; bulk water, 1890 nm; and protein-bound water, 1909 and
1927 nm. The effect of increases in ambient humidity is associated with
increased levels of free water in the skin, while moisturizers containing
hydroxyethyl cellulose, propylene glycol, dipropylene glycol, and glycerol
contribute to a decrease in the light scattering.191 The water observed in such
experiments is primarily in the stratum corneum (SC) since only a small part
of the reflected light comes from the epidermis or below.
Noninvasive measurement of the SC hydration can be performed using
attenuated total reflectance Fourier transform infrared (ATR FTIR)
spectroscopy.192,193 Three absorption bands are relevant for determination
of the water content in the SC: 3300 cm–1 (3030 nm), O–H and N–H vibra-
tions; 1645 cm–1 (6079 nm), amide I band; and 1545 cm–1 (6472 nm), amide II
band. The amide I band intensity is pronounced in the presence of water due
Table 3.4 Diffusion coefficient of glucose and glycerol in rat and human dermis evaluated
on the basis of in vitro and in vivo experimental data.195,199,205,212
OCA Da, cm2/sec (in vitro) Da, cm2/sec (in vivo)
88%-glycerol (rat skin) (5.1 2.3) 10–7 (1.16 0.03) 10–6

40%-glucose (rat skin) (3.2 0.4) 10–6 (1.10 0.15) 10–6
40%-glucose (human skin) – (2.56 0.13) 10–6
∗
These values of diffusion coefficients characterize not only OCA diffusion into tissue, but also diffusion of water from the
tissue due to interaction of these two fluxes.
to the strong absorption of water at 1645 cm–1 and the changes in the carbonyl
absorption under the influence of water, while the amide II band intensity is
due to protein alone. The intensity ratio of the amide I/amide II bands, also
called the moisture factor, is assumed to be a relative measure of SC
hydration.192 Various SC moisturizers based on glycerol, propylene glycol,
sodium lactate, natural moisturizing vegetal liposomes, butylene glycol,
polyglycerylmethacrylate, and urea were used for an in vivo SC hydration
study.192 Depending on the composition and concentration, the maximal
SC hydration could be reached in 0.5–2 hrs after application of the substance
to the skin surface. For some substances, a considerable moisturizing effect
was detectable up to 8 hours following application.
Dual wavelength (1300 and 1450 nm) optical coherence reflectance
measurement is a prospective technique for depth profiling of water
absorption within the skin.194
3.9.3.3 Frequency-domain measurements

The photon-diffusion theory provides independent determination of the
absorption and reduced scattering coefficients from the measurements at a
single modulation frequency (see Chapters 3 and 7 in Vol. 1). The expressions for
the measured quantities as the phase delay q and ac amplitude Aac have been
presented elsewhere.171,188 These expressions depend on the source-detector
separation r, reduced scattering coefficient m0s , and absorption coefficients ma.
The intensity and phase of photon-density waves generated by an NIR
optical source were measured at several source-detector separations. The light
source was a laser diode with a wavelength of 786 nm and power at the end of
a coupled multimode fiber (core diameter 62.5 mm) of about 4 mW.171,188 An
intensity modulation depth of about 80% at 140-MHz frequency was provided
by modulation of injection current of the laser diode. The experimental setup
was designed at the University of Pennsylvania.
For human forearm measurements, the tips of the source and detector
have been mounted in a rubber pad and fastened to the surface of the human
forearm to avoid random moving artifacts. The source-detector separation
was quite large: 2.5 cm.
For source-detection separation measurements that allow for thin tissue
layers examination, a special multichannel fiber optic probe working with
small (1–3 mm) separations was designed. It was used together with the Dicon
multichannel fiber optic switcher. The dynamical response of optical
properties (modulation depth and phase shift of intensity modulation of the
backscattered light) was measured for human skin via the interval of a
chemical agent (solution, gel, or oil) administration.
The multichannel fiber optical probe together with a Dicon multichannel
fiber optical switcher was used for in vivo study of human skin immersion
effects.171,188 The measurement for each separation was done during 10 s and
162 Chapter 3
relative phase shift (degrees)

r=0.35 cm
normalized amplitude
1.00 0.0
r=0.35 cm
r=0.208 cm
0.96 r=0.208 cm
-1.0
r=0.114 cm
r=0.114 cm
0.92
0 500 1000 0 500 1000 1500
time (seconds) time (seconds)
(a) (b)
Figure 3.41 Frequency-domain back-reflectance measurements for the small source-
detector separations.171,188 The time-dependent changes of the (a) amplitude and (b) phase
shift of signal for several source-detector separations (1.14, 2.08, and 3.5 mm) for in vivo
study of human arm under 20 min glycerol administration.
averaged, corresponding to one point in Fig. 3.41. The relative amplitude

(normalized to the initial amplitude) and the phase changes (the current phase
minus the initial phase) during 20 minutes of pure glycerol topical application
are shown in Fig. 3.41. Only scattering changes must be considered due to the
extremely low absorption of glycerol at the measuring wavelength. The
observed amplitude and phase changes are small, reflecting minor permea-
tions of epidermal cell layers to any chemical agent. Nevertheless, these
measurements show sensitivity of the frequency-domain method to the small
changes of the scattering coefficient of the skin.
For a large source-detector separation (2.5 cm), the cosmetic gel with a
refractive index n ¼ 1.403 (Pond’s) has been put on the surface of the arm and
the phase and ac amplitude measurements have been continuously provided.
One sampling point corresponds to one second. The results of measurement
during 30 min of the gel administration are shown in Fig. 3.42(a). The
observed temporal quasi-periodic fluctuations in the phase and amplitude of
the optical signal are caused mainly by heartbeats.
Results of calculations of tissue optical parameters are shown in
Fig. 3.42(b). The initial values of m0s and ma for the human skin were taken
from the literature,158 and the relative changes of these parameters were
calculated with a running averaging procedure for every 5-s interval in order
to exclude the influence of heartbeats. Corresponding temporal evolutions of
the scattering and absorption coefficients have been received. This study
shows that there were no noticeable changes in the absorption during the gel
administration trial. A slight increase in the absorption can probably be
explained by the increase of the water content in the skin due to the moisture
effect of the applied gel. The selected source-detector separation (2.5 cm) and
corresponding measuring volume are too large to make the matching effect a
0.008 33 15.6 0.3
amplitude
absorption coefficient (cm-1)

scattering coefficient (cm-1)
0.008
phase 15.2 μ 0.3
amplitude (rel.units)
s
32
phase (degrees)
0.007
μa
14.8 0.2
0.007
31
14.4 0.2
0.006
0.006 30 14.0 0.1

0 400 800 1200 1600 0 400 800 1200 1600
time (seconds) time (seconds)
(a) (b)
Figure 3.42 Frequency-domain back-reflectance measurements for the large source-
detector separation (2.5 cm).171,188 (a) The raw experimental data of the phase and ac
amplitude of the optical signal and (b) calculation of the absorption and the scattering
coefficients.
useful procedure for topical application of the gel. Only about a 6 % reduction
of the scattering coefficient averaged over the large measuring volume was
observed. This means that the scattering coefficient of the upper (superficial)
layers of the skin changed more effectively. Refractive index matching of the
fiber tips and tissue surface is also important.
In vivo frequency domain measurements for immersed tissues show that
the refractive index matching technique provided by the appropriate chemical
agent or cosmetic preparation application can be successfully used in tissue
spectroscopy and imaging, when reduction of the scattering properties is
needed.
3.9.4 Skin blood flow imaging

Small blood microvessels can be clearly identified visually by the naked eye in
an in vivo study of hamster177,211 and rat210 skin, when a transparent window
in skin was created by glycerol drops to the subdermal side of a native hamster
dorsal skin flap window preparation,211 or by intradermal injection of
glycerol210,211 or 40%-glucose.211 More precise visualization of the vessel
network and immersion agents may have an influence on blood microvessel
functioning, which gives the possibility to control the functioning of tissue
within the area of agent action.
3.9.5 OCT imaging

The result of the optical coherence tomography (OCT) study is the
measurement of optical backscattering or reflectance, R(z), from the tissue
versus axial ranging distance, or depth, z. The backscattering depends on the
164 Chapter 3
optical properties of tissue, i.e., the absorption ma and scattering ms

coefficients, and local tissue reflectivity.158 If the local tissue reflectivity is
not changeable with depth, the total attenuation coefficient mt ¼ ma þ ms can
be obtained from the OCT reflectance measurements at two different depths,
z1 and z2:

1 Rðz1 Þ
mt ¼ ln , (3.59)
ðDzÞ Rðz2 Þ
where Dz ¼ |z1 z2|.
The multiple scattering is a detrimental factor that limits OCT imaging
performances: imaging resolution, depth, and localization. To improve the
imaging capabilities, the multiple scattering of tissue must be reduced. The
immersion technique at the application of biocompatible agents is a
prospective technique for OCT, because the depth of OCT images and their
contrast can essentially be improved at immersion (Fig. 3.43).213
It should be noted that the high sensitivity of OCT signal to immersion of
living tissue by glucose allows one to monitor its concentration in the skin at a
physiological level.214,215
3.9.6 Confocal microscopy

The increase of the upper tissue layers’ transparency can improve the pene-
tration depth, image contrast, and spatial resolution in confocal microscopy
0 0
0.5 0.5
60 60
Reflectance (dB)
Reflectance (dB)
1 1
1.5 1.5
40 40
2 2
0 0.5 1 1.5 0 0.5 1 1.5

20 20
(a) (b)
0 0
0.5 0.5 0 0
1 1 0 0.5 1 1.5 2 0 0.5 1 1.5 2
1.5 1.5
(a) Depth (mm) (b) Depth (mm)
2 2
0 0.5 1 1.5 0 0.5 1 1.5 60 60

Reflectance (dB)
Reflectance (dB)
(c) (d)
0 0
0.5 0.5
40 40
1 1
1.5 1.5
20 20
2 2
0 0
0 0.5 1 1.5 0 0.5 1 1.5
0 0.5 1 1.5 2 0 0.5 1 1.5 2
(e) (f)
(c) Depth (mm) (d) Depth (mm)
Figure 3.43 Ex vivo rat skin OCT imaging in the course of optical clearing using topical
application of 80%-glycerol onto the rat skin: left set of images presents OCT images at
times (a) 0, (b) 3, (c) 10, (d) 15, (e) 20, and (f) 40 min after application of glycerol; all the units
presented are in millimeters, and the vertical axis presents the imaging depth; right set of
graphs presents OCT in-depth reflectance profiles (a) 0, (b) 5, (c) 15, (d) 25 min after
glycerol application.213
as well.216 By Monte Carlo simulations of the point spread function, it was

shown that the signal spatial localization offered by a confocal probe in the
skin tissues during their clearing is potentially useable for the reticular dermis
monitoring. The results of the simulation predict that after 20 min of the
chemical agent diffusion after intradermal glycerol or glucose injection, the
signal from the tissues located twice as deep in the skin can be detected
(Fig. 3.44).
3.9.7 Fluorescence and Raman signal detection

The improvement of the detected fluorescence and Raman signals traveling
through skin in in vitro and in vivo experiments at topical application of
hyperosmotic OCAs, such as anhydrous glycerol (n ¼ 1.47) and pure
DMSO (n ¼ 1.47), and a highly concentrated glucose (n ¼ 1.46), was
demonstrated.217–221 These experiments showed changes of tissue layer
transmittance at optical immersion for light from a fluorescent source placed
behind a tissue layer. A successful transcutaneous Raman spectroscopy of
bone with optical clearing by glycerol for an in vivo animal model was
provided221 (Fig. 3.45).
However, more often, fluorophores are distributed within a tissue layer or
even a multi-layered tissue structure may contain a number of different
fluorophores. In that case, the behavior of the fluorescence signal at tissue
immersion is not so evident, because the cross-section for fluorescence
emission depends on the amount of the absorbed light by fluorescent centers.
Such a bulk cross-section decreases as multiple scattering decreases. Thus, at
skin optical clearing instead of enhancement of a fluorescence signal, one can
see its damping. Evidently that depends on the depth, where the fluorophore
is, and what layer of a skin is optically cleared.222
One of the new directions in tissue spectroscopy is associated with a multi-
photon fluorescence scanning microscopy.158 However, it has been shown that
the effect of light scattering in multi-photon fluorescence scanning microscopy
is to drastically reduce the penetration depth to less than that of the equivalent
single-photon fluorescence while largely leaving the resolution unchanged.
This happens mostly due to excitation beam defocusing (distortion) in the
scattering media. Though some improvement in the penetration depth of two-
photon microscopy can be obtained by optimizing the pulse shape and
repetition rate for the sample under investigation, reduction of scattering is
believed to be more effective in the penetration depth and image contrast
improvement.223 Two-photon fluorescence microscopy provides high-
resolution images of human skin in vivo.
The first demonstration of two-photon in-depth signal improvement using
an optical immersion technique with hyperosmotic agents, such as glycerol,
propylene glycol, and glucose, was done by the authors of Ref. 223 in ex vivo
experiments with human dermis. The enhancement of contrast, as well as the
166 Chapter 3
Figure 3.44 Confocal reflectance microscopy, Monte Carlo (MC) modeling of skin optical
clearing by 40%-glucose solution at intradermal injection: (a), (b), and (c) focusing depth is
600 mm, (a) – in 5 min, (b) – in 10 min, and (c) – in 20 min after injection; left and right down
panels: axial point spread function (PSF) before (left) and 20 min after (right) injection at
focusing (1) – 300 mm, (2) – 600 mm and (3) – 900 mm into the skin.216
increase of penetration depth (from 40 to 80 mm) and total intensity, were

found. The relative contrast at optical clearing has a value of 215% at 40 mm
and dramatically increases with increasing depth.
3.9.8 Second harmonic generation

In skin, second harmonic generation (SHG) is provided mostly within the
dermis due to its main component, collagen, which has an appreciable
nonlinear susceptibility.224–226 Due to optical clearing, less scattering in the
Figure 3.45 Transcutaneous Raman spectroscopy of bone: spectra of rat tibia without
and with skin optical clearing by glycerol.221
epidermis for the incident long wavelength light (800 nm) and especially for
the backward SHG short wavelength light (400 nm) improves the SHG
images of dermis collagen structures. At 100%-glycerol application to rodent
skin dermis and tendon samples, as well as to an engineered tissue model
(raft), a high efficiency of tissue optical clearing was achieved in the
wavelength range from 400 to 700 nm, but the SHG signal was significantly
degraded in the course of glycerol application and it was returned back to its
initial state after tissue rehydration by the application of saline.224 This SHG
signal reduction could be considered as a linear scattering effect.199,225
3.9.9 Skin heating

Long-pulse laser heating induces reversible and irreversible changes in the
optical properties of tissue.182 The total transmittance Tt decreases, and the
diffuse reflectance Rd increases in both fresh and precoagulated human skin.
Signs of nonlinear behavior in the optics of turbid biological media during
pulsed laser heating have also been observed. Possible mechanisms responsi-
ble for this nonlinear optical response are listed in Table 3.5.
A certain reproducible effect of temperature between 25°C and 40°C on
the reduced scattering coefficient of human dermis and subdermis was found
during an ex vivo study in the NIR.183,184 For dermis, the relative change of
the reduced scattering coefficient showed an increase [(4.7 0.5) 10–3 °C–1]
and for subdermis a decrease [( 1.4 0.28) 10–3 °C–1]. A possible
explanation for this effect has been suggested.183,184 The main scattering
components of the subdermis were assumed to be lipids in membranes and
vacuoles. It is known that lipids undergo phase changes at a certain
168 Chapter 3
Table 3.5 Possible mechanisms responsible for inducing reversible changes in tissue
optical response on laser long-pulsed irradiation175
Mechanism Description Optical response
Thermal lensing, n(T) ¼ n Gradient in the index of refraction caused Decrease in Tt and
(273K) þ DT(r, z, t)(dn/dT) by nonuniform heating increase in Rd
Temperature dependence of the Changes in the size and/or shape of Increase in Tt and
reduced scattering coefficient: scatterers due to temperature rise decrease in Rd
m0s (T) ¼ ms(T)[1 g(T)] (as m0s decreases)
Water transport Temporary local dehydration during laser Increase in Tt and
heating decrease in Rd
Thermal expansion Decrease in tissue density and increase in Decrease in Tt and
tissue thickness caused by thermal increase in Rd
expansion of tissue
temperature, which alters their orientation, mobility, and packing order.

Glyco-lipids found in human cell membranes undergo phase transitions in the
temperature range from 25 °C to 45 °C, namely, the transition from a gel
phase through a stable crystalline phase to a liquid-crystalline phase with
increasing temperature. The decrease in scattering coefficient seen experimen-
tally with increasing temperature is, therefore, consistent with an increase in
fluidity known to occur in lipids with increasing temperature. Modifications
of the collagen fiber structure of dermis caused by a temperature increase,
possibly through changes in hydration, are the most plausible explanation of
the increased scattering properties.183,184
3.9.10 UV radiation
UV irradiation causes erythema, stimulates melanin synthesis, and can induce
edema and tissue proliferation if the radiation dose is sufficiently
large.35,44,54,160,196 All these photobiological effects may be responsible for
variations in the optical properties of skin and need to be taken into
consideration when prescribing phototherapy or optical diagnostics. See data
in Figs. 3.9–3.12, 3.14, 3.17, and 3.18.
3.9.11 Applications
The concept of noninvasive blood glucose sensing using the scattering
properties of blood and tissue as an alternative to spectral absorption and
polarization methods for the monitoring of physiological glucose concentra-
tions in blood of diabetic patients is under intensive discussion.215
Noninvasive determination of glucose was attempted using light scattering
of skin tissue components measured by a spatially-resolved diffuse
reflectance173,227 or NIR frequency-domain reflectance techniques.172 Both
approaches are based on changes in glucose concentration, which affect the
refractive index mismatch between the interstitial fluid and tissue fibers, and
hence m0s .
OCT was proposed for noninvasive assessment of glucose concentration
in tissues.214,215 High resolution of the OCT technique may allow high
sensitivity, accuracy, and specificity of glucose concentration monitoring due
to precise measurements of glucose-induced changes in the tissue optical
properties from the layer of interest (dermis).
Reversible tissue optical clearing technology has valuable features that
can be applied to optical diagnostics, where the reduction of scattering
significantly improves information. For example, nanobiophotonic
approaches based on gold nanoparticle interactions with the individual cell
may be beneficial for more precise cell structure imaging at immersion.228
Skin impregnation by a biocompatible clearing agent allows one to
significantly improve the image quality and accuracy of spectroscopic
information, to provide an effective monitoring of trans-epidermal membrane
drug delivery, a precision laser photothermolysis of skin lesions and tattoo
removal229,237,238 (see Fig. 3.46), and an effective phototherapy.
Some other applications, such as near-infrared fluorescence image-guided
surgery239 or luminescence detection of salmonella typhimurium through
porcine skin,219 are under intensive study. Indeed, optical clearing technology
should be applied to account for many factors that can accompany optical
clearing effects. For instance, the authors of Ref. 239 failed in getting a
statistically significant increase in fluorescence signal from Indocyanine Green
(ICG) dye in in vivo studies with the aim of image-guided surgery at topical
application of OCAs. The reasons for that could be the rather prolonged OCA
diffusivity of SC and other skin layers that, as was estimated above, takes a
few hrs (2.7–5.4 hrs for normal human skin) and the impossibility of optical
clearing of underlying tissues as well as washing out of both clearing agents
and ICG (the time is around 1 hr) in living tissues. However, this problem
could be solved by injection of a clearing agent into the tissue layer lying
above the surgery area, which is contrasted by ICG fluorescence. This
injection should be carefully done to prevent OCA from penetrating during
Figure 3.46 Demonstration of enhancement of contrast of tattoo image at skin topical

clearing by 50%-propylene glycol solution for the volunteer; two right images are magnified
images on the border of the tattoo taken before and after 2 hrs of PPG application;
co-polarized images6HHFRORUSODWHV
170 Chapter 3
surgery into the area where ICG is fluorescing, otherwise its fluorescence will
be damped due to less scattering of the surrounding tissue due to the decrease
of ICG molecule bulk absorption cross-section. 222
3.9.12 Conclusion
Control of optical properties of skin is a potential for:
1. a significant improvement of the image quality and accuracy of spectro-
scopic and morphological information using spectrophotometry,197–201
Raman spectroscopy,221 confocal microscopy,216 two-photon-223 and
SHG-225,226 imaging;
2. an effective monitoring of trans-epidermal membrane drug delivery and
glucose sensing;214,215
3. a precision laser photothermolysis of skin lesions and tattoo;229,237,238,240
4. an effective phototherapy.
At topical application of OCAs, fractional ablation of stratum corneum is
one of the prospective techniques.208,241,242 OCT and multiphoton imaging
provide in vivo measurements of agent diffusivity for different skin layers
within the range of their in-depth probing.230–244
Recent achievements of skin optical clearing and the study of OCA
impact on skin and underlying tissues are summarized in Refs. 158, 243–252.
3.10 Conclusion
The optical properties of skin are dynamic, variable, and complex, but they
can be quantitatively determined under certain simplifying assumptions.
Simplified skin models are used for the analysis of spectral reflection and
fluorescence, measured in a certain spectral range, where chromophores
and fluorophores with optical characteristics known a priori (melanin,
hemoglobin, bilirubin, water, collagen, etc.) dominate. These models enable
quantitative assessment of skin chromophores and fluorophores with high
precision and in real time. Combined fluorescence/reflectance techniques
make it possible to obtain additional information on the state of the skin.
Color and polarizing imaging are among the most prospective methods of
skin diagnosis in clinics. The control of optical properties is a prospective
method for the increase of effectiveness of skin disease diagnosis and
therapy.
Acknowledgments
We would like to thank all our colleagues and students from Research-
Educational Institute of Optics and Biophotonics at Saratov State University,
especially A.N. Bashkatov, L.E. Dolotov, E.A. Genina, and A.B. Pravdin.
We are grateful to R.K. Wang, G.B. Altshuler, K.V. Larin, S. Tanev, and
I. Yaroslavsky for collaboration. The support from the Government of the
Russian Federation (grant No. 14.Z50.31.0004), Russian Presidential grant
NSh-703.2014.2, and The Tomsk State University Academic D.I. Mendeleev
Fund Program is appreciated.
References
1. G. J. Müller and D. H. Sliney (eds.), “Dosimetry of laser radiation in
medicine and biology,” Advanced Optical Technology IS5, SPIE Press,
2. S. L. Jacques, “The role of skin optics in diagnostic and therapeutic uses
of lasers,” in Lasers in Dermatology, R. Steiner (ed.), Springer-Verlag,
Berlin, 1–21 (1991).
3. S. Wan, K. F. Jaenicke, and J. A. Parrish, “Comparison of the
erythemogenic effectiveness of ultraviolet-B (290-320 nm) and
ultraviolet-A (320-400 nm) radiation by skin reflectance,” Photochem.
Photobiol. 37, 547–552 (1983).
4. N. Kollias and A. N. Baqer, “On the assessment of melanin human skin
in vivo,” Photochem. Photobiol. 43, 49–54 (1986).
5. N. Kollias and A. N. Baqer, “Quantitative assessment of UV-induced
pigmentation and erythema,” Photodermatology 5, 53–60 (1988).
6. J. B. Dawson, D. J. Barker, E. Grassam, and J. A. Cotterill et al.,
“A theoretical and experimental study of light absorption
and scattering by in vivo skin,” Phys. Med. Biol. 25, 695–702
(1980).
7. P. H. Andersen and P. Bjerring, “Non invasive computerized analysis of
skin chromophores in vivo by reflectance spectroscopy,” Photodermatol.
Photoimmunol. Photomed 7(6), 249–257 (1990).
8. H. Kopola and A. Lahti, “Two-channel fiber optic skin erythema
meter,” Opt. Eng. 32(2), 222–226 (1993).
9. J. W. Feather, K. S. Ryatt, and J. B. Dawson, “Reflectance
spectrophotometric quantification of skin color changes induced by
topical corticosteroid preparations,” Br J. Dermatol. 106, 436–443
(1982).
10. K. S. Ryatt, J. W. Feather, J. B. Dawson, and J. A. Cotrell, “The
usefulness of reflection spectrophotometric measurements during psor-
alens and ultraviolet A therapy for psoriasis,” J. Am. Acad. Dermatol.
85, 558–562 (1985).
11. J. C. Seitz and C. G. Whitmore, “Measurement of erythema and tanning
responses in human skin using a tri-stimulus colorimeter,” Dermatolo-
gica 177(2), 70–75 (1988).
12. C. Edwards and R. Heggie, “A small solid state meter for
measuring melanin pigmentation,” in The Environmental Threat to
172 Chapter 3
the Skin, R. Marks and G. Plewig (eds), Martin Dunitz, 149–154

(1992).
13. S. el-Gammal, K. Hoffman, and P. Steiert et al., “Objective assessment
of intra- and inter-individual skin color variability: an analysis of human
skin reaction to sun and UVB,” in The Environmental Threat to the Skin,
R. Marks and G. Plewig (eds), Martin Dunitz, 99–115 (1992).
14. Y. Mendelson and M. V. J. Solomita, “The feasibility of spectrophoto-
metric measurements of arterial oxygen saturation from fetal scalp
utilizing noninvasive skin-reflectance pulse oximetry,” Biomed. Instrum.
Technol. 26, 215–224 (1992).
15. Y. Mendelson and B. D. Ochs, “Noninvasive pulse oximetry utilizing
skin reflectance photoplethysmography,” IEEE Trans. Biomed. Eng. 35,
798–805 (1988).
16. H. Adhoute, J. de Rigal, J. P. Marchand, Y. Privat, and J. L. Leveque,
“Influence of age and sun exposure on the biophysical properties of the
human skin: an in vivo study,” Photodermatol. Photoimmunol. Photomed.
9, 99–103 (1992).
17. P. H. Andersen, “In vivo cutaneous assays to evaluate topical
corticosteroids and nonsteroidal antiinflammatory drugs using reflec-
tance spectroscopy,” in Bioengineering of the Skin: Cutaneous Blood
Flow and Erythema, E. Barardesca, P. Elsner, and H. I. Maibach (eds),
CRC Press, New York, 281–291 (1995).
18. F. Deleixhe-Mauhin, J. M. Krezinski, G. Rorive, and G. E. Pierard,
“Quantification of skin color in patients undergoing maintenance
hemodialysis,” J. Am. Acad. Dermatol. 6, 950–953 (1992).
19. L. Duteil and J. P. Ortonne, “Colorimetric assessment of the effects of
azelaic acid on light-induced skin pigmentation,” Photodermatol.
Photoimmunol. Photomed. 9, 67–71 (1992).
20. J. W. Feather and M. Haijzadeh et al., “A portable scanning reflectance
spectrophotometer using visible wavelengths for rapid measurement of
skin pigments,” Phys. Med. Biol. 34, 807–820 (1989).
21. N. Kollias and A N. Baqer, “A method for the noninvasive
determination of melanin in human skin in vivo,” in Biological Effects
of UVA Radiation, F. Urbach and R. W. Gange (eds), Praeger Publ.,
New York, 226–230 (1986).
22. N. Kollias and J. M. Al-Hassan, “Evaluation of toxicity of catfish skin
toxin using diffuse reflectance methods,” J. Pharmauceutic. Biomed.
Anal. 9(3), 255–259 (1991).
23. N. Kollias, A. Baqer, I. Sadig, and R. M. Sayer, “In vitro and in vivo
ultraviolet-induced alterations of oxy- and deoxyhemoglobin,” Photo-
chem. Photobiol. 56(2), 223–227 (1992).
24. L. O. Svaasand, L. T. Norvang, E. S. Fiskerstrand, E. K. S. Stopps,
M. W. Berns, and J. S. Nelson, “Tissue parameters determining the
visual appearance of normal skin and port wine stains,” Las. Med. Sci.
10, 55–65 (1995).
25. H. Takiwaki, S. Shirai, Y. Kanno, Y. Watanabe, and S. Arase,
“Quantification of erythema and pigmentation using a videomicroscope
and a computer,” Brit. J. Dermatol. 131, 85–92 (1994).
26. H. Takiwaki, H. Miyamoto, and K. Ahsan, “A simple method to
estimate CIE-L*a*b* values of the skin from its videomicroscope
image,” Skin Res. Technol. 3, 42–44 (1997).
27. G. B. Altshuler and V. V. Tuchin, Physics behind the light-based
technology: Skin and hair follicle interactions with light in Light Based
Systems for Cosmetic Application, ed. Gurpreet Ahluwalia, William
Andrew, Inc., Norwich, NY, USA, 2008.
28. M. F. Yang, V. V. Tuchin, and A. N. Yaroslavsky, “Principles of light
skin interactions,” Light-Based Therapies for Skin of Color, Ed.
E. Baron, Springer, N. Y., 2009
29. D. J. Leffell, M. L. Stetz, L. M. Milstone, and L. I. Deckelbaum, “In
vivo fluorescence of human skin,” Arch. Dermatol. 124, 1514–1518
(1988).
30. H. J. C. M. Sterenborg, M. Motamedi, R. F. Wagner, J. R. M. Duvic,
S. Thomsen, and S. L. Jacques, “In vivo fluorescence spectroscopy and
imaging of human skin tumors,” Lasers Med. Sci. 9, 344–348 (1994).
31. H. J. C. M. Sterenborg, S. Thomsen, S. L. Jacques, and M. Motamedi,
“In vivo autofluorescence of an unpigmented melanoma in mice.
Correlation of spectroscopic properties to microscopic structure,”
Melanoma Res. 5, 211–216 (1995).
32. H. Zeng, C. MacAulay, D. I. McLean, and B. Palcic, “Spectroscopic
and microscopic characteristics of human skin autofluorescence emis-
sion,” Photochem. Photobiol. 61(6), 639–645 (1995).
33. R. R. Anderson, “In vivo fluorescence of human skin [letter, comment],”
Arch. Dermatol. 125, 999–1000 (1989).
34. Yu. P. Sinichkin, S. R. Utz, I. V. Meglinsky, and H. A. Pilipenko,
“In vivo human skin spectroscopy: II. Fluorescence spectra,” Opt.
Spectrosc. 80, 431–438 (1996).
35. Yu. P. Sinichkin, S. R. Utz, A. H. Mavlutov, and H. A. Pilipenko,
“In vivo fluorescence spectroscopy of the human skin: experiments and
models,” J. Biomed. Opt. 3, 201–211 (1998).
36. I. Miksik, R. Struzinsky, and Z. Deyl, “Change with age of UV
absorbance and fluorescence of collagen and accumulation of
e-hexosyllysine in collagen from Wistar rats living of different food
restriction regimes,” Mech. Ageing. Development 57, 163–174 (1991).
37. P. R. Odetti, A. Borgoglio, and R. Rolandi, “Age-related increase of
collagen fluorescence in human subcutaneous tissue,” Metabolism 41(6),
655–658 (1992).
174 Chapter 3
38. B. A. Gilchrest, T. B. Fitzpatrick, and R. R. Anderson et al.,

“Localization of melanin pigmentation with Wood’s lamp,” Brit. J.
Dermatol. 96, 245–248 (1977).
39. W. Lohmann and E. Paul, “In situ detection of melanomas by
fluorescence measurements,” Naturwissenschaften 75, 201–202 (1988).
40. H. J. C. M. Sterenborg, M. Motamedi, R. F. Wagner, S. Thomsen, and
S. L. Jacques, “In vivo fluorescence spectroscopy for the diagnosis of skin
diseases,” Proc. SPIE 2324, 32–38 (1994).
41. E. W. J. van Breggen, A I. Rem, M. M. Christian, C. J. Yang, K. H.
Calhoun, H. J. C. M. Sterenborg, and M. Motamedi, “Spectroscopic
detection of oral and skin tissue transformation in a model for squamous
cell carcinoma autofluorescence versus systematic aminolevulinic acid-
induced fluorescence,” IEEE Select Top. Quant. Electr. 2(4), 997–1007
(1996).
42. L. C. Lucchina, N. Kollias, R. Gillies, S. B. Phillips, J. A. Muccini, M. J.
Stiller, R. J. Trancik, and L. A. Drake, “Fluorescence photography in
the evaluation of acne,” J. Amer. Acad. Dermatol. 35, 58–63 (1996).
43. N. Kollias, R. Gillies, C. Cohen-Goihman, S. B. Phillips, J. A. Muccini,
M. J. Stiller, and L. A. Drake, “Fluorescence photography in the
evaluation of hyperpigmentation in photodamaged skin,” J. Amer. Acad.
Dermatol. 36, 226–230 (1997).
44. R. R. Anderson and J. A. Parrish, “Optical properties of human skin,” in
The Science of Photomedicine, J. D. Regan and J. A. Parrish (eds),
Plenum Press, New York, 147–194 (1982).
45. M. J. C. van Gemert, S. L. Jacques, H. J. C. M. Sterenborg, and W. M.
Star, “Skin Optics,” IEEE Trans. Biomed. Eng. 36(12), 1146–1154
(1989).
46. V. V. Tuchin, Lasers and Fiber Optics in Biomedical Science, 2nd ed.,
Fizmatlit, Moscow (2010).
47. V. V. Tuchin, “Light scattering study of tissues,” Physics-Uspekhi 40(5),
495–515 (1997).
48. N. Kollias and A. N. Baqer, “Spectroscopic characteristics of human
melanin in vivo,” J. Invest. Dermatol. 85, 38–42 (1985).
49. N. Kollias and A. N. Baqer, “Absorption mechanisms of human
melanin in the visible, 400–720 nm,” J. Invest. Dermatol. 89, 384–388
(1987).
50. P. H. Andersen and P. Bjerring, “Remittance spectroscopy: hardware
and measuring principles,” in Bioengineering of the skin: cutaneous blood
flow and erythema, E. Berardesca, P. Elsner, and H. I. Maibach (eds),
CRC Press, 231–241 (1995).
51. R. Marchesini, C. Clemente, E. Pignoli, and M. Brambilla, “Optical
properties of in vivo epidermis and their possible relationship with
optical properties of in vivo skin,” Photochem. Photobiol. 16, 127–140

(1992).
52. F. Hillenkamp, “Interaction between laser radiation and biological
systems,” in Lasers in Biology and Medicine, F. Hillenkamp, R. Pratesi,
and C. Sacci (eds), Plenum Press, New York, 57–61 (1979).
53. N. Ramanujam, M. F. Mitchell, and A. Mahadevan et al., “Fluores-
cence spectroscopy: a diagnostic tool for cervical intraepithelial
neoplasia (CIN),” Gynecologic Oncology 52, 31–38 (1994).
54. Yu. P. Sinichkin, S. R. Utz, and H. A. Pilipenko, “In vivo human skin
spectroscopy: I. Remittance spectra,” Opt. Spectrosc. 80, 260–267
(1996).
55. T. L. Norvang, E. L. Fiskerstrand, and K. Konig et al., “Comparison
between reflectance spectra obtained with an integrating sphere and a
fiber-optic collecting system,” Proc SPIE 2624, 155–164 (1996).
56. L. E. Dolotov and Yu. P. Sinichkin, “Features of applying fiber-optic
sensors in spectral measurements of biological tissues,” Opt. Spectrosc.
115(2), 187–193 (2013).
57. T. J. Farrell, B. C. Wilson, and M. S. Patterson, “The use of a neural
network to determine tissue optical parameters from spatially resolved
diffuse reflectance measurements,” Phys. Med. Biol. 37, 2281–2286
(1992).
58. R. R. Richards-Kortum, R. P. Rava, R. E. Petras, M. Fitzmaurice,
M. Sivak, and M. S. Feld, “Spectroscopic diagnosis of colonic
dysplasia,” Photochem. Photobiol. 53, 777–786 (1991).
59. H. Zeng, C. MacAulay, D. I. McLean, and B. Palcic, “Spectroscopy and
microscopy studies of skin tissue autofluorescence emission,” Proc. SPIE
2324, 198–207 (1995).
60. S. R. Utz, P. Knuschke, and Yu. P. Sinichkin, “In vivo evaluation
of sunscreens by spectroscopic methods,” Skin Res. Technol. V 2(3),
114–121 (1996).
61. S. R. Utz, J. Barth, P. Knuschke, and Yu. P. Sinichkin, “Fluorescence
spectroscopy of human skin,” Proc SPIE 2081, 48–57 (1993).
62. B. Chance, L. Mela, and D. Wong, in Flavins and Flavoproteins, K. Yagi
(ed), University Park Press, 102–121 (1968).
63. S. Kozikowski, L. J. Wolfram, and R. R. Alfano, “Fluorescence
spectroscopy of eumelanins,” IEEE J. Quant. Electr. QE-20(12), 1379–
1382 (1984).
64. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd ed., Plenum
Press, New York (1999).
65. B. Chance, “Pyridine nucleotide as an indicator of the oxygen
requirements for energy-linked functions of mitochondria,” Circ. Res.
38, 31–36 (1976).
176 Chapter 3
66. W. Lohmann, J. Mubmann, C. Lohmann, and W. Kunzel, “Native

fluorescence of the cervix uteri as a marker for dysplasia and invasive
carcinoma,” Eur. J. Ostet Gynecol. Reprod Biol. 31, 249–253 (1989).
67. T. Bocher, J. Beuthan, O. Minet, I. Schmitt, B. Fuchs, and G. Muller,
“Fiberoptical sampling of NADH-concentration in Guinea-pig hearts
during ischemia,” Proc. SPIE 2324, 166–176 (1995).
68. W. Hopewell, “The skin: its structure and response to ionizing
radiation,” Int. J. Radiat. Biol. 57, 751–773 (1990).
69. J. Fuchs, Oxidative Injury in Dermatopathology, Springer Verlag, Berlin
(1992).
70. L. Vitellaro-Zuccarello, S. Cappelletti, V. D. P. Rossi, and M. Sari-
Gorla, “Stereological analysis of collagen and elastic fibers in the normal
human dermis: variability with age, sex and body region,” Anat. Record.
238, 153–162 (1994).
71. S. R. Utz, J. Barth, and P. Knushke, “Fluorescence spectroscopy in
dermatology,” Izvest. Acad. Nauk. (Physical series) 59(6), 156–160 (1995).
72. I. V. Meglinsky, Yu. P. Sinichkin, S. R. Utz, and H. A. Pilipenko,
“Simulation of fluorescent measurements in the human skin,” Proc.
SPIE 2389, 621–631 (1995).
73. K. Konig, A. Ruck, and H. Schneckenburger, “Fluorescence detection
and photodynamic activity of endogenous protoporphyrin in human
skin,” Opt. Eng. 31(7), 1470–1474.
74. A. Nakhaeva, M. R. Mohammed, O. A. Zyuryukina, and Yu. P.
Sinichkin, “The effect of an external mechanical compression on in vivo
optical properties of human skin,” Opt. Spectrosc. 117(3), 506–512
(2014).
75. I. V. Ermakov and W. Gellermann, “Dermal carotenoid measurements
via pressure mediated reflection spectroscopy,” J. Biophotonics, V 5(7),
559–570 (2012).
76. H. Zeng, C. MacAulay, B. Palcic, and D. I. McLean, “Monte Carlo
modeling of tissue autofluorescence measurement and imaging,” Proc
SPIE 2135, 213–219 (1994).
77. G. Zonios, R. M. Cothren, and J. Arendt et al., “Fluorescence
spectroscopy for colon cancer diagnosis,” Proc. SPIE 2324, 9–13
(1995).
78. H. Zeng, C. MacAulay, B. Palcic, and D. I. McLean, “A computerized
autofluorescence and diffuse reflectance spectroanalyzer system for in
vivo skin studies,” Phys. Med. Biol. 38, 231–240 (1993).
79. H. Zeng, C. MacAulay, D. I. McLean, and B. Palcic, “A novel
microspectrophotometer and its biomedical application,” Opt. Eng. 32,
1809–1814 (1993).
80. N. N. Zhadin and R. R. Alfano, “Correction of the internal absorption
effect in fluorescence emission and excitation spectra from absorbing and
highly scattering media: theory and experiment,” J. Biomed. Opt. 3,

171–186 (1998).
81. S. R. Utz, Yu. P. Sinichkin, I. V. Meglinsky, and H. A. Pilipenko,
“Fluorescence spectroscopy in combination with reflectance measure-
ments in human skin examination: what for and how,” Proc. SPIE 2324,
125–136 (1994).
82. H. Zeng, C. MacAulay, D. I. McLean, and B. Palcic, “Spectroscopy and
microscopy studies of skin tissue autofluorescence emission,” Proc. SPIE
2324, 198–207 (1995).
83. S. R. Utz, P. Knuschke, and Yu. P. Sinichkin, “In vivo evaluation of
sunscreens by spectroscopic methods,” Skin Res. Technol. 2(3), 114–121
(1996).
84. S. Wan, K. F. Jaenicke, and J. A. Parrish, “Quantitative evaluation of
ultraviolet induced erythema,” Photochem. Photobiol. 37, 643–648
(1983).
85. A. Knudsen. Prediction of later hyperbilirubinemia by measurement of
skin color on the first postnatal day and from cord blood bilirubin,”
Dan. Med. Bull. 39, 193–196 (1992).
86. M. Strange and G. Cassady, “Neonatal transcutaneous bilirubinome-
try,” Clin. Perinatol. 12, 51–62 (1985).
87. B. L. Diffey, R. J. Oliver, and P. M. Farr, “A portable instrument for
quantifying erythema induced by ultraviolet radiation,” Br. J. Dermatol.
111, 663–672 (1984).
88. N. Kollias and A. Baqer, “An experimental study of the changes in
pigmentation in human skin in vivo with visible and near infrared light,”
Photochem. Photobiol. 39, 651–659 (1984).
89. N. Kollias, “The spectroscopy of human melanin pigmentation,” in
Melanin: Its Role in Human Photoprotection, KS, Valdenmar Publishing
Co., 31–38 (1995).
90. P. Bjerring and P. H. Andersen, “Skin reflectance spectrophotometry,”
Photodermatol. 4, 167–176 (1987).
91. T. Dwyer, L. Blizzard, and R. Ashbolt et al. “Cutaneous melanin density
of Caucasians measured by spectrophotometry and risk of malignant
melanoma, basal cell carcinoma, and squamous cell carcinoma of the
skin.” Am. J. Epidemiol. 155, 614–621 (2002).
92. L. E. Dolotov, Yu. P. Sinichkin, V. V. Tuchin, S. R. Utz, G. B. Altshuler,
and I. V. Yaroslavsky, “Design and evaluation of a novel portable
erythema-melanin-meter,” Lasers Surg. Med. 34(2), 127–135 (2004).
93. G. N. Stamatas, B. Z. Zmudzka, N. Kollias, and J. Z. Beer, “Non-
invasive measurements of skin pigmentation in situ,” Pigment Cell Res.
17, 618–626 (2004).
94. L. E. Dolotov, Yu. P. Sinichkin, V. V. Tuchin, G. B. Al’tshuler, and
I. V. Yaroslavsky, “Specific features of diffuse reflection of human face
178 Chapter 3
skin for laser and non-laser sources of visible and near-IR light,” Quant.
Electron. 41(4), 329–334 (2011).
95. F. F. Jobsis, “Noninvasive, infrared monitoring of cerebral and
myocardial oxygen sufficiency and circulatory parameters,” Science 19,
1264 (1977).
96. H. Liu, D. A. Boas, Yu. Zhang, A. G. Yodh, and B. Chance,
“Determination of optical properties and blood oxygenation in tissue
using continuous NIR light,” Phys. Med. Biol. 40, 1983–1993 (1995).
97. S. J. Matcher and C. E. Cooper, “Absolute quantification of
deoxyhaemoglobin concentration in tissue near infrared spectroscopy,”
Phys. Med. Biol. 39, 1–17 (1994).
98. B. Chance, M. Cope, E. Gratton, N. Ramanujam, and B. Tromberg,
“Phase measurement of light absorption and scatter in human tissue,”
Rev. Sci. Instrum. 69, 3457–3481 (1998).
99. E. M. Sevick, B. Chance, and J. Leigh et al., “Quantitation of time- and
frequency- resolved optical spectra for the determination of tissue
oxygenation,” Anal. Biochem. 195, 330–351 (1991).
100. M. Trorniley, L. Livera, Y. Wickramasinghe, S. A. Spenccer, and
P. Rolfe, “The non invasive monitoring of cerebral tissue oxygenation,”
Adv. Exp. Med. Biol. 277, 323 (1990).
101. I. V. Meglinsky and S. J. Matcher, “The determination of absorption
coefficient of skin melanin in visible and NIR spectral region,” Proc.
SPIE 3907, 143–150 (2000).
102. T. J. Farrell, M. S. Patterson, and B. Wilson, “A diffuse theory model of
spatially resolved, steady-state diffuse reflectance for the noninvasive
determination of tissue optical properties in vivo,” Med. Phys. 19, 879–
888 (1992).
103. N. Kollias, A. Baqer, I. Sadig, and R. M. Sayer, “In vitro and in vivo
ultraviolet-induced alterations of oxy- and deoxyhemoglobin,” Photo-
chem. Photobiol. 56, 223–227 (1992).
104. N. Kollias, R. Gillies, and J. A. Muccini et al., “A single parameter,
oxygenated hemoglobin, can be used to quantify experimental irritant-
induced inflammation,” J. Invest. Dermatol. 90, 421–424 (1991).
105. N. Kollias, R. Gillies, J. A. Muccini, S. B. Phillips, and L. A. Drake,
“Oxyhemoglobin is a quantifiable measure of experimentally induced
chronic tretinoin inflammation and accommodation in photodamaged
skin,” Skin Pharmacol. 10, 97–104 (1997).
106. M. Haijzadeh, J. W. Feather, and J B. Dawson, “An investigation of
factors affecting the accuracy of in vivo measurements of skin pigments
by reflectance spectrophotometry,” Phys. Med. Biol. 35, 1301–1315
(1990).
107. N. Kollias, “Physical basis of skin color and its evaluation,” Clinics
Dermatol. 13, 361–367 (1995).
108. R. D. Kenet, “Digital imaging in dermatology,” Clinics Dermatol. 13,

381–392 (1995).
109. A. Kienle, L. Lilge, and I. A. Vitkin et al., “Why do veins appear
blue? A new look at an old question,” Appl. Opt. 35(7), 1151–1160
(1996).
110. E. H. Land, “Recent advances in retinex theory,” Vision Res. 26, 7–21
(1986).
111. C. Queille, L. Duteil, J. Czernielewski, and H. Schaefer, “Colorimetric
evaluation of the human skin blanching assay,” in Noninvasive Methods
for the Quantification of Skin Functions, P. J. Frosch and A. M. Kligman
(eds), Springer-Verlag, 92–103, Berlin (1993).
112. Yu. P. Sinichkin, S. R. Utz, P. Knushke, A. H. Mavlyutov, and H. A.
Pilipenko, “In vivo human skin autofluorescence: Color perception,”
Proc. SPIE 2927, 217–221 (1996).
113. J. Steinmetz and P. Bjerring, “Video-optical monitoring of wheal and
flare reactions. Effects of topical Na-sucrose-sulphate,” Skin Res
Technol. 1, 90–95 (1995).
114. L. Savolainen, J. Kontinen, J. Roning, and A. Oikarinen, “Application
of machine vision to assess involved surface in patients with psoriasis,”
Br. J. Dermatol. 137, 395–400 (1997).
115. D. L. MacAdam, Color Measurements: Themes, and Variations.
Springer-Verlag, Berlin, 1985.
116. P. Elsner, “Chromametry: hardware, measuring principles, and stan-
dardization of measurements,” in Bioengineering of the Skin: Cutaneous
Blood Flow and Erythema, E. Berardesca, P. Elsner, and H. I. Maibach
(eds), CRC Press, New York, 247–252 (1995).
117. H. Takiwaki, L. Ovengaard, and J. Serup, “Comparison of narrow-band
reflectance spectrophotometric and tristimulus colorimetric measure-
ments of skin color,” Skin Pharmacol. 7, 217–225 (1994).
118. P. Clarys, K. Alawaeters, R. Lambrecht, and A. O. Barel, “Skin color
measurements: comparison between three instruments: the Chroma-
meter, the DermaSpectrometer, and the Mexameter,” Skin Res. Techn.
6, 230–238 (2000).
119. S. B. Park, D. H. Suh, and J. I. Youn, “A long-term time course of
colorimetric evaluation of ultraviolet light-induced skin reactions,” Clin.
Exp. Dermatol. 24, 315–320 (1999).
120. S. Alaluf, D. Atkins, and K. Barrett et al. “The impact of epidermal
melanin on objective measurements of human skin color,” Pigment. Cell
Res. 15, 119–126 (2002).
121. M. D. Shriver and E. J. Parra, “Comparison of narrow-band reflectance
spectroscopy and tristimulus colorimetry for measurements of skin and
hair color in persons of different biological ancestry,” Am. J. Phys.
Anthropol. 112, 17–27 (2000).
180 Chapter 3
122. G. N. Stamatas and N. Kollias, “Blood stasis contributions to the

perception of skin pigmentation,” J. Biomed. Opt. 9, 315–322 (2003).
123. H. Takiwaki, Y. Miyaoka, H. Kohno, and S. Arase, “Graphic analysis
of the relationship between skin colour change and variations in
the amounts of melanin and haemoglobin,” Skin Res. Tech. 8, 78–83
(2002).
124. H. Adhoute, R. Grossman, M. Cordier, and B. Soler, “Chromametric
quantification of pigmentary changes in the solar lentigo after sun-
light exposure,” Photodermatol. Photoimmunol. Photomed. 10, 93–96
(1994).
125. A. Chardon, I. Cretoi, and C. Hourseau, “Skin colour typology and
suntanning pathways,” Int. J. Cosm. Sci., 13, 191–208 (1991).
126. L. Andreassi and L. Casini et al., “Measurement of cutaneous color and
assessment of skin type,” Photodermatol. Photoimmunol. Photomed. 7,
20–24 (1990).
127. T. Igarashi, K. Nishino, and S. K. Nayar, “The appearance of human
skin: a survey,” Foundations and trends in computer graphics and vision,
3(1), 1–95 (2007).
128. E. Tur, “Skin pharmacology,” in Bioengineering of the Skin: Cutaneous
Blood Flow and Erythema, E. Berardesca, P. Elsner, and H. I. Maibach
(eds), CRC Press, New Work, 259–268 (1995).
129. A. O. Barel and P. Clarys et al., “The Visi-Chroma VC-100: a new
imaging colorimeter for dermatocosmetic research,” Skin Res. Tech. 7,
24–31 (2001).
130. B. Jung, B. Choi, and A. J. Durkin et al., “Characterization of port wine
stain skin erythema and melanin content using cross-polarized diffuse
reflectance imaging,” Lasers Med. Sci. 34, 174–181 (2004).
131. L. E. Dolotov, D. A. Zimnyakov, and Yu. P. Sinichkin, “Computer
imaging of spatial distribution of the human skin chromophores,”
Biomedical Technologies and Radioelectronics, no. 5-6, 89–95 (2004).
132. V. V. Tuchin, L. Wang, and D. A. Zimnyakov, Optical Polarization in
Biomedical Applications, Springer-Verlag, Berlin, Heidelberg, N. Y.,
2006.
133. N. Kollias, “Polarized light photography of human skin,” in Bioengi-
neering of the Skin: Skin Surface Imaging and Analysis, K.-P. Wilhelm,
P. Elsner, E. Berardesca, and H. I. Maibach (eds), CRC Press, New
York, 95–106 (1997).
134. S. L. Jacques, J. C. Ramella-Roman, and K. Lee, “Imaging skin
pathology with polarized light,” J. Biomed. Optics 7(3), 329–340
(2002).
135. D. A. Zimnyakov and Yu. P. Sinichkin, “A study of polarization decay
as applied to improved imaging in scattering media,” J. Opt. A: Pure
Appl. Opt. 2, 200–208 (2000).
136. D. A. Zimnyakov, Yu. P. Sinichkin, and V. V. Tuchin, “Polarization

reflectance spectroscopy of biological tissues: diagnostics applications,”
Radiophysics and Quantum Electronics 47(10-11), 860–875 (2004).
137. D. A. Zimnyakov, Yu. P. Sinichkin, I. A. Kiseleva, and D. N. Agafonov,
“Effect of absorption of multiply scattering media on the degree of
residual polarization of backscattered light,” Optics and Spectroscopy,
92(5), 765–771 (2002).
138. R. R. Anderson, “Polarized light examination and photography of the
skin,” Arch Dermatol. 127, 1000–1005 (1991).
139. S. B. Phillips, J. A. Muccini, and P. F. Bilden et al., “Spectroscopic
evaluation of the change in erythema accompanying treatment of
psoriatic plaques with a topical steroid,” J. Invest. Dermatol. 100, 543
(1993).
140. J. A. Muccini, N. Kollias, and S. B. Phillips et al., “Polarized light
photography in the evaluation of photoaging,” J. Am. Acad. Dermatol.
33, 765–769 (1995).
141. P. F. Bilden, S. B. Phillips, N. Kollias, J. A. Muccini, and L. A. Drake,
“Polarized light photography of acne vulgaris,” J. Invest. Dermatol. 98,
606 (1992). (Abstract).
142. S. L. Smith, N. Kollias, and R. Gillies et al., “Improvement of psoriatic
plaques treated with calcipotriol is better visualized using polarized light
photography and spectroscopy,” J. Invest. Dermatol. 104, 689 (1995).
(Abstract).
143. V. Backman, R. Gurjar, and K. Badizadegan et al., “Polarized light
scattering spectroscopy for quantitative measurement of epithelial
cellular structures in situ,” IEEE JSTQE 5, 1019–1027 (1999).
144. X. Feng, R. Patel, and A. N. Yaroslavsky, “Wavelength optimized
cross-polarized wide-field imaging for oninvasive and rapid evaluation of
dermal structures,” J. Biophotonics 1–8 (2015).
145. B. Sennbenn, K. Giese, K. Plamann, N. Harendt, and K. Kolmel, “In
vivo evaluation of the penetration of topically applied drugs into human
skin by spectroscopic methods,” Skin Farmacol. 6, 152–160 (1993).
146. K. F. Kolmer, B. Sennhenn, and K. Giese, “Investigation of skin by
ultraviolet remittance spectroscopy,” British J. Dermatol. 122(2), 209–
216 (1990).
147. J. M. Menter, “Recent developments in UVA photoprotection,” Int. J.
Dermatol. 29, 389–394 (1990).
148. C. A. Elmets, A. Vargas, and C. Oresajo, “Photoprotective effects of
sunscreens in cosmetics on sunburn and Langerhans cell photodamage,”
Photodermatol. Photoimmunol. Photomed.. 9, 113–120 (1992).
149. G. A. Groves and P. D. Forbes, “A method for evaluating the
photoprotective action of sunscreens against UV-A radiation,” Int. J.
Cosmetic Sci. 4, 15–24 (1982).
182 Chapter 3
150. K. H. Kaidbey and A. Barnes, “Determination of UVA protection

factors by means of immediate pigment darkening in normal skin,”
J. Amer. Acad. Dermatol. 25, 262–266 (1991).
151. G. Sauerman and U. Hoppe, “A rapid non-invasive method to evaluate
the light protective potential of sunscreens,” J. Soc. Cosmet. Chem. 36,
125–141 (1985).
152. C. Cole, “Multicenter evaluation of sunscreen UVA protectiveness with
the protection factor test method,” J. Am. Acad. Dermatol. 30, 729–736
(1994).
153. B. L. Diffey and J. Robson, “A new substrate to measure sunscreen
protection factors throughout the ultraviolet spectrum,” J. Soc. Cosmet.
Chem. 40, 127–133 (1989).
154. R. M. Sayer and P. P. Agin, “A method for the determination of UVA
protection for normal skin,” J. Amer. Acad. Dermatol. 23, 429–440
(1990).
155. F. Urbach, “Ultraviolet A transmission by modern sunscreens: is there
a real risk?” Photodermatol. Photoimmunol. Photomed.. 9, 237–241
(1992).
156. P. O. Rol, “Optics for transscleral laser applications,” Dissertation for
the degree of Doctor of Natural Sciences, Institute of Biomedical
Engineering, Zurich, Switzerland (1991).
157. V. V. Tuchin, S. R. Utz, and I. V. Yaroslavsky, “Tissue optics, light
distribution, and spectroscopy,” Opt. Eng. 33, 3178–3188 (1994).
158. V. V. Tuchin, Tissue Optics: Light Scattering Methods and Instruments
for Medical Diagnosis, 3rd ed., PM254, SPIE Press, Bellingham, WA
(2015).
159. A. P. Ivanov, S. A. Makarevich, and A. Ya. Khairulina, “Propagation of
radiation in tissues and liquids with densely packed scatterers,” J. Appl.
Spectrosc. (USSR) 47, 662–668 (1988).
160. S. R. Utz, Yu. P. Sinichkin, and H. A. Pilipenko, “In vivo laser
fluorescence spectroscopy of human skin: the effect of erythema,” Opt.
Spectrosc. 76, 864–868 (1994).
161. E. K. Chan, B. Sorg, and D. Protsenko et al., “Effects of compression on
soft tissue optical properties,” IEEE J. Select. Tops Quant. Electr. 2,
943–950 (1996).
162. J. M. Schmitt, X. Bao, and S. Xiao, “Micro-elastography of tissue with
OCT,” Proc SPIE 3598, 47–55 (1999).
163. J. M. Schmitt, “OCT elastography: imaging microscopic deformation
and strain of tissue,” Optics Express 3, 199–211 (1998).
164. H. Lui and R. Anderson, “Photodynamic therapy in dermatology: recent
developments,” Dermatol. Clinics 11, 1–13 (1993).
165. W.-H. Boehncke, K. Konig, and R. Kaufmann et al., “Photodynamic
therapy in psoriasis: suppression of cytokine production in vitro and
recording of fluorescence modification during treatment in vivo,” Arch.

Dermatol. 286, 300–303 (1994).
166. R. A. Weersink, J. E. Hayward, K. R. Diamond, and M. S. Patterson,
“Accuracy of noninvasive in vivo measurements of photosensitizer
uptake based on a diffusion model of reflectance spectroscopy,”
167. J. Webber, D. Kessel, and D. Fromm, “On-line fluorescence of human
tissues after oral administration of 5-aminolevulinic acid,” J. Photochem.
Photobiol. B: Biology 38, 209–214 (1997).
168. H. A. Green, D. Bua, R. R. Anderson, and N. S. Nishioka, “Burn depth
estimation using indocyanine green fluorescence,” Arch. Dermatol. 128,
43–49 (1992).
169. S. Fickweiler, R.-M. Szeimies, and W. Baumler et al., “Indocyanine
green: Intracellular uptake and phototherapeutic effects in vitro,”
J. Photochem. Photobiol. B. Biol. 38, 178–183 (1997).
170. W. R. Chen, W.-G. Zhu, and J. R. Dynlacht et al., “Long-term tumor
resistance induced by laser photo-immunotherapy,” Int. J. Cancer 81,
808–812 (1999).
171. V. V. Tuchin, J. Culver, and C. Cheng et al., “Refractive index matching
of tissue components as a new technology for correlation and diffusing-
photon spectroscopy and imaging,” Proc. SPIE 3598, 111–120 (1999).
172. J. S. Maier, S. A. Walker, S. Fantini, M. A. Franceschini, and
E. Gratton, “Possible correlation between blood glucose concentration
and the reduced scattering coefficient of tissues in the near infrared,”
Opt. Lett. 19, 2062–2064 (1994).
173. J. T. Bruulsema, J. E. Hayward, and T. J. Farrell et al., “Correlation
between blood glucose concentration in diabetics and noninvasively
measured tissue optical scattering coefficient,” Opt. Lett. 22, 190–192
(1997).
174. R. J. McNichols and G. L. Cote, “Optical glucose sensing in biological
fluids: an overview,” J. Biomed. Opt. 5, 5–16 (2000).
175. R. J. Russell, M. V. Pishko, C. C. Gefrides, M. J. McShane, and G. L.
Cote, “A fluorescence-based glucose biosensor using concanavalin A and
dextran encapsulated in a poly(ethylene glycol) hydrogel,” Anal. Chem.
71, 3126–3132 (1999).
176. E. Lankenau, J. Welzel, R. Birngruber, and R. Engelhardt, “In vivo
tissue measurements with optical low coherence tomography,” Proc.
SPIE 2981, 78–84 (1997).
177. G. Vargas, E. K. Chan, J. K. Barton, H. G. Rylander III, and A. J.
Welch, “Use of an agent to reduce scattering in skin,” Laser. Surg. Med.
24, 133–141 (1999).
178. V. V. Tuchin, “Coherent optical techniques for the analysis of tissue
structure and dynamics,” J. Biomed. Opt. 4, 106–124 (1999).
184 Chapter 3
179. V. V. Tuchin, A. N. Bashkatov, and E. A. Genina et al., “Optics of living

tissues with controlled scattering properties,” Proc. SPIE 3863, 10–21
(1999).
180. V. V. Tuchin, D. A. Zimnyakov, and I. L. Maksimova et al., “Coherent,
low-coherent, and polarized light interaction with tissues undergo the
refractive indices matching control,” Proc. SPIE 3251,12–21 (1998).
181. A. Knüttel and M. Boehlau-Godau, “Spatially confined and temporally
resolved refractive index and scattering evaluation in human skin
performed with optical coherence tomography,” J. Biomed. Opt. 5,
83–92, 2000.
182. W.-C. Lin, M. Motamedi, and A. J. Welch, “Dynamics of tissue optics
during laser heating of turbid media,” Appl. Opt. 35, 3413–3420 (1996).
183. C. R. Simpson, M. Kohl, M. Essenpreis, and M. Cope, “Near-infrared
optical properties of ex vivo human skin and subcutaneous tissues
measured using the Monte Carlo inversion technique,” Phys. Med. Biol.
43, 2465–2478 (1998).
184. J. Laufer, C. R. Simpson, M. Kohl, M. Essenpreis, and M. Cope, “Effect
of temperature on the optical properties of ex vivo human dermis and
subdermis,” Phys. Med. Biol.. 43, 2479–2489 (1998).
185. B. Nemati, A. Dunn, A. J. Welch, and H. G. Rylander III, “Optical
model for light distribution during transscleral cyclophotocoagulation,”
Appl. Opt. 37, 764–771 (1998).
186. A. N. Yaroslavsky, S. R. Utz, S. N. Tatarintsev, and V. V. Tuchin,
“Angular scattering properties of human epidermal layers,” Proc SPIE
2100, 38–41 (1994).
187. V. V. Tuchin, “Immersion effects in tissues,” Proc. SPIE 4162, 1–12
(2000).
188. V. V. Tuchin, “Controlling of tissue optical properties,” Proc SPIE 4001,
30–53 (2000).
189. J. Lademann, H.-J. Weigmann, W. Sterry, and V. Tuchin et al.,
“Analysis of the penetration process of drugs and cosmetic products into
the skin by tape stripping in combination with spectroscopic measure-
ments,” Proc SPIE 3915, 194–201 (2000).
190. P. L. Walling and J. M. Dabney, “Moisture in skin by near-infrared
reflectance spectroscopy,” J. Soc. Cosmet. Chem. 40, 151–171 (1989).
191. K. A. Martin, “Direct measurement of moisture in skin by NIR
spectroscopy,” J. Soc. Cosmet. Chem. 44, 249–261 (1993).
192. K. Wichrowski, G. Sore, and A. Khaiat, “Use of infrared spectroscopy
for in vivo measurement of the stratum corneum moisturization after
application of cosmetic preparations,” Int. J. Cosmet. Sci. 17, 1–11
(1995).
193. G. W. Lucassen, G. N. A. van Veen, and J. A. J. Jansen, “Band analysis
of hydrated human skin stratum corneum Attenuated Total Reflectance
Fourier Transform Infrared Spectra in vivo,” J. Biomed. Opt. 3, 267–280

(1998).
194. J. M. Schmitt, J. Hua, and J. Qu,” Imaging water absorption with
OCT,” Proc SPIE 3598, 36–46 (1999).
195. A. N. Bashkatov, E. A. Genina, and I. V. Korovina et al., “In vivo and
in vitro study of control of rat skin optical properties by acting of
osmotical liquid,” Proc. SPIE 4224, 300–311 (2000).
196. H. J. S. M. Sterenborg and J. C. Van der Leun, “Change in epidermal
transmission due to UV–induced hyperplasia in hairless mice: a first
approximation of the action spectrum,” Photodermatology 5, 71–82
(1988).
197. V. V. Tuchin, “Optical clearing of tissue and blood using immersion
method,” J. Phys. D: Appl. Phys. 38, 2497–2518 (2005).
198. V. V. Tuchin, “Optical immersion as a new tool to control optical
properties of tissues and blood,” Laser Phys. 15(8), 1109–1136 (2005)
199. V. V. Tuchin, Optical Clearing of Tissues and Blood, PM 154, SPIE
Press, Bellingham, WA, (2006).
200. V. V. Tuchin, A clear vision for laser diagnostics, IEEE J. Select. Top.
Quant. Electron. 13(6), 1621–1628 (2007).
201. Special Section on Optical Clearing of Tissues and Cells, V. V. Tuchin,
R. K. Wang, and A. T. Yeh (eds.), J. Biomed. Opt. 13(2), 021101-1
(2008)
202. H. Schaefer and T. E. Redelmeier, Skin Barier: Principles of Percutane-
ous Absorption, Karger, Basel et al. (1996).
203. A. Kotyk and K. Janaček, Membrane Transport: an Interdisciplinary
Approach, Plenum Press, New York (1977).
204. V. V. Tuchin, I. L. Maksimova, D. A. Zimnyakov, I. L. Kon, A. H.
Mavlutov, and A. A. Mishin, “Light propagation in tissues with
controlled optical properties,” J. Biomed. Opt. 2, 401–417 (1997).
205. A. N. Bashkatov, E. A. Genina, I. V. Korovina, Yu. P. Sinichkin, O. V.
Novikova, and V. V. Tuchin, “In vivo and in vitro study of control of rat
skin optical properties by action of 40%-glucose solution,” Proc SPIE
4241, 223–230 (2001).
206. B. Choi, L. Tsu, E. Chen, T. S. Ishak, S. M. Iskandar, S. Chess, and J. S.
Nelson, “Determination of chemical agent optical clearing potential
using in vitro human skin,” Laser. Surg. Med. 36, 72–75 (2005).
207. J. Jiang and R. K. Wang, “Comparing the synergetic effects of oleic acid
and dimethyl sulfoxide as vehicles for optical clearing of skin tissue in
vitro,” Phys. Med. Biol. 49, 5283–5294 (2004).
208. V. V. Tuchin, G. B. Altshuler, A. A. Gavrilova, A. B. Pravdin,
D. Tabatadze, J. Childs, and I. V. Yaroslavsky, “Optical clearing of skin
using flashlamp-induced enhancement of epidermal permeability,”
Laser. Surg. Med. 38, 824–836 (2006).
186 Chapter 3
209. S. P. Chernova, N. V. Kuznetsova, A. B. Pravdin, and V. V. Tuchin,

“Dynamics of optical clearing of human skin in vivo,” Proc. SPIE 4162,
227–235 (2000).
210. E. I. Galanzha, V. V. Tuchin, A. V. Solovieva, T. V. Stepanova, Q. Luo,
and H. Cheng, “Skin backreflectance and microvascular system
functioning at the action of osmotic agents,” J. Phys. D: Appl. Phys.
36, 1739–1746 (2003).
211. G. Vargas, A. Readinger, S. S. Dosier, and A. J. Welch, “Morphological
changes in blood vessels produced by hyperosmotic agents and measured
by optical coherence tomography,” Photochem. Photobiol. 77(5), 541–
549 (2003).
212. V. V. Tuchin, A. N. Bashkatov, E. A. Genina, Yu. P. Sinichkin, and
N. A. Lakodina, “In vivo investigation of the immersion-liquid-induced
human skin clearing dynamics,” Techn. Phys. Lett. 27(6), 489–490
(2001).
213. R. K. Wang, X. Xu, V. V. Tuchin, and J. B. Elder, “Concurrent
enhancement of imaging depth and contrast for optical coherence
tomography by hyperosmotic agents,” J. Opt. Soc. Am. B 18, 948–953
(2001).
214. R. O. Esenaliev, K. V. Larin, I. V. Larina, and M. Motamedi,
“Noninvasive monitoring of glucose concentration with optical coherent
tomography,” Opt. Lett. 26(13), 992–994 (2001).
215. V. V. Tuchin (ed.), Handbook of Optical Sensing of Glucose in Biological
Fluids, and Tissues, CRC Press, Taylor & Francis Group, London, 2009.
216. I. V. Meglinsky, A. N. Bashkatov, E. A. Genina, D. Yu. Churmakov,
and V. V. Tuchin, “Study of the possibility of increasing the probing
depth by the method of reflection confocal microscopy upon immersion
clearing of near-surface human skin layers,” Laser Phys. 13(1), 65–69
(2003).
217. G. Vargas, K. F. Chan, S. L. Thomsen, and A. J. Welch, “Use of
osmotically active agents to alter optical properties of tissue: Effects on
the detected fluorescence signal measured through skin,” Laser. Surg.
Med. 29, 213–220 (2001).
218. Y. He, R. K. Wang, and D. Xing, “Enhanced sensitivity and spatial
resolution for in vivo imaging with low-level light-emitting probes by use
of biocompatible chemical agents,” Opt. Lett. 28(21), 2076–2078 (2003).
219. K. Moulton, F. Lovell, E. Williams, P. Ryan, D. C. Lay, Jr., D. Jansen,
and S. Willard, “Use of glycerol as an optical clearing agent for
enhancing photonic transference and detection of salmonella typhimur-
ium through porcine skin,” J. Biomed. Opt. 11(5), 054027-1–8 (2006).
220. E. D. Jansen, P. M. Pickett, M. A. Mackanos, and J. Virostko, “Effect
of optical tissue clearing on spatial resolution and sensitivity of
bioluminescence imaging,” J. Biomed. Opt. 11(4), 041119-1–7 (2006).
221. M. V. Shulmerich, K. A. Dooley, M. D. Morris, T. M. Vanasse, and S.

A. Goldstein, “Transcutaneous fiber optic Raman spectroscopy of bone
using annular illumination and a circular array of collection fibers,”
J. Biomed. Opt. 11(6), 060502-1–3 (2006).
222. V. V. Tuchin and A. B. Pravdin, “Dynamics of skin diffuse reflectance
and autofluorescence at tissue optical immersion,” in Materials on
European Workshop “BioPhotonics 2002,” October 18-20, 2002,
Heraklion, Crete, Foundation for Research and Technology–Hellas,
Heraklion, CD–edition.
223. R. Cicchi, F. S. Pavone, D. Massi, and D. D. Sampson, “Contrast
and depth enhacement in two-photon microscopy of human skin
ex vivo by use of optical clearing agents,” Opt. Express 13, 2337–2344
(2005).
224. A. T. Yeh, B. Choi, J. S. Nelson, and B. J. Tromberg, “Reversible
dissociation of collagen in tissues,” J. Invest. Dermatol. 121, 1332–1335
(2003).
225. T. Yasui, Y. Tohno, and T. Araki, “Characterization of collagen
orientation in human dermis by two-dimensional second-harmonic-
generation polarimetry,” J. Biomed. Opt. 9(2), 259–264 (2004).
226. S. Plotnikov, V. Juneja, A. B. Isaacson, W. A. Mohler, and
P. Campagnola, “Optical clearing for improved contrast in second
harmonic generation imaging of skeletal muscle,” Biophys. J. 90, 328–
339 (2006).
227. L. Heinemann, U. Kramer, H. M. Klotzer, M. Hein, D. Volz,
M. Hermann, T. Heise, and K. Rave, “Non-invasive task force:
noninvasive glucose measurement by monitoring of scattering coefficient
during oral glucose tolerance tests,” Diabet. Technol. Ther. 2, 211–220
(2000).
228. S. Tanev, V. V. Tuchin, and P. Paddon, “Cell membrane and gold
nanoparticles effects on optical immersion experiments with normal and
cancerous cells: fdtd modeling,” J. Biomed. Opt. 11(6), 025606-1–6
(2006).
229. E. A. Genina, A. N. Bashkatov, V. V. Tuchin, G. B. Altshuler, and I. V.
Yaroslavski, “Possibility of increasing the efficiency of laser-induced
tattoo removal by optical skin clearing,” Quant. Electron. 38(6), 580–587
(2008).
230. M. G. Ghosn, N. Sudheendran, M. Wendt, A. Glasser, V. V. Tuchin,
and K. V. Larin, “Monitoring of glucose permeability in monkey skin in
vivo using Optical Coherence Tomography,” J. Biophoton. 3(1-2), 25–33
(2010).
231. I. A. Nakhaeva, O. A. Zyuryukina, M. R. Mohammed, and Yu. P.
Sinichkin, “The effect of external mechanical compression on in vivo
water content in human skin,” Opt. Spectrosc. 118(5), 834–840 (2015).
188 Chapter 3
232. I. H. Blank, J. Moloney, and A. G. Emslie et al., “The diffusion of water

across the stratum corneum as a function of its water content,” J. Invest.
Dermatol. 82, 188–194 (1984).
233. U. Jacobi, E. Waibler, W. Sterry, and J. Lademann, “In vivo
determination of the long-term reservoir of the horny layer using laser
scanning microscopy,” Laser Phys. 15, 565–569 (2005).
234. E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, V. I. Kochubey, N. A.
Lakodina, G. B. Altshuler, and V. V. Tuchin, “In vitro and in vivo study
of dye diffusion into the human skin and hair follicles,” J. Biomed. Opt.
7, 471–477 (2002).
235. C. L. Gay, R. H. Guy, G. M. Golden, V. H. W. Mak, and M. L.
Francoeur, “Characterization of low-temperature (i.e., ,65°C) lipid
transitions in human stratum corneum,” J. Invest. Dermatol. 104, 233–
239 (1994).
236. M. Haruna, K. Yoden, M. Ohmi, and A. Seiyama, “Detection of phase
transition of a biological membrane by precise refractive index
measurement based on the low coherence interferometry,” Proc. SPIE,
188–193 (2000).
237. M. A. Fox, D. G. Diven, K. Sra, A. Boretsky, T. Poonawalla,
A. Readinger, M. Motamedi, and R. J. McNichols, “Dermal scatter
reduction in human skin: A method using controlled application of
glycerol,” Laser. Surg. Med. 41, 251–255 (2009).
238. V. V. Tuchin, D. Tabatadze, J. Childs, I. Yaroslavsky, A. B. Pravdin,
A. A. Gavrilova, and G. Altshuler, “Optical clearing of skin using
flashlamp-induced permeability enhancement for accelerated delivery of
index-matching agents,” Laser. Surg. Med., Suppl. 17, 5 (March 2005).
239. A. Matsui, S. J. Lomnes, and J. V. Frangioni, “Optical clearing of the
skin for near-infrared fluorescence image-guided surgery,” J. Biomed.
Opt. 14(2), 024019-1–9 (2009).
240. A. N. Bashkatov, E. A. Genina, V. V. Tuchin, and G. B. Altshuler,
“Skin optical clearing for improvement of laser tattoo removal,” Laser
Phys. 19(5), (2009).
241. A. A. Gavrilova, V. V. Tuchin, A. B. Pravdin, I. V. Yaroslavsky, and
G. B. Altshuler, “Skin spectrophotometry under the islet photothermal
effect on the epidermal permeability,” Opt. Spectrosc. 104(1), © Pleiades
Publishing, Ltd., 140–146 (2008).
242. E. A. Genina, A. N. Bashkatov, A. A. Korobko, E. A. Zubkova, V. V.
Tuchin, I. V. Yaroslavsky, and G. B. Altshuler, “Optical clearing of
human skin: comparative study of permeability and dehydration of
intact and photothermally perforated skin,” J. Biomed. Opt. 13(2),
021102 (2008).
243. K. V. Larin, M. G. Ghosn, A. N. Bashkatov, E. A. Genina, N. A.
Trunina, and V. V. Tuchin, “Optical clearing for OCT image
enhancement and in-depth monitoring of molecular diffusion,” IEEE J.

Select. Tops. Quant. Electron. 18(3) 1244–1259 (2012).
244. O. Nadiarnykh and P. J. Campagnola, “SHG and optical clearing,” in
Second Harmonic Generation Imaging, F. S. Pavone and P. J.
Campagnola (Eds.), CRC Press, Taylor & Francis Group, Boca Raton,
London, NY, 169–189 (2014).
245. E. A. Genina, A. N. Bashkatov, and V. V. Tuchin, “Tissue optical
immersion clearing,” Expert Rev. Med. Devices 7(6), 825–842 (2010).
246. E. A. Genina, A. N. Bashkatov, K. V. Larin, and V. V. Tuchin, “Light–
tissue interaction at optical clearing,” in Laser Imaging and Manipulation
in Cell Biology, F. S. Pavone (ed.), Wiley-VCH Verlag GmbH & Co.
KGaA, Weinheim, 115–164 (2010).
247. D. Zhu, K. V. Larin, Q. Luo, and V. V. Tuchin, “Recent progress in
tissue optical clearing,” Laser Photonics Rev. 7(5), 732–757 (2013).
248. R. K. Wang and V. V. Tuchin, “Optical coherence tomography: light
scattering and imaging enhancement,” in Coherent-Domain Optical
Methods: Biomedical Diagnostics, Environmental Monitoring and Mate-
rial Science, 2nd ed., V. V. Tuchin (ed.), Springer-Verlag, Berlin,
Heidelberg, NY, 665–742 (2013).
249. D. Zhu, Q. Luo, and V. V. Tuchin, “Tissue Optical Clearing,” in
Advanced Biophotonics: Tissue Optical Sectioning, R. K. Wang and V. V.
Tuchin (Eds.), CRC Press, Taylor & Francis Group, Boca Raton,
London, NY, 621–672 (2013).
250. V. Hovhannisyan, P.-S. Hu, S.-J. Chen, C.-S. Kim, and C.-Y. Dong,
“Elucidation of the mechanisms of optical clearing in collagen tissue
with multiphoton imaging,” J. Biomed. Opt. 18(4), 046004-1–8 (2013).
251. J. Wang, Y. Zhang, P. Li, Q. Luo, and D. Zhu, “Review: Tissue Optical
Clearing Window for Blood Flow Monitoring (Invited Paper),” IEEE J.
Select. Tops. Quant. Electron. 20(2), 6801112-1–12, 2014.
252. E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, I. Yu. Yanina, and
V. V. Tuchin, Optical clearing of biological tissues: prospects of
application in medical diagnostics and phototherapy [Review],
J. Biomedical Photonics & Eng. 1(1), 22–58 (2015).
Yury P. Sinichkin is a full professor and Chair of Optics and

Biophotonics at Saratov State University. He is head of the
laboratory of Optical Medical Diagnostics of the Research-
Educational Institute of Optics and Biophotonics of SSU and
senior researcher of the Interdisciplinary Laboratory of
Biophotonics at Tomsk State University. His research
interests include biophotonics, tissue optics, spectroscopy,
medical optical diagnostics, and laser medicine. He is a
member of SPIE and IEEE.
190 Chapter 3
Sergey R. Utz is a professor and director of the Dermatolog-

ical Department at Saratov State Medical University. His
research interests include medical optical diagnostics, laser
therapy and photodermatology.
Valery V. Tuchin is a professor and chairman of Optics and

Biophotonics at Saratov National Research State University.
He is also the head of laboratory at the Institute of Precision
Mechanics and Control, RAS, and the supervisor of
Interdisciplinary laboratory of Biophotonics at Tomsk
National Research State University. His research interests
include biophotonics, tissue optics, laser medicine, tissue
optical clearing, and nanobiophotonics. He is a member of
SPIE, OSA, and IEEE, Guest Professor of HUST (Wuhan) and Tianjin
Universities of China, and Adjunct Professor of the Limerick University
(Ireland) and National University of Ireland (Galway). He is a fellow of SPIE
and OSA, and has been awarded Honored Science Worker of the Russia,
SPIE Educator Award, FiDiPro (Finland), Chime Bell Prize of Hubei
Province (China), and Joseph W. Goodman Book Writing Award (OSA/
SPIE).
Chapter 4
Infrared and Raman
Spectroscopy of Human
Skin in vivo
Gerald W. Lucassen
Philips Research, Eindhoven, The Netherlands
Peter J. Caspers and Gerwin J. Puppels

Erasmus University Rotterdam, Rotterdam, The Netherlands
Maxim E. Darvin and Juergen Lademann

Center of Experimental and Applied Cutaneous Physiology (CCP),
Department of Dermatology, University Clinic Charité, Berlin, Germany
4.1 Introduction: Basic Principles of IR and Raman

Spectrosopy
Infrared (IR) and Raman spectroscopy are complementary optical methods
that provide information about the molecular composition, molecular
structures, and molecular interactions in a sample. There exist many books
on the theoretical and experimental aspects of IR and Raman spectroscopy
(see, e.g., Refs. 1, 2, 3 and references therein). Here the methods are only
briefly described.
In both methods, vibrational energy states in molecules are excited by
light. In Fig. 4.1 the IR and Raman processes are depicted in a molecular
energy level diagram. In IR spectroscopy, infrared light from a broadband
source (usually 2.5–25 mm or 4000–400 cm–1) is directly absorbed to excite the
molecules to higher vibrational states. In IR spectra, absorbances are plotted
as a function of wave number (in cm–1), which is the reciprocal of wavelength
(in cm). The absorbance A is equal to log(I0/I), with I0 the incident intensity,
and I the intensity after absorption. When the absorbed energy hn matches the
191
192 Chapter 4
Figure 4.1 Illustration of an IR absorption process and a Raman scattering process in a

molecular energy level diagram.
energy needed for an allowed infrared excitation of a molecular vibration, an

absorption peak is observed in the IR spectrum.
In a Raman scattering event, light is inelastically scattered by a molecule.
In the interaction between the photon and the molecule a small amount of
energy is transferred from the photon to the molecule (or vice versa). This
leads to an excitation of the molecule from its lowest vibrational energy level
in the electronic ground state S0 to a higher vibrational state (in the electronic
ground state). The energy difference between the incident and scattered
photon is expressed in a wave number shift s: (s ¼ 1/ l0 1/l).
If the energy of the Raman-scattered photons is lower than the energy of
the incident photons, the process is called Stokes-Raman scattering. If a
photon interacts with a molecule in a higher vibrational level, anti-Stokes-
Raman scattering can occur, in which the energy of the Raman-scattered
photons is higher than the energy of the incident photons (see Fig. 4.1). The
intensity ratio of the anti-Stokes and Stokes Raman lines for a given vibration
is given by
I anti-Stokes ðn0 þ nvib Þ4

¼ · expðhnvib ∕kTÞ, (4.1)
I Stokes ðn0 nvib Þ4
where n0 is the frequency of incident radiation and nvib is the frequency of the
excited vibrational mode. It follows that at room temperature the intensity of
the Stokes-Raman lines in the most informative spectral region (.400 cm–1) is
much higher than that of the anti-Stokes-Raman lines.
Different selection rules apply for excitation of molecular vibrational
states through absorption of an IR photon or through Raman scattering of an
incident photon. Some vibrations can be excited by both Raman and IR
processes, others can only be excited by either a Raman scattering process or
by IR absorption. For symmetric molecules, the selection rules are mutually
Infrared and Raman Spectroscopy of Human Skin in vivo 193
exclusive for all vibrations. Molecules exhibit IR activity when a change in the
molecular dipole moment occurs during the vibration. Raman activity
occurs when there is a change in polarizability. Therefore, the band intensity
in IR and Raman spectra of the same molecular vibration frequency can be
quite different. As a rule of thumb, symmetric vibrational modes are often
strong in Raman, whereas antisymmetric vibrational modes are strong
in IR. Depending on the polarization state of the incident and analyzed
(observed) light, information on the symmetry of the molecules can also be
obtained.
The subject of this chapter is to analyze IR and Raman vibrational
characteristics of human skin in vivo. In Sec. 4.2, IR spectra of human skin
stratum corneum, of water, and of hydrated skin are analyzed. Section 4.3
deals with the analysis of Raman spectra of human skin constituents
and of water concentration profiles in human skin. Finally, Sec. 4.4 presents
conclusions and outlook for clinical applications of IR and Raman spectro-
scopic diagnostics.
4.2 Fourier Transform Infrared Spectroscopy of Human-Skin

Stratum Corneum in vivo
Noninvasive measurements of water content in skin are important in
dermatology, pharmacology, cosmetics, and the medical sciences for the
monitoring of skin condition and to attain understanding of skin hydration. A
number of techniques4,5 have been applied to measure water in skin, each with
its own resolution, sensitivity, and depth range.
In the 1960s and 1970s, Puttnam et al.6 and Hansen and Yellin7,8 studied
hydration of the skin stratum corneum (SC) using attenuated total reflectance
(ATR), nuclear magnetic resonance, and infrared (IR) spectroscopy. From in
vitro studies, they found three types of water in the SC with different hydrogen
bonding strengths, mobilities, and sorption and desorption kinetics.
These three water types in skin are sometimes referred to as primary
hydration, secondary hydration, and bulk water. At concentrations below
10% (w/w) (water weight to dry weight SC) hydration, water is tightly bound
to the polar sites of the SC keratin proteins (primary hydration). At water
concentrations from 10–40%, less tightly bound water is found, probably
hydrogen bonded to primary water (secondary hydration). At higher concen-
trations in the SC, water behaves more like bulk liquid water. Near-infrared in
vivo spectroscopic studies by Martin9 showed evidence of the three types of
water mentioned above. Martin also introduces water associated to the lipid
phase within the SC, which may be responsible for the evaporative flux across
the SC.
A summary of studies on the effect of humidity on both strength and
number of water binding sites in the SC can be found in Potts.4 A number of
194 Chapter 4
water peaks in the infrared spectra is attributed to different SC water binding

sites.
Potts et al.10 quantitatively determined the water content in the stratum
corneum from combined in vitro and in vivo experiments using attenuated
total-reflectance Fourier-transform infrared spectroscopy (ATR-FTIR). By
this method, Bommannan et al.11 studied the barrier function of the SC
in vivo. Boddé et al.12 and Wichrowski et al.13 applied ATR-FTIR to measure
hydration of the skin applying cosmetic moisturizers using ATR-FTIR.
In this paragraph an ATR-FTIR spectroscopic study on the vibrational
characteristics of normal and hydrated in vivo skin SC14 is described. The
measuring depth of ATR-FTIR in the skin typically is a few microns over the
wave number (n ¼ 1/l) window 4000–650 cm–1. This means that ATR-FTIR
enables one to monitor changes in water content of the outer SC layers only.
Gloor et al.15 obtained the water content from the “moisturizing factor” (MF)
defined as the amide I to amide II band ratio. It was found that this ratio
could be used to measure the water content qualitatively. However, as Potts
pointed out,10 the amide I and amide II bands vary in intensity differently
with increasing water content. Potts10 uses the combination band at 2100 cm–1
to determine the water content. The area under the curve above the line
connecting the spectral intensity at 1900 cm–1 and 2300 cm–1 is considered a
measure of the water content, whereas the area under this baseline is
considered a measure of contact area. The ratio of these two areas is called the
infrared absorption ratio. From a comparison of infrared absorption ratios
obtained in vivo with those measured in vitro, quantitative information on
hydration is obtained. Another measure of water content of stratum corneum
hydration13 is obtained from the amplitude or integrated band intensity of the
OH stretch region around 3300 cm–1. This broadband contains several types
of OH stretch vibration bands.
The three methods mentioned above are of great practical use, since they
can be applied directly using the measured absorbances. However, there are
some problems with each of the methods. First, the amide I and amide II band
absorbance maxima do not always correspond precisely with the actual band
frequency positions in the spectrum. Besides, the moisturizing factor, i.e., the
ratio of the amide I to amide II band, depends on different water band
contributions, as will become clear later. Second, bandshape changes around
the 2100 cm–1 combination band at low water concentrations cause a problem
in determining the actual water content. Third, using integrated intensity in
the OH stretch band region at 3300 cm–1, the situation is hindered by the
presence of an NH-stretch band that varies with water content.
These problems can be circumvented using the spectral band information
of skin and water known from the IR and Raman literature and using direct
band fit analysis of the spectra. This has the advantage of using the entire
spectral range, covering the three regions of interest by the methods
mentioned above. Furthermore, the water band fit parameters (amplitude,

frequency, and bandwidth) directly relate to the water content in the skin SC.
The disadvantage of the fit routine is that it is more complicated to use and
one has to guess the initial parameters of unknown or unassigned bands.
The purpose of this work is to use spectral fit analysis to determine the
contribution of water band amplitudes in the skin SC spectrum. Band
frequencies known from literature are given fixed values in the fits; all other
unknowns are free parameters.
It is shown that the fit method yields detailed information on skin SC and
water band contributions enabling determination of water content. The results
are compared with the MF and IR absorbance ratio methods used by Gloor
and Potts, respectively.
4.2.1 Experimental ATR-FTIR setup

A dry air flushed Nicolet-800 Fourier transform spectrometer with a “high-
top” model ATR with a ZnSe (n ¼ 2.42) or Ge (n ¼ 4.0) crystal of rectangular
shape (10 mm 80 mm) with 45 deg entrance and exit facets is used to record
the ATR-FTIR spectra (see Fig. 4.2). The shape of the crystal limits the
number of reflections inside the crystal to 10. The ATR-FTIR signals are
measured by a liquid N2 cooled Mercury Cadmium Telluride (MCT) detector.
Typically, the ATR-FTIR spectrum is obtained by the Fourier transform of
64 or 128 co-added interferograms, where the Happ-Genzel apodization is
used. The acquisition time for 64 scans at a resolution of 8 cm–1 is about 20 s.
Spectra are recorded in the 4000–650 cm–1 range. Files are converted to
absorbance and stored to disk for further processing and displaying. Water
spectra are recorded from distilled water using a Ge (n ¼ 4.0) crystal instead
of the ZnSe crystal, which gives a limited penetration of the evanescent wave
and prevents signal saturation in the higher wave number range. Skin SC
spectra are recorded on the volar aspect of the forearm by slight pressure on
the ZnSe crystal. Before the measurements, the skin is cleaned by using a 1%
solution of sodium lauryl sulphate in water, rinsing with water, and drying
with paper towels. Hydration of the skin is obtained by occlusion keeping the
forearm on the crystal for about 30 minutes while recording a spectrum each
minute.
Figure 4.2 Simplified schematic of the ATR-FTIR setup. The infrared light is (totally)
reflected inside the ZnSe crystal (n ¼ 2.4). At the skin-crystal interface, part of the light is
absorbed in the skin, which attenuates the beam inside the crystal. An infrared (MCT)
detector picks up the transmitted beam.
196 Chapter 4
0.8
2919 ν CH2 asymm

2850 ν CH2 symmetric
2957νCH3 asymmetric
0.6
3275 νNH
3220 νds OH
Absorbance
3420 νb OH
0.4
νCH ?
2125 δ + ν L H2O
2873 ν CH3 symm
3615 νw OH
0.2
0.0
4000 3600 3200 2800 2400 2000
Wavenumber (cm-1)
1650 AmI
1118 ν CC skeletal trans

1077 ν CC skeletal trans
1515 sh
1035 ν CC skeletal cis

1389 ν CH3 symm
1.0
1403 δC(CH3)2
1585 ν C=C
1640 δ H2O
852 δCCH aromatic

1545 Am II
1245 Am III
1460 δCH2
1455 δCH3
1298 ν CH2
Absorbance
1208
1677 sh
1337
1164
883 ρCH2
917
975
0.5
1740 ν C=O lipid
0.0
1600 1400 1200 1000 800
Wavenumber (cm-1)
Figure 4.3 Measured ATR-FTIR spectrum of in vivo human skin stratum corneum on the
volar aspect of the forearm: (top) the 4000–1750 cm–1 spectral range, (bottom) the 1750–
750cm–1 spectral range. Assignments are indicated in the figure (see also Table 4.1).
4.2.2 Human-skin stratum corneum spectra and band assignments

A typical ATR-FTIR spectrum of untreated in vivo skin SC is displayed in
Fig. 4.3. Band assignments as known from the literature16,17,18 are given in
Table 4.1. Dominant bands are due to the amides at 1650 cm–1 (amide I, C¼O
stretching in O¼C-N-H) and 1545 cm–1 (amide II, N-H bending in O¼C-N-H).
The amide parts of the spectrum carry structural information on the
Table 4.1 FTIR frequencies (cm–1) and assignments of vibrational bands of human stratum
corneum according to Barry et al.18a and Scherer et al.16,17b and frequencies found in our
study (Fig. 4.3).
Frequency (cm–1) Frequency (cm–1) Assignment
Barrya, Schererb This study
3615b vw 3615 vw nw (OH) weak bond of H2O

3420b ms 3420 ms nb (OH) strong bond of H2O
3287a vs, br — n (OH) of H2O
3300a 3275 ms n (NH)
3220b ms 3220 ms nsd (OH) symmetric of H2O
3070a w — 1st overtone amide II at 1548 cm–1
— 3050 w n(CH)?
2957a w 2957 w n(CH3) asymmetric
2919a vs 2919 vs n(CH2) asymmetric
2873a w 2873 w n(CH3) symmetric
2851a s 2851 s n(CH2) symmetric
2125b m 2125 m d þ nL (H2O) combination
1743a w 1740 w n(C=O) lipid
— 1677 sh shoulder
1656a vs — n(C=O) amide I disordered
1650a vs 1650 vs n(C=O) amide I a-helix
— 1640 vs d(H2O) bending
— 1585 w n(C=C) olefinic?
1548a vs 1545 s d(NH) and n (CN) amide II
— 1520 m water combination band
1515a w,sh 1515 m,sh shoulder
1460a vw 1460 m,sh d(CH2)
1451a vw 1455 m d(CH3) asymmetric
1440a vw — d(CH2) scissoring
1401a w 1403 m d[C(CH3)2] symmetric
1389a vw 1389 sh d(CH3) symmetric
1366a vw — d[C(CH3)2] symmetric
— 1337 vw ?
1298a w 1298 w d(CH2)
1247a w 1245 m d(CH2) wagging; n(CN) amide III disordered
— 1164 w n(CC), d(COH)
— 1118 m n(CC) skeletal trans conformation
1076a w 1077 m n(CC) skeletal trans conformation
— 1035 m n(CC) skeletal cis conformation
— 975 vw ?
— 917 vw ?
— 883 vw r(CH2)
— 852 vw d(CCH) aromatic
v¼very; s¼strong; m¼medium; w¼weak; sh¼shoulder; br¼broad; d¼deformation; n ¼stretch; r¼rock
backbone of proteins (keratins) and lipids (ceramides). Human stratum

corneum consists of 75–80% weight of keratin proteins (a-keratin 70%,
b-keratin 10%) and 5–15% weight of lipids (ceramides, free fatty acids,
cholesterol, cholesterol sulfate, etc.). From this it can be inferred that the
protein:lipid contribution to the amide amplitudes is about 8:1. The C¼O
stretching band of lipid ester carbonyl at 1740 cm–1 is indicative of the
198 Chapter 4
presence of sebum in the SC. In spectra taken on the forehead, cheek, or neck,
two peaks with different amplitudes belonging to the two carbonyl groups in
lipid ester appear at 1726 cm–1 and 1740 cm–1 (spectra not shown). Generally,
the bending modes have smaller frequencies as compared to most stretching
frequencies. For example, the C-H bending of lipids is found at 1451 cm–1,
and C-H stretches in asymmetric and symmetric CH2 and CH3 modes of
lipids at 2851 cm–1 and 2919 cm–1, respectively. The broadband around
3300 cm–1 contains OH stretches and an NH stretch. In the normal SC
spectrum, water bands are present, indicating that the normal SC contains a
certain amount of water. These water bands will be described in more detail in
Sec. 4.2.3.
There are not many examples of ATR-FTIR spectra of skin SC in the
literature to compare with. The in vivo skin stratum corneum spectrum shown
in Fig. 4.3 agrees well with that shown by Potts et al. (Fig. 1 in Ref 10). In
comparison with the in vivo spectrum by Wichrowski et al.,13 it is apparent
that their spectrum contains a large double C¼O peak at 1720 and 1740 cm–1,
a situation we only observed in spectra measured on the cheek and forehead
(spectra not shown). The difference between in vitro and in vivo spectra is
obvious from a comparison with the FT-IR spectrum shown by Barry et al.
(Fig. 2 in Ref. 18). Barry et al. assigned the bands in the skin SC spectrum on
the basis of a comparison with in vitro FT-IR and FT-Raman spectra. In our
in vivo ATR-FTIR spectrum (see Fig. 4.3) we observe some bands in the lower
wave-number range that may correspond to some of the assignments of
Raman bands that are not visible in their IR spectrum. For example, the
852 cm–1 dCCH band and the 883 cm–1 rCH2 band, the nCC stretches at
1034 cm–1 and 1118 cm–1, and the 1168 cm–1 nCC stretch and dCOH bending.
Some very weak bands indicated in the figure by their wave numbers are not
assigned in the list of Barry et al., e.g., 917, 975, and 1337 cm–1. Barry et al.
assign two peaks at 1650 and 1656 cm–1 to a-helix and disordered amide I
nC¼O stretches of the carbonyl groups of keratin proteins and ceramide
lipids. We do not observe such a doublet. These bands are probably obscured
by the strong band with a maximum absorbance at 1640 cm–1. The weak
shoulder at 1677 cm–1, which becomes more visible in the derivative of the
spectrum in Fig. 4.3 (not shown), has not been assigned. It could point to a
partly b-sheet conformation of the amide I keratins.
4.2.3 ATR-FTIR spectrum of water

The water band structures have been intensively studied by Draegert et al.,19
Williams,20 and recently by Marechal21,22,23 using infrared spectroscopy, and
by Scherer et al.16,17 and Moskovits et al.24 using Raman spectroscopy. From
comparison of polarized Raman spectra, infrared spectra on H2O, D2O and
HDO mixtures and normal mode calculations by Curnutte et al.25,26,27 many
of the observed bands could be assigned. Still, the details of water at the
1.0
ν(OH)
νb νL
Absorbance
0.5 νds
δ(H2O)
δ-νL
δ+νL(H2O)
νw
0.0
4000 3500 3000 2500 2000 1500 1000
Wavenumber (cm-1)
Figure 4.4 Measured ATR-FTIR spectrum of water at room temperature (thick solid line)
and spectral fit (dashed line) built up from individual vibrational bands (thin solid lines). Most
important bands are the bending mode d(H2O) at 1640 cm–1, the combination band d + nL at
2124 cm–1, the symmetric nsd(OH) stretch at 3225 cm–1, the strong bond nb(OH) stretch at
3420 cm–1, and the weak nw(OH) stretch at 3615 cm–1.
molecular level are not completely understood. The high density of hydrogen
bonds and the quasi-tetragonal symmetry of water makes such a description
very complex.21,22,23,28,29
A measured ATR-FTIR spectrum (solid line) and fit (dashed line) of
water at room temperature are shown in Fig. 4.4. Water band assignments
from the literature and fit parameters are given in Table 4.2.
4.2.3.1 Water bending mode and low-wave-number region

At the low-wave-number edge of the spectrum, the wing of the “libration” band
nL is seen, associated with vibrations around the three rotational axes of the
water molecule. The sharp band at 1640 cm–1 (of the liquid water) corresponds
with the d(H2O) bending mode n2 of the isolated water molecule centered at
1595 cm–1 in vapor. Around 2125 cm–1 a weak band is seen that has been
assigned to the d þ nL combination band of the bending mode at 1640 cm–1
and the libration band nL. Similarly, the d nL combination band is expected
at the low-wave-number side of the 1640 cm–1 band around 1155 cm–1.
However, the latter band is not visible because it is obscured by a broad
feature between the libration band and the bending mode band. These
combinations point to a libration band frequency nL at 485 cm–1. From IR
measurements by Draegert et al.19 only an absorption maximum at
200 Chapter 4
Table 4.2 Fit parameters of the FTIR water spectrum given in Fig. 4.4: band frequency
(in cm–1), band width (in cm–1), amplitude (A in a.u.), and assignments after Scherer.16,17
Bandwidth Amplitude
Frequency (cm–1) (cm–1) Assignment (a.u.)
3615 vw 65 0.04 nw (OH) weak bond stretch

asymmetric di-H-bonded complex
3420 s 166 0.62 nb (OH) strong bond stretch
asymmetric di-H- bonded complex
3220 s 171 0.47 nsd (OH) stretch of symmetric
di-H-bonded complex
2128 w 190 0.04 d þ nL (H2O) combination of
1640 and libration band
1640 s 54 0.29 d (H2O) bending
1520 m 258 0.07 combination band
1150 m 350 0.09 d nL (H2O) combination
487 s 264 2.36 nL libration band
v¼very; s¼strong; m¼medium; w¼weak; d¼deformation; n¼stretch
685 cm–1 ± 15 cm–1 was found that was assigned to “hindered rotation” mode
nR. Also, a translational mode nT at 170 cm–1 ± 15 cm–1 could be identified.
Williams20 proposed that the 2125 cm–1 band would be a combination of
n2 þ nR nT, which is a transition starting from the first excited translational
state. However, the difference nR nT ¼ 515 cm–1 can only be matched to
485 cm–1 if the frequency accuracy ranges of the nR and nT bands are
coadded.
4.2.4 Stratum corneum hydration measurements

Hydration of the skin is obtained by keeping the forearm pressed against the
ATR-FTIR crystal. During occlusion, water in the skin cannot evaporate and
accumulates in the skin SC. A clear increase in the IR signal is observed
during the occlusion period of 30 minutes while recording a spectrum every
minute, see Fig. 4.5. The spectra are not scaled. The observed signal increase
is due to a) increased contact area between skin and crystal, and b) an increase
of water content in the skin SC. Upon hydration, the keratinocytes will be
plastified, which makes it easier to conform the uppermost keratine cells to the
crystal surface. Once this contact area is at a maximum, further signal increase
is due to an increase of water content in the SC. One can clearly see the
influence of the water bands in the skin stratum corneum spectrum.
4.2.4.1 OH stretch region

The literature on the assignments of the broad OH stretch bands of water is
scarce. On the basis of polarized Raman spectra, Scherer et al.16,17 propose a
model of water that classifies two species of hydrogen bonded water
complexes: one symmetric and one asymmetric water complex. In both
complexes, the oxygen of the central water molecule is bonded to two
2
time (min)
34
Absorbance
30
25
20
15
10
9
8
1 7
6
5
4
3
2
1
0
4000 3500 3000 2500 2000 1500 1000
Wavenumber (cm-1)
Figure 4.5 Sequential hydrated human skin stratum corneum spectra measured during
occlusion each minute for half an hour. The thick line represents the water spectrum (scaled)
from Fig. 4.4. Spectral changes can be clearly identified: increased contribution of the water
bending mode at 1640 cm–1 and pronounced increase of the OH stretches in the high wave
number band around 3300 cm–1. Also, the water combination band around 2125 cm–1 is
clearly visible in the hydrated spectra.
hydrogens of two other water molecules. In such complexes the hydrogen

bonds between the molecules can be loose or tight. The symmetric complex
has two strong H bonds, having a symmetric OH stretch nds at 3220 cm–1 and
an asymmetric OH stretch ndas at 3440 cm–1.16,17 The two complexes are IR
active, the first is Raman active in the anisotropic Raman spectrum, the latter
is absent in the isotropic Raman spectrum. The asymmetric water complex
has one strong H bonded OH stretch nb at 3425 cm–1 and one weak H bonded
OH stretch nw at 3615 cm–1. Also, the Fermi resonance of the bending
overtone 2n2 with the nds and nb OH stretches contributes to the broad OH
band region. In Raman spectra it has been shown that this Fermi resonance is
necessary to explain the bandshape of the broad OH stretch bands.16,17 In the
IR spectra, this effect is less pronounced and we were able to obtain good fits
without inclusion of the Fermi resonance. An important conclusion from the
work by Scherer et al.16,17 is that a considerable part of the water molecules is
suggested to have one hydrogen strongly bonded while the other hydrogen is
comparatively free at room temperature.
4.2.4.2 Fit on water spectrum
The simulation of the measured liquid water spectrum shown in Fig. 4.4 has
been achieved with as few as possible (three) water bands. Acceptable fits were
obtained neglecting Fermi resonances between the overtone 2n2 and the nOH
202 Chapter 4
stretches, and without the asymmetric nOH stretch at 3440 cm–1, which has
been taken together with the 3422 cm–1 nbOH stretch band as one band. If
more than three bands are used in the broad OH region it is always possible to
obtain a better fit.30 Some imperfections in the fit are seen in the wings of the
1640 cm–1 bending band and around 2750 cm–1 where intensity is lacking.
These may be due to the presence of another nearby water band or to an
actual lineshape different from a pure Gaussian.31
A Lorentzian component would give broader wings. Seshradi and Jones31
state that the lineshapes of infrared absorbance bands usually are a mixture of
Lorentzian and Gaussian bandshapes with a dominant Lorentzian contribu-
tion. In trying pure Lorentzians, we found that it was impossible to get
acceptable fits, especially in the high wave number wing of the 1640 cm–1
band where the intensity is rather low. As shown, using pure Gaussians, the
error between our measured spectrum and the fit is small and we, therefore,
have chosen to continue using pure Gaussian bandshapes, instead of
introducing three more parameters per band.
The two broadbands at 1150 and 1520 cm–1 are necessary to fit the broad
plateau between the libration band and the bending band. These bands are not
assigned by Scherer. Nonetheless, there are some possibilities of combination
bands that would add to this region.32 For example, the 1150 cm–1 band
matches exactly the d nL combination band. The n3 n2 nL, n2 þ n3 n1
and n1 n2 nL combination bands could possibly contribute to the other
band.
We did not find any information on the intensities of these combinations
in the literature, and we have not studied these further.
4.2.5 Band analysis of hydrated and normal skin

Spectral analysis is performed by band fitting based on a nonlinear least-
squares search using Gaussian band intensity shapes of the form
I i ðxÞ ¼ Ai expf½ðx Vi Þ∕Gi 2 g, (4.2)
where I represents the infrared absorbance, x the running wave-number

variable, Vi the frequency, Gi the width, and Ai the amplitude of the i’th
band. The program enables simultaneous fitting of a chosen number of bands
where band frequency, width, and amplitude can be selected to be “fixed” or
“free running” parameters. Band frequencies known from literature are
fixed. This means that at the time of the experiment, we neglected possible
frequency changes induced by hydration. All other parameters are free-
running parameters. Separate fits of normal skin SC spectra, hydrated
skin SC spectra, and of the pure water spectrum are performed. In the fit
routine, the widths and amplitudes of the skin SC and water bands are
treated as free parameters. The relative water content is obtained from the
2
Absorbance
4000 3000 2000 1000

Wavenumber (cm-1)
Figure 4.6 Fits (thick line) to the ATR-FTIR spectrum (thin line) of hydrated human
stratum corneum: using free water band amplitudes and bandwidths. The residue
(measured fit) is on the same scale, but shifted from zero for better visibility. Parameters
in Table 4.3.
amplitudes in the three water band regions of interest. Good fits are
obtained by interactive control of parameters looking at least squares values
of the fits.
An example of a fit to the fully hydrated SC spectrum of Fig. 4.5 is
presented in Fig. 4.6. Parameters of the fits are listed in Table 4.3. Following
Scherer, we have fitted the broad OH band of water using the symmetric
stretch nsd(OH) at 3220 cm–1, the strong bond nb (OH) stretch at 3422 cm–1,
and the weak bond nw (OH) stretch at 3615 cm–1. The nasd (OH) at 3440 cm–1
has been merged with the strong bond nb (OH) stretch at 3420 cm–1
because these bands are close together. The amplitudes of some of the water
bands (fit results) versus time are plotted in Fig. 4.7. The amplitudes of the
bending mode at 1640 cm–1 and the combination band at 2125 cm–1 are seen
to “saturate,” while the OH stretch amplitude continuously increases with
time. The shapes of the combination band curve and that of the bending
mode and the first part of the OH amplitude-time curve turn out to be
very similar. The accuracy has been estimated from the band amplitudes.
In Fig. 4.7 for example, the relative error in the 1640 cm–1 band amplitude is
less than 2%, which can be inferred from the saturated part after 20 minutes.
The relative error increases for lower signals, e.g., 5% for the 2125 cm–1 band.
The sizes of the symbols in the figure are an indication of the errors.
The spectra in Fig. 4.5 clearly show the increase in infrared signal intensity
with time in the amide regions and in the OH region. The thick line
204 Chapter 4
Table 4.3 Fit parameters of the fits on hydrated stratum corneum spectra of Fig. 4.5: band
frequency (in cm–1) and amplitude (in a.u.) versus time (min).
time (min) Amplitudes (a.u.)
freq.
(cm–1) 1 5 10 15 20 25 30 34
3615 0.000 0.002 0.039 0.059 0.074 0.092 0.100 0.115

3420 0.118 0.682 0.935 1.102 1.197 1.317 1.424 1.497
3275 0.078 0.176 0.196 0.192 0.182 0.157 0.138 0.125
3220 0.114 0.449 0.603 0.717 0.781 0.873 0.954 1.009
3050 0.126 0.345 0.264 0.282 0.309 0.340 0.373 0.396
2957 0.044 0.141 0.162 0.168 0.167 0.161 0.154 0.146
2950 0.127 0.134 0.245 0.250 0.234 0.209 0.171 0.147
2919 0.187 0.452 0.503 0.517 0.515 0.496 0.478 0.460
2873 0.062 0.119 0.132 0.134 0.132 0.121 0.114 0.107
2850 0.127 0.275 0.307 0.316 0.314 0.300 0.290 0.280
2125 0.045 0.114 0.136 0.137 0.137 0.144 0.138 0.137
1740 0.019 0.094 0.117 0.126 0.128 0.125 0.120 0.110
1650 0.137 0.474 0.583 0.642 0.658 0.642 0.652 0.629
1640 0.549 1.219 1.366 1.448 1.482 1.484 1.499 1.519
1585 0.415 0.885 0.965 1.002 1.005 0.953 0.933 0.888
1545 0.303 0.733 0.837 0.881 0.892 0.857 0.837 0.807
1520 0.115 0.277 0.304 0.307 0.311 0.338 0.324 0.315
1515 0.488 0.869 0.897 0.889 0.859 0.767 0.722 0.678
1455 0.257 0.491 0.519 0.528 0.519 0.462 0.448 0.427
1403 0.202 0.394 0.420 0.426 0.417 0.375 0.357 0.335
1337 0.130 0.262 0.278 0.282 0.275 0.243 0.231 0.216
1298 0.016 0.066 0.080 0.086 0.087 0.086 0.083 0.080
1245 0.125 0.273 0.296 0.300 0.294 0.272 0.256 0.240
1164 0.010 0.092 0.118 0.129 0.133 0.129 0.126 0.121
1150 0.473 0.743 0.746 0.735 0.718 0.690 0.664 0.643
1118 0.123 0.279 0.311 0.319 0.319 0.307 0.293 0.274
1077 0.101 0.208 0.229 0.236 0.236 0.226 0.215 0.201
1035 0.148 0.258 0.271 0.270 0.264 0.249 0.234 0.217
489 1.615 3.448 3.989 4.433 4.709 5.068 5.399 5.620
representing the water spectrum (not drawn to scale) is included for

comparison. Initially, the increase is due to the combined effect of water
content increase and enhanced contact area between skin and the crystal.
Hydration plastifies the upper keratin cells, which makes it easier for the skin
to conform to the crystal surface. After some time the contact area between
skin and crystal reaches a maximum and the further increase in signal
intensity is solely due to increased water content in the skin.
The spectral change in the combination band region around 2125 cm–1 is
remarkable. At either side of this band, the intensity decreases with time. Potts
assumed the band area underneath the 1900 and 2300 cm–1 intensity line to be
proportional to the contact area between skin and crystal. In our case, this
area decreases. However, this cannot be due to a decreasing contact area. The
spectral change in intensity is due to the combination of two effects:
1) increased contact area, and 2) increasing contribution from absorption of
1.5
δH2O 1640
Amplitude (a.u.)
1.0
νOH 3420
0.5
δ+νLH2O2125
0.0
0 5 10 15 20 25 30 35
Time (min)
Figure 4.7 Water band amplitudes from the fitting results versus time. Note that the
amplitudes of the bending mode at 1640 cm–1 and the combination band at 2125 cm–1
saturate while the OH stretch amplitude continuously increases with time.
water in the outermost layers of the skin SC, which, in fact, dilutes the skin SC
spectrum with that of water.
One might wonder whether a changed penetration depth with water
content would also add to the spectral changes observed. Hydration of the
outermost skin layers affects the refractive index of these layers since the
refractive index of dry SC is about 1.5533 and that of water is about 1.3334
Therefore, hydration of skin SC continuously decreases the refractive index of
skin SC.
4.2.5.1 Penetration depth of the IR beam
The penetration depth dp of the IR beam in the skin is given by35
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 1
n
d p ðnÞ ¼ 2pnnc sin2 ðuÞ s , (4.3)
nc
with u ¼ 45 deg the angle of incidence, ns the refractive index of the sample
(skin), nc the refractive index of the crystal (nc ¼ 2.42 for ZnSe), n ¼ 1/l the
wave number, and l the wavelength in vacuum. At the penetration depth, the
evanescent field amplitude has dropped to 1/e of its original value. According
to Eq. (2), the penetration depth is dp ¼ 1.317 mm at 650 cm–1 and dp ¼
0.214 mm at 4000 cm–1 for dry SC (using ns ¼ 1.55), and dp ¼ 1.21 mm at
650 cm–1 and dp ¼ 0.197 mm at 4000 cm–1 for water (ns ¼1.33), see Fig. 4.8.
206 Chapter 4
1.2
1.0
0.8
dp (μm)
0.6
0.4
0.2
0.0
4000 3500 3000 2500 2000 1500 1000
Wavenumber ν (cm-1)
Figure 4.8 Penetration depth dp of the IR beam as a function of the wave number
according to Eq. (4.2) for ZnSe crystal nc ¼ 2.42. The solid line represents a sample with
constant refractive index ns ¼ 1.33 (water) and the dashed line denotes a sample with
refractive index ns ¼ 1.55 of dry stratum corneum.
For the Ge crystal (nc ¼ 4.0) the penetration depths are about 50% lower than
those for the ZnSe crystal. If fully hydrated skin SC had a refractive index
close to 1.33, the penetration depth would only be 8% less than the
penetration depth in dry SC.
We tried to find evidence for the presence of a thin layer of water between
the skin and crystal, which we expected to be formed during occlusion. A
water film at the surface would show up in the spectrum immediately,
especially in the OH stretch region and by diminished lipid band
contributions. However, since the CH2 and CH3 bending bands of lipids do
not disappear relative to the OH band, we estimate that such a thin water film
would probably be less than 0.1 or 0.05 times the penetration depth of about
0.2 mm around 3300 cm–1, i.e., less than 20 nm thick.
4.2.5.2 Fits of the hydrated-skin stratum corneum spectra

The fits on the spectra in Fig. 4.7 have been obtained using as few bands as
possible for the skin SC and water. Therefore, not all the bands in Table 4.1
are used in the fit; some less important bands have been omitted. Some
broadbands were required to match the background. In the fit of Fig. 4.6 it is
seen that two such bands at 2950 and 3050 cm–1 are used. The first band is
underneath the symmetrical and asymmetrical CH2 and CH3 lipid stretch
bands. The broad plateau is also present in the spectrum measured on the
forehead skin SC (not shown) and contains a rather large amount of sebum.
The broad band might be attributed to a contribution of stretches of different

CC chain lengths of the ceramides present in the SC. The 3050 cm–1 band has
not been assigned in the list of Barry et al., which might be due to a CH
stretch.
Since the signal in the lower wave number part of the spectrum is built up
from deeper layers compared to the signal in the higher wave number region,
the total measured spectrum depends nonlinearly on the contributions from
different depths of the skin SC. A change in water content in a thin layer close
to the skin SC surface would contribute relatively more to spectral changes in
the higher wave-number region since the penetration depth there is small (see
Fig. 4.8).
The NH bending (amide II) band and C¼O band (amide I) are sensitive to
water by hydrogen bonding, as for example is the case of the hydration of
hairs where the water molecules stack in between C¼O and NH positions of
the a-helical keratin backbone.36 However, the apparent change in amplitudes
of both amides is due to the combination of changed amide I and amide II
bands, the increased water bending mode amplitude, and the plateau at lower
wave numbers (see Table 4.3).
From Fig. 4.7 it can be seen that the amplitudes obtained from the fit of
the water bending mode at 1640 cm–1 and the combination band at 2125 cm–1
saturate, while the OH band amplitudes continuously increase in time. The
different growth of the water band amplitudes with time also means that
the contribution from the water content increase is not homogenous over the
spectrum. This indicates that the actual water contribution to the SC spectrum
is skewed, and is due to the contributions from different depths and the actual
water concentration profile in skin SC depth.
The ratio of the weak stretch band amplitude (3615 cm–1) to the strong
stretch band amplitude (3440 cm–1) is fairly constant after five minutes. This
means that, in the time span measured here, after reaching maximal contact
area, we find no significant change in the contributions of filling from strong
H-bonded sites and weakly H-bonded sites in the SC.
The amplitude of the NH stretch at 3275 cm–1 exhibits a maximum in time
as can be inferred from Table 4.3. The first increase is again due to the
increased contact area between skin and crystal.
4.2.5.3 Comparison with MF and IR absorbance ratio

From the hydration spectra and fits, we calculated the moisturizing factor MF
obtained from the ratio of absorbances at 1640 cm–1 and 1545 cm–1 as used by
Gloor et al.15 Furthermore, we obtained the IR ratio of the 2125 cm–1
combination band following the method used by Potts et al.4 In Fig. 4.9 the
MF, the IR ratio, and the ratio of the amide I (at 1650 cm–1) to amide II (at
1545 cm–1) amplitudes calculated from the fit results are shown. In order to
compare the shapes of the curves, the IR ratio is shifted along the y-axis to
208 Chapter 4
1.5
MF, IR [-]
1.0
0 5 10 15 20 25 30 35
Time (min)
Figure 4.9 Moisturizing factor MF (•) ratio of absorbance at 1640 cm–1 and 1545 cm–1
according to Gloor,15 and infrared absorption ratio (□), shifted in y-axis to match highest MF
value, from the 2125 cm–1 combination band according to Potts4 and the amplitude ratio (○)
scaled 1.8 calculated from the ratio of the amide I at 1650 cm–1 to amide II at 1545 cm–1
amplitudes as a result of the fits.
match the MF curve. From the figure it can be inferred that the shapes of the
MF and IR curves with time are similar. The curve of the amide amplitude
ratio (1.8 scaled to match the MF curve) has a more pronounced slope in
the first part of the curve. This part of the curve is related to the maximizing of
contact area between the skin and crystal during the occlusion. The second
part shows similar slopes for the three ratios and represents the water content
increase after maximal contact area between the skin and crystal has been
reached.
The fit analyses of the hydration spectra have shown the separate
contributions from water bands in the (hydrated) skin SC spectra. For
determining the water content, the fit method bears the advantage that the
water bands (bending mode at 1640 cm–1, combination band at 2125 cm–1,
and the OH stretches) are varied individually and possible variations in, for
example, strong and weak bond stretches can be obtained. However, in the
time span of measurement in this study, no significant variations in strong or
weak bond stretch amplitudes have been observed. Further details on separate
contributions probably need longer (i.e., hours) hydration and corresponding
monitoring.
As compared to the other three methods used, the fit analysis gives more
insight in the changes due to contribution from the individual water bands and
changes of the amide I and amide II bands. In the Potts method using the
combination band region, we have shown that dilution of the skin SC

spectrum by the increasing the contribution of the water spectrum influences
the absorbances at either side of the 2125 cm–1 combination band, especially
at higher water content. The OH region is most sensitive to changes in the
outer SC cells because of the smaller penetration depth as compared to the
low wave number region. For an absolute measure of the water content a
calibration is necessary as in Potts’ method with similar measurements on in
vitro dry skin SC.
We applied ATR-FTIR spectroscopy to study normal and hydrated skin
stratum corneum in vivo. Using band fit analysis, the contribution of the most
important water bands (bending mode at 1640 cm–1, combination band at
2125 cm–1 and three OH stretches) in hydrated skin spectra was determined.
Assignments of skin SC bands and of liquid water are given and compared
with literature. The amplitudes found by the fit method determine the
individual water content contributions. The amplitudes of the bending band
and combination band saturate, whereas the OH band amplitudes continu-
ously increase during hydration time. The total spectrum of hydrated skin SC
depends nonlinearly on contributions from the individual water bands. This is
due to the different penetration depths of the infrared light at different wave
numbers, and to a changed water concentration profile in the outermost SC
layers during hydration.
4.3 Confocal Raman Microspectroscopy of Human Skin in vivo

Raman spectroscopy is widely used in biological studies, ranging from studies
of purified biological compounds to investigations at the level of single cells
(reviews by Tu 1982, Puppels 1999).37,38 The technique possesses certain
characteristics that make it particularly suitable for studying the skin, both in
vitro and in vivo. Raman spectroscopy provides detailed information about
the molecular composition and molecular structures in the skin. Also, because
molecular vibrations are directly influenced by the microenvironment of
functional groups, the vibrational spectrum provides information about
molecular interactions. Moreover, this type of information can be obtained in
a completely noninvasive manner as Raman spectra can be recorded directly
from the skin. When changes in the skin are to be monitored, this is an
advantage over existing techniques that require skin material to be extracted,
for example by tape stripping. An important advantage over FT-IR
spectroscopy is that not only the skin surface, but also the deeper skin layers
can be investigated.
Raman spectroscopic investigation of the skin was initiated by the groups
of Barry and Edwards at the University of Bradford, UK, who were the first
to publish band assignments of in vitro Raman spectra of the human stratum
corneum.39 In their study, the stratum corneum was separated from the
210 Chapter 4
underlying skin layers by heat separation and trypsin digestion. Fluorescence-

free FT-Raman spectra of stratum corneum were acquired in vitro. It was
concluded that Raman spectroscopy is a valuable tool with great potential for
investigating the molecular nature of the stratum corneum barrier. Thereafter,
Raman spectroscopy has been applied in a number of in vitro studies of skin,
addressing various issues such as stratum corneum permeability, molecular
structures of keratotic biopolymers, and characterization of pathological skin
conditions. Barry, Edwards, and Williams have performed Raman spectro-
scopic investigations of mammalian and reptilian (snakes) stratum corneum,
in the light of differences of the diffusion of drugs across these tissues.40–42
Williams et al. have compared Raman spectra of human keratotic biopolymers
in various tissues such as skin, callus, hair and nail. Principal structural
dissimilarities were found in the sulphur content of tissues containing hard
keratin and soft keratin.43 Interactions between the penetration enhancer
dimethyl sulfoxide (DMSO) and human stratum corneum were studied by
Anigbogu et al.44 FT-Raman spectroscopic investigation of stratum corneum
after treatment with aqueous solutions of DMSO revealed changes in the
protein structure of keratin from a-helix to b-sheet conformation, and evidence
was found of interactions of DMSO with stratum corneum lipids. FT-Raman
spectroscopy has also been used for investigations of ancient skin. Comparisons
of FT-Raman spectra of mummy skin with spectra of contemporary skin
have revealed spectral differences indicating changes in protein structure and/
or protein loss.45,46,47 De Faria et al. have demonstrated the feasibility of
measuring Raman spectra of the stratum corneum using visible light excitation.
In this study, the initially strong fluorescent background of the tissue was
photobleached and the residual emission was rejected by spatial filtering.48 The
potential application of FT-Raman spectroscopy for dermatological research
has been investigated by Edwards et al. and by Gniadecka et al. In the first
study by Edwards et al., FT-Raman spectra of normal human stratum
corneum, callus tissue, and hyperkeratotic psoriatic plaques were measured.
Changes in the 1000–1500 and 2800–3100 cm–1 wave number regions were
related to lipid loss in the abnormal tissues, but the keratin component was
found to be structurally unaltered.49 Gniadecka et al. have recorded
FT-Raman spectra from biopsies of various skin lesions, including basal and
squamous cell carcinoma. It was found that the examined skin lesions produced
unique spectral differences as compared to control skin, and it was suggested
that Raman spectroscopy could be useful as a diagnostic tool in dermatol-
ogy.50,51 Lademann et al. have applied Raman spectroscopy to obtain depth
information about the penetration of TiO2 particles in the skin.52 In this study
skin layers were sequentially removed by tape stripping and the removed skin
layers were analyzed by Raman spectroscopic inspection.
The possibility to obtain Raman spectra of human skin in vivo was demons-
trated first by Williams et al.,53 and later by Shim et al.54 and Schrader et al.55
Schallreuter et al. have used in vivo FT-Raman spectroscopy in dermatology to

study oxybenzone oxidation in the skin after solar irradiation,56 changes in
phenylalanine metabolism in the skin of vitiligo patients57 and hydrogen
peroxide accumulation in vitiligo.58
The different skin layers give rise to quite different Raman spectra.
Control over the tissue volume that is sampled can, therefore, be of great help
for the interpretation of Raman spectra of skin. In particular, the in vivo
investigation of the stratum corneum, which is very thin compared to the
underlying skin layers, requires sufficiently high spatial resolution to separate
the stratum corneum signal from that of the viable epidermis and dermis.
Recently, in vivo confocal Raman spectroscopy has been introduced as a
method to gain control over the actual skin layer from which a signal is
collected.59 This technique enables the recording of Raman spectra, from the
skin surface down to several hundred micrometers below the surface, with an
axial resolution of about 5 mm. In the following paragraph, a confocal Raman
setup is described and it is shown how confocal Raman microspectroscopy
can be used to determine concentration profiles of molecular compounds,
such as water and free amino acids, in the stratum corneum in vivo.
4.3.1 Setup for in vivo confocal Raman microspectroscopy

The in vivo Raman experiments described in this chapter were performed on a
home-built experimental inverted confocal Raman microspectrometer (see
Fig. 4.10). The laser source is a tuneable titanium-sapphire laser that is
pumped by an argon-ion laser. During in vivo Raman experiments, the hand
or arm rests on an aluminium table that contains a thin CaF2 or fused silica
PC
RS-232
servo controller
spec
tro
piezo translator
meter
voltage
CCD
filter LVDT sensor
laser filter
Figure 4.10 Experimental confocal Raman microspectrometer used for in vivo measure-
ments of human skin.
212 Chapter 4
window. A microscope objective is located under the window. The microscope

objective focuses the laser light in the skin at a well-defined distance from the
skin surface and also collects the light that is scattered by the tissue. Because
the skin is in contact with this window, it is possible to keep the distance
between the microscope objective and the skin surface fixed to within 2 mm
during the course of a measurement.
An overwhelming fraction of the scattered light is elastically scattered.
The Raman signal is separated from the elastically scattered light by a
dielectric high-pass filter. A holographic notch filter further suppresses the
intensity of the scattered laser light by another factor 106. Light is then
focused onto an optical fiber connected with the spectrometer. The core of this
optical fiber (diameter 100 mm) acts as a confocal pinhole that rejects the light
from out-of-focus regions. The microscope objective is mounted in a PIFOC
piezo nanopositioner enabling rapid computer-controlled axial positioning of
the laser focus. The PIFOC is equipped with an LVDT sensor (Linear
Variable Differential Transformer) and operates in a closed loop. It provides
350-mm axial travel with a full range repeatability of about 60 nm and a
linearity of typically 0.1%. The microscope system thus enables a fully
automated recording of Raman spectra at a range of predefined depths below
the skin surface. The spectrometer is a custom-built dispersive (grating-based)
multichannel system, equipped with a 1024 256 pixel deep depletion back-
thinned CCD detector with extended near-infrared sensitivity. In order to
minimize the influence of thermal noise, the CCD chip is cooled with liquid
nitrogen to 70°C.
The aim of a wave number calibration in multichannel dispersive
instruments is to assign a wave number shift to each individual CCD channel.
For the experiments described in this chapter, atomic lines of neon and argon
and a number of Raman calibration standards were used for wave number
calibration. Raman intensities of all recorded spectra were corrected for the
wavelength-dependent detection efficiency of the setup using the emission
spectrum of a calibrated tungsten band lamp. These procedures are described
in detail by Wolthuis et al.60
Autofluorescence of tissue severely hampers Raman experiments in which
visible laser light is used for excitation (see also de Faria et al.).48 The use of
near-infrared (NIR) light minimizes tissue autofluorescence. However, the
sensitivity of contemporary CCD detectors rapidly decreases at wavelengths
greater than 1000 nm. This puts an upper limit on the excitation wavelength
for which the Raman signal is still in the detection range of the detector.
Therefore, in the experiments described in this chapter, different excitation
wavelengths are employed for measurements in the so-called fingerprint
region (400–1850 cm–1) and for measurements in the CH-stretch region (2500–
3800 cm–1). The laser excitation wavelength for measurements in the
fingerprint region is 850 nm and for measurements in the CH-stretch
730 nm. In this way the Raman signal is always detected in the spectral range
between approximately 880 and 1010 nm.
Conventionally the axial resolution of a confocal microscope is derived
from the signal response of a point source that is moved through the focal
plane in the axial direction. The axial resolution is then defined as the full-
width-at-half-maximum (FWHM) of the response curve. However, for
confocal Raman spectroscopic experiments, a more practical measure of the
axial resolution is the response of the system that would be obtained if it were
scanned through an infinitesimally thin plane with Raman scatterers, rather
than a point source. We, therefore, used the following method to determine
the depth resolution. The Raman signal of oil was recorded, as the laser focus
was scanned across a CaF2-oil interface. The step response function thus
obtained was differentiated to obtain the impulse response of the system (i.e.,
the response of the system that would be obtained if it were scanned through
an infinitesimally thin plane with Raman scatterers). The FWHM of this
impulse response is a measure for the depth resolution. For excitation
wavelengths of 730 and 850 nm, the depth resolution was approximately
5 mm. In other words, an infinitely thin tissue section that is located 2.5 mm
above or below the focal plane contributes half the Raman signal of the
infinitely thin section that is in focus.
In vivo confocal Raman spectra of the skin show a considerable decrease
in the absolute signal intensity if the distance from the laser focus to the skin
surface is increased. This is mainly due to diffuse light scattering, which is a
much stronger effect in the skin than light absorption. Confocal detection is,
therefore, particularly useful to study the outer skin layers, i.e., the stratum
corneum and the viable epidermis. The in vivo Raman signal of the dermis is
strongly reduced due to scattering in the epidermis. Confocal detection of the
dermis Raman signal, therefore, requires considerably longer signal collection
times than for the epidermis.
This is illustrated in Fig. 4.11, showing in vivo confocal Raman spectrum of
the arm, measured in the stratum corneum and in the dermis. The signal
collection time is the same for both spectra, but the dermis signal is
significantly weaker than the signal of the stratum corneum. However, since
the dermis is much thicker than the epidermis (1 4 mm thick) it can easily be
studied using a nonconfocal detection scheme with a detection volume that is
large compared to the thickness of the epidermis. In this case the dermis will be
the dominant source of the Raman signal, which is illustrated by Fig. 4.12. The
figure shows the confocal spectrum from Fig. 4.11 and a nonconfocal spectrum
measured with a fiber-optic probe. The spectra are scaled to equal intensity. It
is clear that the spectrum obtained with the fiber-optic probe is almost entirely
determined by the Raman signal of the dermis. For equal signal collection
time, the S/N of the dermis spectrum obtained with the probe is considerably
higher than that of the spectrum that was measured confocally.
214 Chapter 4
Intensity, arbitrary units
stratum corneum
dermis
600 800 1000 1200 1400 1600 1800

Raman shift, cm-1
Figure 4.11 In vivo confocal Raman spectra of the volar aspect of the forearm. Top trace:
skin surface (stratum corneum). Bottom trace: 85 mm below skin surface (dermis).
Experimental conditions: signal collection time 2 min, laser power 100 mW. Spectra are on
the same scale.
confocal
fiber probe
600 800 1000 1200 1400 1600 1800

Raman shift, cm-1
Figure 4.12 In vivo Raman spectra of the dermis, as obtained by confocal and nonconfocal
detection (fiber-optic probe). Experimental conditions for the confocal spectrum and the
nonconfocal spectrum: signal collection time 2 min, laser power 100 mW. The spectra are
scaled to equal intensity.
4.3.2 Water and natural moisturizing factor in human skin epidermis

The stratum corneum is the body’s main protective barrier against epidermal
water loss and against microbial or chemical assault. Water plays a unique
role in the stratum corneum, affecting both physical and physiological
properties of this layer. From the physical perspective, stratum corneum
hydration is crucial for the pliability of the skin and the integrity of the skin
barrier. If the water content in the stratum corneum drops below about 10%,
it becomes brittle and may easily crack due to movements of the underlying
skin. Physiologically, water plays an important role in the stratum corneum
because it is thought to regulate the activity of specific hydrolytic enzymes
that are involved in the desquamation process of corneocytes at the skin
surface.61 The stratum corneum is exposed to a relatively dry environment,
and without water-retaining mechanisms would easily dehydrate. One of the
most important factors for water retention in the stratum corneum is the
natural moisturizing factor (NMF). This is a highly hygroscopic mixture of
specific free amino acids, derivatives of amino acids and salts, which is
exclusively present in the stratum corneum. NMF is produced in the deeper
cell layers of the stratum corneum by enzymatic hydrolysis of the protein
filaggrin into free amino acids.
A variety of techniques have been employed to study water and free
amino acids in the stratum corneum. However, very little in vivo information
exists about the concentration profiles of these molecular compounds. Warner
et al. have used electron probe microanalysis to determine the water content in
the stratum corneum as a function of depth in vitro.62 The result is widely
accepted as the standard water profile in the stratum corneum. The absolute
and relative amounts of free amino acids in the stratum corneum have been
determined using various techniques, which have in common that the skin
constituents were removed or extracted and then analyzed in vitro. Depth
information about the concentration gradient of NMF constituents in the
stratum corneum has been obtained using sequential tape stripping. With this
technique, adhesive tape is used to sequentially remove cell layers from the
stratum corneum, which can then be analyzed by HPLC.
The experiments described in the following section demonstrate that in
vivo confocal Raman microspectroscopy provides a method to obtain
information about the concentration profiles of molecular compounds such
as water and free amino acids in situ, that is, without removal of any skin
constituents.
4.3.3 Raman spectra of human skin constituents in vitro

Raman spectra of a large number of skin constituents were recorded in vitro.
These spectra were used to extract qualitative and quantitative information
about the chemical composition of the stratum corneum from Raman spectra
216 Chapter 4
of the skin that were obtained in vivo. The first part of this paragraph presents
an overview of the molecular compounds that are relevant for the Raman
signature of the stratum corneum. The second part describes the in vitro
Raman spectra that were used for the analysis of the in vivo experiments.
The major constituents of the stratum corneum are water and keratin. The
water content (grams of water per 100 grams of wet tissue) ranges from
approximately 15% at the outer stratum corneum cells to 70% in the stratum
granulosum.62 The protein keratin represents 75–80% of the stratum corneum
dry weight (see also Sec. 4.2.2.) and accounts for most of the in vivo Raman
signal of SC. Here we use the term keratin to describe the entire class of
keratins, although keratins in fact form a group of similar and closely related
structural proteins. In the literature a multiplicity of terms, such as a- and b-
keratin, hard/soft keratin, low/high-sulfur, is used to refer to different classes
of keratins.
Free amino acids and derivatives of amino acids form another major
group of stratum corneum constituents. This group is commonly referred to as
natural moisturizing factor (NMF). The concentration of NMF can reach as
much as 10% of the stratum corneum dry weight. It has been demonstrated
that a number of NMF constituents can have significant contributions to in
vivo Raman spectra of stratum corneum.59 These are pyrrolidone-5-carboxylic
acid (PCA), arginine (Arg), ornithine (Orn), citrulline (Cit), serine (Ser),
proline (Pro), glycine (Gly), histidine (His), alanine (Ala) and urocanic acid
(UCA). Also lactate and urea, which are excreted in sweat, can strongly
contribute to the Raman signal of stratum corneum.
Lipids comprise approximately 5–15% of the dry weight of the stratum
corneum. About 50% of the stratum corneum lipids are ceramides.
Ceramides, thus represent the most abundant class of lipids in the stratum
corneum. Lipids clearly contribute to the Raman signal of the stratum
corneum; however, on certain body sites such as at the palm of the hand, their
contribution may be small.
Contributions of other stratum corneum compounds, which represent
only minor fractions of the total stratum corneum weight, were not found to
significantly contribute to the Raman signal.
The following section describes the in vitro Raman spectra that were used
for the determination of molecular concentration gradients in the stratum
corneum by in vivo Raman spectroscopy.
Keratin is an insoluble protein and cannot be extracted without destroying
its secondary structure. We, therefore, used the Raman spectrum of stratum
corneum, devoid of lipids and water-soluble compounds, as a model for the
keratin fraction of the stratum corneum. A stratum corneum sample was
obtained from the peeled stratum corneum of a male volunteer who suffered a
mild sunburn. Lipids and water-soluble compounds were largely removed by
soaking the stratum corneum sample in methanol/ethanol (2:1 volume ratio)
amide I
CH2
amide III
Phe
Phe
Tyr
Phe
S-S
400 600 800 1000 1200 1400 1600 1800

Raman shift, cm-1
Figure 4.13 In vitro Raman spectrum of human stratum corneum after delipidization and
extraction of water-soluble compounds.
and water, so that the remaining sample consisted solely of keratin. The
Raman spectrum of the purified stratum corneum is shown in Fig. 4.13.
A number of prominent spectral bands are indicated in the figure. The amide I
vibration at 1650 cm–1, the amide III vibration at approximately 1275 cm–1,
and the CH2 scissoring vibration shift at 1450 cm–1 all give rise to strong
bands in the Raman spectrum. The position of the amide I and III bands
indicate that keratin in human stratum corneum adopts a predominantly a-
helical structure.38,42 A comparison of the Raman spectrum of stratum
corneum to the IR spectrum (see Fig. 4.3) nicely illustrates the complementary
aspect of both techniques. Whereas, for example, the IR spectrum shows a
weak amide III band, it is a prominent feature in the Raman spectrum. On the
other hand, the amide II at approximately 1550 cm–1 is very strong in the IR
spectrum and very weak in the Raman spectrum. Other clear features in the
Raman spectrum are the S-S stretch vibration around 525 cm–1 and the
aromatic ring vibrations of the amino acids phenylalanine (622, 1003, and
1032 cm–1) and tyrosine (644 cm–1).
For the experiments described in this chapter, the Raman spectrum of
ceramide 3 was used as a model for the total stratum corneum lipid content.
This is justified by the fact that ceramides form the most abundant class of
lipids in the stratum corneum. The spectrum of ceramide 3 is shown in
Fig. 4.14. Also shown in Fig. 4.14 are the Raman spectra of the dominant
NMF and sweat constituents. This selection consists of the spectra of lactate,
218 Chapter 4
PCA
Arg
Orn
Cit
Ser
Pro
Gly
Intensity, a.u.
His
Ala
UCA
lactate
urea
ceramide
water
400 600 800 1000 1200 1400 1600 1800
Raman shift, cm-1
Figure 4.14 In vitro Raman spectra of skin constituents. The spectrum of ceramide was
measured in the solid state. All other spectra are from solutions in water at pH 5.5.
urea and of the NMF constituents pyrrolidone-5-carboxylic acid (PCA),

arginine (Arg), ornithine (Orn), citrulline (Cit), serine (Ser), proline (Pro),
glycine (Gly), histidine (His), alanine (Ala) and urocanic acid (UCA). Raman
spectra of these compounds were measured in water solutions at pH 5.5,
which is regarded as the normal skin pH.63
The Raman signal of water only marginally contributes to the total
Raman signal of the stratum corneum when measured in the spectral region
from 400 to 1850 cm–1 (see also Fig. 4.14). However, in the CH-stretch region
between approximately 2500–3800 cm–1, the Raman signal of water is the
dominant factor, showing a broad band between 3100 and 3700 cm–1 due to a
triplet of n(OH) vibrations. Figure 4.15 shows the Raman spectra of pure
water and of dried SC in the spectral range between 2500 and 3800 cm–1. The
prominent features of the stratum corneum spectrum are indicated in the
figure.
4.3.4 Profiling the water content and NMF content in

human skin in vivo
4.3.4.1 Water
A measure of the water content of the stratum corneum is the ratio between
the integrated intensity of the OH stretch bands around 3400 cm–1 of water
and the CH3-stretch band at 2930 cm–1 of protein.64,65 It was mentioned in the
section on ATR-IR (Sec. 4.2) that the N-H vibration at 3275 cm–1 overlaps
with the broad OH-stretch band of water. Figure 4.15 shows that this is also
the case for the Raman spectrum of the stratum corneum. In order to avoid
overlap with the N-H vibration, the signal intensity of water was integrated
from 3350 to 3550 cm–1. The CH3 band of protein was integrated from 2910
to 2965 cm–1. A straight line connecting the spectral intensities at 2500 cm–1
and 3800 cm–1 was subtracted from the spectra in order to correct for
variations in (fluorescent) background signal. In this way a Raman spectrum
that is obtained at a certain distance from the skin surface provides a measure
for the water/protein ratio at that position.
ν(CH3)
ν(C-H)
N-H
dehydrated SC
ν(O-H)
ν(O-H)
ν(O-H)
water
2600 2800 3000 3200 3400 3600 3800

Raman shift, cm-1
Figure 4.15 In vitro Raman spectra in the CH-stretch region of dehydrated stratum
corneum and of water.
220 Chapter 4
The absolute water content in mass-% (gram water per 100 grams of
tissue) was derived as follows. The intensity ratio between the integrated
Raman signals of water and protein (IOH/ICH) is proportional to the mass
ratio between water and protein (mw/mp):
I OH mw
¼ · R: (4.4)
I CH mp
With this proportionality constant R, the absolute water content follows
directly from the Raman ratio (IOH/ICH) using the following equation:
I OH
mw I
water contentð%Þ ¼ · 100% ¼ I OH CH · 100%: (4.5)
mw þ mp I þR
CH
Raman spectra of aqueous solutions of bovine serum albumin (BSA) were

used to determine the proportionality constant R. Concentrations were in the
range of 20–40 weight percent. Equation (4.5) was then used to determine
absolute water concentrations from in vivo Raman measurements. Water
concentration profiles were determined from Raman spectra obtained at a
range of depths below the skin surface.
The use of solutions of BSA for the calibration of water concentration
profiles is based on the assumption that constant R in Eq. (4.4) is independent
of the type of protein used and is, therefore, equal for keratin and BSA. Since
the intensity of the CH-stretch band is insensitive to conformational changes,
this assumption seams reasonable. If, however, the proportionality constant R
shows a small dependence on the type of protein, it would result in a
systematic error in the calculation of the absolute water concentration.
Nonetheless, the systematic error would only influence the absolute water
concentration but not the changes in stratum corneum hydration or the shape
of the water concentration profile.
Figure 4.16 shows the results of a hydration experiment at the volar aspect
of the forearm of a male volunteer (30 years of age). Concentration profiles
were recorded for normal (nonhydrated) skin. Hereafter, the skin was
hydrated for 45 minutes with a wet bandage. Directly after removal of the
bandage the water concentration profiles of hydrated skin were recorded. The
in vivo water concentration profiles for normal and hydrated skin are plotted
in Fig. 4.16.
The water concentration profiles obtained by in vivo Raman spectroscopy
are in agreement with the in vitro profile obtained by electron probe analysis
of thin skin sections.62 The results for normal skin show that the water
concentration in the outer stratum corneum is 20 to 25%, which strongly
increases to 70% of the total tissue mass toward the boundary between the
stratum corneum and the viable epidermis. In hydrated skin the water
concentration in the outer stratum corneum has increased to 50–60%. The
70
hydrated
60
50
Water content
40 normal
30
20
10 viable
SC epidermis
0
0 5 10 15 20 25 30 35 40
Depth, μm
Figure 4.16 In vivo water concentration profiles for normal (solid) and hydrated skin (open)
at the volar aspect of the forearm. The vertical line indicates the estimated boundary
between the stratum corneum and the viable epidermis for normal skin. Experimental
conditions: exposure time: 5 s per data point, 10 s at depths greater than 20 mm. Scan step
size: 2 mm; laser power: 100 mW; excitation wavelength: 720 nm.
signal integration times in this experiment were relatively long (5–10 s). Since
the Raman signals are integrated over large spectral intervals, the quality of the
spectra, in terms of S/N, was much better than was needed for the
determination of water concentration profiles. Therefore, the signal collection
time could be reduced to less than 1 s per spectrum without affecting the results.
These results illustrate that confocal Raman microspectroscopy offers a
refined and subtle method for spatially resolved investigation of water in the
stratum corneum and the effects of hydration on the skin in vivo.
4.3.4.2 NMF
In the CH-stretch region, the Raman spectrum of the stratum corneum is largely
determined by only two compounds: keratin and water. This enables the use of
the simple method to determine the water content. In the fingerprint region
(approximately 400–1850 cm–1) the in vivo Raman spectrum of the stratum
corneum is composed of contributions from a number of compounds with
multiple and partially overlapping bands (see Fig. 4.14). We, therefore, used the
222 Chapter 4
more elaborate method of multiple regression fitting to analyze in vivo spectra of

the stratum corneum. In this analysis, in vivo skin spectra were modelled with a
set of in vitro Raman spectra that represent the dominant skin constituents (see
Sec. 4.3.2). The resulting set of fit coefficients describes how much each of the
skin constituents contributes to the in vivo spectrum of the stratum corneum.
The in vivo Raman spectra of stratum corneum could, to a large extent, be
modelled with the set of fit spectra, yielding residual spectra that showed
mainly noise (not shown here). However, the amide I band around 1650 cm–1
was found to vary and could, therefore, not be completely modelled by the
fixed set of fit spectra. Since the position and bandwidth of the amide I band
are sensitive to changes in the protein secondary structure, this variation may
reflect variations in the secondary structure of keratin and interactions
between keratin and water.
In vivo concentration profiles of NMF constituents were determined for
the stratum corneum of the palm by applying multiple regression fitting to a
series of skin spectra, obtained from a range of depths below the skin surface.
Each individual in vivo spectrum was thus modelled by a set of in vitro spectra
consisting of the spectra of washed/delipidized stratum corneum, ceramide 3,
water, lactate, and urea, and of the NMF constituents PCA, Arg, Orn, Cit,
Ser, Pro, Gly, His, Ala, and UCA. The Raman spectra of these compounds
are plotted in Figs. 4.13 and 4.14.
Figure 4.17 shows a semi-quantitative concentration profile of a number of
stratum corneum constituents, as determined by in vivo Raman spectroscopy.
1.0
Normalized concentration
Ser
0.5
Gly
Pca
Lactate
0.0
0 20 40 60
Depth (μm)
Figure 4.17 In vivo concentration profiles for the stratum corneum of the palm showing the
three dominant NMF constituents and lactate.
The highest concentration of lactate is observed at the surface, where it is

excreted in sweat, and rapidly drops thereafter. The concentrations of the three
dominant NMF constituents Ser, Gly, and PCA are low at the skin surface,
where these highly water-soluble compounds are easily flushed away by
external factors such as washing. Except for the upper cell layers of the stratum
corneum, their concentration shows little variation in most of the stratum
corneum. In the lower part of the stratum corneum, where NMF is produced,
the concentration profiles approach zero. It is clear, therefore, that obtained
in vivo Raman microspectroscopy provides detailed, spatially resolved
information about the molecular composition of the stratum corneum.
4.4 Resonance Raman Spectroscopy of Cutaneous

Carotenoids in vivo
4.4.1 Properties and role of cutaneous carotenoids
The human skin and especially the epidermis contain a high amount of
different antioxidant substances, such as enzymes (superoxide dismutase,
catalase, glutathione peroxidase), vitamins (A, C, D and E), carotenoids
(beta-carotene, lycopene, lutein/zeaxanthin) and others (lipoic acid, uric acid,
selenium, coenzyme Q10, etc.). These substances form an antioxidant network
of living skin. Contrary to enzymes, carotenoids and vitamins (A, C and E)
cannot be synthesized by the human organism, thus they must be fed via
supplements or topically.
Carotenoids are fat-soluble powerful antioxidants,66 which are found in
fruit, vegetables, plants, algae,66 egg yolks,67,68 and in the human skin.69,70
Nature widely uses carotenoids as an effective protection mechanism of the
skin against any negative oxidative action of free radicals produced sub-
sequent to metabolic processes, inflammations, UV, VIS and IR irradiation71
and as a consequence of environmental hazards and toxins.72
The most prevalent carotenoids found in the human skin are beta-
carotene and lycopene.73 Beta-carotene and lycopene are p-electron
conjugated carbon-chain molecules, which consist of alternating double and
single carbon bonds. The absorption spectra of beta-carotene and lycopene
are shown in Figure 4.18. The differences in the absorption spectra for these
carotenoid substances are explained by the quantity of the conjugated carbon-
carbon double bonds in the structure of carotenoid molecules (11 for lycopene
and 9 for beta-carotene).
The efficiency of Raman scattering of the investigated molecules strongly
depends on the absorption spectra and, as a result, on the excitation
wavelength. Taking the absorption spectra of the carotenoids beta-carotene
and lycopene into consideration (see Fig. 4.18), the blue (488 nm) and the
green (514.5 nm) wavelengths of an Ar þ laser are well suited as a source of
resonance Raman excitation of the cutaneous carotenoids.
224 Chapter 4
1,0
Absorption
0,5
0,0
400 450 500 550

Wavelength, nm
Figure 4.18 Absorption spectra of solutions of beta-carotene (solid line) and lycopene
(dotted line) substances in ethanol.
Carotenoids have three prominent Stokes lines, which are shifted to

1523 cm–1, 1156 cm–1 and 1005 cm–1 towards the excitation wavelength.
These Raman peaks corresponded to carbon-carbon double bond vibrations
of the conjugated backbone of the carotenoid molecule (C¼C), carbon-carbon
single bond vibrations (C–C) and rocking motion of the carotenoids methyl
components (C–CH3), respectively.70 The strong Raman line at 1523 cm–1
was used for the determination of carotenoids in the skin. This was done
due to the fact that the C¼C chemical bonds of the carotenoid molecules
are responsible for their antioxidant properties by neutralizing free
radicals.74,75 Thus, the quenching of free radicals induces the destruction of
the carotenoids’ C¼C bonds that are attributed to the decrease of the
1523 cm–1 Raman peak intensity.
The spectrograph was adjusted to receive wavelengths between 524 and
561 nm, which warrants the simultaneous detection of the corresponding
C¼C Raman bands of carotenoids (527 nm under excitation with 488 nm and
558 nm, under excitation with 514.5 nm).
The carotenoids had an extremely low quantum efficiency of luminescence
(10–5–10–4). This is caused by the existence of a second excited singlet
intermediate state, which lies below the first excited energy level of
carotenoids. The interchange of energy between these two excited levels
occurs very rapidly (within 200–250 fs) via non-radiating transitions. The
posterior electronic emission from the second excited level to ground state is
parity-forbidden, which gives rise to the absence of fluorescence between these
1523 cm-1
C=C
0,011
1156 cm-1
C-C
1005 cm-1
C-CH3
0,010
Intensity, a.u.
0,009
0,008
600 800 1000 1200 1400 1600

Wavenumber, cm-1
Figure 4.19 Typical Raman spectrum obtained from human skin under an excitation wave-
length at 514.5 nm (excitation with Ar+ laser, power density on the skin surface 30 mW/cm2).
two energy levels of carotenoid molecules.73,76 Therefore, carotenoids do not

contribute to the signal of high fluorescence measured from the skin. The
fluorescence of the skin may have been caused by cutaneous collagen, elastin,
NAD, NADH, and blood,77 but not by carotenoids.
Thus, only under the resonance conditions of excitation is it possible to
obtain the prominent Raman spectrum of carotenoids from the skin super-
imposed on the large fluorescence background (see Fig. 4.19). The intensities of
the measured Raman peaks are strongly correlated with the absorption spectra
of carotenoids and characterize the concentration of the investigated molecules in
the measured skin volume.
Raman peaks from cutaneous carotenoids obtained under resonance
excitation conditions are superimposed on the large fluorescence background
of the skin and constitute not more than 2% of the total cutaneous intensity of
fluorescence.
4.4.2 Setup for in vivo resonance Raman spectroscopy of carotenoids

An Ar þ multiline laser (1) operated at 488 nm and 514.5 nm was used as a
source of resonance excitation of cutaneous carotenoids. Then light was
expanded, collimated by the lenses of the optical imaging system (3) and
focused onto the skin surface. The special geometry of the receiving channel
cut off the back-reflected light and transmitted the fluorescence and Raman-
scattered light in the range between 524 and 561 nm to the spectrograph (6)
that was coupled to a CCD camera (7). The obtained spectra were analyzed
226 Chapter 4
1
2
3
8
6
7
5
Figure 4.20 Scheme of the setup for in vivo resonance Raman spectroscopy of
carotenoids. (1): Ar+ multiline laser, (2) and (5): optical fibers, (3): optical imaging system,
(4): measuring object (skin), (6): spectrograph, (7): CCD camera, (8): personal computer.
by a personal computer (8). The power density on the skin surface was
adjusted between 30 and 45 mW/cm2 depending on the excitation wavelength
that was within the safety standard. The experimental arrangement for the
Raman measurements of cutaneous carotenoids is presented in Figure 4.20.
To eliminate the negative influence of pigmentation and heterogeneities
of the skin on the optical measurements, the excitation beam was expanded
up to 6.5 mm in diameter.78 The presented value of the excitation beam
size was evaluated experimentally and provided the optimal stability of
the Raman signal obtained from the skin (measured data scattering is
around 10%).
The intensity of the Raman peak was determined with the use of the
Gauss approximation line after the subtraction of fluorescence background.
4.4.3 Selective detection of carotenoids in the human skin

The most prevalent cutaneous carotenoids beta-carotene and lycopene have
different absorption spectra. The difference is observed in the narrow interval
of the spectra, ranged around 495–535 nm. Moreover, under the excitation
with 488 nm, both substances are excited with approximately the same
efficacy, but under the excitation with 514.5 nm, appreciably more lycopene
was excited compared to beta-carotene (see Fig. 4.18). As a result of the
different absorption values for beta-carotene and lycopene, the Raman
efficacies will also be different, strongly reflecting the absorption abilities of
the investigated molecules. This difference is used for the separate detection of
carotenoids in the skin and the following theoretical algorithm is supposed70:
The normalized intensities of Raman peak at 1523 cm–1 under 488 and
514.5 nm excitations (I488n and I514n), divided by the corresponding value of
laser power, are described by the following expressions:
I 488 n ðs488 bC N bC þ s488 L N L Þ (4.6)
I514 n ðs514 bC N bC þ s514 L N L Þ, (4.7)

where NbC and NL are the concentrations of beta-carotene and lycopene,

s488bC, s488L, s514bC, s514L are the resonance Raman scattering cross sections
for beta-carotene and lycopene under 488 nm and 514.5 nm excitations,
respectively.
After combining the equations (4.6) and (4.7), the obtained ratio between
the concentrations of beta-carotene and lycopene will have the following
expression:
N bC ∕N L ¼ ðs488 L s514 L rÞ∕ðs514 bC r s488 bC Þ, (4.8)
where r ¼ I488n / I514n is a measuring parameter.

Taking into consideration the experimentally obtained values for s,
expression (4.8) is transformed to:
N bC ∕N L ¼ ð0.95 0.44 rÞ∕ð0.06 r 1Þ (4.9)
The relative concentrations of beta-carotene and lycopene are obtained

after combining Eqs. (4.6) and (4.9):
N bC ðI 488 n N bC ∕N L Þ∕ðN bC ∕N L þ 0.95Þ (4.10)
N L I 488 n ∕ðN bC ∕N L þ 0.95Þ, (4.11)
which, after calibration, transform to the equations for absolute concentra-

tions, measured in nmol/g:
N abs bC ¼ ð2000 I 488 n N bC ∕N L Þ∕ðN bC ∕N L þ 0.95Þ (4.12)
N abs L ¼ 2000 I 488 n ∕ðN bC ∕N L þ 0.95Þ (4.13)
The most prevalent cutaneous carotenoid antioxidants are alpha-

carotene, beta-carotene, gamma-carotene, lutein, zeaxanthin and lycopene.
Beta-carotene represents more than 50% of the combined carotenes.
Carotenes and lycopene represent more than 70% of the carotenoids in the
human skin.73
All carotenoids measured in the human skin, with the exception of
lycopene, have approximately the same absorption spectra (see beta-carotene
in Fig. 4.18) and, as a result, identical Raman efficacies, which are
superimposed to each other. Therefore, cutaneous carotenes, lutein, zeaxan-
thin, and lycopene were investigated. The measuring algorithm was based on
the assumption that beta-carotene and lycopene represent the majority of the
228 Chapter 4
complete range of cutaneous carotenoids. The detailed description of the

method has been previously published.70
4.4.4 In vivo measurements of the influence of UV irradiation

on human skin
The irradiation of the skin with UV light leads to the formation of free
radicals,79 which react immediately with antioxidants, thus destroying them.
It is well known that cutaneous carotenoids are reduced after the UV
irradiation of the skin.80 These measurements were performed with the use of
the HPLC method (high pressure liquid chromatography). This highly
invasive chemical method can give no information about the real kinetics of
the degradation of carotenoids.
The skin of the flexor forearm of healthy volunteers was exposed to
UVB radiation at a power density of 0.3 mW/cm2 for 3 min. This dose
was sufficient for the formation of a light erythema on the irradiated skin
surface.
Using the non-invasive optical method based on resonance Raman
spectroscopy, it could be clearly shown that the degradation of the cutaneous
carotenoids beta-carotene and lycopene does not occur immediately subse-
quent to the UV irradiation. There is a time delay, which varies from 30 up to
90 minutes for beta-carotene and from 0 up to 30 minutes for lycopene, before
the concentration of carotenoids will be decreased.81
The differences can be explained by different quenching rate constants in
the reaction of neutralization of oxygen free radicals for beta-carotene and
lycopene. Lycopene has the highest quenching rate constant compared to
other carotenoids.82 This means that lycopene reacts first to the free radicals
produced in the skin subsequent to UV irradiation, thus providing the
strongest defense activity.83 The amount of destroyed carotenoids measured in
vivo is 1.5 times higher than the amount of destroyed carotenoids measured on
extracted human skin ex vivo, which is explained by the increased free radical
generation in the skin in vivo due to the higher presence of oxygen in the
tissue.84
The presence of the strong non-linear correlation between the individual
antioxidant level of volunteers and the magnitude of destruction of
carotenoids in the skin showed that, in all volunteers, approximately the
same amount of carotenoids was destroyed after the application of a
definite UV dose. Taking into consideration the different individual levels
of the cutaneous carotenoids of the volunteers, the magnitude of destruction
of carotenoids was less for volunteers with a high individual level
and, vice versa, was higher for volunteers with a low individual level of
cutaneous carotenoids. Thus, volunteers with a high initial level of
carotenoids in the skin have an additional defense against the negative
action of free radicals.
4.4.5 In vivo measurements of the influence of IR irradiation

on human skin
IR irradiation, as a warm irradiation, increases the blood and lymph flows, as
a result of which the rate of metabolism and recovery increases and the wound
healing processes accelerates.85,86 Therefore, IR irradiation is widely used in
medical practice.
It was shown in the in vivo experiments that the cutaneous carotenoids
beta-carotene and lycopene are reduced subsequent to IR irradiation of the
skin.87 The magnitude of destruction was determined at 27% for beta-carotene
and 38% for lycopene on average for all volunteers.
The power density of the IR-irradiation on the skin surface and the time
of irradiation were 190 mW/cm2 and 30 minutes, correspondingly.
The obtained degradation of the cutaneous carotenoids was explained by
the neutralization of free radicals, which can be produced in the skin
subsequent to IR irradiation. This effect was confirmed by direct measure-
ment of the IR-induced free radical generation using electron paramagnetic
resonance spectroscopy.71,88,89
The question as to whether the IR induces directly the formation of ROS
or whether this is a result of the IR-induced heat shock is still open.90,91 The
different signaling pathways of infrared-induced ROS and infrared-induced
heat shock ROS were shown to act independently multiplying the influence on
each other by increasing the doses of irradiation and/or increasing the
temperature.86
As in the case of UV irradiation, a strong correlation between the
individual level of beta-carotene and lycopene in the skin of volunteers and
the magnitude of destruction of carotenoids was found, which also confirms
the previous conclusion about the protective role of the carotenoid
antioxidants against the negative action of free radicals.
4.4.6 In vivo measurements of the influence of VIS irradiation

on human skin
According to the free radical action spectrum published by Zastrow et al.,71
the formation of free radicals in human skin subsequent to irradiation with
visible light should also result in the destruction of cutaneous carotenoids. The
in vivo measurements using resonance Raman spectroscopy show that the
irradiation of the human skin with blue-violet light results in a dose-dependent
significant degradation of the cutaneous carotenoids. A possible influence of
the heat-shock radicals was eliminated in this study. The mean magnitude of
the carotenoid destruction was determined to be 13.5% after irradiation at
50 J/cm2 and 21.2% after irradiation at 100 J/cm2. The restoration time was
measured to be 1 hour for the dose of 50 J/cm2 and 24 hours for the dose of
100 J/cm2. The same effect is expected for irradiation of the human skin in the
green-red range of the spectra.
230 Chapter 4
4.4.7 Factors influencing the concentration of carotenoids

in human skin
In parallel with irradiations, other stress factors are always associated with the
generation of free radicals in the human skin. Using resonance Raman
spectroscopy, the reduction of the carotenoid concentration in the human skin
was clearly observed after alcohol abuse,92 heavy sport loads,93 during
inflammation, illness, fatigue94 and stress.94–96
The human organism cannot synthesize carotenoids, thus they should be
digested systemically or via topically applied formulations containing these
substances.
A variety of foodstuffs contain carotenoids in different mixtures and
concentrations. Fruit and vegetables are naturally rich in different carote-
noids.66 After supplementation, carotenoids are assimilated by the organism
and the increased concentration of carotenoids can be measured in the blood
flow after a carotenoid-rich supplementation. The accumulation of carote-
noids in the tissue including the skin is a slower process.97
It was shown in the in vivo experiments that after the supplementation of
high amounts of fruit and vegetables, as well as tomato paste or ketchup, the
level of cutaneous carotenoids increased.98 The corresponding increase
can be measured on the day after supplementation.99 Moreover, a “seasonal
increase” in the cutaneous carotenoids was detected in all volunteers. This
“seasonal increase” was determined to be 1.26 fold on average and results
from the supplementation of an increased amount of fruit and vegetables in
the German summer and autumn months compared to the winter and spring
months.94 Moreover, carotenoids are able to protect dermal collagen and
elastin against solar irradiation-induced depletion.100
The systemic application of carotenoid-rich supplements also give rise to
the increase of carotenoid concentration in the human skin. This was
confirmed in different studies.97,101–103 The topical application of carotenoid-
containing cosmetic formulations give rise to the immediate increase of
carotenoids in the stratum corneum,97,104 thus providing an additional
protection against environmental hazards.105 Contrary to systemic applica-
tion, topically applied carotenoids are stored in the stratum corneum only for
a short time due to their depletion by desquamation, textile contact, washing,
and the influence of environmental stress.
4.4.8 Distribution of carotenoids in human skin

The carotenoids are differently distributed in the human skin. For example,
the concentration of carotenoids is higher on the inner palm and on the
forehead than on the flexor forearm, neck, and on the back.106 The differences
remained significant after adjustment for age, gender, and diet intake
estimates.
The depth-dependent distribution of carotenoids in the stratum corneum

of the human skin was measured in vivo using confocal Raman microscopy
under non-resonant excitation at 785 nm by tracking the C¼C Raman peak at
1523 cm–1. The highest carotenoid concentration was found in the outermost
layer of stratum corneum. Then the carotenoid concentration reduced deep
inside the stratum corneum.104,107 The obtained carotenoid distribution was
unexpected and was confirmed by the delivery of carotenoids by sweat and
sebaceous glands onto the skin surface.108
4.4.9 Conclusions
Resonance Raman spectroscopy is a well suited, non-invasive and quick
method, which has no analogues for the in vivo determination of the
carotenoid antioxidant substances in the human skin at high sensitivity. It
provides the possibility to perform qualitative and quantitative measurements
on any site of the human skin. The performed measurements show that
cutaneous carotenoids reflect the lifestyle and stress conditions of volunteers.
“Negative” stress factors, such as irradiation, illness, alcohol abuse, smoking,
and others decrease the antioxidant status of the skin, which can be then
recovered only by an antioxidant-rich supplementation, for example, rich in
fruit and vegetables. The amount of the destroyed cutaneous carotenoids can
be well correlated with the amount of the free radicals generated in the skin,
which was confirmed by electron paramagnetic resonance measurements,
showing that carotenoids could serve as marker substances for the whole
antioxidant status of the human epidermis.
Acknowledgments
The authors are indebted to the late Dr. H.E. Boddé for very stimulating
discussions on the FTIR spectra.
References
1. G. Herzberg, Molecular Spectra and Molecular Structure, 2nd Ed., van
Nostrand Reinhold Company, New York (1950).
2. N. B. Colthup, L. H. Daly, and S. E. Wiberley, Introduction to Infrared
and Raman Spectroscopy, 3rd Ed., Academic Press, New York (1990).
3. D. Lin-Vien, The Handbook of Infrared and Raman Characteristic
Frequencies of Organic Molecules, Academic Press, San Diego (1991).
4. R. O. Potts, “Stratum corneum hydration: experimental techniques and
interpretations of results,” J. Soc. Cosmet. Chem. 37, 9–33 (1986).
5. J. Serup and G. B. E. Jemec, Eds., Handbook of Non-invasive Methods
and the Skin, CRC Press, Boca Raton (1995).
232 Chapter 4
6. N. A. Puttnam and B. H. Baxter, “Spectroscopic studies of skin in situ by

attenuated total reflectance,” J. Soc. Cosmet. Chem. 18, 469–472 (1967).
7. J. R. Hansen and W. Yellin, “NMR and infrared spectroscopic studies
of stratum corneum hydration,” pp. 19–28 in Water Structure at the
Water Polymer Interface, H. H. G. Jellinek, Ed., Plenum, New York
(1972).
8. R. L. Anderson, J. M. Cassidy, J. R. Hansen, and W. Yellin, “Hydration
of stratum corneum,” Biopolymers 12, 2789–2802 (1973).
9. K. Martin, “Direct measurements of moisture in skin by NIR
spectroscopy,” J. Soc. Cosmet. Chem. 44, 249–261 (1993).
10. R. O. Potts, D. B. Guzek, R. R. Harris, and J. E. McKie,
“A noninvasive, in vivo technique to quantitatively measure water
concentration of the stratum corneum using attenuated total-reflectance
infrared spectroscopy,” Arch. Dermatol. Res. 277, 489–495 (1985).
11. D. Bommannan, R. O. Potts, and R. H. Guy, “Examination of stratum
corneum barrier function in vivo by infrared spectroscopy,” J. Invest.
Dermatol. 95(4), 403–408 (1990).
12. H. E. Boddé, L. A. R. M. Pechtold, M. T. A. Subnel, and F. H. N.
de Haan, “Monitoring in vivo skin hydration by liposomes using
infrared spectroscopy in conjunction with tape stripping,” pp. 137–149 in
Liposome Dermatics, O. Braun-Falco, H. C. Korting, and H. I. Maibach
Eds., SpringerVerlag, Berlin, Heidelberg (1992).
13. K. Wichrowski, G. Sore, and A. Khaïat, “Use of infrared spectroscopy
for in vivo measurement of the stratum corneum moisturization after
application of cosmetic preparations,” Int. J. Cos. Sci. 17, 1–11 (1995).
14. G. W. Lucassen, G. N. A. van Veen, and J. A. J. Jansen, “Band analysis
of hydrated human skin stratum corneum attenuated total reflectance
Fourier transform infrared spectra in vivo,” J. Biomedical Optics 3(3),
267–280 (1998).
15. M. Gloor, U. Wildebrandt, G. Thomer, and W. Kugerschmid, “Water
content of the horny layer and skin surface lipids,” Arch. Dermatol. Res.
268, 221–223 (1980).
16. J. R. Scherer, M. K. Go, and S. Kint, “Raman spectra and structure of
water in dimethyl sulfoxide,” J. Phys. Chem. 77(17), 2108–2117 (1973).
17. J. R. Scherer, M. K. Go, and S. Kint, “Raman spectra and structure of
water from 10 to 90°,” J. Phys. Chem. 78(13), 1304–1313 (1974).
18. B. W. Barry, H. G. M. Edwards, and A. C. Williams, “Fourier
transform Raman and infrared vibrational study of human skin:
Assignments of spectral bands,” J. Raman Spectrosc. 23, 641–645 (1992).
19. D. A. Draegert, N. W. B. Stone, B. Curnutte, and D. Williams,
“Far-infrared spectrum of water,” J. Opt. Soc. Am. 56(1), 64–69 (1966).
20. D. Williams, “Frequency assignments in infra-red spectrum of water,”
Nature 210, 194–195 (1966).
21. Y. Marechal, “Infrared spectra of water I, Effect of temperature and of

H/D isotopic dilution,” J. Chem. Phys. 95(8), 5565–5573 (1991).
22. Y. Marechal, “Infrared spectra of water II: Dynamics of H2O (D2O)
molecules,” J. Phys. II France, 557–571 (1993).
23. Y. Marechal, “Infrared spectra of a poorly known species: Water 3,”
J. Phys. Chem. 97, 2846–2850 (1993).
24. M. Moskovits and K. H. Michaelian, “A reinvestigation of the Raman
spectrum of water,” J. Chem. Phys. 69(6), 2306–2311 (1978).
25. B. Curnutte and J. Bandekar, “The intramolecular vibrations of the
water molecule in the liquid state,” J. Molec. Spectrosc. 41, 500–511
(1972).
26. J. B. Bryan and B. Curnutte, “A normal coordinate analysis based on the
local structure of liquid water,” J. Molec. Spectrosc. 41, 512–533 (1972).
27. J. Bandekar and B. Curnutte, “A local-structure model for calculation of
lattice vibrations in liquid water,” J. Molec. Spectrosc 58, 169–177
(1975).
28. W. A. P. Luck, “Hydrogen bonds in liquid water,” pp. 1369–1420 in
The Hydrogen Bond, Recent Developments in Theory and Experiments,
P. Schuster et al., Eds., North Holland Pub. Co., Amsterdam (1976).
29. J. Israelachvili and H. Wennerström, “Role of hydration and water
structure in biological and colloidal interactions,” Nature 379, 219–225
(1996).
30. H. G. M. Edwards, D. W. Farwell, A. C. Williams, B. W. Barry, and
F. Rull, “Novel spectroscopic deconvolution procedure for complex
biological systems: vibrational components in the FT-Raman spectra of
ice-man and contemporary skin,” J. Chem. Soc. Faraday Trans. 91(12),
3883–3887 (1995).
31. K. S. Seshadri and R. N. Jones, “The shapes and intensities of infrared
absorption bands-A review,” Spectrochimica 19, 1013–1085 (1963).
32. G. Herzberg, “Molecular spectra and molecular structure II,” in Infrared
and Raman Spectra of Polyatomic Molecules, van Nostrand Reinhold,
New York (1945).
33. P. T. Pugliese and A. J. Milligan, “Ellipsometric measurement of skin
refractive index in vivo,” pp. 291–302 in Bioengineering and the Skin,
R. Marks and P. A. Payne, Eds., MTP Press Limited, Boston (1981).
34. J. E. Bertie, M. K. Ahmed, and H. H. Eysel, “Infrared intensities of
liquids, 5. Optical dielectric constants, integrated intensities, and dipole
moment derivatives of H2O and D2O at 22°C,” J. Phys. Chem. 93,
2210–2218 (1989).
35. J. R. Reitz, F. J. Milford, and R. W. Christy, Foundations of
Electromagnetic Theory, 3rd Ed., Addison-Wesley, Massachusetts (1979).
36. C. Zviak, “The science of hair care,” in Dermatology 7, Marcel Dekker,
Inc. New York, 21 (1984).
234 Chapter 4
37. G. J. Puppels, “Confocal Raman microspectroscopy,” pp. 377–406 in

Fluorescent and Luminescent Probes for Biological Activity, ed. by
W. Mason, Academic Press, London (1999).
38. A. T. Tu, Raman Spectroscopy in Biology, John Wiley & Sons Ltd,
New York (1982).
39. B. W. Barry, H. G. M. Edwards, and A. C. Williams, “Fourier transform
Raman and infrared vibrational study of human skin: assignment of
spectral bands,” J Raman Spectroscopy 23, 641–645 (1992).
40. B. W. Barry, A. C. Williams, and H. G. M. Edwards, “Fourier
transform Raman and IR spectra of snake skin,” Spectrochim. Acta,
Part A 49, 801–807 (1993).
41. H. G. M. Edwards, D. W. Farwell, A. C. Williams, and B. W. Barry,
“Raman spectroscopic studies of the skins of the Sahara sand viper, the
carpet python and the American black rat snake,” Spectrochim. Acta,
Part A 49, 913–919 (1993).
42. A. C. Williams, B. W. Barry, and H. G. M. Edwards, “Comparison of
Fourier transform Raman spectra of mammalian and reptilian skin,”
Analyst 119, 563–565 (1994).
43. A. C. Williams, H. G. M. Edwards, and B. W. Barry, “Raman spectra of
human keratotic biopolymers: skin, callus, hair and nail,” J. Raman
Spectrosc. 25, 95–98 (1994).
44. A. N. C. Anigbogu, A. C. Williams, B. W. Barry, and H. G. M.
Edwards, “Fourier transform Raman spectroscopy of interactions
between the penetration enhancer dimethyl sulfoxide and human stratum
corneum,” Int. J. Pharm. 125, 265–282 (1995).
45. M. Gniadecka, H. C. Wulf, D. H. Christensen, and J. P. H. Hansen,
“Fourier transform Raman spectroscopy of 15th century mummies from
Qilakitsoq, Greenland,” J. Raman Spectrosc. 28, 179–184 (1997).
46. A. C. Williams, H. G. M. Edwards, and B. W. Barry, “The ‘Iceman’ -
molecular structure of a 5200-year-old skin characterized by Raman-
spectroscopy and electron-microscopy,” Biochim. Biophys. Acta 1246,
98–105 (1995).
47. H. G. M. Edwards, D. W. Farwell, A. C. Williams, B. W. Barry, and
F. Rull, “Novel spectroscopic deconvolution procedure for complex
biological systems: vibrational components in the FT-Raman spectra of
ice-man and contemporary skin,” J. Chem. Soc. Faraday. Trans.
91, 3883–3887 (1995).
48. D. L. A. de Faria and M. A. de Souza, “Raman spectra of human skin
and nail excited in the visible region,” J Raman Spectroscopy 30,
169–171 (1999).
49. H. G. M. Edwards, A. C. Williams, and B. W. Barry, “Potential
applications of FT-Raman spectroscopy for dermatological diagnos-
tics,” J. Mol. Struct. 347 379–387 (1995).
50. M. Gniadecka, H. C. Wulf, N. Nymark Mortensen, O. Feurskov

Nielsen, and D. H. Christensen, “Diagnosis of basal cell carcinoma by
Raman spectroscopy,” J Raman Spectroscopy 28, 125–130 (1997).
51. M. Gniadecka, H. C. Wulf, O. Faurskov Nielsen, D. H. Christensen,
and J. Hercogova, “Distinctive molecular abnormalities in benign and
malignant skin lesions: Studies by Raman spectroscopy,” Photochem.
Photobiol. 66, 418–423 (1997).
52. J. Lademann, H. Weigmann, C. Rickmeyer, H. Barthelmes, H. Schaefer,
G. Mueller, and W. Sterry, “Penetration of titanium dioxide
microparticles in a sunscreen formulation into the horny layer and
the follicular orifice,” Skin Pharmacol Appl Skin Physiol 12, 247–256
(1999).
53. A. C. Williams, B. W. Barry, H. G. M. Edwards, and D. W. Farwell,
“A critical comparison of some Raman spectroscopic techniques for studies
of human stratum corneum,” Pharmaceut. Res. 10, 1642–1647 (1993).
54. M. G. Shim and B. C. Wilson, “Development of an in vivo Raman
spectroscopic system for diagnostic applications,” J. Raman Spectros-
copy 28, 131–142 (1997).
55. B. Schrader, B. Dippel, S. Fendel, S. Keller, T. Löchte, M. Riedl, R.
Schulte, and E. Tatsch, “NIR FT Raman spectroscopy - a new tool in
medical diagnosis,” J. Mol. Struct., 408–409, 23–31 (1997).
56. K. U. Schallreuter, J. M. Wood, D. W. Farwell, J. Moore, and H. G.
Edwards, “Oxybenzone oxidation following solar irradiation of skin:
photoprotection versus antioxidant inactivation [see comments],”
J. Invest. Dermatol 106, 583–586 (1996).
57. K. U. Schallreuter, M. Zschiesche, J. Moore, A. Panske, N. A. Hibberts,
F. H. Herrmann, H. R. Metelmann, and J. Sawatzki, “In vivo evidence
for compromised phenylalanine metabolism in vitiligo,” Biochem.
Biophys. Res. Commun. 243, 395–399 (1998).
58. K. U. Schallreuter, J. Moore, J. M. Wood, W. D. Beazley, D. C. Gaze,
D. J. Tobin, H. S. Marshall, A. Panske, E. Panzig, and N. A. Hibberts,
“In vivo and in vitro evidence for hydrogen peroxide (H2O2) accumula-
tion in the epidermis of patients with vitiligo and its successful removal
by a UVB-activated pseudocatalase,” J. Investig. Dermatol. Symp. Proc.
4, 91–96 (1999).
59. P. J. Caspers, G. W. Lucassen, R. Wolthuis, H. A. Bruining, and G. J.
Puppels, “In vitro and in vivo Raman spectroscopy of human skin,”
Biospectroscopy 4, S31–39 (1998).
60. R. Wolthuis, T. Bakker Schut, P. Caspers, H. Buschman, T. Roemer,
H. Bruining, and G. Puppels, “Raman spectroscopic methods for in vitro
and in vivo tissue characterization,” pp. 433–455 in Fluorescent and
Luminescent Probes for Biological Activity, ed. by W. Mason, Academic
Press, London (1999).
236 Chapter 4
61. A. Rawlings, C. Harding, A. Watkinson, J. Banks, C. Ackerman, and

R. Sabin, “The effect of glycerol and humidity on desmosome degradation
in stratum corneum,” Arch. Dermatol. Res. 287, 457–464 (1995).
62. R. R. Warner, M. C. Myers, and D. A. Taylor, “Electron probe analysis
of human skin: determination of the water concentration profile,”
J. Invest. Dermatol. 90, 218–224 (1988).
63. O. Braun-Falco and H. C. Korting, “Normal pH value of human skin,”
Hautarzt 37, 126–129 (1986).
64. N. J. Bauer, J. P. Wicksted, F. H. Jongsma, W. F. March, F. Hendrikse,
and M. Motamedi, “Noninvasive assessment of the hydration gradient
across the cornea using confocal Raman spectroscopy,” Invest.
Ophthalmol. Vis. Sci. 39, 831–835 (1998).
65. A. Huizinga, A. C. Bot, F. F. de Mul, G. F. Vrensen, and J. Greve,
“Local variation in absolute water content of human and rabbit eye
lenses measured by Raman microspectroscopy,” Exp. Eye Res. 48,
487–496 (1989).
66. D. B. Rodriguez-Amaya, “A guide to carotenoid analysis in foods,”
Washington, D. C. 20005-580, ILSI PRESS (2001).
67. G. J. Handelman, Z. D. Nightingale, A. H. Lichtenstein, E. J. Schaefer,
and J. B. Blumberg, “Lutein and zeaxanthin concentrations in plasma
after dietary supplementation with egg yolk,” The American journal of
clinical nutrition 70, 247–251 (1999).
68. K. Hesterberg, S. Schanzer, A. Patzelt, W. Sterry, J. W. Fluhr, M. C.
Meinke, J. Lademann, and M. E. Darvin, “Raman spectroscopic
analysis of the carotenoid concentration in egg yolks depending on the
feeding and housing conditions of the laying hens,” J Biophotonics 5,
33–39 (2012).
69. T. R. Hata, T. A. Scholz, I. V. Ermakov, R. W. McClane, F. Khachik,
W. Gellermann, and L. K. Pershing, “Non-invasive raman spectroscopic
detection of carotenoids in human skin,” The Journal of investigative
dermatology 115, 441–448 (2000).
70. M. E. Darvin, I. Gersonde, M. Meinke, W. Sterry, and J. Lademann,
“Non-invasive in vivo determination of the carotenoids beta-
carotene and lycopene concentrations in the human skin using the
raman spectroscopic method, J. Phys. D. Appl. Phys. 38, 2696–2700
(2005).
71. L. Zastrow, N. Groth, F. Klein, D. Kockott, J. Lademann,
R. Renneberg, and L. Ferrero, ‘The missing link—light-induced (280–
1,600 nm) free radical formation in human skin, Skin pharmacology and
physiology 22, 31–44 (2009).
72. V. Stone, H. Johnston, and M. J. Clift, “Air pollution, ultrafine and
nanoparticle toxicology: Cellular and molecular interactions,” IEEE
transactions on nanobioscience 6, 331–340 (2007).
73. I. V. Ermakov, M. R. Ermakova, W. Gellermann, and J. Lademann,
“Noninvasive selective detection of lycopene and beta-carotene in
human skin using raman spectroscopy,” Journal of biomedical optics 9,

332–338 (2004).
74. W. Stahl and H. Sies, “Photoprotection by dietary carotenoids: Concept,
mechanisms, evidence and future development,” Molecular nutrition &
food research 56, 287–295 (2012).
75. N. I. Krinsky and E. J. Johnson, “Carotenoid actions and their relation
to health and disease,” Molecular aspects of medicine; 26, 459–516
(2005).
76. M. E. Darvin, N. N. Brandt, and J. Lademann, “Photobleaching as a
method of increasing the accuracy in measuring carotenoid concentra-
tion in human skin by raman spectroscopy,” Opt Spectrosc þ 109, 205–
210 (2010).
for Medical Diagnosis,” 3rd edition, PM 254, SPIE Press, Bellingham,
WA (2015).
78. M. E. Darvin, I. Gersonde, S. Ey, N. N. Brandt, H. Albrecht, S. A.
Gonchukov, W. Sterry, and J. Lademann, “Noninvasive detection of
beta-carotene and lycopene in human skin using raman spectroscopy,”
Laser Phys 14, 231–233 (2004).
79. T. Herrling, L. Zastrow, J. Fuchs, and N. Groth, “Electron spin
resonance detection of uva-induced free radicals,” Skin pharmacology
and applied skin physiology 15, 381–383 (2002).
80. H. K. Biesalski, C. Hemmes, W. Hopfenmuller, C. Schmid, and H. P.
Gollnick, “Effects of controlled exposure of sunlight on plasma and skin
levels of beta-carotene,” Free radical research 24, 215–224 (1996).
81. M. E. Darvin, I. Gersonde, H. Albrecht, W. Sterry, and J. Lademann,
In vivo raman spectroscopic analysis of the influence of uv radiation on
carotenoid antioxidant substance degradation of the human skin,” Laser
Phys 16, 833–837 (2006).
82. P. Di Mascio, S. Kaiser, and H. Sies, “Lycopene as the most efficient
biological carotenoid singlet oxygen quencher,” Arch Biochem Biophys
274, 532–538 (1989).
83. J. D. Ribaya-Mercado, M. Garmyn, B. A. Gilchrest, and R. M. Russell,
“Skin lycopene is destroyed preferentially over beta-carotene during
ultraviolet irradiation in humans,” The Journal of nutrition 125, 1854–
1859 (1995).
84. M. C. Meinke, R. Muller, A. Bechtel, S. F. Haag, M. E. Darvin, S. B.
Lohan, F. Ismaeel, and J. Lademann, “Evaluation of carotenoids and
reactive oxygen species in human skin after uv irradiation: A critical
comparison between in vivo and ex vivo investigations,” Experimental
dermatology 24, 194–197 (2015).
85. K. Danno, N. Mori, K. Toda, T. Kobayashi, and A. Utani, “Near-
infrared irradiation stimulates cutaneous wound repair: Laboratory
experiments on possible mechanisms,” Photodermatology, photoimmu-
nology & photomedicine 17, 261–265 (2001).
238 Chapter 4
86. M. Y. Akhalaya, G. V. Maksimov, A. B. Rubin, J. Lademann, and

M. E. Darvin, “Molecular action mechanisms of solar infrared
radiation and heat on human skin,” Ageing research reviews 16, 1–11
(2014).
87. M. E. Darvin, I. Gersonde, H. Albrecht, L. Zastrow, W. Sterry, and
J. Lademann, “In vivo raman spectroscopic analysis of the influence of ir
radiation on the carotenoid antioxidant substances beta-carotene and
lycopene in the human skin. Formation of free radicals,” Laser Phys.
Lett. 4, 318–321 (2007).
88. M. E. Darvin, S. Haag, M. Meinke, L. Zastrow, W. Sterry, and
J. Lademann, “Radical production by infrared a irradiation in human
tissue,” Skin pharmacology and physiology 23, 40–46 (2010).
89. M. E. Darvin, S. F. Haag, J. Lademann, L. Zastrow, W. Sterry, and
M. C. Meinke, “ Formation of free radicals in human skin during
irradiation with infrared light,” The Journal of investigative dermatology
130, 629–631 (2010).
90. T. Jung, A. Hohn, H. Piazena, and T. Grune, “Effects of water-filtered
infrared a irradiation on human fibroblasts,” Free radical biology &
medicine 48, 153–160 (2010).
91. H. Piazena and D. K. Kelleher, “Effects of infrared-a irradiation on
skin: Discrepancies in published data highlight the need for an exact
consideration of physical and photobiological laws and appropriate
experimental settings,” Photochem Photobiol 86, 687–705 (2010).
92. M. E. Darvin, W. Sterry, J. Lademann, and A. Patzelt, “Alcohol
consumption decreases the protection efficiency of the antioxidant
network and increases the risk of sunburn in human skin,” Skin
pharmacology and physiology 26, 45–51 (2013).
93. H. B. Vierck, M. E. Darvin, J. Lademann, A. Reisshauer, A. Baack,
W. Sterry, and A. Patzelt, “The influence of endurance exercise on the
antioxidative status of human skin,” Eur J Appl Physiol 112, 3361–3367,
(2012).
94. M. E. Darvin, A. Patzelt, F. Knorr, U. Blume-Peytavi, W. Sterry, and
J. Lademann, “One-year study on the variation of carotenoid antioxi-
dant substances in living human skin: Influence of dietary supplementa-
tion and stress factors,” Journal of biomedical optics 13 (2008).
95. H. Lademann, B. Gerber, D. M. Olbertz, M. E. Darvin, L. Stauf,
K. Ueberholz, V. Heinrich, J. Lademann, and V. Briese, “Non-invasive
spectroscopic determination of the antioxidative status of gravidae and
neonates,” Skin pharmacology and physiology 28, 189–195 (2015).
96. S. Jung, M. E. Darvin, H. S. Chung, B. Jung, S. H. Lee, K. Lenz, W. S.
Chung, R. X. Yu, A. Patzelt, B. N. Lee, W. Sterry, and J. Lademann,
“Antioxidants in asian-korean and caucasian skin: The influence of
nutrition and stress” Skin pharmacology and physiology 27, 293–302 (2014).
97. M. E. Darvin, J. W. Fluhr, S. Schanzer, H. Richter, A. Patzelt, M. C.
Meinke, L. Zastrow, K. Golz, O. Doucet, W. Sterry, and J. Lademann,
“Dermal carotenoid level and kinetics after topical and systemic

administration of antioxidants: Enrichment strategies in a controlled
in vivo study,” Journal of dermatological science 64, 53–58 (2011).
98. R. X. Yu, W. Kocher, M. E. Darvin, M. Buttner, S. Jung, B. N. Lee,
C. Klotter, K. Hurrelmann, M. C. Meinke, and J. Lademann,
“Spectroscopic biofeedback on cutaneous carotenoids as part of a
prevention program could be effective to raise health awareness in
adolescents,” J. Biophotonics 7, 926–937 (2014).
99. M. E. Darvin, I. Gersonde, H. Albrecht, W. Sterry, and J. Lademann,
“Resonance raman spectroscopy for the detection of carotenolds in
foodstuffs. Influence of the nutrition on the antioxidative potential of the
skin,” Laser. Phys. Lett. 4, 452–456 (2007).
100. M. E. Darvin, H. Richter, S. Ahlberg, S. F. Haag, M. C. Meinke,
D. Le Quintrec, O. Doucet, and J. Lademann, “Influence of
sun exposure on the cutaneous collagen/elastin fibers and carotenoids:
negative effects can be reduced by application of sunscreen,”
J. Biophoton. 7(9), 735–743 (2014).
101. S. T. Mayne, B. Cartmel, S. Scarmo, L. Jahns, I. V. Ermakov, and
W. Gellermann, “Resonance raman spectroscopic evaluation of skin
carotenoids as a biomarker of carotenoid status for human studies,”
Arch Biochem Biophys 539, 163–170 (2013).
102. A. C. Lauer, N. Groth, S. F. Haag, M. E. Darvin, J. Lademann, and
M. C. Meinke, “Dose-dependent vitamin c uptake and radical
scavenging activity in human skin measured with in vivo electron
paramagnetic resonance spectroscopy,” Skin pharmacology and physiol-
ogy 26, 147–154 (2013).
103. A. C. Lauer, N. Groth, S. F. Haag, M. E. Darvin, J. Lademann, and
M. C. Meinke, “Radical scavenging capacity in human skin before and
after vitamin c uptake: An in vivo feasibility study using electron
paramagnetic resonance spectroscopy,” The Journal of investigative
dermatology 133, 1102–1104 (2013).
104. J. Lademann, P. J. Caspers, A. van der Pol, H. Richter, A. Patzelt,
L. Zastrow, M. Darvin, W. Sterry, and J. W. Fluhr, “In vivo raman
spectroscopy detects increased epidermal antioxidative potential with
topically applied carotenoids,” Laser Phys. Lett. 6, 76–79 (2009).
105. M. E. Darvin, J. W. Fluhr, M. C. Meinke, L. Zastrow, W. Sterry, and
J. Lademann, “Topical beta-carotene protects against infra-red-light-
induced free radicals,” Experimental dermatology 20, 125–129 (2011).
106. M. E. Darvin, I. Gersonde, H. Albrecht, S. A. Gonchukov, W. Sterry,
and J. Lademann, “Determination of beta carotene and lycopene
concentrations in human skin using resonance raman spectroscopy,”
Laser Phys 15, 295–299 (2005).
107. M. E. Darvin, J. W. Fluhr, P. Caspers, A. van der Pool, H. Richter,
A. Patzelt, W. Sterry, and J. Lademann, “In vivo distribution of
carotenoids in different anatomical locations of human skin:
240 Chapter 4
Comparative assessment with two different raman spectroscopy

methods,” Experimental dermatology 18, 1060–1063 (2009).
108. J. W. Fluhr, P. Caspers, J. A. van der Pol, H. Richter, W. Sterry,
J. Lademann, and M. E. Darvin “Kinetics of carotenoid distribution in
human skin in vivo after exogenous stress: Disinfectant and wira-induced
carotenoid depletion recovers from outside to inside,” Journal of
biomedical optics 16(3), 035002 (2011).
Gerald Lucassen received a PhD degree (cum laude) in

Physics from the University of Twente, the Netherlands. The
topic of his PhD thesis (1992) was on polarization sensitive
coherent Raman spectroscopy on biomolecules in solutions.
After earning his PhD, he worked at the Laser Centre of the
Amsterdam Medical Centre on the physics and Monte Carlo
modeling of laser removal of portwine stains. He joined
Philips Research, Eindhoven, NL in 1995 where he has since
worked on various topics related to skin characterization using microscopic
and spectroscopic techniques. In 2011, he joined a venture on the photonic
needle in Philips Healthcare.
Maxim E. Darvin studied at the Moscow State Engineering

Physics Institute, the Faculty of Experimental and Theoreti-
cal Physics, Department of Medical Physics, where he
completed his master’s degree in physics. After his postgrad-
uate study, he was awarded the titles Dr. rer. med. in 2007 at
the Humboldt University of Berlin, Germany and Dr. rer.
nat. in 2010 at the Saratov State University, Russia. His
main interests are medical physics, laser techniques, quantum
electronics, spectroscopy, biophysics, skin physiology.
Juergen Lademann studied at the Moscow State University,

the Physical Faculty, Quantum Electronics Department,
where he completed his master’s degree. In the year 2000, he
was appointed Professor of Dermatology at the Charité
University Hospital Berlin.
He is the Editor of the international journal “Skin
Pharmacology and Applied Skin Physiology” and the
President of the “International Society of Skin Pharmacol-
ogy and Physiology.”
Chapter 5
Fluorescence Technologies
in Biomedical Diagnostics
Herbert Schneckenburger
Institute of Laser Technology in Medicine and Metrology, University of Ulm, Ulm,
Germany
Institute of Applied Research, Aalen University, Aalen, Germany
Wolfgang S. L. Strauss, Karl Stock, and Rudolf Steiner

Institute for Laser Technology in Medicine and Metrology, University of Ulm, Ulm,
Germany
5.1 Introduction
5.1.1 Fundamentals
Non-destructive optical techniques are increasingly used in biomedical
diagnostics. Fundamental mechanisms between optical radiation and
biological specimens are absorption, reflection, elastic or inelastic light
scattering, and luminescence. The latter is subdivided into (1) fluorescence,
corresponding to allowed optical transitions with rather high quantum yields
and short (nanosecond) lifetimes and (2) phosphorescence corresponding to
optical transitions with low probability (“forbidden transitions”) and thus
with low quantum yields and long lifetimes in the microsecond to millisecond
range.
Absorption of light is connected with an electronic transition from the
ground state S0 to an excited state of a molecule. Light passing through a layer
of thickness d is thereby attenuated according to the equation
IðlÞ ¼ I 0 10εðlÞcd ¼ I 0 ema d (5.1)
where I (l) is the transmitted light intensity (quantum flux), I 0 is the incident
intensity, ε(l) is the molar extinction coefficient, and c is the concentration of
absorbing molecules. In scattering samples, the absorption coefficient ma and
241
242 Chapter 5
the scattering coefficient ms [omitted in Eq. (5.1)] sum up, thus causing a
further reduction of transmitted light, as described in detail elsewhere1 (and
references therein).
Fluorescence arises upon light absorption and is related to an electronic
transition from the first excited singlet state S1 to the ground state of a
molecule. Its intensity (quantum flux) corresponds to
I F ðlÞ ¼ I 0 ½1 10εðlÞcd hV∕4p (5.2)
with h being the fluorescence quantum yield and V the solid angle of detection
of isotropic fluorescence radiation. In the case of thin samples, e.g., cell
monolayers or biopsies of a few micrometers in diameter, Eq. (5.2) can be
approximated by
I F ðlÞ ¼ I 0 ln 10 εðlÞc dh V∕4p: (5.3)
This implies that fluorescence intensity is proportional to the concentra-

tion and the fluorescence quantum yield of the absorbing molecules. In
scattering media, the path lengths of scattered and unscattered photons within
the sample are different, and Eqs. (5.2) and (5.3) have to be modified.
However, in virtually homogenous thin samples, the linearity between IF, c,
and h is still maintained.
5.1.2 Potential diagram

Energies of the electronic states of a molecule are complex functions of the
nuclear distances of relevant atoms, usually forming “potential wells,” as
shown in Fig. 5.1 for the ground state (S0) and the first excited state (S1). Each
well contains a larger number of vibrational levels ni that further split into
numerous rotational levels (omitted in Fig. 5.1) of the molecule. Electronic
transitions occur in the “vertical direction,” since during their short duration,
nuclear coordinates do not change (Franck-Condon principle). Usually
electronic transitions originate from vibronic ground states (excitation: S0, n0;
fluorescence; S1, n0). The probability of each transition corresponds to the
square of the transition dipole moment and is determined by an overlap of the
corresponding electronic wave functions in the ground state and the excited
state of the molecule. Therefore, absorption and fluorescence spectra originate
from a superposition of several transitions, often resulting in broad spectral
bands. From Fig. 5.1 one can deduce that the so-called 0-0 transition between
the lowest vibrational levels is only a little pronounced, since the overlap
between corresponding wave functions is very low. Therefore, fluorescence
spectra are usually shifted to lower energies DW corresponding to higher
wavelengths l ¼ DW/hc as compared with absorption spectra (h ¼ Planck’s
constant, c ¼ velocity of light). This phenomenon is called “Stokes shift.”
Fluorescence Technologies in Biomedical Diagnostics 243
S1
ν2
energy
ν1
ν0
S0
ν2
ν1
ν0
nuclear distance
Figure 5.1 Potential diagram of electronic states (S0, S1) with their vibrational levels (ni).
Electronic wave functions and optical transitions are indicated (excitation: S0 n0 ! S1 nn;
fluorescence: S1 n0 ! S0 nn).
5.1.3 Jablonski diagram and kinetic rates

When plotting the potential curves without considering the variable nuclear
distances, the different molecular states can be illustrated in a Jablonski
diagram, as shown in Fig. 5.2. Excitation usually occurs from the singlet
ground state S0 to various vibronic levels of the excited singlet states Sn, from
where fast non-radiative transitions (“internal conversion”) occur within the
femtosecond time range to the lowest excited state S1. From S1, various
transitions can be distinguished: fluorescence to the ground state S0 (including
its vibrational states) with a rate kF, internal conversion to the ground state S0
(rate kIC), intersystem crossing from the singlet to the triplet state T1 (rate
kISC), and non-radiative energy transfer to adjacent molecules (rate kET). All
these rates sum up according to
k ¼ k F þ k IC þ k ISC þ k ET ¼ 1∕t, (5.4)
where t is the lifetime of the excited state S1. The ratio kF/k corresponds to the
fluorescence quantum yield h. Although by optical spectroscopy, only
radiative transitions can be monitored, changes of kIC or kET are
often deduced from fluorescence lifetime measurements. For example, an
244 Chapter 5
Singlet Triplet
S
2
kET
S1
kISC T1
kF kIC
S0
Figure 5.2 Jablonski diagram of molecular energy levels and transition rates. Straight
lines: radiative transitions; waving lines: non-radiative transitions.
aggregation of porphyrin molecules (which are used for photodynamic

therapy of tumors) has been deduced from a pronounced shortening of their
fluorescence lifetime,2,3 and impairments in photosynthetic antenna com-
plexes have been related to a prolongation of the fluorescence lifetime of
chlorophyll molecules, when intermolecular energy transfer towards the
reaction centers was impeded.3–6 It should be emphasized that the radiative
transition T1 ! S0 is spin-forbidden, and only within a few specific molecules
is this transition becoming prominent. For example, chelates of lanthanides
are phosphorescence markers with a lifetime of their excited states around
1 millisecond, which can be easily distinguished from the autofluorescence of
cells and tissues.7,8
5.1.4 Fluorescence anisotropy

Transition dipole moments have defined orientations within a molecule. Upon
excitation with polarized light, one preferentially excites those molecules,
whose transition dipoles are parallel to the electric field vector of incident
light. This selective excitation of an oriented population of molecules results in
partially polarized fluorescence, which is described by the degree of
polarization
P ¼ ðI ll I ⊥ Þ∕ðI ll þ I ⊥ Þ (5.5)
or by fluorescence anisotropy
r ¼ ðI ll I ⊥ Þ∕ðI ll þ 2 I ⊥ Þ: (5.6)
with Ill and I⊥ being the fluorescence intensities of light polarized parallel or
perpendicular to the exciting electric field vector, respectively. Usually P and r
depend on the time interval between excitation and fluorescence detection,

since during the lifetime of their excited states, many molecules change their
orientation by rotation (“rotational diffusion”). From time-resolved measure-
ments of fluorescence anisotropy, a time constant tr of rotational diffusion can
be determined that (for spherically shaped molecules) is correlated with the
volume V of the molecule and the viscosity h of its environment according to
tr ¼ h V ∕k B T, (5.7)
where kB is the Boltzmann constant and T the absolute temperature. Time
constants of rotational diffusion of about 13 ns were correlated with a
molecular weight of proteins around 50,000 daltons,9 whereas a time constant
around 300 ps was attributed to an aggregated species of a photosensitizing
porphyrin (protoporphyrin) with a 1.6-nm diameter.10 A pronounced decrease
of fluorescence anisotropy with increasing temperature was recently described
for the membrane marker laurdan, thus indicating an increase of membrane
fluidity.11
5.2 Intrinsic and Extrinsic Fluorescence

5.2.1 Intrinsic fluorophores
Following excitation by ultraviolet light (l ≤ 300 nm), fluorescence of proteins
as well as of nucleic acids can be observed. Fluorescence quantum yields of all
nucleic acid constituents, however, are around 10–4 to 10–5 corresponding to
lifetimes of the excited states in the picosecond time range.12 Since very
selective fluorescent DNA and RNA dyes have recently become available,
autofluorescence of nucleic acids plays only a minor role in biomedical
diagnostics. Autofluorescence of proteins is related to the amino acids
tryptophan, tyrosin, and phenylalanine with absorption maxima at 280 nm,
275 nm, and 257 nm, respectively and emission maxima between 280 nm
(phenylalanine) and 350 nm (tryptophan).9,13,14 Usually the protein spectrum
is dominated by tryptophan. Extracellular autofluorescence arising from
collagen or elastic fibres is excited between 300 nm and 400 nm and shows
broad emission bands between 400 nm and 600 nm with maxima around
400 nm, 430 nm, and 460 nm.15–17 In particular, fluorescence of collagen and
elastin can be used to distinguish various types of tissues, e.g., epithelial and
connective tissue.18
A key role is assigned to the coenzyme nicotinamide adenine dinucleotide
(NAD), whose reduced form (NADH) is excited in a wavelength range
between 330 and 370 nm. NADH is most concentrated within mitochondria
where it is oxidized within the respiratory chain located within the inner
mitochondrial membrane. Malfunction of specific enzyme complexes within
the respiratory chain may have an impact on the concentration ratio between
the oxidized and the reduced form and, therefore, on fluorescence
246 Chapter 5
intensity.18–21 So far, the fluorescence intensity of NADH has been proposed

to be an appropriate parameter for detection of ischemic22,23 or neoplas-
tic24–27 tissues. NADH fluorescence was also correlated with the metabolic
function of glioblastoma28 or muscle cells.29 Therefore, it appears possible to
use this coenzyme in the future for the detection of mitochondrial
myopathies.30–33 Further potential applications may include various dis-
eases33,34 or stress syndromes35 where energy metabolism is affected.
Fluorescence of free and protein-bound NADH has been shown to be
sensitive to oxygen concentration.36 Therefore, both molecular species were
shown to be appropriate parameters for measuring metabolic adaptations of
organisms to environmental changes, e.g., variations of oxygen supply.37 In
addition to NADH, the reduced form of nicotinamide adenine dinucleotide
phosphate (NADPH) contributes to cellular autofluorescence. NADPH
originates mainly from the cytoplasm and exhibits almost identical excitation
and emission spectra as NADH. NADPH fluorescence can be used to
monitor oxygen consumption during respiratory bursts,38 but is generally
supposed to play only a minor role in optical diagnostics. Flavin
mononucleotide (FMN) and dinucleotide (FAD) with excitation maxima
around 380 nm and 450 nm have also been reported to contribute to intrinsic
cellular fluorescence.19,39 Since the oxidized form of flavin molecules is excited
preferentially, their fluorescence can be used in a similar manner as NADH
fluorescence to monitor the redox states of these molecules and to probe the
function of the mitochondrial respiratory chain. When using an excitation
wavelength in the near ultraviolet range, flavin fluorescence of cultivated cells,
however, has been found to be weak in comparison with fluorescence of
NADH.17,19
Porphyrin molecules, e.g., protoporphyrin, coproporphyrin or uropor-
phyrin occur within the pathway of biosynthesis of hemoglobin, myoglobin,
and cytochromes. Usually their accumulation in cells or tissues is very low.
However, abnormalities in heme biosynthesis, occurring in the cases of
porphyrias and some hemolytic diseases, may enhance the porphyrin level
within tissues considerably (Ref. 40 and references therein). In addition,
several bacteria, e.g., Propionibacterium acnes41 or bacteria within dental
caries lesions,42,43 have been reported to accumulate considerable amounts
of protoporphyrin. Therefore, caries detection based on measurements of
intrinsic fluorescence appears to be a promising method.44
5.2.2 Fluorescent markers

Fast progress in organic chemistry since 1850 provided the basis for the
synthesis of various fluorescent probes. As a consequence, fluorescein was
synthesised in 1871 by A. von Baeyer. It was the first dye used in vivo to track
the secretion of aqueous humour in the eye. Fluorescence microscopy was
introduced in 1911 and 1913 by O. Heimstaedt and H. Lehmann, and
experimental cytology with exogenously applied fluorescing dyes started in

1914 with the dye-binding studies of S. von Provazek.45,46 To date, a plethora
of fluorescing dyes covering the entire visible spectral range can be applied for
probing of cell anatomy, cell physiology, or even medical diagnostics.
Generally, photochemical/photophysical properties of ideal fluorescent
probes include high fluorescence quantum yields and high photostability. In
addition, excitation in the visible spectral region is desirable to avoid interference
with autofluorescence from biological material and background fluorescence
from optical components. Until now, fluorescein can be regarded as the most
dominant fluorophore, with an excitation maximum at approximately 494 nm
and an emission maximum around 520 nm. Amine-reactive derivatives of
fluorescein are still among the most common derivatization reagents for the
covalent labelling of proteins, e.g., immunoglobulins used for immunocyto-
chemistry. Fluorescein is characterised by a relatively high absorptivity, excellent
fluorescence quantum yield, and good water solubility. However, major
drawbacks are pronounced photobleaching as well as pH-sensitive fluorescence
emission. Various attempts have been made to overcome the limitations of
fluorescein (derivatives), resulting in the development of photostable
pH-insensitive probes, e.g., the promising Alexa Fluor dyes.47
In vitro diagnosis is often performed with fluorescent labels due to its
superior sensitivity in comparison with established colorimetric assays. Most
common applications are the determination of enzyme activities or the
detection of biomolecules like nucleic acids (e.g., with ethidium bromide or less
mutagenic SYBR stains) or proteins.48,49 In addition, nucleic acids or proteins
labelled with fluorescent dyes (e.g., cyanines Cy3 or Cy5) provide the basis for
emerging fields like genomics or proteomics.50 Fluorescent probes have been
extensively used for the tracking of subcellular organelles, which requires a
highly selective accumulation of fluorophores within these structures.
Commonly used organelle markers are rhodamine 123 (mitochondria), acridine
orange (acidic organelles, e.g., lysosomes) and carbocyanines (endoplasmatic
reticulum).51,52 Specific labels responding to the physiological state of a cell,
e.g., internal ion concentration or membrane potential, usually change their
optical properties in response to variations of their environment. Fluorescent
indicators currently exist for signal transduction (e.g., calcium ion concentra-
tion with calcium probes indo-1 or fluo-3), pH (e.g., BCECF), reactive oxygen
species (e.g., conversion of a reduced leuco form of fluorescein to the
fluorescent molecule) or membrane potential (e.g., styrylpyridinium dyes).51
A further promising technique consists of staining with so-called quantum
dots, i.e., luminescent nanoparticles of semiconductors such as CdS, ZnS or
PbS. These particles had originally been used in material sciences53 and were
recently introduced as biological labels.54 In contrast to most organic dyes,
quantum dots are characterized by low photobleaching rates and by emission
spectra depending on the size of these particles. Therefore, their luminescence
248 Chapter 5
properties can be easily varied, however, their photophysical properties should

be taken into account. For example, photooxidation of CdSe quantum dots
results in some shrinkage of these particles and a concomitant blue shift of the
emission spectrum.55 Typical luminescence lifetimes of quantum dots are in
the range of 150 ns to 25 ms, i.e., considerably longer than those of most
organic dyes as well as of cellular autofluorescence. Therefore, it might be
rather easy to distinguish their emission from the intrinsic fluorescence of cells
or tissues by time-resolving methods.
The most exciting recent development in the use of fluorescent probes for
biological studies has been the introduction of fluorescent proteins as probes.
A green fluorescent protein (GFP) is naturally produced by the jellyfish
Aequorea Victoria.56 After cloning of the GFP gene, various GFP variants
with different excitation and emission properties have been produced.
According to their emission maxima in the blue, yellow, or even red spectral
region, blue, yellow or red fluorescent proteins are distinguished. By fusion of
genes coding for a specific cellular protein and GFP or its variants, functional
as well as fluorescent protein chimera were created, allowing a site-specific
tracking in living cells or even whole organisms.57,58 GFP calmodulin chimera
have further been used for the determination of physiological Ca-ion
concentrations.59
Application of fluorescent probes for diagnostic purposes in humans is
restricted to fluorescein or indocyanine green based fluorescence angiography
of blood volume.60
5.3 Spectroscopic, Microscopic, and Imaging Techniques

5.3.1 Fluorescence spectroscopy
Fluorescence spectra often give detailed information on fluorescent molecules,
their conformation, binding sites, and interaction within cells and tissues.
Fluorescence intensity can be measured either as a function of excitation or
emission wavelength. In the second case the fluorescence spectrum IF(l) is
detected, which is typical for any fluorophore in its microenvironment and,
thus commonly used in fluorescence diagnostics. In the first case the
fluorescence excitation spectrum, whose information is more complex, is
recorded. Often it just reflects the absorption spectrum of a fluorophore. If,
however, the emission spectra of two or more fluorophores overlap at the
detection wavelength, the fluorescence excitation spectrum does not display
the sum of their absorption spectra, but the sum of the absorption spectrum of
each fluorophore multiplied with its fluorescence quantum yield. Fluorescence
excitation spectra are particularly useful for measurements of non-radiative
energy transfer from a molecule A (energy donor, absorbing optical radiation)
to a molecule B (energy acceptor, fluorescent), as reported in detail in
Section 5.6. If fluorescence is measured within the emission band of the
Excitation Emission Photomultiplier

Grating Grating
Lamp Lens Slit 1

Slit 4
Slit 2
Slit 3
Concave
Mirror Sample
Figure 5.3 Set-up for fluorescence excitation and emission spectroscopy.
acceptor, the fluorescence excitation spectrum exhibits the absorption bands

of both donor and acceptor molecules. From the fluorescence ratio detected
for both absorption bands the quantum yield of energy transfer can be
calculated.
The principle of a fluorescence spectrometer is shown in Fig. 5.3. The
excitation light (e.g., xenon high pressure lamp with a continuous spectrum) is
focused to the entrance slit of an excitation monochromator, becomes
spectrally resolved, and is finally used to illuminate the sample. Part of the
sample’s isotropic fluorescence radiation is focused into the slit of the emission
monochromator and is detected as a function of wavelength. Often, in both
monochromators, concave reflection gratings are used for spectral resolution
as well as for focusing the incident light onto the exit slits, thus avoiding
further collimating optics. For both gratings the equation
d ðsin ε sin aÞ ¼ kl (5.8)
has to be fulfilled with d being the grating constant, k an integer number

(0,1,2,...), ε the incident angle, and a the exit angle of light with respect to the
normal of the grating surface. Generally, for excitation spectroscopy the
excitation grating is rotated, whereas for emission spectroscopy various
modalities can be used: either the emission grating is rotated as well, and the
wavelength reaching the detector (photomultiplier) via slit 4 is tuned (scanning
spectrometer), or slit 4 is replaced by a diode array or charge-coupled device
(CCD) camera, such that the whole emission spectrum is registered
simultaneously (optical multichannel analyzer, OMA). The spectral resolution
of a scanning monochromator (1 nanometer or less) is determined by its slit
widths and is usually higher than the resolution of an OMA ($1 nm), which is
determined by the size of individual diodes or pixels. In addition, the
photomultiplier of a scanning monochromator is more sensitive than a diode
250 Chapter 5
array or a CCD camera. The main advantage of OMAs is that optical spectra
can be recorded very rapidly and repeatedly with sequences in the millisecond
range as compared with a few seconds for scanning monochromators. This
permits on-line detection of fast molecular or cellular kinetics, if the
fluorescence intensity is strong enough. Often a compromise between spectral
resolution, sensitivity, and recording time has to be found in order to detect
comparably weak signals of fluorophores in cells or tissues. As a first step
toward that direction, a “Rapid Scan Spectrometer” (RSS) has been recently
developed.61 It is based on the scanning principle using a mirror that oscillates
at 100 Hz. Therefore, fluorescence spectra can be recorded every 10 ms using
the sensitivity of a photomultiplier.
Usually, fluorescence spectra are “uncorrected,” i.e., they contain
wavelength dependent sensitivities of monochromator and detector (emission
spectroscopy) or excitation source and monochromator (excitation spectros-
copy). Correction may not be necessary if only relative changes of individual
emission bands are measured. However, for getting correct spectral intensities
IF (l) a reference source with a well known spectrum S(l) (e.g., a black body)
has to be used. Both signals, IF(l) and S(l), are “folded” by a response
function G(l). When measuring the experimental functions IF0(l) ¼ IF(l)
G(l) and S0(l) ¼ S(l) G(l), the “true” fluorescence spectrum can be calculated
using the algorithm
I F ðlÞ ¼ I 0F ðlÞ SðlÞ∕S 0 ðlÞ: (5.9)
Fluorescence spectrometers for in vivo diagnostics are commonly based on
fiber-optic systems.14,62–64 The excitation light of a lamp, LED, or a laser is
guided to the tissue (e.g., some specific organ) via a glass fiber using appropriate
optical filters (instead of an excitation monochromator). Fluorescence spectra
are usually measured either via the same fiber or via a second fiber or fiber
bundle in close proximity to the excitation fiber. Scanning monochromators or
OMA systems as reported above are used for emission spectroscopy.
5.3.2 Fluorescence microscopy

Due to an increasing number of fluorescent dyes with specific staining of
molecules (e.g., nucleic acids or proteins) or cell organelles, microscopic
techniques are presently gaining more and more importance. In addition to
high-resolution images, fluorescence spectra or fluorescence lifetimes may give
further information. A typical microspectrofluorometer is depicted in
Figure 5.4. Monochromatic excitation light is collimated in the entrance
plane of the microscope, deflected by a dichroic mirror, and focused on the
sample by the objective lens.
Due to its longer wavelength, fluorescence radiation passes through the
dichroic mirror and is again focused in the image plane of the microscope. At
this position, an image detector (e.g., CCD camera) may be placed. For
Fluorescence Technologies in Biomedical Diagnostics
Photomultiplier
Filter
Monochromator 2 Monochromator 1
(380-850nm) (240-850nm)
Variable
Diaphragm
Photometer
M2
XBO-Lamp
Sample
M1
Halogen Lamp
Figure 5.4 Microspectroscopic set-up for fluorescence excitation and emission spectroscopy with additional transillumination (M1 ¼ mirror;
M2 ¼ dichroic mirror).
251
252 Chapter 5
fluorescence spectroscopy, the image detector is replaced by an adjustable

diaphragm (for selection of specific objects, e.g., individual cells) and a
monochromator. A scanning monochromator with a photomultiplier can be
used as well as an OMA system. The entrance slit of the monochromator is
placed either in the exit pupil of the microscope or in the image plane. In the
second case, this slit determines the measured part of the sample, thus
avoiding a separate diaphragm. By transillumination of the sample with a
lamp emitting a continuous spectrum (e.g., from a halogen lamp) specific
areas can be selected for fluorescence measurements. The optical set-up can be
modified in many ways, e.g., by replacing the excitation monochromator by
interference filters or by replacing the excitation lamp by a laser that can be
focused to a diffraction limited spot with a minimum radius
r ¼ 1.22 f lex ∕D, (5.10)
where f represents the focal length and D is the aperture of the objective lens
of the microscope. Assuming f ¼ 1.6 mm, D ¼ 6 mm, and the laser wavelength
lex ¼ 488 nm, a radius r ¼ 160 nm of the focused laser beam is calculated
according to Eq. (5.10). Usually this radius is slightly smaller than the lateral
(diffraction limited) resolution in conventional fluorescence microscopy
d min ¼ 0.61 lem ∕AN , (5.11)
since the excitation wavelength lex is generally shorter than the emission
wavelength lem. (dmin is the minimum object size that can be resolved, whereas
AN corresponds to the numeric aperture of the objective lens). The difference
between r and dmin also explains why the lateral resolution in laser scanning
microscopy is slightly enhanced compared to conventional microscopy.
Considering a depth of focus around or even below 1 mm, large parts of a
sample are out of focus in conventional microscopy. These out-of-focus parts
can be eliminated using specific methods of wide-field 3D microscopy or
confocal laser scanning microscopy as reported in Sections 5.7 and 5.8.
5.3.3 Imaging techniques

Conventional Vidicon cameras have been replaced by charge-coupled device
(CCD) cameras in recent years. A typical 3-chip color CCD camera (Sony,
DXC950P) has 752 582 image elements (pixels) on a surface of 6.4 mm
4.8 mm corresponding to about 8.5 mm 8.5 mm per pixel. For correct
imaging, the pixel size should not be larger than half of the minimum object
size. Considering a microscope objective lens of 100 magnification and a
numerical aperture of 1.30, dmin at l ¼ 550 nm is 212 nm in the object plane
corresponding to 21.2 mm in the image plane. This means that microscopic
resolution is well retained by the camera system.
Sensitivity (i.e., threshold towards noise) of high-performance CCD

cameras with air or Peltier cooling is below 1 lx, corresponding to about
10–7 W/cm2 or 10–13 W/Pixel, in the green spectral range. Assuming an in vitro
experiment with a power density of irradiation of 50 mW/cm2, a cell
monolayer with a diameter d ¼ 20 mm, a concentration of fluorophores c ¼
10–5 mol/L, an extinction coefficient ε ¼ 105 L /(mol cm), and a solid angle of
fluorescence detection V ¼ 1 sr, one can deduce from Eq. (5.3) that the
minimum fluorescence quantum yield of a fluorophore should be around 0.01
for detection by a CCD camera. However, considerable enhancement of the
sensitivity (by a factor 1.00010.000) can be achieved by signal amplification
on the camera chip (electron multiplying CCD cameras65). A higher
sensitivity in comparison with conventional CCD cameras is also attained
by CMOS image sensors, which, in addition, are characterized by an excellent
linearity over many orders of magnitude and high speed image acquisition
down to the nanosecond range.66
In addition, image intensifying systems can be used for detection of
fluorophores with either low quantum yield or at low concentration.
Amplification factors around 103 are attained by multichannel plates. In this
case photoelectrons are emitted from a photocathode and amplified within
some alveolar structure at a voltage of some hundred volts. Each secondary
electron may cause the emission of several photons from a phosphor screen to
which high voltage (up to some kV) is applied again. Thus, amplification
factors of 104 to 107 are attained when using one or two multichannel plates.
This corresponds to an overall sensitivity around 10–14 W/cm² corresponding
to about 10–20 W/pixel. Photon counting is thus possible, if the noise level can
be kept sufficiently low, e.g., by cooling with a Peltier element or liquid
nitrogen. First reports on single photon counting imaging were published
almost 20 years ago.67 More recently, photon counting imaging has become a
valuable technique in single molecule detection.68
Fluorescence imaging can also be performed with high temporal or
spectral resolution. In the first case, fast gating or modulation techniques have
been used (see below), whereas in the second case spectral imaging systems
have been applied successfully. For example, a Sagnac interferometer based
on Fourier spectroscopy in combination with a CCD camera has been
described,69 where fluorescence light is split into two paths of opposite
directions. By moving the beam splitter before bringing the two beams
together, a phase shift occurs that depends on the wavelength of radiation.
Concomitantly, the fluorescence intensity of each pixel is wavelength
dependent and contains a spectral signature. However, several seconds or
minutes are needed for the acquisition of one “three-dimensional” image I(x,y,l).
Multipixel Fourier transform spectroscopy has been applied successfully for
measuring photosensitizers, which are used for photochemotherapy of tumors
within single cells. Monomeric species of protoporphyrin were detected in the
254 Chapter 5
plasma membrane, whereas aggregated species accumulated within endoso-

mal and lysosomal compartments.70 The hydrophilic tetraphenyl porphyrine
TPPS4 was mainly localized within lysosomes, whereas the more hydropho-
bic TPPS1 was concentrated in a Golgi-like complex and in the nuclear
envelope. TPPS4, but not TPPS1, was re-distributed within the cell after light
exposure.71 More recently, different conformations (unfolded/folded) of
green fluorescent protein (GFP) were found in different cell compartments
using multipixel spectral imaging. Thus, pathways of GFP formation,
intracellular transport, and changes of conformation could be studied within
single cells.72
5.4 Time-Resolved Fluorescence Spectrosopy and Imaging

Advantages of time-resolved compared to continuous wave (CW) fluores-
cence measurements are numerous:
• Fluorescence signals of different fluorophores with overlapping
emission spectra might be resolved;
• Autofluorescence of cells or tissues can often be suppressed;
• Kinetic reactions, e.g., intermolecular energy transfer, can be evaluated.
Fluorescence lifetimes of most organic fluorophores usually vary in a time
range between a few hundreds of picoseconds and a few tens of nanoseconds.
Therefore, techniques presented in this article are limited to this time range. In
principle, time-resolved and frequency-resolved techniques provide the same
kind of information.
5.4.1 Time-correlated single photon counting

The most popular time-resolving technique is single photon counting. The
sample is excited by subnanosecond pulses of a laser or a flashlamp at high
repetition rate (kilohertz to megahertz). Single fluorescence photons are
detected by a fast photomultiplier, whose signal is amplified and fed to a
discriminator (“stop signal”). The corresponding pulse of the light source is
usually detected by a fast photodiode and fed to another discriminator;
alternatively, a synchronous electrical trigger signal from the light source can
be used (“start signal”). The time between the start and stop signals is
measured by a time-to-amplitude converter creating a voltage that is
proportional to the delay time of each photon with respect to the exciting
light pulse. Integration over many single photons (typically 105 to 106) gives
the fluorescence decay curve (for details see Refs. 3 and 73). Since after each
excitation pulse only one photon can be measured, the simultaneous detection
of two or more photons should be avoided, e.g., by reduction of the counting
rate to 5% or less of the excitation rate. Alternatively the detection system can
be operated in a “reverse mode,” i.e., each single photon gives a start signal
that is followed by a (delayed) stop signal of the excitation pulse.
Fluorescence intensity (number of photon counts per time unit) excited by
short light pulses usually shows an exponential behavior according to
I ðtÞ ¼ Aekt ¼ Aet∕t , (5.12)
where k is the total rate of deactivation of the excited electronic state and t the
fluorescence lifetime. If several molecular species contribute to the fluores-
cence decay, their intensities sum up according to
I ðtÞ ¼ Si Ai ekit ¼ Si Ai et∕ti (5.13)
with Ai being the amplitude and ti the fluorescence lifetime of an individual

component.
The relative fluorescence intensity of each component, I, out of all
components, j, is given by
X X
I i ¼ ∫Ai et∕ti dt∕ ∫Aj et∕tj dt ¼ Ai ti ∕ A j tj , (5.14)
j j
if the integral is calculated from t ¼ 0 to infinity.

In addition, an average fluorescence lifetime can be calculated according to
X X
, t .¼ A j tj ∕ Aj : (5.15)
j
Figure 5.5 shows the fitted fluorescence decay curve of BKEz-7

endothelial cells from the calf aorta74 incubated with the lysosomal marker
acridine orange (AO; 5 mM, 30 min.) together with the response curve of the
apparatus. The decay curve shows a biexponential behavior with t1 ¼ 2.1 ns
(I1 ¼ 9%) and t2 ¼ 15.8 ns (I2 ¼ 91%) corresponding to a monomeric and a
dimeric (or oligomeric) AO species, respectively.75 AO monomers are located
in the cytoplasm and in the cell nucleus, whereas oligoimers are accumulated
within lysosomes where AO is most concentrated.76
As can be deduced from Fig. 5.5, the measured fluorescence intensity I(t)
is given by convolution of the “real” fluorescence decay curve F(t) and the
response function R(t) of the apparatus corresponding to
I ðtÞ ¼ ∫F ðtÞ Rðt t0 Þdt0 (5.16)
with the integration limits t0 ¼ 0 and t0 ¼ t. F(t) as well as the fluorescence

decay times ti and the amplitudes Ai are commonly determined from a least
squares analysis. A theoretical (e.g., biexponential or triexponential) curve
F(t) is calculated using assumed values Ai and ti. These values are varied in an
256 Chapter 5
600
500
response
400
counts
300
200
BKEz-7 + AO
100
0
0 5 10 15 20 25 30
time / ns
Figure 5.5 Fluorescence decay curve of BKEz-7 endothelial cells incubated with acridine
orange (5 mM, 30 min.) and response curve of the apparatus measured by the single photon
counting method (lex ¼ 390 nm, repetition rate ¼ 1 MHz, fluorescence detected at l $
590 nm).
iterative process until the best fit is obtained. The quality of a fit x2 is
calculated from
X
x2 ¼ vi ½I ðtÞ I C ðtÞ2 (5.17)
i
where I(t) is the measured and IC(t) the calculated fluorescence intensity. vi ¼
1/F(t) is a statistical weighting factor to account for the expected error in each
value of I(t). R(t t0) corresponds to the measured response curve (scattered
light of the picosecond laser diode). A minimum of x2 is obtained for the best
fit. x2 values larger than 2 indicate a poor fit, x2 values less than 1.2 indicate a
good fit.9 A detailed overview on advanced time-correlated single photon
counting techniques, including those with high spatial resolution, has recently
been given.77
5.4.2 Phase fluorometry

Instead of pulsed excitation and time-resolved detection, the sample can be
excited with light whose intensity is modulated sinusoidally (Fig. 5.6). The
intensity of emission is also a sine function, but due to the finite lifetime of
the excited electronic state, the modulated emission is delayed in phase by
an angle w relative to excitation. In addition, the emission is demodulated
relative to the excitation, i.e., the ratio of the variable and the continuous
intensity fractions of emission B/A is smaller than the corresponding ratio
b/a of excitation. When defining a demodulation factor m ¼ Ba/bA, both
the phase angle w and demodulation factor m can be used to calculate the
b
relative intensity
B
ϕ
a A
0
0 50
time [ns]
Figure 5.6 Principle of a phase resolved fluorescence measurement (A,B ¼ continuous
and variable intensities of emission, a,b ¼ continuous and variable intensities of
excitation).
“phase lifetime” tp and the “modulation lifetime” tm according to the

equation
tan w ¼ v tp (5.18)
m ¼ ½1 þ v2 t2m 1∕2 , (5.19)
where v corresponds to 2p the modulation frequency (in Hz).9 Only for
monoexponential decays tp ¼ tm ¼ t, i.e., the lifetimes obtained from phase
shift and demodulation correspond to the real fluorescence lifetime. If the
fluorescence decay is more complex, tp and tm are “apparent” lifetimes with
tp , tm for multi-exponential decays.
Fitting of multi-exponential decays becomes possible if the modulation
frequency is varied, and if the phase angle w or the demodulation factor m is
measured as a function of this frequency, as described in Ref. 78. Frequencies up
to the Gigahertz range are required to resolve decay times in the subnanosecond
range. Since typical electro-optic modulators (e.g., Pockels cells) are usually
limited to some hundred MHz, specific techniques, e.g., frequency multiplication
and cross correlation have to be used, as further described in other work.78,79
Applications of phase and frequency modulation techniques for imaging fluoro-
phores in various samples, cells, and tissues have been described elsewhere.80–82
Frequency-domain fluorometry is also the basis for the so-called “phasor
approach,”83 which permits one to distinguish clusters of different fluorescence
lifetimes or gives some evidence of non-radiative energy transfer (see Section 5.6).
258 Chapter 5
In addition to modulated lamps or lasers, LEDs can be used as excitation

sources for frequency-domain lifetime measurements. For example, a
frequency-domain lifetime fluorometer based on a microscope and a
modulated LED (370/460 nm) was described that operates in the frequency
range 120 Hz 250 MHz.84 LEDs cover the whole visible range and offer
excellent stability, light-noise levels, power efficiency, and economy. Herman
et al. demonstrate that a simple LED excitation source can, for many
applications, successfully replace complex and expensive laser systems that
have been used for frequency-domain lifetime measurements of living cells.84
5.4.3 Time-gated fluorescence spectroscopy

The use of time-resolving methods for fluorescence imaging has been a great
challenge for several scientists in recent years. Comparably little attention,
however, has been attributed to time-gated fluorescence spectroscopy. The
principle of this method is shown in Fig. 5.7, where the decay curve of
autofluorescence of cultivated endothelial cells70 is depicted. This curve can
be fairly fitted by 3 exponentially decaying components with fluorescence
lifetimes of about 0.5 ns, 2.5 ns, and (in some cases) 5.5 ns, which so far have
been attributed to the folded as well as to the extended conformation of
the coenzyme NADH (0.5 ns and 2.5 ns), as well as to flavin molecules
(e.g., flavin mononucleotide, FMN; 5.5 ns).21 The folded conformation is
usually related to “free” NADH, whereas the extended conformation occurs
when NADH molecules are bound to proteins. In Fig. 5.7, time gates are
indicated that can be used for the detection of fluorescence. In an “early”
time gate (i.e., immediately after the exciting laser pulse; gate A) an
emission band at 460 to 470 nm, corresponding to free NADH, was
25
20
A B
I [counts]
15
10
0
0 1 2 3 4 5
t [ns]
Figure 5.7 Decay kinetics of autofluorescence of BKEz-7 endothelial cells from calf aorta
after excitation by short pulses (70 ps) of a laser diode. Excitation wavelength: lex ¼ 375 nm;
fluorescence measured at lem $ 435 nm (reproduced from Ref. 80 with modifications).
predominant, whereas in a “later” time gate (gate B), the emission maxima
around 435 nm (bound NADH) and 515 nm (flavins) became more obvious.
In contrast, CW spectra of autofluorescence were broad and exhibited only
a little substructure.91 Using time-resolved fluorescence detection, free and
bound NADH could be studied separately, e.g., as a function of oxygen
content36,37 or upon inhibition of either the respiratory chain or the
glycolytic pathway.21,85
Further applications of time-gated fluorescence spectroscopy include
studies of the dynamics of various dyes within vesicles,86 as well as the
detection of tumor-localizing porphyrins within tissues of strong autofluor-
escence.87,88 Due to their fluorescence lifetime around 15 ns, these porphyrins
could be easily distinguished from short-lived autofluorescence using a
detection window that was delayed by about 15 ns with respect to the exciting
laser pulse. Monomeric and aggregated porphyrin molecules as well as ionic
species located at different cellular sites were also distinguished by time-gated
fluorescence spectroscopy.89,90 Further applications of this technique include
the detection of membrane markers whose fluorescence is sensitive to a phase
transition of cellular lipids as well as measurements of porphyrin metabolites
in plants, whose fluorescence is strongly superposed by the emission of
chlorophyll.91 Due to the rapid fluorescence decay of chlorophyll (usually
below 1 ns), porphyrins like uroporphyrin or protoporphyrin become
detectable at delay times around 15 ns between excitation and fluorescence
measurement.
5.4.4 Time-resolved fluorescence imaging

With the availability of fast time-gating devices, several time-resolving
fluorescence imaging methods have been applied almost simultaneously since
the early 1990’s. Lakowicz et al.92 reported on phase modulation techniques,
which were used to distinguish free and protein-bound NADH on the basis of
their fluorescence lifetimes of about 0.5 ns and 1–2 ns, respectively. The
intensity of the light source and the gain of an image intensifier (located in
front of a CCD camera) were modulated simultaneously, and the phase shift
between the two modulations was kept variable. The emission of individual
fluorophores was registered most efficiently for appropriate phase shifts. In a
similar way, other authors measured the distribution of various fluorescent
dyes in cultivated cells using the phase modulation technique.93–95
First experiments on one-dimensional and two-dimensional time-gated
laser scanning microscopy were performed by Bugiel at al.96 and Buurman
et al.,97 respectively. In particular, argon ion lasers were used in combination
with electro-optic or acousto-optic modulators and single photon counting
devices. Two different time gates were set simultaneously and synchronized
with the modulator using specific electronic circuits.98 If within time gates A
and B of identical width (see e.g., Fig. 5.7), which were shifted by a delay Dt,
260 Chapter 5
the fluorescence intensities IA and IB were measured, a so-called “effective”

fluorescence lifetime
teff ¼ Dt∕ lnðI A ∕I B Þ (5.20)
could be calculated. In the case of monoexponential decays, teff was identical

to the real fluorescence lifetime. Recently, measurement of fluorescence decay
curves has become possible for each pixel of a laser scanning microscope. This
permits imaging of individual fluorescence lifetimes even of multi-exponential
decays.77 Prolonged measuring times and exposure to high (possibly
phototoxic) light doses, however, should be considered for this kind of
experiment.
As an alternative technique to laser scanning microscopy, time-gated
video microscopy with highly intensifying camera systems has been used
for imaging of tumor-localizing dyes (porphyrins and their derivatives,
phthalocyanines).87,91,99,100 Since fluorescence lifetimes of porphyrin mono-
mers (i.e., those species that are most efficient in photodynamic therapy)
are around 15 ns, their fluorescence can be easily distinguished from the
fluorescence of porphyrin aggregates as well as from autofluorescence of the
tissue with lifetimes below 6 ns. Accumulation of porphyrins could, therefore,
be studied in various kinds of tumors87,99 as well as in the skin and within
cultivated cells.91
Applications of time-gated and frequency-modulated imaging techniques
include the detection of intracellular calcium (using dyes whose fluorescence
lifetimes change upon binding of calcium101–103), oxygen104 or pH values.105
Also during the uptake of the cytostatic drug doxorubicin its fluorescence
lifetime was found to decrease, possibly indicating a change of its cellular
binding site.106 In addition, time-resolving imaging techniques proved to be
useful to study a class of enzymes (caspases) that play an important role in
the initiation and execution of apoptosis. Targeting techniques with mutants
of green fluorescent protein (GFP) and the method of non-fluorescent energy
transfer (see below) were used.107 Meanwhile clinical applications of
fluorescence lifetime imaging have been reported for the skin,108 the ocular
fundus,109 and various organs,110 and specific fundus cameras as well as
endoscopes have been combined with fluorescence lifetime measurements.
A set-up of time-resolved fluorescence imaging (fluorescence lifetime
imaging, FLIM) is shown in Fig. 5.8. Picosecond laser pulses (e.g., from a
laser diode with a repetition rate up to 40 MHz; PicoQuant GmbH, Berlin,
Germany) are used for illumination of the sample, which often is located
within a fluorescence microscope. Laser excitation has to be synchronized
with the image intensifier, which is activated by applying short electric pulses
between about 70 V and 500 V either to the photocathode or to a
multichannel plate of the image intensifier. Usually the time delay between
the laser pulse and the active time of the image intensifier can be varied in
Cooled
CCD
Synchronisation Delay Pulse Image

Generator Intensifier
Fluorescence
Microscope
Sample
Scanner
Picosecond Laser Optical Coupler

Diode + Shutter
Monomode Fiber
Figure 5.8 Experimental set-up for picosecond fluorescence lifetime imaging.
picosecond to nanosecond steps. Time-gated intensified images are usually

detected by a CCD camera. Typical data of a time-gated image-intensifying
camera (Picostar HR12 with thermoelectrically cooled CCD camera;
LaVision GmbH, Göttingen) are: minimum time gate 100 ps; time increments
between 25 ps and several ns; repetition rate up to 110 MHz, which is suitable
for high repetition laser diodes or modelocked lasers. A negative gating pulse
up to –200 V is applied to the photocathode, whereas the sensitivity is adjusted
by a voltage between 0 and þ900 V on the multichannel plate. Due to a
constant voltage around 6 kV on the phosphor screen, about 200 photons are
generated by each incident electron.
Figure 5.9 Fluorescence intensity (a) and fluorescence lifetime (b); scale from black to
white: 0-8 ns) of BKEz-7 endothelial cells incubated with 50 mM Rhodamine 123 (excitation
wavelength: 488 nm: detection range: lem $ 515 nm; image size: 220 mm 160 mm).
Reproduced from Ref. 111 with modifications.
262 Chapter 5
An example for fluorescence lifetime imaging is given in Fig. 5.9 for

cultivated BKEz-7 endothelial cells from the calf aorta incubated with the
mitochondrial marker rhodamine 123 at a concentration of 50 mM. While the
fluorescence intensity image shows an accumulation of this marker within
the mitochondria, the fluorescence lifetime image shows some shortening in
the same organelles due to an increasing rate of non-radiative transitions with
concomitant fluorescence quenching according to Eq. (5.4). This quenching
process possibly results from aggregation of the dye at high concentration
within the mitochondria.
5.5 Total Internal Reflection Fluorescence Spectroscopy

and Microscopy (TIRFS/TIRFM)
In 1981, Axelrod applied for the first time total internal reflection (TIR)
illumination for the selective excitation of fluorophores located in or in close
proximity to the basolateral part of the plasma membrane of cultured cells.112
TIRFM techniques utilize an evanescent electromagnetic field for the
excitation of fluorophores, which arises upon total reflection of the
excitation light at the cell-substrate interface. The evanescent field decays
exponentially with perpendicular distance z; its penetration depth depends
on the wavelength and the incident angle of excitation light as well as on
the refractive indices of the optical media (Fig. 5.10). Typically,
penetration depth can be varied between 100 nm and approximately
350 nm using appropriate technical solutions for variable angle TIRFS/
TIRFM (VA-TIRFS/TIRFM). Therefore, these techniques are especially
well suited for the selective examination of structural aspects of the plasma
n 2 < n1
Θ n1
I0
Figure 5.10 Principle of total internal reflection.

membrane and cell/substrate contacts as well as physiological processes

associated with the plasma membrane. In the future, innovative pulsed
laser systems for TIR illumination in combination with appropriate
detection devices should allow new insights into the morphology and
physiology of living cells with high axial resolution.
5.5.1 Theory of TIRFS/TIRFM

When a beam of light propagating through a medium of refractive index n1
meets an interface with a second medium of refractive index n2 , n1, total
internal reflection occurs at all incidence angles Θ that are greater than a
critical angle ΘC ¼ arcsin n2/n1. While being totally reflected, the incident
beam establishes an electromagnetic field that penetrates a small distance into
the second medium and decays exponentially with perpendicular distance z
from the interface according to I(z) ¼ I0 e–z/d(Θ), where the penetration depth
d(Θ) for light of wavelength l is given by
dðΘÞ ¼ ðl∕4pÞðn21 sin2 Θ n2 2 Þ1∕2 : (5.21)
Intensity of the evanescent electromagnetic field I0 at z ¼ 0 is given by the

intensity of incident light Ie multiplied with the transmission factor T(Θ). If
the electric field vector of the incident light beam is polarized perpendicular to
the plane of incidence (see below), this transmission factor is given by
TðΘÞ ¼ 4 cos2 Θ∕½1 ðn2 ∕n1 Þ2 : (5.22)

For calculating the fluorescence intensity in TIRFS/TIRFM measure-
ments, light absorption dI/dz within thin layers dz has to be considered. With ε
being the molar extinction coefficient and c(z) the concentration of absorbing
molecules, one can calculate
dI ∕dz ¼ ε ln 10 cðzÞ I ðzÞ ¼ ε ln 10 cðzÞ I e ðn2 ∕n1 Þ TðΘÞez∕dðΘÞ : (5.23)

Fluorescence is obtained from Eq. (5.23) by multiplication with the
fluorescence quantum yield h and the solid angle V, as well as by integration
over the layers where fluorophores are located. If emission is assumed to be
isotropic, the result for fluorescence intensity is
I F ðΘÞ ¼ ε ln 10 hðV∕4pÞ I e ðn2 ∕n1 Þ TðΘÞ∫cðzÞez∕dðΘÞ dz, (5.24)

or
I F ðΘÞ ¼ A TðΘÞ∫cðzÞ ez∕dðΘÞ dz, (5.25)

if all factors independent from the angle Θ and the coordinate z are included
within the experimental constant A.
264 Chapter 5
According to Eq. (5.25), the fluorescence intensity IF (Θ) can be calculated

for
• a continuum, where fluorophores are distributed homogeneously (with a
constant concentration c) above the interface. In this case, the integral
has to be calculated from z ¼ 0 to z ¼ `, thus giving IF ¼ A c T(Θ) d(Θ);
• a homogeneous distribution of fluorophores (c ¼ constant) for z $ a,
e.g., within the cytoplasm with cells having a distance a from the
interface. In this case, the integral has to be calculated from z ¼ a to z ¼
`, thus giving IF ¼ A c T(Θ) d(Θ) e–a/d(Θ);
• a distribution of fluorophores within a thin layer of thickness b at a
distance a from the interface, e.g., within cell membranes. In this case,
the integral has to be calculated from z ¼ a b/2 to z ¼ a þ b/2, thus
giving IF ¼ A c T(Θ) b e–a/d(Θ), if the concentration c is considered to be
constant within the layer, and if b is small compared to a.
Figure 5.11(a) displays IF (Θ) for these three conditions with nS ¼ 1.52
(substrate ¼ microscope object slide) and nC ¼ 1.37 (cytoplasm). For cyto-
plasmic (“intracellular”) fluorescence a distance a ¼ 100 nm from the interface
has been assumed and for membrane fluorescence, the parameters a ¼ 100 nm
and b ¼ 8 nm were used. In Fig. 5.11(b,c) an experimental result for the
cytoplasm marker calcein in a single U373-MG glioblatoma cell upon
variable-angle TIRFM (individual images and cell-substrate topology) is
depicted. It should be mentioned that for cells cultivated on glass slides, the
two-phase model with refractive indices n1 and n2 is an approximation, since
in reality four phases interfere: the substrate, the extracellular medium (“water
gap;” refractive index 1.33), the membrane (refractive index 1.45) and the
cytoplasm. However, it has been shown in the literature108 that the refractive
index of the comparatively thin plasma membrane can be neglected for
calculation of IF(Θ) with an error no greater than 2.5%. In the remaining
three-phase model, effective values of the critical angle ΘC, transmission
factor T(Θ) and penetration depth d(Θ) can be used. In the case that the layer
of the extracellular medium is small compared to the wavelength of incident
light, the effective values ΘC and T(Θ) can be approximated by the values
obtained from the two-phase model using the refractive indices of the
substrate and the cytoplasm. In contrast, the effective value of the penetration
depth d(Θ) can be approximated by the two-phase model with the refractive
indices of the substrate and the extracellular medium. Detailed theoretical
background on TIRFS/TIRFM and VA-TIRFS/TIRFM is given in.113,114
5.5.2 Technical set-up

As depicted in Fig. 5.12, two different technical solutions for TIR illumination
can be distinguished. In the most commonly used set-up, the cell substrate is
optically coupled, preferably with immersion oil or glycerol, to a glass or quartz
1000
(a)
Ic
Ie
800 Im
relative fluorescence
600
400
200
0
66 68 70 72 74 76 78 80
angle of incidence [°]
(b)
(c)
Figure 5.11 (a) Fluorescence intensity IF (Θ) for different distributions of fluorophores:
Ic: continuum with homogeneous distribution; Ie: homogeneous distribution in the cytoplasm;
Im: homogeneous distribution in the cell membrane (nS ¼ 1.52 for the substrate and nC ¼
1.37 for the cytoplasm; assumed cell-substrate distance: 100 nm); (b) cytoplasm marker
calcein in a single U373-MG glioblatoma cell upon variable-angle TIRFM; (c) cell-substrate
topology (color scale: 0–200 nm for cell-substrate distances)6HHFRORUSODWHV
prism. The prism usually has the shape of a cube, or a hemisphere or

hemicylinder, respectively. Variation of the angle of incidence and, concomi-
tantly, fluorescence measurements with a z-resolution of a few nanometers
requires a goniometric set-up 115 or a combination of acousto-optic modulators
and scanners.114
266 Chapter 5
For a prismless TIRF configuration, an objective with a high numerical

aperture ($1.4) is used for TIR illumination. Only incident rays traversing a
peripheral annulus116 or focused to an excentric spot117 are allowed to
propagate through the objective lens, while central rays are blocked. The
peripheral rays are incident upon the cell substrate of the sample at
supercritical angles to the cell/substrate interface. Meanwhile this so-called
“objective type” method dominates the TIRF literature and is commercially
available by several companies. However, variation of penetration depth by
different angles of incidence is difficult to perform.
5.5.3 Combination of TIRFS/TIRFM with innovative fluorescence

microscopic techniques
So far, several combinations of TIR illumination and specialized excitation or
detection techniques, e.g., pulsed laser systems and time-gated detection systems
have been realized. For a detailed analysis of the temperature-dependent
fluorescence of the membrane marker laurdan, excitation was performed with
nanosecond pulses of a Nd:YAG laser and time-gated fluorescence detection.118
Using femtosecond pulses of a Ti:Sapphire laser (l ¼ 770 nm), two-photon
excitation of the UV-absorbing calcium ratio marker Indo-1 was achieved under
TIR illumination.119 TIR illumination could also be used for the selective
photobleaching of membrane associated fluorophores in FRAP experiments
(fluorescence recovery after photobleaching).120 By the use of a UV laser as the
TIR illumination light source, selective release of caged compounds, e.g., the
second messenger Ca2 þ , was achieved locally near the plasma membrane.121
1a 1b
Figure 5.12 Technical realization of TIR illumination. 1: optical coupling of the cell
substrate using cube-shaped (1a) or hemispheric prism (1b); 2: prismless configuration
using a high numerical aperture objective lens.
It should be emphasized that meanwhile, TIRFM has become one of the most
powerful techniques in single molecule detection.122,123
5.5.4 Application of TIRFS/TIRFM in cell biology

So far, TIRFS/TIRFM has been used for a multitude of cell biological
examinations. In several papers, detailed experiments for the characterization
of cell-substrate contacts of different cellular models, e.g., amoeba, endothelial
cells, or neuronal cells, are described.115,124–126 Using VA-TIRFM, cell-
substrate distances were calculated for glioblastoma cells prior and subsequent
to activation of tumor suppressor genes. It revealed that the original tumor cells
had almost constant cell-substrate distances around 100 nm (“plane surface”),
whereas these distances varied between about 100 nm and more than 200 nm
for the less malignant cells (“folded surface”).127 Membrane-associated
receptors, especially acetylcholine and epidermal growth factor receptor
(EGF-R), were mainly investigated by Axelrod and co-workers with respect
to spatial distribution and internalization.101,128 In combination with FRAP
techniques, binding kinetics of EGF towards EGF-R as well as diffusion rates
of membrane-associated proteins were measured.120 Recent investigations have
focused on membrane-proximal transport, i.e., endocytosis and exocytosis
vesicles as well as release of the vesicle content.129–131 Using TIRFS/TIRFM,
membrane-proximal calcium concentrations could be measured in neutrophils,
ventricular myocytes, and oocytes.132–134 Force transmission from apical to
basal cell membrane was measured by combining atomic force and TIRFM.135
For the selective examination of biophysical properties of single molecules,
TIR illumination was combined with ultra-sensitive detection methods. Due to
the drastic reduction of background fluorescence, motion of single, fluorophore
labelled motile proteins like myosin and kinesin could be monitored.136,137
One current application of TIRFS/TIRFM comprises the selective
examination of plasma membrane associated photosensitizers used in Photody-
namic Therapy (PDT). Photosensitizers usually are porphyrin-derived com-
pounds that accumulate selectively in tumor tissues. During irradiation,
cytotoxic processes, mainly involving singlet oxygen, are initiated resulting in
tumor cell death and shut-down of tumor-supplying vessels.138 Specific cellular
damage largely depends on the intracellular location of the photosensitizers. Due
to the limited penetration depth of the evanescent electromagnetic field,
photobleaching of membrane-associated protoporphyrin IX (PPIX), a naturally
occurring photosensitizer, was observed and correlated with data on phototoxic-
ity.139 After short incubation times, a huge portion of rapidly bleaching PPIX,
assigned to the photodynamic active monomeric fraction, was detected using
TIR illumination. In contrast, 24 h after incubation, an increased amount of a
slowly bleaching portion, assigned to an aggregated, photodynamically less
active PPIX fraction was found, which correlated with a reduced photodynamic
efficacy. The photobleaching behavior of PPIX induced by 5-aminolevulinic acid
268 Chapter 5
(5-ALA) was slightly different,140 but also in this case, a considerable amount of
photosensitizer was localized in the plasma membrane.140,141 Furthermore,
association of photosensitizers of different hydrophilicity, but comparable
photophysical properties, with the plasma membrane was investigated. Results
obtained using the TIRFS technique provided for the first time direct evidence
for a plasma membrane-associated fraction of the hydrophilic compounds meso-
tetraphenylporphyrine trisulfonate (TPPS3) and meso-tetraphenylporphyrine
tetrasulfonate (TPPS4).142 In contrast, the more lipophilic drugs meso-
tetraphenylporphyrine monosulfonate (TPPS1) and meso-tetraphenylporphyrine
disulfonate (TPPS2) could only be detected using epi-illumination. Further
techniques and applications of TIRFM are summarized in Ref. 143.
5.6 Energy Transfer Spectroscopy

5.6.1 Basic mechanisms
One of the most interesting mechanisms used in luminescence spectroscopy
and cellular analytics is energy transfer between molecules in their lowest
excited electronic state S1 or T1. Three basic mechanisms can be distinguished:
• In the first case, a photon emitted by one molecule is reabsorbed by
another one. In transparent samples this may occur at various distances,
if the difference between the energy levels of the excited state and the
ground state is the same for both molecules. This re-absorption is,
therefore, of rather little analytical interest, but can cause erroneous
measurements. For example, in photosynthetic organisms the chlorophyll
fluorescence arising from an optical transition between vibronic ground
states n0 (0-0 transition; l ¼ 685 nm) is strongly reabsorbed, whereas the
fluorescence radiation arising from a transition between the ground state
of S1 and a vibronic level of S0 (l ¼ 735 nm) is not reabsorbed. Therefore,
the fluorescence ratio measured at 685 nm and 735 nm is affected by
reabsorption and depends on chlorophyll concentration as well as on the
geometry of the sample.144 Only under very well defined experimental
conditions can this ratio be used as a measure of chlorophyll concentra-
tion or of the physiological state of an organism.145
• Energy transfer by direct interaction of optical transition dipoles of a
donor and an acceptor molecule is a mechanism of significant analytical
importance. This dipole-dipole interaction is proportional to r–6 (with r
corresponding to the intermolecular distance) and needs an overlap of the
emission spectrum of the donor and the absorption spectrum of the
acceptor according to
k ET r6 ∫εA ðnÞI D ðnÞn4 dn: (5.26)
with kET being the rate of energy transfer, εA(n) the molar extinction
coefficient of the acceptor, ID(n) the quantum flux of emission of the
donor, and n the frequency of radiation. The principle of this so-called
h ν ex h ν em
k ET
NADH R123
d < 10nm
η NADH ≤ 0.1 ηR123 = 0.9

absorption [a.u.]
intensity [a.u.]
300 350 400 450 500 550 600
wavelength [nm]
Figure 5.13 Principle of non-radiative energy transfer from the coenzyme NADH to the
mitochondrial marker R123 (top); absorption (full lines) and emission (broken lines) spectra
of NADH and R123; spectra of R123 are red-shifted as compared with NADH, with a
pronounced overlap of NADH emission and R123 absorption (bottom).
Förster mechanism146 is shown in Fig. 5.13 (upper part) for the example
of NADH ! rhodamine 123 (R123) as a donor-acceptor pair, where the
donor is a coenzyme of the mitochondrial respiratory chain and the
acceptor a well known mitochondrial marker.147 Light is absorbed by
the donor and emitted by the acceptor, whereas intermolecular energy
transfer is non-radiative.
The absorption and emission spectra of NADH and R123 are depicted in
the lower part of Fig. 5.13. The absorption spectra of NADH (left) and R123
(right) correspond to the full lines, whereas the emission spectra are represented
by the broken lines. A broad overlap of the emission spectrum of NADH
(maximum around 470 nm) and the absorption spectrum of R123 (maximum
around 510 nm) indicates that the condition of resonance is fairly fulfilled.
According to Eq. (5.26), fluorescence resonance energy transfer (FRET) is
limited to short intermolecular distances of a few nanometers. As a
quantitative measure, the so-called Förster radius r0 is used, which is defined
such that the rate of energy transfer kET and the sum of all other rates of
deactivation of the first excited molecular state S1 are equal for r ¼ r0. This
can be described by the equation
k ET ¼ t0 1 ðr∕r0 Þ6 (5.27)
with t0 being the lifetime of the excited electronic state in the absence of
energy transfer. When using non-radiative energy transfer it is an advantage if
270 Chapter 5
the fluorescence quantum yield h of the acceptor is rather high (e.g., h ¼ 0.9
for R123). In principle, kET also depends on the orientation of the transition
dipoles of the donor and the acceptor molecule. In solution as well as in most
cellular systems, however, these dipoles can be assumed to be randomly
oriented. Usually, energy transfer by dipole-dipole interaction is limited to
singlet states, i.e., to a transition S1 ! S0 for the donor and S0 ! S1 for the
acceptor molecule. In nature, non-radiative singlet energy transfer plays an
important role within the antenna complexes of photosynthetic organisms.4–6
For excited triplet states, the rate of this transition is becoming very low due
to a change of electron spins.
Energy transfer rates can be determined from stationary as well as from
time-resolved fluorescence measurements. In the first case, one measures the
fluorescence quantum flux of the donor
I D ¼ I 0 ln 10 d εD cD hD V∕4p, (5.28)
which corresponds to Eq. (5.3), and of the acceptor
I A ¼ I 0 ln 10 dðεD cD hET þ εA cA ÞhA V∕4p, (5.29)
which considers that the acceptor may be excited via energy transfer from
the donor with the quantum yield hET ¼ kET/k as well as by direct
absorption of light.21 The acceptor itself fluoresces with the quantum yield
hA(εD, εA ¼ extinction coefficients, cD, cA ¼ concentrations of donor and
acceptor molecules, d ¼ diameter of the sample). Assuming an almost linear
relationship between kET and the acceptor concentration cA,148 the
quantum flux of acceptor fluorescence is proportional to the acceptor
concentration. Therefore, from the linear relationship between IA and cA,
the concentration of donor molecules can be deduced according to
Eq. (5.29). The energy transfer rate kET ¼ k hET can be obtained by
division of the Eqs. (5.29) and (5.28). If direct absorption of light by
acceptor molecules becomes negligible, and if the total rate k of
deactivation of the excited state is replaced by the lifetime t ¼ 1/k of this
excited state, one can easily calculate
I A ∕I D ¼ tðhA ∕hD Þk ET : (5.30)
If hA and hD are known from the literature, and if t can be approximated

by the literature value t0 of the donor, the energy transfer rate kET can be
calculated from the ratio of acceptor and donor fluorescence at well defined
concentrations of acceptor molecules.
Another possibility of calculating kET is time resolved fluorescence
spectroscopy. According to Eq. (5.4) and to the relation kF þ kIC þ kISC ¼
1/t0 (in the absence of energy transfer), the energy transfer rate can be
calculated according to
k ET ¼ 1∕t 1∕t0 (5.31)
with t and t0 being the fluorescence lifetimes of the donor in the presence and
absence of energy transfer, respectively. Since 1/t and kET depend (almost
linearly) on acceptor concentration, various concentrations cA can be used for
optimal fitting of kET.
• In excited triplet states where dipole-dipole interaction becomes
negligible, non-radiative energy transfer may arise from an electron
exchange mechanism. This process requires a considerable overlap of the
electron orbitals of the excited donor molecule and the acceptor molecule
in the ground state with spectral characteristics being rather irrelevant.
The distance over which this exchange can act is small (typically not more
than 1 nm), so the process requires direct contact between donor and
acceptor molecules, e.g., during diffusion of one of these molecules.
Energy transfer rates kET are often small compared with dipole-dipole
interactions, but since the lifetime t of excited triplet states usually is
rather long (microseconds-milliseconds), the quantum yield of energy
transfer hET ¼ kET/k ¼ kET t may still be large. Energy transfer from the
excited triplet state of porphyrins, chlorins, or related molecules to
oxygen molecules is a well known process that occurs in native cells as
well as in cells incubated with photosensitizers. Oxygen is thereby excited
from the triplet ground state to the singlet state S1. Cytotoxic reactions
initiated by singlet oxygen are well described in the literature and used
e.g., for photodynamic therapy of tumors (for an overview on intra-
cellular reactions see Ref. 149). Quantum yields of singlet oxygen genera-
tion (including S1 ! T1 intersystem crossing of the donor and energy
transfer to oxygen) have been summarized for various photosensitizers in
solution.150 Typical values are 0.57 for protoporphyrin dimethylester,
0.61 for meso-tetrakis(m-hydroxyphenyl)chlorin (m-THPC), 0.36 for
aluminium phthalocyanine or 0.42 for porphycene, i.e., 30 to 60% of
excited molecules of a photosensitizer may generate cytotoxic singlet
oxygen. Damage of plants may also be related to cytotoxic reactions
originating from chlorophyll molecules within the photosystems and
involving singlet oxygen. Some reduced protection against photosensiti-
zation may, therefore, account for light-induced yellowing of plants and
partly explain the phenomenon of “forest decline.” 6,151
5.6.2 FRET applications

Energy transfer processes have been used in cellular biology for many years.
Most of these processes are based on resonant dipole-dipole interaction
272 Chapter 5
according to the Förster mechanism. In the 1980s, microscopic studies on

energy transfer provided information on the architecture and intermolecular
distances in cell membranes.152,153 In addition, Förster resonance energy
transfer (FRET) proved to be useful for measuring the structure and assembly
of actin filaments154 and for the detection of binding sites of enzymes.155
Measurements of non-radiative energy transfer from the coenzyme NADH
to rhodamine 123 (R123; see above) proved to be useful for selective detection
of the mitochondrial fraction of NADH according to Eq. (5.29).21,156 Changes
of mitochondrial NADH upon inhibition of the respiratory chain appeared to
be more pronounced than changes of total cellular NADH. Therefore, energy
transfer spectroscopy may be more sensitive than autofluorescence for
measuring mitochondrial malfunction in various kinds of diseases, as reported
in Section 5.2.1. A problem of energy transfer measurements using R123 was
that intracellular accumulation of the potential sensitive dye R123 decreased
upon inhibition of the mitochondrial respiratory chain. Therefore, energy
transfer NADH ! R123 and intracellular amount of R123 had to be
determined by simultaneous excitation of NADH (at 355 nm) and R123 (at
470 nm). Fluorescence of R123 was measured in a time-gated mode in the first
case and in a frequency modulated mode in the second case. After verification
that the fluorescence signal in the second case was proportional to the
intracellular amount or concentration of R123, the ratio of both signals could
be regarded as a measure of mitochondrial NADH. This amount increased
upon inhibition of specific enzyme complexes of the respiratory chain using
cytotoxic as well as non-cytotoxic concentrations of an inhibitor.156
More recently it was reported that Förster energy transfer (FRET) can be
used to measure selectively mitochondrial depolarization that may precede
mitochondrial autophagy, apoptosis, and necrotic cell death.157 As already
mentioned, FRET is applied increasingly to mutants of green fluorescent
protein (GFP). A direct interaction between the two proteins Bcl-2 and Bax,
which may regulate apoptosis, was proven within individual mitochondria using
GFP-Bax and blue fluorescent protein (BFP)-Bcl-2 fusion proteins coexpressed
within the same cell.158 In addition, it was shown that specific amino acid
sequences located between BFP and GFP were cleaved by the enzyme caspase
upon induction of apoptosis: non-radiative energy transfer BFP ! GFP
disappeared, thus allowing activation of specific caspases to be monitored in
vitro and in vivo.159 Recently a caspase sensor based on FRET between
enhanced cyan fluorescent protein (ECFP) and yellow fluorescent protein
(EYFP) has been anchored to the plasma membrane of living cells and
monitored by TIRFM. Cleavage of the protein complex resulted in a disruption
of FRET that could be monitored with very high sensitivity in the evanescent
electromagnetic field.160 In addition, microscopic FRET techniques could be
transferred to a TIRF reader system for microtiter plates, which permitted high
content screening (HCS) of standardized samples.161 Since different GFP
mutants can be localized on various sites of a protein, conformational changes

of proteins (e.g., of calmodulin upon binding of calcium ions) can be measured
selectively.162 This makes it possible to visualize calcium uptake and distribution
in single cells. Further applications of FRET were dedicated to the detection
and visualization of GFP-tagged receptors in cells that were focally stimulated
by the epidermal growth factor (EGF). Following focal stimulation, energy
transfer from GFP to a fluorescent acceptor was measured and visualized, thus
proving a rapid and extensive propagation of receptor phosphorylation over the
plasma membrane, which finally resulted in full activation of all receptors.163 It
should be added that FRET measurements are used increasingly for studies of
pathogenesis of various diseases, e.g., tumors164 or Alzheimer’s disease.165
5.7 Wide-Field 3D Microscopy

5.7.1 Structured illumination
In modern microscopy, 3D methods are of increasing interest with individual
planes of cells, tissues or even whole organisms being selected by special
illumination or detection techniques prior to calculation of a 3D image. In
addition to confocal laser scanning microcopy (see section 5.8), structured
illumination microscopy is a well established technique. As first reported by
Neil et al.,166 a structure, e.g., an optical grid, is imaged in the plane of the
sample under at least 3 different phase angles. Then, by an appropriate
algorithm, the image from the focal plane is calculated, while out-of-focus
parts of the image are eliminated, as depicted in Fig. 5.14.
More recently, by coherent structured illumination, not only the axial
resolution, but also the lateral resolution has been increased up to a factor of 2
in comparison with Abbe’s criterion (see Section 5.3.2). Thus, resolutions of
Figure 5.14 Cultivated Chinese hamster ovary (CHO) cells after incubation with acridine
orange (5 mM; 30 min.) (a) original image, (b) original image with structured illumination,
(c) image from the focal plane, as calculated from 3 phase-shifted images with structured
illumination6HHFRORUSODWHV
274 Chapter 5
about 100 nm (lateral) and 200 nm (axial) have been obtained.167,168 These
values were further reduced by a combination of structured illumination and
single molecule localization microscopy.169
5.7.2 Light sheet fluorescence microscopy (LSFM)

A disadvantage of structured illumination microscopy (as well as laser scanning
microscopy) is that for measuring each focal plane the whole sample is exposed to
light, and upon recording of numerous planes the light dose sums up and may
damage sensitive organisms like living cells.170,171 This problem may be overcome
by light sheet fluorescence microscopy (LSFM) where a light sheet is created
perpendicular to the observation path either by a cylindrical lens or by scanning
of a laser beam (for a review, see Refs. 172 and 173). In this case, only the plane
under investigation is exposed to light, and when the sample or the light sheet is
shifted in the axial direction, successive measurements can be performed at low
light exposure. The principle of light sheet microscopy is depicted in Fig. 5.15(a)
Figure 5.15 Principle of light sheet fluorescence microscopy (a), and images of various layers
of 10 mm thickness of a CHO-pAcGFP1-Mem cell spheroid recorded at distances of 20 mm,
40 mm, and 60 mm from its edge (b–d); light incidence from top to bottom (excitation wavelength:
488 nm; fluorescence detected at l $ 515 nm). Reproduced from Ref. 174 with modifications
6HHFRORUSODWHV
with a cell spheroid of about 300 mm in diameter being located in a rectangular

microcapillary. Using Chinese hamster ovary cells permanently transfected with
a plasmid encoding for a membrane associated green fluorescent protein (CHO-
pAcGFP1-Mem), the plasma membrane is well visualized, as depicted in
Fig. 5.15(b–d) for individual layers at distances of 20 mm, 40 mm or 60 mm from
the edge of the spheroid. These light sheet images perform the basis of 3D
imaging microscopy.
5.8 Laser Scanning and Multiphoton Microscopy

5.8.1 Introduction
Confocal laser scanning microscopy (CLSM) was undoubtedly the most
important advance in optical microscopy during the last quarter of the 20th
century. It has become a powerful tool for cellular biologists and also for
molecular medical applications. The main advantage, besides a slightly better
lateral resolution, is the fact that out-of-focus blur is essentially absent in
confocal images. Therefore, it became possible to perform direct and non-
invasive serial optical sectioning of intact living specimens and to generate 3D
images of thick transparent or even semi-transparent objects like cells or tissues.
In contrast to conventional epi-illumination in light microscopy where the
entire field is illuminated simultaneously, and where fluorescence is excited
throughout the whole depth of the specimens, a single point in the focal plane
is illuminated in CLSM that is scanned in the x and y direction to produce a
complete image. A confocal imaging aperture is inserted in the optical system
preventing nearly all light emanating from regions above and below the focal
plane not contributing to the observed image. Increased contrast and signal-
to-noise ratio in the final image are due to a reduction of stray light not only
from the out-of-focus object planes, but also from the optical instrument itself.
A CLSM set-up is shown in Fig. 5.16.
The first idea for a confocal microscope was patented by Minsky in
1957.175 Then it took about twenty years until efficient light sources (lasers),
affordable computers, and digital image processing software led to the first
publications of 3D images of biological specimens. At present, most of the
microscope companies offer confocal laser scanning microscopes with com-
fortable image processing packages allowing not only the registration of 3D
fluorescence images, but also time-series of living specimens with temporal and
spatial resolutions superior to video microscopy.176 With a specially designed
Airyscan detector (Zeiss LSM 880 with Airyscan) the resolution still can be
improved by a factor of 1.7 and the confocal volume reduced by a factor of 5.
Spectrally resolved images, time-gated fluorescence images, lifetime images and
spectral lifetime images have become possible using single photon counting
devices. The speed of scanning the light source is limited due to the laser
intensity and fluorescence bleaching. To overcome this limitation, a technical
276 Chapter 5
Laser S
Ph
Figure 5.16 Set-up of a confocal laser scanning microscope with S corresponding to the
scanning device, T to a telecentric optical system, O to the microscope objective lens, Ph to
the confocal aperture, and D to the detector.
modification with an array of beams has been realized using either a line or an
array of pinholes (Nipkow Spinning Disk; Andor Technology). The principle
of image generation by an array of pinholes was first proposed in 1883 by the
German Physicist Paul Nipkow and was realized in 1884.
5.8.2 Performance of confocal laser scanning microscopes

Resolution: The resolution of a microscope can be described by the spatial
frequency cut-off of the imaging system’s optical transfer function (OTF),
which is the Fourier transform of the intensity point spread function (PSF). If
one considers a reflection grating as an object, the zero order of diffraction is
the reflected laser beam. Higher orders of diffraction interfere with the zero
order in such a way that with a small translation of the grating vertical to its
lines, constructive and destructive interference occurs. The signal is modulated
such that the modulation contrast decreases with decreasing numerical
aperture AN ¼ n sina of the microscope objective lens. The smallest distance
between the lines of a grating that can be resolved in conventional trans-
mission or reflection microscopy (with axial illumination) is Dx ¼ l/AN.
In fluorescence microscopy, however, no phase relation between illumination
and emitted light waves exists, and light intensities instead of amplitudes have
to be considered. Therefore, resolution in conventional fluorescence
microscopy is enhanced to Dx ¼ l/2AN. In the case of CLSM, the resolution
of a point object is given by the FWHM (full width at half-maximum) of the
laser intensity, which depends on the wavelength of light and the numerical
aperture of the objective lens. A good approximation177 for lateral resolution

is given by the equation:
FWHM ¼ 0.4l∕AN ¼ 0.4l∕n sin a (5.32)

where l is the wavelength of light, a the aperture angle of the microscope
objective lens, and n the refractive index of the immersion medium. In case the
pinhole is larger than the Airy unit, (AU = 1,22 l/AN), the resolution is the
same as for conventional microscopy
Dx ¼ 0.51 l∕AN : (5.33)

To determine the axial resolution of a microscope, one measures the
variation of the integral intensity of the image of a point object by defocusing
the object. In conventional microscopy, this integral intensity changes very
little when moving away from the focus, thus explaining the poor resolution in
the axial direction. When using a confocal aperture, however, the integral
light intensity resulting from a single point falls off sharply as one moves out
of focus. It is notable that the wavelength dependence of axial resolution is
more pronounced than that of lateral resolution. The axial resolution is given
in a good approximation by
FWHM ¼ 0.45 l∕nð1 cos aÞ: (5.34)

Considering a pinhole diameter larger than an Airy unit, then
Dz 1,67nl∕NA2 : (5.35)
In principle, these values are met within a few percent by high power
objective lenses. Increase of the confocal aperture in low light level
fluorescence microscopy (from a diameter of about 15 mm up to several
hundred mm), however, diminishes the degree of out-of-focus blur rejec-
tion.178 The dependence of the lateral and axial resolution on the numerical
aperture NA is plotted in Fig. 5.17.
In contrast to the conventional microscope where the complete field of
view and the whole object is illuminated, only a diffraction-limited point
source illuminates the object in CLSM. The intensity distribution within this
spot is described by the point spread function (PSF). The intensity distribution
gives the probability that a photon at a certain distance from the center
(normalized to intensity 1) is present to excite a fluorescent molecule. This
probability decreases rapidly with increasing distance from the center. The
probability distribution of a fluorescent molecule that is registered by the
detector is more widespread than the probability of illumination. As an
example, if the probability of illumination at a point P apart from the center
(focal point) is 0.1, and the probability for the detection of an emitted photon
is also 0.1, than the resulting probability of detecting a fluorescence photon is
278 Chapter 5
5
lateral resolution
axial resolution
4
FWHM / μm
0
0.4 0.6 0.8 1.0 1.2 1.4
numerical aperture
Figure 5.17 Plot of the lateral and axial resolution of a confocal laser scanning microscope
according to Eqs. (5.32) and (5.34) for l ¼ 488 nm.
0.01. Multiplication of the two probabilities of all points P leads to the PSF
for fluorescence detection in a confocal microscope, which is roughly the
square of the intensity distribution of illumination. This explains the “narrow”
PSF and the rejection of blur from out-of-focus regions.
Light sources: Laser light sources are well suited for fluorescence
excitation in CLSM. Argon/Krypton ion lasers, HeNe lasers of different
wavelengths, and also blue laser diodes are used and incorporated in the
system. Several wavelengths are necessary when biological objects are labelled
with dyes of different absorption wavelengths. In this case, multiple detection
channels are also needed for simultaneous detection of the fluorescence signals
within one specimen. Beam splitters and filters, or opto-acoustic modulators,
which are adapted to the emission spectrum of each specific dye are used for
spectral discrimination. For lifetime imaging, laser diodes modulated in the
MHz range or femtosecond lasers with multiphoton excitation are used.
Scanners: To create a 3D data set, the illuminating beam must scan all
three dimensions in space. One can distinguish between beam scanner, object
scanner, and objective lens scanner. The velocity to move an object is
limited; therefore, only slow movements (z scan) are performed with the
object. In a single beam CLSM high speed scanning is performed either by
galvo-scanners, piezo-crystals, opto-acoustic modulators, or by electrome-
chanical devices. Beam scanners are classified as single beam scanners,
multiple beam scanners, or slit scanners. The last two options, however, have
a reduced signal-to-noise ratio, because unspecific light from adjacent points
can contribute to the registered signal. The already mentioned rotating
Nipkow disk with numerous small holes in spiral configuration (Paul

Nipkow used this disk in 1884 for the first time in a camera for electrical
transmission of pictures, i.e., a first television system) is a well known
multiple beam scanner.
Photodetectors: For fluorescence imaging, where signal levels usually are
low, the sensitivity and the related noise of the detector are of practical
importance in addition to other specific properties like linearity and dynamic
range. For weak signals, photomultipliers (PMTs) are commonly used,
which may also be combined with single photon counting devices. At higher
signal levels, CCD arrays have different advantages. While the PMT, a
photometric device, has a quantum efficiency around 13%179 and a good
blue sensitivity when equipped with an S20 photocathode, the CCD detector
has a higher quantum efficiency in the range of 30 to 50% (up to 80%
with thinned back-illuminated CCD arrays) and is most sensitive in the
red part of the visible and in the NIR spectrum. Digital cameras exist in
different variations: the popular charge-coupled device (CCD), the electron
multiplying charge-coupled device (EMCCD), the complementary metal-
oxide semiconductor (CMOS) detector camera, and the image intensified
CCD camera (ICCD). Comparing the overall performance of PMTs and
CCD arrays, the combined effects of quantum efficiency h (defined as the
mean number of photoelectrons generated by an incident photon) and noise
Ne (measured in electrons per pixel) must be examined. Considering the
signal-to-noise ratio (SNR), when detecting Np photons, the equation for
SNR is given by:
SRN ¼ h N p ∕ðh N p þ N 2 e Þ1∕2 (5.36)
Thus, SNR is proportional to the square root of the quantum efficiency at

high illumination levels, whereas in the case of a small number of photons, it is
proportional to the ratio of the quantum efficiency and the sensor noise. With
respect to the number of significant grey levels generated by PMTs and CCDs
at very low illumination levels, it is notable that a cooled CCD array detector
is superior at Np . 25 photons/pixel compared to a PMT with S20 cathode.
Below 25 photons/pixel, the PMT has the better performance.
Image processing: Modern and powerful microcomputers (PCs) are part
of the CLSM systems controlling the microscope scanning, image acquisition,
and display synchronization. Up to five images can be acquired synchro-
nously, displayed or combined as a composite pseudocolor image. Image
processing software creates 3D images that can be rotated and evaluated with
respect to size and volume of internal structures within the samples. Time
series of pictures (galleries) in short time intervals show fluorescence kinetics
or motions of cells. Fast computers with sufficient storage capabilities favor
the use of CLSMs.
280 Chapter 5
5.8.3 Applications of CLSM

Confocal laser scanning microscopy is a standard tool for cell biologists,
evolutional biologists, pathologists, and medical or molecular biologists. The
whole range of dyes including green fluorescent protein (GFP) and its mutants
extended the field of applications to molecular medicine, immunology, and
molecular genetics. The methods include
• 3D topography of specimens measured in the reflection mode for
morphological studies of biological samples;
• high-resolution microscopy measured in the transmission mode;
• 3D-fluorescence detection of cellular structures and fluorescence-
bleaching kinetics;
• time-resolved fluorescence kinetics;
• studies of motions of cellular structures;
• time-gated imaging in order to select specific fluorescent molecules or
molecular interactions;
• fluorescence lifetime imaging (FLIM) and spectrally resolved fluores-
cence lifetime imaging;
• spectrally resolved imaging.
As an example, Fig 5.18 shows light-induced fluorescence changes of a
photosensitizer used for photodynamic therapy that is accumulated within
single cells. Using CLSM, fluorescence of volumes of only a few cubic
micrometers within the cell nucleus or the cytoplasm is registered during light
exposure. Parallel to these kinetics, changes of intracellular calcium are
registered from the same sample volumes using the fluorescent calcium
marker fluo-3. The curves show a concomitant increase of intracellular Ca2 þ
with fluorescence changes of the photosensitizer.
Further developments
• New developments permit the combination of morphometry with
spectral imaging providing multi-pixel information from a specimen180
Spectral information (SpectraCubeTM) obtained from a specimen is
stored as a third dimension of an xy-plot. By mathematical analysis a
spectral similarity map (SSM) is created that permits a selection of
specific molecules with well defined intracellular interactions.
• Excitation of fluorescent dyes or of autofluorescence in biological
samples by ultrashort laser pulses gives the possibility of fluorescence
lifetime imaging. This allows one to distinguish between molecules
showing identical fluorescence spectra but exhibiting different lifetimes.
The method also provides information on the microenvironment of
these molecules. The resolution of confocal microscopes can still be
enhanced by a 4Pi-configuration of the microscope optics, as reported,
e.g., in Ref. 181.
170
160 1- Ca2+ nucleus cell 1

2- AlPcS4 cell 1
1
150 3- Ca2+ cytopl. cell 1
4- AlPcS4 cell 2
140 5- Ca2+ cytopl. cell 2
6- Ca2+ nucleus cell 2
130 2
120
6
rel. fluor. int.
110
3 4
100
90
80
70
5
60
50
40
30
0 100 200 300 400
time (s)
Figure 5.18 Dynamics of intracellular Ca2 þ and the photosensitizer tetra-sulphonated

aluminium phthalocyanine (AlPcS4) in RR1022 epithelial cells during light exposure at
488 nm and 632 nm, respectively.179 The curves 1,3,5 and 6 show the transient Ca2 þ
increase in the nucleus as well as in the cytoplasm, whereas the curves 2 and 4 show the
fluorescence increase of AlPcS4 in the cytoplasm during laser irradiation. Reproduced from
Ref. 179 with permission by A. Rück and Photochemistry and Photobiology.
Although the first microscopes were built by van Leeuwenhoek (1632–

1723), over 300 years ago, most of the major developments in light
microscopy have occurred in just the past few decades.191 Several develop-
ments with profound impact on microscopy are under way. These include, for
example, adaptive optics, lens-free microscopy, super lenses, miniaturization,
and combinational microscopy approaches.
5.8.4 Multiphoton microscopy

Multiphoton microscopy using ultra-short laser pulse excitation has several
advantages compared with single photon microscopy: NIR (near infrared)
excitation light is less absorbed, and therefore, penetrates deeper into
biological tissues. No UV light and UV microscope optics are needed, and
multi-fluorophore excitation (up to 24 colors) is possible when using the
fluorescence in situ hybridization (FISH) technology.
282 Chapter 5
Non-destructive fluorescence imaging by multiphoton microscopy needs

excitation wavelengths within the optical window of biological tissues (700 to
1100 nm). Power density of irradiation should be carefully adjusted with
respect to cell viability. Above a certain intensity level, multiphoton laser
microscopy may cause cell damage, e.g., decreased cellular reproduction,182
formation of giant cells, oxidative stress, or apoptotic cell death.
Non-resonant two-photon excitation of electronic states in chromophores
based on simultaneous absorption of these photons was predicted in 1931 by
Göppert-Meyer183 and verified after the invention of lasers in 1961 by Kaiser
and Garret.184 The first two-photon fluorescence image in life sciences was
published again 30 years later by Denk et al.185 These authors used a laser
scanning microscope and focused the laser light to a diffraction limited spot
using an objective lens with high numerical aperture.192
Most applications are dedicated to fluorescent probes with simultaneous
absorption of at least two NIR photons in a wavelength range where single
photons will not be absorbed. Due to the low cross section for two photon
absorption, a photon flux density .1024 photons cm–2s–1 is required.185 This
corresponds to light intensities of some MWcm–2 to GWcm–2. NIR lasers
used, e.g., for laser tweezers with 100-mW light intensity and focused to a
diffraction-limited spot, are potential sources of two-photon excitation.186
Using modern solid state laser-pumped Ti:sapphire lasers with ultra-short
pulses in the 100 fs range and a repetition frequency around 80 MHz, two-
photon absorption is easily achieved. This technology allows for in vivo online
diagnosis of basal cell carcinomas visualizing xy sections of different depths
with image constructions of cross-sectional xz views.190
The efficiency of two-photon excitation is proportional to the square of
the laser power in the focal point. This is also the reason for the high
resolution of two-photon excitation. In Ref. 185 the relation between the
efficiency of two-photon excitation and the fluorescence yield is given by:
N ½P2 a∕ðtf 2 Þxðp2 A4 ∕ðhclÞ2 Þ (5.37)
with N ¼ the number of absorbed photon pairs, P ¼ mean power, a ¼

molecular two-photon cross section, t ¼ pulse width, f ¼ repetition frequency,
A ¼ numerical aperture, and l ¼ laser wavelength.
Eq. (5.37) shows that the fluorescence yield depends on P2/t in two-
photon microscopy. Therefore, the mean power of a laser pulse of 1 ps
duration must be only about 3 times higher than that of a 100-fs pulse in order
to obtain the same two-photon efficiency. This relation is useful for designing
experimental set-ups, where ultra-short light pulses are spread in time when
passing through dispersive transmission optics, e.g., single-mode fibers or
microscope objective lenses: Increasing the laser power often avoids complex
or expensive pulse-compression units. Typical laser pulses of commercially
available femtosecond lasers range between 45 and 200 fs, and those of
picosecond systems between 1 and 10 ps.
When using a femtosecond system (80fs, 80MHz), endogenous fluor-
ophores, e.g., fluorescent coenzymes, can be excited with a mean power of
about 2 mW,186 whereas fluorescence of intracellularly accumulated
fluorophores (DAPI, Hoechst 33258, Rhodamine123) is excited already at
an average power between 100 mW and 1 mW. A power level above 10 mW of
such a laser system leads to cell damage. At mean powers of about 30–50 mW
an optical breakdown can be induced, which is used to drill holes in cellular
structures or to cut chromosomes.
In NIR two-photon microscopy the spatial resolution is about the same as
in one-photon microscopy in spite of doubling the wavelength (lateral
FWHM: 0.3 mm; axial FWHM: 0.9 mm). With the introduction of a pinhole
in the detection path, the resolution can be slightly improved, and in Ref. 187
it was shown that the axial resolution reaches about 100 nm when using 4Pi
two-photon microscopy (with two opposing objective lenses). Recently even
higher resolutions have been achieved with two-photon SNOM (scanning
near-field optical microscope) technology.188,189
One of the advantages of multiphoton NIR microscopy is multifluor-
ophore excitation, where the visible fluorescence of a wide range of
fluorophores is excited simultaneously. This method is used for multicolor
detection of genes and chromosomes in combination with fluorescence in situ
hybridization (FISH). An example deduced from Ref. 182 is depicted in
Fig. 5.19, showing a 3D reconstruction from a set of multicolor images with
intranuclear localization of the FISH-labelled centromeric regions of
Figure 5.19 3D reconstruction of the FISH-labelled centromeric regions of chromosomes 8

and 18 in an interphase nucleus of an amniotic fluid cell indicating the presence of three
chromosomes 18 (Edward’s syndrome).182 Reproduced with permission of K. König and the
Journal of Microscopy6HHFRORUSODWHV
284 Chapter 5
chromosomes 8 and 18 in an interphase nucleus of an amniotic fluid cell,

indicating the presence of three chromosomes 18 (Edward’s syndrome).
NIR two-photon microscopy has a sufficiently good resolution even in
highly scattering tissues such as human skin. Fluorescent skin structures down
to a depth of 100 mm are easily resolved even at a moderate laser power. The
width of the point spread function (PSF) increases with depth from 0.34 mm
(lateral direction) and 0.9 mm (axial direction) on the surface to values of
0.45 mm and 1.53 mm, respectively, at 50 mm skin depth.182 Therefore, in the
future, multiphoton microscopy appears to be a promising technique in
cellular and medical diagnostics.190
5.8.5 Super-resolution and single-molecule detection

In 2014 three scientists, Stefan W. Hell, Eric Betzig, and William E. Moerner,
received the Nobel Prize in Chemistry for their pioneering work of super-
resolution far-field fluorescence microscopy or nanoscopy192–197 breaking the
diffraction barrier (Ernst Abbe: Dr l/(2nsina)) of 200 nm. In confocal
microscopy all fluorescing molecules emit at the same time in the focal area,
which makes their separation virtually impossible. In the 1990s Stefan Hell
developed a concept showing that in fluorescence microscopy, the diffraction
barrier can be broken with normal microscopes making use of the molecular
states of the fluorescent markers, a bright and a dark state. The concepts
established by the Nobel Prize winners differ on whether the sequential
recording of the marker occurs molecule by molecule or in molecule
ensembles.
Breaking Abbe’s barrier is about distinguishing an arbitrary number of
densely packed and similarly labeled features within any distance ,l/2n. This
is possible if the features can be recorded sequentially, for example, by
successively transferring the markers of each feature to a signal-giving
“bright” state A, while keeping the other markers in a “dark” state B.
Reading out the bright ones gives an assembly of a sub-diffraction images
provided that one knows their coordinates ri194,198 This would be possible by
applying an optical transition A ! B that would send all fluorophores to the
dark state B except from those at ri.198 The time-sequential readout from
within the diffraction zone at defined coordinates is a hallmark of “stimulated
emission depletion” (STED) microscopy (Fig. 5.20) and of other concepts
exploiting reversible saturable or photo-switchable transitions A ! B. These
approaches have been generalized under the acronym RESOLFT,194 which
stands for “reversible saturable optically linear fluorescence transitions.”
STED microscopy, which can be regarded as the first concept of the
RESOLFT type, uses the most elementary possible states: the S1 molecular
state as A and the molecular ground state S0 as B. Assuming an intensity
necessary for the transition of Is 3 1025 photons/cm²s (10 MW/cm2),
then applying Imax . Is yields sub-diffraction fluorescent spots with a
resolution: Dr l/(2nsina(1 þ Imax/Is)1/2).
Figure 5.20 Principle of STED microscopy using a regular focused excitation beam (blue)
that is superimposed by a doughnut-shaped STED beam (orange) that instantly quenches
excited molecules at the periphery of the excited spot, thus confining the fluorescence
emission to the doughnut zero (20 nm) (adapted from Ref. 194).
Figure 5.21 Targeted versus stochastic time-sequential readout of fluorophore markers of

a nanostructured object within the diffraction zone l/2n. A and B denote a bright and a dark
state, respectively. In the targeted readout mode all fluorescent molecules in a sub-
diffraction-sized spot are registered. In the stochastic readout mode, a single switchable
fluorophore from a random position of within the diffraction zone is switched to a stable state A,
while the other molecules remain in B. The position is calculated from the centroid of the
GLIIUDFWLRQIOXRUHVFHQFHVSRWE\DSL[HODWHGGHWHFWRUDGDSWHGIURP5HI6HHFRORUSODWHV
Eric Betzig introduced a method for optically imaging intracellular

proteins at nanometer spatial resolution.196 Numerous sparse subsets of
photoactivatable fluorescent protein molecules were activated, localized, and
then bleached. The aggregated position information from all subsets was then
286 Chapter 5
assembled into a superresolution image. The key is the isolation of single

molecules at high densities (up to 105/mm2) based on the serial photo-
activation and subsequent bleaching of numerous sparse subsets of photo-
activatable fluorescent protein (PA-FP) molecules. This technique, termed
photoactivated localization microscopy (PALM), is capable of resolving the
most precisely localized molecules at separations of a few nanometers
(Fig. 5.21, in comparison with STED and Fig. 5.22).
This process is repeated many times (C and D) until the population of
inactivated, unbleached molecules is depleted. Summing the molecular images
Figure 5.22 The principle behind PALM: A sparse subset of PA-FP molecules that are
attached to proteins of interest and then fixed within a cell are activated (A and B) with a brief
laser pulse at lact ¼ 405 nm and then imaged at lexc ¼ 561 nm until most are bleached (C)
6HHFRORUSODWHV
across all frames results in a diffraction-limited image (E and F). However, if

the location of each molecule is first determined by fitting the expected
molecular image given by the PSF of the microscope [(G), center] to the actual
molecular image [(G), left], the molecule can be plotted [(G), right] as a
Gaussian that has a standard deviation equal to the uncertainty sx,y in the
fitted position. Repeating with all molecules across all frames (A through D)
and summing the results yields a superresolution image (E and F) in which the
resolution is dictated by the uncertainties sx,y as well as by the density of
localized molecules. Scale: 1 1 mm2 in (F) and (F0), 4 4 mm2 elsewhere
(adapted from Ref. 196).
William E. Moerner contributed to render more precisely the location of
single emitters in wide-field microscopy by polarization filtering. Many single
nano-emitters, such as fluorescent molecules, produce dipole radiation that
leads to systematic position errors in both particle tracking and super-
resolution microscopy. Via vectorial diffraction equations and simulations,
Moerner showed that imaging only azimuthally polarized light in the
microscope naturally avoids emission from the z-component of the transition
dipole moment, resulting in negligible localization errors for all emitter
orientations and degrees of objective lens misfocus. Furthermore, localization
accuracy is maintained even in the presence of aberrations resulting from
imaging in mismatched media.199,200
5.9 Concluding Remarks

As demonstrated above, fluorescence proved to be one of the most powerful
and versatile techniques in biomedical diagnostics with still a great potential
for the future. Main achievements include resolution in the nanometer range
and highly specific signal detection. Present challenges, in particular in view of
future clinical applications, are label-free detection, real-time 3D imaging, and
avoidance of phototoxic damages of cells and tissues. Previous studies171
showed that the limit of non-phototoxic light doses is between 10 and 200 J/cm2
dependent on relevant fluorophores as well as on the wavelength of
illumination. This is about 100 times less than typical light doses used for
single-molecule techniques and about 10,000 times less than light doses
needed for STED microscopy. Therefore, much effort is still necessary for live
cell fluorescence imaging.
References
1. C. M. Gardner, S. L. Jacques, and A. J. Welch, “Light transport in
tissue: accurate expressions for one-dimensional fluence rate and escape
function based upon Monte Carlo simulation,” Lasers Surg. Med. 18,
129–138 (1996).
288 Chapter 5
2. A. Andreoni, R. Cubeddu, S. de Silvestri, G. Jori, and E. Reddi,

“Hematoporphyrin derivative: experimental evidence for aggregated
species,” Chem. Phys. Lett. 88, 33–36 (1982).
3. H. Schneckenburger, H. K. Seidlitz, and J. Eberz, “Time-resolved fluores-
cence in photobiology,” J. Photochem. Photobiol. B: Biol. 2, 1–19 (1988).
4. A. R. Holtzwarth, “Time-resolved chlorophyll fluorescence - what kind
of information on photosynthetic systems does it provide?” in:
Applications of Chlorophyll Fluorescence (H. K. Lichtenthaler, ed.),
Kluwer Acad. Publ. Dordrecht, 21–31 (1987).
5. E. H. Evans and R. G. Brown, “An appraisal of photosynthetic
fluorescence decay kinetics as a probe of plant function,” J. Photochem.
Photobiol. B: Biol. 22, 95–104 (1994).
6. H. Schneckenburger and W. Schmidt, “Time-resolved chlorophyll
fluorescence of spruce needles after different light exposure,” J. Plant
Physiol. 148, 593–598 (1996).
7. I. Hemmilä, S. Dakubu, V. M. Mukkala, H. Siitari, and T. Lovgren,
“Europeum as a label in time-resolved immunofluorometric assays,”
Anal. Biochem. 137, 335–343 (1984).
8. R. R. deHaas, N. P. Verwoerd, M-P. van der Corput, R. P.
vanGijlswijk, H. Siitari, and H. J. Tanke, “The use of peroxidase-
mediated deposition of biotin-tyramide in combination with time-resolved
fluorescence imaging of europeum chelate in immunohistochemistry
and in situ hybridization,” J. Histochem. Cytochem. 44, 1091–1099
(1996).
9. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd edition,
Springer Science þ Business, New York, 2006.
10. H. K. Seidlitz, H. Schneckenburger, and K. Stettmaier, “Time-resolved
polarization measurements of porphyrin fluorescence in solution and in
single cells,” J. Photochem. Photobiol. B: Biol. 5, 391–400 (1990).
11. P. Weber, M. Wagner, and H. Schneckenburger, “Fluorescence imaging
of membrane dynamics in living cells,” J. Biomed. Opt. 15(4), 046017,
(2010).
12. C. Hélène, “Excited states and photochemical reactions in DNA, DNA-
photosensitizer, and DNA-protein complexes,” Photobiochem. Photo-
biophys. Suppl., 3–22 (1987).
13. W. Schmidt, Optische Spektroskopie, VCH, Weinheim - New York -
Basel - Cambridge - Tokyo, 1994.
14. J. M. Beechem and L. Brand, “Time-resolved fluorescence of proteins,”
Ann. Rev. Biochem. 54, 43–71 (1985).
15. S. Andersson-Engels, J. Johansson, K. Svanberg, and S. Svanberg,
“Fluorescence imaging and point measurements of tissue: applications to
the demarcation of malignant tumours and atherosclerotic lesions from
normal tissue,” Photochem. Photobiol. 53, 807–814 (1991).
16. E. Fujimori, “Cross linking and fluorescence changes of collagen by

glycation and oxidation,” Biochim. Biophys. Acta 998, 105–110
(1989).
17. H. Schneckenburger, M. Gschwend, R.-J. Paul, H. Stepp, K. Rick,
V. Betz, and W. Strauss, “Time-gated spectroscopy of intrinsic
fluorophores in cells and tissues,” in: Optical Biopsy and Fluorescence
Spectroscopy and Imaging (R. Cubeddu, R. Marchesini, S. G. Mordon,
K. Svanberg, H. H. Rinneberg, and G. Wagnières, eds.), Proc. SPIE,
2324, 187–195 (1995).
18. T. Galeotti, G. D. V. vanRossum, D. H. Mayer, and B. Chance, “On the
fluorescence of NAD(P)H in whole cell preparations of tumours and
normal tissues,” Eur. J. Biochem. 17, 485–496 (1970).
19. J.-M. Salmon, E. Kohen, P. Viallet, J. G. Hirschberg, A. W. Wouters,
C. Kohen, and B. Thorell, “Microspectrofluorometric approach to the
study of free/bound NAD(P)H ratio as metabolic indicator in various
cell types,” Photochem. Photobiol. 36, 585–593 (1982).
20. H. Schneckenburger and K. König, “Fluorescence decay kinetics and
imaging of NAD(P)H and flavins as metabolic indicators,” Opt. Eng. 31,
1447–1451 (1992).
21. H. Schneckenburger, M. H. Gschwend, W. S. L. Strauss, R. Sailer,
M. Kron, U. Steeb, and R. Steiner, “Energy transfer spectroscopy for
measuring mitochondrial metabolism in living cells,” Photochem.
Photobiol. 66, 33–41 (1997).
22. K. A. Horvath, D. F. Torchiana, W. M. Daggett, and N. S. Nishioka,
“Monitoring myocardial reperfusion injury with NADH fluorometry,”
Lasers Surg. Med. 12, 2–6 (1992).
23. E. T. Obi-Tabot, L. M. Hanrahan, R. Cachecho, E. R. Berr, S. R.
Hopkins, J. C. K. Chan, J. M. Shapiro, and W. W. LaMorte, “Changes
in hepatocyte NADH fluorescence during prolonged hypoxia,” J. Surg.
Res. 55, 575–580 (1993).
24. W. Lohmann and E. Paul, “In situ detection of melanomas by
fluorescence measurements,” Naturwissenschaften 75, 201–202 (1988).
25. M. Anidjar, O. Cussenot, S. Avrillier, D. Ettori, J. M. Villette, J. Fiet,
P. Teillac, and A. LeDuc, “Ultraviolet laser-induced autofluorescence
distinction between malignant and normal urothelial cells and tissues,”
J. Biomed. Opt. 1, 335–341 (1996).
26. B. Banerjee, B. Miedema, and H. R. Chandrasekhar, “Emission spectra
of colonic tissue and endogenous fluorophores,” Am. J. Med. Sci. 315,
220–226 (1998).
27. L. Rigacci, R. Albertini, P. A. Bernabei, P. R. Feriini, G. Agati, F. Fusi,
and M. Monici, “Multispectral imaging autofluorescence microscopy for
the analysis of lymph-node tissues,” Photochem. Photobiol. 71, 737742
(2000).
290 Chapter 5
28. P. Weber, M. Wagner, P. Kioschis, W. Kessler, and H. Schneckenburger,

“Tumor cell differentiation by label-free fluorescence microscopy,”
J. Biomed. Opt. 17(10), 101508 (2012).
29. M. H. Gschwend, R. Rüdel, W. S. L. Strauss, R. Sailer, H. Brinkmeier,
and H. Schneckenburger, “Optical detection of mitochondrial
NADH content in human myotubes,” Cell. Mol. Biol. 47, OL95-
OL104 (2001).
30. S. DiMauro, E. Bonilla, M. Zeviani, M. Nakagawa, and D. C. DeVivo,
“Mitochondrial myopathies,” Ann. Neurol. 17, 521–526 (1985).
31. R. A. Capaldi, “Mitochondrial myopathies and respiratory chain
proteins,” Trends Biochem. Sci. 13, 144–148 (1988).
32. D. C. Wallace, “Diseases of the mitochondrial DNA,” Annu. Rev.
Biochem. 61, 1175–1212 (1992).
33. R. Luft, “The development of mitochondrial medicine,” Proc. Natl.
Acad. Sci. USA 91, 8731–8738 (1994).
34. A. H. V. Schapira, “Evidence for mitochondrial dysfunction in
Parkinson’s disease - a critical appraisal,” Mov. Disord. 9, 125–138
(1994).
35. P. R. Smith, J. M. Cooper, G. G. Govan, E. A. Harding, and A. H. V.
Schapira, “Smoking and mitochondrial function: a model for environ-
mental toxins,” Q. J. Med. 86, 657–660 (1993).
36. R.-J. Paul and H. Schneckenburger, “Oxygen concentration and the
oxidation-reduction state of yeast: determination of free/bound NADH
and flavins by time-resolved spectroscopy.” Naturwissenschaften 82,
32–35 (1996).
37. R. J. Paul, J. Gohla, R. Föll, and H. Schneckenburger, “Metabolic
changes in Caenorhabditis elegans,” Comp. Biochem. Physiol. B 127,
469–479 (2000).
38. B. Liang and H. R. Petty, “Imaging neutrophil activation: analysis of the
translocation and utilization of NAD(P)H-associated autofluorescence
during antibody-dependent target oxidation,” J. Cell. Physiol. 152,
145–156 (1992).
39. P. Galland and H. Senger, “The role of flavins as photoreceptor”
J. Photochem. Photobiol. B:Biol. 1, 277–294 (1988).
40. K. König and H. Schneckenburger, “Laser-induced autofluorescence for
medical diagnosis,” J. Fluoresc. 4, 17–40 (1994).
41. B. Kjeldstad, A. Johnsson, and S. Sandberg, “Influence of pH on
porphyrin production in Propionibacterium acnes,” Arch. Dermatol.
Res. 276, 296–400 (1984).
42. R. R. Alfano, W. Lam, H. J. Zarrabi, M. A. Alfano, J. Cordero, D. B.
Tata, and C. E. Swenberg, “Human teeth with and without caries
studied by laser scattering, fluorescence and absorption spectroscopy,”
IEEE-QE 20, 1512–1515 (1984).
43. K. König, G. Flemming, and R. Hibst, “Laser-induced autofluorescence

spectroscopy of dental caries lesions,” Cell. Mol. Biol. 44, 1293–1300
(1998).
44. K. König, H. Schneckenburger, and R. Hibst, “Time-gated in vivo auto-
fluorescence imaging of dental caries,” Cell. Mol. Biol. 45, 233–239 (1999).
45. H. Roempp and J. Falbe, Roempp-Chemie-Lexikon. 9th Edition, Thieme
Verlag, Stuttgart - New York (1995).
46. H. Beyer and H. Riesenberg, Handbuch der Mikroskopie. 3d Edition
VEB Verlag Technik, Berlin (1988).
47. N. Panchuk-Voloshina, R. P. Haugland, J. Bishop-Stewart, M. K. Bhalgat,
P. J. Millard, F. Mao, W. Y. Leung, and R. P. Haugland, “Alexa dyes, a
series of new fluorescent dyes that yield exceptionally bright, photostable
conjugates,” J. Histochem. Cytochem. 47, 1179–1188 (1999).
48. V. L. Singer, T. E. Lawlor, and S. Yue, “Comparison of SYBR Green I
nucleic acid gel stain mutagenicity and ethidium bromide mutagenicity
in the Salmonella/mammalian microsome reverse mutation assay (Ames
test),” Mutat. Res. 439, 37–47 (1999).
49. W. F. Patton, “A thousand points of light: the application of
fluorescence detection technologies to two-dimensional gel electrophore-
sis and proteomics,” Electrophoresis 21, 1123–1144 (2000).
50. P. Fortina, K. Delgrosso, T. Sakazume, R. Santacroce, S. Moutereau,
H. J. Su, D. Graves, S. McKenzie, and S. Surrey, “Simple two-color
array-based approach for mutation detection,” Eur. J. Hum. Genet. 8,
884–894 (2000).
51. I. Johnson, “Fluorescent probes for living cells,” Histochem. J. 30, 123–
140 (1988).
52. J. M. Mullins, “Overview of fluorochromes,” Methods Mol. Biol. 115,
97–105 (1999).
53. M. G. Bawendi, M. L. Steigerwald, and L. E. Bruns, “The quantum
mechanics of larger semiconductor clustres (“quantum dots”),” Annu.
Rev. Phys. Chem. 41, 477–496 (1990).
54. M. Bruchez, M. Moronne, P. Gin, S. Weis, and A. P. Alivisatos,
“Semiconductor nanocrystals as fluorescence biological labels,” Science
281, 2013–2016 (1998).
55. W. G. J. H. M. van Sark, P. L. T. M. Frederix, D. J. van den Heuvel,
A. A. Bol, J. N. J. van Lingen, C. de Mello Donega, H. C. Gerritsen, and
A. Meijerink, “Time-resolved fluorescence spectroscopy study on the
photophysical behaviour of quantum dots,” J. Fluoresc. 12, 69–76
(2002).
56. C. W. Cody, D. C. Prasher, W. M. Westler, F. G. Prendergast, and
W. W. Ward, “Chemical structure of the hexapeptide chromophore of
the Aequorea green-fluorescent protein,” Biochemistry 32, 1212–1218
(1993).
292 Chapter 5
57. R. Rizzuto, M. Brini, P. Pizzo, M. Murgia, and T. Pozzan, “Chimeric

green fluorescent protein as a tool for visualizing subcellular organelles
in living cells,” Curr Biol 5, 635–642 (1995).
58. M. Ikawa, S. Yamada, T. Nakanishi, and M. Okabe, “Green fluorescent
protein (GFP) as a vital marker in mammals,” Curr. Top. Dev. Biol. 44,
1–20 (1999).
59. A. Miyawaki, J. Llopis, R. Heim, J. M. McCaffery, J. A. Adams,
M. Ikura, and R. Y. Tsien, “Fluorescent indicators for Ca2 þ based on
green fluorescent proteins and calmodulin,” Nature 388, 882–887
(1997).
60. R. Brancato and G. Trabucchi, “Fluorescein and indocyanine green
angiography in vascular chorioretinal diseases,” Semin. Ophthalmol. 13,
189–98 (1998).
61. W. Schmidt, S. Koppenhöfer, and H. Schneckenburger, “Rapid scan
spectrometer (RSS),” GIT Laborfachzeitschrift 44(4), 441–443 (2000).
62. H. Schneckenburger, M. Lang, T. Köllner, A. Rück, M. Herzog,
H. Hörauf, and R. Steiner, “Fluorescence spectra and microscopic
imaging of porphyrins in single cells and tissues,” Lasers Med. Sci. 4,
159–166 (1989).
63. J. Hung, S. Lam, J. C. LeRiche, and B. Palcic, “Autofluorescence of
normal and malignant bronchial tissue,” Lasers Surg. Med. 11, 99–105
(1991).
64. H. J. C. M. Sterenborg, S. Thomsen, S. L. Jacques, and M. Motamedi,
“In vivo autofluorescence of an unpigmented melanoma in mice.
Correlation of spectroscopic properties in microscopic structure,”
Melanoma Res. 5, 211–216 (1995).
65. C. G. Coates, D. J. Denvir, N. G. McHale, K. D. Thornbury, and M. A.
Hollywood, “Optimizing low-light microscopy with back-illuminated
electron multiplying charge-coupled device: enhanced sensitivity, speed
and resolution,” J. Biomed. Opt., 9(6), 1244–1252 (2004).
66. M. El-Desouki, M. J. Deen, Q. Fang, L. Liu, F. Tsen, and
D. Armstrong, “CMOS image sensors for high speed applications,”
Sensors 9(1), 430–444 (2009).
67. W. Mueller-Klieser, S. Walenta, W. Paschen, F. Kallinowski, and
P. Vaupel, “Metabolic imaging of tumours and normal tissues with
bioluminescence and photon counting,” J. Natl. Cancer Inst. 80, 842–848
(1988).
68. Y. Sako, S. Minoguchi, and T. Yanagida, “Single-molecule imaging of
EGFR signalling on the surface of living cells,” Nature Cell. Biol. 2, 168–
172 (2000).
69. Z. Malik, D. Cabib, R. A. Buckwald, A. Talmi, Y. Garini, and S. G.
Lipson, “Fourier transform multipixel spectroscopy for quantitative
cytology, “ J. Microsc. 182, 133–140 (1996).
70. Z. Malik, M. Dishi, and Y. Garini, “Fourier transform multipixel

spectroscopy and spectral imaging of protoporphyrin in single mela-
noma cells,” Photochem. Photobiol. 63, 608–614 (1996).
71. Z. Malik, I. Amit, and C. Rothmann, “Subcellular localization of
sulfonated tetraphenyl porphines in colon carcinoma cells by spectrally
resolved imaging,” Photochem. Photobiol. 65, 389–396 (1997).
72. L. Greenbaum, C. Rothmann, J. Haniana, and Z. Malik, “Multi-pixel
spectral imaging of green fluorescent protein (GFP) in COS-7 cells:
Folding kinetics and chromophore formation,” in: Laser microscopy
(K. König, H. J. Tanke, and H. Schneckenburger, eds.), Proc. SPIE,
4164, 48–52 (2000).
73. D. V. O’Connor and D. Philipps, “Time-correlated single photon
counting,” Academic Press, London (1984).
74. W. Halle, W.-E. Siems, K. D. Jentzsch, E. Teuscher, and E. Göres, “Die
in vitro kultivierte Aorten-Endothelzelle in der Wirkstofforschung -
Zellphysiologische Charakterisierung und Einsatzmöglichkeiten der
Zellinie BKEz-7,” Pharmazie 39, 77–81 (1984).
75. N. Miyoshi, K. Hara, K. Yokoyama, G. Tomita, and M. Fukuda,
“Fluorescence lifetime of acridine orange in sodium dodecyl sulfate
premicellar solutions,” Photochem. Photobiol. 47, 685–688 (1988).
76. A. C. Allison and M. R. Young, “Uptake of dyes and drugs by living
cells in culture,” Life Sci. 3, 1407–1414 (1964).
77. W. Becker, “Advanced time-correlated single photon counting techni-
ques,” Springer, Berlin-Heidelberg-New York (2005).
78. J. R. Lakowicz, G. Laczko, I. Gryczinski, H. Szmacinski, and W. Wiczk,
“Gigahertz frequency domain fluorometry: resolution of complex
decays, picosecond processes and future developments,” J. Photochem.
Photobiol. B:Biol. 2, 295–311 (1988).
79. E. Gratton and M. Linkeman, “A continuously variable frequency cross-
correlation phase fluorometer with picosecond resolution,” Biophys.
J. 44, 315–324 (1984).
80. G. Wagnières, J. Mizeret, A. Strudzinski, and H. Van den Bergh,
“Frequence-domain fluorescence lifetime imaging for endoscopic clinical
cancer photodetection: apparatus design and preliminary results,”
J. Fluoresc. 7, 75–83 (1997).
81. A. Squire, P. J. Verveer, and P. I. Bastiaens, “Multiple frequency
fluorescence lifetime imaging microscopy,” J Microsc. 197, 136–49
(2000).
82. K. Suhling, P. M. French, and D. Phillips, “Time-resolved fluorescence
microscopy,” Photochem Photobiol Sci. 4, 13–22 (2005).
83. M. A. Digman, V. R. Caiolfa, M. Zamai, and E. Gratton, “The phasor
approach to fluorescence lifetime imaging analysis,” Biophys. J. 94(2),
L14–L16 (2008).
294 Chapter 5
84. P. Herman, B. P. Maliwal, H.-J. Lin, and J. R. Lakowicz, “Frequency-

domain fluorescence microscopy with the LED as a light source,”
J. Microsc., 203: 176–181 (2001).
85. H. Schneckenburger, M. Wagner, P. Weber, W. S. L. Strauß, and
R. Sailer, “Autofluorescence lifetime imaging of cultivated cells using a
novel uv picosecond laser diode,” J. Fluoresc. 14, 649–654 (2004).
86. R. Hutterer, F. W. Schneider, and M. Hof, “Time-resolved emission
spectra and anisotrpy profiles for symmetric diacyl- and dietherpho-
sphatidylcholines,” J. Fluoresc. 7, 27-3 (1997).
87. M. Kohl, J. Neukammer, U. Sukowski, H. Rinneberg, D. Wöhrle,
H.-J. Sinn, and E. A. Friedrich, “Delayed observation of laser-
induced fluorescence for imaging of tumors,” Appl. Phys. B 56, 131–138
(1993).
88. H. Schneckenburger, K. König, T. Dienersberger, and R. Hahn, “Time-
gated microscopic imaging and spectroscopy in medical diagnosis and
photobiology,” Opt. Eng. 33, 2600–2606 (1994).
89. H. Schneckenburger, M. H. Gschwend, R. Sailer, A. Rück, and W. S. L.
Strauss, “Time-resolved pH dependent fluorescence of hydrophilic
porphyrins in solution and in cultivated cells,” J. Photochem. Photobiol.
B:Biol. 27, 251–255 (1995).
90. W. S. L. Strauss, R. Sailer, H. Schneckenburger, N. Akgün,
V. Gottfried, L. Chetwer, and S. Kimel, “Study of the photodynamic
efficacy of naturally occurring porphyrins in endothelial cells in vitro and
microvasculature in vivo,” J. Photochem. Photobiol. B:Biol. 39, 176–184
(1997).
91. H. Schneckenburger, M. H. Gschwend, R. Sailer, H.-P. Mock, and
W. S. L. Strauss, “Time-gated fluorescence microscopy in molecular and
cellular biology,” Cell. Mol. Biol. 44, 795–805 (1998).
92. J. R. Lakowicz, H. Szmacinski, K. Nowaczyk, and M. L. Johnson,
“Fluorescence lifetime imaging of free and protein-bound NADH,”
Proc. Natl. Acad. Sci USA 89, 1271–1273 (1992).
93. C. G. Morgan, A. C. Mitchell, and J. G. Murray, “Prospects for
confocal imaging based on nanosecond fluorescence decay time,”
J. Microsc. 165, 49–60 (1992).
94. T. W. J. Gadella, T. M. Jovin, and R. M. Clegg, “Fluorescence lifetime
imaging microscopy (FLIM): Spatial resolution of microstructures on
the nanosecond time scale,” Biophys. Chem. 48, 221–239 (1993).
95. T. W. J. Gadella, A. van Hoek, and A. J. W. G. Visser, “Construction
and characterization of a frequency-domain fluorescence lifetime
imaging microscopy system,” J. Fluoresc. 7, 35–43 (1997).
96. I. Bugiel, K. König, and H. Wabnitz, “Investigation of cells by
fluorescence laser scanning microcopy with subnanosecond resolution,”
Lasers Life Sci. 3, 47–53 (1989).
97. E. P. Buurman, R. Sanders, A. Draijer, H. C. Gerritsen, J. J. F. van

Veen, P. M. Houpt, and Y. K. Levine, “Fluorescence lifetime imaging
using a confocal laser scanning microscope,” Scanning 14, 155–159
(1992).
98. R. Sanders, A. Draijer, H. C. Gerritsen, P. M. Houpt, and Y. K. Levine,
“Quantitative pH imaging in cells using confocal fluorescence lifetime
imaging microscopy,” Anal. Biochem. 227, 302–308 (1995).
99. R. Cubeddu, G. Canti, P. Taroni, and G. Valentini, “Time-gated
fluorescence imaging for the diagnosis of tumours in a murine model,”
100. A. D. Scully, A. J. MacRobert, S. Botchway, P. O’Neill, A. W. Parker,
R. B. Ostler, and D. Philipps, “Development of a laser-based
fluorescence microscope with subnanosecond time resolution,”
J. Fluoresc. 6, 119–125 (1996).
101. J. R. Lakowicz, H. Szmacinski, and M. L. Johnson, “Calcium imaging
using fluorescence lifetimes and long-wavelength probes,” J. Fluoresc. 2,
47–61 (1992).
102. B. Herman, P. Wodnicki, K. Seongwook, A. Periasamy, G. W. Gordon,
N. Mahajan, and X. F. Wang, “Recent developments in monitoring
calcium and protein interactions in cells using fluorescence lifetime
microscopy,” J. Fluoresc. 7, 85–91 (1997).
103. A. V. Agronskaia, L. Tertoolen, and H. C. Gerritsen, “Fast fluorescence
lifetime imaging of calcium in living cells,” J. Biomed. Opt. 9 (6) 1230–
1237 (2004).
104. H. C. Gerritsen, R. Sanders, A. Draaijer, C. Ince, and Y. K. Levine,
“Fluorescence imaging of oxygen in living cells,” J. Fluoresc. 7, 11–15
(1997).
105. H. C. Gerritsen, “Confocal fluorescence lifetime imaging,” In: Fluores-
cence Microscopy and Fluorescent Probes, J. Slavik (ed.), Plenum Press
New York - London, 35–46 (1996).
106. P. Weber, M. Wagner, and H. Schneckenburger, “Cholesterol dependent
uptake and interaction of doxorubicin in MCF-7 breast cancer cells,”
Int. J. Mol. Sci. 14, 8358–8366 (2013).
107. B. Herman, M. Sun, M. Qiu, and V. Centonze, “Protein interaction of
enzymatic activities monitores using FRET,” Cell. Mol. Biol. 46, 93
(2000).
108. B. R. Masters, B. T. C. So, and E. Gratton, “Multiphoton excitation
microscopy of in vivo human skin: functional and morphological optical
biopsy based on three-dimensional imaging, lifetime measurements and
fluorescence spectroscopy,” Ann. New York Acad. Sci. 838, 58–67
(1998).
109. D. Schweitzer, M. Hammer, F. Schweitzer, R. Anders, T. Doebecke,
S. Schenke, and E. R. Gaillard, “In vivo measurement of time-resolved
296 Chapter 5
autofluorescence at the human fundu” J. Biomed. Opt. 9, 1214–1222

(2004).
110. I. Munro, J. McGinty, N. Galletly, J. Requejo-Isidro, P. N. P. Lanigan,
D. S. Elson, C. Dunsby, M. A. A. Neil, M. J. Lever, G. W. H. Stamp,
and P. French, “Toward the clinical application of time domain
fluorescence lifetime imaging,” J. Biomed. Opt. 10(5), 051403 (2005).
111. H. Schneckenburger, K. Stock, M. Lyttek, W. S. L. Strauss, and
R. Sailer, “Fluorescence lifetime imaging (FLIM) of rhodamine 123 in
living cells,” Photochem. Photobiol. Sci. 3, 127–131 (2004).
112. D. Axelrod, “Cell-substrate contacts illuminated by total internal
reflection fluorescence,” J. Cell Biol. 89, 141–145 (1981).
113. J. S. Burmeister, G. A. Truskey, and W. M. Reichert, “Quantitative
analysis of variable-angle total internal reflection fluorescence micros-
copy (VA-TIRFM) of cell / substrate contacts,” J. Microsc. 173, 39–51
(1994).
114. B. P. Ölveczky, N. Periasamy, and A. S. Verkman, “Mapping
fluorophore distributions in three dimensions by quantitative multiple
angle-total internal reflection fluorescence microscopy,” Biophys. J. 73,
2836–2847 (1997).
115. K. Stock, R. Sailer, W. S. L. Strauss, M. Lyttek, R. Steiner, and
H. Schneckenburger, “Variable-angle total internal reflection fluores-
cence microscopy (VA-TIRFM): realization and application of a
compact illumination device,” J. Microsc. 211, 19–29 (2003).
116. A. L. Stout and D. Axelrod, “Evanescent field excitation of fluorescence
by epi-illumination,” Applied Optics 28, 5237–5242 (1989).
117. F. Schapper, J. T. Goncalves, and M. Oheim, “Fluorescence imaging
with two-photon evanescent wave excitation,” Eur. J. Biophys. 32, 635–
643 (2003).
118. H. Schneckenburger, K. Stock, W. S. L. Strauss, J. Eickholz, and
R. Sailer, “Time-gated total internal reflection fluorescence spectroscopy
(TG-TIRFS): application to the membrane marker laurdan,” J. Microsc.
211, 30–36 (2003).
119. I. Gryczynski, Z. Gryczynski, and J. R. Lakowicz, “Two-photon
excitation by the evanescent wave from total internal reflection
fluorescence,” Anal. Biochem. 247, 69–76 (1997).
120. E. H. Hellen and D. Axelrod, “Kinetics of epidermal growth factor/
receptor binding on cells measured by total internal reflection/
fluorescence recovery after photobleaching,” J. Fluoresc. 1, 113–128
(1991).
121. T. Suga, M. Hirano, M. Takayanagi, H. Koshimoto, and A. Watanabe,
“Restricted photorelease of biologically active molecules near the
plasma membrane,” Biochem. Biophys. Res. Commun. 253, 423–430
(1998).
122. Y. Sako and T. Uyemura, “Total internal reflection fluorescence

microscopy for single-molecule imaging in living cells,” Cell Struct.
Funct, 27, 357–356 (2002).
123. Y Suzuki, T Tani, K Sutoh, and S Kamimura, “Imaging of the
fluorescence spectrum of a single fluorescent molecule by prism-based
spectroscopy,” FEBS Lett 512, 235–239 (2002).
124. I. Todd, J. S. Mellor, and D. Gingell, “Mapping cell-glass contacts of
Dicyostelium amoeba by total internal reflection aqueous fluorescence
overcomes a basic ambiguity of interference reflection microscopy,”
J. Cell Sci. 89, 107–114 (1988).
125. W. M. Reichert and G. A. Truskey, “Total internal reflection
fluorescence (TIRF) microscopy (1) Modelling cell contact region
fluorescence,” J. Cell Sci. 96, 219–230 (1990).
126. H. Tatsumi, Y. Katayama, and M. Sokabe, “Attachment of growth
cones on substrate observed by multi-mode light microscopy,” Neurosci.
Res. 35, 197–206 (1999).
127. M. Wagner, P. Weber, H. Baumann, and H. Schneckenburger,
“Nanotopology of cell adhesion upon variable-angle total internal reflec-
tion fluorescence microscopy (VA-TITFM),” J. Vis. Exp. 68, e4133 (2012).
128. M. D. Wang and D. Axelrod, “Time-lapse total internal reflection
fluorescence video of acetylcholine receptor cluster formation on
myotubes,” Dev. Dyn. 201, 29–40 (1994).
129. M. Oheim, D. Loerke, R. H. Chow, and W. Stühmer, “Evanescent wave
microscopy: a new tool to gain insight into the control of transmitter
release,” Phil. Trans. R. Soc. Lond. B 354, 307–318 (1999).
130. A. Llobet, V. Beaumont, and L. Lagnado, “Real-time measurements of
exocytosis and endocytosis using interference of light,” Neuron 40, 1075–
1086 (2003).
131. S. S. Licht, A. Sonnleitner, S. Weiss, and P. G. Schultz, “A rugged
energy landscape mechanism for trapping of transmembrane receptors
during endocytosis,” Biochemistry 42, 2916–2925 (2003).
132. G. Omann and D. Axelrod, “Membrane-proximal calcium transients in
stimulated neutrophils detected by total internal reflection fluorescence,”
Biophys. J. 71, 2885–2891 (1996).
133. L. Cleemann, G. DiMasa, and M. Morad, “Ca2 þ sparks within 200 nm
of the sarcolemma of rat ventricular cells: evidence from total internal
reflection fluorescence microscopy,” Adv. Exp. Med. Biol. 430, 57–65
(1997).
134. A. Demuro and I. Parker, “Imaging the activity and localization of
single voltage-gated Ca2 þ channels by total internal reflection fluores-
cence microscopy,” Biophys J. 86, 3250–3259 (2004).
135. A. B. Matur, G. A. Truskey, and W. M. Reichert, “Atomic force and
total internal reflection fluorescence microscopy for the study of
298 Chapter 5
force transmission in endothelial cells,” Biophys. J. 78, 1725–1735

(2000).
136. T. Funatsu, Y. Harada, M. Tokunaga, K. Salto, and T. Yanagida,
“Imaging of single fluorescent molecules and individual ATP turnovers
by single myosin molecules in aqueous solution,” Nature 374, 555–559
(1995).
137. R. D. Vale, T. Funatsu, D. W. Pierce, L. Romberg, Y. Harada, and
T. Yanagida, “Direct observation of single kinesin molecules moving
along microtubules,” Nature 380, 451–453 (1996).
138. J. D. Spikes, “Photosensitization,” In: KC Smith (Hrsg) The Science of
Photobiology, 2. Edition, Plenum Press, New York, 79–110 (1989).
139. W. S. L. Strauss, R. Sailer, M. H. Gschwend, H. Emmert, R. Steiner,
and H. Schneckenburger, “Selective examination of plasma associated
photosensitizers using total internal reflection fluorescence spectroscopy
(TIRFS) - correlation between photobleaching and photodynamic
efficacy of protoporphyrin IX,” Photochem. Photobiol. 67, 363–369
(1998).
140. R. Sailer, W. S. L. Strauss, M. Wagner, H. Emmert, and
H. Schneckenburger, “Relation between intracellular location and
photodynamic efficacy of 5-aminolevulinic acid-induced protoporphyrin
IX in vitro – comparison between human glioblastoma cells
and other cancer cell lines,” Photochem. Photobiol. Sci. 6, 145–151
(2007).
141. H.-P. Lassalle, H. Baumann, W. S. L. Strauss, and H. Schneckenburger:,
“Cell-substrate topology upon ALA-PDT using variable-angle total
internal reflection fluorescence microscopy (VA-TIRFM),” J. Environ.
Pathol. Toxicol Oncol. 26, 83–88 (2007).
142. R. Sailer, W. S. L. Strauss, H. Emmert, K. Stock, R. Steiner, and H.
Schneckenburger, “Plasma membrane associated location of sulfonated
meso-tetraphenylporphyrins of different hydrophilicity probed by total
internal reflection fluorescence spectroscopy,” Photochem. Photobiol. 71,
460–465 (2000).
143. H. Schneckenburger, “Total internal reflection fluorescence microscopy:
technical innovations and novel applications,” Curr. Opin. Biotechnol.
16, 13–18 (2005).
144. H. Schneckenburger and M. Frenz, “Time-resolved fluorescence of
conifers exposed to environmental pollutants,” Radiat. Environ. Biophys.
25, 289–295 (1986).
145. H. K. Lichtenthaler, C. Buschmann, U. Rinderle, and G. Schmuck,
“Application of chlorophyll fluorescence in ecophysiology,” Radiat.
Environ. Biophys. 25, 297–308 (1986).
146. T. Förster, “Zwischenmolekularer Übergang von Elektronenanregung-
senergie,” Z. Elektrochem. 64, 157–164 (1960).
147. L. V. Johnson, M. L. Walsh, and L. B. Chen, “Localization of

mitochondria in living cells with rhodamine 123,” Proc. Natl. Acad. Sci.
USA 77, 990–994 (1980).
148. H. Port, H. Schneckenburger, and H. C. Wolf, “Host-guest energy
transfer via dipole-dipole interaction in doped fluorene crystals,” Z.
Naturforsch. 36a, 697–704 (1981).
149. A. C. E. Moor, “Signaling pathways in cell death and survival after
photodynamic therapy,” J. Photochem. Pbotobiol. B: Biol. 57, 1–13 (2000).
150. R. Bonnett, “Photosensitizers of the porphyrin and phthalocyanine series
for photodynamic therapy,” Chem. Soc. Rev. 19–33 (1995).
151. D. Siefermann-Harms, “The yellowing of spruce in polluted atmo-
spheres,” Photosynthetica 27, 3223–342 (1992).
152. P. S. Uster and R. E. Pagano, “Resonance energy transfer microscopy:
observations of membrane-bound fluorescent probes in model mem-
branes and in living cells,” J. Cell Biol. 103, 1221–1234 (1986).
153. J. Szöllösi, S. Damjanovich, S. A. Mulhern, and L. Tron, “Fluorescence
energy transfer and membrane potential measurements monitor dynamic
properties of cell membranes: a critical review,” Prog. Biophys. Molec.
Biol. 49, 65–87 (1987).
154. D. L. Taylor, J. Reidler, A. Spudich, and L. Stryer, “Detection of actin
assembly by fluorescence energy transfer,” J. Cell Biol. 89, 65–87 (1981).
155. T. C. Squier, D. J. Bigelow, J. G. deAncos, and G. Inesi, “Localization
of site-specific probes on the Ca-ATPase of sarcoplasmic reticulum using
fluorescence energy transfer,” J. Biol. Chem. 89, 362–367 (1987).
156. H. Schneckenburger, M. H. Gschwend, R. Sailer, W. S. L. Strauss,
M. Lyttek, K. Stock, and P. Zipfl, “Time-resolved in situ measurement
of mitochondrial malfunction by energy transfer spectroscopy,”
J. Biomed. Opt. 5, 362–366 (2000).
157. J. J. Lemasters, A. L. Nieminen, T. Qian, L. C. Trost, S. P. Elmore,
Y. Nishimura, R. A. Crowe, W. E. Cascio, C. A. Brandham, D. A.
Brenner, and B. Herman, “The mitohcondrial permeability transition in
cell death: a common mechanism in necrosis, apoptosis and autophagy,”
Biochim. Biophys. Acta 1366, 177–196 (1998).
158. N. P. Mahajan, K. Linder, G. Berry, G. W. Gordon, R. Heim, and
B. Herman, “Bcl-2 and Bax interactions in mitochondria probed with
green fluorescent protein and fluorescence resonance energy transfer,”
Nat. Biotechnol. 16, 547–552 (1998).
159. N. P. Mahajan, D. C. Harrison-Shostak, J. Michaux, and B. Herman,
“Novel mutant green fluorescent protein protease substrates reveal the
activation of specific caspases during apoptosis,” Chem. Biol. 6, 401–409
(1999).
160. B. Angres, H. Steuer, P. Weber, M. Wagner, and H. Schneckenburger,
“A membrane-bound FRET-based caspase sensor for detection of
300 Chapter 5
apoptosis using fluorescence lifetime and total internal reflection

microscopy,” Cytometry 75A, 420–427 (2009).
161. T. Bruns, B. Angres, H. Steuer, P. Weber, M. Wagner, and
H. Schneckenburger: “Förster resonance energy transfer-based total
internal reflection (TIR) fluorescence reader for apoptosis,” J. Biomed.
Opt. 14(2), 021003 (2009).
162. S. Brasselet, E. J. G. Peterman, A. Miyawaki, and W. E. Moerner, “Single-
molecule fluorescence resonant energy transfer in calcium concentration
dependent cameleon,” J. Phys. Chem. B 104, 3676–3682 (2000).
163. P. J. Verveer, F. S. Wouters, A. R. Reynolds, and P. I. Bastiaens,
“Quantitative imaging of lateral ErbB1 receptor signal propagation in
the plasma membrane,” Science 290, 1567–1570 (2000).
164. E. Kiyokawa, S. Hara, T. Nakamura, and M. Matsuda, “Fluorescence
(Förster) resonance energy transfer imaging of oncogene activity in
living cells,” Cancer Sci. 97, 8–15 (2006).
165. C. A. F. von Arnim, B. von Einem, P. Weber, M. Wagner,
D. Schwanzar, R. Spoelgen, W. S. L. Strauss, and H. Schneckenburger,
“Impact of cholesterol level upon APP and BACE proximity
and APP cleavage,” Biochem. Biophys. Res. Commun. 370, 207–212
(2008).
166. M. A. Neil, R. Juskaitis, and T. Wilson, “Method of obtaining optical
sectioning by using structured light in a conventional microscope,” Opt.
Lett. 22, 1905–1907 (1997).
167. M. G. L. Gustafsson, L. Shao, P. M. Carlton, C. J. R. Wang, I. N.
Golubovskaya, W. Z. Cande, D. A. Agard, and J. W. Sedat, “Three-
dimensional resolution doubling in wide-field fluorescence microscopy
by structured illumination,” Biophys. J. 94, 4957–4970 (2008).
168. G. Best, R. Amberger, D. Baddeley, T. Ach, S. Dithmar,
R. Heintzmann, and C. Cremer, “Structured illumination microscopy
of autofluorescent aggregations in human tissue,” Micron. 42(4), 330–
335 (2011).
169. S. Rossberger, G. Best, D. Baddeley, R. Heintzmann, U. Birk,
S. Dithmar, and C. Cremer, “Combination of structured illumination
and single molecule localization microscopy in one setup,” J. Opt. 15,
094003 (2013).
170. M. Wagner, P. Weber, T. Bruns, W. S. L. Strauss, R. Wittig, and
H. Schneckenburger, “Light dose is a limiting factor to maintain cell
viability in fluorescence microscopy and single molecule detection,” Int.
J. Mol. Sci. 11, 956–966 (2010).
171. H. Schneckenburger, P. Weber, M. Wagner, S. Schickinger, V. Richter,
T. Bruns, W. S. L. Strauss, and R. Wittig, “Light exposure and
cell viability in fluorescence microscopy,” J. Microsc. 245, 311–318
(2012).
172. J. Huisken and D. Y. R. Stainier, “Selective plane illumination

microscopy techniques in development biology,” Development 136,
1963–1975 (2009).
173. P. A. Santi, “Light sheet fluorescence microscopy: a review,”
J. Histochem. Cytochem. 59, 129–138 (2011).
174. T. Bruns, S. Schickinger, R. Wittig, and H. Schneckenburger, “Prepara-
tion strategy and illumination of 3D cell cultures in light—sheet based
fluorescence microscopy,” J. Biomed. Opt. 17(10), 101518 (2012).
175. M. Minsky, “Memoir on inventing the confocal scanning microscope,”
Scanning 10, 128–138 (1988).
176. K. R. Spring and S. Inoné, “Video Microscopy: The Fundamentals,
New York: Plenum Press (1997).
177. J. Pawley, Handbook of Biological Confocal Microscopy, Plenum Press,
New York (1990).
178. C. J. R. Sheppard and D. M. Shotton, Confocal Laser Scanning
Microscopy, BIOS Scientific Publishers Limited, Springer-Verlag
New York Berlin Heidelberg (1997).
179. A. Rück, K. Heckelsmiller, R. Kaufmann, N. Grossmann, E. Haseroth,
and N. Akgün, “Light-induced apoptosis involves a defined sequence of
cytoplasmic and nuclear calcium release in AlPcS4-photosensitized rat
bladder RR 1022 epithelial cells,” Photochem. Photobiol. 72, 210–216
(2000).
180. C. Rothmann, I. Bar-Am, and Z. Malik, “Spectral imaging for
quantitative histology and cytogenetics,” Histol. Histopathol. 13, 921–
926 (1998).
181. M. Schrader, H. T. M. van der Voort, and S. W. Hell, “Three-
dimensional super-resolution with 4Pi-confocal microscope using image
restoration,” J. Appl. Phys. 84, 4033–4042 (1998).
182. K. König, “Multiphoton microscopy in life science,” J. Microscopy, 200,
83–104 (2000).
183. M. Göppert-Meyer, “Über Elementarakte mit zwei Quantensprüngen,”
Göttinger Dissertation, Ann. Phys. 9, 273–294 (1931).
184. W. Kaiser and C. Garret, “Two-photon excitation in CaF2:Eu2 þ ,” Phys.
Rev. Lett. 7, 229–231 (1961).
185. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning
microscope,” Science, 248, 73–76 (1990).
186. K. König, “Two-photon near-infrared excitation in living cells,” J. Near
Infrared Spectrosc., 5, 27–34 (1997).
187. S. W. Hell, M. Schrader, and H. T. M. van der Voort, “Far-field
fluorescence microscopy with three-dimensional resolution in the 100 nm
range,” J. Microsc. 187, 1–7 (1997).
188. S. W. Hell, M. Booth, S. Wilms, C. M. Schnetter, A. K. Kirsch, D. J.
Arndt-Jovin, and T. Jovin, “Two-photon near and far-field fluorescence
302 Chapter 5
microscopy with continuous-wave excitation,” Opt. Lett. 23, 1238–1240

(1998).
189. A. Jenai, A. K. Kirsch, V. Subramaniam, D. Arndt-Jovin, and T. M.
Jovin, “Picosecond multiphoton scanning near-field microscopy,”
Biophys. J. 76, 1092–1100 (1999).
190. M. Balu et al., “In vivo Multiphoton Microscopy of Basal Cell
Carcinoma,” JAMA Dermatology, E1–E7, April 24, 2015.
191. S. J. M. Wollman et al., “From Animaculum to single molecules:
300 years of light microscope,” Open Biol. 5:150019 (2015). http://dx.doi.
org/10.1098/rsob.150019.
192. P. T. C. So, “Two-photon Fluorescence Light Microscopy,” Encyclope-
dia of Live Sciences (2002), Nature Publishing Group, www.els.net.
193. S. W. Hell and Jan Wichmann, “Breaking the diffraction resolution limit
by stimulated emission: stimulated-emission-depletion fluorescence
microscopy,” Optics Letters 19, 780–782 (1994).
194. S. W. Hell, “Far-Field Optical Nanoscopy,” SCIENCE 316, 1153–1158
(2007).
195. A. Chmyrov et al., “brief communications,” Nature methods 10, 737–740
(2013).
196. E. Betzig et al., “Imaging Intracellular Fluorescent Proteins at
Nanometer Resolution,” Science 313, 1642–1645 (2006).
197. T. A. Klar et al., “Fluorescence microscopy with diffraction resolution
barrier broken by stimulated emission,” PNAS 97, 8206–8210 (2000).
198. S. W. Hell, “Toward fluorescence nanoscopy,” Nature Biotechnology 21,
1347–1355 (2003).
199. M. D. Lew and W. E. Moerner, “Azimutal Polarization Filtering for
Accurate, Precise, and Robust Single-Molecule Localization Micros-
copy,” NANO Letters 14, 6307–6413 (2014).
200. Hsiao-Iu D. Lee et al., “The double-helix microscope super-resolves
extended biological structures by localizing single blinking molecules in
three dimensions with nanoscale precision,” Applied Physics Letters, 100,
153701 (2012) (doi: 10.1063/1.3700446).
Herbert Schneckenburger is a professor of Physics, Optics

and Biophotonics at Aalen University and a private lecturer
at the Medical Faculty of the University of Ulm. He received
his PhD in Physics from the University of Stuttgart in 1979
and his habilitation in Biomedical Technology from the
University of Ulm in 1992. His research is focused on the
fields of biomedical optics, 3D optical microscopy, and laser
spectroscopy, for which he has published about 250 scientific
papers and received 6 patents.
Wolfgang Strauß studied Chemistry at the University of

Ulm. Since 1989 he has been working at the Institut für
Lasertechnologien in der Medizin und Meßtechnik, Ulm,
Germany. His main research interests are focused on
photodynamic therapy (mechanisms of action, structure
activity relationships) and fluorescence diagnosis, drug
targeting (including cellular transport mechanisms to opti-
mize drug delivery) as well as applications of various
fluorescence microscopic and spectroscopic methods in the above mentioned
fields.
Karl Stock studied Precision Engineering at the University of

Applied Sciences in Aalen. He received his PhD in Human
Biology at the University of Ulm. Since 1990 he works at the
Institut für Lasertechnologien in der Medizin und
Meßtechnik, Ulm, Germany, since 2008 as the vice director
and head of development. His main interests and experiences
are in the development and introduction of new medical and
dental laser applications and of novel optical measurement
methods for medical and industrial applications.
Rudolf Steiner studied physics at the Technical University

Munich (TUM). He earned his PhD at the C.N.R.S. in
Montpellier, France, and passed his thesis at TUM in 1972.
After a one-year NATO fellowship in France, he joined
Heinrich Heine University (HHU) in Düsseldorf to build up
a laser laboratory for medical diagnostics at the Institute of
Clinical Physiology. His habilitation for biophysics was
1979. In 1985 he became Professor at HHU Düsseldorf. In
1986 he became director of the new “Institute of Laser Technologies in
Medicine and Metrology (ILM) at the University of Ulm” and a faculty
member of the Medical Faculty. After he retired in 2008 he continued
international research collaborations at ILM with Russia, Egypt, and South
Africa. He received several awards for innovations in medical technologies
2001 and 2003 and the “Germany Land of Ideas” award in 2006. He is also
member of several scientific societies.
Part IV: Coherent-Domain
Methods for Biological Flows
and Tissue Structure
Monitoring
Chapter 6 discusses Doppler, laser speckle, and other imaging techniques for
blood and lymph flow monitoring. After a brief description of blood, lymph,
and how they flow through vessels in the body, the authors present an
extensive review of existing methods of monitoring blood and lymph flow
in vivo. They begin with a discussion of the details and the interrelations
between Doppler and speckle techniques for the measurement of flow in single
vessels. This leads to an interesting discourse on two-wavelength near-infrared
speckle imaging. This technique allows for the assessment of not only blood
flow but also blood concentration based on spectroscopic reflectance
measurements. Starting with Section 6.5, the authors begin to describe
approaches for the direct, quantitative measurement of blood flow using
Doppler techniques and micro-particle image velocimetry, both by itself and
combined with intravital microscopy. In vivo fluorescent flow cytometry is
then discussed in some detail for both blood and lymph monitoring. The
remainder of the chapter is devoted to direct biomedical applications of
monitoring primarily lymph flow and optical lymphography.
Chapter 7 discusses the challenges of imaging the microstructure and
function of living tissue. Light scattering is so severe that good images are very
difficult to achieve using conventional imaging techniques. Spatial gating
techniques, such as confocal microscopy, produce some improvement, but
temporal gating methods are superior. Of these, coherence gating is the most
economical. Several techniques are available, but the most cost effective is
probably optical coherence tomography (OCT). The authors begin the
chapter with a thorough description of the principles of both time-domain and
frequency-domain OCT, including a discussion of the equipment needed to
produce quality OCT images. Functional imaging using various OCT-based
305
306 Part IV
methods is then addressed. The specific techniques discussed include Doppler

OCT and polarization-sensitive OCT. The former technique is useful for
quantifying motions and flows using OCT, and the latter is useful for imaging
changes in the scattering and birefringent properties of tissue. This permits for
the identification of abnormal or damages tissue based on variances in the
tissue architecture. The authors then describe some applications of OCT in
ophthalmology, cardiology (particularly related to the imaging of atheroscle-
rotic plaques), and oncology. Several clinical examples are given.
Chapter 8 describes some dynamic speckle techniques for imaging and
monitoring tissue. Diffusing-wave spectroscopy is an application of the
techniques of photon correlation spectroscopy to the time-varying multiply
scattered speckle fields produced by the motion of blood cells in vessels
illuminated with laser light. The authors describe this technique in some
detail, together with its successful implementation for diagnosing the depth of
damage in burned tissue. Another technique, laser speckle contrast analysis,
uses the spatial statistics of time-integrated images of the speckle fields. This
allows a full-field map of flow to be produced effectively in real time and is
cost-effective compared with correlation and scanning Doppler methods. It
has been applied to the mapping of blood flow in many scenarios, but perhaps
most significantly in the monitoring of cerebral blood flow during
neurological research and procedures. The authors describe a number of
modifications to the original technique, including the use of multiple
exposures. The effect of partial coherence is also described. Finally, speckle
techniques are combined with fiber optics for the monitoring of the thermal
modification of tissue.
Chapter 9 is devoted to the optical assessment of tissue mechanics,
specifically of strain measurement using laser speckle (optical elastography).
This has been applied to the detection of subsurface skin tumors, to the
diagnosis of skin diseases such as psoriasis, to the investigation of wound
closure after plastic surgery, and to tissue engineering such as cartilage repair.
A theory based on a constitutive model is presented, with the modifications
necessary for biological tissue. Speckle techniques applied to biological tissue
suffer from the problems of decorrelation and depolarization (the latter
caused by multiple scattering). There is therefore a need for speckle methods
that are insensitive to polarization properties, which leads the authors into a
discussion of speckle statistics and especially the effects on the statistics of
adding two speckle patterns incoherently—the result is poor fringe contrast in
speckle interferometry and speckle photography. Many of the problems are
avoided by using a so-called laser-speckle strain gauge. This technique is
described in some detail, together with improvements introduced by using a
Fourier transform approach. Applications to bone mechanics (e.g.,osteopo-
rosis effects) and arteries are described. Other methods of calculating the
speckle shift, both nonparametric (cross-correlation) and parametric
Coherent-Domain Methods for Biological Flows and Tissue Structure Monitoring 307
(minimum mean square error and maximum entropy methods) are

introduced, and all of these techniques are then compared. The use of
OCT-based elastography is greatly expanded upon in this edition and
examples of the application of elastographic techniques in ophthalmology,
dermatology, oncology, and tissue engineering are presented.
Chapter 10 takes an in-depth look at the theory and applications of
optical clearing of tissue and how optical clearing of tissue can be used to
improve optical imaging, diagnostics and light-based therapeutics in biology
and medicine. As the authors point out early in the chapter, the main
limitation of optical diagnostic methods is the strong scattering of light by
biological tissues and blood. Optical clearing seeks to reduce the deleterious
effects of excessive scattering. In this chapter, the authors discuss the physical
and molecular mechanisms of various optical clearing methods, including
methods based on tissue immersion, compression of the tissue and approaches
based on photodynamic and photothermal actions. The chapter starts with a
discussion of the efficacy of various optical clearing agents used in immersion
methods in increasing light penetration depth in various tissues. Following
this, the authors proceed to explain the details of compression optical clearing
and subsequently, photochemical, thermal and photothermal optical clearing
methods. The bulk of the remainder of the chapter is devoted to
demonstrating applications of optical clearing in diverse imaging modalities
such as OCT, fluorescent imaging, and photoacoustic imaging. Finally, the
challenging issue of determining diffusion coefficients of optical clearing
agents in tissues is discussed in some detail.
J. David Briers
Sean J. Kirkpatrick
Co-editors
Chapter 6
Laser Speckles, Doppler, and
Imaging Techniques for Blood
and Lymph Flow Monitoring
Ivan V. Fedosov
Yoshihisa Aizu
Muroran Institute of Technology, Muroran, Japan
Valery V. Tuchin
Saratov, Russia
Naomichi Yokoi
Asahikawa National College of Technology, Asahikawa, Japan
Izumi Nishidate
Tokyo University of Agriculture and Technology, Tokyo, Japan
Vladimir P. Zharov and Ekaterina I. Galanzha

University of Arkansas for Medical Sciences, Little Rock, USA
6.1 Introduction
Blood is a body fluid that circulates through the body through a network of
vessels to support the vital functions (e.g., nutrition, oxygenation, immunity)
of all organs. Blood is a two-phase suspension: (1) the liquid plasma, an
aqueous solution of organic molecules, proteins, and salt, and (2) solid
corpuscles such as normal and abnormal cells, their microparticles, and
aggregates. Normal blood consists of erythrocytes (so-called red blood cells
[RBCs]; 4.6–5.1 106/1 ml for humans], leukocytes (so-called white blood
309
310 Chapter 6
cells [WBCs]; 4,000–10,000/1 ml for human) and platelets [150,000–300,000/

1 ml for human]. Blood may also carry a variety of abnormal cells
(e.g., circulating tumor cells [CTCs], bacteria, leukemic cells), and viruses
and, thus, disseminate a disease from one organ to another.1–5
Owing to the overwhelming majority of RBCs, their red color, and their
ability to absorb light at specific wavelengths, blood vessels and blood flow can
be visualized and detected without labeling in a whole body.6–12 This advantage
of blood has been extensively used to develop in vivo assays for diagnosis of blood
vessels and blood flow using different modifications of optical (e.g., transmission
and fluorescent microscopies; multiphoton microscopy) and laser-based
(e.g., Doppler, laser speckle, and photoacoustic methods) techniques.10,13–17
Laser speckle and Doppler effect based methods addressed to blood flow
dynamics are based on measurements of the correlation and spectral charac-
teristics of fluctuations of scattered light at the diffraction region of the focused
Gaussian beam and their application to measurements of blood or lymph flow
in a single vessel. On one hand, this scattering phenomenon is treated as
inhomogeneous dynamic biospeckles or, with some conditions, speckled bio-
speckles by means of the interpretation of speckle fluctuations. On the other
hand, the same phenomenon is discussed with relation to the spectral
broadening of Doppler shifts by means of the interpretation of the Doppler
effect. This means that there are different possible approaches that are used to
interpret the above-mentioned scattering phenomenon. Therefore, to consider
the relation of the Doppler effect and the speckle fluctuations is meaningful for
understanding the phenomenon from different points of view. In this section,
we briefly discuss the interrelation between the Doppler and speckle techniques
for general cases of measurements of blood flows in a single vessel.18,19
As is well known, both the Doppler effect and speckle fluctuations are
based on dynamic light scattering and are applied to blood flow monitoring.
The terms “laser Doppler” and “laser speckle” are, then, carelessly mingled
very often in the studies of blood flow measuring techniques. Due to their
different principles, however, what is detected from each is originally different
from each other: periodic beat signals in the Doppler technique and random
intensity fluctuations in the speckle technique. In some aspects, there is
equivalence between them.20
The absolute rate of blood flow in macro- and microvessels is one of the most
important functional characteristics of the cardiovascular system. Specifically,
microvascular blood flow is critical for early diagnostics of glaucoma and
diabetic retinopathy21–24 as well as for the understanding of local blood flow
regulation mechanisms in organs and tissues at cerebral stroke25 or myocardial
ischemia.26 Only the absolute blood flow rate can be considered as a charac-
teristic of the normal and pathological status of an organism and as a
quantitative measure of the transport and exchange functions of microcircula-
tion. Despite the practical and fundamental importance of the volumetric flow
Techniques for Blood and Lymph Flow Monitoring 311
rate, it is difficult to measure it noninvasively because most of the optical

techniques for blood microcirculation diagnostics like laser Doppler flowmetry
(LDF),27–29 laser speckle contrast analysis (LASCA),25,30,31 diffusing wave
spectroscopy (DWS),32,33 and fluorescent34 and optical coherent angiography35
are aimed for the imaging of blood vessels and for detection of relative variations
of blood perfusion.
Currently only laser Doppler anemometery (LDA)21–23,36–38 and Doppler
optical coherent tomography (DOCT)24,39–41 are capable of quantitative flow
velocity measurements in blood vessels up to several hundred mm in diameter,
thus they can be used to determine the absolute flow rate. Absolute flow rate
in the smallest capillaries less than 10 mm in diameter can also be measured
using intravital microscopy, capable of tracking movement of individual red
blood cells in superficial capillaries.42,43 Although the potential of these
techniques is accurate measurements of in vivo absolute blood flow rate, they
are still challenging.
The use of Doppler effect based techniques for this purpose faces two
principal issues: 1) angular uncertainty because of unknown blood vessel
direction and 2) scattering of light along its path through a living tissue.
A straightforward solution for the first problem is the measurement of two or
three components of the flow velocity vector. The method is referred to as bi-
directional LDA37 and is used for measurements of blood flow velocity in
retinal arterioles and venules.21–23,37 As was shown in Ref. 37 the specific
location of these vessels at the interface between the retina and transparent
ocular media minimizes the ambiguity of the scattering vector caused by
multiple scattering of light. Therefore, a Doppler frequency shift (DFS) cutoff
frequency corresponding to a centerline flow velocity for each scattering
direction can be assessed with a simple fundus camera based LDA
arrangement.37 Nevertheless, the effect of multiple scattering in the blood and
retina prevents precision measurements, and the filtering of multiply scattered
light with pinhole aperture is required to measure the flow velocity profile.36
When implemented in DOCT, the capability for measurements of three
components of the velocity vector require extremely complicated hardware.41
In this chapter, we discuss a way to handle these issues with the use of a
differential LDA arrangement38,44 that is less sensitive to angular uncertainty
and enables to suppress the effect of multiple scattering with confocal detection
and a recently proposed data acquisition and processing method.38,45
Intravital microscopy addresses measurements of red blood cell velocity in
narrow superficial capillaries in that the movement of individual cells can be
tracked with imaging techniques. Currently, this technique is mostly based on
confocal microscopy. It enables high-resolution 3D imaging of a capillary
network and circulating cells, but requires fluorescent labeling of plasma for
fluorescent angiography34 or the tagging of individual cells.42,43,46 The
labeling complicates the use of the method and makes it invasive or requires
312 Chapter 6
the use of a specially prepared experimental environment. The field of view of

the confocal microscope is small because of the scanning operation principle.
However, from the aspect of studying blood microcirculation, the main
interest in the study of a native capillary network is the use of label free
microscopy for capillary networks.47 Another problem with intravital
microscopy concerns involuntary movements of living objects during
microscopic examination. This results in displacement of observed capillaries
over the microscope field of view. That motion could not be compensated
completely with any mechanical fixation of the living object because of the
softness of living tissue and native cardiac and respiratory activity.48 In this
chapter, we discuss the use of a software image stabilization method for label-
free quantitative measurements of capillary blood flow with microparticle
image velocimetry (mPIV) technique.
The great interest in blood testing and its clinical relevance is determined by
the well-established facts that blood flow and its composition can accurately
diagnose many severe human diseases (e.g., cardiovascular diseases, cancer,
atherosclerosis, inflammation, infections) and reflect the efficacy of therapeutic
interventions (e.g., infusion of heparin or warfarin).49–65 For example, in vivo
detection of apoptotic cells is crucial for understanding the fundamentals of
metabolic and immune functions, optimization of radiation therapy and
chemotherapy, diagnosis of many diseases (e.g., rheumatoid arthritis,
metastasis development, and Alzheimer’s, Huntington’s, and Parkinson’s
diseases) and assessment of acute organ transplant rejection or the effect of
immunosuppressive drugs.56–58 Tumor cells are continuously shed from
primary solid tumors (e.g., melanoma, breast cancer) into the blood flow and
disseminate through the body to develop deadly metastasis.16,17,59–65
Lymph is another common medium in the human body, but it is currently
poorly diagnosed.66–68 Lymph, similar to blood, consists of plasma and cells.
However, since the majority of lymphatic cells are optically transparent
WBCs, lymph is a colorless suspension66,67,69,70 As a result, finding the lymph
channel in vivo is not easy and requires additional labeling (lymphography). In
addition, lymph sampling is impractical because it yields only a few
microliters at a time and requires long-term cannulation.69 This has
represented a big diagnostic challenge for many years.
At the same time, considerable data underscore the clinical significance of
routine evaluation for lymph flow for staging, prognosis, and disease
recurrence after surgery, radiation, and chemotherapy.66–132 Along with the
key role of lymphatic drainage in diagnosis, prognosis and resolving
lymphedema, numerous recent studies have demonstrated the high impor-
tance of cell trafficking by lymph flow for the functioning of a whole
organism. Transport of normal and abnormal cells by lymph flow can induce
positive (e.g., transport of lymphocytes to initiate immune responses) or
negative (e.g., dissemination of cancer cells to initiate metastasis) effects. For
Figure 6.1 Schematic of blood and lymph circulation6HHFRORUSODWHV
example, the features of intra- and peri-tumor lymphatics (e.g., gaps in the
lymphatic wall due to tumor-induced interstitial pressure and dilation) suggest
relatively easy entry of tumor cells into lymph vessels.66,133 However, only a
few studies (vs. hundreds of studies of blood CTCs) have demonstrated CTCs
in lymph.119,134–137
Furthermore, blood and lymph systems have close relationships with
multiple cross-pathways for lymph and blood, allowing passage of cells and
other compounds from one system to another (Fig. 6.1).138–140
Thus, detection and imaging of blood and lymph flows and circulating
individual cells, especially in vivo in their native state, is highly important for
the early diagnosis and therapy (theranostic) of many diseases (e.g., cancer,
diabetes) and for the study of the impact of environmental and therapeutic
agents (e.g., radiation, drugs) on living organisms at the single-cell level.
Because of this, both blood and lymph flows and their composition should be
examined to provide the right understanding of disease and, in turn, lead to
the development of new advanced diagnoses and therapies. In this chapter we
discuss optical approaches used to date for characterization of blood flow at
various structure levels of an organism. On a tissue or an organ scale, multiple
scattering of light typically prevents exact characterization of blood
independent of the whole tissue structure. But on this scale, laser Doppler
and laser speckle based techniques provide useful ways for quantification of
relative blood flow and concentration changes under various physiological
conditions as outlined through Secs. 6.1 to 6.4. At the level of individual
microvessels, multiple scattered light can be filtered out, thus the absolute
flow becomes accessible with laser Doppler techniques as well as with particle
image velocimetry (PIV) methods, both of which are discussed in Sec. 6.5.
And finally, at the cellular level corresponding to individual blood and
lymphatic vessels accessible with numerous methods of intravital microscopy,
a detailed characterization of circulating cells becomes available as it is
discussed in Secs. 6.6 to 6.12 and Sec. 6.13, overviewing clinical applications
of intravital microscopy of blood and lymphatic vessels.
314 Chapter 6
6.2 Doppler and Speckle Techniques

6.2.1 Laser Doppler technique
Let us briefly review the basic principle of the laser Doppler technique that
uses a reference beam with a backscattering configuration.18,71 An incident
beam is focused onto a blood vessel and scattered by a moving RBC. The
scattered light is Doppler-shifted in frequency and then received in a specified
direction. The non-shifted reference light is mixed with the scattered light in
the same direction, thus the heterodyne detection is realized at the detector
surface, which produces the well-known Doppler beat signals. The beat
frequency F0 of the signals is given by
1
F0 ¼ ðk k0 Þ · V, (6.1)
2p s
where k0 and ks denote the wave vectors of the incident and scattered light,
and V is the velocity vector of a moving RBC. The velocity can be determined
by measuring the Doppler beat frequency providing that the geometry of k0,
ks, and V is specified. It should be noticed that this principle originally
assumes a single scattering by a single RBC. Measurements with this principle
may also be possible for a very dilute suspension of RBCs. As illustrated in
Fig. 6.2(a), the laser Doppler technique works quite well in the case of a single
RBC or a very low concentration of RBCs, and there is no speckle produced
here because of very few scatterers. The output signal I(t) shows a periodic
waveform with a speckle-like envelope, the frequency spectrum P(f ) contains
(a)
(b)
(c)
Figure 6.2 Schematic illustration of detected signals, frequency spectra, and correlation
functions for typical cases of (a) low, (b) moderate, and (c) high RBC concentration.
a Doppler-shift peak at fD, and the correlation function g(t) shows a

periodically oscillating curve. This condition is, however, usually not satisfied
for in vivo blood flow measurements.
The next step is to consider the moderate concentration of RBCs in a little
more realistic case. There could be a transient situation from a single scattering
to multiple scattering, which may produce only fractionally time-varying
speckle fluctuations while the Doppler beat components still remain.19 The
detected signals are possibly the Doppler beat with the envelope of speckle-like
random fluctuations as shown in Fig. 6.2(b). The new type of manifestation of
the Doppler effect72 generated by the strongly focused Gaussian beam seems to
be of this type. When the incident laser beam is divergent (not focused at the
scattering point) and the aperture of a receiving lens is enlarged, the coherence
condition73 for the optical beating can be corrupted due to the ranges of both
the incident and scattering angles. In addition to this, the number of scattered
waves having different optical path lengths increases, and that may enhance the
random interference. The heterodyne component is, then, degraded and the
speckle formation may be promoted in this situation. Thus, the Doppler-shift
peak of the frequency spectrum becomes somewhat unclear.
A high concentration of RBCs is the situation that is treated in actual blood
flow measurements in a vessel with a larger diameter. Increased concentration
causes multiple scattering. Random distribution of many RBCs, random
directions of numerous scattered waves, and a range of velocities may generate
randomly distributed Doppler-shifted frequencies. Random positions of RBCs
yield randomness in the optical path lengths of scattered waves. All of these
effects fully randomize the phases of scattered waves in the observation plane,
then their phase-consistent time or the correlation time ts is significantly
shortened. When the phase-consistent time becomes shorter than the period tD
of one cycle of the Doppler beat signals, the Doppler heterodyne components
are significantly deteriorated and finally wiped out by randomized phase
fluctuations. A typical detected signal, frequency spectrum, and correlation
function are schematically illustrated in Fig. 6.2(c). This situation may be better
interpreted as speckle fluctuations. If one would interpret the situation still
from the standpoint of the Doppler technique, it may be said that the
“homodyne” or mixing of various Doppler-shifted waves with themselves, may
be regarded as an extended scheme of the ordinary Doppler technique. The
homodyne or speckle fluctuations still carry information on the blood flow
velocity, thus relative measurement of the velocity is possible.
6.2.2 Laser speckle technique

The use of a finite size of an incident beam spot and a range of scattering angles
for a moving diffuser makes it easy for the optical system to produce a moving
speckle pattern. Here we consider the blood flow in a single microvessel instead
of the diffuser. An incident beam illuminates a relatively extended area so that
316 Chapter 6
the area should contain a number of randomly distributed RBCs inside. The light
waves scattered by moving RBCs are coherently superposed in random phases
and interfere with each other in the observation plane. This results in a biospeckle
pattern that is time-varying with the RBCs’ motion. Consider a scattered wave Aj
coming from j-th RBC (j ¼ 1 N). The resultant amplitude A(x, t) detected at a
certain point x and a time t in the observation plane is given by
XN
Aðx,tÞ ¼ jAj ðx,tÞj exp½ifj ðx,tÞ, (6.2)
j¼1
where fj(x, t) is a random phase. The randomness in the phase originates from
the randomized optical path length of scattered waves due to multiple
scattering by randomly distributed moving RBCs and their random
interference. Equation (6.2) means that the biospeckle intensity in the pattern
shows space-time random fluctuations. The increased concentration of RBCs
may enhance the randomization, and the phase-consistent or phase-correlating
time is more shortened than that for the ordinary speckle dynamics obtained
with the diffuser. Therefore, this situation usually generates higher-frequency
speckle fluctuations than the ordinary one, and it may be regarded as an
extended scheme of the ordinary speckle technique. In this optical system, the
heterodyne beat component can hardly be detected even if the reference beam
is introduced and the concentration of RBCs is low, since the incident and
scattered wave vectors have ranges in their directions and the coherence
condition is unsatisfied. In spite of the random phenomena, biospeckle
fluctuations reflect the blood flow velocity, and their autocorrelation function
or power spectrum can be used for monitoring the blood flow.
When the incident beam is strongly focused, the number of scatterers
becomes less than that for the above case. Coherent addition of the
insufficient number of scattered waves with random phases produces an
inhomogeneous speckle pattern. As the diameter of the blood vessel becomes
large, the number of scatterers that contribute to the speckle formation
increases and the phase randomization is further enhanced. This results in,
again, higher-frequency speckle fluctuation or possibly speckled speckles.
Even in these different types of speckles, their dynamics originates from the
motion of the RBCs. Thus, their correlation or spectral properties carry
information on the velocity of the RBCs.
6.2.3 Interrelation
As far as blood flow measurements are concerned, the heterodyne Doppler
beat component cannot be clearly obtained. Recent techniques using optical
coherence tomography have ingeniously provided a solution to this problem.
In conventional techniques, however, high-frequency speckle fluctuations, or
possibly homodyne components, are usually dominant in blood flow
measurements. Providing that the Doppler technique reserves only the
heterodyne component, whether the phenomenon considered here is Doppler

or speckle may depend on the optical geometry employed, the concentration
of RBCs, and the physical structure of vessels and surrounding tissues.
The ordinary Doppler technique for a single scattering and the ordinary
speckle technique for a single diffuser can clearly be compared, and their
equivalence and difference can easily be described.19,20 However, the
scattering phenomena from blood and lymph flows are not placed directly
under the categories of these ordinary techniques. The above scattering
phenomena can be interpreted as an extended scheme of ordinary Doppler or
ordinary speckle. It should be noted here that such an extension is possible
both from Doppler and speckle. We consider that there is no clear border
between the two extended schemes. Therefore, a choice between the two terms
“Doppler” and “speckle” for the scattering phenomena being considered is
unproductive and meaningless. What is most important is to understand the
phenomena from different points of view, which promotes the correct
interpretation of their measurement characteristics.
Recent blood flow monitoring techniques are based on imaging modality.
This is being encouraged by advanced image sensing technology, including
image processing software. In this modality, the terms “Doppler” and “speckle”
are still used ambiguously. Serov and his co-workers have utilized a
complementary metal-oxide semiconductor (CMOS) image sensor in the
scheme of laser Doppler perfusion imaging.74–76 The group of Fujii developed
their laser speckle flowgraphy (LSFG) system using a commercially-available
charge-coupled device (CCD) camera.77,78 Choi et al. have also reported speckle
imaging based on a CCD camera.79 However, it should be noted that some
recent studies treat the relationship of laser Doppler and speckle phenomena in
blood flow imaging. Serov et al. studies Doppler-induced speckle fluctuations.80
Rajan et al. discusses speckle effects on laser Doppler perfusion imaging, with a
relation to the concentration of scattering particles.81,82 As was mentioned in
Sec. 6.2.1, the particle concentration influences the degree of scattering and the
coherence condition, which governs a phenomenon predominating between the
Doppler effect and speckle fluctuation.
6.3 Two-Wavelength Near-Infrared Speckle Imaging

6.3.1 Optical system
Figure 6.3(a) illustrates the basic optical system used for detecting speckle
patterns. Two laser diodes (LDs) with wavelengths of 780 nm and 830 nm are
employed as laser sources. The reason why two different wavelengths in the
near infrared range are employed is that they are effective for measuring not
only blood flow but also blood concentration change by means of the
spectroscopic reflectance measurement, which is described in Sec. 6.3.5. Light
rays from two LDs simultaneously illuminate the subject to be measured.
318 Chapter 6
CCD camera B Observation area

Camera lens
Interference filter
(@830nm) Polarizer
CCD camera A
LD @780nm
BS
PBS
Interference filter
(@780nm)
LD @830nm PC
Measurement area
Subject
(b)
(a)
Sphygmo-
manometer
Normal Put into water Recovery process
RSD
Pulse oxy. a b c d e f g h
0 60 120 180 240 300 360 420 480
Time [s]
(c)
Figure 6.3 (a) Schematic diagram of the optical system for detecting speckle patterns,
(b) picture of the volunteer’s left wrist being subjected to measurements of blood flow and
blood pressure, and (c) illustration of a time table of measurements.
The scattered light from the subject is divided into two light fields with
equivalent power. Each of the two fields passes through the polarizer,
enters an interference filter whose center wavelength is 780 nm or 830 nm,
respectively, passes through a camera lens and, then reaches the CCD camera,
at which a speckle pattern corresponding to a wavelength of 780 nm or 830 nm
is separately observed.
6.3.2 Frame-rate analysis of blood flow

To visualize the local speed of blood flow, estimation parameters such as the
square blur rate (SBR)77 and the average derivative (AD)83 are available. The
reciprocal spatial difference (RSD),84,85 which is based on the spatial contrast of
the speckle image, is especially useful for frame rate analysis of blood flow. For
calculating RSD, a processing unit, which is a square of pixels, is introduced into
image data. RSD at the processing unit of size p p pixels can be given as
X
p X
p 2
I x,y,n ∕p2
x¼1 y¼1
RSD ¼ 2 , (6.3)
p X
X p
ðI x,y,n I 0.5ðpþ1Þ,0.5ðpþ1Þ,n Þ ∕ðp 1Þ
2
x¼1 y¼1
where Ix,y,n is the signal intensity at a certain pixel in the nth frame whose
horizontal and vertical coordinates inside the processing unit are x and y,
respectively, and p is the pixel number forming the processing unit in x and y.
RSD is expected to be increased as the speed of the blood flow becomes high.
6.3.3 Blood flow measurements in humans

Figure 6.3(b) shows a picture of a volunteer’s left wrist, on which the cold
stimulation was applied by cold water. To detect changes in blood flow and
blood pressure caused by the cold stimulation, temporal changes in the
maximum values of RSD and sphygmomanometer output were investigated
simultaneously. Figure 6.3(c) shows a time table of measurements performed
in the present study. We imaged the blood flow in the observation area as
shown in Fig. 6.3(b), in which the artery was located in the left side portion.
Figure 6.4(a) shows an example of temporal variation of the RSD value
averaged over the measurement area on the artery as shown in Fig. 6.3(b)
before putting the hand in cold water. As seen from Fig. 6.4(a), the RSD value
demonstrates periodical peaks, and the time intervals between adjacent
maximum values of RSD seem to be regular. Similar results were also
obtained during and after the cold stimulation, although the maximum values
of RSD were different from each other. The correlation between temporal
changes in RSD and the sphygmomanometer output due to the cold
stimulation was further investigated. Figure 6.4(b) shows plots of the
maximum values of RSD and sphygmomanometer output during the periods
of a–h in Fig. 6.3(c). As seen from Fig. 6.4(b), the maximum value of
the sphygmomanometer output increases while the maximum value of the
RSD decreases during the period of 60–240 s corresponding to b, c, and d in
Fig. 6.3(c). That is, the systolic blood pressure rises while the blood flow
decreases during the period when the hand is in cold water. Generally, a rise in
systolic blood pressure is thought to be caused by contraction and hardening
of the peripheral blood vessels. In our experiments, peripheral blood vessels
Blood pressure (high) [mmHg]
250 150 230

RSD
200 140
RSD [arb. unit]
RSD [arb. unit]
220
150 130
100 120
210
50 110 Output of
sphygmomanometer
0 100 200
5 10 15 20 25 30 0 60 120 180 240 300 360 420 480
Time [s] Time [s]
(a) (b)
Figure 6.4 (a) Temporal variation of RSD value averaged over the area on the artery and
(b) temporal variations of the maximum values of RSD and sphygmomanometer output.
320 Chapter 6
[arb. unit] [arb. unit]

255 255
Observation area
0 0
(b) (c)
[arb. unit] [arb. unit]
255 255
Medication place
0 0
(a) (d) (e)
Figure 6.5 (a) Picture of an anesthetized rat used for observation of blood flow and
(b)–(e) blood flow images of the ear of the anesthetized rat obtained by RSD prior to dosing,
immediately after dosing, 5 min. after dosing, and 10 min. after dosing, respectively.
inside the wrist were naturally contracted and hardened by cold stimulation,
thus the systolic blood pressure was raised. At the same time, blood flow in the
artery was decreased by the rise of systolic blood pressure due to the increase
of peripheral vascular resistance. Thus, results in Fig. 6.4(b) are found to be
true to the physiologic knowledge mentioned above.
6.3.4 Blood flow measurements in rats

Figure 6.5(a) shows a picture of the anesthetized rat used for observation of
blood flow. We imaged the blood flow on the ear of the rat as shown in
Fig. 6.5(a), in which the inflammation was induced. Here, a place of
medication is indicated by a circle in the figure. Each measurement was done
for 30 s, and a total of four measurements were made: prior to dosing,
immediately after dosing, 5 min. after dosing, and 10 min. after dosing.
Figures 6.5(b)–(e) show examples of the RSD images obtained prior to
dosing, immediately after dosing, 5 min. after dosing, and 10 min. after dosing,
respectively, at the observation area as shown in Fig. 6.5(a) under illumination of
LD with wavelength of 780 nm. Similar results were obtained under illumination
of LD with a wavelength of 830 nm. Figure 6.5(c) shows that a gray level around
the place of medication, which is indicated by the circle again, is clearly increased
in comparison with that in Fig. 6.5(b). On the other hand, Figs. 6.5(d) and (e)
show that the gray level around the place of medication is decreased in
comparison with that in Fig. 6.5(c). The above phenomena suggest that RSD
was increased with dosing and then decreased in a recovery process.
6.3.5 Simultaneous monitoring of blood flow and concentration

Simultaneous monitoring of blood flow and hemoglobin concentration is
important for evaluating the cerebral metabolic rate of oxygen (CMRO2).
CMRO2 is the rate of oxygen consumption by the brain, and is thought to be
a direct index of energy homeostasis and brain health.86 CMRO2 is a valuable

index of tissue viability and neuronal functions in the brain. Many conditions
are related to alterations in oxygen metabolism, such as Huntington’s
disease,87 Alzheimer’s disease,88 and normal aging.89 In addition, quantitative
measurement of CMRO2 is useful in understanding normal cerebral
physiology during the resting state, sleep, and brain activation.
CMRO2 has been described as the product of cerebral blood flow (CBF)
with oxygen extraction fraction (OEF) as90
CMRO2 ¼ CBF OEF, (6.4)
OEF is given by the fractional difference between the arterial oxygen
saturation and venous oxygen saturation, SA and SV, respectively.
SA SV
OEF ¼ : (6.5)
SA
Assuming that SA ¼ 1, Eq. (6.5) can be simply expressed as
HbRV
OEF , (6.6)
HbTV
The above expression assumes that oxygen extraction has occurred in the
capillaries and arterioles, and that there is no oxygen extraction in venules.
Considering the combination of Eqs. (6.4) and (6.5), relative change in
CMRO2 can be calculated as 91

DCMRO2 DCBF DHbR DHbT 1
1þ ¼ 1þ 1 þ gr 1 þ gt , (6.7)
CMRO2;0 CBF0 HbR0 HbT0
where the subscript 0 indicates baseline values. The parameters gr and gt
represent vascular weighting constants and are defined as
DHbRV DHbR
gr ¼ ∕
HbRV,0 HbR0
, (6.8)
DHbTV DHbT
gt ¼ ∕
HbTV,0 HbT0
: (6.9)
Since the optical measurements usually average the changes in hemoglobin

over the arteriole, capillary, and venule compartment, and do not give a direct
measure of the changes in hemoglobin for the venule compartment, two
parameters gr and gt are required to be assumed and have been tested for a
broad range (0.1–5)92 and (0.5–2)91 and for the more physiologically plausible
ranges of (0.75–1.25).92 The relative change in cerebral blood flow (1 þ DCBF/
CBF0) in Eq. (6.7) can be determined by calculating the changes in the speckle
322 Chapter 6
contrast in a series of laser speckle images. The relative change in the

concentrations of oxygenated and deoxygenated hemoglobin can be estimated
from the measured diffuse reflectance images based on the Monte Carlo
simulation-based modified Lambert-Beer law.91–93 CMRO2 obtained from the
simultaneous measurements of laser speckle contrast images and diffuse
reflectance images have been used to evaluate cerebral physiology during
ischemia,94–96 cortical spreading depression,96,97 and neuronal activity.91,92
By using speckle patterns at two wavelengths, not only blood flow but also
blood concentration change can be measured simultaneously by means of the
spectroscopic reflectance measurement.98–100 A diffuse reflectance R(l), which
is obtained on an area illuminated by the laser source with wavelength l, can
be converted to an absorbance OD(l) by the modified Beer-Lambert’s law, as
log RðlÞ ¼ ODðlÞ ¼ εoxy ðlÞC oxy l þ εdeoxy ðlÞ C deoxy l þ OD0 ðlÞ þ SðlÞ,
(6.10)
where ε(l) is the molar extinction coefficient, C and l are the molar
concentration of tissue and the mean path length of light rays inside tissue,
respectively. OD0(l) is attenuation due to the other minor pigments such as
melanin and bilirubin, and S(l) denotes scattering loss in tissue. A difference
in absorbance is defined between the two states before and after some
physiological change in tissue and can be expressed by
DOD ¼ εoxy DðC oxy lÞ þ εdeoxy DðC deoxy lÞ: (6.11)
Here we assume that changes in OD0(l) and S(l) are small between the two
states before and after the change in comparison with those in the
concentrations of HbO and Hb and, thus can be neglected. D(Cl) means a
difference of each concentration between the two states, but in the form of a
product with the mean path length l. By expressing the above equation in the
two wavelengths and solving the two equations by linear transform, we derive
two parameters D(Coxyl) and D(Cdeoxyl) for concentration changes in oxy- and
deoxy-hemoglobin. The concentration change of the total hemoglobin is then
expressed by Eq. 6.12:
DðC total lÞ ¼ DðC oxy lÞ þ DðC deoxy lÞ: (6.12)
Though these parameters include a factor of the mean path length l, they are
expected to give a useful measure for detecting a change in blood concentration.
6.3.6 Measurements for humans

Figure 6.6 shows a position of occlusion and measurement area on a finger.
Image data acquisition was started during the normal state and continued
during moderate occlusion and after release. Figure 6.7 demonstrates typical
images of hemoglobin concentration changes and blood flow (SBR)77
Occlusion Measurement area

Figure 6.6 Position of occlusion and measurement area on finger.
(SBR) 0 150 300 [-]
(ΔCl) -1.5 0 1.5 [mM*mm]

1 sec 90 sec 260 sec 280 sec 660 sec
(Normal) (Occlusion) (Occlusion) (Released) (Finished)
ΔCtotal l
ΔCdeoxy l
ΔCoxy l
SBR
Figure 6.7 Measured images of hemoglobin concentration changes and blood flow (SBR)
on human finger in occlusion6HHFRORUSODWHV
measured simultaneously on the finger. The values of (Ctotall) and (Cdeoxyl)

increase with occlusion and then gradually decrease after release. The value of
(Coxyl) decreases with occlusion and then slightly increases after release. The
blood flow clearly reduces with occlusion and rapidly increases back after
release. These four parameters demonstrate a kind of delay or shift
mechanism in their temporal changes. They seem to be quite useful for
analyzing blood circulation dynamics in tissue.
6.3.7 Experiments on rats

Figure 6.8(a) shows a picture of the anesthetized rat used for experiments. The
total measurement period was set to be 30 s: the first 10 s with no stimulation,
the next 10 s with stimulation by supplying an electrical pulse signal to the
pneumogastric nerve, and the final 10 s with no stimulation again. We imaged
blood flow and blood concentration change within the exposed pharyngeal
area as shown in Fig. 6.8(a), in which the carotid artery was located.
324 Chapter 6
Measurement area [arb. unit]

Observation area 255
0
(b) [arb. unit]
255
(a) 0
(c)
Figure 6.8 (a) Picture of the exposed pharyngeal area of an anesthetized rat, and (b) and
(c) blood flow images of an anesthetized rat obtained by RSD for the observation at t ¼ 3.9
and 18.0 s, respectively.
150
RSD [arb .unit]
100
50
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Time[s]
(a)
1
Output [arb. unit]
0.8
0.6
0.4
0.2
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Time[s]
(b)
Figure 6.9 (a) Temporal variation of the RSD value averaged over the area on the carotid
artery and (b) electrocardiogram in the exposed pharyngeal area of the anesthetized rat.
Figures 6.8(b) and (c) show examples of the RSD images at the
observation area as shown in Fig. 6.8(a) at t ¼ 3.9 and 18.0 s, respectively,
under illumination of LD with a wavelength of 830 nm. By comparing
Figs. 6.7 (b) and (c), the brightness on the carotid artery, which is vertically
located at the center of both images, is clearly changed. Figures 6.9(a) and (b)
[mmol.mm]
+1.6
Δ C total䝿l [mmol䝿mm]
0
0.5
0
–1.6 –0.5
(a)
[mmol.mm] –1
Artery
+1.6 –1.5
Tissue
–2
0 5 10 15 20 25 30
0 Time [s]
(c)
–1.6
(b)
Figure 6.10 (a) and (b) Examples of imaging of blood concentration change at t ¼ 10 and
20 s with median filtering in 5 5 pixels, respectively, and (c) blood concentration change
versus time relations with median filtering in 5 5 pixels.
show temporal variations of RSD value averaged over the measurement area
shown in Fig. 6.8(b) and the corresponding electrocardiogram, respectively.
Over 0–10 s and 20–30 s, the RSD value in Fig. 6.9(a) shows periodic
fluctuations. The electrocardiogram in Fig. 6.9(b) also shows periodic pulses
according to the heartbeat. The peaks of the RSD value and pulses in the
electrocardiogram are almost synchronized with each other. In contrast, the
RSD value during 10–20s in Fig. 6.9(a) significantly decreases and shows no
periodical fluctuation due to the reduction of blood flow speed according to
the stimulation.
Figures 6.10(a) and (b) demonstrate images of blood concentration
change at t ¼ 10 and 20 s, respectively, obtained by applying median filtering
in 5 5 pixels to the speckle patterns in advance. As seen from Fig. 6.10, the
blood concentration change on the carotid artery is significantly decreased in
(b). This is due to the reduction of the blood flow according to the stimulation.
Figure 6.10(c) shows temporal variations of the blood concentration change
on the artery and the surrounding tissue. Results in Fig. 6.10(c) indicate that
the temporal variation in blood concentration change on both the artery and
tissue seems to be reproduced in detail. This temporal variation is considered
to be mainly derived from the change in blood flow according to the
heartbeat.
6.4 Low-Coherence Speckle Interferometry

Figure 6.11(a) shows a schematic diagram of a low-coherence speckle
interferometer.101 A superluminescent diode (SLD) is used as a low-coherence
light source. Light from the SLD is divided into two equivalent beams.
326 Chapter 6
Moving object
Scattering medium 720μm

900μm
1054μm
Reference mirror
Pinhole
BS Collimated beam
SLD 10mm
L0 L1 ND
L2 filter Stage
Pinhole Controller
L3
Diffusive plate B
PC Intralipid solution 2mm
Diffusive plate A
CCD
(a) (b)
Figure 6.11 (a) Illustration of a low-coherence speckle interferometer for detecting speckle
images and (b) sample model used for distinguishing between objects moving at different
velocities in the same plane under the depth-resolved manner.
One beam illuminates a reference mirror and the other illuminates a moving
object behind a scattering medium. The light fields from the reference mirror
and the object are superimposed again, then reach the CCD camera by which
a speckle pattern is observed. The speckle contrast is enhanced only when the
optical pathlength difference between two arms of the interferometer is within
a coherence length of the SLD. This allows us to make a depth-resolved
enhancement of speckle patterns by scanning the reference mirror along the
optical axis of the interferometer. Figure 6.11(b) illustrates a sample model
used for depth-resolved motion imaging of two moving objects in layers
behind a scattering medium. A diffusive plate A has a 2-mm-wide painted bar
of 1% Intralipid solution that was dried on a glass plate and was vertically
oscillated. A diffusive plate B was composed of two horizontally separated
aluminum diffusive plates moving with different velocities to each other. Both
plates were illuminated by a beam via 4% Intralipid solution.
Figure 6.12(a) shows a speckle difference pattern obtained for the
diffusive plate A, and (b) and (c) show patterns obtained for the diffusive plate
B with the upper-plate velocities of 33 and 100 mm/s, respectively. When the
interference position is adjusted to plate A, the speckle difference pattern is
found only along the vertical bar on plate A, as seen in Fig. 6.12(a), without
the effects of both the Intralipid solution and plate B. When plate B is selected
by the interferometer, it is imaged with almost no effect for both the Intralipid
solution and plate A, as seen in Figs. 6.12(b) and (c). Moreover, as seen in the
upper plate in Figs. 6.12(b) and (c), the magnitude of the speckle difference is
increased as the upper-plate velocity becomes large. Thus, the present method
is useful not only for imaging an object motion in a depth-resolved manner
but also for discriminating between objects moving at different velocities in
the same plane.
[arb. unit] [arb. unit] [arb. unit]

255 255 255
0 0 0
(a) (b) (c)
Figure 6.12 Speckle difference patterns obtained for (a) diffusive plate A and (b) and
(c) upper diffusive plate B with velocities of 33 and 100 mm/s, respectively, while the lower
plate has a velocity of 11 mm/s.
6.5 Quantitative Characterization of Blood Flow Rate

6.5.1 The use of laser Doppler anemometry for measurements
of absolute blood flow velocity
Blood flow velocity measurements with laser Doppler anemometry
(LDA)21–23,36–38 and Doppler OCT24,39–41 are based on detection of DFS of
light scattered on a moving red blood cell. The DFS is directly proportional to
a projection of cell velocity on the scattering vector, defined as the vector
difference of incident and scattered wave vectors.29,37,44 The absolute flow
velocity can be measured when DFS is detectable and the angle between the
light scattering vector and flow velocity are known.
Since the angle could not be found directly during intravital measure-
ments, the absolute flow velocity can be measured by detection of the DFSs of
the radiation scattered simultaneously in two or three directions. The method,
referred to as bi-directional LDA,37 is used for measurements of blood flow
velocity in retinal arterioles and venules.21–23,37
Doppler optical coherent tomography (DOCT) provides efficient filtering
of multiple scattered light for visualization of tissue layers and spatially
resolved detection of blood flow at depths up to 1 mm. Similar to
ophthalmological LDA, the scattering vector of DOCT is nearly perpendicu-
lar to the tissue surface and thus to the axis of superficial blood vessels.
Therefore, the result of measurements are strongly dependent on the angle of
incidence, and thus the DOCT is capable of detecting only relative flow
velocity changes when the exact angle cannot be determined.
Because the complex optical arrangement of a conventional DOCT
makes simultaneous detection of DFS of light scattered in two or more
directions102 very difficult, the approaches based on the scattering angle
assessment by means of the scanning of two adjacent vessel cross sections103
or integration of a blood flow rate104 were proposed. Simultaneous
measurements of three velocity vector components has become available
recently with the progress of high-speed Fourier DOCT at the cost of
increased complexity of the setup.24,41
328 Chapter 6
Another approach of LDA is based on the illumination of a probe volume

with two intersecting laser beams.29,44 The intensity of light scattered in any
direction by a particle intersecting the probe volume is modulated with the
frequency proportional to a scalar product of the particle velocity and LDA
sensitivity vector. The latter factor is defined as a vector difference of incident
laser beam wave vectors, and it is perpendicular to the optical axis of LDA.
The direction of the sensitivity vector minimizes the effect of angular
uncertainty when the blood vessel is nearly perpendicular to the LDA axis.
The additional advantage of the setup is that the modulation frequency is not
dependent on the refractive index of a medium surrounding the particle.44
The method referred to as differential LDA has been widely employed in
experimental fluid dynamics.44 But until now, it was not used for quantitative
measurements of blood microcirculation except for narrow applications in
biological microscopy.105
As with ophthalmological bi-directional LDA, the key problem here is
related to the multiple light scattering preventing unambiguous detection of
the DFS. The principal difference between bi-directional and differential
LDA schemes is the independence of the latter on the angle at which scattered
light is collected, while for the former, both the angle of incidence and
that of scattering must be exactly defined for quantitative flow velocity
measurements.
The effect of multiple scattering of light on optical imaging can be
described as a distribution of light energy between two fractions29: “ballistic”
that retains its initial direction of propagation and phase, and “diffuse”
carrying no useful information. Reaching a light detector, the diffuse fraction
of light produces on it a halo that degrades the contrast of an image produced
with the ballistic fraction.29 Because of light scattering, the ballistic fraction
degrades exponentially during propagation along its straight path.
The model shows that an image can be detected when the ballistic fraction
carries reasonable energy, e.g., one photon per image point per second, and
certain measures have been taken to suppress the diffuse light and to enable
detection of this minute ballistic fraction. For example, OCT employs
selective registration of the ballistic fraction of light with gating based on
temporal coherence, and the minimal detectable energy determines the
fundamental limitation of the probing depth of the method.
Like optical imaging, the quantitative laser Doppler anemometry is based
on rectilinear propagation of light between the object and detector,29 thus it
requires selection of the ballistic fraction of scattering light with coherent
gating or with confocal detection of the scattered light. The former approach
is known as DOCT39–41 and can be described as bi-directional LDA with
coherent gating of scattered light.
As was mentioned above, differential LDA is capable of measuring a flow
velocity perpendicular to its optical axis; therefore, it is less sensitive to the
Figure 6.13 LDA for blood flow velocity measurements.45
incident angle. It provides better localization of the probe volume as an

intersection of laser beams and is insensitive to the optical path of light
scattered by the moving object. But because of its operation principle, it could
not be employed in DOCT and requires another approach for the selection of
ballistic light. Below we discuss the recently proposed method for suppression
of the multiple light scattering with the differential LDA used for blood flow
velocity measurements.38,45
A differential LDA setup is schematically shown in Fig. 6.13. Collimated
laser radiation emitted a red cw diode laser module with a 15-mW output at
650 nm (ML-09, Skat-R, Russia) is divided into a pair of parallel beams by a
beamsplitter prism.
Beams are focused into a flow with an objective lens with a 100-mm focal
length and 30-mm clear aperture. Intersection of the laser beams with an angle
of 14 deg in air form the probe volume of the setup. Because of interference,
the probe volume is filled with an interference pattern. The maxima of the
pattern corresponds to a set of planes perpendicular to the X direction in
Fig. 6.1329,44 separated with a 2.7-mm distance. The probe volume has a
nearly circular cross-section of 60-mm maximal diameter in the XY plane and
110-mm length along the Z axis (Fig. 6.13).38,45
When a scattering object passes through the probe volume, it traverses the
interference pattern and scatters modulated light radiation that is detected
with a photodetector placed behind the field stop optically conjugated with
the probe volume by means of lenses 1 and 2. The signal of photodetector is
amplified, digitized with an AD converter at 44.1 kS/s with a 16-bit resolution,
and then processed with a PC using homemade software.
In terms of the Doppler effect, a particle that moves across the probe
volume scatters toward the detector light radiation of the first beam and the
second beam with shifted frequencies (Fig. 6.14):38
v1 ¼ v0 ðks ki1 Þu ¼ v0 K1 u, (6.13)
v2 ¼ v0 ðks ki2 Þu ¼ v0 K2 u, (6.14)
where ki1ki2 are the wave vectors of Beam 1 and Beam 2, respectively, and ks,
defines the scattering direction towards the detector, u is a particle velocity,
330 Chapter 6
Figure 6.14 Scattering geometry of differential LDA.38
and v0 is the frequency of the incident beam. Superposition of both waves at

the detector produces oscillating light intensity38,29:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I d ðtÞ ¼ I 1 ðtÞ þ I 2 ðtÞ þ 2 I 1 ðtÞI 2 ðtÞ cosðvd t þ dÞ, (6.15)
where I1(t), I2(t) are intensities of each wave, d is a phase constant determined
by the setup geometry, and
vd ¼ v1 v2 ¼ ðki1 ki2 Þu ¼ Ku: (6.16)
K denotes here a sensitivity vector of the LDA that is perpendicular to its
optical axis (Fig. 6.14). The third term of Eq. (6.15) corresponds to oscillations
of the light intensity with frequency vd proportional to the velocity of a
scattering particle (Eq. 6.16). In most LDA applications at low concentrations
of flow tracing particles, the frequency can be determined with spectral
analysis of the photodetector signal as a center of a clearly detectable spectral
component.29,44
In highly scattering media, the effect of scattered light intensity
fluctuations makes detection of vd difficult because of the high impact of
the first and second terms of Eq. (6.15) into the power spectrum. Figure 6.15
shows the power spectra of LDA signals corresponding to a centerline flow of
a fluid with an extinction of 1.5 mm–1 (kaolin suspension in water) through a
glass channel with a 1- 2-mm rectangular cross-section. Line 1 represents
Figure 6.15 Power spectra of LDA38: 1 – P12(f ); 2 – P1(f ); 3 – P2(f ); 4 –|P 012(f )|
the spectrum P12(f ) of a signal obtained when the probe volume is illuminated
with both beams of a differential LDA as shown in Fig. 6.13. The black box
denotes fd ¼ vd/2p. Lines 2 and 3 are spectra P1(f ) and P2(f ) obtained when
only beam 1 or 2 was used. These are the spectra of the first and the second
terms of Eq. 6.15. that obviously form a low-frequency pedestal for P12(f ).
Because each term of Eq. 6.15 additively impacts each spectral component
of P12(f ), it was recently proposed to minimize the low-frequency pedestal by
calculating the difference spectrum:38,45
P012 ðf Þ ¼ P12 ðf Þ P1 ðf Þ P2 ðf Þ: (6.17)
The modulus of the resulting spectrum |P0 12(f )| is shown in Fig. 6.15 with line
4. The frequency component centered around fd is clearly visible, and the flow
velocity can be measured. Broadening of the spectral component is
determined with a nonuniform velocity distribution over the probe volume
and the size of the probe volume, as has been discussed in the extensive
literature on LDA metrology.44 We should note here the impact of multiple
light scattering into the broadening of the spectrum that must be also taken
into consideration.38
Technical implementation of the proposed approach is shown in Figs. 6.13
and 6.16. It includes a simple mechanical modulator rotating with a motor.
A modulator blade blocks light at 1/3 of the circumference, and three
commutations can be performed during a full turn of the modulator: 1) both
laser beams are open; 2) beam 1 is closed; 3) beam 2 is closed. The light
Figure 6.16 The optical modulator scheme

332 Chapter 6
Figure 6.17 Superposition of raw photodetector signal (Channel 1) and the blade position
encoder data (Channel 2): 1 is the opening of both beams; 2 is the closing of beam 1; 3 is the
closing of beam 2.45
emitting diode and reference photodetector are used to encode the modulator
position. Figure 6.17 shows superposition of the raw detector signal and the
blade position encoder. The signal processing procedure includes calculation
of three periodograms per cycle of the modulator. Periodograms were
calculated with FFT using a Hanning data window over 521 samples.106 Then
the estimates of P12(f ), P1(f ), and P2(f ) were calculated over a set of
corresponding periodograms and finally, P012(f ) was calculated using Eq. (6.17).
Figure 6.18 shows spectra of the LDA signal registered at the centerline
flow of whole rat blood in a 250-mm-thick rectangular channel at a depth of
125 mm from the glass window surface. A blood sample of 2 mL was taken
Figure 6.18 Power spectra of LDA signal for whole blood flow38 at centerline of 250 mm
rectangular channel: 1 is P12(f ); 2 is |P012(f )|; 3 is positive values of P012(f ); 4 – P012(f ) 10
immediately prior to the experiment from the anesthetized rat’s aorta. Red
blood cells’ concentration in the sample was 7.5 106 ml–1, which is normal
for the animal.107 A syringe pump was used to drive blood flow through the
channel.
Spectra were calculated over 170 periodogams of data sequence acquired
during 13 seconds. Because of the multiple scattering of light in a blood power
spectrum, P12(f ) [Fig. 6.18(1)] decreases monotonically over the amplifier
bandwidth of 0.5to 20 kHz. Modulus of the difference spectrum |P0 12(f )| and
especially the positive values of P0 12(f ) [Fig. 6.18(2) and (3)] have a maximum
centered near 3 kHz, that corresponds to an average blood flow velocity of
4 mm/s in the channel.
Line 4 on Fig. 6.18 shows positive values of P0 12(f ) magnified ten times with
respect to line 3. Because the probe volume length is comparable to a channel
thickness spectrum, P0 12(f ) is broad. It contains numerous DFS components
corresponding to all velocities of RBCs over the LDA probe volume.
The frequency of the optical modulator rotation was 13 Hz and it was
enough for measurement in a flow with a relatively constant velocity. In
fact, the flow velocity in biological tissues varies with time. The velocity of
blood flow in arteries varies periodically in accordance with the cycles of the
heart contractions. Therefore, the frequency of the Doppler shift can change
over time. For example, a heartbeat rate of a rat is 300–600 per minute, so
the period (cycle) of the heart is 0.2–0.1 sec.107 To resolve blood flow
velocity changes in different phases of the cardiac cycle (systolic and
diastolic), it is necessary to have at least 10 measurements at regular
intervals for one cycle.
6.5.2 Intravital particle image velocimetry of capillary blood flow

Automated tracking of particles seeded in a fluid flow by means of digital
image processing, referred to as microparticle image velocimetry (mPIV), is
widely used in fluidics for measurements of liquid and gas flow velocities in
the framework of various applications.108,109 The mPIV technique has been
applied for studies of blood or blood phantoms’ circulation in aortas and their
models,110,111 blood rheology,112,113 and in vivo monitoring of blood
circulation in small blood vessels of animals and humans.114
The main difficulty of capillary blood flow imaging is related to
involuntary movements of the living object that results in random displace-
ments of the capillary image across the microscope field of view.48 These
displacements superpose over the regular movement of blood cells, introducing
significant error in the PIV measurement of the capillary blood flow velocity.
To minimize the effect of involuntary movement, either software or a hardware
image stabilization technique can be used. When the exposure time is short
enough to ensure smear-free images of moving red blood cells, the software
image stabilization is enough to exclude the effect of involuntary movement of
334 Chapter 6
Figure 6.19 Experimental arrangement for imaging of blood capillaries.48
an object within the object plane. Displacements along the microscope optical
axis cause defocusing of the image, thus they can be handled only with a
hardware autofocus arrangement.
The effect of involuntary movement on mPIV measurements was
demonstrated on a capillary loop of a human nailfold.48 Figure 6.19 shows
the experimental setup for in vivo imaging and mPIV measurements of
microcapillary blood flow.48 It consists of a digital microscope, an adjustable
optical table for alignment of the object under study, and a light source. The
object is illuminated with a 3W white light emitting diode with a condenser
incorporated into the light source module. The image of the capillary is
captured by a microscope built of an objective (magnification 10 and
numerical aperture NA ¼ 0.24) and digital monochrome CMOS camera
DCC1545M (Thorlabs, Germany) with an adjustable microscope tube
adaptor. The microscope can be focused on the object with a precise linear
translation stage (sensitivity 1 mm).
The arrangement was used for in vivo imaging of nailfold capillaries of a
human hand. The volunteer’s right forearm in a relaxed position was located
on a semi-rigid armrest at chest height. The right hand was placed palm down
on the adjustable table of the microscope. The ring finger was fixed in a
lodgment of modeling clay with all the necessary hygiene certificates. The
lodgment, attached to the microscope stage, was tightly fitted to the finger
shape, but was not squeezing it. Such a fixation has no significant effect on
capillary blood flow. It does not totally exclude the involuntary movement of
the finger as a result of breathing, palpitations, tremors, etc., but was very
useful to avoid relatively large translational movements of the finger.
Positioning of capillary vessels within the microscope field of view was
performed by translation of a stage with lodgment in a horizontal plane
perpendicular to the optical axis of the microscope with micrometer screws.
To reduce the light scattering by the epidermis of the skin layer, glycerol as an
Figure 6.20 Image of human nailfold capillary loop and stabilized window position.48
Figure 6.21 Blood flow velocity field calculated for stabilized (a) and nonstabilized
(b) series of capillary loop images.48
immersion agent was applied.48 The recorded image of a capillary loop is

shown in Fig. 6.20. Stabilization of an interactively selected region (white
rectangle) of the image throughout a series of 100 frames was performed using
the cross correlation feature detection method. Then the stabilized sequence
was analyzed using homemade mPIV software.48 The effect of image
stabilization is clearly seen in Fig. 6.21.
336 Chapter 6
Figure 6.22 Rat stomach microvasculature image (a) and the same image superposed
with flow velocity color map (b)6HHFRORUSODWHV
The same technique can be applied for precise measurements of blood

flow velocity in superficial capillaries of internal organs of laboratory animals.
Figure 6.22(a) shows an image of blood capillaries of the outer surface of rat
stomach. Although the stomach of the anesthetized animal was surgically
removed from the abdomen and placed on the object stage, image stabiliza-
tion was needed to track RBC movement. Figure 6.22(b) shows calculated
flow velocity values coded with color points superposed on the image of blood
vessels.47
Optical methods are the only way for non-contact and non-invasive
detection and characterization of blood flow. But quantitative measure-
ments of blood flow velocity and volumetric flow rate are still challenging
because of the high scattering of light in tissue. Probing depth of both laser
Doppler and imaging techniques based on rectilinear propagation of light
can be estimated in terms of exponential attenuation of ballistic fraction of
light along its path in tissue, and it does not exceed a few millimeters.29
Typically, the depth is much shorter than the limit because of the effect of
scattered light that introduces ambiguity in either the object position or
velocity detection. The limitation addresses optical velocimetry of
superficial blood vessels that are parallel to the tissue surface and thus,
are examined with optical instruments in a direction perpendicular to blood
flow velocity. Therefore, the optical imaging techniques sensitive to
transversal displacement of the object with respect to the optical axis are
more useful for blood flow velocity measurement than an arrangement
based on a Michelson interferometer like Doppler OCT. The latter
technique ensures efficient suppression of multiply scattered light and
provides near to maximal possible probing depth and is capable for
measurements of the velocity component parallel to the instrument axis.
Thus it is not suitable for unambiguous absolute flow velocity in measure-
ments without its modification41 for bi- or three-directional detection. In
contrast to this, a differential laser Doppler anemometer that is equivalent
to an imaging system29 can be used for absolute flow velocity even in its
simplest implenetation.38,45
6.6 Intravital Microscopy (IM) for Monitoring Blood and Lymph

Flows
In principle, there is very little difference between conventional micros-
copy and IM. In practice, however, there are tremendous differ-
ences.6–12,16,116,129–130,141–175 First, the setup for IM is optimized for
in vivo work and the handling of live animals or even humans rather than
fixed tissue or cell cultures. The next major difference is the way image
acquisition occurs. In most cases, IM systems monitor physiological
functions in real time as opposed to the frizzed tissue sample in
conventional microscopy. This requires the use of fast, sensitive detectors
such as CCD cameras, intensified charge-coupled device (ICCD)
cameras, CMOS cameras, electronic multiplying charge-coupled devices
(EMCCDs), or photomultipliers (PMTs). The optical design usually, but
not necessarily, needs to be optimized for the use of long working distance
lenses since the target samples are thick. The available options in the use of
the IM technique include methods that vary according to their technical
complexity from conventional microscopy that utilizes transillumination,
reflected light, or fluorescence, up to modern multiphoton micros-
copes (MPMs), allowing acquisition of high-quality images up to a depth
of 0.7 mm in tissue,157,163–167 and high-resolution photothermal
microscopy170–173 with nanoscale resolution at depths down to a few
millimeters.
338 Chapter 6
6.7 Intravital Transmission Digital Microscopy (ITDM)

One of the most widely used techniques is ITDM using absorption and
scattering phenomena with forward (transillumination) or backward
(reflected light) schemes. An example of a conventional setup for ITDM is
shown in Figure 6.23.146 ITDM based on an upright or inverted microscope
provides the following functions6,7,9–12,129,130,143,144,147–149:
1. real-time monitoring of relatively large structures such as lymph and
blood vessels with relatively low magnification (4 to 10 );
2. quantitative dynamic evaluation of blood and lymph vessel diameter,
parameters of lymphatic phasic contractions and valve activity;
3. measuring of cell velocity in flow by video recording cell movement [so-
called PIV];
4. single-cell analysis of moving cells at high magnification (40 , 60 , and
100 , water immersion) including time-resolved changes of cell shape
(e.g., deformability), real-time tracking of circulating cells, and cell to cell
interactions (e.g., aggregation and adhesion).
In our in vivo [so-called intravital (IV)] studies, we used Olympus IX81
inverted and Olympus BX51 upright microscopes that were equipped with
various digital video cameras. The set of technical parameters (e.g., type of
camera; frame per second [fps]; length of video recording, magnification of an
objective) are chosen depending on the biomedical tasks of the particular
experiments. Speeds of up to 25 fps and an exposure time of 0.04 s are quite
well suited for the imaging of relatively slow-moving individual cells, such as
rolling WBCs (30–70 mm/s) and pre-stopping or stopping RBCs and platelets
(e.g., during initial stage of clot formation) (Fig. 6.24).144,168
However, to image cells in high-speed blood and lymph flow (e.g., several
millimeters per second) without optical distortion, high-speed and highly
sensitive cameras should be used. In particular, the imaging of moving cells in
blood vessels with flow velocities of 5–10 mm/s requires imaging speeds in the
range of 5,000–10,000 fps [Fig. 6.25(a)]. This can be, for example, achieved
Figure 6.23 Schematic of intravital microscopy.10

Figure 6.24 Imaging blood flow at 25 fps. (a, b) Rolling WBCs in the small mesenteric
veins of healthy rats at low (10 ) magnification (a) and at high (100 , water immersion)
magnification (b). (c) Platelets in slow blood flow (100 , water immersion). (d) slow moving
RBCs and WBCs (100 , water immersion).
Figure 6.25 High-speed high resolution ITDM. (a) Three subsequent images of RBCs in
small artery at 10,000 fps; blood flow velocity up to 2.5.mm/s. (b) 1D image of a blood cell
using 512 1 pixels at 40,000 fps.
with a highly sensitive CCD camera (Cascade:512; Photometrics/Roper

Scientific, Inc., Tucson, AZ, USA) at speeds of up to 500 fps and with a
CMOS camera (model MV-D1024-160-CL8; Photonfocus, Lachen, Switzer-
land), with speeds of 10,000 fps for an area of 128 128 pixels. Cells with
higher velocity can be imaged at 40,000 fps with a linear pixel array [e.g.,
40,000 fps for 512 1 pixels, Fig. 6.25(b). In this case, further reconstruction
of 2-D cell images was required, Fig. 6.25(b), bottom].10,12
Thus, integration of high-resolution and high-speed monitoring improves
PIV in the dynamic range and enhances spatial resolution and measurement
accuracy.
6.8 Intravital Fluorescent Digital Microscopy (IFDM)

Another modification of IM is IV fluorescent digital microscopy
(IFDM).10,16,65,129,130,144–146,151,153–162,168,169,174–183 The applications of
IFDM for blood and lymphatic flow research include, but are not limited to:
1. Mapping blood (angiography) and lymph (lymphography) vessels using
fluorescent dyes and quantum dots (QDs);
340 Chapter 6
2. Detection and counting circulating cells in blood and lymph flows, and
measurement of cell to cell interactions in vivo by labeling cells of interest
with different fluorescent markers. Marked cells can be of animal or
human origin, or infectious pathogens including circulating tumor cells
(CTCs), their aggressive subpopulations such as tumor-initiating cancer
stem CTCs, circulating apoptotic cells, and circulating sepsis-induced
bacteria (e.g., staphylococcus aureus);
3. Exploring physiological processes at the cellular level in living animals,
using special fluorescent indicators sensitive to changes in the microenvi-
ronment, such as pH, ion fluxes, membrane electrical potential, redox
levels, protease activity, and more;
4. Studying the anatomical structure and function of blood or lymphatic
vessels (e.g., angiogenesis);
5. Monitoring in vivo blood and lymphatic responses to therapy, including
pharmacological drugs and therapeutic laser irradiation.
Finally, obtained images and video records can be acquired, combined (if
necessary), and processed in Adobe Photoshop 7.0.1 software (Adobe Systems,
San Jose, CA, USA), Adobe Premiere Pro CS6 software and Image J 1.46 for
Windows.
Despite the promising capability of IV microscopy, the high-resolution
imaging of individual flowing cells, especially in label-free mode, can be
obtained in relatively transparent animal models (e.g., the mesentery of the
frog, mouse, and rat.9,10,144,149,184–190 High-resolution imaging of blood and
lymph flow in other widely used animal models (e.g., mouse ear, skin-fold,
tumors at different locations) is limited because of significant light scattering
from surrounding tissue (e.g., skin) and/or the relatively deep location of
vessels from the skin (or other tissue).
6.9 Optical Clearing

This problem can be, at least, partly solved by a clearing technique, whereby a
chemical agent is topically applied into the skin to dramatically reduce light
scattering and thus improve image quality.130,191–193 With this approach that
is pioneered by Prof. V. Tuchin, we obtained high-resolution images of skin
vessels in rats and mice with the use of a dorsal skin-flap window.144,194 In the
mouse-ear model, high-resolution IV microscopy with optical clearing
provides the conditions for monitoring the shape of individual WBCs
[Fig. 6.26(a), arrows] and even RBCs [Fig. 6.26(b)].
6.10 In vivo Flow Cytometry

In the last decade, a significant leap forward in noninvasive single cell analysis
of blood and lymph flow in vessels was made because of the development of
in vivo flow cytometry (FC). In vivo FC was pioneered by Profs. Zharov
Figure 6.26 Imaging rolling WBCs in small vein (10 ) and RBCs in capillaries (100 ) of
mouse ear using integration of ITDM and optical clearing with glycerol.144
(photothermal and photoacoustic modes)141 and Lin (fluorescent mode)153 in

2004.
The technical platform of in vivo FC is based on principles of conventional
FC in vitro.10,16,195 In conventional flow cytometry, cells are introduced into a
high speed (up to few m/s) laminar artificial flow.195 After focusing the cells
into a single file, laser-induced fluorescence from single fast moving cells, and/
or forward and sideways scattered lights emitted from the cells are detected
using photodetector arrays with spectral filters. This highly accurate
technology provides fast (a few million cells in a minute), multiparameter
quantification of the biological properties of individual cells at subcellular and
molecular levels, including their functional states, morphology, composition,
proliferation, and protein expression. Nevertheless, the crucial limitation of
in vitro FC is low sensitivity for detection of rare CTCs, bacteria, sickle cells,
and clots due to a small blood sample volume.16,157,169 Other limitations
include the discontinuity of sampling with limited, discrete time points and
that invasive extraction of cells from a living system may alter the cell
properties (e.g., clot formation, morphology, and marker expression) and
prevent the long-term study of cells (e.g., cell to cell interactions, aggregation,
or rolling) in their natural biological environment.
In vivo FC overcomes these limitations: compared to in vitro FC, in vivo
FC has significantly increased sensitivity due to monitoring of almost an
entire blood volume; it can be performed noninvasively, in a natural cell
environment; and it is able to monitor single-cell behavior for a long time over
disease development. To date, our and others’ numerous preclinical studies
have successfully shown that in vivo FC of cells in blood and lymph flows
(e.g., rare CTCs at the earliest stage of metastatic disease; see examples below)
have an unprecedented sensitivity and a high specificity, compared to ex vivo/
in vitro CTC assays. To detect cells, a wide range of contrast agents have been
used, including intrinsic chromophores (e.g., melanin; hemoglobin), geneti-
cally encoded fluorescent proteins (e.g., GFP), and bioconjugated fluorescent
dyes, quantum dots or nanoparticles.10,11,16,17,64,65,129,141–143,153–161,170,174,175
In this chapter, we focus on fluorescent flow cytometry (FFC). Specifically,
fluorescence detection schematics have been used in different modifications
342 Chapter 6
(e.g., confocal or two-photon) employing standard fluorescent labels as in

conventional fluorescent microscopy in vitro. In the confocal scheme,
fluorescent signals from the cell populations of interest are recorded as the
cells passed through a slit of laser (e.g., He-Ne) light focused across 20- to
50-mm mouse ear blood vessels.153,154 Emitted fluorescence is collected by
microscope objectives and directed through a dichroic splitter and mirrors to
photomultiplier tubes. Compared to single-photon fluorescence microscopy,
multiphoton fluorescent technique can increase the depth of light penetration in
microvessels located deeper in tissue (few hundred lm) and reduce out-of-focus
photodamage.143 Recent advances have significantly extended the capability of
FFC by development of two color schematics, linear configuration of laser
beams completely covering a vessel’s cross-section, fiber-delivered laser light to
deep vessels and advanced integration with other in vivo FC approaches such as
photoacoustic FC.143,158,169,174,175
Our two-color FFC module (Fig. 6.27) as a part of an integrated FC is
based on a customized Nikon Eclipse E400 microscope platform (Nikon
Instruments Inc., Melville, New York, USA) that is converted into an inverted
system.161,169,174
The setup is equipped with continuous wave (CW) laser(s) for fluorescence
excitation. For example, we use CW diode 488-nm laser (IQ1C45 (488-60)
G26, Power Tech., Alexander, Arkansas, USA) with 2 mW in the sample for
fluorescence excitation and, thus detection of CTCs genetically encoded with
GFP. Laser beams are focused into the mouse vessel by a 40 microobjective
(Plan Fluor, NA 0.75; Nikon Instruments, Inc.). The dichroic mirrors and
microobjective are used to collect fluorescence from cells. Additionally, the
emission bandpass filter with a spectral band centered at 520 nm and a
Figure 6.27 (a) Schematic of two-color in vivo FFC. (b) Signal traces in green and red
channels; each signal is associated with single fluorescently labeled cells travelling in blood
(or lymph) flow. (c) laser beam focused on a lymph vessel of mouse ear; EB dye was used
for mapping lymph vessel (see details below).
bandwidth of 15 nm is used (Semrock, Inc., Rochester, New York, USA). A

variable-width slit in the front of the photomultiplier tube ([PMT] R928,
Hamamatsu, Co., Bridgewater, New Jersey, USA) is used to control the axial
resolution to provide detection of cells of interest in a whole vessel, while
efficiently filtering out-of-plane autofluorescence of the tissues. Cylindrical
lenses (f ¼ 250 mm) provide linear configuration (e.g., 10 80 mm for ear
vessels) for laser beams covering the whole vessel diameter, which allowed
detection of all CTCs passing through the vessel cross section.
In general, an FFC system is controlled using a workstation (Precision
690, Dell Inc., Round Rock, Texas, USA) and custom software (LabView,
8.5, National Instruments, Austin, Texas, USA). Specifically, the signals after
PMT are continuously sampled at 4 MHz by a high-speed digitizer (PCI-5124,
National Instruments) and downsampled to 10 kHz by averaging 400 points.
Fluorescence signals are then combined into traces (Fig. 6.27, right), displayed
in real-time, and saved for later processing. Traces are analyzed to identify
peaks over the selected threshold and measurement location, amplitude, and
width of each peak.
6.11 In vivo Lymph Flow Cytometry (LFC)

In vivo lymph flow cytometry (LFC) was invented by our team based on
principles of 1) in vitro flow cytometry and 2) natural cell-focusing phenomena
in lymph vessels.129,169 The complex cell motion (e.g., fluctuations in the
position and velocity of individual cells10,11,129,144) in unstable, turbulent,
oscillating lymph flow makes it difficult to detect individual cells compared to
FC in vitro in which sheath fluids act as an artificial nozzle to provide a
hydrodynamic cell focusing nozzle and well organized stable single file
flow.195 Nevertheless, we discovered that nature has also created a single file
(or close to it) cell flow in a localized zone near a lymphatic valve.129 For
example, in vessels with diameters of 136 ± 10 mm, the valve leaflets formed
natural nozzles, measuring approximately 44 ± 7 mm in diameter, that
provided an approximately threefold constriction of flow with a significant
increase in flow velocity. As a result, cell acceleration, together with sheering
forces, led to hydrodynamic focusing of cells into a single file cell flow, with
radial cell fluctuations of just 5 to 10 mm [Fig. 6.28(a) and (b)]. The phasic
contractility can also provide auxiliary cell focusing [Fig. 6.28(c)]; however,
because of the relatively low degree of constriction (20 to 30%), single file cell
flow was only observed in small diameter vessels (,80 mm).129
6.12 Animal Models

In general, the best targets for ITDM are animals with relatively transparent
structures, such as zebrafish (Brachydanio rerio), and the vascular nets of a
hamster cheek pouch; however, these models are not ideal for studying lymph
344 Chapter 6
Figure 6.28 Principle of lymph FC in vivo. (a) General schematic of natural cell focusing
with lymphatic valve. (b) ITDM imaging of valve-induced cell focusing in real lymph flow
in vivo at 10 (top) and 100 (bottom) magnifications. (c) Imaging of cell focusing (bottom)
induced by natural lymphatic contractility: top image is non-focused cell flow in the relaxed
lymph vessel; bottom image is the same vessel during constriction with the focused cell flow.
or blood vessels.144,196–198 The hamster cheek pouch model, for example, is

good for visualizing blood vessels, but its lymphatics are not well-developed.
To date, the best optical images of both lymph and blood vessels have been
obtained in the mesentery of small, cold-blooded animals (e.g., frog) and
mammals (e.g., mouse, cat, rabbit, guinea pig, rat). Of the mammalian
models, the mouse and rat are excellent models in terms of size, physiology,
and pharmacokinetics for broadening medical applications, including single-
cell diagnostics. Additionally, many mice models and some rat models are
able to mimic certain human diseases (e.g., metastatic cancer). The unique
advantages of rat mesentery are its very thin (8 to 17 mm) and transparent
duplex connective tissue with its single layer of blood and well-developed
lymph vessels (Fig. 6.29 top).9,10,144,149,184–190 Specifically, a light is slightly
attenuated in the mesentery, mainly in the relatively thin vessel wall, without
any influence from other tissues, as occurs in other models. The mesentery
refractive index of 1.38 (for a rat) in the typical spectral range of 400 to
Figure 6.29 Animal models of rat mesentery and nude mouse ear6HHFRORUSODWHV
700 nm is lower than the refractive index for skin (1.40–1.42 for rats) and,
especially for epidermis (1.55 for humans); thus, it is close to the refractive
index of water (1.33). As a result, these optical and geometric features
significantly reduce unwanted scattered light, allowing the use of a microscope
objective with a high numerical aperture (up to 1.4) and high magnification
(60 –100 ). An additional advantage of the rat mesenteric model is the
good penetration of reagents into the mesenteric tissue. Thus, the responses of
the microvascular network to the impact of different environmental factors
can be studied relatively easily with the use of a simple topical application.
In general, on the basis of our experience and the experiences of other
groups, this easy-to-access mesenteric model is a very promising model for
real-time monitoring of individual static, migrating, and circulating cells (e.g.,
WBCs, RBCs, CTCs, and many others) with the highest optical resolution.
Such a model is essential for studying cell traffic in the blood and the lymph
systems under normal and pathologic (lymphedema, metastasis, and many
other lymph-related diseases) conditions. The capability of this model for
blood and lymphatic research is demonstrated in detail below. The limitation
of the mesentery model is invasiveness.
To overcome this problem, most FFC studies have been performed
noninvasively on the thin ear (±250 mm) of the nude mice that has well-
distinguished blood vessels (e.g., 30 to 70 mm in diameter) and lymph vessels
(Fig. 6.29 bottom).129,153–161,169,174,175
We have also used numerous animal models to study metastatic tumors,
including melanoma and breast cancer.16,17,64,65,129,130,161,169,175 To accurately
reflect the clinical problems of metastatic progression, it is necessary to
employ an animal model in which the disease is similar to that in humans.
Clearly, to date, there is no model that faithfully reproduces all aspects of
human cancer. Among others, a well-established model of choice is an
orthotopic xenograft metastatic mouse model that has already made major
346 Chapter 6
contributions to cancer research.199,201 The advantages of orthotopic xenograft

animal models is that they produce a primary tumor from human cells in the
correct anatomical site (e.g., mammary gland in breast cancer). To prevent
rejection of the human (xenogeneic) tumor cells, this model requires the use of
immunodeficient mice. There are two main types of immunodeficient mice: nude
mice or mice homozygous for the scid mutation (SCID mice). The nude mice are
preferable because the relatively low-light-scattering quality of the hairless skin
increases the accuracy of noninvasive detection and counting with FFC.
6.13 Biomedical Applications

6.13.1 Optical lymphography
The progress in in vivo mapping of lymph vessels and lymph nodes with
reasonable precision has been made with many advanced technologies such as
optical lymphography, photoacoustic lymphography with nanoparticles or
dyes, computer tomography, positron emission tomography, magnetic
resonance imaging (MRI) with magnetic nanoparticles and radio-lymphos-
cintigraphy with 99mTc-sulfur-colloid-, albumin- or gadolinium-labeled
contrast agents.129,130,144,179–183,202–212 Among others, the optical lymphogra-
phy is the study of lymphatic vessels and lymph nodes through an injection of
optical blue dyes (e.g., Evans, Blue [EB], isosulfan, or lymphazurin) or
fluorescent contrast agents (e.g., fluorescein, QDs, Indocyanine Green
[ICG]).129,130,144,180–183,202–204
For mapping skin lymphatics in animals, we used lymphography with
well-established EB dye, fluorescein isothiocyanate (FITC; molecular weight
[MW], 40 kDa; excitation/emission wavelengths, 492 nm/518 nm; green) or
tetramethyl rhodamine isothiocyanate (RITC; MW, 150 kDa; excitation/
emission wavelengths, 495 nm/595 nm; red) [Fig. 6.30(a) and (b) and
Fig. 6.30(c)].129,130,144,146
Figure 6.30 Optical lymphography. (a) Mapping of mouse ear lymph vessels with EB (5 ml
of 1% EB dye in 0.9% NaCl was injected intradermally into the ear tip with a 10-ml Hamilton
solution syringe). (b) Mapping of mouse skin lymph vessels and their visualization using skin
chamber. (c) Fluorescent mapping of blood (green; FITC) and lymph (red; RITC) vessels in
mouse ear.
6.13.1.1 Indocyanine Green (ICG) lymphography

In fact, most contrast agents cannot be used in humans due to the concerns of
toxicity (e.g., quantum dots). In addition, some of these contrast agents are
limited in their ability to visualize small lymphatics (e.g., lymphazurin) and
provide insufficient depth resolution (e.g., a few mm for FITC). Among
others, the most promising translational contrast agent for optical lymphog-
raphy is ICG, for which application for mapping of human lymph vessels has
been already demonstrated in clinical trials in Japan.180,182
In our early preclinical studies, we injected ICG (0.2 mL/100 g) in a rat’s
tail vein, and we successfully mapped vessels of rat mesentery using TDM and
fluorescent microscopy. Excitation at a wavelength of 805 nm, 0.25 mW/cm2,
was provided by continuous diode laser, and the re-emitted fluorescence was
filtered at 830 nm and then detected using an intensified highly sensitive
camera (PentoMAX, Roper Scientific).129–144
6.13.1.2 Integrated fluorescent angio- and lymphography

Tracking a fluorescent dye after intravenous (i.v.) injection over time can be
used for diagnosis of some diseases with disturbing lymphatic drainage such as
lymphedema. In our studies, we demonstrated the feasibility of this approach
with FITC and ICG in healthy rats. After i.v. injection of ICG (tail vein
injection), the visualization of the vascular network was started, as expected,
with veins (70 to 80 s after injection); then in a short period of time (first 2 min
after injection), dye appeared in the arteries; and, finally, ICG mapped lymph
vessels (40 to 50 min) after injection (Fig. 6.31).144
After i.v. injection of an FITC-dextran, the first fluorescent image of the
rat mesenteric veins appeared within 1 minute. Fifteen minutes after injection,
the dye accumulated both in blood and lymph vessels.54,129 Two hours after
injection, FITC-dextran disappears from vascularity (Fig. 6.32).
6.13.1.3 Monitoring lymph flow profile

To visualize a lymph flow profile, we injected FITC-dextran into mesenteric
lymphatic vessels of rats with the FemtoJet microinjection system (Eppendorf
Figure 6.31 ICG blood and lymph angiography: ITDM image before ICG injection (left) and
fluorescence images (excitation 805 nm; emission 830 nm) at the 5th (middle; no dye in
lymph flow) and 45th min (right; dye in lymph flow) after ICG injection; LV – lymph vessel,
BVs – blood vessels.
348 Chapter 6
Figure 6.32 FITC-dextran blood and lymph angiography: ITDM image before injection
(left) and fluorescence images at the 15th (middle; dye in lymph flow) and 120th min (right; no
dye in lymph flow) after injection; LV – lymph vessel, BVs – blood vessels.
Figure 6.33 Monitoring lymph flow profile after injection of a bolus of FITC-dextran solution
into a lymph vessel. (a) imaging of microinjection in lymph flow. (b) Changes of lymph flow
profile over time
North America) (Fig. 6.33). This local injection allowed us, for the first time,
to estimate lymph plasma flow velocity in individual lymph vessels in vivo and
measure the real-time dynamics of the lymph flow profile in vivo.
6.13.2 In vivo label-free imaging of lymphatic function

With ITDM and mesenteric animal models, the following lymphatic functions
can be explored: (1) lymph flow velocity; (2) vessel diameter; (3) contractile
function and (4) valve activity.9–11,129,144,148,149,187–189 Depending on a
structure’s size, different magnifications ( 4, 8, 20, 10) were used to
image the relatively large whole lymphangion (fragment of lymph vessel between
closely located input and output valves), as well as single cells ( 40, 60,
100), in lymph and blood flow (Fig. 6.34). In addition, ITDM images of rat
mesentery allowed visualizing initial lymphatic vessels [Fig. 6.34(b) and (c)].
Mesenteric lymph vessels of healthy rats that are available for imaging by
ITDM (i.e., located in a transparent part of the mesentery) have mean
diameters of 130 to 150 mm (diameter range, 50 to 270 mm). One half of them
had spontaneous phasic contractions with a mean amplitude of 29 ± 9%.
Lymphangions with larger diameters had lower amplitudes. The majority of
Figure 6.34 In vivo imaging of mesenteric lymph vessels. (a) Schematic of vascular
network in tissue. (b,c) Initial lymphatic (4 and 100 ); (d) Mesenteric tissue with valvular
lymph vessel and surrounding blood vessels (10 ). (e) High-resolution imaging of single
cells in lymph flow (100 ).
lymphatics with phasic contractions (78%) had active valves that periodically
opened and closed. Cells moved in 85% of the lymphangions.9,144
6.13.2.1 Lymph flow
Using the principles of PIV approach, we calculated lymph flow velocity as an
average cell velocity (Fig. 6.35):9,10,129,144
Flow velocity ¼ Distance travelled ðmmÞ∕Time ðsecÞ
The recording of lymph flow in real time showed that cell motion had an
oscillatory character. Lymph usually moved forward for a short time; the
motion was then interrupted, and lymph movement stopped for up to 1 to
1.5 s. After that, the lymph started to flow in the reverse direction. Lymph
usually oscillated at a mean rate of 64 ± 8 oscillations/min.
Figure 6.35 Real-time dynamics of cell velocity in axial lymph flow of the mesenteric lymph
vessel with a mean diameter of 170 ± 5 mm.
350 Chapter 6
Lymph flow velocity in noncontracting microvessels were 20 to 30% lower

than in contracting lymphangions. Depending on lymph flow velocity, cell
distribution in a cross section of a lymphangion varied. Most often, cell
distribution was relatively uniform in vessels with relatively low velocities and
nonactive valves (80% of cases). However, in vessels with high velocities and
functioning valves (20% of cases), most cells were concentrated near the vessel
axis. On average, concentration of cells in lymph, especially in prenodal
lymph, is lower than blood. From our estimations, the mean concentration of
cells in the lymph flow of intact mesenteric vessels is approximately
1 105 cells/ml. From these data, the average cell fraction in lymph was
5.5%. This parameter is analogous to the hematocrit in the blood system and
may be called the “lymphocrit.” It is interesting that a hematocrit of 5 to 6%
may be found only in blood capillaries, compared to a hematocrit of 20 to
30% in 60 to 70-mm-diameter arterioles. The cell concentration in flow was
higher in vessels with higher lymph flow velocity, and was somewhat
correlated with the amplitude of phase contractions and the valve activity.
Furthermore, we discovered the unique natural feature of a lymphatic
valve and contractile activity to provide hydrodynamic focusing of cells
moving in lymph flow that can significantly advance FC in vivo.129
Thus, ITDM with appropriate animal model(s) provides studying
lymphatic function in different diseases and under the impact of therapeutic
drugs and laser radiation.
6.13.2.2 Experimental lymphedema

Lymphedema is a complication of lymphatic drainage decompensation that may
happen during congenital lymphatic dysplasia, hepatic cirrhosis, venous
insufficiency, obstruction of lymph nodes due to tumor infiltration or filariasis,
or damage of lymph vessels and nodes following surgical extirpation or radiation
therapy.66,67,118,213–221 In particular, postmastectomy lymphedema (PML)
develops in 25 to 50% of women after breast cancer treatment.218–220 The key
mechanisms of lymphatic disturbances are associated with dilation of lymph
vessels, valve insufficiency, and abnormal lymph flow. In our studies we created
new animal models of lymphedema in mesentery that allowed us to use the
power of high-resolution ITDM to monitor the unknown before changes of
lymphatic function, including lymphatic dysfunction at the latent stage of
lymphedema.221 Specifically, lymphedema was produced in rats by lymphade-
nectomy through microsurgical removal of all regional (mesenteric) lymph
nodes. The same lymph vessel has been imaged by ITDM before (intact state)
and at four time points after surgery (30 min, 1 week, 4 weeks, and 11 weeks). At
the same time points, the degree of edema was evaluated by the water amount in
tissues that was measured through weighing of the excised mesentery before and
after being dried at 100°C. The volume of fluid in the mesentery was calculated
as grams of water per 1 g of dry tissue. Our data showed that significant edema
Figure 6.36 Extension of lymphatic vessels at the experimental lymphedema. (a) Mesenteric
lymph vessel before lymphedema. (b) The same vessel after 1 week of lymphedema.221
was developed in one half of the animals after lymphadenectomy and was
accompanied by lymphatic dilation (increasing diameter) and inhibition of
contractility and lymph flow (Fig. 6.36). Notably, the rest of the animals without
tissue edema also demonstrated lymphatic dysfunction, which may indicate a
latent stage of lymphedema. From this, a possible preventive treatment might be
directed at the compensation of vascular disturbances.
We also obtained experimental evidence involving blood vessels in the
development of lymphedema after lymphadenectomy. The dilation of blood
microvessels and expansion of the microvascular net without significant
hemorrhage into the interstitium was detected at the stage of well developed
tissue edema and significant lymphatic disturbances. Obtained links between
functions of lymph vessels, blood vessels, and tissue edema in vivo are
important for understanding basic lymphedema and may be used for
developing new treatment strategies.
6.13.2.3 Nicotine intoxication

We studied lymphatic disturbances induced by nicotine, an important
component of cigarette smoke that is responsible for inducing pathology in
many tissues in both humans and animals.148,222–224 Using ITDM and the rat
mesentery model, we demonstrated that nicotine induces lymphatic dysfunction
in vivo.144,148 The topical effect of nicotine is concentration dependent and
includes disturbances in contractility (e.g., asynchronous motion of the lymphatic
wall) and inhibition of lymph flow. Inhibition of lymph flow was also monitored
at acute nicotine intoxication induced by i.v. injection.
Surprisingly, the 14-day chronic intoxication using a mini–osmotic pump
with a 10-mM nicotine solution (0.5 ml/h delivery rate) did not markedly
change the function of lymphatic vessels. The absence of effects may be the
result of adaptation of the lymphatic system to nicotine.
These findings suggest that the nicotine-induced lymphatic dysfunction is
a part of an integrative quick response of an organism to cigarette smoke that
might contribute to vascular abnormalities and tissue edema.
352 Chapter 6
6.13.2.4 Nitric oxide

The essential regulator of mesenteric lymph vessel function in vivo is nitric
oxide (NO).225–228 In our experiments,9,144,228 the direct local effect of Sodium
Nitroprusside (SN; topical application; 10–5 М; 30 min), an exogenous donor
of NO, on lymphatic function included the dilation of lymph vessels by 25 ±
2.5 mm with no changes in lymph flow. The responses of lymph vessels to i.v.
injection of SN (total introductory dose, 100 mg/kg; i.e., the drug concentration
in blood was 5 10–5 M) were similar to this but less intense.
The direct effect of L-NAME (10–4 M; 30 min) that blocks endothelial
NO synthase affects lymphatic contractility by an two times decrease of the
amplitude of phasic contractions.
6.13.2.5 High-power laser–induced thermal effects on lymph vessels

We estimated the effect of high-power laser radiation (585 nm, 10-ms pulse
duration, 0.5–30 J/cm2) on mesenteric lymph vessels as a model of optimizing
laser treatment of different vascular anomalies (port wine stains, lymphatic
malformation, etc.).144 At laser exposure, we monitored significant constric-
tion of lymph vessels up to their complete obliteration that was associated
with the inhibition and stopping of lymph flow (Fig. 6.37).
Simultaneously, irradiation induced localized hemorrhages around blood
vessels, likely due to rupture of vascular walls. These hemorrhages led to the
Figure 6.37 Laser-induced thermal effects on blood and lymph vessels in vivo. (a) Intact
mesenteric vein (V) and arteries (A1 and A2) with fast blood flow. (b) Damage of these
vessels immediately after a laser pulse: localized hemorrhage (arrow) around the vein (V)
and stasis in small arteries (A2). (c) Intact lymph vessel (L) before a laser pulse (black
dashed line, internal margin of lymphatic wall). (d) Laser-induced constriction of a lymph
vessel, which coincided with stasis in neighboring veins. Pulse parameters: wavelength,
585 nm; pulse duration, 10 ms; radiant exposure, 0.5–30 J/cm2.
Figure 6.38 High-resolution image (100 , water immersion) of RBC aggregates in lymph
flow.
entry of many RBCs into lymph flow, which formed aggregates that inhibited
lymph flow up to stopping (Fig. 6.38).
6.13.3 In vivo flow cytometry

Imaging and counting of individual cells in blood and lymph flows in vivo is
potentially important for studying cell-to-cell and cell-to-vessel-wall dynamic
interactions, blood transport (e.g., oxygen delivery), the response of cells to
different interventions (e.g., drugs, smoking, radiation), and disease diagnosis and
prevention (e.g., metastases, heart attack or stroke alert, diabetic shock, sickle-cell
crisis, etc.).1–4,10–12,67,70,116,120,122,123,126,129–130,134,140,142–144,148,151,169,171,173,175
6.13.3.1 Label-free image flow cytometry
Using high-speed and high-resolution ITDM in the bright-field mode and
selected animal models, the combination of light absorption and scattering
effects on cells made it possible to visualize and identify blood and lymphatic
cells without conventional labeling and vital staining. In particular, due to
relatively strong light absorption by RBCs, these single cells in flow appeared
mostly as dark objects in the transillumination mode, while weakly absorbing
WBCs and platelets appeared either as light objects (e.g., in the presence of
many more strongly absorbing RBCs in blood flow or with dominant
scattering effects) or, in contrast, as slightly dark objects (e.g., in the
transparent plasma without RBCs) (Figs. 6.24 and 6.26). In particular, our
technique enables us to measure the velocity of individual cells up to 10 mm/s
without marked optical distortion of cell images in packed multi-file blood
flow (see also Section 6.7).10–12,129,144
6.13.3.1.1 High-speed and high-resolution imaging of circulating lymphatic cells
WBCs. On the basis of aforementioned notes and differences in size and
shape, LFC can detect individual WBCs (the majority of lymphatic cells) and
distinguish them from some other lymphatic cells (e.g., RBCs, melanoma
CTCs) in lymph flow of mesenteric vessels without conventional labeling and
vital staining (Fig. 6.39).129
354 Chapter 6
Figure 6.39 High speed imaging (2,500 fps) of a WBC in the mesenteric lymph flow.
RBCs. Rare RBCs can transport by lymph flow in humans and other
mammals under physiological conditions.66,67,229–233 The features of initial
lymphatics66,234–238 suggest easy entry for highly deformable and relatively
small-diameter (5 to 7 mm) RBCs1–3 when they extravasate into tissue from
blood vessels [Figs. 6.34(b) and (c)]. Compared to the high-speed packed flow
of RBCs in blood that induced their significant deformability, the relatively
slow lymph flow with small cell concentration prevents transient deformation
of the rare-moving RBCs and allows keeping their original biconcave shape.
(Fig. 6.40).10,11,129,144
The number of RBCs in the peripheral lymph increases as a result of aging,
muscular exercises, exposure to heat/cold and radiation as well as in some
pathologies associated with increasing extravasation of RBCs in tis-
sues.144,221,233,243 In experimental venous insufficiency (ligation of the collecting
vein) and after high-power laser irradiation, we monitored hemorrhages into the
interstitium and, as a result, many RBCs entered into lymph flow; they can
change their behavior in flow and affect lymphatic function (Fig. 6.38).144–221
CTCs. It has been well established that the metastatic cascade for many
tumors involves the lymphatic system.117,119,129–131,134,135,137,139,140,169,244–248
However, the role of lymphatic CTCs in metastatic progression are poorly
understood. To solve this problem, we have developed different FC-based in
vivo approaches and demonstrated their feasibility on animal cancer models to
count and image lymphatic CTCs naturally and spontaneously shedding from
Figure 6.40 In vivo high-resolution, high-speed imaging of the single RBC in lymph flow of
a healthy rat (indexed colors).
Figure 6.41 Image lymph flow cytometry in vivo. (a) Metastatic melanoma tumor in mouse
ear. (b) High-resolution, high-speed images of melanoma CTCs (aggregate of two cells), a
partly lysed necrotic melanoma CTC and normal WBC in lymph flow (magnification, 40
and 100 , water immersion; 500–2,500 fps).
a primary tumor.16,64,129,140,169,175 Specifically, in mice with metastatic

melanoma [Fig. 6.41(a)], we obtained high-resolution label-free images of
rare spontaneous lymphatic CTCs [Fig. 6.41(b), left].54,129 CTCs were
distinguished by their larger size (approximately 1.5 to 2 times more than
WBCs, Fig. 6.41(b), right) and black localized pigmentation (melanin). Cells
with round shapes were probably alive or apoptotic, while pigmented moving
objects with irregular forms were likely related to the fragments of dead
CTCs, Fig. 6.41(b), middle. In contrast to WBCs, which moved typically as
single cells, CTCs often formed aggregates [Fig. 6.41(b), left and right].
6.13.3.1.2 Label-free high-speed imaging RBCs in fast blood flow

Because of the high endogenous absorption of Hb and low absorption of
plasma proteins, label-free, high-speed (up to 10,000 fps), high-resolution (250
to 300 nm) transmission imaging of thin tissue, such as ear and mesentery in
animal models, enables time-resolved determination of the shape of single
RBCs in a capillary with single file and in a small vessels with multi-file flow
(e.g., with diameters of 20 to 70 mm and flow velocity up to 0.5 cm/
sec).10–12,144,173 This approach can provide real-time measurements of rapidly
changing shapes, and, thus deformability of RBCs in vivo.
We demonstrated the feasibility of this technical platform to monitor:
(1) high deformability of parachute-like RBCs as they squeezed at 0.6 mm/sec
through a narrow gap between the vessel wall and adherent cell [Fig. 6.42(a)];
(2) how quickly relatively fast-flowing RBCs (1 mm/sec) changed shape as
they interacted with much more slowly moving (so-called rolling) cells, likely
WBCs (0.1 mm/sec) [Fig. 6.42(b)]; (3) significant dynamic deformation of
two RBCs in merging flow streams in a bifurcation zone [Fig. 6.42(c)] and
(4) extremely fast stretching (10,000 fps) of initially discoid RBCs to 0.7 mm
[Fig. 6.42(d)].12
This approach can also be used for real-time monitoring changes of RBC
deformability under drug impacts.144,173 For example, we showed the
presence of low-deformable RBCs after an injection of Diamide or
356 Chapter 6
Figure 6.42 High-resolution, high-speed monitoring of cells in blood flow. RBCs are
indicated by conventional arrows and arrowheads, rolling cells by arrows originating from
filled circles, and direction of flow by arrows with dashed lines. (a) Parachute-shaped RBC
traveling at 0.4 mm/s; 1,250 fps. (b) Interaction of fast-moving RBC with rolling cell;
2,500 fps. (c) Two RBCs in an area of merging flow streams with a velocity of 0.3 mm/s;
2,500 fps. (d) Fast-moving RBCs, 2,500 fps; magnification 40 .
Figure 6.43 High-speed (1,000–5,000 fps), high-resolution imaging of changes of RBC

deformability before (top images: normal deformability) and after (bottom images: low
deformability) administration of Chlorpromazine (a) and Diamide (b).173
Chlorpromazine (Fig. 6.43) compared to control measurements before

injection. This effect was profound in a specific localized area in curved
vessels, where maximum centrifugal forces were acting on cells, and in
bifurcation zones with higher RBC acceleration.
In general, in vivo image FC permits the study of the dynamics of cell
deformability in vivo, including early diagnosis of diseases accompanied by
changes in cell deformability (e.g., sickling diseases, anemia, cardiovascular

pathologies), which can serve as biological markers of disease development or
the response to drugs.
6.13.3.2 In vivo lymph and blood fluorescent flow cytometry

6.13.3.2.1 Normal and apoptotic WBCs in lymph flow
IFDM is promising in context of animal models for the study of lymphatic
cells.129,134,144,163,249 To study circulating lymphatic cells, we used in vivo
lymph FFC. Specifically, to count WBC (majority of lymphatic cells), the
fluorescence (FITC) labeled antibodies specific to the CD45 receptor of
WBCs (a common receptor of WBCs250,251) were injected into rat peritoneum
in vivo followed by time-resolved fluorescence monitoring of mesenteric
lymphatics.129,169 Figure 6.44(a) shows the CD45þ WBC moving in a lymph
flow with a velocity of 1720 ± 18 mm/s.
To capture an image of apoptotic cells, apoptosis was induced by
intraperitoneal injection of 1 mM dexamethasone (Sigma-Aldrich, St Louis,
Missouri) followed 6 h later by the injection of Annexin V-FITC (emission,
480 nm; exitation, 530 nm) and then by fluorescence monitoring of apoptotic
cells. Only a few apoptotic (green) cells were detected in lymph flow during the
1-h observation [Fig. 6.44(b)].129 Some of them were adhered to a vessel wall. In
vivo data were verified by in vitro tests: when dexamethasone-treated WBCs
were labeled by a conventional apoptotic kit of Annexin-FITC and Propedium
Iodide (PI), the majority of cells were green (i.e., apoptotic) and only rare cells
were unlabeled (i.e., viable) or red (i.e., necrotic) [Fig. 6.44(c)].
To our knowledge, it was the first demonstration of detection of
circulating apoptotic cells in lymph flow in vivo, and holds promise to be
significant since apoptotic cells in vivo is the crucial point for diagnosis and
therapy of many diseases56–59 (e.g., efficacy of anti-tumor drugs).
6.13.3.2.2 RBCs, WBCs, and apoptotic cells in blood flow

For counting fast-moving cells in blood flow without imaging, Lin and
colleagues153,154 successfully adapted a single-photon fluorescence technique
to monitor in vivo the kinetics of fluorescently labeled circulating cells in the
Figure 6.44 Lymph image FFC: (a) In vivo CD45 þ WBC exiting a valve aperture. (b) In
vivo FITC-labeled apoptotic cell in the lymph flow; dashed lines indicate lymph vessel
structure. (c) Apoptotic (green) and necrotic (red) WBCs in suspension in vitro.
358 Chapter 6
Figure 6.45 (a) Apoptotic (arrowhead, red) and normal (two arrow triangles, green) cells in
two blood vessels (dashed lines). (b) Apoptotic cells (arrowhead, red) in interstitium (arrow
shows one vessel). (c) Normal (arrow, green) and apoptotic (red) cells in a lymph node
(dashed line).
blood vessels (50 mm deep) of a mouse ear. Later Lin’s, ours, and other
groups used one- and two-color FFC for counting various types of fluorescently
tagged circulating blood cells, including RBCs, WBCs, apoptotic cells, and
CTCs.153–161,169,174,175,252
Furthermore, using a mesentery model, we performed counting and
imaging apoptotic cells in the blood flow (Fig. 6.45).144 Fluorescently labeled
apoptotic WBCs (red) were injected into a rat’s tail vein. The high-speed IFDM
was used to image apoptotic cells in blood flow [Fig. 6.45(a)] and tissues, and,
additionally, the PIV approach was used to count these cells in small mesenteric
vessels. Our results demonstrated rapid clearing of apoptotic cells from the
circulation, with a half-life of 8 minutes, which is in line with data from C.
Lin.252 We also observed (1) rolling apoptotic cells in veins within 10 minutes
after the injections [Fig. 6.45(a), bottom vessel], (2) the appearance of apoptotic
cells in the interstitium at the 15th min, and (3) accumulation of apoptotic cells
in mesenteric lymph nodes [Fig. 6.45(b)] at 30th min.
6.13.3.2.3 Circulating tumor cells (CTCs)

To date, the most important target of in vivo FC is
CTCs.10,16,17,64,65,129,130,140,143,154–157,159–161,169,175 Approximately 90% of all
cancer deaths are caused by metastases produced by CTCs.59–63,244–245,253–259
Thus, to develop effective, targeted diagnosis and therapy, we need to identify
them as early as possible when CTC concentration in blood is suggested to be
extremely low; ideally, a few cells in whole blood volume. Toward this goal,
substantial efforts have been made to develop methods to examine CTCs in
blood samples in vitro (e.g., Cell Search system, CTC-iChip, RT-PCR) to
identify molecular and genetic properties of metastatically aggressive CTCs
(e.g., tumor-initiating cancer stem cells, CTCs with epithelial-mesenchymal
transition) and to characterize the interactions of CTCs with host cells in the
primary tumor (e.g., fibroblasts and tumor-associated macrophages) and
blood cells (e.g., platelets).59–63,255–259 However, as mentioned above
(Section 6.10 In vivo Flow Cytometry), the sensitivity of these tools is limited
for detecting rare cells. Additionally, blood sampling, which is required for
any in vitro test, only provides snapshots of CTCs and thus, prevents
monitoring the behavior of individual CTCs over disease progression. As a
result, the multistep process of CTC dissemination in the entire body at the
single-cell level is far from being understood, thus many early mechanisms of
metastasis remain elusive.
The problem can be solved by in vivo CTC detection16,143,169 as a first
step, in preclinical animal models of human disease. One of the advanced
technical platforms to study CTCs in animal models in vivo is
FFC.153–161,174–175 As in conventional fluorescent microscopy, to be detect-
able by FFC, cells should be labeled by fluorescent contract agents. The most
promising approach involves using CTCs with fluorescent proteins, in
particular green fluorescent protein (GFP).16,145,154,156,161–162 Genetic encod-
ing of GFP results in fluorescence in all cancer cells produced during disease
progression. This enables easy and noninvasive detection of CTCs by in vivo
FFC, and of primary tumor and metastases by intravital whole-body imaging.
More specific molecular targeting, involving exogenous labels bioconjugated
with antibodies against a cell-surface marker of CTCs, can identify a specific
cell subpopulation (e.g., stem CTCs).
Over the last decade, in vivo FFC has been widely used in preclinical
cancer research to study i.v. injected and spontaneous (i.e., naturally shed
from primary tumor) CTCs.17,153–161,169,175 It was shown that 80 to 90% of
tumor cells injected in the bloodstream of healthy animals were depleted from
circulation within 0.5 to 4 h, while a low amount of CTCs continued to
circulate over days. Furthermore, cells from highly metastatic cell lines
disappeared faster than low metastatic ones.16,159
Recently, we used FFC to find links between counts of spontaneous
CTCs, primary tumor size, and metastasis progression. For this, immunode-
ficient nude mice were inoculated in the mammary glands with human breast
cancer cells expressing GFP and luciferase (MDA-MB-231-luc2-GFP), which
caused orthotopic primary tumor growth and metastatic disease development
(Fig. 6.46).161
With FFC, we were able to detect bulk CTCs that were expressing GFP
with an excitation wavelength of 488 nm and an emission wavelength of
509 nm. Luciferase was used to identify metastasis by whole-body imaging
(e.g., with IVIS system). In the majority (.90%) of mice, CTCs appeared
starting at week 1 of tumor development and were monitored during the entire
course of the disease. The in vivo whole-body imaging, ex vivo high-resolution
fluorescence imaging, and in vitro histology, as expected, demonstrated the
earliest metastasis in the sentinel lymph node (SLN). In week 2 after tumor
inoculation, metastatic lesions of the SLN affected 20% of the mice, and CTC
colonies in the lung were detected in 40% of the mice [Fig. 6.46(b)]. In weeks 5
360 Chapter 6
Figure 6.46 Development of metastatic breast cancer. (a) Growth of primary tumor;
(b) Colony of cancer cells (green) in the lung at week 2 after tumor inoculation; (c) A CTC
(green) in the blood sample obtained from a mouse at week 1 after tumor inoculation;
(d,e) Lung metastases at week 8 after tumor inoculation confirmed by two independent
methods: fluorescence image of lung ex vivo (d) and histological staining (H&E) of a lung
section (e). Black arrow and green line in (e) show metastasis6HHFRORUSODWHV
and 6, metastatic disease progression was characterized by the development of

micrometastases (mainly in the lungs) in one half of the animals. In week 8,
micro- and macrometastases (e.g., lungs, lymph nodes, liver, and bones) were
observed in 88.9% of tumor-bearing mice [Figs. 6.46(d) and (e)]. Monitoring
CTCs afterwards revealed interesting facts. In most animals (83%), the highest
CTC rate was detected when the primary tumor was small and slowly growing
[Fig. 6.47(a)]. Later, when the primary tumor started progressively growing,
the CTC rate decreased. The occurrence of this phenomenon varied from
mouse to mouse, likely due to specific features depending on the host
environment. In a few cases when the tumor was large, we observed a second
increase in CTC number, but this CTC peak was smaller than the earlier one
[Fig. 6.47(b)].161
To determine whether an aggressive primary tumor releases CTCs with
the aforementioned phenomena, we used conventional flow cytometry to sort
breast cancer cells for CD44 þ /EpCam þ /CD24 that characterized tumor-
initiating cancer stem cells (CSCs). Sorted cells (i.e., CSCs) were inoculated
into a mammary gland of the first group of mice, and non-sorted cells (i.e.,
bulk tumor cells) were inoculated into a mammary gland of the second group
of mice. Although the number of inoculated CSCs (1.7 105 cells per mouse)
Figure 6.47 Correlation of CTC kinetics with tumor growth in breast cancer based on FFC
data and whole body imaging, respectively. (a) Individualized CTC dynamics maximize
before the primary tumor starts progressively growing; (b) Individualized CTC dynamics with
two peaks. Tumor volume was calculated by the equation: ½ long diameter short
diameter.161
was almost 30 times less than the number of inoculated bulk tumor cells (5
106 cells per a mouse), all mice from the first group developed metastatic
disease (e.g., lung, liver) within 4 weeks after inoculation. In this group, CTCs
were detected in circulation starting from week 1 after inoculation, and their
dynamic was characterized by the maximum CTC number at an early stage of
disease and their gradual decrease during primary tumor growth and
metastasis progression (Fig. 6.48).161
In summary, our findings hold promise to provide new insights on
mechanisms of metastatic diseases and may have diagnostic implication
towards developing advanced diagnosis, allowing for well-timed therapy that
is more effective. Further studies with comprehensive statistical analysis and
trials in humans will provide the insight on these phenomena.
Figure 6.48 Dynamics of CTCs in mice after inoculation of CSCs. (a) A typical example of
the dynamics of CTCs with their maximum number before the primary tumor starts growing
progressively; (b) Lung micrometastases (green).
362 Chapter 6
6.14 Summary
Optical techniques provide minimal invasive insight into blood and lymph
circulation systems physiology and biophysics. At tissue level, coherent
radiation scattering can be employed to detect blood perfusion that causes
specific spatiotemporal variations of scattered light intensity known as
dynamic laser speckles. Because of high light scattering in tissues, the laser
speckle technique does not provide quantitative data related to volumetric
blood flow rate, and it is applicable for the monitoring and mapping of short-
term relative variations of blood perfusion caused by external factors, as
demonstrated in Section 6.3. The effect of light scattering within a tissue can
be decreased with certain optical techniques. Section 6.4 illustrates a possible
way to improve the laser speckle technique by means of a coherent selection of
light scattered at a specific depth. Quantitative optical measurements of the
blood flow velocity within an individual blood vessel are available at length
scales less than 1 mm. As discussed in Section 6.5.1, in this case one can detect
the fraction of light that retains its rectilinear propagation while passing
through thin tissue layers. The further decrease of length scale toward cellular
levels makes the effect of light scattering negligible, and superficial blood
vessels can be visualized in a common way. At the capillary-network level, the
issue concerns complex dynamics of a living object. The problem of
involuntary movements' effect on blood velocity measurements is discussed
in Section 6.5.2. Finally, at very high magnification, intravital microscopy
provides a plenty of data at cellular level as it was overviewed through
Sections 6.7 to 6.13. Although intravital microscopy at the single-cell level
became a routine tool with its highest impact in physiology and biophysics,
the authors hope that the optical technique will find a challenging application
in understanding life functions at larger scales of tissues, organs, and whole
organisms.
Acknowledgments
This work was supported by the grants R01 EB000873, R01 CA131164, R01
EB009230, R21 CA139373; R01EB017217 from the National Institutes of
Health (NIH); grant IDBR 085 2737 from the National Science Foundation
(NSF); grant W81XWH-11-1-0129 from the Department of Defense (DOD),
grant UL1TR000039 from the Arkansas Breast Cancer Research Program,
and grants from the Arkansas Biosciences Institute and the Translational
Research Institute at UAMS.
This work has been performed in parts of the in the framework of RF
governmental contract №2014/203, №1490, competitive contract of The
Ministry of Science and Education of RF, №3.1340.2014/K, grant of The
President of RF NSh-703.2014.2, The Tomsk State University Academic D.I.
Mendeleev Fund Program, and grant RFBR 14-02-00526a.
References
1. K. S. Saladin, “Anatomy and Physiology: the unity of form and
function,” 6th Edition, McGraw-Hill Higher Education, New York (2012).
2. C. Pallister, “Blood: physiology and pathophysiology,” Oxford Univer-
sity Press, Oxford, 585 (1994).
3. K. Rogers (Ed.), “Blood: physiology and circulation (The Human
body).” Rosen Education Service, Britannica, ISBN-10: 1615301216,
239 (2010).
4. S. E. Charm and G. S. Kurland, “Blood flow and microcirculation,”
John Wiley & Sons, Boston (1974).
5. H. H. Lipowsky, “Microvascular rheology and hemodynamics,”
Microcirculation (N.Y.). 12, 5–15 (2005).
6. R. Skalak and P. I. Branemark, “Deformation of red blood cells in
capillaries,” Science 164, 717–719 (1969).
7. J. C. Firrell and H. H. Lipowsky, “Leukocyte margination and
deformation in mesenteric venules of rat,” Am. J. Physiol. 256,
H1667–H1674 (1989).
8. H. Minamitani, K. Tsukada, E. Sekizuka, and C. Oshio, “Optical
bioimaging: from living tissue to a single molecule: imaging and
functional analysis of blood flow in organic microcirculation,”
J. Pharmacol. Sci.. 93, 227–233 (2003).
9. E. I. Galanzha, G. E. Brill, Y. Aizu, S. S. Ulyanov, and V. V. Tuchin,
“Speckle and Doppler Methods of Blood and Lymph Flow Monitor-
ing,” In: V. V. Tuchin (Ed.), Handbook of Optical Biomedical
Diagnostics, SPIE Press, PM107, Bellingham, USA, 875–937 (2002)
10. V. V. Tuchin, E. I. Galanzha, and V. P. Zharov, “In vivo image flow
cytometry,” In: V. V. Tuchin (Ed.), Advanced Optical Flow Cytometry:
Wiley-VCH Verlag GmbH & Co. KGaA, 387–433 (2011).
11. E. I. Galanzha, V. V. Tuchin, and V. P. Zharov, “In vivo integrated flow
image cytometry and lymph/blood vessels dynamic microscopy,”
J. Biomed. Opt. 10, 54018 (2005).
12. V. P. Zharov, E. I. Galanzha, Yu. Menyaev, and V. V. Tuchin, “In vivo
high-speed imaging of individual cells in fast blood flow,” J. Biomed.
Opt. 11 (5), 054034 (2006).
13. D. E. Sosnovik, “Molecular imaging in cardiovascular magnetic
resonance imaging: current perspective and future potential,” Top.
Magn. Reson. Imaging. 19, 59–68 (2008).
14. H. Markus, “Transcranial Doppler detection of circulating cerebral
emboli. A review,” Stroke 24, 1246–1250 (1993).
15. Y. Yang, D. G. Grosset, Q. Li, A. Shuaib, and K. R. Lees. “Turbulence
and circulating cerebral emboli detectable at Doppler ultrasonography: a
differentiation study in a stenotic middle cerebral artery model,” Am. J.
Neuroradiol. 23, 1229–1236 (2002).
364 Chapter 6
16. E. I. Galanzha and V. P. Zharov, “Photoacoustic flow cytometry,”

Methods 57, 280–296 (2012).
17. E. I. Galanzha, E. V. Shashkov, T. Kelly, J.-W. Kim, L. Yang, and V. P.
Zharov, “In vivo magnetic enrichment and multiplex photoacoustic
detection of circulating tumour cells” Nat. Nanotechn. 4(12), 855–860
(2009).
18. Y. Aizu and T. Asakura, “Coherent optical techniques for diagnostics of
retinal blood flow,” J. Biomed. Opt. 4(1), 61–75 (1999).
19. Y. Aizu and T. Asakura, “Bio-speckle phenomena for blood flow
measurements: speckle fluctuations and Doppler effects,” in Optics and
Lasers in Biomedicine and Culture, OWLS V - Optics Within Life
Sciences 5, C. Fotakis, T. Papazoglou, and C. Kalpouzos (Eds.),
297–300, Springer, Berlin (2000).
20. J. D. Briers, “Laser Doppler and time-varying speckle: A reconcilia-
tion,” J. Opt. Soc. Am. A 13, 345–350 (1996).
21. G. Garhofer, R. Werkmeister, N. Dragostinoff, and L. Schmetterer,
“Retinal blood flow in healthy young subjects,” Inv. Opht. & Vis. Sci. 53(2),
698–703 (2012).
22. P. Cherecheanu, G. Garhofer, D. Schmidl, R. Werkmeister, and
L. Schmetterer, “Ocular perfusion pressure and ocular blood flow in
glaucoma,” Curr. Opin. Pharmacol. 13, 36–42 (2013).
23. B. Pemp, E. Polska, G. Garhofer, M. Bayerle-Eder, A. Kautzky-Willer,
and L. Schmetterer “Retinal blood flow in type 1 diabetic patients with
no or mild diabetic retinopathy during euglycemic clamp,” Diabetes
Care 33, 2038–2042 (2010).
24. V. Doblhoff-Dier, L. Schmetterer, W. Vilser, G. Garhöfer, M. Gröschl,
A. L. Rainer, and R. M. Werkmeister, “Measurement of the total retinal
blood flow using dual beam Fourier-domain Doppler optical coherence
tomography with orthogonal detection planes,” Biomed. Opt. Express
5(2), 630–642 (2014).
25. D. Feuerstein, M. Takagaki, M. Gramer, A. Manning, H. Endepols,
S. Vollmar, T. Yoshimine, A. J. Strong, R. Graf, and H. Backes,
“Detecting tissue deterioration after brain injury: regional blood flow
level versus capacity to raise blood flow,” J. Cereb. Blood Flow Metab.
34(7), 1117–1127 (2014).
26. J. S. Silvestre, D. M. Smadja, and B. I. Lévy, “Postischemic
revascularization: from cellular and molecular mechanisms to clinical
applications,” Physiol. Rev. 93(4), 1743–1802 (2013).
27. M. J. Leahy (Ed.) Microcirculation Imaging, Wiley-VCH Verlag GmbH &
Co. KGaA, Weinheim, 411 (2012).
28. Y. Tajima, H. Takuwa, H. Kawaguchi, K. Masamoto, Y. Ikoma,
C. Seki, J. Taniguchi, I. Kanno, N. Saeki, and H. Ito, “Reproducibility
of measuring cerebral blood flow by laser-Doppler flowmetry in mice,”

Front. Biosci. 6, 62–68 (2014).
29. I. V. Fedosov and V. V. Tuchin “Laser Doppler and speckle techniques
for bioflow measurements” in Coherent-Domain Optical Methods:
Biomedical Diagnostics, Environmental Monitoring and Material Science,
Ed. by V. V. Tuchin, Second edition. Berlin, Heidelberg, NY: Springer-
Verlag, 487–564 (2013).
30. I. Sigal, R. Gad, M. A. Caravaca-Aguirre, Y. Atchia, D. B. Conkey,
R. Piestun, and L. Ofer, “Laser speckle contrast imaging with extended
depth of field for in vivo tissue imaging,” Biomed. Opt. Express 5(1), 123–
135 (2014).
31. D. Briers, D. D. Duncan, and E. R. Hirst et al., J. Biomed. Opt. 18(6),
066018 (2013).
32. I. Meglinski and V. Tuchin, “Diffusing Wave Spectroscopy: Application
for Blood Diagnostics,” in Coherent-Domain Optical Methods: Biomedi-
cal Diagnostics, Environmental Monitoring and Material Science, 2nd
ed., V. V. Tuchin (ed.), Springer-Verlag, Berlin, Heidelberg, N.Y.,
149–166 (2013).
33. I. Meglinski, V. Kal’chenko, Y. Kuznetsov, B. Kuznik, and V. Tuchin,
“Towards the nature of biological zero in the dynamic light scattering
diagnostic modalities,” Doklady Physics 58(8), 323–326 (2013).
34. A. Pinhas, M. Dubow, N. Shah, T. Y. Chui, D. Scoles, Y. N. Sulai,
R. Weitz, J. B. Walsh, J. Carroll, A. Dubra, and R. B. Rosen, “In vivo
imaging of human retinal microvasculature using adaptive optics
scanning light ophthalmoscope fluorescein angiography,” Biomed. Opt.
Express 4(8), 1305–1317 (2013).
35. Y. Jia, J. C. Morrison, J. Tokayer, O. Tan, L. Lombardi, B. Baumann,
C. D. Lu, W. J. Choi, J. G. Fujimoto, and D. Huang, “Quantitative
OCT angiography of optic nerve head blood flow,” Biomed. Opt.
Express 3(12), 3127–3137 (2012).
36. E. Logean, L. Schmetterer, and C. E. Riva, “Velocity profile of red
blood cells in human retinal vessels using confocal scanning laser
Doppler velocimetry,” Laser Physics 13(1), 45–51 (2003).
37. C. E. Riva, “Laser Doppler Techniques for Ocular Blood Velocity and
Flow” in Ocular Blood Flow, L. Schmetterer and J. Kiel (Eds.), Springer-
Verlag, Berlin, Heidelberg (2012).
38. M. A. Borozdova, I. V. Fedosov, and V. V. Tuchin, “Laser Doppler
anemometer signal processing for blood flow velocity measurements,”
Quantum Electronics 45 (3), 275–282 (2015).
39. Y. Huang, Z. Ibrahim, D. Tong, S. Zhu, Q. Mao, J. Pang, W. P. A. Lee,
G. Brandacher, and J. U. Kang, “Microvascular anastomosis guidance
and evaluation using real-time three-dimensional Fourier-domain
366 Chapter 6
Doppler optical coherence tomography,” J. Biomed. Opt. 18(11), 111404

(2013).
40. M. S. Mahmud, D. W. Cadotte, B. Vuong, C. Sun, T. W. H. Luk,
A. Mariampillai, and V. X. D. Yang, “Review of speckle and phase
variance optical coherence tomography to visualize microvascular
networks,” J. Biomed. Opt. 18 (5), 050901 (2013).
41. W. Trasischker, R. M. Werkmeister, S. Zotter, B. Baumann,
T. Torzicky, M. Pircher, and C. K. Hitzenberger, “In vitro and in vivo
three-dimensional velocity vector measurement by three-beam spectral-
domain Doppler optical coherence tomography” J. Biomed. Opt. 18 (11),
116010 (2013).
42. G. Egawa, S. Nakamizo, Y. Natsuaki, H. Doi, Y. Miyachi, and
K. Kabashima, “Intravital analysis of vascular permeability in mice
using two-photon microscopy,” Sci. Rep. 3, 1932 (2013).
43. L. Vieira de Moraes, C. E. Tadokoro, I. Gómez-Conde, D. N. Olivieri,
and C. Penha-Gonçalves, “Intravital placenta imaging reveals
microcirculatory dynamics impact on sequestration and phagocytosis
of plasmodium-infected erythrocytes,” PLoS One 9(1), e1003154
(2013).
44. Z. Zhang, LDA Application Methods, Springer-Verlag, Berlin, Heidelberg
(2010).
45. M. A. Borozdova, I. V. Fedosov, and V. V. Tuchin “Laser Doppler
anemometer: new algorithm for signal processing at high light
scattering” Proc. SPIE 9448, 94481N (2015).
46. V. V. Tuchin, “In vivo optical flow cytometry and cell imaging, Rivista
Del Nuovo Cimento,” 37(7), 375–416 (2014).
47. M. A. Kurochkin, P. A. Timoshina, I. V. Fedosov, and V. V. Tuchin,
“Advanced digital methods for blood flow flux analysis using mPIV
approach,” Proc. SPIE 9448, 94481A (2015).
48. M. A. Kurochkin, I. V. Fedosov, and V. V. Tuchin, “In vivo study of
blood flow in capillaries using mPIV method,” Proc. SPIE 9031, 903107–1
(2014).
49. The Clinical Use of Blood (Handbook). World Health Organization.
Blood Transfusion Safety. Geneva. 2002: 221pp. http://www.who.int/
bloodsafety/clinical_use/en/Handbook_EN.pdf
50. D.K. Dressler, “Death by clot: acute coronary syndromes, ischemic
stroke, pulmonary embolism, and disseminated intravascular coagula-
tion,” AACN Adv Crit Care 20(2), 166–176 (2009).
51. V. L. Feigin, C. M. Lawes, D. A. Bennett, S. L. Barker-Collo, and
V. Parag, “Worldwide stroke incidence and early case fatality reported in
56 population-based studies: a systematic review,” Lancet Neurol. 8,
355–369 (2009).
52. N. M. Bambace and C. E. Holmes, “The platelet contribution to cancer

progression” J. Thromb Haemost. 209, 237–249 (2011).
53. Z. M. Ruggeri, “Platelets in atherothrombosis” Nat. Med. 8, 1227–1234
(2002).
54. A. F. Shorr, Y. P. Tabak, A. D. Killian, V. Gupta, L. Z. Liu, and M. H.
Kollef, “Healthcare-associated bloodstream infection: A distinct entity?
Insights from a large,” Crit. Care Med. 34, 2588–2595 (2006).
55. E. I. Galanzha, E. Shashkov, M. Sarimollaoglu, K. E. Beenken, A. G.
Basnakian, M. E. Shirtliff, J. W. Kim, M. S. Smeltzer, and V. P. Zharov,
“In vivo magnetic enrichment, photoacoustic diagnosis, and photother-
mal purging of infected blood using multifunctional gold and magnetic
nanoparticles,” PLoS One 7(9), e45557 (2012).
56. M. Brauer, “In vivo monitoring of apoptosis,” Prog. Neuropsychophar-
macol Biol. Psychiatry 27(2), 323–331 (2003).
57. P. Yang, J. R. Smith, K. S. Damodar, S. R. Planck, and J. T.
Rosenbaum, “Visualization of cell death in vivo during murine endotoxin-
induced uveitis,” Invest. Ophthalmol. Vis. Sci 44(5), 1993–1997 (2003).
58. S. D. Yan and D. M. Stern, “Mitochondrial dysfunction and
Alzheimer’s disease: role of amyloid-beta peptide alcohol dehydrogenase
(ABAD),” Int. J. Exp. Pathol. 86(3), 161–171 (2005).
59. T. A. Yap, D. Lorente, A. Omlin, D. Olmos, and J. S. de Bono,
“Circulating tumor cells: a multifunctional biomarker,” Clin. Cancer
Res. 20, 2553–2568 (2014).
60. C. Alix-Panabières, H. Schwarzenbach, and K. Pantel, “Circulating
tumor cells and circulating tumor DNA,” Annu. Rev. Med. 63, 199–215
(2012).
61. M. Cristofanilli, G. T. Budd, M. J. Ellis, A. Stopeck, J. Matera, M. C.
Miller, J. M. Reuben, G. V. Doyle, W. J. Allard, L. W. Terstappen, and
D. F. Hayes, “Circulating tumor cells, disease progression, and survival
in metastatic breast cancer,” N. Engl. J. Med. 351, 781–791 (2004).
62. S. C. Williams, “Circulating Tumor Cells,” Proc. Natl. Acad. Sci. USA
110(13): 4861 (2013).
63. M. Yu, S. Stott, M. Toner, S. Maheswaran, and D. A. Haber,
“Circulating tumor cells: approaches to isolation and characterization
(review),” J. Cell Biol. 192, 373–382 (2011).
64. E. I. Galanzha, E. V. Shashkov, P. Spring, J. Y. Suen, and V. P. Zharov,
“In vivo label-free detection of circulating metastatic melanoma cells by two-
color photoacoustic flow cytometry,” Cancer Res. 69, 7926–7934 (2009).
65. E. I. Galanzha, J.-W. Kim, and V. P. Zharov, “Nanotechnology-based
molecular photoacoustic and photothermal flow cytometry platform for
in vivo detection and killing of circulating cancer stem cells,”
J. Biophotonics 2, 725–735 (2009).
368 Chapter 6
66. M. Foldi and E. Kubik Foldi. (Eds.) Textbook of Lymphology. Urban &
Fischer, Munchen, Germany. (2003).
67. W.L. Olszewski, “The lymphatic system in body homeostasis: physio-
logical conditions,” Lymphat. Res. Biol. 1, 11–21 (2003).
68. P. Brown, “Lymphatic system: unlocking the drains,” Nature 436(7050),
456–458 (2005).
69. T. Godal and A. Engeset, “A preliminary note on the composition of
lymphocytes in human peripheral lymph,” Lymphology 11, 208–10
(1978).
70. J. G. Hall, J. G. Hall, and B. Morris, “The origin of the cells in the efferent
lymph from a single lymph node,” J. Exp. Med. 1121, 901–10 (1965).
71. Y. Aizu, K. Ogino, T. Sugita, T. Yamamoto, N. Takai, and T. Asakura,
“Evaluation of blood flow at ocular fundus by using laser speckle,”
Applied Optics 31(16), 3020–3029 (1992).
72. S. S. Ulyanov, “A new type of manifestation of doppler effect. An
application to blood and lymph flow measurements,” Opt. Eng. 34(10),
2850–2855 (1995).
73. L. E. Drain, The Laser Doppler Technique, John Wiley & Sons,
New York (1980).
74. A. Serov, W. Steenbergen, and F. de Mul, “Laser Doppler perfusion
imaging with a complimentary metal oxide semiconductor image
sensor,” Opt. Lett. 27, 300–302 (2002).
75. A. Serov, B. Steinacher, and T. Lasser, “Full-field laser Doppler
perfusion imaging and monitoring with an intelligent CMOS camera,”
Opt. Exp. 13, 3681–3689 (2005).
76. A. Serov and T. Lasser, “High-speed laser Doppler perfusion imaging
using an integrating CMOS image sensor,” Opt. Exp. 13, 6416–6428
(2005).
77. N. Konishi, Y. Tokimoto, K. Kohra, and H. Fujii, “New laser speckle
flowgraphy system using CCD camera,” Opt. Rev. 9, 163–169 (2002).
78. M. Lee, N. Konishi, and H. Fujii, “Blood flow analysis of skin tissue
under the sacrum using laser speckle flowgraphy,” Opt. Rev. 10, 562–566
(2003).
79. B. Choi, J. C. Ramirez-San-Juan, J. Lotfi, and J. S. Nelson, “Linear
response range characterization and in vivo application of laser speckle
imaging of blood flow dynamics,” J. Biomed. Opt. 11, 041129-1–7 (2006).
80. N. Serov, W. Steenbergen, and F. de Mul, “Prediction of the
photodetector signal generated by Doppler-induced speckle fluctuations:
theory and some validations,” J. Opt. Soc. Am. A 18, 622–639 (2001).
81. V. Rajan, B. Varghese, T. G. van Leeuwen, and W. Steenbergen, “Speckles
in laser Doppler perfusion imaging,” Opt. Lett. 31, 468–470 (2006).
82. V. Rajan, B. Varghese, T. G. van Leeuwen, and W. Steenbergen,
“Influence of tissue optical properties on laser Doppler perfusion
imaging, accounting for photon penetration depth and the laser speckle
phenomenon,” J. Biomed. Opt. 13, 024001-1–9 (2008).
83. N. Konishi and H. Fujii, “Real-time visualization of retinal microcircu-
lation by laser flowgraphy,” Opt. Eng. 34, 753–757 (1995).
84. N. Yokoi, T. Maeda, Y. Shimatani, M. Kyoso, H. Funamizu, and
Y. Aizu, “Improvement of estimation parameter for frame-rate analysis
of blood flow using laser speckle image sensing,” Opt. Lasers Eng. 27,
156–166 (2014).
85. N. Yokoi, J. Sato, Y. Shimatani, M. Kyoso, H. Funamizu, and Y. Aizu,
“Frame-rate analysis of arterial blood flow in human and rat using laser
speckle image sensing,” Opt. Rev. 21, 345–352 (2014).
86. P. Liu, H. Huang, N. Rollins, L. F. Chalakd, T. Jeon, C. Halovanic, and
H. Lu, “Quantitative assessment of global cerebral metabolic rate of
oxygen (CMRO2) in neonates using MRI,” NMR Biomed. 27, 332–340
(2014)
87. K. L. Leenders, A. J. Palmer, N. Quinn, J. C. Clark, G. Firnau, E. S.
Garnett, C. Nahmias, T. Jones, and C. D. Marsden, “Brain dopamine
metabolism in patients with Parkinson’s disease measured with positron
emission tomography,” J. Neurol. Neurosurg. Psychiatry 49, 853–860
(1986).
88. R. L. Buckner, A. Z. Snyder, B. J. Shannon, G. LaRossa, R. Sachs,
A. F. Fotenos, Y. I. Sheline, W. E. Klunk, C. A. Mathis, J. C. Morris, and
M. A. Mintun, “Molecular, structural, and functional characterization of
Alzheimer’s disease: evidence for a relationship between default activity,
amyloid, and memory,” J. Neurosci. 25, 7709–7717 (2005).
89. M. D’Esposito, L. Y. Deouell, and A. Gazzaley, “Alterations in the
BOLD fMRI signal with ageing and disease: a challenge for
neuroimaging,” Nat. Rev. Neurosci. 4, 863–872 (2003).
90. F. Hyder, “Neuroimaging with calibrated FMRI,” Stroke 35, 2635–2641
(2004).
91. K. Dunn, A. Devor, A. M. Dale, and D. Boas, “Spatial extent of oxygen
metabolism and hemodynamic changes during functional activation of
the rat somatosensory cortex,” NeuroImage 27, 279–290 (2005).
92. M. Jones, J. Berwick, D. Johnston, and J. Mayhew, “Concurrent optical
imaging spectroscopy and Laser-Doppler flowmetry: The relationship
between blood flow, oxygenation, and volume in rodent barrel cortex,
NeuroImage 13, 1002–1015 (2001).
93. M. Kohl, U. Lindauer, G. Royl, M. Kuhl, L. Gold, A. Villringer, and
U. Dirnagl, “Physical model for the spectroscopic analysis of cortical
intrinsic optical signals,” Phys. Med. Biol. 45, 3749–3764 (2000).
94. P. B. Jones, H. K. Shin, D. A. Boas, B. T. Hyman, M. A. Moskowitz,
C. Ayata, and A. K. Dunn, “Simultaneous multispectral reflectance
imaging and laser speckle flowmetry of cerebral blood flow and oxygen
370 Chapter 6
metabolism in focal cerebral ischemia,” J. Biomed. Opt. 13(4), 044007

(2008).
95. W. B. Baker, Z. Sun, T. Hiraki, M. E. Putt, T. Durduran, M. Reivich,
A. G. Yodh, and J. H. Greenberg, “Neurovascular coupling varies with
level of global cerebral ischemia in a rat model,” J. Cereb. Blood Flow
Metab. 33(1), 97–105 (2013).
96. M. Gramer, D. Feuerstein, A. Steimers, M. Takagaki, T. Kumagai,
M. Sué, S. Vollmar, M. Kohl-Bareis, H. Backes, and R. Graf, “Device for
simultaneous positron emission tomography, laser speckle imaging and
RGB reflectometry: validation and application to cortical spreading
depression and brain ischemia in rats,” NeuroImage 94, 250–262 (2014).
97. I. Yuzawa, S. Sakadžić, V. J. Srinivasan, H. K. Shin, K. Eikermann-
Haerter, D. A. Boas, and C. Ayata, “Cortical spreading depression
impairs oxygen delivery and metabolism in mice,” J. Cereb. Blood Flow
Metab. 32(2), 376–386 (2012).
98. Y. Aizu, T. Hirata, T. Maeda, I. Nishidate, and N. Yokoi, “Simulta-
neous imaging of blood flow and hemoglobin concentration change in
skin tissue using NIR speckle patterns,” Proc. SPIE 7371, 73711D
(2009).
99. N. Yokoi, Y. Shimatani, M. Kyoso, H. Funamizu, and Y. Aizu, “Imaging
of blood flow and blood concentration change in a frame rate using laser
speckle: Methods for image analysis,” Opt. Laser Tech. 64, 352–362 (2014).
100. N. Yokoi, Y. Shimatani, M. Kyoso, H. Funamizu, and Y. Aizu,
“Improvement of temporal resolution in blood concentration imaging
using NIR speckle patterns,” Proc. SPIE 8798, 87980W (2013).
101. N. Yokoi and Y. Aizu, “Motion imaging of objects in layers hidden by
scattering media using low-coherence speckle interferometry,” Opt.
Laser Technol. 40, 52–57 (2008).
102. C. J. Pedersen, D. Huang, M. A. Shure, and A. M. Rollins,
“Measurement of absolute flow velocity vector using dual-angle, delay-
encoded Doppler optical coherence tomography” Opt. Lett. 32(5),
506–508 (2007).
103. Y. Wang, B. A. Bower, J. A. Izatt, O. Tan, and D. Huang, “In vivo total
retinal blood flow measurement by Fourier domain Doppler optical
coherence tomography” Journal of Biomedical Optics 13(6), 064003
(2008).
104. V. J. Srinivasan, S. Sakadzić, I. Gorczynska, S. Ruvinskaya, W. Wu,
J. G. Fujimoto, and D. A. Boas, “Quantitative cerebral blood flow
with optical coherence tomography,” Optics Express 18(3), 2477–2494
(2010).
105. M. Nagai, K. Matsuda, J. Ohtsubo, K. Homma, and K. Shimizu,
“Microscopic laser Doppler velocimeter for blood velocity measure-
ment,” Opt. Eng. 32(1), 15–20 (1993).
106. P. D. Welch, “Application of the fast Fourier transform to computation

of Fourier integrals, Fourier series, and convolution integrals” IEEE
Trans. Acoust. Audio and Electro acoustics AU-15(2), 70–73 (1967).
107. C. Bernardi, D. Moneta, M. Brughera, M. Di Salvo, D. Lamparelli,
G. Mazu, and M. J. Iatropoulos, “Haematology and clinical chemistry
in rats: comparison of different blood collection sites,” Comp. Haematol.
Int. 6, 160–166 (1996)
108. Y. Sugii, S. Nishio, and K. Okamoto, “In vivo PIV measurement of red
blood cell mesentery motion,” Physiol. Meas. 23, 403–416 (2002).
109. R. Lindken, M. Rossi, S. Große, and J. Westerweel, “Micro-Particle
Image Velocimetry (mPIV): Recent developments, applications, and
guidelines,” Lab. Chip 9, 2551–2567 (2009).
110. M. Shakeria, I. Khodarahmia, M. S. Keith, and A. A. Aminia, “Optical
Imaging of Steady Flow in a Phantom Model of Iliac Artery Stenosis:
Comparison of CFD Simulations with PIV Measurements,” Proc. SPIE
7626, (2010).
111. A. Kazemzadeh, P. Ganesan, F. Ibrahim, S. He, and M. J. Madou, “The
Effect of Contact Angles and Capillary Dimensions on the Burst Frequency
of Super Hydrophilic and Hydrophilic Centrifugal Microfluidic Platforms,
a CFD Study,” PLoS One 8(9), e73002 (2013).
112. R. Lima, S. Wada, M. Takeda, K. Tsubota, and T. Yamaguchi, “In vitro
confocal micro-PIV measurements of blood flow in a square micro-
channel: The effect of the haematocrit on instantaneous velocity
profiles,” J. Biomechanics 40, 2752–2757 (2007).
113. R. Lima, S. Wada, M. Takeda, K. Tsubota, and T. Yamaguchi,
“Confocal micro-PIV measurements of three dimensional profiles of
cell suspension flow in a square microchannel,” Meas. Sci. Techn. 17,
797–808 (2006).
114. P. Vennemann, R. Lindken, and J. Westerweel, “In vivo whole-field blood
velocity measurement techniques,” Exp. Fluids 42, 495–511 (2007).
115. J. A. Chalela, C. S. Kidwell, L. M. Nentwich, M. Luby, J. A. Butman,
A. M. Demchuk, M. D. Hill, N. Patronas, L. Latour, and S. Warach,
“Magnetic resonance imaging and computed tomography in emergency
assessment of patients with suspected acute stroke: a prospective
comparison,” Lancet 369, 293–298 (2007).
116. W. L. Olszewski and A. Tárnok, “Photoacoustic listening of cells in
lymphatics: research art or novel clinical noninvasive lymph test,”
Cytometry A. 73A, 1111–1113 (2008).
117. I. Carr, “Lymphatic metastasis,” Cancer Metastasis Rev 2, 307–317 (1983).
118. W. L. Olszewski, “Atlas of the lymphatics of the lower limbs,” Servier
International. 22Rue Garnier, 92200, Neuilly-sur-Seine, France (2003).
119. S. Y. Wong and R. O. Hynes, “Lymphatic or hematogenous dissemination:
how does a metastatic tumor cell decide?” Cell Cycle 5(8), 812–817 (2006).
372 Chapter 6
120. P. Friedl and B. Weigelin, “Interstitial leukocyte migration and immune

function,” Nat. Immunol. 9, 960–989 (2008).
121. B. Ristevski, H. Becker, M. Cybulsky, T. Seabrook, S. Bak, E. Chan,
M. Johnston, and J. B. Hay, “Lymph, lymphocytes, and lymphatics,”
Immunol. Res. 35, 55–64 (2006).
122. M. D. Cahalan and I. Parker, “Imaging the choreography of lymphocyte
trafficking and the immune response,” Curr. Opin. Immunol. 18, 476–482
(2006).
123. C. R. Mackay, “Moving targets: cell migration inhibitors as new anti-
inflammatory therapies,” Nat. Immunol. 9, 988–998 (2008).
124. L. A. Johnson, S. Clasper, A. P. Holt, P. F. Lalor, D. Baban, and DG.
Jackson “An inflammation induced mechanism for leukocyte transmi-
gration across lymphatic vessel endothelium,” J. Exp. Med. 203, 2763–
2777 (2006).
125. J. L. Gowans and H. W. Steer, “The function and pathways of
lymphocyte recirculation,” Ciba. Found. Symp. 71, 113–126 (1980).
126. G. Azzali, M. L. Arcari, and G. F. Caldara. “The “mode” of lymphocyte
extravasation through HEV of Peyer’s patches and its role in normal
homing and inflammation,” Microvasc. Res. 75, 227–237 (2008).
127. D. G. Jackson, “Biology of the lymphatic marker LYVE-1 and
applications in research into lymphatic trafficking and lymphangiogen-
esis,” APMIS. 112, 526–538 (2004).
128. G. Garcia-Tsao, F. Y. Lee, G. E. Barden, R. Cartun, and A. B. West,
Bacterial translocation to mesenteric lymph nodes is increased in
cirrhotic rats with ascites, “Gastroenterology 108, 1835–1841.
129. E. I. Galanzha, E. V. Shashkov, V. V. Tuchin, and V. P. Zharov,
“In vivo multiparameter, multispectral lymph flow cytometry with
natural cell focusing, label-free detection and multicolor nanoparticle
probes,” Cytometry A. 73A, 884–894 (2008).
130. E. I. Galanzha, M. S. Kokoska, E. V. Shashkov, J-W Kim, V. V.
Tuchin, and V. P. Zharov, “In vivo fiber-based multicolor photo-
acoustic detection and photothermal purging of metastasis in sentinel
lymph nodes targeted by nanoparticles,” J. Biophotonics 2, 528–539
(2009).
131. J. P. Sleeman, I. Nazarenko, and W. Thiele, “Do all roads lead to Rome?
Routes to metastasis development,” Int. J. Cancer 128, 2511–2526
(2011).
132. S. Podgrabinska and M. Skobe, “Role of lymphatic vasculature in
regional and distant metastases,” Microvasc. Res. 95, 46–52 (2014).
133. B. Fisher and E. R. Fisher, “The interrelationship of hematogenous and
lymphatic tumor cell dissemination,” Surg. Gynecol. Obstet. 122, 791–
798 (1966).
134. K. Hayashi, K. Yamauchi, N. Yamamoto, H. Tsuchiya, K. Tomita, and

R. M. Hoffman, “Real-time imaging of tumor-cell shedding and
trafficking in lymphatic channels,” Cancer Res. 67, 8223–8228 (2007).
135. P. O. Van Trappen and M. S. Pepper, “Lymphatic dissemination of
tumour cells and the formation of micrometastases,” Lancet Oncol. 3,
44–52 (2002).
136. Y. Hüsemann, J. B. Geigl, F. Schubert, P. Musiani, M. Meyer, G. Forni,
R. Eils, T. Fehm, G. Riethmüller, and C. A. Klein, “Systemic spread is
an early step in breast cancer,” Cancer Cell. 13, 58–68 (2008).
137. J. Carr, I. Carr, B. Dreher, and K. Betts, “Lymphatic metastasis:
invasion of lymphatic vessels and efflux of tumour cells in the afferent
popliteal lymph,” J. Pathol. 132, 287–305 (1980).
138. R. R. Langley and I. J. Fidler, “Tumor cell-organ microenvironment
interactions in the pathogenesis of cancer metastasis,” Endocr. Rev. 28,
297–321 (2007).
139. G. Azzali, “Tumor cell transendothelial passage in the absorbing
lymphatic vessel of transgenic adenocarcinoma mouse prostate,” Am.
J. Pathol. 170(1), 334–346 (2007).
140. E. I. Galanzha, “Blood and lymph circulating cells: well-known systems,
well-forgotten interdependence,” (Editorial). J. Blood Lymph 1, 1–2.
doi: 10.4172/2165-7831.1000e104. Free full text at: http://www.omicsgroup.
org/journals/2165-7831/2165-7831-1-e104.digital/2165-7831-1-e104.html
141. V. Zharov, E. Galanzha, and V. Tuchin, “Photothermal imaging of
moving cells in lymph and blood flow in vivo animal model,” Proc SPIE
5320, 256–263 (2004).
142. V. P. Zharov, E. I. Galanzha, E. V. Shashkov, J. W. Kim, N. G.
Khlebtsov, and V. V. Tuchin, “Photoacoustic flow cytometry: principle
and application for real-time detection of circulating single nanoparticles,
pathogens, and contrast dyes in vivo,” J. Biomed. Opt. 12, 051503 (2007).
143. V. V. Tuchin, A. Tarnok, and V. P. Zharov, “In vivo flow cytometry:
A horizon of opportunities,” Cytometry A 79A, 737–745 (2011).
144. E. I. Galanzha, V. V. Tuchin, and V. P. Zharov, “Advances in small
animal mesentery models for in vivo flow cytometry, dynamic microscopy,
and drug screening (review),” World J. Gastroenterol. 13, 192–218 (2007).
145. L. S. Sasportas and S. S. Gambhir, “Imaging circulating tumor cells in
freely moving awake small animals using a miniaturized intravital
microscope,” PLoS One 9, e86759 (2014).
146. V. Kalchenko, A. Harmelin, I. Fine, V. Zharov, E. Galanzha, and
V. Tuchin, “Advances in intravital microscopy for monitoring cell flow
dynamics in vivo,” Proc. SPIE 6436, (2007).
147. I. V. Fedosov, S. S. Ulyanov, E. I. Galanzha, V. A. Galanzha, and V. V.
Tuchin, in Coherent-Domain Optical Methods: Biomedical Diagnostics,
374 Chapter 6
Environmental and Material Science 1, Chapter 10 (Ed. V. V. Tuchin),

Kluwer Academic Publishers, Boston, MA, 397–435 (2001).
148. E. I. Galanzha, P. Chowdhury, V. V. Tuchin, and V. P. Zharov,
“Monitoring of nicotine impact on microlymphatics of rat mesentery
with time-resolved microscopy,” Lymphology 38, 181–192 (2005).
149. J. B. Dixon, D. C. Zawieja, A. A. Gashev, and G. L. Coté, “Measuring
microlymphatic flow using fast video microscopy,” J. Biomed. Opt. 10,
064016 (2005).
150. S. A. Japee, R. N. Pittman, and C. G. Ellis, “A new video image analysis
system to study red blood cell dynamics and oxygenation in capillary
networks,” Microcirculation 12, 489–506 (2005).
151. D. Kedrin, B. Gligorijevic, J. Wyckoff, V. V. Verkhusha,
J. Condeelis, J. E. Segall, and J. van Rheenen, “Intravital imaging of
metastatic behavior through a mammary imaging window,” Nature
Methods 5, 1019–1021 (2008).
152. S. Langer, F. Born, R. Hatz, P. Biberthaler, and K. Messmer,
“Orthogonal polarization spectral imaging versus intravital fluorescent
microscopy for microvascular studies in wounds,” Ann. Plast. Surg. 48,
646–653 (2002).
153. J. Novak, I. Georgakoudi, X. Wei, A. Prossin, and C. P. Lin, “In vivo
flow cytometer for real-time detection and quantification of circulating
cells,” Opt. Lett. 29, 77–79 (2004).
154. I. Georgakoudi, N. Solban, J. Novak, W. L. Rice, X. Wei, T. Hasan,
and C. P. Lin, “In vivo flow cytometry: a new method for enumerating
circulating cancer cells,” Cancer Res. 64, 5044–5047 (2004).
155. C. M. Pitsillides, J. M. Runnels, J. A. Spencer, L. Zhi, M. X. Wu, and
C. P. Lin, “Cell labeling approaches for fluorescence-based in vivo flow
cytometry,” Cytometry A 79, 758–765 (2011).
156. D. Hwu, S. Boutrus, C. Greiner, T. DiMeo, C. Kuperwasser, and
I. Georgakoudi, “Assessment of the role of circulating breast cancer cells
in tumor formation and metastatic potential using in vivo flow
cytometry,” J. Biomed. Opt. 16, 040501 (2011).
157. W. He, H. Wang, L. C. Hartmann, J. X. Cheng, and P. S. Low,
“In vivo quantitation of rare circulating tumor cells by multiphoton
intravital flow cytometry,” Proc. Natl. Acad. Sci. USA 104, 11760–11765
(2007).
158. S. Boutrus, C. Greiner, D. Hwu, M. Chan, C. Kuperwasser, C. P. Lin,
and I. Georgakoudi, “Portable two-color in vivo flow cytometer for real-
time detection of fluorescently-labeled circulating cells,” J. Biomed. Opt.
12, 020507 (2007).
159. Y. Li, J. Guo, C. Wang, Z. Fan, G. Liu, C. Wang, Z. Gu, D. Damm,
A. Mosig, and X. Wei, “Circulation times of prostate cancer and
hepatocellular carcinoma cells by in vivo flow cytometry,” Cytometry A
79, 848–854 (2011).
160. Z. C. Fan, J. Yan, G. D. Liu, X. Y. Tan, X. F. Weng, W. Z. Wu,

J. Zhou, and X. B. Wei, “Real-time monitoring of rare circulating hepato-
cellular carcinoma cells in an orthotopic model by in vivo flow cytometry
assesses resection on metastasis,” Cancer Res. 72, 2683–2691 (2012).
161. M. A. Juratli, M. Sarimollaoglu, D. A. Nedosekin, A. V. Melerzanov,
V. P. Zharov, and E. I. Galanzha, “Dynamic Fluctuation of Circulating
Tumor Cells during Cancer Progression,” Cancers (Basel) 6(1), 128–142
(2014).
162. R. M. Hoffman, “Orthotopic mouse models of tumor metastasis
expressing fluorescent reporters produce imageable circulating tumor
cells,” Cancer Microenviron. 7, 133–138 (2014).
163. T. P. Padera, B. R. Stoll, P. T. So, and R. K. Jain, “Conventional and
high-speed intravital multiphoton laser scanning microscopy of micro-
vasculature, lymphatics, and leukocyte-endothelial interactions,” Mol.
Imaging 1, 9–15 (2002).
164. M. Rubart, Two-photon microscopy of cells and tissue, “ Circ. Res. 95,
1154–1166 (2004).
165. E. R. Tkaczyk, C. F. Zhong, and J. Y. Ye et al.. “In vivo monitoring of
multiple circulating cell populations using two-photon flow cytometry,”
Opt. Commun. 281, 888–894 (2008).
166. C. F. Zhong, E. R. Tkaczyk, and T. Thomas et al. “Quantitative
two-photon flow cytometry– in vitro and in vivo,” J. Biomed. Opt 13,
(2008).
167. E. R. Tkaczyk, A. H. Tkaczyk, and S. Katnik et al. “Extended cavity
laser enhanced two-photon flow cytometry,” J. Biomed. Opt. 14,
(2009).
168. E. I. Galanzha, M. Sarimollaoglu, D. A. Nedosekin, S. G. Keyrouz,
J. L. Mehta, and V. P. Zharov, “In vivo flow cytometry of circulating
clots using negative phototothermal and photoacoustic contrasts,”
Cytometry 79A, 814–824 (2011).
169. E. I. Galanzha and V. P. Zharov, “Circulating tumor cell detection and
capturing using photoacoustic flow cytometry in vivo and ex vivo
(review),” Cancers 5, 1691–1738 (2013).
170. V. P. Zharov, E. I. Galanzha, and V. V. Tuchin, “Photothermal Image
Flow Cytometry in Vivo,” Opt. Lett. 30, 628–630 (2005).
171. V. P. Zharov, E. I. Galanzha, and V. V. Tuchin “In vivo photothermal
flow cytometry: imaging and detection of individual cells in blood and
lymph flow,” J. Cell Biochem. 97(5), 916–930 (2006).
172. D. A. Nedosekin, E. I. Galanzha, E. Dervishi, A. S. Biris, and V. P.
Zharov, “Super-resolution nonlinear photothermal microscopy,” Small
10(1), 135–42 (2014).
173. E. I. Galanzha and V. P. Zharov, “In vivo photoacoustic and
photothermal cytometry for monitoring multiple blood rheology
parameters (review),” Cytometry A 79(10), 746–757 (2011).
376 Chapter 6
174. D. A. Nedosekin, M. Sarimollaoglu, E. I. Galanzha, R. Sawant, V. P.

Torchilin, V. V. Verkhusha, J. Ma, M. H. Frank, A. S. Biris, and V. P.
Zharov, “Synergy of photoacoustic and fluorescence flow cytometry of
circulating cells with negative and positive contrasts,” J. Biophotonics
6(5), 425–34 (2013).
175. D. A. Nedosekin, V. V. Verkhusha, A. V. Melerzanov, V. P. Zharov,
and E. I. Galanzha, “In vivo photoswitchable flow cytometry for direct
tracking of single circulating tumor cells,” Chem. Biol. 21, 792–801
(2014).
176. S. R. Taylor and J. B. Jorgensen, “Use of fluorescent angiography to
assess donor site perfusion prior to free tissue transfer,” Laryngoscope
(2015 Feb 13). doi: 10.1002/lary.25190. [Epub ahead of print]
177. M. R. Zenn, “Fluorescent angiography,” Clin. Plast. Surg. 125(6),
E192–E197 (2015).
178. K. Waseda, P. J. Fitzgerald, and M. Takahashi, “Intraoperative
assessment of coronary grafts with fluorescent angiography,” BMJ Case
Rep. bcr2006109421 (2009).
179. Y. Hama, Y. Koyama, Y. Urano, P. L. Choyke, and H. Kobayashi,
“Simultaneous two-color spectral fluorescence lymphangiography with
near infrared quantum dots to map two lymphatic flows from the breast
and the upper extremity,” Breast Cancer Res. Treat. 103, 23–28 (2007).
180. T. Yamamoto, H. Yoshimatsu, M. Narushima, N. Yamamoto,
A. Hayashi, and I. Koshima, “Indocyanine green lymphography
findings in primary leg lymphedema,” Eur. J. Vasc. Endovasc. Surg.
49(1), 95–102 (2015). doi: 10.1016/j.ejvs.2014.10.023.
181. C. Hirche, H. Engel, Z. Hirche, S. Doniga, T. Herold, U. Kneser,
M. Lehnhardt, and M. Hünerbein, “Real-time lymphography by
indocyanine green fluorescence: improved navigation for regional lymph
node staging,” Ann. Plast. Surg. 73(6), 701–5 (2014).
182. N. Unno, K. Inuzuka, M. Suzuki, N. Yamamoto, D. Sagara,
M. Nishiyama, and H. Konno, “Preliminary experience with a novel
fluorescence lymphography using indocyanine green in patients with
secondary lymphedema,” J. Vasc. Surg. 45(5), 1016–1021 (2007) Epub
2007 Mar 28.
183. B. Zhu and E. M. Sevick-Muraca, “A review of performance of near-
infrared fluorescence imaging devices used in clinical studies,” Br. J.
Radio l88(1045), 20140547 (2015).
184. N. Ono, R. Mizuno, H. Nojiri, and T. Ohhashi “Development of an
experimental apparatus for investigating lymphatic pumping activity of
murine mesentery in vivo,” Jap. J. Physiol. 50, 25–31 (2000).
185. M. J. Sanz, B. Johnston, A. Issekutz, and P. Kubes, “Endothelin-1
causes P-selectin-dependent leukocyte rolling and adhesion within rat
mesenteric microvessels,” Am. J. Physiol. 277, H1823–H1830 (1999).
186. K. Ley, G. Linnemann, M. Meinen, L. M. Stoolman, and P. Gaehtgens,

“Fucoidin, but not yeast polyphosphomannan PPME, inhibits leuko-
cyte rolling in venules of the rat mesentery,” Blood 81, 177–185
(1993).
187. J. N. Benoit, “Relationship between lymphatic pump flow and total
lymph flow in the small intestine,” Am. J. Physiol. 261, H1970–H1978
(1991).
188. Y. u. Shirasawa, F. Ikomi, and T. Ohhashi, “Physiological roles of
endogenous nitric oxide in lymphatic pump activity of rat mesentery
in vivo,” Am. J. Physiol. 278, G551–G556 (2000).
189. E. Sekizuka, C. Ohshio, and H. Minamitani, “Automatic analysis of
moving images for the lymphocyte velocity measurement,” Microcircu-
lation Annual, M. Tsuchiya, M. Asano, and A. Kamiya, (Eds.), 107–108
(1995).
190. G. Horstick, T. Kempf, M. Lauterbach, M. Ossendorf, L. Kopacz,
A. Heimann, H. A. Lehr, S. Bhakdi, J. Meyer, and O. Kempski,
“Plastic foil technique attenuates inflammation in mesenteric intravi-
tal microscopy,” J. Surg. Res. 94, 28–34 (2000).
191. V. V. Tuchin, “Optical clearing of tissue and blood using the immersion
192. V. V. Tuchin, “Optical immersion as a new tool for controlling the
optical properties of tissues and blood,” Laser Physics 15(8), 1109–1136
(2005).
193. Y. A. Menyaev, D. A. Nedosekin, M. Sarimollaoglu, M. A. Juratli, E. I.
Galanzha, V. V. Tuchin, and V. P. Zharov, “Optical clearing in
photoacoustic flow cytometry,” Biomed. Opt. Express 4(12), 3030–3041
(2013 Nov 27).
194. V. Zharov, E. Galanzha, E. Shashkov, N. Khlebtsov, and V. Tuchin,
“In vivo photoacoustic flow cytometry for monitoring circulating singe
cancer cells and contrast agents,” Opt. Lett. 31, 3623–3625 (2006).
195. H. M. Shapiro, Practical Flow Cytometry, 4th ed., Wiley-Liss, New
York (2003).
196. Y. Zeng, B. Yan, Q. Sun, S. He, J. Jiang, Z. Wen, and J. Y. Qu, “In vivo
micro-vascular imaging and flow cytometry in zebrafish using two-
photon excited endogenous fluorescence,” Biomed. Opt. Express 5(3),
653–63 (2014).
197. K. Stoletov, V. Montel, R. D. Lester, S. L. Gonias, and R. Klemke
“High-resolution imaging of the dynamic tumor cell vascular interface in
transparent zebrafish,” Proc. Natl. Acad. Sci. USA 104(44), 17406–11
2007).
198. V. Schacht, D. Berens von Rautenfeld, and C. Abels, “The lymphatic
system in the dorsal skinfold chamber of the Syrian golden hamster
in vivo,” Arch. Dermatol. Res. 295, 542–548 (2004).
378 Chapter 6
199. M. de Jong, J. Essers, and W. M. van Weerden, “Imaging preclinical

tumour models: improving translational power,” Nat. Rev. Cancer 14,
481–493 (2014).
200. L. Hennighausen, “Mouse models for breast cancer,” Breast Cancer Res.
2(1), 2–7 (2000).
201. P. D. Bos, D. X. Nguyen, and J. Massagué, “Modeling metastasis in the
mouse,” Curr. Opin. Pharmacol. 10, 571–577 (2010).
202. T. Barrett, P. L. Choyke, and H. Kobayashi, “Imaging of the lymphatic
system: new horizons,” Contrast Media Mol. Imaging 1, 230–245 (2006).
203. H. Kobayashi, Y. Hama, Y. Koyama, T. Barrett, C. A. Regino,
Y. Urano, and P. L. Choyke, “Simultaneous multicolor imaging of five
different lymphatic basins using quantum dots,” Nano. Lett. 7, 1711–
1716 (2007).
204. S. Kim, Y. T. Lim, E. G. Soltesz, A. M. De Grand, J. Lee,
A. Nakayama, J. A. Parker, T. Mihaljevic, R. G. Laurence, D. M. Dor,
L. H. Cohn, M. G. Bawendi, and J. V. Frangioni, “Near-infrared
fluorescent type II quantum dots for sentinel lymph node mapping,”
Nat. Biotechnol. 22, 93–97 (2004).
205. R. Guimaraes, O. Clément, J. Bittoun, F. Carnot, and G. Frija, “MR
lymphography with superparamagnetic iron nanoparticles in rats:
pathologic basis for contrast enhancement,” Am. J. Roentgenol. 162(1),
201–207 (1994).
206. M. G. Harisinghani, W. T. Dixon, M. A. Saksena, E. Brachtel, D. J.
Blezek, P. J. Dhawale, M. Torabi, and P. F. Hahn, “MR lymphangiog-
raphy: imaging strategies to optimize the imaging of lymph nodes with
ferumoxtran-10,” Radiographics 24, 867–878 (2004).
207. F. Réty, O. Clément, N. Siauve, C. A. Cuénod, F. Carnot, M. Sich,
A. Buisine, and G. Frija, “MR lymphography using iron oxide
nanoparticles in rats: pharmacokinetics in the lymphatic system after
intravenous injection,” J. Magn. Reson. Imaging 12, 734–739 (2000).
208. R. A. Heesakkers, A. M. Hövels, G. J. Jager, H. C. van den Bosch, J. A.
Witjes, H. P. Raat, J. L. Severens, E. M. Adang, C. H. van der Kaa, J. J.
Fütterer, and J. Barentsz, “MRI with a lymph-node-specific contrast
agent as an alternative to CT scan and lymph-node dissection in patients
with prostate cancer: a prospective multicohort study,” Lancet Oncol. 9,
850–856 (2008).
209. K. Yamashita and K. Shimizu, “Video-assisted breast surgery and
sentinel lymph node biopsy guided by three-dimensional computed
tomographic lymphography,” Surg. Endosc. 22, 392–397 (2008).
210. R. D. White, J. R. Weir-McCall, M. J. Budak, S. A. Waugh, D. A.
Munnoch, and T. A. Sudarshan, “Contrast-enhanced magnetic reso-
nance lymphography in the assessment of lower limb lymphoedema,”
Clin. Radiol. 69(11), e435–444 (2014).
211. C. Li, S. Meng, X. Yang, D. Zhou, J. Wang, and J. Hu, “Sentinel lymph
node detection using magnetic resonance lymphography with conven-
tional gadolinium contrast agent in breast cancer: a preliminary clinical
study,” BMC Cancer 15, 213 (2015).
212. M. Nakagawa, M. Morimoto, H. Takechi, Y. Tadokoro, and
A. Tangoku, “Preoperative diagnosis of sentinel lymph node (SLN)
metastasis using 3D CT lymphography (CTLG),” Breast Cancer. [Epub
ahead of print] PubMed PMID: 25814093 (2015).
213. C. Shah, D. Arthur, J. Riutta, P. Whitworth, and F. A. Vicini, “Breast-
cancer related lymphedema: a review of procedure-specific incidence
rates, clinical assessment aids, treatment paradigms, and risk reduction,”
Breast J. 18, 357–361 (2012).
214. N. L. Browse and G. Stewart, “Lymphoedema: pathophysiology and
classification,” J. Cardiovasc. Surg. (Torino) 26, 91–106 (1985).
215. “The diagnosis and treatment of peripheral lymphedema,” Consensus
Document of the International Society of Lymphology. Lymphology 36,
84–91 (2003).
216. A. A. Ramelet, “Pharmacologic aspects of a phlebotropic drug in CVI-
associated edema,” Angiology 51, 19–23 (2000).
217. P. A. Hurst and J. M. Edwards, “Chylous ascites and obstructive
lymphoedema of the small bowel following abdominal radiotherapy,”
Br. J. Surg. 66, 780–781 (1979).
218. R. L. Ahmed, K. H. Schmitz, A. E. Prizment, and A. R. Folsum,
“Risk factors for lymphedema in breast cancer survivors, the Iowa
Women’s Health Study,” Breast Canc. Res. Treatment 130, 981–991
(2011).
219. J. L. Bevilacqua, M. W. Kattan, Y. Changhong, S. Koifman, I. E.
Mattos, R. J. Koifman, and A. Bergmann, “Nomograms for predicting
the risk of arm lymphedema after axillary dissection in breast cancer,”
Ann. Surg. Oncol. 19, 2580–2589 (2012).
220. S. A. Norman, A. R. Localio, M. J. Kallan, A. L. Weber, H. A. Simoes
Torpey, S. L. Potashnik, L. T. Millers, K. R. Fox, A. DeMichele, and
L. J. Solins, “Risk factors for lymphedema after breast cancer
treatment,” Cancer Epidemiol. Biomarkers Prev. 19, 2734–2746
(2010).
221. E. I. Galanzha, V. V. Tuchin, and V. P. Zharov, “Optical monitoring of
microlympatic disturbances at experimental lymphedema,” Lymphat.
Res. Biol. 5, 11–27 (2007).
222. Report of Surgeon General: The Health Consequences of Smoking.
Department of Health & 13 Human Services, Public Health Services,
Centers for Disease Control & Prevention, National Center for Chronic
Disease Prevention and Health Promotion, Office of Smoking & Health,
Washington, D.C. (2004).
380 Chapter 6
223. P. Chowdhury, P. L. Rayford, and L. W. Chang, “Induction of acinar

cell pathology via inhalation of nicotine,” Proc. Soc. Exp. Biol. Med. 20
159–164 (1992).
224. A. K. Armitage, C. T. Dollery, and C. F. George et al., “Absorption
and metabolism of nicotine from cigarettes,” Br. Med. J. 4 313–316
(1975).
225. R. Mizuno, A. Koller, and G. Kaley, “Regulation of the vasomotor
activity of lymph microvessels by nitric oxide and prostaglandins,” Am.
J. Physiol. 274(3 Pt 2), R790–R796 (1998).
226. Y. Shirasawa, F. Ikomi, and T. Ohhashi, “Physiological roles of endoge-
nous nitric oxide in lymphatic pump activity of rat mesentery in vivo,” Am.
J. Physiol. Gastrointest Liver Physiol. 278(4), G551 (2000).
227. J. P. Scallan, M. A. Hill, and M. J. Davis, “Lymphatic vascular integrity
is disrupted in type 2 diabetes due to impaired nitric oxide signaling,”
Cardiovasc Res. 107, 89–97. [Epub ahead of print]
228. E. I. Galanzha, G. E. Brill, A. V. Solov’eva, and A. V. Stepanova,
“Nitric oxide in the lymphatic microvessel regulation,” Ross. Fiziol. Zh.
im. I.M Sechenova 88(8), 983–989 (2002).
229. K. Aukland, “Arnold Heller and the lymph pump,” Acta Physiol. Scand.
185, 171–180 (2005).
230. C. M. Hogg, O. Reid, and R. J. Scothorne, “Studies on hemolymph
nodes. III. Renal lymph as a major source of erythrocytes in the
renalhemolymph node of rats,” J. Anat. 135, 291–299 (1982).
231. M. F. Abu-Hijleh and R. J. Scothorne, “Studies on haemolymph nodes.
IV. Comparison of the route of entry of carbon particles into parathymic
nodes after intravenous and intraperitoneal injection,” J. Anat. 188,
565–573 (1996).
232. W. Andrade, M. G. Johnston, and J. B. Hay, “The exit of lymphocytes
and RBCs from the peritoneal cavity of sheep,” Immunobiology 195,
77–90 (1996).
233. A. Engeset, J. Sokolowski, and W. L. Olszewski, “Variation in output of
leukocytes and erythrocytes in human peripheral lymph during rest and
activity,” Lymphology 10, 198–203 (1977).
234. J. E. Skandalakis, L. J. Skandalakis, and P. N. Skandalakis, “Anatomy
of the lymphatics,” Surg. Oncol. Clin. N. Am. 16, 1–16 (2007).
235. G. W. Schmid-Schonbein, “Microlymphatice and lymph flow,” The Am.
Physiol. Soc. 70, 987–1028 (1990).
236. K. Aukland and R. K. Reed, “Physiological Reviews. Interstitial-
lymphatic mechanisms in the control of extracellular fluid volume,” The
Am. Physiol. Soc. 73, 1–78 (1993).
237. M. Jeltsch, T. Tammela, K. Alitalo, and J. Wilting, “Genesis and
pathogenesis of lymphatic vessels,” Cell Tissue Res. 314, 69–84 (2003).
238. J. R. Casley-Smith, “The functioning and interrelationships of blood
capillaries and lymphatics,” Experientia 32, 1–12 (1976).
239. M. F. Abu-Hijleh and R. J. Scothorne, “Studies on haemolymph nodes.

IV. Comparison of the route of entry of carbon particles into parathymic
nodes after intravenous and intraperitoneal injection,” J. Anat. 188, 565–
73 (1996).
240. W. Andrade, M. G. Johnston, and J. B. Hay, “The exit of lymphocytes
and RBCs from the peritoneal cavity of sheep,” Immunobiology 195, 77–
90 (1996).
241. I. Aursnes, “Appearance of red cells in peripheral lymph during
radiation-induced thrombocytopenia,” Acta Physiol. Scand. 88, 392–
400 (1973).
242. S. Levine and A. Saltzman, “Retrosternal hemorrhage: an experimental
model for study of lymphatic leakage,” Lymphology 21, 105–9 (1988).
243. M. Oehmichen and V. Schmidt, “Erythrocytes in cervical lymph nodes
of the human as a sequel of stasis and/or lymph drainage. Questionable
diagnostic significance in strangulation and mechanical injuries of the
head,” Z. Rechtsmed. 103, 33–41 (1989).
244. G. Christofori, “New signals from the invasive front,” Nature 441, 444–
450 (2006).
245. C. A. Klein, “Parallel progression of primary tumours and metastases,”
Nat. Rev. Cancer. 9, 302–12 (2009).
246. P. M. Schlag and V. S. Verone, “Lymphatic Metastasis and Sentinel
Lymphonodectomy,” Springer (1986).
247. A. F. Chambers, A. C. Groom, and I. C. MacDonald, “Dissemination
and growth of cancer cells in metastatic sites,” Nat. Rev. Cancer. 2, 563–
72 (2002).
248. F. J. Gujam, J. J. Going, J. Edwards, Z. M. Mohammed, and D. C.
McMillan, “The role of lymphatic and blood vessel invasion in
predicting survival and methods of detection in patients with primary
operable breast cancer,” Crit. Rev. Oncol Hematol. 89, 231–41 (2014).
249. S. Mandl, C. Schimmelpfennig, M. Edinger, R. S. Negrin, and C. H.
Contag, “Understanding immune cell trafficking patterns via in vivo
bioluminescence imaging,” J. Cell Biochem Suppl. 39, 239–48 (2002).
250. A. Bikoue, G. Janossy, and D. Barnett, “Stabilised cellular immuno-
fluorescence assay: CD45 expression as a calibration standard for human
leukocytes,” J. Immunol Methods. 266, 19–32 (2002).
251. L. S. Pelan-Mattocks, B. A. Pesch, and M. E. Kehrli, Jr., “Flow
cytometric analysis of intracellular complexity and CD45 expression for
use in rapid differentiation of leukocytes in bovine blood samples,” Am.
J. Vet. Res. 62, 1740–1744 (2001).
252. X. Wei, D. A. Sipkins, C. M. Pitsillides, J. Novak, I. Georgakoudi, and
C. P. Lin, “Real-time detection of circulating apoptotic cells by in vivo
flow cytometry,” Mol. Imaging. 4(4), 415–6 (2005 Oct–Dec).
253. R. Siegel, J. Ma, Z. Zou, and A. Jemal, ‘Cancer statistics, 2014,” CA
Cancer J. Clin. 64, 9–29 (2014).
382 Chapter 6
254. J. Scott, P. Kuhn, and A. R. Anderson, “Unifying metastasis—

integrating intravasation, circulation and end- organ colonization,”
Nat. Rev. Cancer. 12, 445–446 (2012).
255. T. Shibue and R. A. Weinberg, “Metastatic colonization: settlement,
adaptation and propagation of tumor cells in a foreign tissue
environment,” Semin. Cancer Biol. 21, 99–106 (2011).
256. I. J. Fidler, “The pathogenesis of cancer metastasis: the ’seed and soil’
hypothesis revisited,” Nat. Rev. Cancer 3, 453–458 (2003).
257. T. Shibue and R. A. Weinberg, “Metastatic colonization: settlement,
adaptation and propagation of tumor cells in a foreign tissue
environment,” Semin. Cancer Biol. 21, 99–106 (2011).
258. D. X. Nguyen, P. D. Bos, and J. Massagué, “Metastasis: from dissemina-
tion to organ-specific colonization,” Nat. Rev. Cancer 9, 274–284
(2009).
259. B. Rack, C. Schindlbeck, J. Jückstock, U. Andergassen, P. Hepp,
T. Zwingers, T. W. Friedl, R. Lorenz, H. Tesch, P. A. Fasching, T. Fehm,
A. Schneeweiss, W. Lichtenegger, M. W. Beckmann, K. Friese, K. Pantel,
and W. Janni; SUCCESS Study Group, “Circulating tumor cells predict
survival in early average-to-high risk breast cancer patients,” J. Natl.
Cancer Inst. 106(5). pii: dju066 (2014). Erratum in J. Natl. Cancer Inst.
(2014) 106(9): doi/10.1093/jnci/dju273.
Ivan V. Fedosov has been an associate professor of the

Department of Optics and Biophotonics of Saratov State
University since 2004. He received his PhD degree in bio-
photonics from Saratov State University in 2002. His current
research activities are in the field of biomedical imaging, laser
based blood flow measurements, super-resolution microscopy,
micro-anemometry, and optical micromanipulation.
Yoshihisa Aizu received his Dr. Eng. in electronics from

Hokkaido University in 1985. From 1985 to 1989, he was
with Kowa Company Limited, Tokyo. In 1989, he joined
Hokkaido University as a research associate. In 1990, he was
an associate professor at Muroran Institute of Technology,
and in 2006 he was promoted to professor. From 1992 to
1993, he was a visiting researcher in University Erlangen,
Germany. His current research activities are in the fields of
biomedical applications of spectroscopy, optical imaging, colorimetry, laser
speckle imaging, and laser light scattering. He is a member of SPIE and OSA.

and OSA, and has been awarded Honored Science Worker of the Russia, SPIE
Educator Award, FiDiPro (Finland), Chime Bell Prize of Hubei Province
(China), and Joseph W. Goodman Book Writing Award (OSA/SPIE).
Naomichi Yokoi received a Dr. Eng in mechanical engineer-

ing from Muroran Institute of Technology in 1999. From
2000 to 2001, he was in Muroran Institute of Technology as
a research fellow of the Japan Society for the Promotion of
Science. In 2002, he joined Asahikawa National College of
Technology as a research associate, and in 2005 he was
promoted to an associate professor. His current research
activities are in the fields of biomedical applications of
optical imaging, laser speckle imaging, laser light scattering, optical particle
sizing, and optical particle manipulation. He is a member of OSA.
Izumi Nishidate is an associate professor in the Graduate

School of Bio-Applications and Systems Engineering, Tokyo
University of Agriculture and Technology. He received his
MS and PhD degrees from Muroran Institute of Technology,
Japan. His research interests include diffuse reflectance
spectroscopy, light transport in biological tissues, multispec-
tral imaging, and functional imaging of skin and brain tissues.
Vladimir P. Zharov is the director of the Arkansas

Nanomedicine Center and a Professor of Biomedical
Engineering at the University of Arkansas for Medical
Sciences, USA. He received his PhD and DSc degrees from
the Bauman Moscow State Technical University (BMSTU),
completed a postdoctoral fellowship at Lawrence Berkeley
National Laboratory of the University of California and
served as the Chairman of the Biomedical Engineering
Department at BMSTU. He is the author of 5 books,
384 Chapter 6
52 patents, and more than 200 papers in the field of laser spectroscopy,
biophotonics, and nanomedicine including five publications in the Nature
family journals. Dr. Zharov has been the principle investigator on 16 NIH,
NSF, DoD and other agency grants including 5 R01 awards. He is one of the
pioneers of high resolution photoacoustic spectroscopy and the inventor of
photoacoustic tweezers, pulse nanophotothermolysis of infections and cancer,
and in vivo multicolor flow cytometry for detection, magnetic capturing, and
photothermal elimination of circulating tumor cells (CTCs), bacteria and
virus with the unprecedented sensitivity down to a few CTCs (1 CTC/500 mL)
in whole blood volume (5 liter in adult). These technologies can provide
breakthroughs in the diagnosis of cancer, infections, and cardiovascular
disorders at an early stage when well-timed therapy is more effective. Dr.
Zharov is the State Prize Winner, the most prestigious national award in
Russia, and the first recipient of the U.S. Maiman Award, named after the
inventor of the first laser.
Ekaterina I. Galanzha is an Associate Professor in the

University of Arkansas for Medical Sciences (UAMS), USA.
She received her MD, PhD, and DSc degrees from Saratov
Universities in Russia. Dr. Galanzha has interdisciplinary
skills in medicine, biology, and biomedical engineering and
her interdisciplinary expertise includes: single cell analysis of
circulating cells in vivo, experimental medicine, biophotonics,
and nanobiotechnology with a focus on lymphatic and
cancer research. Early in her career, she discovered mechanisms of lymphatic
disturbances in an animal model of lymphedema and staphylococcal infection
induced by S. aureus. Later, her studies focused on circulating tumor cells
(CTCs) and lymphatic-related mechanisms of cancer metastasis in vivo.
Dr. Galanzha is the coauthor of 4 book chapters on in vivo flow cytometry
and bioimaging and 56 peer-reviewed papers in high-impact journals (Nature
Nanotechnology, Proceedings of the National Academy of Sciences U S A,
Cancer Research, Nano Letters and Chemistry & Biology [Cell publishing
group journal]). Her work has been featured in several respected sources,
including the NIBIB’s e-Advances online newsletter (“Listening for One Cell
in a Billion,” February 25, 2009) and Medical Daily online journal (http://
www.medicaldaily.com/how-do-you-know-if-you-have-cancer-fluorescent-
protein-may-illuminate-spread-cancer-cells-281250; “How Do You Know
If You Have Cancer? Fluorescent Protein May Illuminate Spread Of
Cancer Cells,” by Chris Weller, May 8, 2014). Dr. Galanzha is a co-
inventor of the in vivo multicolor multifluid photoacoustic flow cytometry
of blood, lymph & cerebrospinal fluid.
Chapter 7
Real-Time Imaging of
Microstructure and Function
Using Optical Coherence
Tomography
Christine P. Hendon
Columbia University, New York, USA
Andrew M. Rollins
Case Western Reserve University, Cleveland, USA
7.1 Introduction
Optical coherence tomography (OCT) is a noninvasive imaging modality that
provides high-resolution, depth-resolved imaging of tissue microstructure in
real time.1,2 Images are generated by detecting back-reflected light, where
contrast is generated by optical index changes in the sample. By measuring
singly backscattered light as a function of depth, OCT fills a valuable niche in
the imaging of tissue microstructure, providing subsurface imaging to depths
of 1 to 3 mm with high spatial resolution (10 mm) in three dimensions and
high sensitivity (.110 dB) in vivo with no contact needed between the probe
and the tissue. With high imaging speeds, high resolution, optical fiber-based
implementations, and functional extensions for measuring tissue birefringence
and flow, OCT has made a significant impact in clinical and biomedical
applications, including ophthalmology, cardiology, and oncology. In these
applications OCT provides the physician with near-histological resolution
imaging of subsurface tissue morphology, potentially aiding in monitoring
treatment, biopsy site selection, or even approaching the goal of “optical
biopsy.” With these specifications, OCT is quickly being translated into
commercial products readily available to the clinical community, and in some
cases, they are becoming the standard of care.
385
386 Chapter 7
7.2 Optical Coherence Tomography Principles

In OCT, depth is gated by measuring interference between the sample and a
reference using a low coherence interferometer (Fig. 7.1). Reflection and
scattering sites are localized with a resolution corresponding to the coherence
length of the illumination source, lc, given by
2 ln 2 l20
lc ¼ : (7.1)
p Dl
Here, l0 is the center wavelength of the source, with full-width at half-
maximum (FWHM) bandwidth Dl. Hence, low-coherence (broadband)
sources are employed in OCT to achieve microstructural imaging approaching
the cellular level.3 On the other hand, the lateral resolution is determined by
the focused beam spot size in the tissue. This decoupling of axial and lateral
resolution is an advantage of OCT, allowing for high axial resolution in
applications where high-numerical-aperture optics is challenging, such as
ophthalmic and catheter-based imaging. Using conventional optics (assuming
Figure 7.1 Michelson interferometer. a) Interferometer schematic. b) Interference pattern

for light source with a narrow bandwidth. c) Interference pattern for light source with broad
bandwidth. Interference occurs when the path length difference of the sample, ls, and
reference arm, lr, is within the coherence length, lc, of the light source.
Real-Time Imaging of Microstructure and Function Using OCT 387
a Gaussian beam), there is a tradeoff between the lateral resolution and

imaging depth. As shown in (Eq. 7.2), the depth of focus, b, is proportional to
the square of the spot size, Dx [Eq. (7.3)]. Therefore, OCT traditionally uses
low-numerical-aperture lenses to maintain a long depth of focus:
Dx2
b¼p , (7.2)
2l
where the spot size

4l f
Dx ¼ (7.3)
p d
is determined by the focal length f and aperture diameter d of the focusing

optics.
A single reflectivity profile as a function of depth is referred to as an
A-scan [Fig. 7.2(a)], analogous to ultrasound A-mode imaging. A two-
dimensional image (B-scan) is built by collecting many A-scans while
scanning the probe beam laterally across the sample [Fig. 7.2(b)]. With the
use of two scanning mirrors, multiple B-scan images can be acquired by raster
scanning the probe beam to obtain volumetric image sets [Fig. 7.2(c)].
Figure 7.2 OCT image generation of mouse epicardium. (a) 1-D axial scan. (b) 2-D B-scan
image generated by transverse scanning and collection of multiple axial scans. (c) 3-D
volume reconstruction from raster scanning across surface, collecting a series of B-scan
images.
388 Chapter 7
7.2.1 Time-domain OCT

OCT imaging can be performed either in the time domain (TDOCT) or the
frequency domain (FDOCT, also called Fourier-domain OCT). The optical
configuration for a fiber-based TDOCT system is illustrated schematically in
Fig. 7.3. TDOCT uses a scanning optical delay line in the reference arm to
generate axial scans.
As an illustration of generating an axial scan, Fig. 7.3 is a sample with
three reflection sites in depth. Light returning from the sample and reference
arms is recombined and interferes at the detector. The interference signal is
processed and recorded as the reference arm delay line is scanned. Because
interference only occurs when the optical path lengths of the sample and
reference arms are matched to within the coherence length of the light source,
reflection and scattering sites are localized within a resolution corresponding
to the coherence length [Fig. 7.3(c)]. By monitoring the envelope of the
detected interferometric pattern (i.e., photodiode current) as a function of the
reference arm delay, a profile of sample reflectivity versus depth is obtained
[Fig. 7.3(d)].
Figure 7.3 a) Time-domain OCT system, in which axial scans are generated by
mechanical scanning of reference mirror. (b) A sample with three reflecting sites results in
depth, z. (c) Interference fringes are localized to within the coherence length of light source.
(d) The envelope of the fringe pattern results in an axial reflectivity profile, or A-scan.
The recorded signal at the detector, Io, is proportional to a DC term and

an autocorrelation term, I o jE r j2 þ jE s j2 þ 2E r E s cosð2 kDzÞ. The DC term
is the sum of the reference, Er, and sample arm, Es, electric fields squared. The
autocorrelation term is produced by the path length difference, Dz, between
the reference and sample arm.
In OCT, as in any optical heterodyne detector, the detected signal-to-noise
ratio (SNR) is approximately proportional to the optical power illuminating
the sample and is inversely proportional to the detection bandwidth.4,5 In the
shot noise limit,
rPs Rs
SNRTDOCT ¼ , (7.4)
2eB
where Ps is the power incident on the sample, Rs is the power reflectivity of the
sample, e is the electronic charge, B is the detection bandwidth, and r is the
detector responsivity given by r ¼ hl0e/hc. Here, h is the detector quantum
efficiency, l0 is the optical source center wavelength, h is Planck’s constant,
and c is the free space speed of light. In TDOCT, the detected signal
bandwidth is proportional to the image acquisition rate.5 This expression
assumes that the optical power returned from the sample is much less that that
returned from the reference arm of the interferometer, which is generally true
when imaging biological tissues. Therefore, an increase in the image
acquisition rate will increase the signal bandwidth and decrease the SNR.
In order to maintain the SNR while detecting the entire signal bandwidth, any
increase in the image acquisition rate must be accompanied by a proportional
increase in source optical power, which may be limited by the maximum
permissible exposure for safe biological imaging.
7.2.2 Frequency-domain OCT

As illustrated by Fig. 7.3, TDOCT discriminates light from a single reflection
site within the sample while illuminating its entire depth. Therefore, TDOCT is
inherently inefficient, which has led to innovations of parallel detection for
improving sensitivity and frame rate. These innovations are broadly referred to
as frequency- (or Fourier-) domain OCT (FDOCT). The principle of FDOCT is
capturing a spectral interference pattern6,7 instead of the temporal interference
pattern to generate an A-scan. Parallel detection in FDOCT increases the
integration time at each A-scan and thus improves the sensitivity of FDOCT
about 20dB over TDOCT without a trade-off in the imaging speed.8–10
FDOCT can be implemented using a broadband light source and
spectrometer in the detector arm of the interferometer, called spectral domain
OCT (SDOCT) [Fig. 7.4(a)], or using a single detector in conjunction with a
swept-frequency optical source, called swept source (SSOCT) or optical
frequency-domain imaging (OFDI) [Fig. 7.4(b)]. In FDOCT the reference
arm is stationary and the captured spectral interferogram represents the
390 Chapter 7
Figure 7.4 Fourier-domain OCT can be implemented using either (a) a broadband source
and spectrometer, (spectral domain OCT) or (b) a rapidly tunable laser (swept source OCT
or optical frequency domain imaging).
Fourier transform of the backscatter profile as a function of depth, or OCT

A-scan (Fig. 7.4). The recorded FDOCT signal must be inverse-Fourier
transformed to generate an OCT image. The photodetector signal as a
function of wavenumber, known as the spectral interferogram, is composed of
three terms [Eq. (7.5). The first term is a constant DC offset. The second term
is composed of a sum of sinusoidal terms, where each cosine is proportional to
the square root of sample reflectivity. This is the signal of interest. The depth
of the scattering event is encoded in the frequency of the sinusoidal term. The
third term is composed of a sum of auto-correlation terms, due to mutual
interference from each reflection site in the sample:
I D ðkÞ ¼ DC þ AC þ MI
XN
r
DC ¼ SðkÞ RR þ RSn
4 n¼1
XN pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r
AC ¼ SðkÞ RR RSn cosð2 kðzR zSn ÞÞ
2 n¼1
XN pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r
MI ¼ SðkÞ RSn RSm cosð2 kðzSn zSm ÞÞ (7.5)
4 n≠m¼1
where S(k) is the wavenumber spectrum, RR is the reflectivity of the reference

mirror, RSn is the reflectivity of the nth reflection site in the sample, zR is the
position of the reference mirror, zSn is the position of the nth reflection site in
the sample, and r is the detection sensitivity.
The first step taken after obtaining the spectral interferogram is to remove
the DC component of the signal. This can be accomplished, for example, by
blocking the sample arm and capturing a frame to subtract from all
subsequent frames. Because spectrometers typically do not sample the
spectrum linearly in wavenumber (k), and fast-sweeping lasers do not
typically scan linearly in k, the next step is to resample the spectrum so that it
is linearly spaced in k. The resulting signal is inverse-Fourier transformed,
transforming the data from the k-domain to the z-domain (distance), to obtain
the axial scan.
An example of obtaining an axial scan with a FDOCT system of a sample
with three reflection sites is shown in Fig. 7.5. An axial scan is generated by
taking the inverse-Fourier transform of the spectra interferogram. Since the
measured spectral interferogram is real, the Fourier transform is an even
function, resulting in a mirrored image on the opposite side of the zero-path
length. Samples that are equidistant from the zero path length cannot be
resolved because cos(2kDz) ¼ cos(2kDz), a problem that is referred to as the
complex conjugate artifact. If the entire sample being imaged is located
entirely on one side of the zero-path length, there is no complex conjugate
Figure 7.5 Axial scan generation with FDOCT. (a) Three reflection sites within the sample.
(b) Spectral interferogram represents summation of sinusoidal modulations of spectrum.
(c) Axial scan obtained by computing the inverse-Fourier transform of a spectral
interferogram that is evenly spaced in wavenumber.
392 Chapter 7
Figure 7.6 Full range imaging, complex conjugate ambiguity resolved in the anterior
chamber of the eye. Reprinted from Sarunic et al.17
artifact as only the positive or negative distances are displayed. However, if a

portion of the sample is located above the zero path length position, the
mirrored image overlaps with the actual image and cannot be removed with
standard imaging processing. Several techniques have been proposed to solve
this problem.11–16 Resolving the complex conjugate ambiguity will result in a
doubling of the usable imaging range, as shown in Fig. 7.6.17
The imaging range of an FDOCT system is fixed based on the optical and
detection design. The frequency of the modulation of the spectral interfero-
gram increases as the reflector site increases in distance from the zero path
length. With a finite number of sample points sampling the spectrum, the
frequencies that can be unambiguously resolved are limited by the Nyquist
theorem. As shown in Fig. 7.6, the imaging range, DD, is a function of the
center wavelength, lc, the spectral range covered by the spectrometer (or
sweep range of the tunable laser), Δl, and the number of pixels on the detector
(or number of samples of the spectral interferogram), N. If the complex

conjugate ambiguity is resolved, this equation is multiplied by two:
1 l2c
DD ¼ N: (7.6)
4 Dl
The SNR for a FDOCT system, assuming it is shot-noise limited, is
rPs Rs Dt
SNRF DOCT ¼ : (7.7)
2e
This is comparable to Eq. (7.4), replacing the detector bandwidth with the
inverse of the integration time of the camera or sweep time of the swept
source, Δt. The increase in sensitivity of the FDOCT systems is approximately
a factor of N/2, and allows significantly increased A-line rates, enabling
acquisition of 3-D image sets in short time periods. For many clinical and
biomedical applications, a high imaging speed is critical to reduce motion
artifacts and observe dynamic biological processes.
Within FDOCT systems, the sensitivity degrades with increasing distance
from the zero path length match.18,19 This sensitivity falloff is related to the
spectrometer optics and the pixel width of the detection cameras in SDOCT or
the instantaneous linewidth of the swept light source. The theoretical 6-dB
fall off can be calculated using Eq. (7.8), where drk is the spectral resolution:
ln 2
Dz6 dB ¼ : (7.8)
pdr k
7.2.2.1 Spectrometers
SDOCT uses a spectrometer to detect the broadband interference spectrum
between the sample and reference. The spectral interferogram recorded by the
spectrometer must be inverse Fourier transformed to generate an OCT image.
Spectrometers used for FDOCT typically consist of a diffraction grating,
collimating and objective optics, and a line-scan camera with pixels
approximately evenly spaced as a function of wavelength. The diffraction
angle of light dispersed by the grating is a nonlinear function of wavenumber
k ¼ 2p/l. Therefore, the spectrum recorded by the line-scan camera is
unevenly spaced in k. However, a spectral interferogram that is a linear
function of k is needed to inverse Fourier transform the data into the spatial
domain. One common practice is to interpolate the nonlinearly spaced
spectral interferogram and rescale the data into the wavenumber-domain
prior to the inverse-Fourier transform. Another option is to use a
spectrometer with a dispersion that is linear in wavenumber (linear k), which
improves sensitivity falloff and reduces computing time as compared to an
equivalent conventional spectrometer.20 To linearize the spectral dispersion in
394 Chapter 7
Figure 7.7 Linear-k spectrometer. a) Optical schematic. b) Representative B-scan of finger

nail fold. Reprinted from Hu and Rollins.20
wavenumber, an appropriately designed prism is inserted between the grating

and the objective lens of the spectrometer (Fig. 7.7).
The other key element in SDOCT spectrometers is an array detector. The
imaging range is directly proportional to the number of detecting elements N
l2c
in the line scan camera DD ¼ 14 Dl N. As an example, using a SDOCT setup
with a 100 nm spectral range, center wavelength of 1310 nm, and detector
with a 1024-element line scan camera, results in an imaging range of 4.3 mm.
7.2.2.2 Light sources

The optimal center wavelength is dependent on the sample and application.
Systems developed for retinal imaging typically use light sources around
800 nm (due to the low water absorption in this range) or 1 mm (due to an
advantageous trade-off between water absorption and light scattering).
Longer wavelengths (usually 1 mm or 1.3 mm) are often employed for
imaging of highly scattering tissues, since lower scattering at these
wavelengths increases penetration depth.
In order to achieve ultrahigh axial resolution (, 5 mm in air), several
broadband light sources have been employed, including broadband solid state
lasers,3 supercontinuum generation in highly nonlinear fiber,21,22 and multi-
plexed SLEDs.23 However, there are additional factors to consider when
developing a high-resolution imaging system. There is a tradeoff between the
axial resolution and imaging range in SD-OCT, as the imaging range is
inversely proportional to the bandwidth of the spectrometer, assuming a fixed
number of pixels. Water absorption and dispersion are other important
factors to consider when designing OCT systems with ultra-high axial
resolution. Figure 7.8 is an example spectrum of a broadband continuum light
source developed by pumping a photonic crystal fiber with a 1.059-mm
Figure 7.8 Broadband continuum light source: (a) Measured optical spectrum of photonic
crystal fiber (PCF) output (solid line), input pump laser spectrum (dotted blue line) and
(b) numerically simulated spectrum generated by using the same parameters as used in the
experiment. Simulations including (solid line) and not including (dashed line) stimulated
Raman scattering. Reprinted from Wang et al.25
femtosecond laser. This resulted in a smooth spectrum centered at 1.15 mm

that can be used for ultra-high-resolution OCT.24 The emission spectrum
ranges from 800 to 1300 nm, resulting in a measured axial resolution of
2.8 mm in air.
SS-OCT or OFDI systems are implemented similar to the fiber-based
interferometers described in Section 7.2.1, but the reference mirror is fixed and
the broadband light source is replaced by a frequency-swept laser light source.
Key factors in swept laser sources are the repetition rate, tuning range, and
instantaneous linewidth. These factors determine the acquisition line rate,
axial resolution, and range fall-off, respectively. Swept source lasers have been
demonstrated using scanning filters implemented using gratings and polygon
mirrors26 and Fabry-Pérot tunable etalons.27,28 High tuning rates of 115 kHz,
80-nm tuning range, centered at 1325 nm have been demonstrated with
polygon filters.29 Fourier-domain mode-locking (FDML) has been demon-
strated to enable high OCT imaging speeds (.1 MHz axial scan rates).30–32
FDML lasers achieve increased performance by synchronizing the sweep rate
of the tunable filter to the round-trip time of the light in the laser and allow for
both broad tuning range and high imaging speeds. Recently, swept light
sources have been developed with microelectromechanical systems (MEMS)
tunable vertical-cavity surface-emitting lasers (VCSELs), resulting in imaging
line rates up to 1 MHz, with greater than a 100-nm tuning range, and long
coherence lengths for sources centered at 1310 nm.33,34
396 Chapter 7
7.3 Functional Imaging

7.3.1 Doppler OCT
Most applications of OCT concentrate on imaging the static micro-anatomy
of near-surface tissue structures. In physical terms, this corresponds to
measuring the magnitude of the interferometric signal detected as a function
of depth and lateral position in the sample under investigation. Detection of
the complex OCT signal also provides interferometric phase information,
which is discarded in most conventional implementations but may be used to
assess tissue function in addition to imaging structure. Doppler optical
coherence tomography (DOCT) is an extension of OCT that performs
micron-scale-resolution velocity flow mapping simultaneous with anatomical
imaging.35,36 The technique employs coherent detection to monitor the
velocity of moving scatterers within the sample. The mean velocity in a
localized region is estimated from the phase of the depth-resolved back-
scattered light from the specimen.
Initial demonstrations of time-domain DOCT imaging used time-
frequency analysis methods such as the short-time Fourier transform
(STFT) or the wavelet transform to resolve the complex interferometric
spectrum as a function of depth.35 However, the velocity resolution scales
with 1/T, where T is the observation time used to calculate velocity,
limiting the velocity sensitivity in high-speed imaging applications. In
order to overcome this limitation, the interferometric phase is measured
across sequential scans, resulting in a high T and thus high-velocity
resolution, without compromising high-speed image acquisition. Real-time
implementation of DOCT has been demonstrated by many groups.37–39
FDOCT has allowed increased velocity dynamic range in Doppler
measurements, as the dynamic range is dependent on the line scan rate.
The absolute value of velocity can be quantified by multiple angle
detection.40,41 DOCT enables one to measure blood flow in capillaries
down to 10–100 um/s,42 and has been used successfully for imaging blood
flow, retinal vessels 43–35 (Fig. 7.13), skin,47,48 gastrointestinal (GI)
track,48,49 and small animal models.50–52 Recently, methods have been
developed to measure absolute flow in vessels without the need to measure
vessel dimensions or orientation.53–56
Speckle variance is another method to detect moving scatters within OCT
images without the need for analyzing the phase information. Using the OCT
structural image, speckle variance between sequential A-lines or B-scans are
used to detect motion. With this, vascular mapping is possible, with a range of
applications including analyzing angiogenesis within a tumor or for
angiography, or vascular mapping in the retina to enable volumetric
mosaicing.57–60 Figure 7.10 show examples of three applications for vascular
mapping using speckle variance.
Figure 7.9 (a) Doppler and (b) intensity OCT image of patient with proliferative diabetic
retinopathy. Reprinted from Wang et al.44
Figure 7.10 Three different applications of speckle variance measurements for vascular
PDSSLQJ DUH VKRZQ LQ WKH EUHDVW EUDLQ DQG VNLQ 6HHFRORUSODWHV5HSULQWHG IURP 9DNRF HW DO
7.3.2 Polarization-sensitive OCT

Polarization-sensitive OCT (PSOCT) provides information on the polariza-
tion state of the light reflected from the samples under study.62 Changes in the
polarization state of light are dominated by two mechanisms: scattering and
birefringence. Scattering changes the polarization state of light in a random
manner. Birefringence is a material property exhibited in highly organized
tissue such as collagen, where anisoptropic indices of refraction are observed.
Beyond structural imaging, PSOCT provides additional contrast to identify
organized tissue architecture and abnormal or damaged tissue. PSOCT has
been used to evaluate collagen content within intravascular plaques,63–65
normal fiber organization within the myocardium of animal models,66 tissue
398 Chapter 7
Figure 7.11 PS-OCT images of retina with age related macular degeneration
(a) Reflectivity; (b) retardation (color bar: 0–90 deg); (c) degree of polarization uniformity
(color bar: 0 – 1); (d) reflectivity overlaid with segmented retinal pigment epithelium. Image
VL]H GHJ KRUL]RQWDO [ PP YHUWLFDO6HHFRORUSODWHV 5HSULQWHG IURP *RW]LQJHU HW DO
damage due to ablation therapy,67–69 infarction,66 and the nerve fiber layer
within the retina70–73 (Fig. 7.11). When imaging through fiber catheters, the
rotation of the fiber may cause stress-induced birefringence, which can
adversely affect PSOCT measurements. Through the use of frequency
multiplexing, an OFDI system has been developed that allows catheter-based
PSOCT independent of the fiber birefringence.74 Spectral binning has also
been implemented for reducing polarization mode dispersion artifacts in
catheter-based imaging.75
7.4 Applications of OCT

7.4.1 Ophthalmology
The first successful clinical application of OCT was for high-resolution
imaging of ocular structure.1,76–81 OCT is well suited to ophthalmology
because it is non-contact, easily adaptable to existing ophthalmic instrumen-
tation, and most importantly, the axial imaging resolution is independent of
the working distance. In the anterior eye, the micron-scale resolution of OCT
imaging permits accurate biometry of large scale ocular structures and the
evaluation of morphological changes associated with pathologies of the
cornea,82 iris, and lens.
Figure 7.12 High-resolution image of the anterior chamber angle. Clearly visible is the
trabecular meshwork and Schlemm’s canal. Reprinted from Sarunic et al.17
OCT has the capability of imaging anterior segment structures and

providing precise measurements of clinically useful parameters such as corneal
thickness, anterior depth, and angle width (Fig. 7.12). The cornea appears as a
hyporeflective structure in which two zones can be identified. The posterior
wider zone represents the corneal stroma, Descemet’s membrane, and
endothelium. The anterior narrow zone, consisting of a surface interface
reflection and an underlying dark band, represents the corneal epithelial
structures. In the pupillary region, the lens capsule can be identified. Central
corneal thickness as well as epithelial and stromal thickness can be measured
from this image. Central anterior chamber depth is measured from the inner
surface of the cornea to the lens capsule.
In the posterior pole of the eye, OCT enables unprecedented, high-
resolution access to retinal and subretinal layers, for detecting and monitoring
therapy of diseases such as macular degeneration, macular holes, glaucoma,
and age-related macular degeneration. OCT is the only technique capable of
resolving retinal substructure in cross section in the living eye. The
development of broadband light sources has enabled ultra-high-resolution
OCT imaging to approach “optical biopsy,” capable of delineating nearly all
of the layers within the retina (Fig. 7.13).
FDOCT enables high speed, volumetric imaging of the retina and anterior
chamber. High acquisition speeds, focusing targets, and imaging registration
methods have all been used to reduce the effect of patient motion. En face
projections of these 3-D volumes can provide views of retinal fibers and blood
vessels84 (Fig. 7.14) comparable with standard diagnostic techniques such as
fundus photography.
400 Chapter 7
Figure 7.13 Comparison of histology (a) with in vitro ultra-high-resolution OCT (b) in the
monkey (Macaca Fascicularis) fovea. Ganglion cell axons (gglc ax); ganglion cell body layer
(gglc); inner plexiform layer (ipl); inner nuclear layer (inl); outer plexiform layer (transition
between Henle fibers and inl; not l/abeled); outer nuclear layer (onl); foveal cone inner
segments (cis); foveal cone outer segments (cos); pigment epithelium þ processes (pe);
choriocapillaris (chc); choroid (ch). Reprinted from Drexler and Fujimoto.83
Adaptive optics has been implemented in OCT systems to compensate for

ocular and system aberrations85 using a closed-feedback mechanism with a
wavefront sensor.86–88 Adaptive optics in combination with broad bandwidth
light sources provides ultra-high resolution imaging of the retina. Example
stitched C-scans are shown in Fig. 7.15. A high resolution enables
segmentation of the retinal layers within B-scans and visualization of fibers
and loss of fibers within en face images.
7.4.2 Cardiology
Cardiovascular disease is the leading cause of morbidity and mortality in the
United States.90 Imaging has played a vital role for early diagnosis of
cardiovascular diseases, monitoring and guidance of procedures, and
characterization of preclinical models of disease. The unique features of
OCT have made it a powerful tool for cardiovascular imaging, from basic
scientific research to clinical applications. In particular, cardiovascular OCT
is a potentially disruptive technology in the detection and treatment of
atherosclerotic plaques. Postmortem studies of patients who experienced a
Figure 7.14 C-mode slab images of the outer plexiform layer of the perimacular retina
provide a detailed image of the retinal capillaries. Reprinted from Kagemann et al.84
myocardial infarction have identified common features of plaques including

the presence of a large lipid, a necrotic core, a thin fibrous cap (,65 micron),
microcalcification, and inflammatory cells.91 The resolution of conventional
imaging modalities has limited the ability to visualize these features in vivo.
With resolution on the order of 10 mm, and imaging depth of approximately
2 mm, OCT can visualize the thin fibrous cap92 and accurately differentiate
between major components of the atherosclerotic plaque (fibrotic, lipid,
calcium, and collagen).63,93,94 OCT has also demonstrated the ability to
visualize and quantify macrophage density in atherosclerotic plaques.95–97 In
addition to evaluating plaque composition, OCT shows great promise for
assessing vascular response to stents.98–101 OCT has been used to assess stent
apposition, coverage, and the presence of thrombosis.
Broader adoption of intravascular OCT in the clinical setting was initially
limited by the need of temporary vessel occlusion and saline flushing to
provide an optically clear field of view during image acquisitions. With the
introduction of high-speed FDOCT systems, these limitations are disappear-
ing. An optically clear field of view is still necessary. However, with a single
402 Chapter 7
Figure 7.15 Mosaicking of ultrahigh resolution C-scans. En face images show retinal
nerve fibers and in some areas, loss of fibers, with corresponding decrease in retinal nerve
ILEHU OD\HU WKLFNQHVV PHDVXUHV 6HHFRORUSODWHV5HSULQWHG IURP +RRG HW DO
four-s saline flush without vessel occlusion, volumetric imaging of an entire

coronary segment is possible. This has been shown in vivo in large animals and
humans102 (Fig. 7.16). Clinically, OCT imaging of coronary vessels is being
used for evaluation of stent apposition (Fig. 7.17), stent coverage, thrombosis
formation, and vessel composition. Commercial devices have been FDA
approved, including stand alone OCT catheter based imaging systems in
addition to OCT integrated with fractional flow reserve (FFR).
Catheter based multimodality imaging is becoming increasingly common
for research evaluation through integration with ultrasound104,105 and other
optical modalities. To enable simultaneous OCT and optical modalities, the
use of double clad fibers are used, where the single-mode core transmits
the OCT sample arm signal, while the inner cladding collects multimode
signals. To make the multimodality systems optically efficient, double-clad
Figure 7.16 (a) Perspective cutaway view of entire 3D volume OFDI data set,
demonstrating the stent, a side branch, and a large calcific nodule. (b) Longitudinal section
through a portion of the dataset. The side branch and calcific nodule are evident. The scale
bar in B is 1.0 mm. Reprinted from Tearney et al.102
couplers were developed to separate returning single mode and multimode

signals.106,107 Systems have been demonstrated with OCT integrated
with fluorescence,108–110 spectroscopy,111 and autofluoresence.112 Figure 7.18
shows example co-registered images of a human coronary artery ex vivo
obtained using an integrated OCT and near-infrared fluorescence catheter.
Recent research has demonstrated that OCT has great potential for
studying the myocardium, with applications in cardiac electrophysiology and
heart failure. The structure of the myocardium is important to both electrical
conduction and mechanical contractility. OCT has been demonstrated to
visualize critical structures related to electrical conduction, including the
purkinje network,113 and imaging the fast and slow pathways in the atrial-
ventricular (AV) node,114,115 myofiber organization115–117 in animal models,
and in vitro preparations of human tissue.118 In addition, OCT can be used to
image dynamic changes to the myocardium during the application of radio
frequency energy for the treatment of abnormal electrical conduction.119–122
An example of cardiovascular OCT in an animal model116 is shown in
Fig. 7.19. With the use of optical clearing, the entire ventricular wall of a
404 Chapter 7
Figure 7.17 Measurements of stents to identify stent coverage and volumetric

reconstruction of stent geometry. Reprint from Bezerra et al.103
mouse heart can be imaged with OCT, allowing investigators to analyze

the fiber organization in the entire ventricle.123 Polarization-sensitive OCT
has also been shown to provide additional contrast for analyzing fiber
organization without image processing.124–126
Due to its high spatial and temporal resolution, imaging penetration
depth, and non-contact nature, OCT fills a valuable need for imaging
the structure and function of the developing cardiovascular system. OCT has
been used to image the embryonic chick heart over the first few days of
development, where the heart transitions from a tube to a four-chamber
heart. OCT has promise to be a tool to enable studies of normal and abnormal
heart development and to evaluate surgical or pharmacological interventions.
Three-dimensional OCT imaging of fixed127 or living31,128–130 embryonic
hearts allows for analysis of morphological, tissue strain,131 and blood
flow53,132–137 changes as a function of genetic deficiencies127 or environmental
perturbations.136
Figure 7.18 Correlation of optical coherence tomography catheter based pullback images
(a,b) within a human cadaveric coronary artery compared with near infrared fluorescence
(c) cylindrical rendering of near infrared fluorescence signal and (d) validation with
IOXRUHVFHQFH UHIOHFWDQFH LPDJLQJ WR LGHQWLI\ UHJLRQV ZLWK LQFUHDVHG ILEULQ6HHFRORUSODWHV
5HSULQWHG IURP<RR HW DO
High-speed image acquisition has allowed visualization of the beat-

ing avian and murine hearts in three dimensions (Fig. 7.20). Longitudinal
imaging of embryonic heart development has been demonstrated by
integrating the sample arm into a microscope inside a custom incubator.
The incubator maintains oxygen, temperature, and humidity levels, which
allows five-dimensional imaging of avian heart development.138
406 Chapter 7
Figure 7.19 (A1–A3) 3D OCT imaging of fiber orientation in the rabbit right ventricular free
wall. (B1–B3) The fiber structure is visible within the en face slices. (C1–C3) Quantification
of the fiber orientation in the plane parallel to the wall. Slightly modified from Fleming et al.116
7.4.3 Oncology
The use of OCT in the field of oncology is increasing, including clinical
applications such as endoscopy and tumor margin detection, and small animal
imaging for the evaluation of tumor growth and response to drugs. The
potential for clinical application of OCT in gastro-intestinal (GI) endoscopy is
particularly promising since many common GI lesions occur within the depth
range of OCT imaging (1–2 mm). Endoscopic ultrasound (EUS), which is
currently used in clinical practice for evaluating larger lesions, lacks the
resolution to image small lesions in this depth range.
OCT catheters have been extensively developed for minimally invasive
procedures such as gastrointestinal endoscopy and intravascular imaging.
For endoscopic imaging, OCT light is delivered to the GI tract via small
diameter catheter probes that are often passed through the accessory
channel of standard GI endoscopes. Probes are usually developed with
a specialized shaft, which is axially flexible and torsionally rigid to
mechanically support the optical elements within the probe housing. The
probe sheath materials are selected for biocompatibility, compatibility with
standard cleaning and sterilization procedures, and optical and mechanical
properties. A variety of designs have been employed, including forward
Figure 7.20 OCT imaging in a developing embryo. (a) 2D OCT image in diastole cut
coronal to the body of the embryo. (b) 2D OCT image in systole cut coronal to the body of the
embryo. (c) Sum voxel projection image in diastole. (d) Sum voxel projection image in
systole. The OCT images clearly show the compact myocardial and endocardial borders of
the heart. CM is the compact myocardium, endo is the endocardium, and CJ is the cardiac
jelly. Reprinted from Jenkins et al,128
imaging,139–142 side-viewing,143 combined forward- and side-viewing,144 and

catheters based on MEMs actuated mirrors.145,146 Some OCT catheters use
a magnetic garnet, single-crystal 45-deg Faraday rotator at the probe tip147
to compensate for variations in the state of polarization of the probe light
due to bend-induced fiber birefringence as the probe is manipulated within
the body. To take advantage of the technology developments in high-speed
imaging and broadband light sources, recent catheter probes have been
developed with both rotational and pullback capabilities for volumetric
imaging of internal lumens. Specialized optical designs have been
developed to maintain a circular spot profile and reduce astigmatism
(Fig. 7.21)148 and accommodate broad bandwidth light sources.149
High-speed imaging with FDOCT systems enables comprehensive
imaging of entire lumens.148,150–152 An example volumetric image obtained
of a pig esophagus in vivo is shown in Fig. 7.22. The normal esophagus is
composed of stratified layers that are identifiable within OCT images,
408 Chapter 7
Figure 7.21 Catheter schematics of (A) a compound-lens configuration and (B) a single
gradient-index (GRIN) lens configuration. A cylindrical reflector used to redirect the light by
90 deg and to correct the astigmatism caused by the inner balloon lumen. (C) and (D) are the
calculated focused spot size and the maximum beam diameter within the GRIN lens versus
the GRIN lens pitch number for the compound- and single-lens configurations, respectively.
Reprinted from Xi et al.148
Figure 7.22 Representative real-time OCT images of a pig esophagus in vivo acquired
with the balloon imaging catheter in conjunction with an FDML SS-OCT system. (A) 2-D
snapshot and (B) 3-D cutaway view. Reprinted from Xi et al.148
including epithelium (E), lamina propria (LP), muscularis mucosae (SM),

submucosae (SM), muscularis propria (MP), and the submucosal gland
(Fig. 7.22).
Clinical studies using EOCT systems have been performed in the
esophagus, stomach, duodenum, ileum, colon, and rectum of human
subjects. In vivo EOCT images clearly delineate the mucosa and
submucosa in all endoscopically accessible GI organs.153 EOCT imaging
of the esophagus typically reveals a five-layer pattern (Fig. 7.23). The
uppermost layers correspond to mucosal substructures. Gland-like and
vessel-like structures seen in the fourth layer are interpreted as the
submucosa. In stomach, structures resembling gastric pits are evident,
and in duodenum and ileum, villi are clearly observed. EOCT imaging in
the colon delineates the mucosa, muscularis mucosae, and submucosa,
and colonic crypts.
Clinical observations during a study in Barrett’s esophagus resulted in the
identification of several distinguishing characteristics of the OCT appearance
of dysplasia.154 The most significant characteristics appeared to be hypore-
flectivity and lack of tissue organization. It has been hypothesized that the
characteristics of hyporeflectivity and tissue disorganization could distinguish
dysplasia from normal or benign tissue. It was further hypothesized that these
characteristics are universal in dysplasia in gastrointestinal mucosa. In one
study, volumetric images acquired with an OFDI system enabled an average
Figure 7.23 Barrett’s esophagus with dysplasia. (a) Videoendoscopic image.

(b) Histopathologic image of biopsy. (c) Cross-sectional OFDI image demonstrating regions
without dysplasia and with HGD. (d, e) Expanded view of regions in c. (f) Longitudinal slice
highlights the transition from gastric cardia, through a 9-mm segment of SIM and finally
into squamous mucosa. Scale bars and tick marks represent 1 mm. Reprinted from
Suter et al.151
410 Chapter 7
Figure 7.24 Capsule-based optical coherence tomography imaging of the GI tract.

Reprinted from Gora et al.156
coverage of 5.7 cm within twelve patients (Fig. 7.23).151 Endoscopic

OCT shows the potential capability to distinguish dysplastic from benign
mucosal tissue in the gastrointestinal tract. This capability is potentially
useful for screening or surveillance applications and potentially guiding
biopsies.
More recently, there have been technical advances towards capsule based
comprehensive imaging of the esophagus. Distal probe optics are packaged
into a capsule that patients can swallow.155,156 This allows for 3-D imaging of
the esophagus in patients without the need for sedation (Fig. 7.24).156
Another application of OCT for oncology is margin assessment to aid
tumor resection surgery (Fig. 7.25). In this application, an OCT system may
be located within or near the surgical suite and resections imaged to confirm
that the borders are completely normal and free of cancerous cells.
Figure 7.25 Example of positive tumor margins from patient diagnosed with ductal
carcinoma in situ. Foci of high back reflection (arrows) indicate areas with cancerous cells.
Reprinted from Nguyen et al.157
This would provide rapid feedback to the surgeon as to whether more tissue
needs to be removed.
7.5 Conclusions
Optical coherence tomography has advanced rapidly over the last two
decades. From its origins as a laboratory tool to its present status in a variety
of clinical and scientific applications, the technique has enjoyed rapid
technological growth and commercialization.
Several extensions of OCT technology have been developed that supple-
ment morphological imaging with functional imaging capability. In addition to
Doppler flow sensing and measuring polarization described above, functional
imaging research using OCT also includes technologies for analyte concentra-
tions or scatter size determination by spectral analysis,158–162 and tissue
electrical activity.163–166 Ongoing work to develop molecular sensitive/contrast
OCT,167–168 with either absorption or scattering agents, may highlight the pre-
sence of molecular labels in a manner comparable to fluorescence microscopy.
Multi-modality imaging has been shown to be an increasing trend, with
the development of integrated systems with techniques such as multiphoton
microscopy,169 second harmonic generation,170,171 or Raman spectroscopy.172,173
Future applications of OCT in biomedical research, disease diagnosis, and
procedural guidance are vast. With continued and rapid technology
developments, increased imaging speed, high resolution, small-diameter
catheters, automated analysis tools, and complementary functional data, the
future of OCT biomedical and clinical imaging looks bright. The rapid
technological development that has characterized OCT research and its
extension into clinical and basic science applications since its inception shows
no sign of abating.
412 Chapter 7
References
1. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson,
W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G.
Fujimoto, “Optical Coherence Tomography” Science 254(5035), 1178–
1181 (1991).
2. “Optical Coherence Tomography: Technology and Applications,”
ed. W. Drexler and J. G. Fujimoto, Springer (2008).
3. W. Drexler, U. Morgner, F. X. Kartner, C. Pitris, S. A. Boppart,
X. D. Li, E. P. Ippen, and J. G. Fujimoto, “In Vivo Ultrahigh-
Resolution Optical Coherence Tomography,” Optics Letters 24, 1221–
1223 (1999).
4. E. A. Swanson, D. Huang, M. R. Hee, J. G. Fujimoto, C. P. Lin, and
C. A. Puliafito, “High-speed optical coherence domain reflectometry,”
Optics Letters 17, 151–153 (1992).
5. A. M. Rollins and J. A. Izatt, “Optimal interferometer designs for
optical coherence tomography,” Optics Letters 24, 1484–1486
(1999).
6. A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat,
“Measurement of intraocular distances by backscattering spectral
interferometry,” Optics Communications 117(1-2), 43–48 (1995).
7. M. Wojtkowski, R. Leitgeb, A. Kowalczyk, T. Bajraszewski, and A. F.
Fercher, “In vivo human retinal imaging by Fourier domain optical
coherence tomography,” J Biomed Optics 7(3), 457–463 (2002).
8. M. Choma, M. Sarunic, C. Yang, and J. Izatt, “Sensitivity advantage of
swept source and Fourier domain optical coherence tomography,” Opt.
Express 11(18), 2183–2189 (2003).
9. R. Leitgeb, C. Hitzenberger, and A. Fercher, “Performance of fourier
domain vs. time domain optical coherence tomography,” Opt. Express,
11(8), 889–894 (2003).
10. J. F. d. Boer, B. Cense, B. H. Park, M. C. Pierce, G. J. Tearney, and
B. E. Bouma, “Improved signal-to-noise ratio in spectral-domain
compared with time-domain optical coherence tomography,” Optics
Express 28(21), p 2067–2069 (2003).
11. M. Wojtkowski, A. Kowalczyk, R. Leitgeb, and A. F. Fercher, “Full
range complex spectral optical coherence tomography technique in eye
imaging” Opt. Lett. 27(16), 1415–1417 (2002).
12. R. A. Leitgeb, C. K. Hitzenberger, A. F. Fercher, and T. Bajraszewski,
“Phase-shifting algorithm to achieve high-speed long-depth-range
probing by frequency-domain optical coherence tomography,” Opt.
Lett. 28(22), 2201–2203 (2003).
13. E. Götzinger, M. Pircher, R. Leitgeb, and C. Hitzenberger, “High speed
full range complex spectral domain optical coherence tomography,” Opt.
Express 13(2), 583–594 (2005).
14. Y. K. Tao, M. Zhao, and J. A. Izatt, “High-speed complex conjugate

resolved retinal spectral domain optical coherence tomography using
sinusoidal phase modulation,” Opt. Lett. 32(20), 2918–2920 (2007).
15. M. Sarunic, M. A. Choma, C. Yang, and J. A. Izatt, “Instantaneous
complex conjugate resolved spectral domain and swept-source OCT
using 3x3 fiber couplers, ” Opt. Express 13(3), 957–967 (2005).
16. M. V. Sarunic, B. E. Applegate, and J. A. Izatt, “Real-time quadrature
projection complex conjugate resolved Fourier domain optical coherence
tomography,” Optics Letters 31(16), 2426–2428 (2006).
17. M. V. Sarunic, S. Asrani, and J. A. Izatt, “Imaging the ocular anterior
segment with real-time, full-range Fourier-domain optical coherence
tomography,” Arch. Ophthalmol. 126(4), 537–542 (2008).
18. T. Bajraszewski, M. Wojtkowski, M. Szkulmowski, A. Szkulmowska,
R. Huber, and A. Kowalczyk, “Improved spectral optical coherence
tomography using optical frequency comb,” Opt. Express 16(6), 4163–
4176 (2008).
19. Z. Hu, Y. Pan, and A. M. Rollins, “Analytical model of spectrometer-
based two-beam spectral interferometry,” Appl. Opt. 46(35), 8499–8505
(2007).
20. Z. Hu and A. Rollins, “Fourier domain optical coherence tomography
with a linear-in-wavenumber spectrometer,” Opt Lett 32(24), 3524–7
(2007).
21. I. Hartl, X. D. Li, C. Chudoba, R. K. Ghanta, T. H. Ko, J. G. Fujimoto,
J. K. Ranka, and R. S. Windeler, “Ultrahigh-resolution optical
coherence tomography using continuum generation in an air silica
microstructure optical fiber,” Optics Letters 26(9), 608–610 (2001).
22. S. Ishida and N. Nishizawa, “Quantitative comparison of contrast and
imaging depth of ultrahigh-resolution optical coherence tomography
images in 800-1700 nm wavelength region,” Biomedical Optics Express
3(2), 282–294 (2012).
23. H. Wang, M. W. Jenkins, and A. M. Rollins, “A combined multiple-
SLED broadband light source at 1300 nm for high resolution optical
coherence tomography,” Optics Communications 281(7), 1896–1900
(2008).
24. H. Wang, C. Fleming, and A. M. Rollins, “Ultrahigh-resolution optical
coherence tomography at 1.15 mm using photonic crystal fiber with no
zero-dispersion wavelengths,” Optics Express 15(6), 3085–3092 (2007).
25. H. Wang, C. P. Fleming, and A. M. Rollins, “Ultrahigh-resolution
optical coherence tomography at 1.15 mm using photonic crystal fiber
with no zero-dispersion wavelengths,” Optics Express 15(6), 3085 (2007).
26. S. H. Yun, C. Boudoux, G. J. Tearney, and B. E. Bouma, “High-speed
wavelength-swept semiconductor laser with a polygon-scanner-based
wavelength filter,” Optics Letters 28(20), 1981–1983 (2003).
414 Chapter 7
27. M. A. Choma, K. Hsu, and J. A. Izatt, “Swept source optical coherence

tomography using an all-fiber 1300-nm ring laser source,” J Bio. Optics
10(4), 44009 (2005).
28. R. Huber, M. Wojtkowski, and J. Fujimoto, “Fourier Domain Mode
Locking (FDML): A new laser operating regime and applications for
optical coherence tomography,” Optics Express 14(8), 3225–3237
(2006).
29. W. Y. Oh, S. H. Yun, G. J. Tearney, and B. E. Bouma, “115 kHz tuning
repetition rate ultrahigh-speed wavelength-swept semiconductor laser,”
Optics Express 30(23), 3159–3161 (2005).
30. D. C. Adler, R. Huber, and J. G. Fujimoto, “Phase-sensitive optical
coherence tomography at up to 370, 000 lines per second using buffered
Fourier domain mode-locked lasers,” Optics Letters 32(6). 626–628
(2007).
31. M. W. Jenkins, D. C. Adler, M. Gargesha, R. Huber, F. Rothenberg,
J. Belding2, M. Watanabe, D. L. Wilson, J. G. Fujimoto, and A. M.
Rollins, “Ultrahigh-speed optical coherence tomography imaging and
visualization of the embryonic avian heart using a buffered Fourier
Domain Mode Locked laser,” Optics Express 15(10), 6251–6267
(2007).
32. W. Wieser, T. Klein, D. C. Adler, F. Trépanier, C. M. Eigenwillig,
S. Karpf, J. M. Schmitt, and R. Huber, “Extended coherence length
megahertz FDML and its application for anterior segment imaging,”
Biomed Optics Express 3(10), 2647–2657 (2012).
33. B. Potsaid, V. Jayaraman, A. G. Fujimoto, J. Jiang, P. J. S. Heim, and
A. E. Cable, “MEMS tunable VCSEL light source for ultrahigh speed
60kHz - 1MHz axial scan rate and long range centimeter class OCT
imaging,” in Proc. SPIE 8213, Optical Coherence Tomography and
Coherence Domain Optical Methods in Biomedicine XVI San
Francisco, CA (2012)
34. J. J. Vijaysekhar Jayaraman, Hua Li, Peter Heim, Garrett Cole, Ben
Potsaid, James G. Fujimoto, and Alex Cable. “OCT Imaging up to
760 Khz Axial Scan Rate Using Single-Mode 1310 nm MEMs-Tunable
VCSELs with .100 nm Tuning Range,” in CLEO (2011) - Laser
Applications to Photonic Applications (2011).
35. J. A. Izatt, M. D. Kulkarni, S. Yazdanfar, J. K. Barton, and A. J. Welch,
“In Vivo Bidirectional Color Doppler Flow Imaging of Picoliter Blood
Volumes Using Optical Coherence Tomography,” Optics Letters 22(18),
1439–1441 (1997).
36. Z. Chen, T. E. Milner, S. Srinivas, X. Wang, A. Malekafzali, M. J. C.
van Gemert, and J. S. Nelson, “Noninvasive imaging of in vivo blood
flow velocity using optical Doppler tomography,” Optics Letters 22,
1119–1121 (1997).
37. V. Westphal, S. Yazdanfar, A. Rollins, and J. Izatt, “Real-time, high

velocity-resolution color Doppler optical coherence tomography,”
Optics Letters 27(1), 34–36 (2002).
38. Y. Zhao, Z. Chen, Z. Ding, H. Ren, and J. S. Nelson, “Real-time phase-
resolved functional optical coherence tomography by use of optical
Hilbert transformation,” Opt Lett. 27(2), 98–100 (2002).
39. Z. Ding, Y. Zhao, H. Ren, J. Nelson, and Z. Chen, “Real-time phase-
resolved optical coherence tomography and optical Doppler tomogra-
phy,” Optics Express 10(5), 236–245 (2002).
40. D. P. Davé and T. E. Milner, “Doppler-angle measurement in highly
scattering media,” Opt. Lett. 25(20), 1523–1525 (2000).
41. C. J. Pedersen, D. Huang, M. A. Shure, and A. M. Rollins,
“Measurement of absolute flow velocity vector using dual-angle, delay-
encoded Doppler optical coherence tomography,” Opt. Lett. 32(5), 506–
508 (2007).
42. R. A. Leitgeb, L. Schmetterer, C. K. Hitzenberger, A. F. Fercher,
F. Berisha, M. Wojtkowski, and T. Bajraszewski, “Real-time measure-
ment of in vitro flow by Fourier-domain color Doppler optical coherence
tomography,” Opt. Lett. 29(2), 171–173 (2004).
43. S. R. Yazdanfar, AM, and J. A. Izatt, “In vivo imaging of human retinal
flow dynamics by color Doppler optical coherence tomography,”
Archives of Ophthalmology 121(2), p. 235–139 (2003).
44. Y. Wang, A. Fawzi, O. Tan, J. Gil-Flamer, and D. Huang, “Retinal
blood flow detection in diabetic patients by Doppler Fourier domain
optical coherence tomography,” Optics Express 17(5), 4061–4073 (2009).
45. Y. Wang, A. Lu, J. Gil-Flamer, O. Tan, J. A. Izatt, and D. Huang,
“Measurement of total blood flow in the normal human retina using
Doppler Fourier-domain optical coherence tomography,” Br.
J. Ophthalmol. 93(5), 634–637 (2009).
46. J. K. Barton, J. A. Izatt, M. D. Kulkarni, S. Yazdanfar, and A. J. Welch,
“Three-Dimensional Reconstruction of Blood Vessels from in vivo
Color Doppler Optical Coherence Tomography Images,” Dermatology
198(4), 355–361 (1999).
47. M. C. Pierce, J. Strasswimmer, B. H. Park, B. Cense, and J. F. de Boer,
“Advances in optical coherence tomography imaging for dermatology,”
J investigative dermatology 123(3), 458–463 (2004).
48. V. X. D. Yang, M. L. Gordon, B. Qi, J. Pekar, S. Lo, E. Seng-Yue,
A. Mok, B. C. Wilson, and I. A. Vitkin, “High speed, wide velocity
dynamic range Doppler optical coherence tomography (Part III): in vivo
endoscopic imaging of blood flow in the rat and human gastrointestinal
tracts,” Optics Express 11(19), 2416–2424 (2003).
49. V. X. Yang, S. J. Tang, M. L. Gordon, B. Qi, G. Gardiner, M. Cirocco,
P. Kortan, G. B. Haber, G. Kandel, I. A. Vitkin, B. C. Wilson, and
416 Chapter 7
N. E. Marcon, “Endoscopic Doppler optical coherence tomography in

the human GI tract: initial experience,” Gastrointestinal Endoscopy
61(7), 879–890 (2005).
50. R. K. Wang and L. An, “Doppler optical micro-angiography for
volumetric imaging of vascular perfusion in vivo,” Optics Express
17(11), 8926–8940 (2009).
51. A. Mariampillai, B. A. Standish, E. H. Moriyama, M. Khurana, N. R.
Munce, M. K. Leung, J. Jiang, A. Cable, B. C. Wilson, I. A. Vitkin, and
V. X. Yang, “Speckle variance detection of microvasculature using
swept-source optical coherence tomography,” Optics Letters 33(13),
1530–1532 (2008).
52. M. Khurana, E. H. Moriyama, A. Mariampillai, and B. C. Wilson,
“Intravital high-resolution optical imaging of individual vessel response
to photodynamic treatment,” J of Biomed Opt 13(4), 040502 (2008).
53. M. W. Jenkins, L. Peterson, S. Gu, M. Gargesha, D. L. Wilson,
M. Watanabe, and A. M. Rollins, “Measuring hemodynamics in the
developing heart tube with four-dimensional gated Doppler optical
coherence tomography,” J of Biomed Opt 15(6), 066022 (2010).
54. V. J. Srinivasan, S. Sakadzić, I. Gorczynska, S. Ruvinskaya, W. Wu,
J. G. Fujimoto, and D. A. Boas, “Quantitative cerebral blood flow with
Optical Coherence Tomography,” Optics Express 18(3), 2477–2494
(2010).
55. C. Blatter, B. Grajciar, L. Schmetterer, and R. A. Leitgeb, “Angle
independent flow assessment with bidirectional Doppler optical coher-
ence tomograph,” Opt Lett 38(21), 4433–4436 (2013).
56. L. M. Peterson, S. Gu, M. W. Jenkins, and A. M. Rollins, “Orientation-
independent rapid pulsatile flow measurement using dual-angle Doppler
OCT.,” Biomed Opt Express 5(2), 499–514 (2014).
57. J. Enfield, E. Jonathan, and M. Leahy, “In vivo imaging of the
microcirculation of the volar forearm using correlation mapping optical
coherence tomography (cmOCT),” Biomed Opt Express 2(5), 1184–1193
(2011).
58. A. Mariampillai, M. K. K. Leung, M. Jarvi, B. A. Standish, K. Lee,
B. C. Wilson, A. Vitkin, and V. X. D. Yang, “Optimized speckle
variance OCT imaging of microvasculature,” Opt Lett 35(8), 1257–1259
(2010).
59. N. Sudheendran, S.H.S., M. E. Dickinson, I. V. Larina, and K. V. Larin,
“Speckle variance OCT imaging of the vasculature in live mammalian
embryos,” Laser Physics Letters 8(3), 125–135 (2011).
60. H. C. Hendargo, R. Estrada, S. J. Chiu, C. Tomasi, S. Farsiu, and J. A.
Izatt, “Automated non-rigid registration and mosaicing for robust
imaging of distinct retinal capillary beds using speckle variance optical
coherence tomography,” Biomed Opt Express 4(6), 803–821 (2013).
61. B. J. Vakoc, D. Fukumura, R. K. Jain, and B. E. Bouma, “Cancer

imaging by optical coherence tomography: preclinical progress and
clinical potential,” Nature Reviews Cancer 12, 363–368 (2012).
62. J. F. de Boer and T. E. Milner, “Review of polarization sensitive optical
coherence tomography and Stokes vector determination,” J. Biomed.
Opt. 7(3), 359–371 (2002).
63. S. Nadkarni, M. Pierce, B. Park, J. de Boer, P. Whittaker, B. Bouma,
J. Bressner, E. Halpern, S. Houser, and G. Tearney, “Measurement of
collagen and smooth muscle cell content in atherosclerotic plaques using
polarization-sensitive optical coherence tomography,” J Am Coll
Cardiol. 49(13), 1474–81 (2007).
64. S. Nadkarni, B. Bouma, J. de Boer, and G. Tearney, “Evaluation of
collagen in atherosclerotic plaques: the use of two coherent laser-based
imaging methods,” Lasers Med Sci (2008).
65. W.-C. Kuo, M.-W. Hsiung, J.-J. Shyu, N.-K. Chou, and P.-N.
Yang, “Assessment of arterial characteristics in human atheroscle-
rosis by extracting optical properties from polarization-sensitive
optical coherence tomography,” Optics Express 16(11), 8117–8125
(2008).
66. C.-W. Sun, Y.-M. Wang, L.-S. Lu, C.-W. Lu, I.-J. Hsu, M.-T. Tsai,
C. C. Yang, Y.-W. Kiang, and C.-C. Wu, “Myocardial tissue
characterization based on a polarization-sensitve optical coherence
tomography system with an ultrashort pulsed lase,” J Biomed Opt
11(5), 054016 (2006).
67. J. F. de Boer, S. M. Srinivas, A. Malekafzali, Z. Chen, and J. S. Nelson,
“Imaging thermally damaged tissue by polarization sensitive optical
coherence tomography,” Opt Express 3(6), p. 212–8 (1998).
68. K. Schoenenberger, J. Bill, W. Colston, D. J. Maitland, L. B. D. Silva,
and M. J. Everett, “Mapping of birefringence and thermal damage in
tissue by use of polarization-sensitive optical coherence tomography”
Applied Optics 37(25), 6026–6036 (1998).
69. X. Fu, Z. Wang, H. Wang, Y. T. Wang, M. W. Jenkins, and A. M.
Rollins, “Fiber-optic catheter-based polarization-sensitive OCT for
radio-frequency ablation monitoring,” Opt Lett. 39(17), 5066–5069
(2014).
70. E. Götzinger, M. Pircher, W. Geitzenauer, C. Ahlers, B. Baumann,
S. Michels, U. Schmidt-Erfurth, and C. K. Hitzenberger, “Retinal
pigment epithelium segmentation by polarization sensitive optical
coherence tomography,” Opt Express 16(21), 16410–16422 (2008).
71. B. Cense, T. C. Chen, B. H. Park, M. D. Pierce, and J. F.d. Boer,
“In vivo depth-resolved birefringence measurements of the human retinal
nerve fiber layer by polarization-sensitive optical coherence tomogra-
phy,” Opt Lett 27(18), 1610 (2002).
418 Chapter 7
72. M. Yamanari, M. Miura, S. Makita, T. Yatagai, and Y. Yasuno,

“Phase retardation measurement of retinal nerve fiber layer by
polarization-sensitive spectral-domain optical coherence tomography
and scanning laser polarimetry,” J of Biomed Opt 13(1), 014013
(2008).
73. E. Götzinger, M. Pircher, and C. K. Hitzenberger, “High speed spectral
domain polarization sensitive optical coherence tomography of the
human retina,” Opt Express 13(25), 10217–10229 (2005).
74. W. Oh, S. Yun, B. Vakoc, M. Shishkov, A. Desjardins, B. Park,
J. de Boer, G. Tearney, and B. Bouma, “High-speed polarization
sensitive optical frequency domain imaging with frequency multi-
plexing,” Opt Express 16(3), 1096–1103 (2008).
75. M. Villiger, E. Z. Zhang, S. K. Nadkarni, W.-Y. Oh, B. J. Vakoc, and
B. E. Bouma, “Spectral binning for mitigation of polarization mode
dispersion artifacts in catheter-based optical frequency domain imag-
ing,” i Express 21(14), 16353–16369 (2013).
76. A. F. Fercher, K. Mengedoht, and W. Werner, “Eye-Length Measure-
ment by Interferometry with Partially Coherent Light,” Opt. Lett. 13,
186–188 (1988).
77. E. A. Swanson, J. A. Izatt, M. R. Hee, D. Huang, C. P. Lin, J. S.
Schuman, C. A. Puliafito, and J. G. Fujimoto, “In vivo retinal
imaging by optical coherence tomography,” Opt. Lett. 18, 1864–1866
(1993).
78. J. A. Izatt, M. R. Hee, D. Huang, J. G. Fujimoto, E. A. Swanson, C. P.
Lin, J. S. Schuman, and C. A. Puliafito, “Ophthalmic Diagnostics using
Optical Coherence Tomography,” Proc. Soc. Photo-Opt. Instrum. Eng.
1877, 136 (1993).
79. J. A. Izatt, M. R. Hee, E. A. Swanson, C. P. Lin, D. Huang, J. S.
Schuman, C. A. Puliafito, and J. G. Fujimoto, “Micrometer-Scale
Resolution Imaging of the Anterior Eye with Optical Coherence
Tomography,” Arch. Ophthalmol. 112, 1584–1589 (1995).
80. C. A. Puliafito, C. P. Lin, M. R. Hee, J. S. Schuman, J. S. Duker,
E. Rachel, J. A. Izatt, E. A. Swanson, D. Huang, and J. G. Fujimoto,
“Diagnostic of Macular Diseases with Optical Coherence Tomography,”
Ophthalmol 102, 217–229 (1995).
81. J. S. Schuman, M. R. Hee, C. A. Puliafito, C. Wong,
T. Pedut-Kloizman, C. P. Lin, E. Hertzmark, J. A. Izatt, E. A. Swanson,
and J. G. Fujimoto, “Quantification of Nerve Fiber Thickness in Normal
and Glaucomatous Eyes using Optical Coherence Tomography: A Pilot
Study,” Arch. Ophthalmol. 113, 586–596 (1995).
82. J. Ramos, Y. Li, and D. Huang, “Clinical and research applications of
anterior segment optical coherence tomography - a review,” Clinical &
experimental ophthalmology 37(1), 81–89 (2009).
83. W. Drexler and J. G. Fujimoto, “State-of-the-art retinal optical

coherence tomography,” Progress in Retinal and Eye Research 28, 45–
88 (2008).
84. L. Kagemann, H. Ishikawa, G. Wollstein, M. Gabriele, and J. S.
Schuman, “Visualization of 3-D high speed ultrahigh resolution optical
coherence tomographic data identifies structures visible in 2D frames,”
Opt Express 17(5), 4208–4220 (2009).
85. R. J. Zawadzki, S. S. Choi, S. M. Jones, S. S. Oliver, and J. S. Werner,
“Adaptive optics-optical coherence tomography: optimizing visualiza-
tion of microscopic retinal structures in three dimensions,” J Optical
Society of America A 24(5), 1373–1383 (2007).
86. R. J. Zawadzki, S. M. Jones, S. S. Olivier, M. Zhao, B. A. Bower, J. A.
Izatt, S. Choi, S. Laut, and J. S. Werner, “Adaptive-optics optical
coherence tomography for high-resolution and high-speed 3D retinal
in vivo imaging,” Opt Express 13(21), 8532–8546 (2005).
87. Y. Zhang, J. Rha, R. S. Jonnal, and D. T. Miller, “Adaptive optics
parallel spectral domain optical coherence tomography for imaging the
living retina,” Opt Express 13(12), 4792–4811 (2005).
88. Y. Zhang, B. Cense, J. Rha, R. S. Jonnal, W. Gao, R. J. Zawadzki,
J. S. Werner, S. Jones, S. Olivier, and D. T. Miller, “High-speed
volumetric imaging of cone photoreceptors with adaptive optics
spectral-domain optical coherence tomography” Opt Express 14(10),
4380–4394 (2006).
89. D. C. Hood, M. F. Chen, D. Lee, B. Epstein, P. Alhadeff, R. B. Rosen,
R. Ritch, A. Dubra, and T. Y. P. Chui, “Confocal Adaptive Optics
Imaging of Peripapillary NerveFiber Bundles: Implications for Glauco-
matous Damage Seen on Circumpapillary OCT Scans,” Translational
Vision Science & Technology 4(2), 12 (2015).
90. W. Rosamond, K. Flegal, K. Furie, A. Go, K. Greenlund, N. Haase,
S. M. Hailpern, M. Ho, V. Howard, B. Kissela, S. Kittner,
D. Lloyd-Jones, M. McDermott, J. Meigs, C. Moy, G. Nicho,
C. O'Donnell, V. Roger, P. Sorlie, J. Steinberger, T. Thom, M. Wilson,
and Y. Hong, “Heart Disease and Stroke Statistics 2008 Update: A Report
From the American Heart Association Statistics Committee and Stroke
Statistics Subcommittee,” Circulation 117, e25–e146 (2008).
91. A. P. Burke, A. Farb, G. T. Malcom, Y.-h. Liang, J. Smialek, and
R. Virmani, “Coronary Risk Factors and Plaque Morphology in Men
with Coronary Disease Who Died Suddenly,” N Engl J Med 336(18),
1276–1282 (1997).
92. K. Teruyoshi, A. Takashi, K. Takahiro, O. Hiroyuki, W. Nozomi,
T. Eiji, N. Yoji, S. Renan, S. Yoshito, and Y. Kiyoshi, “Measurement of
the thickness of the fibrous cap by optical coherence tomography,”
American heart journal 152(4), 755.e1-755.e4 (2006).
420 Chapter 7
93. H. Yabushita, B. E. Bouma, S. L. Houser, H. T. Aretz, I. K. Jang, K. H.

Schlendorf, C. R. Kauffman, M. Shishkov, D. H. Kang, E. F. Halpern,
and G. J. Tearney, “Characterization of human atherosclerosis by
optical coherence tomography,” Circulation 106(13), 1640–5 (2002).
94. M. Cilingiroglu, J. H. Oh, B. Sugunan, N. J. Kemp, J. Kim, S. Lee,
H. N. Zaatari, D. Escobedo, S. Thomsen, T. E. Milner, and M. D.
Feldman, “Detection of vulnerable plaque in a murine model of
atherosclerosis with optical coherence tomography,” Catheter
Cardiovasc Interv. 67(6), 915–923 (2006).
95. G. J. Tearney, H. Yabushita, S. L. Houser, H. T. Aretz, I.-K. Jang,
K. H. Schlendorf, C. R. Kauffman, M. Shishkov, E. F. Halpern, and
B. E. Bouma, “Quantification of Macrophage Content in Atheroscle-
rotic Plaques by Optical Coherence Tomography,” Circulation 107(1),
113–119 (2003).
96. O. C. Raffel, G. J. Tearney, D. D. Gauthier, E. F. Halpern, B. E.
Bouma, and I.-K. Jang, “Relationship Between a Systemic Inflamma-
tory Marker, Plaque Inflammation, and Plaque Characteristics Deter-
mined by Intravascular Optical Coherence Tomography,” Arterioscler
Thromb Vasc Biol 27(8), 1820–1827 (2007).
97. J. Oh, M. D. Feldman, J. Kim, P. Sanghi, D. Do, J. J. Mancuso,
N. Kemp, M. Cilingiroglu, and T. E. Milner, “Detection of
macrophages in atherosclerotic tissue using magnetic nanoparticles
and differential phase optical coherence tomography,” J of Biomed Opt
13(5), 054006 (2008).
98. P. Moore, P. Barlis, J. Spiro, G. Ghimire, M. Roughton, C. Di Mario,
W. Wallis, C. Ilsley, A. Mitchell, M. Mason, R. Kharbanda, P. Vincent,
S. Sherwin, and M. Dalby, “A randomized optical coherence tomogra-
phy study of coronary stent strut coverage and luminal protrusion with
rapamycin-eluting stents,” JACC Cardiovasc Interv. 2(5), 437–444
(2009).
99. N. Gonzalo, H. M. Garcia-Garcia, P. W. Serruys, K. H. Commissaris,
H. Bezerra, P. Gobbens, M. Costa, and E. Regar, “Reproducibility of
quantitative optical coherence tomography for stent analysis,” i5(2),
224–232 (2009).
100. G. T. Bonnema, K. O. Cardinal, S. K. Williams, and J. K. Barton, “An
automatic algorithm for detecting stent endothelialization from volu-
metric optical coherence tomography datasets,” Physics in Medicine and
Biology 53(12), 3083–3098 (2008).
101. H. Otake, J. Shite, J. Ako, T. Shinke, Y. Tanino, D. Ogasawara,
T. Sawada, N. Miyoshi, H. Kato, B. K. Koo, Y. Honda, P. J. Fitzgerald,
and K. Hirata, “Local determinants of thrombus formation following
sirolimus-eluting stent implantation assessed by optical coherence
tomography,” JACC Cardiovasc Interv. 2(5), 459–466 (2009).
102. G. J. Tearney, S. Waxman, M. Shishkov, B. J. Vakoc, M. J. Suter, M. I.

Freilich, A. E. Desjardins, W.-Y. Oh, L. A. Bartlett, M. Rosenberg, and
B. E. Bouma, “Three-Dimensional Coronary Artery Microscopy by
Intracoronary Optical Frequency Domain Imaging,” JACC: Cardiovas-
cular Imaging 1(6), 752–761 (2008).
103. H. G. Bezerra, M. A. Costa, G. Guagliumi, A. M. Rollins, and D. I.
Simon, “Intracoronary optical coherence tomography: a comprehensive
review clinical and research applications,” JACC Cardiovasc Interv
2(11), 1035–46 (2009).
104. X. Li, J. Yin, C. Hu, Q. Zhou, K. K. Shung, and Z. Chen, “High-
resolution coregistered intravascular imaging with integrated ultrasound
and optical coherence tomography probe,” Applied Physics Letters
97(13), 133702 (2010).
105. J. Yin, X. Li, J. Jing, J. Li, D. Mukai, S. Mahon, A. Edris, K. Hoang,
K. K. Shung, M. Brenner, J. Narula, Q. Zhou, and Z. Chen, “Novel
combined miniature optical coherence tomography ultrasound probe for
in vivo intravascular imaging,” J of Biomed Opt Lett 16(6), 060505
(2011).
106. S. Lemire-Renaud, M. Rivard, M. Strupler, D. Morneau, F. Verpillat,
X. Daxhelet, N. Godbout, and C. Boudoux, “Double-clad fiber coupler
for endoscopy,” 18(10), 9758–9764 (2010).
107. S.-Y. Ryu, H.-Y. Choi, M.-J. Ju, J.-H. Na, W.-J. Choi, and B.-H. Lee,
“The Development of Double Clad Fiber and Double Clad Fiber
Coupler for Fiber Based Biomedical Imaging Systems,” J Optical
Society of Korea 13(3), 310–315 (2009).
108. S. Liang, A. Saidi, J. Jing, G. Liu, J. Li, J. Zhang, C. Sun, J. Narula, and
Z. Chen, “Intravascular atherosclerotic imaging with combined fluores-
cence and optical coherence tomography probe based on a double-clad
fiber combiner,” J of Biomed Opt 17(7), 070501 (2012).
109. S. Y. Ryu, H. Y. Choi, J. Na, E. S. Choi, and B. H. Lee, “Combined
system of optical coherence tomography and fluorescence spectroscopy
based on double-cladding fiber,” Opt Lett 33(20), 2347–2349 (2008).
110. H. Yoo, J. W. Kim, M. Shishkov, E. Namati, T. Morse,
R. Shubochkin, J. R. McCarthy, V. Ntziachristos, B. E. Bouma,
F. A. Jaffer, and G. J. Tearney, “Intra-arterial catheter for simulta-
neous microstructural and molecular imaging in vivo,” Nature
Medicine, 17(12), 1680–1685 (2011).
111. A. M. Fard, P. Vacas-Jacques, E. Hamidi, H. Wang, R. W. Carruth,
J. A. Gardecki, and G. J. Tearney, “Optical coherence tomography–
near infrared spectroscopy system and catheter for intravascular
imaging,” Opt Express 21(25), 30849–30858 (2013).
112. H. Wang, J. Gardecki, G. Ughi, P. Jacques, E. Hamidi, and G. Tearney,
“Ex vivo catheter-based imaging of coronary atherosclerosis using
422 Chapter 7
multimodality OCT and NIRAF excited at 633 nm,” Biomed Opt

Express 6(4), 1363–1375 (2015).
113. M. W. Jenkins, R. S. Wade, Y. Cheng, A. M. Rollins, and I. R. Efimov,
“Optical Coherence Tomography Imaging of the Purkinje Network,”
J Cardiovasc Electrophysiol (2005).
114. M. Gupta, A. M. Rollins, J. A. Izatt, and I. R. Efimov, “Imaging of the
atrioventricular node using optical coherence tomography,” J Cardio-
vasc Electrophysiol 13(1), 95 (2002).
115. W. Hucker, C. Ripplinger, C. P. Fleming, V. Fedorov, A. M. Rollins,
and I. R. Efimov, “Bimodal Biophotonic Imaging of the Structure-
Function Relationship in Cardiac Tissue,” J Biomed Opt 13(5), 054012
(2008).
116. C. P. Fleming, C. Ripplinger, B. Webb, I. R. Efimov, and A. M. Rollins,
“Quantification of Cardiac Fiber Orientation Using Optical Coherence
Tomography,” J Biomed Opt, 13(3), 030505 (2008).
117. Y. Gan and C. P. Fleming, “Extracting three-dimensional orientation
and tractography of myofibers using optical coherence tomography,”
Biomed Opt Express 4(10), 2150–2165 (2013).
118. C. M. Ambrosi, N. Moazami, A. M. Rollins, and I. R. Efiomv, “Virtual
histology of the human heart using optical coherence tomography,”
i 14(5), 054002 (2009).
119. C. P. Fleming, N. Rosenthal, A. M. Rollins, and M. Arruda, “First
in vivo Real-Time Imaging of Endocardial Radiofrequency Ablation by
Optical Coherence Tomography: Implications on Safety and The Birth
of ‘‘Electro-structural’’ Substrate-Guided Ablation,” Innovations in
Cardiac Rhythm Management 2, 199–201 (2011).
120. C. P. Fleming, K. J. Quan, H. Wang, G. Amit, and A. M. Rollins,
“In vitro characterization of cardiac radiofrequency ablation lesions using
optical coherence tomography,” Opt Express, 18(3), 3079–3092 (2010).
121. C. P. Fleming, K. J. Quan, and A. M. Rollins, “Toward guidance of
epicardial cardiac radiofrequency ablation therapy using optical coher-
ence tomography,” J Biomed Opt 15(4), 041510 (2010).
122. C. P. Fleming, H. Wang, K. J. Quan, and A. M. Rollins, “Real-time
monitoring of cardiac radio-frequency ablation lesion formation using
an optical coherence tomography forward-imaging catheter,” J Biomed
Opt 15(3), 030516 (2010).
123. C. J. Goergen, H. Radhakrishnan, S. Sakadžić, E. T. Mandeville,
E. H. Lo, D. E. Sosnovik, and V. J. Srinivasan, “Optical coherence
tractography using intrinsic contrast,” Opt Lett 37(18), 3882–3884
(2012).
124. C. Fan and G. Yao, “Imaging myocardial fiber orientation using
polarization sensitive optical coherence tomography,” Biomed Opt
Express 4(3), 460–465 (2013).
125. Y. Wang, K. Zhang, N. B. Wasala, X. Yao, D. Duan, and G. Yao,

“Histology validation of mapping depth-resolved cardiac fiber orienta-
tion in fresh mouse heart using optical polarization tractography,”
Biomed Opt Express 5(8), 2843–2855 (2014).
126. Y. Wang and G. Yao, “Optical tractography of the mouse heart using
polarization-sensitive optical coherence tomography,” Biomed Opt
Express 4(11), 2542–2545 (2013).
127. M. W. Jenkins, P. Patel, H. Deng, M. M. Montano, M. Watanabe, and
A. M. Rollins, “Phenotyping transgenic embryonic murine hearts using
optical coherence tomography,” Applied Optics 46(10), 1776–1781
(2007).
128. M. W. Jenkins, O. Q. Chughtai, A. N. Basavanhally, M. Watanabe, and
A. M. Rollins, “In vivo gated 4D imaging of the embryonic heart using
optical coherence tomography,” J Biomed Opt 12(3), 030505 (2007).
129. M. Gargesha, M. W. Jenkins, D. L. Wilson, and A. M. Rollins, “High
temporal resolution OCT using image-based retrospective gating,” Opt
Express 17(13), 10786–10799 (2009).
130. J. Männer, L. Thrane, K. Norozi, and T. M. Yelbuz, “High-resolution
in vivo imaging of the cross-sectional deformations of contracting
embryonic heart loops using optical coherence tomography,” Develop-
mental Dynamics 237(4), 953–961 (2008).
131. P. Li, A. Liu, L. Shi, X. Yin, S. Rugonyi, and R. K. Wang, “Assessment
of strain and strain rate in embryonic chick heart in vivo using tissue
Doppler optical coherence tomography,” Physics in Medicine and
Biology 56(22), 7081–7092 (2011).
132. S. Yazdanfar, M. D. Kulkarni, and J. A. Izatt, “High Resolution
Imaging of in vivo Cardiac Dynamics Using Color Doppler Optical
Coherence Tomography,” Opt Express 1, 424–431 (1997).
133. V. X. D. Yang, M. L. Gordon, B. Qi, J. Pekar, S. Lo, E. Seng-Yue,
A. Mok, B. C. Wilson, and I. A. Vitkin, “High speed, wide velocity
dynamic range Doppler optical coherence tomography (Part II):
Imaging in vivo cardiac dynamics of Xenopus laevis,” Opt Express
11(14), 1650–1658 (2003).
134. A. Davis, J. Izatt, and F. Rothenberg, “Quantitative Measurement of
Blood Flow Dynamics in Embryonic Vasculature Using Spectral
Doppler Velocimetry,” The Anatomical Record 292, 311–319 (2009).
135. I. V. Larina, N. Sudheendran, M. Ghosn, J. Jiang, A. Cable, K. V.
Larin, and M. Dickinson, “Live imaging of blood flow in mammalian
embryos using Doppler swept-source optical coherence tomography,”
J Biomed Opt. 13(6), 060506 (2008).
136. S. Rugonyi, C. Shaut, A. Liu, K. Thornburg, and R. K. Wang,
“Changes in wall motion and blood flow in the outflow tract of chick
embryonic hearts observed with optical coherence tomography after
424 Chapter 7
outflow tract banding and vitelline-vein ligation,” Physics in Medicine

and Biology 53(18), 5077–5091 (2008).
137. I. V. Larina, S. Ivers, S. Syed, M. E. Dickinson, and K. V. Larin,
“Hemodynamic measurements from individual blood cells in early
mammalian embryos with Doppler swept source OCT,” Opt Lett 34(7),
986–988 (2009).
138. M. Liebling, A. S. Forouhar, M. Gharib, S. E. Fraser, and M. E.
Dickinson, “Four-dimensional cardiac imaging in living embryos via
postacquisition synchronization of nongated slice sequences,” J Biomed
Opt 10(5) 054001–10 (2005).
139. N. R. Munce, A. Mariampillai, B. A. Standish, M. Pop, K. J. Anderson,
G. Y. Liu, T. Luk, B. K. Courtney, G. A. Wright, I. A. Vitkin, and V. X.
D. Yang, “Electrostatic forward-viewing scanning probe for doppler
optical coherence tomography using a dissipative polymer catheter,” Opt
Lett 33(7), 657–9 (2008).
140. J. Wu, M. Conry, C. Gu, F. Wang, Z. Yaqoob, and C. Yang, “Paired-
angle-rotation scanning optical coherence tomography forward-imaging
probe,” Opt Lett 31(9), 1265–7 (2006).
141. X. Liu, M. J. Cobb, and Y. Chen, “Rapid-scanning forward-imaging
miniature endoscope for real-time optical coherence tomography,” Opt
Lett 29(15), 1763–5 (2004).
142. X. D. Li, X. M. Liu, Y. C. Chen, M. J. Cobb, and M. B. Kimmey.
“Development of a Fast Scanning Miniature Probe and Methods of
Dispersion Management for High-resolution Optical Coherence Tomog-
raphy,” in IEEE EMBS. San Francisco, CA USA (2004).
143. P. R. Herz, Y. Chen, A. D. Aguirre, K. Schneider, P. Hsiung, and J. G.
Fujimoto, “Micromotor endoscope catheter for in vivo, ultrahigh
resolution optical coherence tomography,” Opt Lett 29(19), 2261–3
(2004).
144. Y. Wang, M. Bachman, G.-P. Li, S. Guo, B. J. F. Wong, and Z. Chen,
“Low-voltage polymer-based scanning cantilever for in vivo optical
coherence tomography,” Opt Lett 30(1), 53–5 (2005).
145. K. H. Kim, B. H. Park, G. N. Maguluri, T. W. Lee, F. J.
Rogomentich, M. G. Bancu, B. E. Bouma, J. F. de Boer, and J. J.
Bernstein, “Two-axis magnetically-driven MEMS scanning catheter for
endoscopic high-speed optical coherence tomography,” Opt Express
15(26), 18130–40 (2007).
146. P. Tran, D. Mukai, M. Brenner, and Z. Chen, “In vivo endoscopic
optical coherence tomography by use of a rotational microelectrome-
chanical system probe,” Opt Express 29(11), 1236–1238 (2004).
147. A. M. Rollins, R. Ung-Arunyawee, A. Chak, R. C. K. Wong,
K. Kobayashi, J. M.V. Sivak, and J. A. Izatt, “Real-time in vivo
imaging of human gastrointestinal ultrastructure using endoscopic
optical coherence tomography with a novel efficient interferometer

design,” Opt Lett 24, 1358–1360 (1999).
148. J. Xi, L. Huo, Y. Wu, M. J. Cobb, J. H. Hwang, and X. Li, “High-
resolution OCT balloon imaging catheter with astigmatism correction,”
Opt Lett 34(13), 1943 (2009).
149. A. R. Tumlinson, B. Povazay, L. P. Hariri, J. McNally, A. Unterhuber,
B. Hermann, H. Sattmann, W. Drexler, and J. K. Barton, “In vivo
ultrahigh-resolution optical coherence tomography of mouse colon with
an achromatized endoscope,” J Biomed Opt 11(6), 064003 (2006).
150. B. J. Vakoc, M. Shishko, S. H. Yun, W.-Y. Oh, M. J. Suter, A. E.
Desjardins, J. A. Evans, N. S. Nishioka, G. J. Tearney, and B. E.
Bouma, “Comprehensive esophageal microscopy by using optical
frequency-domain imaging,” Gastrointestinal Endoscopy 65(6), 898–905
(2007).
151. M. J. Suter, B. J. Vakoc, P. S. Yachimski, M. Shishkov, G. Y. Lauwers,
M. Mino-Kenudson, B. E. Bouma, N. S. Nishioka, and G. J. Tearney,
“Comprehensive microscopy of the esophagus in human patients with
optical frequency domain imaging,” Gastrointest. Endosc. 68(4), 745
(2008).
152. D. C. Adler, C. Zhou, T.-H. Tsai, J. Schmitt, Q. Huang, H. Mashimo,
and J. G. Fujimoto, “Three-dimensional endomicroscopy of the human
colon using optical coherence tomography,” Opt Express 17(2), 784
(2009).
153. M. V. SivakJr., K. Kobayashi, J. A. Izatt, A. M. Rollins, R. Ung-
Runyawee, A. Chak, R. C. Wong, G. A. Isenberg, and J. Willis, “High-
resolution endoscopic imaging of the GI tract using optical coherence
tomography,” Gastrointest Endosc 51(4 Pt 1), 474–9 (2000).
154. A. Das, M. V. Sivak, A. Chak, R. C. K. Wong, V. Westphal, A. M.
Rollins, J. A. Izatt, G. A. Isenberg, and J. Willis, “Role of High-
Resolution Endoscopic Imaging using Optical Coherence Tomography
(OCT) in Patients With Barrett’s Esophagus (BE),” Gastrointestinal
Endoscopy 51, AB93, Part 2 (2000).
155. K. Liang, G. Traverso, H.-C. Lee, O. O. Ahsen, Z. Wang, B. Potsaid,
M. Giacomelli, V. Jayaraman, R. Barman, A. Cable, H. Mashimo,
R. Langer, and J. G. Fujimoto, “Ultrahigh speed en face OCT capsule
for endoscopic imaging,” Biomed Opt Express 6(4), 1146–1163 (2015).
156. M. J. Gora, J. S. Sauk, R. W. Carruth, K. A. Gallagher, M. J. Suter,
N. S. Nishioka, L. E. Kava, M. Rosenberg, B. E. Bouma, and G. J.
Tearney, “Tethered capsule endomicroscopy enables less invasive
imaging of gastrointestinal tract microstructure,” Nature Medicine 19,
238–240 (2013).
157. F. T. Nguyen, A. M. Zysk, E. J. Chaney, J. G. Kotynek, U. J. Oliphant,
F. J. Bellafiore, K. M. Rowland, P. A. Johnson, and S. A. Boppart,
426 Chapter 7
“Intraoperative Evaluation of Breast Tumor Margins with Optical

Coherence Tomography,” Cancer Research 69(22), 8790–8796 (2009).
158. C. P. Fleming, J. Eckert, E. F. Halpern, J. A. Gardecki, and G. J.
Tearney, “Depth Resolved Detection of Lipid using Spectroscopic
Optical Coherence Tomography,” Biomed Opt Express 4(8), 1269–1284
(2013).
159. M. Kulkarn and J. A. Izatt, “Spectroscopic optical coherence
tomography,” Conference on Lasers and Electro-Optics, CLEO 59–
60 (1996).
160. R. N. Graf, F. E. Robles, X. Chen, and A. Wax, “Detecting
precancerous lesions in the hamster cheek pouch using spectroscopic
white-light optical coherence tomography to assess nuclear morphology
via spectral oscillations,” J Biomed Opt 14(6), 064030 (2009).
161. A. E. Desjardins, B. J. Vakoc, G. J. Tearney, and B. E. Bouma,
“Backscattering spectroscopic contrast with angle-resolved optical
coherence tomography,” Opt Lett 32(21), 3158–3160 (2007).
162. C. Xu, P. S. Carney, and S. A. Boppart, “Wavelength-dependent
scattering in spectroscopic optical coherence tomography,” Opt Express
13(14), 5450–5462 (2005).
163. M. Lazebnik, D. Marks, K. Potgieter, R. Gillette, and S. Boppart,
“Functional optical coherence tomography for detecting neural
activity through scattering changes,” Opt Letters 28(14), 1218–1220
(2003).
164. X. Yao, A. Yamauchi, B. Perry, and J. George, “Rapid optical
coherence tomography and recording functional scattering changes from
activated frog retina,” Applied Opt 44(11), 2019–2023 (2005).
165. B. Graf, T. Ralston, H. Ko, and S. Boppart, “Detecting intrinsic
scattering changes correlated to neuron action potentials using optical
coherence imaging,” Opt Express 17(16), 13447–13457 (2009).
166. T. Akkin, C. Joo, and J. de Boer, “Depth-resolved measurement of
transient structural changes during action potential propagation,”
Biophysical Journal 93(4), 1347–1353 (2007).
167. S. A. Boppart, “Advances in contrast enhancement for optical coherence
tomography,” Conference Proceedings IEEE Engineering in Medicine
and Biology Society 1, 121–4 (2006).
168. S. Boppart, A. Oldenburg, C. Xu, and D. Marks, “Optical probes and
techniques for molecular contrast enhancement in coherence imaging,”
Jf Biomed Opt 10(4), 41208 (2005).
169. S. Tang, T. Krasieva, Z. Chen, and B. Tromberg, “Combined
multiphoton microscopy and optical coherence tomography using a
12-fs broadband source,” J Biomed Opt 11(2), 020502 (2006).
170. S. Yazdanfar, L. Laiho, and P. So, “Interferometric second harmonic
generation microscopy,” Opt Express 12(12), 2739–2745 (2004).
171. B. E. Applegate, C. Yang, A. M. Rollins, and J. A. Izatt, “Polarization

resolved second harmonic generation optical coherence tomography in
collagen,” Opt Lett 29(19), 2252–2254 (2004).
172. C. A. Patil, N. Bosschaart, M. D. Keller, T. G. van Leeuwen, and
A. Mahadevan-Jansen, “Combined Raman spectroscopy and optical
coherence tomography device for tissue characterization,” Opt Lett
33(10), 1135–1137 (2008).
173. J. Evans, R. Zawadzki, R. Liu, J. Chan, S. Lane, and J. Werner,
“Optical coherence tomography and Raman spectroscopy of the ex vivo
retina,” J Biophotonics 2(6-7), 398–406 (2009).
Christine P. Hendon is an assistant professor in the

Department of Electrical Engineering at Columbia Univer-
sity. She is also the principal investigator of the Structure
Function Imaging Laboratory. Her research interests include
biomedical optical imaging and spectroscopy system
development, image processing, and cardiac electrophysiol-
ogy. Dr. Hendon is a member of SPIE and BMES.
She was awarded the NSF CAREER and NIH New
Innovator Awards.
Andrew M. Rollins is a professor of Biomedical Engineering

at Case Western Reserve University. His research interests
include the development and application of advanced
biomedical optical technologies, especially optical coherence
tomography (OCT), and including optical stimulation and
imaging of electrophysiology. His recent projects apply these
technologies to the study of developmental cardiology and
endoscopic imaging of cardiovascular disease and cancer.
He is a member of SPIE, OSA, BMES, AHA, and IEEE, and a Fellow of
AIMBE.
Chapter 8
Speckle Technologies for
Monitoring and Imaging
Tissues and Tissue-Like
Phantoms
Dmitry A. Zimnyakov and Olga V. Ushakova
Saratov Technical University, Saratov, Russia
David J. Briers
Kingston University, London, UK
Valery V. Tuchin
Saratov, Russia
8.1 Introduction
The appearance of random interference patterns, or speckle fields, in the spatial
distributions of laser light scattered by weakly ordered media such as tissues, is
probably the most evident and simply observed manifestation of coherence
phenomena in tissue-light interactions. In the case of dynamic light scattering,
when temporal speckle intensity fluctuations are induced by stochastic or
regular motions of the scattering structure elements in the probed tissue volume,
it is reasonable to try to apply the statistical analysis of these fluctuations to the
monitoring or functional imaging of the tissue structure or the dynamics. As a
result, laser speckle methods have recently become one of the most universally
adopted and familiar technologies in biology and medicine (e.g., laser Doppler
flowmetry is an example of such an analysis of dynamic laser speckle as applied
to the monitoring and functional imaging of in vivo blood microcirculation).
429
430 Chapter 8
The most typical applications of the statistical analysis of static and

dynamic laser speckles for the monitoring and functional imaging of tissue
structure and dynamics are considered in this chapter; these applications are
based on the various physical principles that determine the relations between
the statistical characteristics of the detected speckle intensity fluctuations and
the dynamical and structural properties of the probed tissue.1–3
Early pioneering works on the use of diffusing-wave spectroscopy
technologies for blood flow monitoring and burned tissue diagnostics are
discussed in Section 8.2. These technologies are based on the correlation
analysis of the time-varying multiply scattered speckle fields induced by the
motions of erythrocytes in the probed tissue volume. Another basic technique
for monitoring and imaging tissue motion such as blood flow that is
concerned with the statistical analysis of the spatial fluctuations of the time-
averaged dynamic speckles is considered in Section 8.3. This full-field method,
known as laser speckle contract analysis (LASCA), offers certain advantages
such as high performance of preliminary data processing and functional image
reconstruction due to the “quasi-parallel” mode of the analyzed speckle image
formation. Section 8.4 is dedicated to full-field techniques with the additional
spatial filtration of the scattered optical field; this approach gives the
possibility to improve the depth resolution of the blood microcirculation
analysis. Such techniques can also be used for the characterization of
the optical properties of tissue phantoms and tissues. Various modifications
of the LASCA method and corresponding data processing algorithms
recently applied in the biomedical diagnostics are discussed in Section 8.5.
Potential applications of the contrast analysis of speckles produced by the
multiple scattering of partially coherent light to the visualization of
macroscopically inhomogeneous weakly ordered media or characterization
of their optical properties are discussed in Section 8.6. Finally, the possibility
of using a full-field speckle instrument with fiber-optical light-delivering
channels for the monitoring of thermally-induced tissue changes is considered
in Section 8.7.
8.2 Diffusing-Wave Spectroscopy (DWS) as a Tool for Tissue

Structure and Cell Flow Monitoring
Random phase shifts, accumulated by each component of a coherent optical
field propagating in disordered nonstationary scattering media due to
statistically independent sequences of the scattering events, will cause complex
spatial-temporal distributions of scattered light intensity. These distributions,
or random time-varying interference patterns, or dynamic speckle patterns,
actually mirror the inner dynamic and structure peculiarities of the scattering
media and thus can be used for the purposes of noninvasive diagnostics and
imaging of various scattering systems probed by coherent light. In the case of
small relative velocities of scattering sites that form the probed media (this is
Speckle Technologies for Monitoring and Imaging Tissues and Tissue-Like Phantoms 431
typical for the vast majority of biological systems such as individual organelles
and cells, cell aggregates, and whole tissues), the corresponding Doppler shifts
of the individual components of the scattered light will be negligibly small
compared with the probe light frequency. Thus, the resulting spectral
broadening of the probe light passing through the scattering system is usually
classified as a result of quasi-elastic light scattering (QELS) and can be
detected only by means of so-called light-beating spectroscopy.4
QELS methods as applied to the monitoring of dynamic systems are based
mainly on the correlation or spectral analysis of the temporal fluctuations of
the speckle intensity at a fixed detection point; these fluctuations are caused by
the motions of scattering centers in the probed volume. In the case of single
scattering disordered media (i.e., when the scattered optical field can be
considered as the result of the superposition of the statistically independent
components and each of them appears due only to one scattering event), the
spectral width of the speckle intensity fluctuations depends on the scattering
angle that characterizes the detector position with respect to the probe beam
axis and the dynamic parameters of the scattering ensemble. In particular, for
disordered systems of scattering particles that undergo Brownian motion, the
spectral density of the speckle intensity fluctuations has a typical Lorentzian
form with its halfwidth evaluated as DT jq̄j2 ; DT is the self-diffusion coefficient
of the scattering particles; and jq̄j is the momentum-transfer module equal to
(4pn/l)sin(u/2), where n is the refractive index of the scattering medium, u is
the scattering angle, and l is the wavelength of the probe light in free space.
Thus, measurement of the spectral width of the speckle intensity fluctuations
for a given scattering angle allows the determination of the value of the self-
diffusion coefficient of the scattering centers. A similar technique, known as
light-beating spectroscopy or correlation spectroscopy (if the correlation time of
the speckle intensity fluctuations is determined during the QELS experiment) is
widely used for various biophysical and medical applications.4,5
When laser light is used to probe an optically thick medium consisting of
relatively closely packed scattering centers, each contribution to the scattered
optical field should be considered as a result of at least a few scattering
events and, consequently, this single-scattering approach cannot be used to
determine the parameters of the scattering system. In this case, to find the
relationship between the parameters of the experimentally obtained speckle
intensity fluctuations I(t) and the parameters describing the dynamics of the
scattering sites in the probed volume, the following assumptions are usually
made:
1. The temporal autocorrelation function of the scattered field fluctuations
g1(t) ¼ 〈E(t þ t)E*(t)〉/〈|E(t)|2〉 and the corresponding normalized inten-
sity autocorrelation function g2(t) ¼ 〈I(t þ t)I(t)〉/〈I(t)2〉 are related to
each other by the Siegert relation4
g2 ðtÞ ¼ 1 þ bjg1 ðtÞj2 , (8.1)
432 Chapter 8
where the coefficient b is determined by the detection conditions; for an

“ideal” detector with a “pinhole” aperture, i.e., one that is significantly
smaller than the characteristic correlation area of the spatial fluctuations
of the scattered field associated with the average speckle size, b ¼ 1.
The Siegert relation [Eq. (8.1)] is valid for Gaussian scattered fields with
zero mean values of the field amplitude induced by multiply scattering
disordered systems consisting of a large number of noninteracting
elementary scatterers.
2. During the observation time, the scattering object can be considered as a
stationary and ergodic stochastic system.
3. The volume density of the elementary scatterers is relatively small, i.e., the
mean values of the scattering length and the transport mean-free path
characterizing the scattering medium are significantly larger than the
wavelength used (l,l*≫l) and also l,l*≫a, where a is the effective size of
the scattering particles; thus, we can neglect the contributions of the effects
of spatial correlations to the scattered field formation.
For these assumptions, in terms of a discrete scattering model, the
amplitude of the scalar scattered field at the arbitrary selected observation
point can be expressed as the sum of partial contributions (Fig. 8.1); each
contribution appears as a result of a sequence of statistically independent
scattering events (see, e.g., Ref. 6),
Y
Mi
E si ðtÞ ¼ expðjvtÞ aik expfj q̄ik r̄ik ðtÞg, (8.2)
k
where E0,v are the probe incident light amplitude and frequency, respectively,
Mi is the number of scattering events for the i’th scattering sequence, aij is
determined by the amplitude of the scattered light for the j’th scattering
Figure 8.1 Typical scheme of a QELS experiment on a multiply scattering medium.

event in the i’th scattering sequence; q̄ij is the momentum transfer for the j’th
scattering event; and r̄ij ðtÞ is the current position of the j’th scattering site.
It is necessary to note that each partial contribution can be associated with
the effective optical path in the scattering volume and can, therefore, be
characterized by the path length value defined as si Mil.
Expressing the path-dependent field correlation function as
Gs1 ðtÞ ¼ hE si ðt þ tÞE si ðtÞi,
and carrying out the averaging over the pairs of noncorrelated scattering
events, one can obtain
Y
Mi
G s1 ðtÞ ¼ jaik j expðjvtÞ
2
expfjDq̄ik Dr̄ik ðtÞg .
k
For each scattering event, the momentum transfer Dq̄ik and the
displacement of the elementary scatterer Dr̄ik for the delay time t can be
considered as statistically independent vectors and, consequently, we can
express g1(s)(t) ¼ G1s(t)/G1s(0) as

ðsÞ 1 2 2 s
g1 ðtÞ ¼ exp k 0 hDr ðtÞi , (8.3)
3 l
where the factor exp(jvt) is excluded without loss of generality, and k0 is the
wavenumber of the probe light in the scattering medium.
To obtain the temporal correlation function of the observed field,
averaging over the ensemble of partial contributions should be done; the final
form of the normalized temporal correlation function can be obtained by
introducing the probability density of the effective optical paths and replacing
the discrete path statistics by continuous statistics:
`
g1 ðtÞ ¼ G1 ðtÞ∕G1 ð0Þ ¼ ∫ rðsÞ exp k 0 hDr ðtÞi ds.
1 2 2 s
(8.4)
0 3 l
The path length statistics are determined by the scattering medium

geometry as well as by the illumination and detection conditions; for the
diffusion mode of light propagation through the probed object, and for a
limited number of scattering systems, the temporal correlation function of the
field fluctuations can be obtained in analytical form. In this case, the path
length density can be obtained from the solution of the nonstationary
diffusion equation describing the propagation of a d-like light pulse emitted by
the source with a given geometry.
The simplest and most basic solution is that describing the pulse
propagation in an infinite isotropic statistically homogeneous scattering
434 Chapter 8
medium between a point-like source and a detector separated by a distance jr̄j

(see, e.g., Ref. 7),

3 3∕2
rðsÞ ¼ exp½3jr̄j2 ∕4l s: (8.5)
4pl s
This expression, considered as the Green function of the light diffusion
equation, allows the determination of the path distributions for the given
illumination and detection conditions; substitution in Eq. (8.4) leads to the
following expression for the field correlation function:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
g1 ðtÞ ¼ expð 6tDT k 20 jr̄j∕l Þ. (8.6)
In the absence of absorption, g1(t) exhibits non analytical behavior

(“nondiffererentiability”) in the vicinity of t ¼ 0; this is a direct consequence
of the presence of a slowly decaying “tail” of the path length distribution for
such a scattering system [r(s) s1.5 for large s]. As a result, this leads to
unlimited values for the mean path length and the higher-order statistical
moments of the path length distribution. Additional absorption in the
scattering system, expressed as the presence of a Bougier factor exp(mas) in
the expression for the path length density (ma is the absorption coefficient of
the scattering medium), causes the suppression of the long-path tail in the s
distribution and, consequently, leads to the following form of the field
correlation function:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
g1 ðtÞ ¼ expð 3ma l þ 6k 20 DT tjr̄j∕l Þ. (8.7)
In the presence of boundaries and the use of extended coherent light

sources for media probing, the probability density r(s) can be obtained by a
modification of the above solution for an infinite medium and a point-like
source [see Eq. 8.5)]. Thus, for an infinite slab and a semi-infinite medium
illuminated by a point-like source, r(s) is expressed by using the images
method as a series of Green functions for the diffusion equation for an infinite
medium. A wide range of analytical forms of path length densities obtained
as the corresponding solutions of the light diffusion equation for given
geometries of the scattering medium are presented in Ref. 8. Here, we present
only the three most typical r(s) distributions that are the most important from
the viewpoint of applications in biology and medicine:
1. semi-infinite scattering medium and point-like illuminating source;
backscattered light is detected:

lzc 3jr̄j2
rðsÞ ¼ exp ma s þ ;
ð4pl ∕3Þ1.5 s2.5 4l s
2. infinite scattering slab and point-like illuminating source; forward

scattered light is detected:
h i
c exp ma s þ 3ðjr̄j4lþl X̀
2 2Þ
z
s 3ðzþk Þ2
rðsÞ ¼ zþk exp
ð4pl ∕3Þ1.5 s2.5 k¼0
4l s

3ðzk Þ2
zk exp ;
4l s
zþk ¼ ð2 k þ 1ÞL þ l z ; zk ¼ ð2 k þ 1ÞL l z ;
3. infinite scattering slab and point-like illuminating source; backscattered

light is detected:
h i
cl z exp ma s þ 3ðjr̄j4l þl X̀
2 2Þ
z
s 3ðzþk0 Þ2
rðsÞ ¼ zþk0 exp
ð4pl ∕3Þ1.5 s2.5 k¼0
4l s

3ðzk0 Þ2
zk0 exp ; zþk0 ¼ 2kL þ l z ; zk0 ¼ 2kL l z :
4l s
Here, L is the slab thickness, c is the velocity of light in the scattering

medium, and lz is the depth of the source position in the scattering medium,
which is determined by the boundary conditions and is of the order of the mean
scattering length for the probed medium. It can be shown that the path length
density distribution for the probe light backscattered from an infinite slab
asymptotically approaches that for the case of backscattering from a semi-
infinite medium as the slab thickness increases. This is a simple manifestation of
the fact that r(s) becomes insensitive to the presence of the second boundary
when the slab thickness becomes much larger than the source-detector
separation. To find the dependence of the field correlation function on the
time delay, the appropriate dynamic model should be chosen to describe the
dynamics of the scattering particles. Usually two classical models are used for
the analysis of QELS results from living biological systems:
1. Brownian-like motions of scattering particles; in this case
hDr̄2 ðtÞi ¼ 6DT t, where DT is the translation diffusion coefficient of the
scattering particle;
2. “Random flow” of the scattering particles; this kind of motion
corresponds to hDr̄2 ðtÞi ¼ v2 t2 , where v is the velocity of the scattering
particles.
For a wide variety of real nonstationary systems, the motion of scattering
particles can be considered as combinations of these types of dynamics with
differing time scales.
436 Chapter 8
Figure 8.2 Typical schemes of QELS probing of structural or dynamic inhomogeneities

embedded in a multiply scattering non-stationary medium, for the cases of (a) backscattering
and (b) transillumination geometry.
Thus, the location of dynamic or structural macro-inhomogeneities

embedded in a highly scattering medium by means of the DWS technique is
based on the evaluation of the correlation time of the scattered field
fluctuations or the slope of the normalized correlation function g1(t), taking
into account the mutual positions of the light source, the detector, and the
located inhomogeneity. Typical schemes of speckle correlation probes for
biomedical applications are presented in Fig. 8.2 for the cases of (a) back-
scattering and (b) transillumination geometry. Laser diode 1 is used as an
illumination source; laser light is delivered to the probed object 3 through the
multimode fiber 2; traveling into the object, it passes through the macro-
inhomogeneity 4; scattered light is collected by the single-mode fiber 5 and
detected by the photomultiplier tube 6 operating in a photon-counting mode.
The detected signal is processed by the digital autocorrelator 7 to obtain the
intensity correlation function g2(t) for a given separation between source and
detector and their position with respect to the located inhomogeneity.
The field correlation function g1(t) can be calculated from the measured
intensity correlation function by use of the Siegert relation of Eq. (8.1). The
solution of the inverse problem based on correlation measurements can be
carried out in two stages:
• reconstruction of the probability density function of the effective optical
paths as the inverse Laplace transform of the experimentally obtained
field correlation function g1(t);
• solution of the tomographic problem based on the reconstructed set of

probability density distributions rj(s) for various relative positions of the
located inhomogeneity, the light source, and the light-collecting fiber.
It is necessary to note that the second stage of this problem (reconstruction
of the form and evaluation of the size and position of the located
inhomogeneity with respect to the chosen origin) is closely related to
similar problems associated with various diffuse-light methods developed for
biomedical applications (such as time-domain and frequency-domain
techniques, NIR transillumination imaging with CW light sources, etc.).
Characteristic changes of the field correlation function of the detected
speckle intensity fluctuations in the case of DWS probing of dynamic
inhomogeneities can be illustrated by experimental results obtained for a
phantom scatterer and presented in Ref. 9 (Fig. 8.3).
A semi-infinite, highly scattering slab of TiO2 suspended in resin was used
as the scatterer. The slab contained a spherical cavity filled with a turbid,
fluctuating suspension of 0.296-mm polystyrene balls. An Ar-ion laser
was used as the illumination source. The graphs in Fig. 8.3 show the
experimentally obtained temporal-field autocorrelation functions and the
corresponding theoretical curves obtained by using the diffusion approxima-
tion for different positions of the source fiber and the detector fiber with
respect to the spherical cavity (S1,D1; S2;D2;S3,D3). The source–detector
separation was kept fixed at 1.75 cm. With respect to an xy coordinate system
whose origin lies directly above the center of the spherical cavity, the source–
detector axis was aligned to the y axis with the source at y ¼ 1.0 cm and the
detector at y ¼ 0.75 cm.
Figure 8.3 Experimental measurements of the normalized temporal field autocorrelation

function for a phantom scatterer with dynamic inhomogeneity (reprinted from Ref. 9).
438 Chapter 8
Figure 8.4 Burn depth diagnostics by means of speckle correlation measurements.

1: source fiber; 2: detector fiber; 3: burned tissue; 4: normal tissue.
Measurements were made at x ¼ 0 cm (S1,D1), x ¼ 1.0 cm (S2,D2), and

x ¼ 2.0 cm (S3,D3), indicated as o, þ , and □. It is clear that larger and more
rapid decays are observed when the source and detector are closest to the
dynamic sphere.
A typical example of the potential applications of the DWS technique in
medicine is burn depth diagnostics. The main idea of this approach is based
on the well-known dependence of the penetration of the light propagation
paths on the source–detector separation in the case of backscattering
(Fig. 8.4).
For such a configuration, the regions of maximum density of light paths
have a typical “banana-like” shape and the penetration depth for each
“banana” can be expressed as10
pffiffiffi
z d∕2 2, (8.8)
where d is the source–detector separation.

When the probe light propagates through the burned layer, the lack of
blood microcirculation means that the scattering is predominantly from
stationary scatterers. As a result, there is only a slow decay of the correlation
function of scattered light with increasing t. Thus, in the case of surface burn
diagnostics by a pair of closely adjacent light-emitting and light-collecting
fibers (as shown in Fig. 8.5) only the upper, necrotic layer of the burned tissue
will be probed and the DWS technique will show a slowly decaying g1(t). But
with an increase in the distance between source and detector, the “banana-
shaped” region of the maximum concentration of the photon paths will reach
Figure 8.5 Schematic of burn depth diagnostics (by Boas et al. taken from Ref. 9).
the underlying layers of tissue where there is blood flow. This will be
manifested as an increase in the slope of the intensity correlation function and,
correspondingly, g1(t).
This promising possibility of DWS burn depth diagnostics was
demonstrated by D. Boas et al.9 using the pig burn model suggested by
Nishioka and Schomacker at the Wellman Institute in Boston (Fig. 8.5).
A He-Ne laser with an output power of 8 mW was used as the light source
in their experiments. The laser light was coupled into a multimode optical
fiber with a core diameter of 200 mm. This fiber delivered the probe light to
the burn surface being diagnosed; after passing through the layer of burned
tissue, the backscattered laser light was collected by the light-collecting system
consisting of a single-mode fiber assembly: several single-mode fibers were
positioned at different distances varying from 0.2 mm to 2.4 mm with respect
to the source fiber. The light-collecting fibers were connected to a photo-
detector [photon-counting photomultiplier tube (PMT)] via an electronically
controlled fiber optical switch. The PMT output signal was processed by a
digital autocorrelator to obtain the temporal correlation function for a given
burn depth and source-detector separation. The burn depth was controlled by
applying a hot metal block (100°C) to the surface of a pig skin for a given
duration. In this experiment, five different durations of burn with
consequently increasing depth (60–100 mm, 400–500 mm, 500–600 mm,
1500–2000 mm, and 2100–2200 mm) were used. The strong dependence of the
440 Chapter 8
Figure 8.6 Temporal field correlation functions obtained from 48-hour-old burns for a
source–detector separation of 800 mm (by Boas et al., reprinted from Ref. 9); the correlation
functions for burn duration times of 3 s (solid line), 5 s (dotted line), 7s (dashed line), 12 s
(dot-dash line) and 20 s (dot-dot-dot-dash line) are presented.
decay rate of the correlation function on the burn depth for a given source–
detector separation allows the different grades of tissue burns to be
distinguished (Fig. 8.6).
To summarize the data for all source–detector separations and to
produce the criteria for burn depth estimations, the following technique was
suggested: decay rates of the field correlation functions were determined
for 0 , t , 100 ms by fitting a line to the data, and these values of decay rate
were plotted against the source-detector separation. The tendencies of the
decay rate behavior can be summarized as follows (Ref. 9, Fig. 8.7): for
shallow burns, the decay rate increases linearly with the source–detector
separation as observed for healthy tissue and as would be expected for a
homogeneous system, i.e., the shallow burn does not perturb the correlation
function. On the contrary, for deeper burns, the decay rate is smaller and no
longer increases linearly with the source–detector separation.
One of the advantages of the developed DWS technique is that it makes it
possible to distinguish burns with thicknesses that vary by 100 mm. However,
it is necessary to have an adequate calibration technique in order to obtain
reliable diagnostic results. One possibility is to use the DWS technique in
combination with other optical technologies (e.g., with diffuse reflectance
measurements in the time domain and/or frequency domain).
In the detection of backscattered light, some analytical problems related
to the so-called diffusion approximation breakdown (see, e.g., Ref. 11) and
connected with an underestimation of ballistic and low-step scattered
contributions and the nonphysical behavior of the diffusion equation solution
cause strong discrepancies between experimentally observed and calculated
correlation functions of the detected intensity fluctuations.
Figure 8.7 Dependence of the decay rates of the field correlation functions on the source–
detector separations for different burn depths.
One of the promising approaches in the analytical calculation of the path

length statistics for a nondiffusing mode of light propagation in turbid media
is the application of a three-dimensional extension of the telegrapher equation
proposed by Lemieux et al. It has been found12 that this form of transport
equation can describe the behavior of a random walk in the vicinity of
boundaries and source more adequately than the “conventional” diffusion
equation.
In this case, the second-order time derivative on the right side of the
modified telegrapher equation can be associated with the nonscattered
(“ballistic”) and the low-step scattered components of the optical field
propagating in the disordered medium. To improve the agreement between
the analytical solutions of the modified telegrapher equation and the “exact”
solution of the radiative transfer equation obtained by a Monte-Carlo
simulation, Lemieux et al. considered the concept of an exponentially
distributed light source.
In the presence of boundaries (e.g., in the case of backscattering from a
semi-infinite medium or transmission through a slab), a corresponding
solution can be obtained by using the conventional method of images. It
should be noted that for large distances between source and detector as well as
for large propagation times, the path length density r(s) asymptotically
approaches the well-known solution of the diffusion equation describing the
path statistics of the diffuse components emitted by a point-like isotropic
source embedded in an infinite medium.
442 Chapter 8
8.3 Laser Speckle Contrast Analysis (LASCA) for Measuring

Blood Flow
8.3.1 Statistical properties of laser speckle
Although a detailed account of laser speckle statistics is outside the scope of
this chapter, it is worth mentioning at this point that the size of the individual
speckles has, in general, nothing to do with the structure of the surface
producing them. It is determined entirely by the aperture of the optical system
used to observe the speckle pattern. If the speckle pattern is being observed
directly by the human eye, it is the pupil of the eye that determines the speckle
size. More importantly, if a camera is used, it is the setting of the aperture stop
that determines the speckle size. This can have a serious effect if the aperture is
used to control the exposure of the photograph.
8.3.2 Time-varying speckle

When an object moves, the speckle pattern it produces changes. For small
movements of a solid object, the speckles move with the object, i.e., they
remain correlated. For larger motions, the speckles “decorrelate” and the
speckle pattern changes completely. Decorrelation also occurs when the light
is scattered from a large number of individual moving scatterers, such as
particles in a fluid. An individual speckle appears to “twinkle” like a star.
This phenomenon has come to be known as “time-varying speckle.” It is
frequently observed when living organisms are observed under laser light
illumination.
It is reasonable to assume that the frequency spectrum of the fluctuations
should be dependent on the velocity of the motion. It should, therefore, be
possible to obtain information about the motion of the scatterers from a study
of the temporal statistics of the speckle fluctuations. This analysis can be
based on the techniques of either photon correlation spectroscopy or laser
Doppler velocimetry. It is not intuitively obvious that time-varying speckle and
Doppler-induced fluctuations are identical. The theory of time-varying
speckle starts with the classical (though random) interference pattern
produced when light beams of the same frequency interfere. The fluctuations
are caused by the changes in optical path lengths of the interfering beams
caused by the movement of the scatterers. Doppler fluctuations, on the other
hand, are explained by the beating effect that occurs when two waves of
slightly different frequency are superimposed, the difference being due to the
frequency shift induced by the Doppler effect when light is scattered by a
moving object. Thus the speckle explanation is based on the superposition of
waves of the same frequency, whereas in the Doppler explanation, the
superimposed waves have different frequencies. Despite these apparent
differences in approach, it can be shown mathematically that the two
explanations lead to identical equations linking the fluctuations to the velocity
distribution of the scatterers.13 Thus, the two approaches are just different
ways of looking at the same physical phenomenon.
8.3.3 Full-field methods

One problem of using the temporal statistics of time-varying speckle is that
measurements are made at only one point in the speckle pattern (a single
speckle). If an area is to be analyzed, it is necessary to scan the detector over
the field. If a map of velocity distribution is required, some method of
scanning the area of interest is necessary. Such a map is of particular
importance if blood flow is to be used as a diagnostic tool. Fujii et al. used a
scanning technique in their application of time-differentiated speckle to
measure skin capillary blood flow.14 They used a linear charge-coupled device
(CCD) array to monitor simultaneously a line of speckles, and a scanning
mirror to extend this to a 2-D area. They later applied the technique to the
mapping of retinal blood flow.15–17 They produced a microcirculation map of
the retina by illuminating the retina with light from a diode laser, scanning
and storing the speckle images, and then calculating the differences between
successive images. The parameter they use is the ratio of the mean intensity to
the intensity difference, a quantity they call “normalized blur” and that is used
as a measure of velocity. Experiments on a rabbit eye showed good
correlation with invasive methods.16
Scanning has also been applied to the laser Doppler technique,18,19 and
commercial scanning Doppler systems are now on the market that can
provide full-field monitoring of capillary blood flow over quite large areas of
the body. The problem of either processing or storing vast amounts of data
demands a compromise between cost and performance, and the systems
offered suffer from both long scanning times and a loss of spatial resolution.
Nevertheless, they are useful as a diagnostic tool and offer easily interpreted
false-color maps of blood velocity and/or flow.
The scanning necessary to apply time-varying speckle techniques to map
velocity distributions results in the collection of a vast amount of data and
usually takes an appreciable time (typically five minutes for 256 256 pixels).
Ideally, a full-field technique would avoid the need for scanning. One such
technique is the “global Doppler” method.20 This converts velocity directly to
intensity (or false color) by means of the ingenious device of measuring how
much of the return light is absorbed by a substance of known absorption/
frequency properties. Unfortunately, resolution problems limit the technique
to fairly high velocities, and it is not suitable for biomedical applications.
There is another way to avoid the need to scan: by using the spatial
statistics (specifically the contrast) of time-integrated fluctuating speckle
patterns. If the integration time is comparable with the period of the
fluctuations, it is clear that the effect will be a reduction in the speckle
contrast.
444 Chapter 8
8.3.4 Single-exposure speckle photography

The use of such time-integrated speckle led in the early 1980s to a technique
for flow visualization that simultaneously achieved full-field operation
(without scanning) and very simple data gathering and processing. It was
originally called “single-exposure speckle photography”21 to distinguish it
from the then well-known technique of double-exposure speckle photography
for measuring displacement and strain. Single-exposure speckle photography
was developed primarily for the measurement of retinal blood flow.22,23 The
basic technique is simply to photograph the retina under laser illumination
using an exposure time that is of the same order as the decorrelation time of
the intensity fluctuations. It is clear that a very short exposure time will
“freeze” the speckle and result in a high-contrast speckle pattern, whereas a
long exposure time will allow the speckles to average out, leading to a low
contrast. In general, the velocity distribution in the field of view is mapped as
variations in speckle contrast. In the original technique, high-pass optical
spatial filtering of the resulting photographs converted these contrast
variations to more easily seen intensity variations.24 Later work introduced
digital image processing of the speckle photographs, including a color-coding
of the velocities.25
8.3.5 Laser speckle contrast analysis (LASCA)

By 1990, digital techniques were sufficiently advanced to allow the
photographic stage to be eliminated. By measuring the speckle contrast
directly and converting the data to a false-color image, a fully digital, real-
time technique for the mapping of skin capillary blood flow was developed.26
The technique was called “laser speckle contrast analysis” or LASCA.27,28
Some researchers still use this name,29,30 but others have introduced names
such as “laser speckle contrast imaging,”31 or “laser speckle imaging.”32 For
the sake of simplicity, we shall use the term LASCA in this paper, as this was
the name used in the original 1996 work.27
The experimental setup for LASCA is very simple, as shown in Fig. 8.8.
Diverging laser light illuminates the object under investigation, which is
imaged by a CCD camera (or equivalent). The image is captured by a frame
grabber (or equivalent) and the data passed to a personal computer for
processing by custom software. The operator usually has several options at his
disposal. In the original LASCA technique,27 this included the exposure time,
the number of pixels over which the local contrast was computed, the scaling
of the contrast map, and the choice of colors for coding the contrast. The
choice of the number of pixels over which to compute the speckle contrast is
important: too few pixels and the statistics will be compromised; too many
and spatial resolution is sacrificed. In practice, it is found that a square of
7 7 or 5 5 pixels is usually a satisfactory compromise. (A square with
Figure 8.8 Basic setup for laser speckle contrast analysis (LASCA).
sides of an odd number of pixels was chosen so that the computed contrast
could be assigned to the central pixel.) The speckle contrast is quantified
by the usual parameter of the ratio of the standard deviation to the mean
(s/,I.) of the intensities recorded for each pixel in the square [see Eq (8.9)].
The pixel square is then moved along by one pixel and the calculation
repeated: this overlapping of the pixel squares results in a much smoother
image than would be obtained by using contiguous squares, and at little cost
in terms of additional processing time. It must be remembered, though, that
this overlapping of the squares does not lead to an increase in resolution,
which is determined by the size of the square used: there is a trade-off between
spatial resolution and reliable statistics.
If the object under investigation contains moving scatterers, such as blood
cells, each speckle will be fluctuating in intensity. A time-integrated image,
therefore, shows a reduction in speckle contrast because of the averaging of
the intensity of each speckle over the integration time. In practice, the
exposure time can be very short, typically 0.02 seconds, and the processing
time is less than one second for the whole frame,33 making it effectively a real-
time technique.
8.3.6 The question of speckle size

A complicating factor for the technique is that of speckle size. As mentioned
in Section 8.3.1, this is determined entirely by the aperture of the imaging
system used (in this case the camera). The pixels of the detector are effectively
sampling the speckle pattern. If the speckles are smaller than the pixels, some
averaging will occur and the technique rapidly becomes less effective. If the
speckles are much larger than the pixels, only a few speckles will be sampled
by the square of pixels used, and the statistics will become unreliable. The best
446 Chapter 8
compromise is to arrange for the speckle size to be equal to the pixel size.34
This is achieved by choosing an appropriate setting for the aperture of the
camera lens. However, this removes the possibility of using the aperture to
control the amount of light reaching the detector, the usual function of a
variable lens aperture. Further, the shutter speed cannot be used to do this, as
it is determined by the need to match the velocities being measured, and some
other method of controlling the amount of light reaching the detector must be
found.
8.3.7 Theory
The principle of LASCA is very simple. A time-integrated image of a moving
object exhibits blurring. In the case of a laser speckle pattern, this appears as a
reduction in the speckle contrast, defined (and measured) as the ratio of
the standard deviation of the intensity to the mean intensity. This occurs
whatever is the “movement” of the speckle. For random velocity distribu-
tions, each speckle fluctuates in intensity. For lateral motion of a solid object,
on the other hand, the speckles also move laterally and become “smeared” on
the image, but a reduction in speckle contrast still occurs. For fluid flow, the
situation might be a combination of both these types of “movement.” In each
case, the problem for quantitative measurements is the establishment of a
relationship between speckle contrast and velocity (or velocity distribution).
It is clear that there must be a link between the flow velocity and the
amount of blurring. The higher the velocity, the faster are the fluctuations and
the more blurring occurs in a given integration time. By making certain
assumptions, the following mathematical relationship can be established
between speckle contrast and the temporal statistics of the fluctuating
speckle:21
¼ ∫ g̃2 ðtÞdt,
1
s2s ðTÞ (8.10)
T 0
where s2s is the spatial variance of the intensity in the speckle pattern, T is the
integration time, and g̃2 ðtÞ is the autocovariance of the temporal fluctuations
in the intensity of a single speckle. The autocovariance is a normalized version
of the autocorrelation function and is defined as follows:
g̃2 ðtÞ ¼ G̃ 2 ðtÞ∕G̃2 ð0Þ, (8.11)
where G̃ 2 ðtÞ ¼ hðI ðtÞ hI it ÞðI ðt þ tÞ hIit Þit .

Equation (8.11) defines the relationship between laser speckle contrast
imaging and those techniques that use the intensity fluctuations in laser light
scattered from moving objects or particles. LASCA measures the quantity on
the left side of Eq. (8.11); photon correlation spectroscopy, laser Doppler, and
time-varying speckle techniques measure the quantity on the right side. (It is
also worth noting that LASCA uses image speckle, whereas most of the
temporal techniques use far-field speckle. However, this does not detract from
the fundamental equivalence of the two approaches expressed in Eq. (8.11)).
Provided the assumptions made in establishing Eq. (8.11) are valid,
LASCA is now on an equal footing with all of the temporal techniques
(photon correlation, Doppler, time-varying speckle) so far as linking the
measurements to actual velocities is concerned. All the techniques now allow
the correlation time tc to be determined. In the case of photon correlation, this
parameter is measured directly. In the case of LASCA, some further
assumptions must be made in order to link the measurement of speckle
contrast (defined as ss/〈I〉;) with tc.
Various models can be used, depending on the type of motion being
monitored. For the case of a Lorentzian velocity distribution, for example, the
equation becomes21
1
ss tc 2T 2
¼ 1 exp : (8.12)
hI i 2T tc
Equation (8.12) (and similar equations for other types of scatterer motion)
relates the speckle contrast, defined as the ratio of the standard deviation of
the speckle pattern intensity to its mean intensity, to the correlation time tc
for a given integration time T. The relationship is plotted in Fig. 8.9.
Figure 8.9 shows that the speckle contrast rises from near zero to near its
maximum value of 1 over about three orders of magnitude of tc (and hence of
velocity). (For a single exposure, of course, T is a constant.) For flow
velocities corresponding to values of tc less than about 0.04T, the speckle
contrast is very low, i.e., the speckles are completely blurred out by the
Figure 8.9 Variation of speckle contrast with the ratio decorrelation time to integration time
for the Lorentzian model of LASCA
448 Chapter 8
motion. For velocities corresponding to values of tc greater than about 4T, the
speckle pattern has high contrast. Between these limits, flow velocity will be
mapped as variations in speckle contrast. If a different velocity distribution is
assumed, for example, a Gaussian or a uniform distribution, the curve is
slightly different but still shows this characteristic S-shape. The dynamic range
– the range of velocities that can be covered – remains similar.
8.3.8 Practical considerations

From this point on, LASCA experiences the same problem as all the temporal
frequency measurement techniques—photon correlation spectroscopy, laser
Doppler, and time-varying speckle. This is the problem of relating the
correlation time tc to the velocity distribution of the scatterers. It is not
straightforward. Problems include the effects of multiple scattering, the size of
the scattering particles (blood cells in the present case), the shape of the
scatterers, non-Newtonian flow, non-Gaussian statistics resulting from a low
number of scatterers in the resolution cell, spin of the scatterers, etc. Much
work is being done on these effects, and the question is far from settled.
Because of the uncertainties caused by these factors, it is common in all of
these techniques to rely mainly on calibration rather than on absolute
measurements.
Another problem that occurs with LASCA is the difficulty in achieving the
full range of contrasts that should theoretically be available. A stationary
object should give a speckle contrast of 1 (s ¼ ,I., in accordance with well-
established speckle statistics theory and experiment). A fully blurred speckle
pattern produced by rapidly moving scatterers should have zero contrast. The
Lorentzian model of Eq. (8.12), for example, predicts the relationship between
contrast (s/,I.) and the ratio tc/T presented in Fig. 8.9. This suggests that, for
a given integration time T, the dynamic range of the technique corresponding
to contrasts between 0.1 and 0.9 should be about three orders of magnitude in
tc (and hence in velocity). In practice, contrasts of only 0.6 were being
measured even for stationary random diffusers.26 One possible cause of this
problem is the CCD camera. Most CCD cameras are designed to produce a
direct-current (DC) offset from zero called the pedestal. This is to avoid cutting
off the negative peaks produced during any processing. It is formed because of
the dark current and other anomalies in the video circuitry of the camera. By
carrying out some pre-processing on the data, it is possible to remove the effect
of the dark current. When this was done,28 the measured speckle contrast on
the image of a stationary diffusing surface rose to 0.95, very close to the
theoretical value of 1.0 for a fully developed speckle pattern. Similarly, the
measured contrast in the speckle pattern recorded from a hand increased from
0.33 to 0.68. This lower contrast (compared with the stationary diffuser) is due
to the scatterers in the hand (the red blood cells) being in motion.
Other problems with the statistics occur as a result of such effects as the
Gaussian profile of the laser beam and the nonlinearity of the CCD camera.28
However, many of these problems also affect the laser Doppler, photon
correlation, and other time-varying speckle techniques.
Other parameters can be varied. The effect of using different exposure
times (integration times) has been investigated and times as short as 1 ms have
been used. Wavelengths other than 633 nm could also be used. There is a need
in some clinical situations, for example, eczema and other dermatological
problems, to measure the superficial dermal blood flow.35 This has prompted
several groups to use other wavelengths with the laser Doppler technique.35–37
Penetration of tissue is highly wavelength-dependent, with infrared light
penetrating the furthest (several mm) and green and blue hardly at all
(typically 150 mm for green light).
8.3.9 Early applications of the LASCA technique

The original LASCA technique was developed specifically for monitoring
capillary blood flow in the skin. Figure 8.10 shows an example from the mid-
nineties. It shows part of a forearm with a superficial hot-water burn. The
increased perfusion around the burn is clearly visible.
Some researchers have continued this theme of using LASCA and related
techniques to measure skin and other microcirculation blood flow.38–41 Others
have used it to characterize atherosclerotic plaques.42 There has also been a
return to the area that first prompted the development of single-exposure
speckle photography in the 1980s – ophthalmology.43–45 However, the most
important application of the LASCA technique is currently in the area of
neurological research, for monitoring cerebral blood flow.30,34,46–55 A specific
investigation was into the causes of migraine.56 (It is stressed that these
Figure 8.10 LASCA image of part of a forearm, showing increased perfusion around the
site of a hot-water burn (Briers and Webster, 1996276HHFRORUSODWHV
450 Chapter 8
Figure 8.11 Raw image of part of a rat cortex (left) and its processed version (right)
(source: D. Boas, Harvard Medical School).
Figure 8.12 Blood-flow changes during stroke: relative cerebral blood flow ten minutes
after occlusion of the middle cerebral artery in a rat, demonstrating the spatial gradient in the
EORRG IORZ GHILFLW GXH WR WKH LVFKHPLD VRXUFH $ . 'XQQ 8QLYHUVLW\ RI 7H[DV6HHFRORU
SODWHV
examples are just a selection from the large amount of work that is going on -
the list is not meant to be exhaustive. We are aware of more than forty groups
in at least eighteen countries who are working on or with the LASCA
technique.)
The intensive activity involving the LASCA technique in recent years has,
of course, been accompanied by improvements in the images produced. Two
examples are reproduced here. Figure 8.11 is from David Boas’ group at
Harvard Medical School, and Fig. 8.12 is from Andrew Dunn at the
University of Texas. In both cases the improvement in the quality of the
images is clear.
8.3.10 Important developments of the basic LASCA technique

In general, researchers using LASCA have developed their own version of the
technique. This has naturally led to some significant improvements.
Developments include optimization of the exposure time,31 noise reduction,57
and some significant contributions to the theory.32,58 Some workers59 have

identified errors in the original theoretical formulations [such as Eq. (8.12)].
However, it was shown that the model still forms a reliable basis on which to
quantify velocities from the contrast measurements.55
A major development took place in late 2006 with the announcement by
the UK company Moor Instruments Ltd of a commercial instrument that uses
the LASCA technique. Moreover, it has a real-time video capability. For the
first time, this allows the researcher or clinician to follow changes of blood
flow in real time. Of course, some resolution is sacrificed in order to achieve
the high-speed processing required, but the instrument also has a still facility
for those cases where resolution must be maintained. Typical images from the
instrument are shown in Figs. 8.13 and 8.14.
One of the important questions is related to taking into account an effect
of multiple scattering in probed tissues on the LASCA imaging and blood
flow characterization. This question is directly connected with the problem of
the imaging depth in the full-field speckle contrast analysis. It should be noted
that, despite the widespread use of the LASCA technique, the probing depth
and the average number of scattering events in the course of laser light
propagation in a probed medium are not exactly defined. Both values (the
probing depth and the average number of scattering events) are necessary for
Figure 8.13 This image, produced by the Moor Instruments laser perfusion imager (FLPI),
shows the mesenteric vessels in a frog. The area shown is approximately one centimeter
square, and the gut runs around the right periphery of the image. In this investigation, the
user is assessing angiogenic growth in mesenteric vessels caused by a virus. This image
was taken before injection of the virus. (source: Moor Instruments Ltd: www.moor.co.uk)
6HHFRORUSODWHV
452 Chapter 8
Figure 8.14 A composite image produced by the Moor Instruments laser perfusion
imager (FLPI). The series of images of the hand shows the base-line blood flow, the
changes during a partial occlusion using a pressure cuff, and the flow following pressure
release. The images were taken at 3-s intervals. The single image window is from a video
sequence taken at 25 frames per second with a time constant of 0.3 s. The single point
measurement is taken from a finger tip. (source: Moor Instruments Ltd: www.moor.co.uk)
6HHFRORUSODWHV
reliable interpretation of the local contrast values in different zones of an

analyzed image.
A possible way to solve this problem is to use a numerical simulation of
dynamic light scattering in tissue structure. In particular, this approach was
used for the speckle contrast analysis blood flow in the cortex.60 In this case,
3-D voxelized geometries of brain microvasculature can be used in
combination with the Monte Carlo technique to generate spatial probability
distributions of photon travel inside vessels. The sampling depth and the
average number of scattering events can then be derived from the probability
distributions of photon travel in the intravascular space.
8.3.11 Conclusions
To summarize, the LASCA technique (also known as laser speckle contrast
imaging or laser speckle imaging) offers a full-field, real-time, noninvasive,
and non-contact method of mapping flow fields such as capillary blood flow.
In its basic form it uses readily available off-the-shelf equipment. Laser
Doppler, photon correlation spectroscopy, and time-varying speckle are
related techniques, but work by analyzing the intensity fluctuations in
scattered laser light. As they are essentially methods that operate at a single
point in the flow field, some form of scanning must be used if a full-field
velocity map of the flow area is required. Typical scanning laser Doppler
systems take some minutes to complete this scan. LASCA achieves this goal in
a single shot by utilizing the spatial statistics of time-integrated speckle. The
technique produces a false-color map of blood flow in a fraction of a second,
without the need to scan. Thus, LASCA is a truly real-time technique. In
addition, the LASCA system is much more cost effective and very simple to
operate. The main disadvantage of LASCA compared with the other
techniques is the loss of resolution caused by the need to average over a
block of pixels in order to produce the spatial statistics used in the analysis. In
principle, at least, the time-domain techniques can operate on a single pixel.
However, in practice, even this disadvantage is not always present, as many of
the temporal methods use sampling in order to speed up the processing. In any
case, the advantage of real-time operation without scanning probably
outweighs the problem of loss of resolution, as demonstrated by the current
global interest in the technique. Finally, a more complete account of the
history and development of LASCA can be found in some review papers.61,62
Recent advances in laser speckle contrast analysis for tissue functional
imaging and characterization are related to modifications of the speckle data
acquisition procedures and implementation of novel data processing
algorithms. These approaches will be discussed in Sec. 8.5.
8.4 Modification of Speckle Contrast Analysis to Improve Depth

Resolution and to Characterize Scattering Properties of a
Probed Medium
One of the interesting applications of speckle contrast analysis, whether in the
diffraction plane or in the image plane, is also concerned with burn depth
analysis. The central point of this approach is the use not of an extended but
of a localized light source for the illumination of the tissue being probed. This
could be a focused laser beam or an optical fiber tip placed close to the tissue
surface. In the case of nonstationary layered media with strongly differing
dynamic properties of moving scatterers from layer to layer, this leads to the
formation of nonstationary, statistically inhomogeneous speckle patterns on
the interface between the probed tissue and free space. The peculiarities of this
speckle induced by the stochastic interference of backscattered light can be
obtained by consideration of Fig. 8.15, where a simple two-layered model of
the burned tissue is presented.
For each observation point on the tissue surface, the ensemble of effective
optical paths that the probe light travels between the light source and this
point is characterized by the variance of the path lengths, and corresponding
trajectories occupy a spatial region that has the familiar “banana-like” shape.
454 Chapter 8
Figure 8.15 Burn depth diagnostics by means of a localized probe light source; optical
properties (scattering coefficient, absorption coefficient, and mean cosine of the scattering
angle) for both layers are identical.
The axial cross-section of such a “banana” for a given separation between

source and observation point is illustrated in Fig. 8.16. Here we use the results
of a comparison of a Monte Carlo simulation and approximate analytical
calculations of the photon migration paths in the homogeneous semi-infinite
media carried out by Feng, Zeng, and Chance.10 For certain illumination and
detection conditions (highly scattering medium with small absorption;
separation between light source and observation point at least three times
larger than the elastic mean-free path for the scattering medium), the diffusion
approximation can be used to evaluate the statistical properties of the
ensemble of effective optical paths. In particular, the most probable effective
Figure 8.16 Comparison of simulated photon trajectories with analytical “banana”

geometry (reprinted from Ref. 10). A pair of optical fibers delivers the probe light into the
medium and collects the scattered light from the observation point. All the dimensions are
given in millimeters; fiber–fiber separation is 10 mm. Optical parameters of the scattering
medium: ma ¼ 0.02 mm-1; ms ¼ 10.0 mm-1; g ¼ 0.85.
optical path, associated with the center line of the “banana” region, in the case
of the weak absorption limit r 3ma ms ð1 gÞ ≪ 1, can be expressed as
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
zmod ðxÞ ðx2 þ ðr xÞ2 Þ2 þ 32x2 ðr xÞ2 x2 ðr xÞ2 : (8.13)
8
Here r is the separation between the source and the observation point; the
x axis is directed along the surface, and the z axis is oriented normally to the
surface.
Central line of the “banana”-shaped region is shown as the dotted line.
The maximum penetration depth of the pmost ffiffi probable photon trajectory is
reached at x ¼ r/2 and is equal to zmod 4 . On the other hand, in the strong
max 2r
absorption limit r 3ma ms ð1 gÞ ≫ 1, the form of the most probable
trajectory can be expressed as
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2xðr xÞ
zmod ðxÞ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi , (8.14)
3ma ms ð1 gÞr
pffiffiffiffi
mod 2k at x ¼ r/2.
r
and the maximum penetration depth is equal to zmax
It is obvious that for the burn model used (Fig. 8.16), consisting of a layer
of motionless, stationary scatterers (upper layer, burned tissue with suppressed
blood microcirculation) and an underlying layer of normal tissue with normal
blood microcirculation, the scattered light at the observation point is a
mixture of Doppler-shifted components induced by the deeper penetration of
the probe light into the tissue layers with moving scatterers (erythrocytes), and
nonshifted components propagating nearer the surface. The fraction f of
Doppler-shifted photons in the total observed optical signal will determine the
ratio of the fluctuating part of the detected speckle intensity to its average
value (i.e., the modulation depth of the detected optical signal). In particular,
the relation between the fraction of the Doppler-shifted component of the
scattered light and the modulation depth of the detected signal has been
studied by Serov, Steenbergen, and De Mul63; it was found that the
modulation depth, estimated as the ratio of the variance of the fluctuating
part of the detected signal to the square of its mean value, can be evaluated
as f ð2 f Þ N1 , where N is the number of speckles within the detector aperture.
A corresponding dependence obtained in experiments with phantom layered
scattering media consisting of an aqueous suspension of Intralipid (used as the
nonstationary scattering medium) with different concentrations, and layers of
clear tape as the static scatterer, is shown in Fig. 8.17.
The dependence of the penetration depth on the distance between the
localized light source and the observation point leads to a strong influence of
the observation point position on the fraction of Doppler-shifted photons and,
456 Chapter 8
Figure 8.17 Value of modulation depth normalized by N against the fraction of the
Doppler-shifted component of the scattered optical field (reprinted from Ref. 63). Object
under study—aqueous suspension of Intralipid in container plus a few layers of clear tape.
Open symbols correspond to different concentrations of Intralipid in suspension; solid line:
theoretical curve. Probe used: commercial blood perfusion monitor (Periflux 4000, Perimed)
with a standard optical fiber probe (Probe 408, Perimed). Source-detector separation:
0.25 mm.
correspondingly, on the temporal correlation function of the speckle intensity

fluctuations. The characteristic value of the Doppler shift or the decorrelation
time for each component of the scattered field that propagates between the
light source and the observation point can be estimated from the number of
statistically independent scattering events caused by the interactions with
moving scatterers; this number can be found as the ratio of the path length in
the nonstationary medium (underlying tissue for the model used) to the
mean scattering length characterizing the ensemble of moving scatterers
(e.g., erythrocytes).
Thus, if the analysis of time-integrated speckle patterns (as in the LASCA
technique) is used to diagnose the layered medium with differing dynamic
properties of each layer, this leads to the existence of a non-zero speckle
contrast (“residual contrast”) for exposure times significantly exceeding the
characteristic decorrelation time of the speckle intensity fluctuations. This
residual contrast is caused by the contribution of the static layers to the
speckle pattern formation. The effect of the residual contrast on the time-
integrated speckle patterns induced by laser light scattering in human skin is
illustrated by Fig. 8.18, where two fragments of speckle patterns obtained with
strongly differing exposure times are shown.
The experimental setup for recording the speckle patterns is shown in
Fig. 8.19. The beam of the single-mode He-Ne laser was focused on the
fingernail bed surface by a lens of 150-mm focal length. The surface of the nail
bed was covered by an opaque screen with a circular diaphragm exposing part
of the skin near the illuminated area. Varying the diameter of the diaphragm
Figure 8.18 Fragments of the time-integrated speckle patterns induced by blood

microcirculation in the upper layers of human skin. Patient: 23-year-old male, right hand,
fingernail bed (open area). (a) 8-ms exposure; (b) 25-ms exposure.
Figure 8.19 The experimental setup for recording time-integrated speckle patterns with a
varying ratio of the Doppler-shifted and non-shifted contributions induced by laser light
scattering in human skin. 1: localized light source (focused laser beam or optical fiber);
2: layered tissue with differing levels of blood perfusion for each level; 3: variable diaphragm;
4: backscattered light; 5: CCD camera used to record dynamic speckle pattern in the
diffraction plane.
458 Chapter 8
Figure 8.20 Time-dependent contrast vs. exposure time for different diameters of the
exposed area of probed skin. Object: fingernail bed; patient: 23-year-old male.
allows us to select part of the components of the backscattered optical field

and to vary the ratio of the Doppler-shifted and nonshifted components in the
observed speckle pattern. The scattered light from the whole exposed area of
the probed skin formed the speckle pattern in the registration plane, which
was recorded by a CCD camera without an imaging lens. A polarizing filter
was used to improve the contrast of the recorded speckle pattern by selecting a
linearly polarized component of the multiply scattered optical field.
The exposure-dependent contrast of the time-integrated speckle patterns is
plotted against the exposure time in Fig. 8.20 for different values of
diaphragm diameter. The increase in speckle contrast with decreasing exposed
area of skin surface can be easily explained by the predominant role of the
static component in the formation of the recorded speckle pattern. This static
component is induced mainly by the probe light propagating in the upper
layers of the epidermis of the skin where there is no blood microcirculation.
When the diameter of the diaphragm increases, the mean value of the
penetration depth also increases in accordance with the equation for zmax mod in
the case of the weak absorption limit, and we have an increasing contribution
of the Doppler-shifted components of the backscattered light induced by the
light propagating in the deeper layers of skin where there is a developed
microcapillary network; these components cause the blurring of the time-
integrated speckle patterns, which increases with an increase in the exposure
time.
It should also be noted that a significant increase in the exposure time does
not lead to the total suppression of the spatial fluctuations of the speckle
Figure 8.21 “Residual” contrast as a function of the diaphragm diameter. Object: forearm;
patient: 44-year-old male.
intensity. An example of the behavior of this “residual” contrast, estimated for

an exposure time of 20 ms with increasing area of the probed skin surface, is
shown in Fig. 8.21.
The dependence of the modulation depth of the time-varying speckles on
the position of the observation point with respect to the illuminated area of
probed tissue leads to statistically inhomogeneous intensity distributions of
the time-integrated speckle patterns in the image plane (obtained, e.g., by
using a lens to form an image of the tissue surface on the CCD sensor). Such
inhomogeneity is manifested as significant changes in the speckle intensity
variance (and the corresponding local estimations of the time-dependent
contrast) across the image of the scattering zone on the tissue surface. Applied
to burn depth diagnostics, this approach allows us to estimate the thickness of
the burned tissue by analyzing the local values of the time-dependent contrast
as a function of position with respect to the central part of the speckle pattern
image. The modification of this approach was discussed earlier in Ref. 64. The
basic principle of the technique is illustrated in Fig. 8.22.
Experiments made with a burned tissue phantom (static layer: Teflon
layer with different thickness; modulated layer: aqueous suspension of
Intralipid-10% with resulting lipid concentration equal to 1%) probed by a
focused beam from a He-Ne laser (spot diameter was equal to 80 mm) have
shown a linear dependence of the speckle pattern diameter on the static layer
thickness. The results obtained allow us to conclude that “the developed
method is attractive because of its simplicity, requiring only a coherent light
source, a CCD camera, and a computer. However, because burn wounds are
much more complex than the two-layer model, further studies are needed to
determine the applicability of this technique in vivo. In addition, this technique
460 Chapter 8
Figure 8.22 Determination of the depth of burned tissue by measurement of the diameter
of the time-averaged image of the speckle pattern (reprinted from Ref. 64). Photons that
travel to depth x1 in the static layer produce a speckle pattern upon returning to the surface,
whereas photons that travel to depth x2 in the modulated layer produce a time-averaged
blurred pattern at the surface. The probability of a photon returning to the surface further
from the point of incidence increases with the depth that it travels. Integrating over depth
yields the speckle line profiles P(r), which change from speckle to a blurred pattern at some
value r. This value defines the speckle pattern radius that is dependent on the thickness of
the static layer. The transition point is found from an empirically defined threshold.
might be useful in other clinical applications in which the assessment of

perfusion is important, such as vascular ulcers of the skin and skin graft
viability.” 64
Speckle contrast analysis with the use of a localized source of probe light
and spatial filtering of dynamic speckle patterns in the image plane can also be
applied for the characterization of optical properties of multiple scattering
random media.65 These features (the use of a localized source of probing
radiation (laser beam focused on the surface of medium) and the selection of
partial components of the scattered radiation field with respect to path length
(by means of spatial filtration of speckle modulated images of the surface with
the aid of a programmed ring filter)) are illustrated by Fig. 8.23.
In the case of a moving probed sample, the modulus of a normalized
autocorrelation function of the scattered field intensity fluctuations at an
arbitrary point on the surface of the medium can be expressed as follows:
`
jg1 ðtÞj ¼ jhEðtÞE ðt þ tÞi∕hjEðtÞj ij ¼ ∫ expðk 2 hDr̄2 ðtÞiS∕3ÞPðSÞdS,

2
0
(8.15)
where E(t) is the amplitude of the scattered field at time t; k is the wavenumber
of laser radiation in the medium; hDr̄2 ðtÞi is the mean square displacement for
Figure 8.23 Scheme of a full-field speckle correlometer with a localized source of probing
radiation and spatial filtration of scattered radiation in the image plane of the optical
system:65 (1) laser; (2) lens; (3) probed sample; (4) objective; (5) CMOS or CCD camera.
time t averaged over the ensemble of scatterers; and P(S) is the probability
density function of the paths of scattered field components S ¼ s̄∕l normalized
to l . For stationary and ergodic scattering systems, the normalized
autocorrelation function of the scattered field intensity fluctuations g2(t) ¼
〈{I(t)〈I〉}{I(t þ t)〈I〉}〉/〈{I(t)〈I〉}2〉 is related to |g1(t)| by the Siegert
relation: g2(t) ¼ b|g1(t)2, where b is a constant factor determined by the
conditions of detection. The use of a localized source of probing radiation and
the spatial filtration of detected radiation make possible discrimination of the
partial components with respect to the path length s. Using Eq. (8.15), we can
eventually estimate the correlation time of intensity fluctuations for a uniform
pffiffiffiffi
translational motion of the medium in the selected zone as tc 1.22∕kv S̄ ,
where v is the velocity of the medium, and S̄ is the average weighed
normalized path for the analyzed ring segment of the surface. In the case of
probing light wavelengths l1 and l2, the ratio of correlation times for the
qffiffiffiffiffiffiffi qffiffiffiffiffiffiffi
identical detection conditions is estimated as tc,l1 ∕tc,l2 l1 S̄ l2 ∕l2 S̄ l1 . In
the case of a ring zone with a radius obeying the condition R . l , where r is
the effective radius of the source of diffuse radiation generated in the volume
of the medium, we can express the optical path in the diffusion regime of
radiation propagation as follows: s̄ ¼ S̄l K 1 R2 ∕l , where K1 is the
normalization coefficient; accordingly, tc,l1 ∕tc,l2 l1 l l1 ∕l2 l l2 . For the
optical scheme presented in Fig. 8.15, the correlation time of the fluctuations
of subjective speckles in the image plane is related to the value of tc
determined in the object plane (near the surface of the medium) by the
transform t0c ¼ K 2 tc , where K2 is a factor determined by the parameters of the
pffiffiffi
optical system. It can be shown that K 2 ¼ K 3 F ðM þ 1Þ∕ 2pqM, where K3 is
the normalization coefficient, F is the focal distance, q is the pupil radius, and
M is the magnification employed. In order to determine the absolute value of
l* from the measured correlation time t0c , it is necessary to know S̄ for a given
462 Chapter 8
R and, hence, to know r. This problem was solved by Monte Carlo

simulations of the probing radiation transfer in a medium with fitting
parameters l* and r in comparison to experimental data on the radial
distribution of intensity over the spot of scattering on the sample surface.
Experimental verification of the proposed approach was carried using
laser radiation with wavelengths l ¼ 633 nm (He-Ne laser GN-5P) and
405 nm (blue FLEXPOINT laser modules). The model medium was
polytetrafluoroethylene (PTFE) cylinder with a height of 30 mm and a
diameter of 25 mm, which was rotated by a controlled drive (Fig. 8.23). The
rotation frequency was selected within an interval from 0.00244 to
0.00735 Hz. The laser beam was focused by a microobjective (MIM-9, F ¼
16 mm, NA ¼ 0.30) into a spot on the cylinder edge surface at a distance of
7.5 mm from the axis of rotation. The sequences of speckle-modulated
images of the surface region, which were monitored using a CMOS camera,
were used to determine t0c . The width of the analyzed ring zone was 0.18 mm
for all R values.
Figure 8.24(a)shows the t0c ðRÞ values normalized to kv for l ¼ 633
and 405 nm. Figure 8.24(b) presents the radial distributions of the
normalized intensity I(R/r)/I(0) of backscattered radiation for media with
various l*/r, which were obtained by numerical simulations. The values of
the refractive index (1.35) and g 0.9 corresponded to published data for
PTFE.66 For the comparison, Fig. 8.24(b) also shows the plots of I(R/r)/I(0)
constructed on the basis of experimentally measured I(R)/I(0) distributions
by selecting r values ensuring the best fit of experimental data to the results
of simulations.
Figure 8.24 Plots of t0c kv vs. R65 for (1) l ¼ 633 nm, v ¼ 0.1 mm/s; (2) l ¼ 633 nm, v ¼
0.19 mm/s; (3) l ¼ 633 nm, v ¼ 0.29 mm/s; (4) l ¼ 405 nm, v ¼ 0.1 mm/s; (5) l ¼ 405 nm,
v ¼ 0.19 mm/s; (6) l ¼ 405 nm, v ¼ 0.29 mm/s(ðt0c,633 k 633 vÞ∕ðt0c,405 k 405 vÞ ¼ l 633 ∕l 405 2.87
at R ¼ 1.875 mm). (b) Radial distributions I(R/r)/I(0): (1–4) simulations for r/l* ¼ 4,8,16
and 32, respectively; (5, 6) experiment for l ¼ 633 and 405 nm, respectively. Best fit
parameters r633 1120 ± 70 mm and r405 1050 ± 60 mm yield l 633 195 10 mm and
l 405 68 4 mm.
The values of m0s ¼ 1∕l for PTFE obtained from the analysis of full-field
speckle-correlometric data and radial distributions of scattered intensity have
a good agreement with methods of diffuse transmission measurements using
an integrating sphere. It should be noted that, in addition to determining m0s ,
the proposed approach can be used for evaluating other parameters (in
particular, refractive index) from an analysis of the dependences of the
correlation time and average intensity of scattered radiation on the radius of
the ring filter. The characteristics of a probed medium that can be evaluated
using this approach also include the parameters of mobility of scatterers
(average velocity or coefficient of translation diffusion of particles in the
medium).
8.5 Various Modifications of Laser Speckle Contrast Imaging

A brief analysis of recent approaches to speckle contrast analysis as a tool
for imaging and characterization of scatter motions in random media
(in particular, blood flow in tissues) was provided in the review paper by
M. Draijer et al.67
The following speckle contrast techniques can be enumerated:
1. the classical LASCA technique; speckle contract analysis is provided in
the spatial domain; the contrast is determined for a single image over a
group of pixels (usually 5 5 or 7 7 pixels);
2. the laser speckle imaging (LSI) or the laser speckle temporal contrast
analysis (LSTCA); these techniques are described in Refs. 68 and 69; the
contrast is analyzed in the temporal domain and determined for one pixel
over a sequence of frames (25 or 49 frames in the case of LSI);
3. the laser speckle perfusion imaging (LSPI, Refs. 70 and 71); the speckle
contrast is analyzed in the spatial-temporal domain and this technique is
actually a combination of LASCA and LSI;
4. the laser speckle flowgraphy (LSFG, Ref. 72); the spatial-temporal
domain is used to calculate the speckle contrast and it is calculated for a
group of pixels over a given number of speckle images (typically, 3 3
pixels over 3 speckle images);
5. the technique of spatially derived contrast with averaging (SDCav,
Ref. 73); analysis is provided in the spatial domain and the speckle
contrast is evaluated using averaging over a sequence of LASCA
images;
6. the temporal laser speckle contrast analysis, or tLASCA (Ref. 74);
analysis is provided in the temporal-spatial domain and the contrast is
determined based on averaging over a sequence of LSI images;
7. the spatial laser speckle contrast analysis, or sLASCA (Ref. 74); analysis is
provided in the spatial domain and the contrast is determined based on
averaging over a sequence of LASCA images.
464 Chapter 8
It can be seen that these techniques are similar to each other in basic
principles and differ in details of raw data processing. Thus, we should not
expect any qualitative advantages from one modification over other
modifications; the choice of the most appropriate for a given case should be
done on the basis of consideration of the particular conditions.
One of the original approaches to the full-field speckle contrast analysis
(the so-called multi-exposure speckle imaging, MESI, Ref. 75) is based on
application of an illuminating pulse laser with the variable pulse duration. In
this case the exposure time is kept constant and the speckle integration time is
determined by the pulse duration.
Further development of various modifications of the laser speckle contrast
analysis in spatial and/or temporal domain is related to the development of
more sophisticated data processing algorithms. The typical example is the
adaptive computation of contract for laser speckle contrast analysis
(adLASCA) technique considered in Ref. 76. This technique is based on
adaptive computation of contrast. In the preprocessing stage, histogram
equalization is implemented to increase the dynamic range of input intensity.
As an example, the raw speckle image of mice skin vasculature and its
preprocessed version are shown in Figs. 8.25 (a) and (b).
Next, image registration is performed to overcome the internal inevitable
disturbances due to respiration and heart beating of the animal. The first
frame is considered as the reference image, and subsequently all other frames
Figure 8.25 (a) Raw image of mice skin vasculature, (b) histogram equalized image,
(c) zoomed time series speckle images, before registration (up) and after registration (down).
Taken from Ref. 76.
are registered by a single step DFT approach.77 Initial cross-correlation peak

is estimated using simple FFT with an upsampling factor of 2. A normalized
root mean square error metric (NRMSE) [Eq. (8.4)] is computed to optimize
the algorithm as
maxx0 ,y0 jrf f i ðx0 ,y0 Þj2

NRMSE 2 ¼ 1 P P , (8.16)
jf ðx,yÞj2 jf i ðx,yÞj2
x,y x,y
where
X
rf f i ðx0 ,y0 Þ ¼ f ðx,yÞf i ðx x0 ,y y0 Þ
x,y
X
ux0 vy0
¼ F ðu,vÞF i ðu,vÞ exp i2p þ :
x,y
M N
Here, rffI is a cross-correlation between the first frame f(x,y) and the rest of
the frames f(xi,yi) for i ¼ 2, 3,. . . , n in time series image data. (x0,y0) denotes the
translation between each set of images. M and N are image dimensions and
uppercase letters represent the discrete Fourier transform (DFT) of their
lowercase counterparts. The algorithm uses a matrix multiplication implemen-
tation of the 2D DFT to refine the initial peak location.77 It helps to compute
an upsampled version of rff(x0,y0) without zero padding of F(u,n)F*(u,v).
Upsampled cross-correlation (by a factor k) is derived within a 1.5 x 1.5 pixel
neighborhood by multiplying three matrices with dimensions (1.5K, N),(N, M)
and (M,1.5k).77 Rigid translational parameters (Dxi,Dyi) for each frame (i ¼
2,3,. . . , n) are computed to align each of the histogram equalized images.
It should be also noted that simple contrast analysis is influenced by the
presence of corrupting speckles. The effect of these unwanted speckles
becomes very hard to reduce as the vascular information also appears in the
form of speckles. Contrast analysis algorithm, which reduces the effect of
corrupting speckles, restraining structural and functional integrity of
microvasculature is based on a hypothesis—spatial locations of high-
frequency corrupting speckle (coherent speckle noise) alter in successive time
frames, without maintaining any statistical correlation. In contrast to this,
changes in dynamic speckle pattern for slowly moving scatterers bear a strong
correlation in local volume of pixels. For faster movements, contrast gets
reduced and the areas become blur in time-integrated speckle images. In a
local volume of pixels, statistical properties of such corrupting speckles are
different with respect to the dynamic speckle pattern. Adaptive computation
of contrast helps to reduce such unwanted speckles based on local speckle
statistics in a spatiotemporal domain. It helps to maintain the integrity of the
dynamic speckle pattern in successive speckle images. In general, a low-pass
466 Chapter 8
filter can reduce the energy of speckles. A modified Gaussian low-pass filter
function is given as follows:
H u,v ¼ gl þ ðgh gl Þ expfcðD2u,v ∕D20 Þg, (8.17)
where D2u,v ¼ ðu M∕2Þ2 þ ðv N∕2Þ2 , and D0 is the cut-off frequency of the

filter. (M, N) are the dimensions of the image. gl ¼ 0.6 and gh ¼ 1.4 are the
gains of low and high frequencies, respectively. However, this operation
reduces the energy of speckles globally, without considering the changes in
local image statistics. The dynamic speckle pattern can be characterized by a
group of scatterers moving in a similar pattern and velocity. The
spatiotemporal adaptive computation of contrast helps to highlight such a
group of scatterers. The filter function is convolved with the registered image
fx,y as
u,v ¼ fK 1 þ K 2 H u,v gF u,v ,

F out (8.18)
where F(u,v) is the Fourier transform of fx,y. K1 and K2 are the filtering
weights, where K1 is kept at 1, and K2 is derived as the average of Kx,y,t over a
predefined number of time frames. Kx,y,t for the tth time frame is computed
adaptively as
1 ḡ2i,j,t s2i,j,t
K x,y,t ¼ , (8.19)
s2i,j,t ð1 þ s2t Þ
where s2i,j,t and s2t are the local noise variance in the moving window and
global noise variance for the tth time frame, respectively. ḡi,j,t is the average
intensity in a 5 5 pixel moving window. The global noise variance in the tth
time frame is computed from
X s2i,j,t
s2t ¼ : (8.20)
i,j,t ḡ2i,j,t
A temporally averaged LASCA is implemented over these speckle time

frames. Contrast value of a pixel at (i,j) can be represented as
1 Xiþ1 X jþ1
di,j,t
C tLASCAði,jÞ ¼ (8.21)
9 r¼i1 c¼j1 hI i,j,t i
where di,j,t and 〈Ii,j,t〉 are the standard deviation and mean intensity,
respectively, of all pixels at (i,j) over n number of frames.
As a result, this approach allows the reconstruction of sufficiently less
noisy speckle contrast images in comparison with other techniques (see
Fig. 8.26).
Figure 8.26 Comparative study of different contrast analysis techniques: (a) classical
LASCA, (b) LSPI, (c) t-LASCA, and (d) adLASCA for retinal microvasculature in normal
condition of mice.77 It is clearly evident from marked regions that distinguishability of fine
vessels is superior in the adLASCA image more than the other contrast analysis techniques.
8.6 Imaging Using Contrast Measurements of Partially

Developed Speckles
One of the possible approaches to the location of optical macro-inhomogenei-
ties embedded in a highly scattering medium is the evaluation of the statistical
moments of multiply scattered speckles induced by partial coherent illumina-
tion of the probed object. The different effective optical paths of the various
components of the scattered optical field will cause different magnitudes of
modulation depth of the individual random interference patterns depending on
the relation of the path length difference for each pair of interfering components
to the coherence length of the light source used. If the path length difference is
significantly larger than the coherence length, then random interference
modulation of the partial optical field induced as a superposition of two
arbitrarily chosen propagating components will be suppressed due to the
averaging by a detector with a finite response time. Thus, an increase in the
modal value of the effective optical paths and, correspondingly, in the variance
of the path length distribution due to the decrease in the transport mean-free
path or to the increase in the scattering media thickness will cause a reduction in
the modulation depth of the observed multiply scattered partially coherent
speckle pattern. On the other hand, the cut-off of the long path part of the
individual contributions due to, e.g., an increasing absorption of the scattering
medium, will cause an increase in the “visibility” of the detected speckle pattern,
which can be evaluated as the speckle contrast.
Imaging of spatially inhomogeneous multiply scattering media by use of
speckle contrast measurements with quasi-monochromatic illumination was
discussed by Thompson, Webb, and Weiner.78 Considering the formation of
the random optical field due to the multiple scattering of probe light from a
light source with a finite spectral bandwidth, they obtained the following
expression for the contrast of partially coherent, multiply scattered speckles
(scalar approximation):
` `
0 0

0

V ¼ ∫ ∫ SðlÞSðl Þ exp½2pjs 0
1 1
l l

dldl
0
1∕2
∕∫`
0
SðlÞdl, (8.22)
468 Chapter 8
where S(l) is the spectral density of the light source used, and s is the value of
the effective optical path of the individual component of the scattered
field. The expected value of the phase term, or characteristic function
hexp½2pjsðl1 l10 Þi, can be expressed by introducing the probability density
function of the effective optical paths r(s) as
`
¼ ∫ rðsÞ exp 2pj
1 1 1 1
exp 2pjs ds. (8.23)
l l0 0 l l0
The path length statistics described by r(s) depend on the illumination and
detection conditions as well as on the geometry of the probed scattering
system. In order to analyze its influence on the contrast of partially coherent
speckles as a visualization parameter, a random walk approach can also be
used.78,79 In this case, one can interpret random interference patterns such as
speckle fields as the result of the superposition of noncorrelated contributions
caused by sequences of scattering events. In this case, the coordinate-
dependent part of the complex amplitude of the scattered field at an
arbitrarily chosen point can be expressed by using the scalar approximation as
X
Eðr̄Þ ¼ ak expfjfk ðr̄Þg, (8.24)
k
where the phase of each contribution can be written as fk ¼ 2psk/l. Here, sk is

the effective optical path of the kth contribution, and l is the wavelength of
the probe light in the scattering medium. Using the simplifying assumption
ak ¼ 1, we can obtain the value of the intensity at the observation point as
XX
I ðr̄Þ ¼ jEðr̄Þj2 ¼ Eðr̄ÞE ðr̄Þ ¼ cosf2pðsk sk0 Þ∕lg. (8.25)
k k0
It is easy to see that averaging over all possible combinations of

statistically independent sk and sk0 gives the value of the mean intensity at the
observation point , I ðr̄Þ . as equal to N, where N is the number of
contributions. Replacing the summation over k and k0 by a summation over a
new index l, we express the “instantaneous” value of the intensity
corresponding to the current configuration of scattering sites as
X
Iðr̄Þ ¼ N þ 2 cosð2pDsl ∕lÞ, (8.26)
l
where the summation over l is carried out from 1 to (N2!N)/2. In this notation,
the statistical properties of the intensity fluctuations are controlled by the
distribution of the path length differences Dsl. The second term in Eq. (8.26)
can be considered as the result of interference between the k and k0
components of the scattered field. Thus, if a partially coherent source is used
for the probing of the scattering medium and this source is characterized by
the normalized coherence function g(Ds), then Eq. (8.26) should be replaced
by the following equation:
X
I ðr̄Þ ¼ N þ 2 cosð2pDsl ∕lÞjgðDsl Þj. (8.27)
l
Taking into account the statistical independence of the path length

differences and calculating the mean value of I 2 ðr̄Þ, we can obtain
X
hI ðr̄Þi ¼ N þ 4N
2 2
cosð2pDsl ∕lÞjgðDsl Þj
l
X
þ4 cos ð2pDsl ∕lÞjgðDsl Þj
2 2
l
X X
þ4 cosð2pDsl ∕lÞ cosð2pDsl 0 ∕lÞjgðDsl ÞjjgðDsl0 Þj
l l0
X
¼N þ4 2
cos ð2pDsl ∕lÞjgðDsl Þj ,
2 2
where l ≠ l0 .
Taking the number of contributions as infinite, N ! `, and considering
the transition from the discrete model of a scattering medium that is
characterized by a finite number of contributions to the “continuous”
description, we can introduce the probability density function of the path
length differences r(Ds) and write the second-order statistical moment of the
scattered light intensity as
`
hI 2 ðr̄Þi hI i þ 2ðhI i hI iÞ ∫ cos2 ð2pDs∕lÞjgðDsÞj2 rðDsÞdðDsÞ.

2 2
(8.28)
0
The probability density function of the path length differences r(Ds) is

related to the probability density function of the effective optical paths r(Ds) by
`
rðDsÞ ∫ rðjÞrðj þ DsÞdj,

0
i.e., r(Ds) can be expressed as the autocorrelation of r(Ds). Thus, the contrast
of the multiply scattered speckle pattern can be expressed as
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u`
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u
t
V ¼ hI 2 ðr̄Þi∕hI ðr̄Þi2 1 ∫f1 þ cosð4pDs∕lÞgjgðDsÞj2rðDsÞdðDsÞ.
0
(8.29)
470 Chapter 8
For a fully coherent illumination source with |g(Ds)| equal to 1 for all
possible values of the path length differences, the contrast value introduced in
Eq. (8.29) is equal to 1 independently of the path length statistics. On the other
hand, for broadband light sources, V 0. In the intermediate case, when the
coherence length of the light source used is comparable to the spatial scales
characterizing the ensemble of the path length differences (e.g., root mean
square or modal value), the speckle contrast varies between 1 and 0.
The statistics of the path length differences described by the probability
density function r(Ds) can be analyzed by using the diffusion approximation,
or, in the case of nondiffuse regimes of the probe light propagation in the
scattering object, when the diffusion approximation fails,11 by means of a
Monte Carlo simulation. In our case, the latter approach has been used to
study the influence of the scattering and detection conditions on the contrast
of the partially coherent speckles induced by the probe light being scattered by
macroscopically inhomogeneous media.
One of the possible ways of controlling the coherence length of a probe
light source is the use of a laser diode with the varying pumping current below
the generation threshold.79 Another way is connected with the application of a
frequency-modulated laser diode above the generation threshold (for example,
based on the Littrow scheme with a piezo electric actuator). If the speckle
integration time is much larger than the period of laser frequency modulation,
the variations of modulation depth provide an effect similar to that in the case
of a decrease of the coherence length for a continuous waveform (CW) source.
This approach was defined as the “blink speckle spectroscopy.”80
It should be noted that the question about the potential applications of
partially coherent speckle analysis to medical imaging is still open. But it is
necessary to note that such an analysis plays an important role in other
speckle techniques in biomedical applications, especially those based on the
measurements of the time-dependent contrast of speckles obtained with a light
source of finite coherence length. The evaluation of the additional contrast
decay due to the multiple scattering of the partially coherent probe light
improves the accuracy of speckle-analysis-based diagnostic techniques (e.g., in
the case of blood flow monitoring).
8.7 Monitoring Tissue Thermal Modification with a

Bundle-Based Full-Field Speckle Analyzer
Among other biomedical applications, the high potential of the full-field
speckle technique was demonstrated in the case of the monitoring of laser-
mediated modification of fibrous tissues (such as, e.g., a cartilage81). The
contrast of speckle-modulated images of the treated zone of tissue, which are
partially blurred due to the exposure time comparable with the speckle
decorrelation time, appears to be an adequately sensitive indicator of
characteristic changes in the cartilage structure at various stages of

modification.
Application of fiber optic units (for instance, a fiber optic bundle) in a full-
field speckle instrument can be considered as a further step from the benchtop
studies toward the practical use of the speckle monitoring of tissues. In
particular, because of its mechanical flexibility, such a bundle-based speckle
monitor can be combined with a fiber optics laser system for thermal treatment
of tissues and used as a feedback control system of treatment. In addition, the
incorporation of a multiple-channel fiber optic unit in the speckle-based
instrument can provide an additional potential for the processing of dynamic
speckle patterns due to the specific filtering, sampling, and phase-modulating
properties of such a unit. In comparison with classical fiber optic Doppler
systems, which are applied to monitor the blood perfusion in tissues, the bundle-
based full-field speckle instrument has the additional advantage of multispeckle
analysis of detected optical signals. This advantage is beneficial in the case of a
dynamic scattering probe of nonergodic scattering systems characterized by the
time-varying scatter dynamics (for instance, such nonergodic behavior is typical
for thermally modified cartilage and is manifested in time-varying nonuniform
distributions of the statistical characteristics of image-modulating speckles across
the probed area, see Ref. 81). Application of the multispeckle principle to
analysis of the speckle correlation decay (i.e., by evaluation of the exposure-
dependent speckle contrast related to the ensemble-averaged integrand of the
normalized autocorrelation function of speckle intensity fluctuations) can
diminish the negative effect of nonergodicity of analyzed speckles. In particular,
the potential of multispeckle diffusing light spectroscopy for the study of
nonergodic scattering systems (such as, e.g., the aged, soft, glassy materials) is
discussed in detail in Ref. 82. In this sense, the method of full-field speckle
monitoring with the use of a fiber optic bundle and CCD array in the detection
unit, which is discussed in this work, is in some points similar to the two-cell
technique applied for the diffusing-wave spectroscopy of nonergodic scattering
systems 83–85 (the probed nonergodic and nonstationary medium acts like the first
cell, and the detection unit with a fiber optic bundle is similar to the second cell in
the two-cell scheme). Note that from the viewpoint of nonergodicity suppression,
the application of a fiber optic bundle in the speckle instrument instead of a
multimode optical fiber (as applied in classical fiber optic Doppler monitors) has
a principal significance because of the necessity to collect and mix the optical
signals from a great number of statistically independent coherence areas
(speckles) across the non ergodic speckle-modulated image of a probed region.
Note that the important problem arising in this case is the noiselike
disturbance of speckle dynamics, which can appear because of unexpected
movements of the bundle. This problem can be particularly avoided by
providing mechanical stability of the speckle instrument during the time
required to capture a set of speckle-modulated images.
472 Chapter 8
Figure 8.27 Multicascade transformation of optical field by the bundle-based full-field

analyzer; 1 – object, L1 - image-transferring component, 2 – fiber optic bundle, 3 – detector,
x̄ ¼ ðx,yÞ - object plane, X̄ ¼ ðX ,Y Þ - image plane (input plane of the bundle), x̄j ¼ ðx j ,y j Þ -
output plane of the bundle, x̄2 ¼ ðx 2 ,y 2 Þ - detector plane.
Here we discuss some peculiarities of the speckle contrast analysis with the
use of a bundle-based full-field instrument applied to monitor the thermal
modification of collagenous tissue (cartilage),86
Figure 8.27 displays the scheme of the considered full-field speckle
analyzer with a fiber optic bundle as a light-delivering unit and speckle
detection in the diffraction zone. A probed object is illuminated by a
collimated laser beam. The area of interest on the object surface is imaged by
lens L1 on the input tip of the bundle. The dynamic speckle pattern is captured
with a CCD camera without a lens, which is placed behind the output tip of
the bundle at the distance Z≫d2/l (d is the diameter of the single fiber in the
bundle, l is the wavelength of light). For these conditions, the transformation
of the scattered light in the optical system is the result of:
1. spatial filtration of the optical field in the object plane by the image-
transferring component L1,
2. spatial sampling and random-phase modulation of propagating light
by the fiber optic bundle, and
3. diffraction in free space between the output tip of the fiber and the
CCD sensor.
Such sequential transformation of the speckle field causes changes in the
correlation properties of detected light with respect to that of light leaving
the object. The problem of correlation transformation can be addressed by the
general problem of coherence transfer in optical systems.87 We consider the
effect of light transfer in the optical system (Fig. 8.27) on the spatial-temporal
correlations of detected light in the framework of scalar approach.
Despite the limitations of the scalar model in the case of depolarization of
incident radiation by the probed medium and the fiber optic bundle, it will
allow us to find basic relationships between the transfer properties of the
optical system and the correlation decay parameters of speckles in the

detection plane. In addition, the results can be applied for the case of
polarization-sensitive speckle diagnostics (when incident light is linearly
polarized and speckle statistics are analyzed using the polarization
discrimination of detected light). Considering the transformation of speckle-
modulated optical field by the L1 component, we can write (see, for instance,
Refs. 88 and 89):
G 1 ðX̄ 1 ,X̄ 2 ,t1 ,t2 Þ ¼ hEðX̄ 1 ,t1 ÞE ðX̄ 2 ,t2 Þi
¼ ∫∫hEðx̄1 ,t1 ÞE ðx̄2 ,t2 ÞiKðX̄ 1 ,x̄1 ÞK ðX̄ 2 ,x̄2 Þd x̄1 d x̄2 . (8.30)
Here, G1 ðX̄ 1 ,X̄ 2 ,t1 ,t2 Þ is the correlation function of the field fluctuations
in the image plane, Eðx̄,tÞ is the field amplitude in the object plane,
and KðX̄ ,x̄Þ is the point-spread function of the image-transferring
component under coherent illumination; we assume the statistical homoge-
neity and ergodicity of spatial fluctuations of Eðx̄,tÞ. In the case of Gaussian
statistics of RefEðX̄ ,tÞg,ImfEðX̄ ,tÞg and hEðx̄,tÞi ¼ 0, the Siegert equation
is valid, and the normalized temporal correlation functions of the intensity
and field fluctuations in an arbitrarily chosen point X̄ are related to each
other as90
hI ðX̄ ,tÞI ðX̄ ,t þ tÞi
g2 ðtÞ ¼ ¼ 1 þ jg1 ðtÞj2 , (8.31)
hI ðX̄ ,tÞi2
where g1 ðtÞ ¼ G1 ðtÞ∕G 1 ð0Þ ¼ hEðX̄ ,tÞE ðX̄ ,t þ tÞi∕hjEðX̄ ,tÞj2 i.
In the case of a conventional full-field speckle technique, the time
averaging of speckle patterns in the image plane is used and the contrast of the
averaged speckle pattern is introduced for characterization of the speckle
decorrelation due to scatter dynamics:
T̃ 0.5
V T̃ ¼
1
∫ðg2 ðtÞ 1Þdt , (8.32)
T̃ 0
where T̃ is the exposure time used to capture speckles. Usually, the fixed value
of T̃ is set comparable with the speckle decorrelation time, and the local
estimates of V T̃ reflect the relative changes in scatter dynamics across the
analyzed region.
Thus, analysis of Eqs. (8.30)–(8.32) for the given detection conditions
allows us to characterize the influence of the transfer properties of the
image-transferring component (L1) on the decorrelation rate of dynamic
speckles in the image plane. It is obvious that for the ideal conditions
(KðX̄ ,x̄Þ dðX̄ x̄Þ) this influence is absent. Conversely, with the low-
quality imaging component the role of correlation transformation becomes
474 Chapter 8
significant. In the extreme case of d-correlated speckles in the object plane ðx̄Þ,
the spatial-temporal speckle dynamics in the image plane is controlled by the
transfer properties of a speckle-imaging system.88,89,91
The important feature is the strong effect of the type of speckle dynamics
in the object plane on the speckle decorrelation rate in the image plane ðX̄ Þ for
the given transfer properties of L1. Considering the extreme cases of speckle
dynamics in the object plane (the pure translation or the pure “boiling”) we
obtain in the latter case the relatively weak dependence of the speckle
decorrelation time tc,i in the image plane, which is determined by a 1/e decay
of g2(t)1, on the full width at half maximum (FWHM) value DV of the
square of the point-spread function modulus jK L1 ðX̄ ,x̄Þj2 . For “boiling”
speckles, the decrease in tc,i plane with respect to that in the object plane (tc,o)
is controlled not only by the properties of the imaging system, but also the
peculiarities of hEðx̄1 ,t1 ÞE ðx̄2 ,t2 Þi; decay with the increasing jx̄1 x̄2 j; and |
t1t2|. The rigid and detailed consideration of the spatial-temporal correlation
properties of multiply scattered optical fields in the object plane is beyond the
framework of our consideration, but we can choose the reasonable model of
the decay in G 1 ðx̄1 ,x̄2 ,t1 ,t2 Þ for further analysis of the influence of correlation
transfer in the considered full-field speckle instrument on the exposure-
dependent speckle contrast in the image plane.
Figure 8.28 displays the calculated dependencies of V T̃ for partially blurred
speckles in the ðX̄ Þ plane, which are imaged by L1 with the “soft” apodized
aperture (the Gaussian function jK L1 ðX̄ ,x̄Þj2 ), on the normalized exposure time
T̃∕tc,o . Different curves correspond to different relationships between DV
and the average speckle size r̄o in the object plane (the high-resolution
Figure 8.28 Theoretical dependencies of V T̃ in the image plane on T̃∕tc,o for the case of a
soft aperture of L1. Solid line: DV ≪ r̄ 0 ; dashed line: DV ¼ r̄ 0 ; dotted line: DV ≫ r̄ 0 . Inset:
dependencies of the characteristic exposure time on DV∕r̄ 0 ; solid line: uniform distribution of
td ; dashed line: quadratic model; dotted line: linear model.
image-transferring element with DV∕r̄o ≪ 1, the medium-resolution element

with DV∕r̄o ¼ 1, and the low-resolution element with DV∕r̄o ≫ 1). The
Gaussian second-order statistics of the spatial fluctuations and the Lorentzian
spectrum of the temporal fluctuations of speckles in the object plane were
assumed in the framework of the considered scatter model. Also, the linearly
decreasing decorrelation time td of G 1 ðx̄1 ,x̄2 ,tÞ was assumed with the increasing
D ¼ jx̄1 x̄2 j : td tc,o f1 ðD∕2.12r̄0 Þg. The value of td for D ¼ 0 was taken
equal to tc,o and equal to 0 for D exceeding the cutoff value 2.12r̄o , which
corresponds to exp(4.5) decay of the Gaussian correlation function
G1 ðx̄1 ,x̄2 ,0Þ ¼ expfðD∕r̄0 Þ2 g). This model gives tc,i close to tc,o in the case of
the narrow point-spread function of the image-transferring element. With the
increasing DV, the value of tc,i monotonically decreases and approaches the
value characteristic for the used linear model of td decay. The inset in Fig. 8.28
shows the dependencies of the characteristic exposure time (which corresponds
to a twofold decay in V T̃ with respect to V T̃¼0 ) normalized by tc,o on the ratio
DV∕r̄o for apodized L1 and various models of td decay: the constant value of td
across the coherence area, the linear decay of td, and the quadratic decay:
td tc,o f1 ðD∕2.12r̄0 Þ2 g.
Consideration of the simulation data obtained in the framework of the
above-mentioned models shows that the extreme value of the decay factor tc,o/
tc,i at large DV∕r̄o , which is estimated from the speckle-contrast analysis,
strongly depends on the characteristic rate (Dtd ∕r̄0 ) of a decrease in td for D
varying from 0 to r̄0 . Thus, tc,o/tc,i ¼ 1 for Dtd ∕r̄0 ¼ 0, tc,o/tc,i 1.32 for
Dtd ∕r̄0 0.78tc,0 ∕r̄0 (the quadratic model), and tc,o/tc,i 1.86 for Dtd ∕r̄0
0.53tc,0 ∕r̄0 (the linear model). A similar analysis provided for the case of L1
with a circular pupil and uniform transmittance has shown the insignificant
deviations of the extreme values of tc,o/tc,i from that obtained for the case of
an apodized aperture. Multi-cascade transformation of a speckle-modulated
optical field by image-transferring elements with DV∕r̄o ≫ 1 will cause the
gradual decrease in tc,o/tc,i decay factor.
The situation dramatically changes if the speckle translation occurs in the
object plane. In this case, the increasing FWHM value of jK L1 ðX̄ ,x̄Þj2 leads to
an increasing tc,i. With image-modulating speckles much larger in comparison
with speckles in the object plane (i.e., for DV∕r̄o ≫ 1), the correlation time tc,i
is defined by DV/vt,i, where vt,i is the velocity of the speckle translation in the
image plane. Note that a similar effect of the transfer properties of optical
system on the decorrelation of “translating” speckles was discussed in Ref. 89.
Further transformation of the speckle-modulated optical field in the
considered optical system is controlled by the spatial sampling in regular or
stochastic fashion (depending on the fiber optic bundle that is used), the
random-phase modulation of light propagating in separate fibers, and
the propagation in free space between the output tip of the bundle and the
476 Chapter 8
detector. We present the complex amplitude in the arbitrarily chosen detection

point (x2,y2) as
X
N
pj
Eðx2 ,y2 ,tÞ ∝ E i ðtÞ expfj½fi ðtÞ þ fsi g exp ½fx2 xji g2 þ fy2 yji g2 ,
i
Zl
(8.33)
where N is the number of fibers in the fiber optic bundle, Ei(t),fi(t) are the
amplitude and phase of fluctuating field at the output tip of i-th fiber caused
by time-varying speckle-modulated optical field at the input tip of the given
fiber, fsi is the phase delay related to the optical length of i-th fiber, and xji and
yji characterize the position of the i-th fiber in the j plane. Here, we assume
formation of speckles in a far diffraction zone as a result of superposition of
contributions from the ensemble of randomly fluctuating point-like sources.
The instantaneous complex amplitude Ei(t)exp{jfi(t)} of the i-th effective
point-like source is taken into consideration by assuming in the generalized
form the effects related to the modal structure of optical fields propagating in
each fiber and the antenna properties of fibers excited by speckle-modulated
light at the input tip of the bundle. This model is valid under the condition
Z≫d2/l.
The time-varying intensity in the detection point can be expressed as
I ðx2 ,y2 ,tÞ ∝ jEðx2 ,y2 ,tÞj2
X
N X
∝ jE i ðtÞj2 þ E i ðtÞE i0 ðtÞ cos½fwi ðtÞ wi0 ðtÞg þ fFi Fi0 g;
i¼1 i≠i0
(8.34)
where wi(t) is the fluctuating component of phase related to speckle
fluctuations at the input tip, and Fi is the stationary random component
determined by the transfer properties of the i-th fiber and free space between
the bundle and the detector. In the case of deep phase modulation (Fi ¼ 1,. . . ,N
≫2p) and for large values of N, the outgoing light induces fully developed
speckle patterns with the circular Gaussian statistics and the zero mean value
of the complex amplitude.
In this case, after some simple calculations, the temporal correlation
function of the intensity fluctuations in the arbitrarily chosen detection point
is obtained as
* +
XN X
N
G2 ðtÞ ¼ hI ðx2 ,y2 ,tÞI ðx2 ,y2 ,t þ tÞi ¼ I i ðtÞI i0 ðt þ tÞ , (8.35)
i¼1 i0 ¼1
where Ii(t) is the intensity related to the contribution from the i-th fiber in the
detected signal. Ensemble averaging, which was carried out under the
conditions of ergodicity and spatial homogeneity of the ensemble of point-like

sources in the j plane, leads to
k maxX
¼N1
G 2 ðtÞ NG2,s ðtÞ þ G2,k ðtÞPðkÞ, (8.36)
k¼1
P
where G 2,s ðtÞ ¼ ð1∕NÞ N i¼1 hI i ðtÞI i ðt þ tÞi and the second cross-correlation
term is caused by temporal correlations between optical signals sampled
from different regions of the same speckle in the input plane. G2,k(t) are
the cross-correlation functions of Ii(t) and Ii0 (t) with k ¼ |ii0 | and P(k)
are
Pkmaxthe weight factors satisfying the following normalization condition:
¼N1
k¼1 PðkÞ ¼ NðN 1Þ. The influence of the cross-correlation term on
G2(t) rises up with an increase of the average speckle size in the input plane
above the distance between the neighboring sampling zones determined by
the antenna properties of the fibers. Thus, the effect of light propagation in the
bundle and free space on decorrelation of detected speckles is related to the
averaging of td in the input plane over the discrete groups of partially
correlated optical signals sampled from the same coherence area in the input
plane. Therefore, the two-stage transformation of “boiling” speckles (by the
image-transferring element and by the bundle in combination with free space)
will be accompanied by a sequential decrease in the speckle decorrelation time
with respect to that in the object plane.
In the experimental studies, the full-field speckle analyzer with the bundle-
based light delivery unit (Fig. 8.29, a) was applied to monitor laser-mediated
modification of cartilage. The laser beam emitted by a single-mode He-Ne
laser (1) (632.8 nm, linear polarization, 5-mW output power) was expanded by
a 50 telescopic system (2) and fell on the surface of a sample (3) of ex vivo
bovine nasal septum cartilage. The samples under study were taken from
freshly slaughtered animals and prepared as square pieces with 25 25 mm2
lateral dimensions and 1.5-mm thickness. The diffuse reflected light was
collected by the imaging lens 4 (focal length of 16 mm and diameter of 6 mm)
and the speckle-modulated image of the area of interest on the sample surface
was formed on the input tip of a fiber optic bundle (5). The bundle with the
hexagonal packing of fibers (the fiber diameter is 25 mm, and the separation
between the fibers in a row is 3 mm) was 1.2 m in length. Light from the
output tip was detected by a CCD sensor (6) (SONY ICX415 chip-based unit,
25-fps frame rate, 10-bit analog-to-digital conversion) co-axially placed at a
distance of 300 mm from the output tip. In the experiments, the capture of
sub-frames (450 pixels 450 pixels) at the frame rate of 25 fps was used. This
allowed us to set the constant value of exposure time equal to 20 ms with the
value of CCD readout time approximately equal to 22 ms.
To characterize the optical system of the bundle-based speckle monitor,
we estimated the average speckle size in the image plane of lens 4 and in the
478 Chapter 8
Figure 8.29 (a) scheme of the experimental setup with the bundle-based full-field analyzer;
1 – He-Ne laser, 2 – telescopic system, 3 – object, 4 – image-transferring element, 5 – fiber-
optic bundle, 6 – CCD detector, 7 – erbium laser, 8 – light-delivering fiber, 9 - thermograph.
(b) scheme of the experimental setup with the LASCA modality; 1 – He-Ne laser, 2 –
telescopic system, 3 – object, 4 – CCD camera, 5 – erbium laser, 6 – light-delivering fiber, 7 -
thermograph. Inset: the temperature distributions across the surface of treatment zone at
various stages of modification; 1 – tissue heating, 2 – quasi-stationary phase of modification.
detector plane. A Teflon plate (10 mm thick) was placed in the object plane as
a source of backscattered speckle-modulated light. The pixel size of the CCD
array is 8.3 mm 8.3 mm, and the average speckle size in the detection plane
was evaluated as 65 mm. Thus, the average coherence area in the detection
plane covered approximately 50 CCD pixels. Also, the quantification of
“intermediate” speckle pattern was carried out by positioning the CCD

module without a lens instead of the input tip of the fiber optic bundle. In this
case the average speckle size appeared comparable with the pixel size of the
CCD array, which resulted in distortions of detected intensity distributions
(due to the effect of integration of speckle intensity fluctuations over the area
of each pixel). Statistical analysis of the captured speckle pattern allowed us to
estimate the correlation radius of intensity fluctuations in the image plane of
lens 4 as 11 mm. Thus, we can conclude that the average speckle size in the
image plane did not exceed 22 mm and was comparable with the diameter
of fibers in the fiber optic bundle.
Thermal treatment of the samples was carried out by radiation of an
erbium fiber optic laser (7) (l ¼ 1.56 mm, adjustable output power up to 5 W)
arranged with multimode quartz fiber (8) (0.6 mm in diameter) as a radiation-
delivering unit. The diameter of the treatment zone on the sample surface was
approximately equal to 15 mm. The wavelength of 1.56 mm provides the bulk
absorption of laser radiation by tissue.82
In our experiments the recording of sequences of speckle-modulated
images with the use of the CCD camera was started synchronously with the
acquisition of the temperature of the tissue surface with the use of the
thermograph, whereas the laser treatment of tissue was started manually after
the beginning of image recording. Such an experimental procedure resulted in
the random delay (typically, about 2 – 4 seconds) of the alterations in speckle
contrast and tissue temperature with respect to the initial moment of each
experiment. This feature is not important for further data analysis because of
synchronization of the temperature and contrast time series and application of
delay-independent procedures of data processing (in particular, estimates of
the contrast decrement).
For comparison, a similar experiment was carried out with the
conventional LASCA modality [Fig. 8.29(b)] arranged with the CCD sensor
of the same type and high-quality imaging lens (LMZ13A5M). The
magnification of the tissue surface image at the CCD sensor was 0.4 .
Similarly to the bundle-based speckle instrument, the average speckle size in
the detector plane was estimated for the LASCA modality. In this case the
focal length and the aperture of the CCD camera objective [Fig. 8.29(b)] equal
to 75 mm and 5 mm were chosen to provide the optimal detection conditions
(the sufficient brightness of speckle patterns and the appropriate speckle size;
see Ref. 82 for more details). The speckle size estimated using the Teflon plate
was equal to 15 mm. An experiment with the conventional LASCA modality
was carried out only with the single value of IR laser power (4.5 W) in order to
demonstrate the higher efficiency of the bundle-based speckle instrument for
cartilage modification monitoring.
Simultaneously with the capture of dynamic speckle patterns, the spatial-
temporal distributions of the tissue surface temperature were monitored with
480 Chapter 8
the use of a thermograph (IRTIS-200, manufactured in Russia). Typical

profiles of the instantaneous temperature, which correspond to different
stages of cartilage modification, are illustrated by the inset in Fig. 8.29(b).
Analysis of speckle dynamics in the course of modification has shown the
pure “boiling” type of speckle motions in the detector plane (without any
noticeable translations). Characterization of the dynamics type can be carried
out using the analysis of the normalized spatial–temporal correlation function
of speckle intensity fluctuations in the detector plane:
g2 ðDx2 ,Dy2 ,tÞ

¼ hfI ðx2 þ Dx2 ,y2 þ Dy2 ,t þ tÞ hI igfI ðx2 ,y2 ,tÞ hI igi∕fhI 2 i hIi2 g:
With the zero time lag t ¼ 0, the 1/e decay of g2(Dx2,Dy2,0) characterizes
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
the average speckle size hrs i 2 ðDx2 Þ2 þ ðDy2 Þ2 g2 ¼1∕e . The presence of a
noticeable translational component in speckle dynamics should manifest itself
in the appearance of a visible correlation peak for the given sufficiently non-
zero value of t (t . tc). This peak should be shifted in the coordinate plane
(Dx2,Dy2) to the position (vx2t,vy2t), where vx2,vy2 are the x2– and y2–
components of the speckle translation velocity. Figure 8.30(a) displays the
normalized temporal correlation functions g2(t) of the detected optical signal
at various stages of modification; spatial correlations of captured speckle
patterns are illustrated in Fig. 8.30(b) for the cases of the same frame (g2(Dx2,
Dy2,0)) and two frames (g2(Dx2,Dy2,t)) captured with the time delay t slightly
exceeding the speckle decorrelation time in the detector plane [Fig. 8.30(c)].
Note the absence of any correlation peak in the latter case. This indicates the
absence of any remarkable translational dynamics of speckles in the detector
plane. Also, the direct observation of the sequences of captured speckle
patterns showed only the “boiling” of speckles at the stages of tissue heating
and modification.
The boiling type of speckle dynamics is obviously caused by the absence
of macroscopic regular motions of scattering sites in the thermally treated
tissue. Because of the backscattering geometry and contributions of multiply
scattered diffuse components, the detected light is characterized by the
significantly more broadband intensity fluctuations in comparison with the
case of transillumination monitoring of thermally treated ex vivo cartilage.81
The remarkable blur of speckle patterns occur at the frame rate of 25 fps (with
the above-mentioned value of the exposure time equal to 20 ms). Conse-
quently, the value of V T̃ was estimated for each frame in the recorded
sequence without any additional time averaging (as it was carried out in
Ref. 81). Figure 8.31 displays the dependencies of the temperature in the
central region of the treatment zone (a) and the normalized speckle contrast
V T̃ ðtÞ∕V T̃ ð0Þ (b) on the time lapse t in the course of the treatment procedure
provided with the different values of the output power of IR radiation.
The appearance of some artifacts on the experimental dependencies
plotted in Fig. 8.31(b) (in particular, the minimal value of contrast at 4.8 W
Figure 8.30 (a) normalized temporal correlation functions of speckle intensity fluctuations
in the detector plane; 1 – tissue heating (the temperature in the central region of treatment
zone is 50°C), 2 – the quasi-stationary phase of tissue modification (the temperature is
68°C). (b), (c) normalized spatial correlation functions of dynamic speckle patterns in the
detector plane; (b) the generalized spatial autocorrelation function; (c) the cross-correlation
function for two frames with the time delay Dt approximately equal to 2.5tc,detector; the scale
for (b) and (c) are the same.
482 Chapter 8
Figure 8.31 (a) The dependencies of the tissue temperature in the central region of
treatment zone on the time lapse; inset displays the “loop-like” behavior of V T̃ ðT Þ (the output
power of the erbium laser is 4.5 W; I – tissue heating; II- modification; III – thermal
relaxation); (b) the dependencies of V T̃ on the time lapse (bundle-based analyzer); (c) the
same as in (b) for LASCA modality (IR laser output power is 4.5 W; low-amplitude
fluctuations related to non ergodic behavior of the speckle-modulated image are marked by
arrows).
appears later than the minimal values of V T̃ ðtÞ∕V T̃ ð0Þ at 4.2 W and 4.5 W) is
related to the arbitrary shift of the origin for each data sequence because of
the above-mentioned random delay in the start of laser treatment. As
mentioned above, this circumstance is insufficient for further data analysis.
Similar to previously reported results on the transillumination speckle
monitoring of cartilage modification,81 the loop-like behavior of
V T̃ ðtÞ∕V T̃ ð0Þ occurs with the changes in tissue temperature. This peculiarity
is related to the observed non-monotonic behavior of the contrast: the
abrupt decrease in Ṽ at the stage of tissue heating from room temperature to
65°C changes to a slight increase at the stage of modification [Fig. 8.31(b)].
This results in the larger value of Ṽ at the final phase of modification (when
the treatment is terminated) in comparison with the initial phase (when T
approaches 65°C) and, correspondingly, in the “loop-like” form of Ṽ ðTÞ
dependence. This “loop-like” feature is displayed by the inset in Fig. 8.31(a).
Fig. 8.31(c) shows the dependence V T̃ ðtÞ∕V T̃ ð0Þ obtained with the LASCA
modality. In calculation of the speckle contrast from raw images obtained
with the LASCA unit, the area of the analyzed zone of speckle-modulated
images was 0.7 0.7 mm2, which corresponded to the analyzed area of the
central part of the treatment zone with a size of 1.75 1.75 mm2. The area of
the same size was analyzed with the bundle-based modality.
Considering the contrast decrement DV T̃ ðtÞ ¼ V T̃ ðtÞ V T̃ ðt þ T̃Þ, we
can assume the first-order approximation of DV T̃ for the small values of
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
T̃∕tc : DV T̃ fðdg2 ðT̃Þ∕d T̃Þ∕2 g2 ðT̃Þ 1gT̃ KðT̃∕tc Þ and, respectively,
tc ∝ 1∕DV T̃ . The factor K is controlled by the asymptotic behavior of the
temporal correlation function of intensity in the object plane and the transfer
properties of the bundle-based analyzer as well. The static character of
captured speckle patterns at room temperature and the abrupt increase in the
contrast decrement with the increasing tissue temperature allow us to consider
the decay in tc at the first stage of tissue modification as a thermally activated
process with a certain value of the activation energy.
Figure 9.32 displays the values of lnðDV T̃ Þ estimated for the pairs of
sequential frames and plotted against the instantaneous inverse absolute
temperature in the central part of treatment zone. This plot corresponds to
data obtained with the conventional LASCA modality (with the single value
of output power of an IR laser) and the bundle-based speckle analyzer. For
the relatively narrow interval of T, the plotted data can be approximated with
reasonable accuracy by the linear dependencies of lnðDV T̃ Þ on T1, which are
also shown in Fig. 8.32.
This allows us to assume the Boltzmann-like dependence of the inverse
speckle decorrelation time on the absolute temperature of modified tissue:
t1
c,o K 1 expðE a ∕RTÞ, where the parameter K1 depends on various factors
such as the frequency factor of the basic thermally activated process with the
activation energy of Ea and the transfer properties of the speckle-imaging
system. Note that both data sets are characterized by the close values of
484 Chapter 8
Figure 8.32 The values of lnðDV T̃ Þ obtained with the bundle-based analyzer (1–3) and the
LASCA modality (4) and plotted against the inverse absolute temperature in the central
region of treatment zone. The output power of the erbium laser is: 4.5 W (1), 4.2 W (2),
4.8 W (3), 4.5 W (4).
Ea (61 ± 4.5) kJ/mol for data obtained with the LASCA modality and
(64 ± 6) kJ/mol for the bundle-based speckle analyzer, respectively,), but
significantly differ in the values of the normalization parameter (K1,LASCA/
K1,bundle 2.9). This difference is obviously caused by the difference in the
transfer properties of the conventional LASCA system and the bundle-based
speckle analyzer. In the latter case, the influence of the transfer properties of
the light-delivering channel is more expressed because of the cascade
transformation of speckle patterns and the lower quality of the image-
transferring component in comparison with the examined LASCA modality.
The above-considered linear and quadratic models of td decay predict the
extreme value of the ratio tc,o/tc,detector in the case of sequential two-stage
transformation of “boiling” speckle patterns and varies from 1.74 (the
quadratic model) to 3.24 (the linear model). Thus, under the semi-
qualitative assumptions used in theoretical analysis and without knowledge of
the correlation properties of dynamic speckles in the object plane, the
experimentally obtained ratio K1,LASCA/K1,bundle 2.9 reasonably agrees with
the above-presented values.
Let us consider some probable mechanisms of the laser-induced cartilage
modification, which cause dynamic light scattering in the treated tissue. One
mechanism can be related to the temperature-dependent changes in the rate of
scatter diffusion in the treated volume. For many condensed media, the
dependence of the diffusion coefficient on the temperature is described as:
D ¼ D0 exp(E0/RT), where E0 is the activation energy of transition of a
scattering particle from one stable position to another one. Among other
factors, the contrast decrement is determined by the average displacement
pffiffiffiffiffiffiffiffi
of scatters L over the observation time T̃ (i.e., DV T̃ ∝ L ∝ DT̃ ∝
expðE 0 ∕2RTÞ). Thus, this consideration leads to the Boltzmann dependence
of the contrast decrement on the characteristic temperature of tissue. Another
probable mechanism is related to the thermally induced alterations in tissue

structure, which are accompanied by the generation of new effective scattering
sites in the treated volume. Assuming the mono-molecular process of
thermally induced transformation of one of the basic components of cartilage
and the insignificant changes in its concentration during the observation time
T̃, we can obtain the contrast decrement proportional to the temperature-
dependent transformation rate DV T̃ ∝ Ñ˙ (where Ñ˙ is the rate of transforma-
tion of the scattering centers in the probed volume). With these assumptions,
the value of Ñ˙ will be controlled by the factor exp(Ea/RT). Thus, both
mechanisms lead to the Boltzmann-like dependence of DV T̃ on T.
The obtained effective activation energy can correspond to the thermally
induced transfer of tissue liquid, which causes the changes in local configurations
of scattering sites, or to the low-energy modification of the collagen matrix. Note
that the thermal denaturation of collagen itself can be excluded from the
considered mechanisms, which cause the pronounced structural changes at the
initial phase of modification. Typically, the expressed denaturation of collagen in
cartilage occurs at the temperatures above 70°C (Ref. 92). However, the
thermally induced dynamics of collagen matrix can be related to the partial
denaturation of proteoglycan aggregates (PGA)93 or to the conformational
transitions in glycosaminoglycans as one of the PGA components. Note that
each disaccharide unit in the glycosaminoglycan macromolecule is stabilized by
three hydrogen bonds94 with a total energy of ~ 60-75 kJ/mol,95 which is close to
the found value of the effective activation energy.
The comparison of V T̃ ðtÞ dependencies obtained with the LASCA
modality and the bundle-based analyzer allows us to emphasize the expressed
non-monotonic behavior of the exposure-dependent speckle contrast in the
first case. The “stationary” phase of thermal modification, which corresponds
to the plateau-like region of T(t) curve, and the phase of thermal relaxation
(after switching off the erbium laser) are featured by the remarkable
oscillations in V T̃ ðtÞ [Fig. 8.31(c)]. Such non monotonic behavior is related
to macroscopically heterogeneous non ergodic spatial distributions of
modulating speckles in the image plane, which are associated with
inhomogeneities of the cartilage structure. Similar statistical inhomogeneities
of image-modulating speckles varying in the course of tissue modification
were observed in the case of transillumination monitoring.81 On the contrary,
dynamic speckle patterns obtained with the use of the bundle-based modality
exhibit significant suppression of the contrast oscillations due to the specific
conditions of speckle formation (random phase modulation of partial optical
signals sampled by the bundle and their mixing in the diffraction zone; these
features result in the formation of ergodic dynamic speckle patterns in the
detector plane). This gives us the opportunity for more detailed analysis of
peculiarities in the behavior of the exposure-dependent speckle contrast at the
486 Chapter 8
various stages of thermal treatment. In particular, the existence of inverse

peaks on V T̃ ðtÞ dependencies [marked by arrows in Fig. 8.31(b)] and dramatic
changes in the contrast behavior at the “quasi-stationary” phase of
modification (the trend to increase with the increasing tissue temperature)
can be emphasized. The latter peculiarity is obviously manifested for data
obtained with the higher values of treatment power [curve 3 in Fig. 8.31(b)]
and is correlated with previously reported loop-like behavior of V T̃ ðTÞ in the
transillumination mode.81 The presumable interpretation of such behavior is
related to the change in a predominating mechanism of dynamic light
scattering in the thermally treated tissue at the temperatures above 70°C (from
generation of new scattering centers due to partial denaturation of the basic
components of cartilage structure toward migration of accumulated denatur-
ation products in the treatment zone). Note that in the case of too high a value
of treatment power [curve 3 in Fig. 8.31(a)] the heat consumption in the
treated tissue is insufficient and overheating of the tissue takes place [see
marked region in Fig. 8.31(a)]. This situation is characterized by the
remarkably lower values of V T̃ at the stage of thermal modification itself
[the time interval from 20 s to 40 s, Fig. 8.31(b)] in comparison with the
smoother regimes of the tissue treatment (curves 1 and 2).
Thus, we can summarize that the analysis of the exposure-dependent
speckle contrast in the backscattering mode with the developed bundle-based
analyzer is the appropriate tool for the monitoring of the laser-mediated
cartilage modification. The specific conditions of speckle pattern formation in
the developed full-field speckle unit make it possible to obtain the statistically
reliable estimates of the instantaneous values of diagnostical parameter (the
exposure-dependent contrast) by processing the small fragments of speckled
images (subframes). In turn, this will provide a sufficient decrease in the data
processing time as a further step toward the design of the real-time speckle
monitor, which is appropriate for clinical conditions.
8.8 Summary
In this chapter, we have considered some typical examples of laser speckle
techniques applied to diagnostics and functional imaging in biology and
medicine. The common feature of almost all of the methods presented is the
use of correlation analysis of the intensity fluctuations of time-varying speckle
patterns to evaluate the dynamic parameters of a scattering system (such as,
e.g., ensembles of erythrocytes moving through the microcapillary net).
Correlometry as an optical signal processing technique is closely related to
another method of analysis of dynamic speckle, the spectral analysis widely
used in laser Doppler flowmetry. In the latter case, values of the spectral
moments of the time-dependent speckle intensity fluctuations are used to
characterize the blood perfusion level for diagnosed tissues and organs.
Actually, the two techniques of signal processing are different ways of

describing the same physical phenomenon and the choice of the more
appropriate signal processing technique depends on the features of the
formation of the analyzed dynamic speckle pattern. In particular, correlation
analysis of the speckle intensity fluctuations is preferable because of the
light detection technique used (PMTs operating in photon-counting mode) in
the case of optically dense multiply scattering objects characterized by low
levels of detected intensity and wide spectra of the speckle intensity
fluctuations.
In addition, the statistical analysis of spatial speckle intensity variations
using time-averaged electronic images of the dynamic speckle patterns (LASCA
and related techniques) can give additional opportunities for the design of blood
perfusion monitors and imagers due to the high speed of the “quasi-parallel”
information processing and the possibility of fast reconstruction of the
functional images of the analyzed area without mechanical scanning.
Finally, the functional imaging of multiple scattering macroscopically
inhomogeneous media by means of contrast analysis is more an example of
the possibilities for the application of various coherence phenomena to the
probing of scattering media rather than a diagnostic technique that would be
acceptable for real clinical applications (due to its relatively low spatial
resolution and the low quality of the images obtained).
Acknowledgments
David J. Briers is very grateful to David Boas of the Harvard Medical School,
Andrew Dunn of the University of Texas, and Moor Instruments Ltd, for
their cooperation and permission to use some of their images.
Preparation of this chapter was possible due to the support from the
Russian Foundation for Basic Research (grants # 13-02-00440), basic tasks of
the state higher education institutions of research, the Ministry of Education
and Science of the Russian Federation (grant № 14.Z56.15.7102-MK).
Valery V. Tuchin is grateful for support from the Russian Presidential
grant NSh-703.2014.2, the Government of the Russian Federation grant 14.
Z50.31.0004, the Tomsk National Research State University Academic D.I.
Mendeleev Fund Program, and RFBR 14-02-00526a.
References
1. G. Mueller, B. Chance, R. Alfano, S. Arridge, J. Beuthan, E. Gratton,
M. Kaschke, B. Masters, S. Svanberg, and P. Van der Zee, Medical
Optical Tomography: Functional Imaging and Monitoring, SPIE Press,
IS11, Bellingham, WA (1993).
2. V.V. Tuchin, Tissue Optics: Light Scattering Methods and Instruments for
Medical Diagnostics, 3rd ed. PM254, SPIE Press, Bellingham, WA (2015).
488 Chapter 8
3. O. Minet, G. Mueller, and J. Beuthan (eds.), Selected Papers on Optical

Tomography, Fundamentals and Applications in Medicine, SPIE Press,
MS147, Bellingham, WA (1998).
4. H. Z. Cummins and E. R. Pike (eds.), Photon Correlation and Light
Beating Spectroscopy, Plenum Press, New York (1974).
5. H. Z. Cummins and E. R. Pike (eds.), Photon Correlation Spectroscopy
and Velocimetry, Plenum Press, New York (1977).
6. G. Maret and P.E. Wolf, “Multiple light scattering from disordered
media. The effect of Brownian motion of scatterers,” Z. Phys. B 65, 409–
413 (1987).
7. F.C. MacKintosh and S. John, “Diffusing-wave spectroscopy and
multiple scattering of light in correlated random media,” Phys. Rev. B
40(4), 2383–2406 (1989).
8. S.R. Arridge, M. Cope, and D.T. Delpy, “The theoretical basis for the
determination of optical pathlengths in tissue: temporal and frequency
analysis,” Phys. Med. Biol. 17(7), 1531–1560 (1992).
9. D.A. Boas and A.G. Yodh, “Spatially varying dynamical properties of
turbid media probed with diffusing temporal light correlation,” JOSA A
14(1), 192–215 (1997).
10. S. Feng, F. Zeng, and B. Chance, “Monte Carlo simulations of photon
migration path distributions in multiple scattering media,” Proc. SPIE
1888, 78–89 (1993).
11. I. Freund, M. Kaveh, and M. Rosenbluh, “Dynamic light scattering:
ballistic photons and the breakdown of the photon-diffusion approxi-
mation,” Phys. Rev. Lett. 60, (12), 1130–1133 (1988).
12. P.-A. Lemieux, M.U. Vera, and D.J. Durian, “Diffusing-light spectro-
scopies beyond the diffusion limit: The role of ballistic transport and
anisotropic scattering,” Phys. Rev. E 57(4), 4498–4515 (1998).
13. J.D. Briers, “Laser Doppler and time-varying speckle: a reconciliation,”
J. Opt. Soc. Am. A 13, 345–350 (1996).
14. H. Fujii, K. Nohira, Y. Yamamoto, H. Ikawa, and T. Ohura,
“Evaluation of blood flow by laser speckle image sensing,” Appl. Opt.
26, 5321–5325 (1987).
15. H. Fujii, “Visualization of retinal blood flow by laser speckle
flowgraphy,” Med. & Biol. Eng. & Comput. 32, 302–304 (1994).
16. Y. Tamaki, M. Araie, E. Kawamoto, S. Eguchi, and H. Fujii, “Non-
contact, two-dimensional measurement of retinal microcirculation using
laser speckle phenomenon,” Inv. Ophthalmol. & Vis. Sci. 35, 3825–3834
(1994).
17. N. Konishi and H. Fujii, “Real-time visualization of retinal microcircu-
lation by laser flowgraphy,” Opt. Eng. 34, 753–757 (1995).
18. T.J.H. Essex and P.O. Byrne, “A laser Doppler scanner for imaging
blood flow in skin,” J. Biomed. Eng. 13, 189–194 (1991).
19. G. Nilsson, A. Jakobsson, and K. Wårdell, “Tissue perfusion monitoring

and imaging by coherent light scattering,” Proc. SPIE 1524, 90–109
(1991).
20. H. Komine, S.J. Brosnan, A.B. Litton, and E.A. Stappaerts, “Real-time
Doppler Global Velocimetry,” AIAA Aerospace Sciences Meeting,
Nevada, January 1991, Paper AIAA-91-0337 (1991).
21. A.F. Fercher and J.D. Briers, “Flow visualization by means of single-
exposure speckle photography,” Opt. Commun. 37, 326–329 (1981).
22. J.D. Briers and A.F. Fercher, “A laser speckle technique for the
visualization of retinal blood flow,” Proc. SPIE 369, 22–28 (1982).
23. J.D. Briers and A.F. Fercher, “Retinal blood-flow visualization by
means of laser speckle photography,” Inv. Ophthalmol. Vis. Sci. 22,
255–259 (1982).
24. J.D. Briers, “Optical filtering techniques to enhance speckle contrast
variations in single-exposure speckle photography,” Optik 63, 265–276
(1982).
25. A.F. Fercher, M. Peukert, and E. Roth, “Visualization and measure-
ment of retinal blood flow by means of laser speckle photography,” Opt.
Eng. 25, 731–735 (1986).
26. J.D. Briers and S. Webster, “Quasi-real time digital version of single-
exposure speckle photography for full-field monitoring of velocity or
flow fields,” Opt. Commun. 116, 36–42 (1995).
27. J.D. Briers and S. Webster, “Laser speckle contrast analysis (LASCA): a
nonscanning, full-field technique for monitoring capillary blood flow,”
J. Biomed. Opt. 1, 174–179 (1996).
28. J.D. Briers, G. Richards, and X.W. He, “Capillary blood flow
monitoring using laser speckle contrast analysis (LASCA),” J. Biomed
Opt. 4, 164–175 (1999).
29. H. Cheng et al., “Modified laser speckle imaging method with improved
spatial resolution,” J. Biomed. Opt. 8, 559–564 (2003).
30. J.S. Paul, A.R. Luft, E. Yew, and F.S. Sheu, “Imaging the development
of an ischemic core following photochemically induced cortical
infarction in rats using laser speckle contrast analysis (LASCA),”
Neuroimage 29(1), 38–45 (2006).
31. S. Yuan, A. Devor, D.A. Boas, and A.K. Dunn, “Determination of
optimal exposure time for imaging of blood flow changes with laser
speckle contrast imaging,” Appl. Opt. 44, 1823–1830 (2005).
32. P. Zakharov, A. Völker, A. Buck, B. Weber, and F. Scheffold,
“Quantitative modelling of laser speckle imaging,” Opt. Letts. 31,
2465–3467 (2006).
33. X.W. He and J.D. Briers, “Laser speckle contrast analysis (LASCA): a
real- time solution for monitoring capillary blood flow and velocity,”
Proc. SPIE 3337, 98–107 (1998).
490 Chapter 8
34. A.K. Dunn, H. Bolay, M.A. Moskowitz, and D.A. Boas, “Dynamic
imaging of cerebral blood flow using laser speckle,” J. Cereb. Blood Flow
Metab. 1, 195–201 (2001).
35. A.N. Obeid, D.M. Boggett, N.J. Barnett, G. Dougherty, and P. Rolfe,
“Depth discrimination in laser Doppler skin blood flow measurement
using different lasers,” Med. And Biol. Eng. And Comput. 26, 415–419
(1988).
36. L. Duteil, J.C. Bernengo, and W. Schalla, “A double-wavelength laser
Doppler system to investigate skin microcirculation,” IEEE Trans.
BME-32, 439–447 (1985).
37. R.J. Gush and T.A. King, “Discrimination of capillary and arterio-
venular blood flow in skin by laser Doppler flowmetry,” Med. And Biol.
Eng. And Comput. 29, 387–392 (1991).
38. M.M. Gonik, A.B. Mishkin, and D.A. Zimnyakov, “Visualization of
blood microcirculation parameters in human tissues by time-integrated
dynamic speckles analysis,” Ann. NY Acad. Sci. 972, 325 (2002).
39. K. Forrester, C. Stewart, J. Tulip, C. Leonard, and R. Bray,
“Comparison of laser speckle and laser Doppler perfusion imaging:
Measurement in human skin and rabbit articular tissue,” Med. Biol.
Engg. Comp. 40, 687–697 (2002).
40. B. Choi, N. Kang, and J. Nelson, “Laser speckle imaging for monitoring
blood flow dynamics in the in vivo rodent dorsal skin fold model,”
Microvasc. Res 68, 143–146 (2004).
41. H. Cheng, Q. Luo, Q. Liu, Q. Lu, H. Gong, and S. Zeng, “Laser speckle
imaging of blood flow in microcirculation,” Phys. Med. Biol. 49, 1347–
1357 (2004).
42. S.K. Nadkarni et al., “Characterization of atherosclerotic plaques by
laser speckle imaging,” Circ. 112, 885–892 (2005).
43. K. Yaoeda et al., “Measurement of microcirculation in optic
nerve head by laser speckle flowgraphy in normal volunteers,” Am. J.
Opthalmol. 130, 606–610 (2000).
44. M. Nagahara, Y. Tamaki, M. Araie, and T. Umeyama, “The acute
effects of stellate ganglion block on circulation in human ocular fundus,”
Acta Ophthalmol. Scand. 79, 45–48 (2001).
45. J. Flammer et al., “The impact of ocular blood flow in glaucoma,” Prog.
Retin. Eye Res. 21, 359–393 (2002).
46. B. Weber, C. Burger, M. Wyss, G. von Schulthess, F. Scheffold, and
A. Buck, “Optical imaging of the spatiotemporal dynamics of cerebral
blood flow and oxidative metabolism in the rat barrel cortex,” Eur. J.
Neurosci. 20, 2664–2670 (2004).
47. T. Durduran et al., “Spatiotemporal quantification of cerebral blood
flow during functional activation in rat somatosensory cortex using laser
speckle flowmetry,” J. Cereb. Blood Flow Metab. 24, 518–525 (2004).
48. D. Atochin et al., “Mouse model of microembolic stroke and

reperfusion,” Stroke 35, 2177–2182 (2004).
49. A. Kharlamov, B. Brown, K. Easley, and S. Jones, “Heterogeneous
response of cerebral blood flow to hypotension demonstrated by laser
speckle imaging flowmetry in rats,” Neurosci. Lett 368, 151–156 (2004).
50. C. Ayata, A. Dunn, Y. Gursoy-Ozdemir, Z. Huang, D. Boas, and
M. Moskowitz, “Laser speckle flowmetry for the study of cerebrovascu-
lar physiology in normal and ischemic mouse cortex,” J. Cereb. Blood
Flow Metab. 24, 744–755 (2004).
51. H. K. Shin, A.K. Dunn, P.B. Jones, D.A. Boas, M.A. Moskowitz, and
C. Ayata, “Vasoconstrictive neurovascular coupling during focal
ischemic depolarizations,” J. Cereb. Blood Flow Metab. 25, 1–13 (2005).
52. A.K. Dunn, A. Devor, A.M. Dale, and D.A. Boas, “Spatial extent of
oxygen metabolism and hemodynamic changes during functional
activation of the rat somatosensory cortex,” Neuroimage 27, 279–290
(2005).
53. A. Strong, E. Bezzina, P. Anderson, M. Boutelle, S. Hopwood, and
A. Dunn, “Evaluation of laser speckle flowmetry for imaging cortical
perfusion in experimental stroke studies: quantitation of perfusion and
detection of periinfarct depolarisations,” J. Cereb. Blood Flow Metab.
26, 645–53 (2006).
54. K. Murari, N. Li, A. Rege, X. Jia, A. All, and N. Thakor, “Contrast-
enhanced imaging of cerebral vasculature with laser speckle,” Appl. Opt.
46, 5340–5346 (2007).
55. Z. Wang, S. Hughes, S. Dayasundara, and R.S. Menon, “Theoretical
and experimental optimization of laser speckle contrast imaging for high
specificity to assess brain microcirculation,” J. Cereb. Blood Flow Metab.
27, 258–269 (2007).
56. H. Bolay, U. Reuter, A.K. Dunn, Z. Huang, D.A. Boas, and M.A.
Moskowitz, “Intrinsic brain activity triggers trigeminal meningeal
afferents in a migraine model,” Nature Med. 8, 136–142 (2002).
57. A.C. Völker, P. Zakharov, B. Weber, A. Buck, and F. Scheffold, “Laser
speckle imaging with an active noise reduction scheme,” Opt. Express 13,
9782–9787 (2005).
58. A. Serov, W. Steenbergen, and F. de Mul, “Prediction of the
photodetector signal generated by Doppler-induced speckle fluctuations:
theory and some validations,” J. Opt. Soc. A 18, 622–630 (2001).
59. R. Bandyopadhyay, A.S. Gittings, S.S. Suh, P.K. Dixon, and D.J.
Durian, “Speckle-visibility spectroscopy: a tool to study time-varying
dynamics,” Rev. Sci. Instrum. 76, 093110 (2005).
60. M.A. Davis, S. M. Shams Kazmi, and A. K. Dunn “Imaging depth
and multiple scattering in laser speckle contrast imaging,” Journal of
Biomedical Optics 19(8), 086001–1-10 (2014).
492 Chapter 8
61. J.D. Briers, “Laser speckle contrast imaging for measuring blood flow,”
Optica Applicata 37, 139–152 (2007).
62. D. Briers, D.D. Duncan, E. Hirst, S.J. Kirkpatrick, M. Larsson,
W. Steenbergen, T. Stromberg, and O.B. Thompson, “Laser speckle
contrast imaging: theoretical and practical limitations,” J Biomed Opt.
18(6), 066018 (2013).
63. A. Serov, W. Steenbergen, and F. De Mul, “A method for estimation of
the fraction of Doppler-shifted photons in light scattered by mixture of
moving and stationary scatterers,” Proc. SPIE 4001, 178–189 (2000).
64. A. Sadhwani, K.T. Schomacker, G.J. Tearney, and N.S. Nishioka,
“Determination of Teflon thickness with laser speckle. I. Potential for
burn depth diagnosis,” Applied Optics 35(28), 5727–5735 (1996).
65. D.A. Zimnyakov, A.A. Isaeva, E.A. Isaeva, O.V. Ushakova, and
R.A. Zdrazhevskii, “Speckle Correlometry Method for Evaluating the
Transport Scattering Coefficient of a Randomly Inhomogeneous
Medium,” Technical Physics Letters 38(10), 935–937 (2012).
66. L.F. Rojas, M. Bina, G. Cerchiari, M.A. Escobedo-Sanchez, F. Ferri,
and F. Scheffold, “Photon path length distribution in random media
from spectral speckle intensity correlations,” Eur. Phys. J. ST. 199(1).
167–180 (2011).
67. M. Draijer, E. Hondebrink, T. Leeuwen, and W. Steenbergen, “Review
of laser speckle contrast techniques for visualizing tissue perfusion,”
Lasers in Medical Science, 639–651 (2009).
68. H. Cheng, Q. Luo, S. Zeng, S. Chen, J. Cen, and H. Gong, “Modified
laser speckle imaging method with improved spatial resolution,”
J. Biomed. Opt. 8, 559-564 (2003).
69. P. Li, S. Ni, L. Zhang, S. Zeng, and Q. Luo, “Imaging cerebral blood
flow trough the intact rat skull with temporal laser speckle imaging,”
Opt. Lett. 31, 1824–182 (2006).
70. K.R. Forrester, C. Stewart, J. Tulip, C. Leonard, and R.C. Bray,
“Comparison of laser speckle and laser Doppler perfusion imaging:
measurement in human skin and rabbit articular tissue,” Med. Biol. Eng.
Comput. 40, 687–697 (2002).
71. K.R. Forrester, J. Tulip, C. Leonard, R.C. Bray, and C. Robert,
“A laser speckle imaging technique for measuring tissue perfusion,”
IEEE Trans. Biomed. Eng. 51, 2074-2084 (2004).
72. N. Konishi, Y. Tokimoto, K. Kohra, and H. Fujii, “New laser
speckle flowgraphy system using CCD camera,” Opt. Rev. 9, 163-196
(2002).
73. Y.K. Tan, W.Z. Liu, Y.S. Yew, S.H. Ong, and J.S. Paul, “Speckle image
analysis of cortical blood flow and perfusion using temporally derived
contrasts,” In: International conference on image processing ICIP 2004.
Proc. IEEE 5, 3323–3326 (2004).
74. T.M. Le, J.S. Paul, H. Al-Nashash, A. Tan, A.R. Luft, F.S. Sheu, and
S.H. Ong “New insights into image processing of cortical blood flow
monitors using laser speckle imaging,” IEEE Trans. Biomed. Eng. 26,
833–842 (2007).
75. A.B. Parthasarathy, W.J. Tom, A. Gopal, X. Zhang, and A.K. Dunn
“Robust flow measurement with multi-exposure speckle imaging,” Opt.
Express. 16, 1975–1989 (2008).
76. K. Basak, G. Dey, M. Mahadevappa, M. Mandal, and P.K. Dutta,
“In vivo laser speckle imaging by adaptive contrast computation for
microvasculature assessment,” Optics and laser in Engineering 62,
87–94 (2014).
77. M.C. Sicairos, S.T. Thurman, and J.R. Fienup, “Efficient subpixel
image registration algorithms,” Opt Lett. (2), 156–8 (2008).
78. C.A. Thompson, K.J. Webb, and A.M. Weiner, “Imaging in scattering
media by use of laser speckle,” JOSA A 14(9), 2269–2277 (1997).
79. D.A. Zimnyakov, Jung-Taek Oh, Yu.P. Sinichkin, V.A. Trifonov, and
E.V. Gurianov, “Polarization-sensitive speckle spectroscopy of scatter-
ing media beyond the diffusion limit,” J. Opt. Soc. Am. A 21, 59–70
(2004).
80. D.A. Zimnyakov and M.A. Vilensky, “Blink speckle spectroscopy of
scattering media,” Optics Letters 31(4), (2005). 31, 429–431 (2006).
81. D.A. Zimnyakov, D.N. Agafonov, A.P. Sviridov, A.I. Omel’chenko,
L.V. Kuznetsova, and V.N. Bagratashvili, “Speckle-contrast monitoring
of tissue thermal modification,” Applied Optics 41, 5989–5996 (2002).
82. V. Viasnoff, F. Lequeux, and D.J. Pine, “Multispeckle diffusing-wave
spectroscopy: A tool to study slow relaxation and time-dependent
dynamics,” Review of Scientific Instruments 73, 2336–2344 (2002).
83. S.E. Skipetrov and R. Maynard, “Dynamic multiple scattering of light in
multilayer turbid media,” Phys. Lett. A 217, 181–185 (1996).
84. S. Romer, F. Scheffold, and P. Schurtenberger, “Sol-gel transition of
concentrated colloidal suspensions,” Phys. Rev. Lett. 85, 4980–4983
(2000).
85. F. Scheffold, S. E. Skipetrov, S. Romer, and P. Schurtenberger,
“Diffusing-wave spectroscopy of non-ergodic media,” Phys. Rev. E 63,
061404 (2001).
86. D.A. Zimnyakov, A.P. Sviridov, L.V. Kuznetsova, S.A. Baranov, and
N. Yu. Ignatieva, “Monitoring of tissue thermal modification with a
bundle-based full-field speckle analyzer,” Applied Optics 45, 4480–4490
(2006).
87. L. Mandel and E. Wolf, Optical Coherence and Quantum Optics,
Cambridge University Press, 1995.
88. T. Yoshimura, “Statistical properties of dynamic speckles,” J. Opt. Soc.
Am. A 3, 1032–1054 (1986).
494 Chapter 8
89. T. Yoshimura, K. Nakagawa, and N. Wakabayashi, “Rotational and

boiling motion of speckles in a two-lens imaging system,” J. Opt. Soc.
Am. A 3, 1018–1022 (1986).
90. H.Z. Cummins and E.R. Pike, eds. Photon Correlation and Light-Beating
Spectroscopy. NATO Advanced Study Institute Series B: Physics,
Plenum Press, New York (1974).
91. N. Takai, T. Iwai, and T. Asakura, “Correlation distance of dynamic
speckles,” Applied Optics 22, 170–177 (1983).
92. N. Yu. Ignatieva, V.V. Lunin, S.V. Averkiev, A.F. Maiorova, V.N.
Bagratashvili, and E.N. Sobol, “DSC investigation of connective tissues
treated by IR-laser radiation,” Thermochimica Acta 422, 43–48 (2004).
93. A.M. Jamieson, J. Blackwell, H. Reihanian, H. Ohno, R. Gupta, D.A.
Carrino, A.I. Caplan, L.H. Tang, and L.C. Rosenberg, “Thermal and
solvent stability of proteoglycan aggregates by quasielastic laser light-
scattering,” Carbohydrate Research 160, 329–341 (1987).
94. J.E. Scott, “Secondary and tertiary structures of hyaluronan in aqueous
solution. Some biological consequences,” In: Science of Hyaluronan
Today. V.C. Hascall and M. Yanagishita, eds. (1998). http://www.
glycoforum.gr.jp/science/hyaluronan/hyaluronanE.html
95. G.C. Pimentel and A.L. McClennan, The Hydrogen Bond. (W.H.
Freeman, San Francisco, 189 (1960) 189.
Dmitry A. Zimnyakov is a professor and chairman of Physics

Department at Yuri Gagarin Saratov State Technical
University. He is also the chief researcher of the laboratory
at the Institute of Precision Mechanics and Control, RAS.
He is a member of international scientific societies SPIE and
OSA, as well as the Soros Professor in 2000–2001. His
research interests include biophotonics, statistical optics,
laser physics, and biomedical optics. He is the author, editor,
or the co-author of over 20 books, chapters in books, special issues of
journals, conference proceedings, textbooks, lecture manuals, and brochures,
and more than 200 scientific articles and reviews.
Olga V. Ushakova is an associate professor of the Electronics

and Telecommunications Department of Yuri Gagarin
Saratov State Technical University. She is also director of
the Research and Education Center “Fundamental and
Applied Photonics” in Yuri Gagarin Saratov State Technical
University. She is a member of SPIE. Her research and
educational interests include biophotonics and nanobiopho-
tonics, laser physics, and biomedical optics. She is author
and co-author of more than 30 articles and reviews.
David J. Briers is a scientific consultant and Emeritus

Professor of Applied Optics at Kingston University, London,
UK, where he was Head of the School of Applied Physics
from 1990 to 1999. He had previously held posts in UK
industry (1964–69 and 1983–89) and a government standards
laboratory in New Zealand (1969–80), and had been a
Visiting Professor and Research Fellow at the University of
Essen, Germany (1980–83). Since 1980, his main research
interest has been biomedical applications of laser speckle techniques,
especially the use of laser speckle contrast analysis to monitor blood flow
and perfusion. He is a Fellow of both the UK Institute of Physics and the
European Optical Society.

and OSA, and has been awarded Honored Science Worker of the Russia,
SPIE Educator Award, FiDiPro (Finland), Chime Bell Prize of Hubei
Province (China), and Joseph W. Goodman Book Writing Award (OSA/
SPIE).
Chapter 9
Optical Assessment of Tissue
Mechanics
Sean J. Kirkpatrick
Department of Biomedical Engineering, Michigan Technological University,
Houghton, USA
Donald D. Duncan
Department of Electrical and Computer Engineering, Portland State University,
Portland, USA
Brendan F. Kennedy
Optical þ Biomedical Engineering Laboratory, School of Electrical, Electronic, and
Computer Engineering, The University of Western Australia, Perth, Australia
David D. Sampson
Optical þ Biomedical Engineering Laboratory, School of Electrical, Electronic,
and Computer Engineering, and Center for Microscopy, Characterization, and
Analysis, The University of Western Australia, Perth, Australia
Additional nomenclature of definitions

uij Stress tensor
Cijkl Constitutive constants
ekl Strain tensor
se Eulerian stress
L Stretch ratio
se Specific stress value on stress–stretch curve
L* Specific stretch value on stress–stretch curve
Io Unstretched length of a sample
I Instantaneous stretched length of a sample
ro Tissue density
W Strain energy
ro W Strain energy density function
497
498 Chapter 9
eij Green’s strain components

eu Green’s strain in the circumferential direction
ez Green’s strain in the axial direction
C0 Constitutive constant with units of stress
an,bn Dimensionless material constants
T Stress tensor
S Strain rate tensor
Kijkl Material constant
qijkl Material constant
ε Strain
mε microstrain; 10–6 strain
s Stress
E Young’s modulus
t0 Retardation time
D(t) Creep compliance
p(I) Intensity probability distribution function
dsp Mean objective speckle size
d 0sp Mean subjective speckle size
d 00sp Speckle size on scattering surface
M Magnification
D Diameter of uniformly illuminated spot or system pupil
di Distance from lens to the imaging plane
uo Observation angle
us Illumination angle
Lo Observation distance
Ls Source distance
ax In-plane motion
az Out-of-plane motion
εxx In-plane strain
Vy Rotation about the y axis
dx Shift parameter
dA Differential speckle motion
m1,m2 Slopes of focused line in frequency space
s Standard deviation
a Probability level
ft Temporal frequency
fs Spatial frequency
X Second-order constant
n Refractive index
g(x) One-dimensional speckle record
R(t) Cross-correlation sequence
ε2j Mean square error
dj(xi) Speckle signal comprised of a deterministic signal plus noise
Optical Assessment of Tissue Mechanics 499
p(nj) Noise probability density function

Rj(xi) Quotient of oppositely shifted records
Hj(dx) Entropy in Rj(xi)
r2j Cross-correlation coefficient
S(y) Spectral density
v(z, t) Depth-resolved tissue velocity
Gi Grayness value
I Average intensity
9.1 Introduction
Optical elastography, as the mechanical characterization of biological tissues
using optical methods has come to be known, has developed substantially
since the original printing of this handbook in 2002. Advances in both
hardware and in data-processing algorithms have permitted these develop-
ments. Full-field 2-D and 3-D elastograms can now be generated in only a few
seconds. These developments have led to an increased focus on applying
elastography to clinical and biological applications, including in dermatology,
oncology, ophthalmology, and cardiology. This updated chapter discusses
several of the recent advances in optical elastography, including a discussion
of what is really measured, mechanically speaking, with optical elastography.
A discussion of some recent advances in OCT-based elastography, optical
coherence elastography (OCE), is also presented.
9.2 Introduction to Prior Edition

Noncontacting strain measurement is frequently required in many biomedical
situations where the more conventional methods of strain measurement,
such as contact strain gauges and extensometers, cease to function
effectively. Examples of these situations include the evaluation of displace-
ment derivatives in soft tissues for biomechanical investigations or for
medical diagnostics where the micro- and macro-mechanics of pathological
tissue change as a function of disease progression. The goal of this chapter is
to present an overview of a few coherent light-based techniques that are
particularly suitable for the mechanical interrogation of biological tissues
and to demonstrate practical applications of these techniques in biomedical
diagnostics. As an analogy to parallel efforts by researchers in the
ultrasound community,1 we refer to this general field of study as optical
elastography.
The use of laser speckle techniques for the mechanical characterization of
materials is well known in the non destructive evaluation community.2–5 One
particular application is to infer strain by monitoring the motion of the
speckle pattern that results from coherently illuminating a stressed object.
Typically, a reference image of the speckle pattern is acquired before
500 Chapter 9
deformation of the object. Motion (with respect to this reference image) of

subsequent speckle patterns, which occurs as the object undergoes deforma-
tion, is used to infer the corresponding strain. A problem experienced in using
this technique for measurements of hydrated tissues is the rapid decorrelation
of the speckle patterns.6 Thus, application of speckle techniques to assessment
of strain in biological tissues often relies upon rapid sampling of the speckle
patterns and the use of processing algorithms that are aimed at inferring strain
rates rather than absolute strains.
9.3 Tissue Mechanics and Medicine

The goal of optical elastography is to noninvasively, or minimally invasively,
quantify meaningful mechanical constants of tissues in a manner that provides
clinically relevant information. It is well known that many disease processes,
such as tumors of the breast and prostate, manifest as stiff nodules relative to
the surrounding tissue. This characteristic often allows for their detection
through manual palpation, for example, in breast cancer and skin cancer.
However, soft-tissue palpation is not only qualitative but also highly
subjective. Furthermore, it provides information on a relatively large spatial
scale. Thus, the ability to detect small tumors by palpation is limited.
Palpation is also limited to those areas of the body that are accessible to
touch. To overcome some of the limitations of manual palpation and to allow
for a noninvasive mechanical probe deep into the body, a number of non-
optical imaging modalities for visualizing the distributions of tissue stiffness in
vivo have been devised and are reviewed in the literature.7 Examples of such
methods include vibration amplitude sonoelastography,8 vibration phase
gradient sonoelastography,8 elastography,1 spectral tissue strain measure-
ment,9 and various methods employing MRI data.10 These methods have
been applied to a range of clinical applications, most notably breast cancer
and fibrotic liver diseases, such as liver fibrosis and cirrhosis. Critical issues,
such as imaging low-contrast elastic modulus distributions in tissue,7 out-of-
plane displacements, and image decorrelation,11 are still being addressed in
these methods.
Optical elastography offers increased spatial resolution (i.e., smaller
lesions may be detected) and the potential for better strain resolution (low-
contrast elastic modulus distributions may be visualized) than other
elastographic methods. Strain resolution on the order of a single microstrain
has been evaluated in biological tissue optically.12 However, this increased
spatial and strain resolution is gained at the cost of decreased probing depth,
as optical methods are limited to the outer few millimeters of tissue. To
partially overcome this issue, endoscopic13 and interstitial (needle-based)
optical elastography14 probes have been proposed. An additional consider-
ation is that most optical elastography methods are limited, in that only
relatively small areas or volumes of tissues may be probed at any one time.
Nevertheless, optical methods still can be useful in the early detection of

neoplastic changes because many of these early changes occur in the mucosa
and submucosa of the affected organs. In addition, in certain surgical
procedures it may be feasible to use optical elastography methods after the
tissue has been cut, providing access to tissue well below the tissue surface.
In the following subsections, we describe a number of application areas in
which optical elastography is being developed and tested, and speculate on
future areas of interest for this technology.
9.3.1 Dermatology
Skin imaging is among the most attractive application areas for optical
elastography, as skin is amenable both to optical imaging and mechanical
loading. A number of examples of in vivo skin imaging have been
demonstrated in optical elastography.15–19 Importantly, several diseases and
conditions alter the mechanical properties of skin. For example, subsurface
skin tumors present as objects with distinct mechanical properties relative to
the surrounding normal tissue. The displacement of fibrillar papillary dermis
by the cellular mass of a growing melanoma is one example of this.
Preliminary measurements using an acousto-optical elastography technique
have demonstrated mechanical contrast in vivo on the micro-scale in both a
mouse melanoma lesion and a human melanocytic nevi.15 Optical elasto-
graphic techniques may provide a means of probing these masses to determine
their state of progression and thereby help to determine a proper means of
disease management. Other skin conditions, such as psoriasis, scleraderma,
and icthyosis, also present as localized tissue areas with distinct mechanical
properties that can be delineated optically. Additionally, it is well established
that scarring significantly alters the mechanical properties of skin.20,21 Indeed,
pliability is one of the four parameters used in the Vancouver scar scale, an
assessment scale commonly used by clinicians. Preliminary measurements,
based on optical palpation, a derivative of optical coherence elastography
(OCE), have demonstrated mechanical contrast between scars and normal
tissue.22 An example of optical elastography on skin performed using an
acousto-optical approach is shown in Fig. 9.1.
9.3.2 Oncology
Optical elastography holds great promise for improving the treatment of
cancer. A number of preliminary results have demonstrated this using both
OCE23–25 and digital holography-based elastography.26 Ultrasound elasto-
graphy and, to a lesser extent, magnetic resonance elastography have been
widely used clinically in oncology, mainly for the detection of suspicious lesions
in breast cancer. The higher spatial resolution and limited penetration depth of
optical elastography suggest that it is better suited to other aspects in the
treatment of cancer patients. For example, in breast cancer a main problem
502 Chapter 9
Figure 9.1 Skin optical elastography. (a) White light photograph, (b) elastogram of the region
indicated by the black box in (a). The less-stiff nevus seen in the photograph presents as a
region of relatively high strain in the elastogram. Reproduced with permission from Ref. 15
6HHFRORUSODWHV
facing the surgeon is in identifying during the surgery if all malignant tissue
has been removed. Techniques such as frozen section histology and imprint
cytology are in use, but re-excision rates remain very high (20–60%).27 Optical
elastography could potentially be used to scan the perimeter of the excised
tumor mass sufficiently rapidly to provide feedback to the surgeon during the
initial surgery. Preliminary results from OCE have demonstrated visualization
of microstructure within the breast and have indicated the promise of optical
elastography in distinguishing malignant tissue from surrounding healthy
tissue.25 An example of OCE on human breast tissue is presented in Fig. 9.2.
Recently, optical elastography has also been shown to detect prostate cancer
in biopsy samples with sensitivity and specificity of 98% and 91%,
respectively.28 The next steps needed for optical elastography in oncology are
larger-scale studies to assess the diagnostic accuracy in detecting cancer.
9.3.3 Ophthalmology
As OCT is widely used in ophthalmic applications, it is a natural area in which
to develop optical elastography. Coupled with this, several diseases of the eye
are linked with changes to its mechanical properties, including glaucoma,
keratoconus, and presbyopia. Initial efforts in optical elastography have
focused on corneal biomechanics and, in particular, on keratoconus, a disease
that reduces corneal stiffness. A number of OCE techniques have been
proposed, based on both contact29,30 and noncontact31,32 mechanical loading
of the cornea. Noncontact mechanical loading is of particular importance in
this delicate tissue and has been achieved using both an air-puff port and
photothermal expansion via a pulsed laser. Figure 9.3 shows an example of
OCE performed on a mouse cornea. As well as measuring corneal stiffness,
several papers have proposed OCE as a technique to assess the treatment of
keratoconus with collagen cross-linking33 induced by UV light exposure.34
Figure 9.2 Optical elastography of malignant human breast tissue. (a) OCT image, (b)
OCE elastogram, and (c) co-registered histology. A, adipose; D, duct; M, smooth muscle;
T, region densely permeated with tumor; and V, blood vessel. In the micro-elastogram, the
scale is in millistrain. The insets show a 2.5X magnification of the blue-dotted boxes.
Reproduced with permission from Ref. 256HHFRORUSODWHV
504 Chapter 9
Figure 9.3 Ophthalmic optical elastography. (a) 3D-OCT image of a mouse eye. The white
dots indicate the regions subjected to mechanical loading. (b) Graph showing the variation in
phase velocity of surface waves propagating in the cornea as a function of age. Reproduced
with permission from Ref. 134.
In addition to OCE, Brillouin microscopy shows great promise as a non-

contact mechanical imaging technique for estimating corneal elasticity, and
has also been shown to map stiffening of the cornea as a result of collagen
crosslinking. Brillouin microscopy also shows potential in measuring age-
related stiffening of the crystalline lens.35
9.3.4 Cardiology
In cardiovascular disease, the mechanical properties of atherosclerotic plaques
have been linked to plaque rupture, and have been investigated using
intravascular ultrasound elastography.36 However, the spatial resolution of
ultrasound elastography is relatively low and optical elastography may
provide higher resolution measurements on a scale more relevant to the
disease morphology. A number of preliminary studies have been reported,
using both OCE37–39 and laser speckle-based elastography.12,40 A preliminary
measurement performed on excised human coronary artery is presented in
Fig. 9.4. A key requirement for cardiology is the development of an
intravascular optical elastography probe to enable in vivo imaging. While this
presents technical challenges, not least the incorporation of mechanical
loading in a small footprint probe, optical elastography will likely take
advantage of advances in closely related intravascular OCT imaging probes.41
9.3.5 Other application areas

Optical elastography is also beginning to play a role in the field of tissue
engineering.42–44 For tissues that must sustain dynamic loads, understanding
the mechanics of the neotissue may provide insight into the ontogenetic
evolution of the mechanical properties of tissues and may also provide
clinically relevant data. For example, in the growth of new bone, it may be of
value to track the development of the strength and stiffness of the tissue in
order to determine the appropriate time for allowing the new tissue to be
physiologically loaded, or to develop a model that correlates the rate of
Figure 9.4 Optical elastography in cardiology. (a) 3D-OCT image of a human cadaver
coronary artery, (b) corresponding OCE image. The red region was confirmed by histology to
correspond to an atherosclerotic lesion. Scale bars, 1 mm. Reproduced with permission from
Ref. 396HHFRORUSODWHV
resorption of the tissue scaffold with the development of mature tissue. A

further example lies in tissue engineering for the repair of cartilage that is
designed to bear compressive loads. Immature cartilage does not respond to
compressive loads in the same manner as mature cartilage. Because of this
differential response to compressive loading between mature and neocartilage,
premature loading of the immature tissue may lead to failure of the
developing tissue.45 Furthermore, it has been established that the mechanical
properties of the scaffold upon which the engineered cartilage is grown and
the mechanical environment in which the cartilage is grown have direct
influences on the mechanics of the resultant cartilage. Thus, it is of use to be
able to quantitatively evaluate these variables46 and optical elastography may
provide the only nondestructive method to do this with sufficient spatial and
mechanical resolution. With a view to providing a tool to assist surgeons
during laser construction and regeneration, preliminary measurements using
OCE have demonstrated the temperature dependence of the stiffness of
porcine cartilage.47
Optical elastography may also find application to the measurement of
cellular-scale mechanics. The impairment of a cell’s capacity to generate and
respond to mechanical forces is central to the pathogenesis of diseases such as
asthma, atherosclerosis, cancer, glaucoma, and muscular dystrophy.48 Tools
such as atomic force microscopy (AFM) have been used to probe cell
mechanics and have provided insights into these disease processes. However,
these tools will be challenging to employ in situ. Optical elastography could
overcome this limitation to reveal the cell mechanics of living tissue, thereby
generating new knowledge central to the genesis, onset, and manifestation of
disease.
9.4 Constitutive Relations in Biological Tissues

Constitutive equations of materials describe the macroscopic stress response
(or, equivalently, heat flux vector) to an applied strain. These equations are
506 Chapter 9
specific for a particular material under very restrictive conditions.

Nevertheless, there are a few general approaches to determining constitutive
models that can be applied to nearly all biological tissues. It must be
emphasized, however, that constitutive models are just models, a series of
equations that approximate the behavior of specific tissue types under very
well defined and controlled conditions. Therefore, they should be viewed
with some caution and with the full realization that the predictions of a
particular constitutive model are strictly limited to the input parameters of
the model. Constitutive models do remain, however, the best method for
relating theoretical predictions of material behavior to experimental results.
In fact, proper determination of a constitutive model demands input from
both theoretical considerations and empirical data. Ideally, models will be
developed that describe the constitutive behavior of these tissues not only in
vitro but also in vivo. A complete discussion of the processes involved in
determining appropriate constitutive models, and a listing of the axioms of
constitutive theory can be found in most textbooks dealing with the subject
of continuum mechanics; the reader is strongly encouraged to consult such a
reference.49
In the case of most traditional engineering materials (i.e., metals), the
behavior can be described adequately under certain conditions with a simple,
linear constitutive model:
uij ¼ C ijkl ekl , (9.1)
where the stress tensor, uij, is linearly related to the strain tensor, ekl. The
variables Cijkl are known as constitutive constants. In this case, the con-
stitutive constants are simply the elastic moduli, and Eq. (9.1) is a simplified
expression of Hooke’s law. One of the main reasons for the simplicity of this
model is that, in these materials, the constitutive constants are assumed to
be independent of both the direction of the applied strain and the rate of
strain application. Neither of these assumptions is truly physically valid,
however, they are assumed to be valid for simplicity and based on evidence
from empirical data. Equation (9.1) also assumes linearity over a range of
strains. This is not a physically valid assumption because linear elasticity
assumes that the stress tensor is a linear function of the strain tensor only
for infinitesimal strains. However, empirical data again show that this
assumption is sufficient for describing the macroscopic behavior of these
materials within their elastic range. This simple discussion demonstrates the
synergistic relationship between theory and experiment in the development of
constitutive equations.
Biological tissues, however, are nonlinear, nonuniformly inhomogeneous,
demonstrate a distinct dependence upon time and rates (viscoelasticity),
poroelastic, anisotropic, and capable of undergoing very large strains. In the
very simple case of a tissue in a uniaxial stress state, Tanaka and Fung50
proposed an exponential form as a constitutive model of soft tissue in
tension:
se ¼ ðse þ bc Þ exp½ac ðL L Þ bc , (9.2)
where se is Eulerian stress, L is known as the stretch ratio, which is defined as

I/Io, where I is the stretched length of a tissue sample and Io is the unstretched
length of the sample,51 se and L are specific points on the stress–stretch
curve, and ac and bc are constitutive constants. An assumption in this
equation is that the tissue is incompressible. In order to model multiaxial
stress states, much more complicated equations are needed. Unfortunately,
because of the nonlinear behavior of tissue, there is no unique formulation of
the constitutive relationships and any model that relates stress to strain with
sufficient accuracy for the application is probably sufficient. Four general
models (polynomial, exponential, logarithmic, and power law) appear to be
used most frequently, and all rely upon the form of the strain energy density
function, or a pseudo-elastic potential,52 for the tissue being modeled. The
strain energy density function, roW, is a function purely of the (Green’s) strain
components, eij, and is a measure of the internal strain energy per unit volume
for an isotropic process (see Sec. 9.9 for a definition of Green’s strain
components). The stress can be obtained from the strain energy function, W,
by a partial differentiation procedure:
∂W
sij ¼ ði, j ¼ 1,2,3Þ. (9.3)
∂eij
Much of the work on developing constitutive equations for biological

tissues has focused on cardiovascular tissues, with much less emphasis on
other tissue types, such as skin,53 ligament,54 and bone.55–58 In the
development of these relations for vascular tissue, many simplifying
assumptions have been made in order to produce equations that relate well
to empirical data. The assumptions include cylindrical geometry, homogene-
ity, incompressibility, cylindrical orthotropism, lack of shear stresses and
strains, and hyperelasticity.59,60 Viashnav61,62 has advocated the use of a
polynomial form of the strain energy density function and experimentally
demonstrated that a seven-constant form of the polynomial is sufficient to
describe the behavior of canine thoracic aorta, while only a three-constant
form is necessary to describe the behavior of peripheral arterial tissue.63 The
polynomial takes the following form:
ro W ¼ Ae2u þ Beu ez þ Ce2z þ De3u þ Ee2u þ E z þ F eu e2u þ Ge3z þ · · · ,

(9.4)
508 Chapter 9
where eu and ez are the Green’s strains in the circumferential and axial
directions, respectively, and A, B, etc., are constitutive constants. By taking
the partial derivatives of W with respect to the strains, one can evaluate the
stresses in the u and z directions.
Fung52,64 proposed a simpler expression for the elastic properties of
arterial tissue based on an exponential function that yields the relationship
between the average circumferential stresses and axial stresses and strains in
the same directions (denoted by the subscripts 1 and 2, respectively):
ro W ¼ C 0 expða1 e21 þ a2 e22 þ 2a4 e1 e2 Þ. (9.5)
The constants C0 (with units of stress), a1, a2, and a4 (dimensionless) are
material constants. For soft tissues, Eq. (9.5) implies that the incremental
moduli increase with increasing stresses.
Takamizawa and Hayashi65 proposed a logarithmic form of the function.
This form fits experimental data from pressure-diameter and axial force-
diameter experiments of carotid arteries at least as well as Fung’s exponential
equation, but with fewer constants. Both models outperform the polynomial
models.65 The general form of the logarithmic model is:
ro W ¼ C lnð1 auu e2u ∕2 azz e2u ∕2 auz eu ez Þ, (9.6)
where C, auu, azz, and auz characterize the mechanical properties of the tissue.
A fourth equation, which is most appropriately applied in a strictly
uniaxial case, is a power law stress–strain model.66,67 This model has the
distinct advantage of being the simplest equation for soft tissue. The model
can relate uniaxial stress–strain data reasonably well in a variety of tissue
types, including both cardiac and smooth muscle, fascia, fibrocartilage, and
cortical bone. The power law model is generally given in terms of the stress
tensor, T, the strain or strain rate tensor, S, and material constants, K and q:
T ¼ K ijkl S qijkl . (9.7)
The original implementation was for modeling the constitutive behavior of
blood and is known as the Walburn-Schneck power law.68
None of the above models, however, adequately fits experimental data
throughout the entire range of observable physiological stresses. At the lower
strain range, any of the models (linear, exponential, logarithmic, power law)
can be made to fit experimental data reasonably well, while none fit the
remainder of the strain range. Fung et al.69 proposed a model that combines
the exponential form with the polynomial form in an attempt to model the
entire range of physiological strains and beyond. Using this model, the strain
energy is expressed as
q
W ¼ CðeQ Q 1Þ þ ,
2
where
Q ¼ a1 a2 þ a2 b2 þ 2a4 ab q ¼ b1 a2 þ b2 b2 þ 2b4 ab. (9.8)

This model has been applied to canine carotid artery,69 and the model fits the
experimental data reasonably well. However, the standard deviations about
the seven constitutive constants are quite large, even for this one tissue type. It
is likely that the constants are considerably different in other arteries and
other tissue types.
Two important ideas arise from this discussion on constitutive modeling
of soft tissue. First, and implicit in Eqs. (9.5) and (9.8), is that the incremental
moduli of the tissue increase with increasing stress and, second, that at low
strain levels, any of the models, including the linear model, i.e,, Hooke’s law
[Eq. (9.1)], adequately describe the stress-strain relationships in soft tissue.
This means that as a tissue or tissue sample is subjected to small changes in
strain, the corresponding changes in stress are also small and the relationship
between these small stresses and strains can be linearized. This approximation
provides a sound theoretical basis for the subsequent development of the
remainder of this chapter.
A key challenge for optical elastography is to provide experimental values
of constitutive constants for incorporation into new constitutive models. This
includes models that predict the changes in the mechanics of tissue as a
function of age and disease and can even provide a framework for failure
prediction in healthy, diseased, and engineered tissues and organs.70
Of particular interest in understanding time dependent tissue mechanics
are the retardation (creep response to a load) and relaxation spectra.
Understanding these spectra aids in interpreting the micro-scale and mole-
cular processes that underlie the mechanical behavior, may provide insight
into pathological processes, and may also provide some guidance for the
development of synthetic materials or engineered tissues to replace damaged
or pathological tissue. Thus, there is a need for methods that directly and
noninvasively assess the time rate of the mechanical response of tissue to an
imposed load and the frequency dependent response of tissue to a dynamic
loading situation.
The retardation spectrum is essentially a description of a creep recovery
process in response to an imposed, constant load. The following argument is
given in more detail in several texts51,71 and is presented here in abbreviated
form in order to point out some of the more salient features of viscoelastic
behavior that can be evaluated optically. If a strain, ε, is imparted at some rate
to a Hookean body then the stress, s, response is given by
dε 1 ds
¼ , (9.9)
dt E dt
510 Chapter 9
where E is the Young’s modulus of the tissue. If, however, we assume that
viscous flow of the tissue occurs while it is loaded, then the strain will tend to
change at a rate that depends upon its initial, loaded value:
dε 1 ds ε
¼ þ , (9.10)
dt E dt t0
where t0 has the dimensions of time and is called the retardation time of the
material. By integrating Eq. (9.10) for ds∕dt ¼ 0, we can obtain the strain at
time t:
εðtÞ ¼ ε0 expðt∕t0 Þ, (9.11)
and t0 is thus the time required for the strain to increase (or decrease) to 1/e of
its original value. Assuming that the strain is being subtracted from its
original value, we may write:
εðtÞ ¼ ε0 ½1 expðt∕t0 Þ. (9.12)

Real tissues, however, consist of many individual structural elements, n, each
possessing a unique retardation time. Assuming that the elements are
arranged in series, we can then write from the above equations
εðtÞ X 1
¼ DðtÞ ¼ ½1 expðt∕t0n Þ, (9.13)
so En
where D(t) is the creep compliance. Equation (9.13) indicates that the
contribution of each individual element to the equilibrium compliance is 1/En.
Thus, each compliance is associated with a unique retardation time, resulting
in a spectrum of retardation times. The retardation times are often close
together and numerous in real biological tissues, so replacing the summation
sign of Eq. (9.13) with an integral, we see that
εðtÞ
so
¼ DðtÞ ¼ ∫ E1 ½1 expðt∕t Þdt .
0 0
(9.14)
Because of the long times frequently associated with some materials, it has
become convention to express D(t) as a function of ln t0 . Doing so results in
∫ Et ½1 expðt∕t Þdðln t Þ,
0
εðtÞ 0 0
¼ DðtÞ ¼ (9.15)
so
where t0 ∕E is the contribution to the creep compliance of retardation times in
the range d(ln t0 ), and the integral covers all values of t0 . The quantity t0 /E is
called the retardation spectrum.71
Implicit in the above argument is the idea that each element in the series
has its own frequency dependent behavior that results in a spectrum of
compliances for the tissue. Thus, the retardation spectrum completely defines
the behavior of a tissue and, in principle, can be determined through inversion
of the integral that defines ε(t). Through an understanding of the spectrum,
the molecular processes that determine the mechanical behavior may be
elucidated, which may be of use in the investigation of early neoplastic
changes in tissues. Many optical elastography techniques lend themselves
particularly well for investigating retardation spectra and, therefore, may be
valuable in such studies. However, this has not been investigated in any
detail.72 Of equal importance, data from different types of experiments can be
compared directly by transforming the results into a spectrum that can then be
transformed into the results of any other experiment conducted over a range
of frequencies (or temperatures).
It is apparent, then, that measures of the time rate of strain and the
frequency response of tissue are vital to understanding the mechanics of both
healthy and diseased tissue and may provide information on pathological
processes.
9.5 Laser Speckle Patterns Arising from Biological Tissues

Although this subject is covered in Chapters 6 and 8 of this volume, a
short synopsis of the main properties of speckle patterns as they relate to
optical elastography is appropriate. We assume illumination of the tissue
with a polarized, spatially coherent, narrowband laser beam. Speckle
patterns arising from biological tissues from such a beam not only exhibit
rapid decorrelation,6,73 but also tend to be depolarized relative to the
incident beam. In most tissues, generally considered to be randomly
inhomogeneous media, the randomness of the tissue structure results in
multiple scattering leading to rapid depolarization of light that has
propagated into the tissue. Some exceptions to this do exist. For example,
in certain transparent tissues, such as eye tissues, mucous membranes,
and very superficial skin layers, the degree of polarization of reflected or
transmitted light is measurable, even when the tissue layer reaches
considerable thickness.74 Nevertheless, for most elastography applica-
tions, the observed speckle patterns, either reflected or transmitted, are
randomly polarized. These characteristics of speckle patterns originating
from biological tissues, that is, rapid decorrelation and randomness of
polarization, lead to challenges in optically assessing tissue mechanics
with traditional speckle techniques, such as speckle interferometry or
speckle photography.2 Speckle tracking methods must be devised that are
insensitive to the polarization properties of the speckle patterns and that
rely upon rapid sampling of the speckle patterns. Furthermore, data-
processing algorithms that are aimed at directly inferring strain rates
rather than absolute strains75 are useful.
512 Chapter 9
The speckle phenomenon is a 3-D interference effect that exists in all

points of space where the reflected or transmitted waves from an optically
rough surface or volume intersect. Most practical applications are concerned
with the first- and second-order statistics of the speckle pattern at a plane.
First-order statistics are those that are concerned with the properties of the
speckle pattern at a single point in space, while second-order statistics, as the
name implies, are concerned with the joint statistical properties of the speckle
pattern at two or more points and gives measures of the granularity of the
speckle pattern. Speckle size is an example of a second-order statistic.
9.5.1 First-order statistics

For first-order statistics, it is the probability density function (PDF) of inten-
sity that is of primary interest.76 Laser speckle patterns originating from most
biological tissues are not “fully developed” in the sense that their brightness
distribution does not follow the negative exponential relationship as derived
by Goodman42
pðI Þ ¼ ð1∕I Þ expðI ∕I Þ, (9.16)
where pðI Þ is the probability that a speckle has brightness between I and
(I þ dI), and I is the average brightness. Equation (9.16) is plotted as curve 1
in Fig. 9.5, and it can be seen that the most probable speckle is dark. The
brightness distribution dictated by Eq. (9.16) can be produced only by the
interference of light that is polarized all in the same manner, resulting in a
similarly polarized speckle pattern.77 Thus, to produce such a distribution, the
scattering surface cannot depolarize the scattered light or a polarizer must be
Figure 9.5 Theoretical intensity probability distribution functions of a fully developed

speckle pattern (curve 1) and the incoherent combination of two speckle fields (curve 2).
used to detect a single polarization. Lightly roughened metals, or other

materials into which the light does not penetrate and is scattered only a single
time, do generally produce speckle patterns that have a brightness distribution
in accord with Eq. (9.16). On the other hand, materials into which the light
penetrates and is subject to multiple scattering, such as most biological tissues,
tend to depolarize the interfering light. These speckle patterns have a distinctly
different brightness distribution, one that is best thought of in terms of an
incoherent combination of two orthogonally polarized speckle fields.
Many speckle interferometers function by allowing two independent
speckle patterns to interfere.2,77 These speckle patterns can interfere either
coherently or incoherently. In the case of coherent combination, the statistical
properties of the resulting third speckle pattern remain fundamentally the
same as the two original patterns, typically following Eq. (9.16). However, in
the case of an incoherent combination of two speckle fields, the final
brightness distribution does not obey negative exponential statistics, but
instead follows the equation
pðI Þ ¼ 4ðI ∕Ī 2 Þ expð2I ∕Ī Þ. (9.17)

The shape of this relationship is shown as curve 2 in Fig. 9.5. The brightness
distribution of individual speckle patterns arising from most biological tissues
obeys Eq. (9.17). The reason is as follows: coherent light scattered from most
biological tissues produces randomly polarized speckle patterns and any two
orthogonally polarized components of scattered light are incoherent with one
another. Thus, single speckle patterns arising from biological tissues that
randomly polarize the speckle pattern can be considered to be the incoherent
combination of two or more speckle patterns. Another way of thinking about
this is that the shape of Eq. 9.17 arises from the convolution of two
independent negative exponential brightness distributions yielding a Rayleigh
distribution. Figure 9.6 displays the measured brightness distribution of a
backscattered speckle pattern arising from illuminating a sample of porcine
skin with an expanded singly polarized He-Ne laser beam. Superimposed
upon the measured PDF is a plot of Eq. (9.17). It is clear that the intensity
probability distribution function of the scattered light from the skin more-or-
less follows that predicted by Eq. (9.17). Now, in the case of an interferometer
that operates through the interference of two speckle patterns that
individually follow Eq. (9.17), the resultant intensity distribution will simply
be the autocorrelation of Eq. (9.17). Diagrammatically, the shape of the
resultant brightness distribution is fundamentally the same as that of curve 2
in Fig. 9.5, with perhaps a small amount of broadening due to the
autocorrelation procedure.
The practical implication of the above discussion for optical elastography
of soft tissue is that when attempting to apply classical speckle interferometry
to soft biological tissues, fringe visibility will be extremely poor, if one can
514 Chapter 9
Figure 9.6 Measured intensity probability distribution function of the speckle field
generated by illuminating a sample of porcine skin with a He-Ne laser compared to that
predicted by an incoherent combination of two fields.
produce fringes at all. Conceptually, this can be readily envisioned because

when two intensity speckle patterns that obey the first order statistics
described by Eq. (9.17) are physically overlaid, there is a high probability that
a dark area in one pattern will overlay a bright area in the other.77 Heterodyne
detection may help to overcome this issue in certain instances.
9.5.2 Second-order statistics

Speckle size, the statistical average of the distance between adjacent regions of
maximum and minimum brightness, is an important parameter in the optical
elastography techniques described subsequently in this chapter. For proces-
sing algorithms that rely upon frequency transforming the speckle data (e.g.,
the transform method, below), and for techniques that rely upon calculating
speckle statistics, such as acoustically modulated speckle imaging (amSI),78 it
is important in the optical design to consider the size of the speckle relative to
that of the detector element for a point detector or the pitch of the detector
array in the case of a multi-element detector such as a CCD camera.
The minimum speckle size in a speckle field ultimately depends upon the
aperture of the viewing system. In the case of an objective (or nonimaged)
speckle, the effective aperture is the diameter of the uniformly illuminated
area on the scattering surface, D. The smallest speckle size, dsp, at a screen
some distance z from the scattering plane, is given by2,76
d sp lo z∕D, (9.18)
where lo is the wavelength of the incident light. If, however, a lens is used to
image the speckle pattern (imaged or subjective speckle), then the spatial
distribution of the speckle in the image plane is ultimately determined by the
diffraction limit of the imaging system. Here, the mean speckle size, d 0sp , is
approximated by
lo d i
d 0sp , (9.19)
D
where di is the imaging distance, and D is the pupil diameter. Equation (9.19)
can be generalized to account for image magnification, M:
lo f
d 0sp ð1 þ MÞ , (9.20)
D
where f is the focal length of the imaging system. In certain instances, such as
in classical speckle interferometry and speckle photography, it is important
to be able to approximate the size of the speckles on the scattering surface.
In this case, the approximate speckle size referred to the object plane, d 00sp , is
given by

1 lo f
d 00sp 1þ . (9.21)
M D
Equations (9.19)–(9.21) assume that the lens actually images the scattering
surface. This is, however, not a requirement.77 It should be noted that while
the spatial distribution of the speckles differs between objective and subjective
speckle, the intensity distributions are the same for the two cases.
For speckle patterns that have a minimum size of a single pixel or larger,
the minimum speckle size in the pattern can be estimated by examining the
FWHM of the autocovariance function via the Wiener-Khintchine theorem
or, for speckle patterns that have a minimum speckle size of at least two
pixels, by examining the width of the power spectral density of the speckle
intensity.
9.6 Elastography Measurements by Tracking and Translating

Laser Speckle: The Transform Method
Based on the preceding discussion on the statistics of speckle arising from
biological tissues, and in light of the fact that speckle patterns from tissues are
largely depolarized and subject to extremely rapid decorrelation,6,12,73,79 it
becomes apparent that the more traditional methods of speckle interferometry
and speckle photography are perhaps not the best approaches to optical
elastography for medical diagnostics. In addition, the anisotropic behavior of
516 Chapter 9
biological tissues necessitates elastographic methods in which there is as little

directional ambiguity as possible.
The conventional strategy behind most one- and two-dimensional speckle
techniques that are aimed at measuring surface displacements is to compute
the correlation between a reference and displaced (sample) speckle pattern.
This approach is straightforward and has been used successfully in processing
data from speckle strain gauges.80–82 This approach, however, is subject to
errors and has certain limitations for biomedical diagnostics. For example, if
the test is subject to vibrations, there will always be some question as to the
validity of the reference exposure. Furthermore, if the speckle shifts are
substantial, the speckle pattern will decorrelate from the reference and a new
reference must be chosen. This can lead to compounding of errors.
However, since the goal of these conventional approaches is simply to
quantify a lateral shift in a “noisy” signal, a number of other data collection
and data analysis options present themselves. One data collection scheme that
has proven to be very useful for evaluating strains in biological tissues begins
with the fundamental concept of the laser speckle strain gauge as described by
Yamaguchi.80–82 The configuration is appropriate for either objective or
subjective laser speckle.
The scheme is based upon observing translating laser speckle with a linear
array CCD camera through an observation angle, uo, as the specimen is
sequentially illuminated through two equal but opposite illumination
angles, us, by a collimated laser beam (Fig. 9.7). Alternatively, a two
detector, single beam configuration is also possible.82–84 Due to the fast
decorrelation of the observed speckle patterns as a result of random
movements within the tissues (water, blood, etc.), the switching of the
illumination angle and subsequent triggering of the CCD array must be rapid
enough to acquire at least several records with minimal decorrelation between
subsequent exposures. Typically, the switching must be on the order of
50 Hz.12,73 This frequency is beyond the range of most mechanical shutters, so
electro-optical devices, such as ferroelectric crystals (FLC), in combination
with polarizing beam splitters have been used in the past.12,73 In this case, the
FLC acts as a binary, switchable half-wave plate, and the beam is thus
transmitted through, or reflected by, the polarizing beam splitter, depending
upon the polarization of the light exiting the FLC.
Using a physical optics approach, Yamaguchi80 has shown that, for an
object undergoing strain, the speckle motion observed through uo for
illumination angle us is given by

Lo cos2 us Lo cos us sin us
dxðuo ,us Þ ¼ ax þ cos uo az þ sin uo
Ls cos uo Ls cos uo

sin us cos us
Lo εxx þ tan uo Vy þ1 , (9.22)
cos uo cos uo
Figure 9.7 Optical arrangement for collecting speckle data.
where ax is an in-plane motion, az is an out-of-plane motion, εxx is the linear

strain in the plane of the detector and laser beams (ultimately, the desired
term), Vy is a rotation about the axis perpendicular to the measurement plane,
Ls is the radius of curvature of the illuminating wavefront (i.e.,, the source
distance), and Lo is the observation distance. By using the illustrated
configuration where u0 ¼ 0, by using collimated beams (Ls ! `), and by
subtracting the speckle motions as observed from the two equal but opposite
illumination angles, a relation describing the differential speckle motion, dA,
can be derived from Eq. (9.22):
dA ≡ dxð0, þus Þ dxð0, us Þ ¼ 2Lo εxx sin us . (9.23)
It can be seen, then, that the in-plane strain term can readily be isolated. Had
the complementary configuration (2 cameras, 1 laser beam) been employed,
Eq. (9.23) would have taken the form
dA ≡ dxðþuo ,0Þ dxðuo ,0Þ ¼ 2Lo εxx tan uo 2az sin uo (9.24)
and the desired strain term could only be isolated if the term containing az was
made negligible compared to the strain term. This can be accomplished
through a proper choice of Lo and uo.
518 Chapter 9
The goal of this strain measurement concept is to determine the shift in a

speckle pattern resulting from an applied load. Towards this end, the 1-D
records are stacked into what is termed a “stacked speckle history.”83,84
Stacked speckle histories are time series of 1-D views of the speckle patterns
combined in a spatio-temporal array such that the spatial dimension (camera
pixel) is along the abscissa and the temporal axis is the ordinate. In the
configuration shown in Fig. 9.3, two stacked speckle histories are generated,
one for each us. Figure 9.8 is a gray scale display of one such stacked speckle
history taken from an object undergoing a slow linear strain. The figure shows
a sequence of 200 one-dimensional speckle patterns stacked one atop another.
The desired information, that is, the shift of the speckle pattern is reflected in
the tilt of the structure.
Duncan et al.84 noted a striking similarity between the data displayed in
the form of a stacked speckle history and unprocessed synthetic aperture radar
(SAR) data.85 The analogy is particularly strong between the fast (range) axis
and the slow (Doppler) axis for the SAR data and the fast (speckle) direction
and the slow (temporal) direction of the speckle history. The standard method
of processing SAR data, therefore, suggests a simple method for processing
the data of Fig. 9.8, that is, a 2-D Fourier transform implemented with a fast-
Fourier algorithm. Figure 9.9 is an example of such an operation on the data
of Fig. 9.8. In Fig. 9.9, spatial frequency is along the horizontal axis, and
temporal frequency is along the vertical axis. The origin, DC, is in the center.
Inspection of Fig. 9.9 reveals that the desired information, that is, the time
Figure 9.8 Stacked speckle history of a sample undergoing a linear strain. Time is along
the y axis, and space is along the x axis.
Figure 9.9 Two-dimensional FFT of Fig. 9.8.

rate of speckle pattern shift, dx,_ is given by the slope, m, of the bright line
running through DC. The slopes are determined by fitting a weighted least
squares regression to the transformed data, using the signal-to-noise ratio
(SNR) of each point as the weight. Thus, by taking the time derivative of
Eq. (9.24), and rearranging to isolate the desired strain term, we get a simple
expression for directly estimating the time rate of in-plane strain, ε_ xx
(confidence intervals):
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
m2 m1 s22 þ s21
ε_ xx ¼ tðN 2,aÞ, (9.25)
2Lo sin us 2Lo sin us
where the subscripts are associated with the positive and negative illumination
angles, s is the standard deviation about each slope, respectively, and t is the
critical value of the Student’s t-distribution with N 2 degrees of freedom at a
probability level of a. Absolute strains can be determined by an integration
over the time course of the experiment. Figure 9.10 summarizes the scheme.
In practice, only a subset of records is transformed into the frequency
domain. Typically, somewhere between 10 and 30 records are analyzed. The
stacked speckle histories are visually inspected and matching sets of records
from each history that display minimal decorrelation, or for some other
experiment-driven reason, are selected and transformed. While it is possible to
perform the transform with a 2-D FFT, other spectral estimators can also be
used. For example, good results have been obtained using a fast Fourier
Figure 9.10 Summary of transform method.

520 Chapter 9
algorithm in the spatial dimension and an autoregressive (AR) spectral

estimator (modified covariance, 3–5 poles) in the temporal direction.12,73,84,86
The advantage of using parametric spectral estimators, such as autoregressive
estimators, is that they make use of a priori knowledge of the process. The
specific knowledge that is known in this case is that the focused line image will
be highly localized. That is, after transforming in the spatial direction using an
FFT, it is known that subsequent transformation in the temporal direction
would produce a narrow line spectrum at each spatial frequency. Furthermore,
it is the nature of parametric estimators to produce “spiky” spectra, making
them particularly useful for processes that display a lack of stationarity. In
biological tissues, this lack of stationarity is reflected by the rapid decorrelation
of the speckle patterns. Thus, by using a parametric estimator in the temporal
direction, good temporal resolution can be obtained without extended
observation times. Using an AR spectral estimator in the temporal direction
slightly changes the process by which the position of the focused line is
determined. Instead of using the SNR of each point as the weight, as is done
when a 2-D FFT is employed, for the AR algorithm, the weight is the value of
the image at the peak for each spatial frequency.84 Note that to avoid aliasing,
the speckle pattern should be sampled at or above the Nyquist rate, that is, dsp
2p, where p is the detector pitch.
The preceding discussion has implicitly assumed that the strain rate is
uniform over the surface of the object in question. That is, there is no scale-
size dependence of strain rate. There is no reason to believe that this will
always be the case. Any stress concentrator, such as an inclusion or a hole, will
result in a nonuniform strain field. For example, at the interface between
healthy and pathological tissue, such as subsurface skin tumors, one would
expect to observe a very nonuniform strain field. Clearly, if this approach is to
be useful in medical diagnostics, the ability to quantify nonuniform strain
fields is quite important. Under the assumption of no scale-size dependence of
strain rate, the regression model to be fit is
f t ðf s Þ ¼ mf s , (9.26)
where ft and fs are the temporal and spatial frequencies, respectively. One model
that can incorporate the possibility of a scale-size dependence on strain rate is
f t ðf s Þ ¼ mf s þ Xf 2s . (9.27)
Statistical significance of the second term dictates the appropriateness of this

model. Using this second-order polynomial model, the resulting strain estimate is
given by

m2 m1 X2 X1
ε_ xx ðf s Þ ¼ 2 þ f : (9.28)
2Lo sin us 2Lo sin us s
By taking advantage of the relationship between scale-size and spatial

frequency, we can determine the separation, s, between the scatterers on the
surface, giving rise to the nonuniform speckle shift:
s cos us
fs ¼ . (9.29)
lo Lo
This model has been applied to billets of stainless steel84 and to rectangular
samples of mandibular cortical bone,87 each with a round hole drilled
into them. Strain rate estimates were made in a direction parallel to the
direction of the load application, along a stripe immediately adjacent to the
circular holes. The results were roughly what one would expect based on
elasticity theory for the steel samples and viscoelasticity theory for the bone
samples. This is an area that requires much further development before
optical elastography implemented in this way will become a meaningful
diagnostic tool.
9.6.1 Potential error sources

Both numerical and physical limitations are imposed upon the transform
method. A potential source of numerical error in the algorithm is that it is a
differential technique and the strain rate estimates are arrived at from
subtracting two numbers that may be of the same order of magnitude. The
resulting rounding errors may limit the precision of the approach. In addition,
the rigid body terms of Eq. (9.22) must be kept on the same order as the
desired strain term.86 The magnitude of the rigid body terms can be
investigated by adding the slopes of the data in the frequency domain, rather
than subtracting them. This yields

Lo cos us _ y ðcos us þ 1Þ,
_
dxð0,u _
s Þ þ dxð0, us Þ ¼ 2a_ x þ 1 þ 2Lo V (9.30)
Ls
where the first term on the right side of the equation is a result of in-plane
translation, and the second term is a result of rotation about the axis
perpendicular to the measurement plane. The total speckle motions as
observed for each illumination angle are then

m1 m2 m1 þ m2
dxð0,us Þ ¼ þ T,
2 2

m1 m2 m1 þ m2
dxð0, us Þ ¼ T, (9.31)
2 2
522 Chapter 9
where T is the total measurement time. The total speckle motion associated
with in-plane strain, dxε, is then
dxð0,us Þ dxð0, us Þ

dxε ¼ , (9.32)
2
and that with rigid body motion, dxrb, is
dxð0,us Þ þ dxð0, us Þ

dxrb ¼ : (9.33)
2
It has been shown73 that with proper care in the experimental design, dxε and
dxrb can be kept within 10% of each other, and thus the numerical error
associated with the differencing technique can be minimized.
As with any metrological method, a variety of physical factors can each be a
source of error. Such sources include: Temperature transients in the laser causing
laser wavelength drift, temperature transients in the camera structure causing
apparent speckle shifts, barometric pressure fluctuations, changes in the
refractive index through which the scattered light propagates, and deviations
from symmetry in the experimental configuration. Most of these potential noise
sources can be mitigated by allowing adequate warm-up time for the
instrumentation, using a temperature-stabilized laser, and through care in the
experimental set-up. Slight deviations from symmetry have been shown to be
benign,86 but care must be taken to limit these deviations. Special consideration
has been paid to errors associated with changes in refractive index, n, resulting in
apparent speckle shifts. Two separate cases have been examined in detail:
changes in n resulting from atmospheric barometric pressure fluctuations,86 and
changes in n resulting from temperature changes in a water bath in which sample
human arteries under test were submerged.12 Slightly different approaches were
used in each case, but the conclusions were the same: the differential speckle
motions caused by strain and by fluctuations in n were comparable when
dn
εxx . (9.34)
n
Since the typical strains measured in these experiments are on the order of a
few to tens of microstrain, the maximum allowable change in n over the
course of the experiment is on the order of a few ppm or less.
9.6.2 Applications of laser speckle elastography to hard

and soft tissues
The transform method of processing speckle strain measurements has been
applied successfully to both hard and soft tissues, as well as to synthetic
biomaterials. None of the studies, however, have been used for diagnostic
purposes and have been mostly of a proof-of-concept nature.
From a diagnostic standpoint, understanding the mechanics of bone is

quite important in the study of bone degenerative diseases, such as
osteoporosis. Furthermore, knowledge of the true, physiological performance
of bone is important in the development of traditional and biomimetic
materials designed for the replacement or augmentation of diseased or
damaged bone tissue. Traditionally, the most common method for arriving at
stress–strain relationships in cortical bone was to rely on contact strain gauge
data. While this approach has led to much of what is known about bone
mechanics, it is not at all appropriate for diagnostic purposes. Furthermore,
these strain gauges are attached with a cyanoacrylate adhesive that must
penetrate the porous surface of the bone to some depth. This bonding
procedure may distort the apparent structure-property relationships, particu-
larly when investigating the micromechanics of the tissue.
Various speckle interferometry,88 speckle photography,89 and holographic
interferometry90 methods have been used to evaluate displacements in bone
tissue.91 Most of these studies have required that the bone either be dry (or
mostly dry) or coated with a reflective surface, such as talc or silver.89
Kirkpatrick and Brooks73 and Kirkpatrick and Duncan42 have applied the
transform method to bone mechanics using fully hydrated bone samples with
no reflective coatings. In the first of these studies, samples of cortical bone
were machined under constant irrigation from the mid-diaphysis of porcine
femurs. The bone samples were mounted into a tensile testing machine,92
preloaded to about 20 N, and dynamically loaded over a 1.5-N range. The
speckle measurements were made on the fly and resulted in stacked speckle
histories and transforms such as that shown in Figs. 9.11(a)–(d). The loading
waveform for this same test is shown in the next figure (Fig. 9.12). It is clear
that the preferential tilt of the speckle history follows the direction of the
loading waveform; both the loading waveform curve and the tilt of the speckle
history reverse directions at the same record number. Subsequent analysis of
the speckle histories using the FFT/AR algorithm yielded estimated strain
rates ranging between 4.61 and 23.84 mε/s, depending upon the sample. Using
the data from the load cell on the tensile testing instrument, the mean Young’s
modulus of the samples was calculated to be 9 GPa, which, considering the
low loads and extremely slow strain rates, is a reasonable estimate.71 Total
strains [calculated by the integration of Eq. (9.25)] for the slowest strain rate
were calculated to be on the order of only 3.7 mε. The total speckle movement
as seen by the camera was approximated as
Movement ≅ Lo ε_ t sin us , (9.35)
where t is the time over which the data were analyzed (,1 s). Again, for the
slowest strain rate recorded, the total speckle movement was approximately
1.15 mm. The camera pixel pitch was 10.5 mm, yielding a total movement in
terms of pixels at 0.11 pixels. While these numbers by no means indicate the
524 Chapter 9
Figure 9.11 (a and b) Stacked speckle histories from a cortical bone sample undergoing a
sawtooth waveform loading; (c and d) FFT-AR transform of data records #160–180 of (a)
and (b).
resolution of the algorithm, they do provide a feel for the high sensitivity of
the approach.
Using a similar approach, Kirkpatrick and Duncan42 simultaneously
recorded strains in cortical bone samples with contact strain gauges,
extensometers, and with the transform method. Even with the strain ranges
being at the very lower limit of the sensitivity of contact strain gauges, the
results generated by the methods were in good agreement with each other,
with the correlation between the extensometer strain data and the speckle
strain data being within 4% of each other. Comparisons between the
transform method and other mechanical methods have also been made using
copper wire,93 stainless steel billets,84 and stainless steel wires.92 All of these
studies reflect the accuracy of the transform method.
Experimental measures of vascular mechanics, specifically human tibial
arteries and rat vena cava, have also been performed using the transform
method.12 Strain rates and total strains on the order of tens of microstrain
were recorded from the tissues that were both superfused and perfused to
Figure 9.12 Loading waveform for bone experiment.
physiological pressure with a glucose-infused physiological saline solution.

This strain resolution is at least an order of magnitude higher than other soft-
tissue strain measurement methods, such as video dimensional analysis
(VDA). Of particular interest in this study was a demonstration of how the
same basic optical arrangement and data processing algorithm could be used
to evaluate pressure–diameter relations in vessels. Changes in the dilational
properties of vascular tissue as a function of intraluminal pressure, disease,
and drug application are of great interest to vascular biologists and surgeons.
Traditionally, such measures were made using VDA techniques. However,
by using the optical arrangement shown in Fig. 9.13, and observing the
speckle motion in the geometrically specular direction (or in the exact
forward direction), dilational data about the vessels can be acquired. These
arrangements make use of the fact that the bidirectional reflectance distri-
bution function (BRDF) of a rough surface is centered around these two
directions, resulting in a higher intensity of light scattered at these angles.
Using the formalisms of Eqs. (9.22)–(9.25), it can readily be shown that
the differential speckle motion observed using the arrangement of Fig. 9.13 is
a function of out-of-plane (az) motion only. The differential speckle motion is
given by

L0
dA ≡ dxðus ,us Þ dxðus , us Þ ¼ 2az sin us 1 . (9.36)
Ls
526 Chapter 9
Figure 9.13 Optical arrangement for dilation measurements on vascular tissue.
Taking the time derivative, as was done for the in-plane strain derivations,
rearranging, and assuming uniform dilational behavior, the total out-of-plane
motion (dilation) rate is given by
_
dxðu _ s , us Þ
s ,us Þ dxðu
a_ ðz,zÞ ¼ . (9.37)
sin us ð1 LLos Þ
The factor of 2 arises because the dilation is in both the þz and z directions.
Integration of Eq. (9.37) over the time for which the data were analyzed will
yield the total dilational movement along the z axis.
9.7 Alternative Processing Algorithms for Calculating

Speckle Shift
There are numerous alternative approaches that can be employed for
determining speckle movement. Generally, these approaches can be catego-
rized as either nonparametric or parametric. The fundamental difference
between the two types of approaches is that nonparametric estimators make
no a priori assumptions regarding the nature of the speckle shift. An example
of a nonparametric scheme is the correlation approach.94–96 In this approach,
the cross-correlation of sequential records is calculated, and the shift of the
correlation peak determines the amount of speckle pattern shift. Examples of
parametric approaches include a minimum mean square error estimator, a
maximum likelihood estimator, and a maximum entropy estimator. Each of
these processing schemes is discussed in the following sections.
9.7.1 Non parametric speckle shift estimators

The traditional approach to processing speckle data makes use of the
crosscorrelation between successive records. Regardless of the dimensionality
of the data the concept is the same; we, therefore, illustrate with data in one
dimension. Two sequential speckle records are modeled as
gðxÞ, gðx dxÞ, (9.38)
where one record is shifted with respect to the other by the amount dx. The
cross-correlation between these two records is given by
`
RðtÞ ¼ ∫ dxgðx tÞgðx dxÞ: (9.39)

`
It is straightforward to show that this cross-correlation reaches its maximum

when t ¼ dx.97 This is the basic concept: we seek the lag (t) for which the
correlation is maximum. It is usually implemented with the fast Fourier
transform (FFT)98 through use of the (auto) correlation theorem,
RðtÞ ¼ F 1 fF ½gðx tÞF ½g ðx dxÞg, (9.40)
where the forward (inverse) Fourier transform is denoted by F(F1).

One of the fundamental assumptions in this (and other) approaches is that
the two speckle records are simple translates of one another. With analogy to
one of the basic tools of the theory of propagation through atmospheric
turbulence, Taylor’s frozen turbulence hypothesis,99 we might call this the
“frozen speckle” model.
9.7.2 Parametric speckle shift estimators

The previously discussed speckle motion estimation techniques are termed
nonparametric because they do not rely on any a priori knowledge of the
experiment. We contrast this class with the class of approaches that do
incorporate a priori knowledge. The objective, of course, in using such
model-based estimators is to gain a degree of sensitivity. By making use of
information about the conditions of the measurement, we can eliminate
possible outcomes that are nonphysical. While this often results in greater
resolution, this increased resolution sometimes comes at a price. Typically,
nonparametric approaches are more robust. Because the parametric
approaches implicitly eliminate estimates that are “non physical,” when
data are encountered that are less than ideal, these estimators sometimes
break down.
In the material that follows, we explore the relative merits of a couple of
specific parametric approaches to estimation of speckle pattern shift or
movement. The first is a minimum mean square error (MMSE) technique
that, for small amounts of speckle motion, requires no search for an optimum
solution. Additionally, when the noise component in the data is Gaussian, it is
shown that this approach leads to the maximum likelihood estimate of the
528 Chapter 9
g ( x)
xo
g ( x − βt )
xo
Figure 9.14 Illustration of shifting frozen speckle pattern.
speckle pattern shift. Another approach makes use of maximum entropy

concepts. This method, like the maximum likelihood technique, is shown to
have the ability to estimate subpixel motion with temporal resolution that is
much finer than the nonparametric approaches.
9.7.2.1 A minimum mean square error estimator
We begin our discussion of parametric approaches by adopting a “frozen
speckle” model. We assume that, over a time on the order of several sequential
exposures of the camera, the structure of the speckle pattern is fixed. The only
change with time is its lateral motion (Fig. 9.14). Thus, the speckle motion can
be modeled as
gjþ1 ðx1 Þ ¼ gj ðxi dxÞ, (9.41)
where the subscript i denotes the pixel (spatial dimension), and the subscript j
represents the record (temporal dimension). We assume that the shift, dx, is
small compared to a pixel so that Eq. (9.41) can be approximated as
gjþ1 ðxi Þ gj ðxi Þ þ dxg0j ðxi Þ: (9.42)
This is simply the first two terms of the Taylor series expansion for g. To
introduce a degree of symmetry into the problem, we inspect the two speckle
records on either side of the record of interest (Fig. 9.15),
½gjþ1 ðxi Þ, gj1 ðxi Þ: (9.43)
We then determine the dx that minimizes the error,
X
N
ε2j ¼ ½gjþ1 ðxi þ dxÞ gj1 ðxi dxÞ2 , (9.44)
i¼1
space, x = ip
record j − 1
record j
record j + 1
time, t = j Δt
Figure 9.15 Central differencing technique.
where the summation is over all pixels in the array. We are thus seeking
the dx that brings these two records into registration. Equation (9.44) may
be solved numerically by making use of a gradient search algorithm.100 If,
however, we make use of the approximation in Eq. (9.42) (small speckle
motions), then differentiation with respect to dx and rearranging yields the
following equation:
PN 0
j¼1 ½gjþ1 ðxi Þ gj1 ðxi Þ½gjþ1 ðxi Þ þ g0j1 ðxi Þ
dxj ¼ PN 0 0
: (9.45)
j¼1 ½gjþ1 ðxi Þ þ gj1 ðxi Þ
2
The term in the first square bracket in the numerator is simply the first central
difference approximation101 to the derivative
∂gj ðxi Þ gjþ1 ðxi Þ gj1 ðxi Þ

: (9.46)
∂tj 2
The spatial derivatives may be approximated similarly:
∂gjþ1 gjþ1 ðxiþ1 Þ gjþ1 ðxi1 Þ

,
∂xi 2
∂gj1 gj1 ðxiþ1 Þ gj1 ðxi1 Þ
: (9.47)
∂xi 2
Note that the shift parameter, dx, is the time rate at which the speckle pattern
shifts; units are pixels/record.
Although the means by which we arrived at our estimate for dx was quite
specific, this estimation approach is more general than it would seem. For
instance, the term approximating the first central difference for the estimate of
the temporal derivative arose because we chose to inspect speckle records on
either side of the record of interest. We could just as easily have included
additional records, weighting their contributions appropriately to estimate
530 Chapter 9
higher-order approximations to the derivative. For example, instead of using

the weights
1
½1,0, þ1, (9.48)
2
we could use102
1
½1, 8,0,8, 1: (9.49)
12
In this case, the formulation for the mean square error would be
X
N
ε2j ¼ ½gjþ2 ðxi þ 2dxÞ 8gjþ1 ðxi þ dxÞ þ 8gj1 ðxi dxÞ gj2 ðxi 2dxÞ2 ,
i¼1
(9.50)
the temporal derivative term of Eq. (9.46) would be
∂gj ðxi Þ
½gj2 ðxi Þ þ 8gj1 ðxi Þ 8gjþ1 ðxi Þ þ gjþ2 ðxi Þ 12 , (9.51)
∂tj
and the term involving the spatial derivatives would be
½2g0j2 ðxi Þ þ 8g0j1 ðxi Þ þ 8g0jþ1 ðxi Þ 2g0jþ2 ðxi Þ. (9.52)
Note that this is simply the weighted average

0 0 0 0
4 gj1 þ gjþ1 1 gj2 þ gjþ2
12 (9.53)
3 2 3 2
of the terms on either side of the record of interest. Further, these higher-order
operators for the derivative can be used to estimate the spatial derivatives:
∂gj ðxi Þ 1
½gj ðxi2 Þ þ 8gj ðxi1 Þ 8gj ðxiþ1 Þ þ gj ðxiþ2 Þ: (9.54)
∂xi 12
These higher-order approximations to the derivative have somewhat
better noise characteristics.101 This improvement, however, comes at the
expense of reduced temporal resolution. Nevertheless, by making use of these
higher-order approximations, the processing can be tailored to the demands of
the experiment.
Up to this point, we have made no specific assumptions about the
statistics of the speckle pattern or associated noise. The only a priori
knowledge that we have introduced is that the speckle shift is small with
respect to the pixel size. If we now make the assumption that the measured
speckle signal comprises a deterministic speckle signal plus noise, the

measured signal can be modeled as
d j ðxi Þ ¼ gj ðxi Þ þ nj ðxi Þ, (9.55)
where n is a zero-mean noise term. If we calculate the central difference about
the j-th record, we get
d jþ1 ðxi þ dxÞ d j1 ðxi dxÞ ¼ gjþ1 ðxi þ dxÞ gj1 ðxi dxÞ
þ njþ1 ðxi þ dxÞ nj1 ðxi dxÞ. (9.56)
We assume that the sequential speckle patterns have a constant mean.

Further, we assume that the noise is described by a statistically independent
Gaussian variable with zero mean and constant variance. As a result, the
noise probability density function can be written as

1 XN
pðd j jgj Þ ¼ pðnj Þ ¼ C exp 2 ½gjþ1 ðxi þ dxÞ gj1 ðxi dxÞ ,
2
2s i¼1
(9.57)
where C is a constant. Equation (9.57) is commonly referred to as the
likelihood function.103 To choose the dx that maximizes this likelihood, we set
∂
ln pðd j jgj Þ ¼ 0: (9.58)
∂ðdxÞ
Carrying out this calculation leads to the equation in Eq. (9.45). It is
straightforward to show that this is an unbiased estimate of the speckle
pattern shift and that the variance of the estimate attains the Cramér-Rao
lower bound.104
Finally, we note that the MMSE solution [Eq. (9.44)] can be made more
efficient using the shift theorem and a gradient search algorithm. In this case,
we need not restrict our interest to small speckle motions.
9.8 Expanding to Higher Dimensions

Expanding the MMSE solution to higher dimensions for small speckle motion
estimation is a straightforward n-dimensional implementation of the one-
dimensional maximum-likelihood estimator that we described above. Here,
we demonstrate this expansion in two dimensions. Such an approach allows
the formation of images that encode the local deformation of the tissue. In this
specific example, the problem is to assess the motion, from frame to frame of
the features in a sequence of xz OCT images, where the x direction is a
lateral dimension along the surface, and the z direction is the depth dimension.
We denote the intensity or gray level variations in this k-th image as g(x, z, k),
532 Chapter 9
and wish to determine the time progression of the lateral shift of features in
this image. To accomplish this, we inspect this shift based on the pair of
images prior to and after the k-th image. For this image, we define a mean
square difference between a portion of the pair of adjacent images:
XX
kεk k2 ¼ ½gðxi þ f x ,zj þ f z ,k þ 1Þ gðxi f x ,zj f z ,k 1Þ2 . (9.59)
i j
The summation is over a small neighborhood, say, of dimension 3 3 pixels,

of the images.
The objective is to determine the local velocity, f̄ ¼ ðf x , f z Þ, that will bring
the features in this neighborhood pair into registration. Note that the units of
velocity are implicit, i.e., pixels/record. Obviously, the shift estimates are
specified for the neighborhood chosen by the summation in Eq. (9.59). As
detailed previously, expanding each image in terms of a 2-D Taylor series
results in the equation
XX
kεk k2 ¼ ½f x gx ðxi ,zj ,kÞ þ f z gz ðxi ,zj ,kÞ þ gt ðxi ,zj ,kÞ2 , (9.60)
i j
where
1 ∂
gx ¼ ½gðxi ,zj ,k 1Þ þ gðxi ,zj ,k þ 1Þ,
2 ∂x
1 ∂
gz ¼ ½gðxi ,zj ,k 1Þ þ gðxi ,zj ,k þ 1Þ,
2 ∂z
1
gt ¼ ½gðxi ,zj ,k þ 1Þ gðxi ,zj ,k 1Þ: (9.61)
2
In the above, gx and gz are, respectively, the average x and z gradients at the
k-th record, and gt is the central difference approximation to the temporal
gradient. Minimization of this error, by taking the partial derivatives with
respect to the velocity components, results in a pair of simultaneous
equations:

hgx gx i hgx gz i f x
¼ hgx gt i , (9.62)
hgz gx i hgz gz i f z hgz gt i
where 〈⋯〉 represents the local averaging operation. These equations are easily
inverted to yield estimates of the velocity, f̄ . Note that each element in
Eq. (9.62) is a full-size N M array. Solution of these linear simultaneous
equations produces velocity estimates for each pixel of each plane of the image
cube.
More formally, Eq. (9.61) is expressed as
kεk ðx̄Þk2 ¼∫ ∫ wðx̄ x̄0Þ½ð∇gÞT f̄ þ gt 2d x̄0 , (9.63)

`
where the window function, w, denotes the local averaging operation. In its
simplest form, this window function is a simple boxcar that evenly weights the
residuals within the region of interest.
An alternative derivation of this equation is based on the concept of
optical flow. This concept relies on the assumption that the brightness,
g(x, z, t), is conserved (only its spatial distribution changes with time) so that
the total time derivative, dg/dt, is zero, i.e.,
ð∇gÞT f̄ þ gt ¼ 0, (9.64)
where f ¼ [ fx fz]T is the optical flow. Note that this equation (commonly
known as the brightness change constraint equation, or BCCE) characterizes
a single point on the image. We presume that the image gradient, ∇g, and
the partial derivative with respect to time, gt, are known, i.e., measured.
We have a single equation [Eq. (9.64) represents a single constraint] and
two unknowns, the velocities fx and fz. This is referred to as the aperture
problem of motion estimation. Minimizing the residual within the region
specified by the function, w (the aperture), leads to the previous result,
Eq. (9.63).
Once the velocities (fx, fz) are estimated, they can be integrated over time
and mapped on a pixel-by-pixel basis to create an image whose gray values
encode displacements in either the x or z direction, or the vector sum via the
Pythagorean theorem. Normalizing these displacement maps by the initial
dimensions of the sample yields elastograms that encode strain in the limit of a
small deformation (i.e., infinitesimal strain).
Visualization of regions that display a different strain response is
enhanced through the use of a neighborhood operator in the form of a
discrete convolution filter. Pixel operations as described above are not
particularly suitable for discriminating regions of interest in images because
the gray value of each pixel is determined with no consideration of the
gray values of the neighboring pixels. In contrast, neighborhood opera-
tions, such as convolution filters, analyze the spatial relations of the gray
values. Thus, in effect, this operator converts the gray scale elastograms
into feature-based elastograms. The convolution kernel (mask) may be
adaptively generated directly from the gray values of a small background
region of the strain-encoded elastogram. The elastogram can then be
convolved with this kernel in two dimensions to emphasize the local
magnitude of the strains.
534 Chapter 9
9.9 What is Really Measured in Laser Speckle-Tracking

Elastography?
By tracking the shifts in the backscattered speckle patterns, speckle-tracking
optical elastography (OE) in general provides a Lagrangian estimation of the
tissue deformation. However, depending upon the form of the processing
algorithm used, OE can provide what can be called an ‘updated Lagrangian
estimate,’ which may be more appropriate for soft tissue deformations that
exceed the limits of small deformation approximations. The difference
between the two approaches is that whereas a traditional Lagrangian
approach describes the deformation of a body in reference to an initial
unloaded state, an updated Lagrangian approach continuously updates the
“reference” state.72 A Lagrangian, or material, description of deformation
employs an ordered pair of numerical values of spatial coordinates for each
particle in the body in its reference state and the spatial coordinates of that
same particle in the deformed state are “referred” back to the reference
coordinates in order to estimate the displacement of that particle. This is
opposed to an Eulerian, or spatial, description in which one is interested in a
physical quantity at a particular point in the spatial domain. In this case, the
time history of the physical quantity at a given location need not match that of
a particular particle.
9.9.1 Lagrangian description of motion of particles in object space

Consider a block of tissue subjected to a deforming force in compression.
Since soft tissues can, in general, be considered to be incompressible, then no
volumetric change occurs during the deformation, thus,
∇ •~
u ¼ 0, (9.65)
where ~u is the displacement vector. Equation (9.65) is a formal descrip-
tion of the divergence of the displacement vector and describes the
volumetric change in a material under loading. Further, assuming that
the out-of-plane deformation ! 0, then we can make the plane strain
approximations, thereby reducing the 3-D deformation state of Eq. (9.65)
to a 2-D problem:
∂u ∂v
þ ¼ 0, (9.66)
∂x ∂y
where u is the component of ~ u in the x direction and v is the component of ~
u
in the y direction. Rewriting in terms of the normal components of the Cauchy
infinitesimal strain tensor, we get
∂u ∂v
εxx þ εyy ¼ 0, where εxx ¼ , and εyy ¼ : (9.67)
∂x ∂y
Integration of Eq. (9.67) can provide an estimate of the displacement at any

(x,y) position:
x
uðx,yo Þ ¼ ∫½εxx ðx0 ,yo Þdx0 þ uðxo ,yo Þ: (9.68)

x0
Taking advantage of the incompressibility of the tissue [Eqs. (9.65)–(9.68)], we

can write
x
uðx,yo Þ ¼ ∫½εyy ðx0 ,yo Þdx0 þ uðxo ,yo Þ: (9.69)

x0
Equations (9.68) and (9.69), then, describe in a Lagrangian fashion the

motion of particles in the object (tissue) space as the tissue is deformed. The
issue now is to relate tissue particle motions u(x, yo) to the observed speckle
motion. The velocities, fx, fy, measured in the image space [Eq. (9.62)] are
directly related to the time derivatives of the displacements in the object
(tissue) space, u0 (x, yo), by

0 f ðx,yo Þ
u ðx,yo Þ ¼ M, (9.70)
2Lo sin u
where M is the imaging system magnification, u is the angle of illumination

relative to the tissue normal, and Lo is a factor that accounts for any image
misfocus. With the appropriate variable substitutions, the identical argument
can be made for the v(xo, y) component of ~ u.
Once the velocities (fx, fy) are estimated, they can be integrated over time
and mapped on a pixel-by-pixel basis to create an image whose gray values
encode the Lagrangian displacements, u or v, in either the x or y direction,
respectively, or the vector sum via the Pythagorean theorem. Normalizing
these displacement maps by some original reference dimension of the sample
yields elastograms that encode the normal strain components [as defined in
Eq. (9.67)], εxx or εyy in the limit of small deformations (i.e., infinitesimal
strains).
Each image in the processed image sequence then can encode in its gray
values either the Lagrangian displacements or the Cauchy infinitesimal strain
components, depending upon the normalization. If we denote the intensity or
gray level variations in the reference image as gL(x, y, jr), then the Lagrangian
deformation image can be defined as
I u,L ðx,yÞ ¼ gL ðx,y, j r þ nÞ gL ðx,y, j r Þ: (9.71)
In terms of an updated Lagrangian descriptor, Eq. (9.71) may be rewritten as
I u,L ðx,yÞ ¼ gL ðx,y, j r þ nÞ gL ½x,y, j r þ ðn 1Þ, (9.72)
536 Chapter 9
where the reference image is constantly updated to the image prior to the
image of interest. Equation (9.71) yields images (elastograms) of cumula-
tive displacement (over n images), whereas Eq. (9.72) provides images of
instantaneous displacements between subsequent images. Normalization of
these images by the initial or updated reference specimen dimensions yields
the desired components of the Cauchy infinitesimal strain tensor. Because
many of the cumulative tissue deformations of interest are large relative to the
restrictions of the small deformation approach, it is possible to implement
the updated Lagrangian model in the analysis. Using this approach, one can
describe the tissues using a hyperelastic material description. That is, the
stress-strain relationships in the tissues can be described via strain-energy
density functions (SEDFs). The exact forms of these functions will depend
upon the nature of the data. In real tissues, the stress–strain relations will
change as a function of time, as well as the mechanical, biochemical, and
structural history of the tissue, so the SED functional form may also change
with these parameters.
9.9.2 Relationship between elastograms and SEDFs

When developing constitutive models of tissue deformation, it is common
to employ the Green-Lagrange strain tensor, Eij, instead of the Cauchy
infinitesimal strain tensor. Two key reasons for this are that the Green-
Lagrange strain tensor is symmetric (i.e., Eij ¼ Eji) and conjugate with the
second Piola-Kirchhoff stress tensor, Sii ¼ (1/Li)Tii. Here, Li is the stretch
ratio, ∂xi/∂ai, where (a1, a2, a3) are the coordinates of the tissue particle in an
undeformed state, (x1, x2, x3) are the stressed-state coordinates, and Tii is the
Green’s (Lagrangian) stress tensor. Note that for finite strains in an
incompressible material, L1L2L3 ¼ 1. The Green’s stress tensor (conjugate
with the Lagrange strain tensor), in general, is not symmetric. Thus, we need
to describe the relationship between the Cauchy infinitesimal strain tensor
(which we measure via OE) and the Green-Lagrange strain tensor, which we
need in order to derive the conjugate stress tensor, Sij, through a strain energy
function. Formally, we can define the Green-Lagrange strain tensor by first
defining the square of the distance a tissue particle moved from a reference
state to a deformed state:
ds2 ds2o ¼ 2E ij dai daj , (9.73)
where

1 ∂xk ∂xk 1 ∂ui ∂uj ∂uk ∂uk
E ij ¼ dlk ¼ þ þ , (9.74)
2 ∂ai ∂aj 2 ∂aj ∂ai ∂ai ∂ai
dlk is the Kronecker delta, and the subscripts are dummy variables. Note that
if the derivative ∂uk/∂ai is infinitesimally small, its square can be neglected and
the Green’s strain tensor [Eq. (9.73)] reduces to

1 ∂ui ∂uj
E ij ¼ þ ¼ εij , (9.75)
2 ∂aj ∂ai
which is simply a restatement of Cauchy’s infinitesimal strain tensor. In

practical implementation, the two strains are related through the stretch ratio:
Dl 1 ðl 2 l 2o Þ
ε¼ ðCauchy infinitesimalÞ E ¼ ðL2 1Þ ¼ ðGreen’sÞ:
lo 2 2l 2o
(9.76)
Now, in order to arrive at a constitutive equation to quantify the relationship
between the imposed stresses and the resultant strains (measured with OE), we
define a strain energy function in terms of the 2nd Piola-Kirchhoff stress and
the Green-Lagrange strain as
∂ðro W Þ
S ij ¼ , (9.77)
∂E ij
where ro is the tissue density, and W is the strain energy function (strain
potential or work). From this definition, it is seen that because the strain
energy function (SEF) is a function of strain, then the stresses can be obtained
by differentiation with respect to strain. It is difficult to justify the existence of
an SEF in biological tissues based solely upon thermodynamic considerations,
as this requires that stress be purely a function of strain and variables such as
strain rate do not influence the derived stress values. In order to justify
applying the SEF approach to biological tissues, we can take the usual
approach and assume that, after preconditioning, strain rate has minimal
influence on the SEF.
The issue then becomes what form does the strain energy density function
take? Historically, there have been essentially two schools of thought on the
SEDF form: either a polynomial form or an exponential form. Perhaps the
most commonly used 2D SEDF is a combination of the two, so that the strain
energy function may take the form (in the absence of shear, and thus no
E3 terms)
C
ro W ¼ exp½a1 E 211 þ a2 E 222 þ 2a4 E 11 E 22 : (9.78)
2
This form has been applied successfully to a variety of tissue types including
vascular, heart, skin, and connective. Thus, it serves as a reasonable starting
538 Chapter 9
place. It can be seen, then, that in elastography, we can progress from the
collection of speckle images to the generation of elastograms and the
formulation of SEDFs in a relatively straightforward manner.
9.10 In vivo Laser-Speckle-Tracking Optical Elastography

Outside of OCE experiments, only a very few in vivo optical elastography
measurements have been published in the literature using the speckle tracking
approach. Following the approach described above, Kirkpatrick et al.15
produced elastograms that displayed the relative stiffness of an in vivo murine
melanoma lesion and two types of in vivo human melanocytic nevi. They used
a low-frequency acoustic source as the driving stimulus that produced a
surface acoustic wave in the skin. The acoustic pressure served as the
deforming force, and the motion of the speckle patterns arising from light
scattered from the skin surfaces was tracked and mapped in two dimensions.
9.11 Performance Comparisons

In evaluating the relative merits of the various estimators, there are several
criteria. One, of course, is the sensitivity, i.e., the ability to estimate very small
speckle motions. On the other hand, the technique should be capable of
dealing with very large speckle motions as well. Another evaluation criterion
is robustness. This is the ability of the algorithm to cope with other than ideal
data. Dynamic range, that is, the ability to reliably estimate both large and
small motions, is also a critical consideration for these estimators.
Our first example considers relatively large speckle motions. In this case, a
simple nonparametric approach such as correlation works quite well. An
example speckle history for large motion is shown in Fig. 9.16. The speckle
motions predicted using a correlation approach are shown in Fig. 9.17. For
this calculation, a central differencing technique was used; records to either
side of the record of interest were cross-correlated. Of course, the speckle shift
per record is half the total, so that the minimum speckle shift is one half pixel.
Figure 9.16 Speckle history displaying large speckle shifts.

1
Shift(pixels/record)
-1
-2
-3
-4
-5
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.17 Performance of correlation estimator for large speckle shifts.
2.5
1.5
1
0.5
-0.5
-1
-1.5
-2
-2.5
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.18 Performance of closed-form max likelihood estimator for large speckle shifts.
Speckle shifts estimated using the maximum likelihood estimator (assuming a

small speckle shift, which is not the case) are illustrated in Fig. 9.18.
This result shows some differences with those derived using correlation
processing. On the other hand, the maximum likelihood estimate for arbitrary
speckle shift (using a gradient search algorithm to determine the shift),
shown in Fig. 9.19, is in excellent agreement with the correlation results.
540 Chapter 9
1
-1
-2
-3
-4
-5
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.19 Gradient search for max likelihood solution for large speckle shifts. Compare
with the results shown in Fig. 9.17.
Figure 9.20 Example of speckle history for small shifts.
The correlation between these two results is in excess of 0.99. Obviously, the
small shift requirement is violated by these data. These results simply show
that estimating speckle shifts for large motions is a relatively straightforward
problem.
Now, we consider an example displaying small speckle motions
(Fig. 9.20). Correlation and maximum likelihood processing produce the
results shown, respectively, in Figs. 9.21 and 9.22. Clearly, the speckle motion
in these data is below the resolution of the correlation processing algorithm (it
cannot perceive motions of less than a half pixel), whereas the maximum
likelihood estimator performs quite well.
0.8
0.6
0.4
0.2
-0.2
-0.4
-0.6
-0.8
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.21 Correlation estimator results for small speckle motions.
0.8
0.6
0.4
0.2
-0.2
-0.4
-0.6
-0.8
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.22 Max likelihood estimate of small speckle motions.
9.12 Generalizations
In the discussion so far, it has been assumed implicitly that we are dealing with
objective (nonimaged) speckle. In this case, a particular portion of the speckle
record cannot be identified with a particular spot on the object. Most of
these techniques, however, are appropriate for subjective speckle as well. The
bone example previously discussed employed objective speckle. However, the
542 Chapter 9
0.8
0.6
0.4
0.2
-0.2
-0.4
-0.6
-0.8 first 128 columns

last 128 columns
-1
0 20 40 60 80 100 120 140 160 180 200
Record
Figure 9.23 Speckle motion estimates based on first and last 128 columns.
experiment was carefully contrived so that the specimen (a cortical bone sample)
was uniformly illuminated and uniformly strained. Success of this effort is
confirmed by the plots of speckle motion shown in Fig. 9.23. These two traces
represent estimates of speckle motion made with the max likelihood estimator
using the first and the last 128 columns of the speckle history. Although there
are some minor differences, which we attribute to the fact that these are two
distinct speckle realizations, the speckle motions are very similar (r ¼ 0.97).
With this concept in mind, it is easy to see that these algorithms can be used for
a full 2-D sequence of images by separately processing a 2-D speckle history for
each of the image dimensions, as in Fig. 9.24.
We have discussed some illustrative parametric processing algorithms.
Using the basic approach, one could imagine a variety of others. For example,
the conventional processing approach searches for the lag for which the cross-
correlation is a maximum. Rather than using the FFT, one might search
directly for the shift that maximizes the cross-correlation. Specifically, in the
notation of this chapter, we could seek the speckle pattern shift (in one
dimension), dx, that maximizes the cross-correlation coefficient,
X
N
p2j ∝ gj1 ðxi dxÞgjþ1 ðxi þ dxÞ: (9.79)
i¼1
Subject to the appropriate restrictions on the magnitude of dx, an

approximate closed-form expression for the shift may be obtained readily.
Otherwise, a straightforward gradient search algorithm could be used to
determine a solution.
Figure 9.24 Generalization of processing approaches to higher-dimensional problem.
For any of the differential speckle measurement techniques, whether using

a single detector and two laser beams or a single laser beam and two detectors,
one arrives at a pair of stacked speckle histories. These histories display the
time rate of movement of the individual speckles whether they are subjective
or objective. The relationship between these speckle motions and the strain
undergone by the test specimen is dictated by the actual measurement
configuration. As an example, consider the configuration shown in Fig. 9.7. If
dx1 and dx2 represent the speckle motions derived from the two speckle
histories (one for laser beam #1, and the other for laser beam #2), then the
object strain in the plane of the page is given by
dx1 dx2
εxx ¼ , (9.80)
2Lo sin us
where Lo is the effective object difference, and us is the illumination angle. For
objective speckle, the effective object distance is the physical distance between
the object and the detector focal plane. In the case of subjective speckle, it is
the misfocus distance. For the complementary configuration of one normally
incident laser beam and two cameras, the corresponding relationship is
dx1 dx2
εxx ¼ , (9.81)
2Lo tan us
where uo is the observation angle. Note that these are the same results as
Eqs. (9.23)–(9.25).
544 Chapter 9
9.13 Elastography of Tissues with Optical Coherence

Tomography
While optical elastography by laser speckle tracking is an elegantly simple
approach, it is limited to probing tissue surfaces, and potentially confounded
by multiple scattering from deeper in tissues. Performing such measurements
using optical interference with a low coherence reference beam, e.g., as per-
formed by OCT, provides advantages in sensitivity, through heterodyne
detection, and in depth sectioning, enabling assessment of strain that is
spatially localized in three dimensions.
Schmitt105 initially proposed OCT elastography, now usually referred to
as optical coherence elastography (OCE), in 1998, demonstrating the ability
to quantify very small internal strains in tissue phantoms, soft tissue (pig)
samples, and skin in vivo. In this initial pioneering demonstration, a quasi-
static mechanical load was applied to the sample using an external
compression plate. To estimate internal displacements, the shift in the speckle
patterns of sequential 2-D images was tracked using either a cross-correlation
or optical flow algorithm. Displacements as small as a few microns and strains
of less than 1% were measured to a depth of about 1 mm. Subsequent to this
initial demonstration, a number of closely related speckle-tracking-based
OCE techniques were proposed, mainly focused on developing OCE as a tool
to assess arterial mechanical properties.106–108
As these early OCE techniques employed speckle-tracking techniques
similar to those described in Section 9.6 and 9.7, they can be classified as
intensity-based methods. However, with the advent of Fourier-domain OCT,
ready access to the phase of the detected speckle pattern became possible. This
led to a shift from speckle-tracking-based techniques to phase-sensitive
techniques. Phase-sensitive detection in OCE was first proposed in several
papers in 2006.109,110 In Fourier-domain OCT (FdOCT), after Fourier
transformation of the detected spectral fringes, the depth-resolved complex
amplitude is obtained, providing access to the phase of the detected signal.
The OCT phase appears to be random in soft tissue; however, it is temporally
invariant if the sample is stationary. If a sample is subjected to mechanical
loading, its small axial displacement between two A-scans acquired from the
same lateral position results in a phase shift. Such phase-sensitive methods
were first developed to measure flow in Doppler OCT.111,112
Here, we describe phase-sensitive OCE through the use of a FdOCT
system based on the spectral-domain (spectrometer-based) implementation.
However, it should be appreciated that the theory also applies to swept-source
and time-domain OCT systems.
In spectral-domain OCT, the interference signal between the reference
light and the scattered light from within a sample is spectrally resolved by a
spectrometer and a linear array detector, usually a line-scan CCD or CMOS

camera. The light intensity incident on each element of the linear array
detector is proportional to the spectral density, Id(v), of the combined
reference and sample light, which can be expressed as
h pffiffiffiffiffiffiffiffiffiffi i
I d ðvÞ ¼ SðvÞ 1 þ ∫RðtÞdt þ 2Rself þ 2∫ RðtÞ cosð2pvt þ f0 Þdt , (9.82)
where v is the optical frequency. Range information is determined from the

differential propagation time t of the light backscattered by the scatterers
within the sample relative to the reference mirror, R(t) is the fraction of
the intensity backscattered from the scatterer at time t, and S(v) is the
spectral density of the light source. The third term within the square
brackets represents the self interference, Rself, of the backscattered light
from the different scatterers in the sample. The reflection in the reference
arm is assumed to be unity for clarity. The fourth term contained within
the brackets is the spectral interference between the sample and the
reference and is the term used to generate OCT images. Neglecting all but
the signal term, we write, for convenience,
pffiffiffiffiffiffiffiffiffiffi
I d ðvÞ ¼ 2 SðvÞ∫ RðtÞ cosð2pvt þ f0 Þdt: (9.83)
The depth information regarding the local scatterers in the sample is obtained
by Fourier-transforming the spectral interference signal above to obtain:
I ðzÞ ¼ F ½I d ðvÞ ¼ AðzÞ exp½iFðzÞ: (9.84)
Note that Eq. (9.84) is a complex function in which the amplitude describes
the OCT structural image. The phase F(z) appears to be pseudo-random with
depth, z, and is temporally invariant, i.e., it is constant versus time for a given
position for (relatively) static scatterers. A translation of the scatterer at
depth z by a distance Dd(z) during the time interval Dt between two
successive A-scans will induce a change in the measured phase of the
reflected light given by113
DFðz,tÞ ¼ 2nkDdðzÞ, (9.85)
where n is the refractive index of the sample, and k is the wavenumber of the
light source (k ¼ 2p/l ¼ 2pv/c). Calculating this phase difference at each
depth z yields depth-resolved measurements of both the magnitude and
direction of the axial (parallel to the imaging beam) displacement of the tissue
at the time t,113
DFðz,tÞl
Ddðz,tÞ ¼ : (9.86)
4pn
546 Chapter 9
Similar to phase-resolved optical Doppler tomography, where the concern is

the determination of flow velocity, the depth-resolved tissue velocity v(z,t) in
the beam direction is given by111
DFðz,tÞl
vðz,tÞ ¼ : (9.87)
4pnDt
After the depth-resolved instantaneous displacement and velocity are
obtained, the strain rate map ε_ ðz,tÞ representing the localized elastic properties
of tissue can be generated:
vðz,tÞ DFðz,tÞl
ε_ ðz,tÞ ¼ ¼ , (9.88)
z0 4pnz0 Dt
where zo is the initial depth in the sample before the tissue movement, i.e.,
before the compression, and the dot over the character denotes the time
derivative. So far, we have derived the depth-resolved instantaneous
displacement [Eq. (9.86)] and strain [Eq. (9.87)] of the sample at the time t.
The total displacement and strain over a time period T can, therefore, be
obtained by integration of the instantaneous displacement and strain over
the elapsed time, respectively. As a consequence, the depth-resolved displace-
ment d(z) and strain ε(z) maps of the sample over the time duration T can be
written as
DF ðz,tÞl
dðzÞ ¼ ∫0 Ddðz,tÞdt ¼ ∫0
T T
dt, (9.89)
4pn
DFðz,tÞl
εðzÞ ¼ ∫0 ε_ ðz,tÞdt ¼ ∫0
T T
dt: (9.90)
4pnz0 Dt
To ensure correlation between the phase measurements made at a given depth
in successive A-scans while simultaneously transversally scanning, the
transverse displacement of the imaging beam between A-scans must be small
enough relative to the probe beam size to avoid spatial speckle decorrelation.
This constraint can be met by densely sampling in the transverse direction.
Above, we describe phase-sensitive detection in the context of comparing
consecutive A-scans acquired within a B-scan. A similar approach was
outlined by Wang et al.114 in which the phase difference was calculated
between A-scans acquired in consecutive B-scans. Recently, a technique was
proposed in which the phase difference was calculated between A-scans
acquired in consecutive C-scans, enabling rapid acquisition of OCE volumes
in as little as 5 s.115
Phase-sensitive detection is the most prominent displacement-estimation
technique in contemporary OCE systems. However, speckle-tracking methods
are also being utilized.116 Advantages of phase-sensitive detection include
higher displacement sensitivity (in the picometer range) and a larger dynamic
range, particularly if phase-unwrapping techniques are used to extend the
maximum measureable displacement.117 However, it is anticipated that
speckle-tracking techniques will continue to be utilized. Speckle tracking has
the distinct advantage that it can readily be used to measure displacement in
both axial and transverse dimensions. This feature is needed to allow accurate
quantification of tissue stiffness in OCE. Phase-sensitive detection, on the
other hand, measures only the axial component of displacement. Extensions
of phase-sensitive detection, developed mainly for OCT flow imaging, have
extended the technique to enable the detection of transverse displacements;
however, they require either additional hardware to provide illumination of
the sample at multiple angles or they necessitate a degradation in spatial
resolution.118 Additionally, speckle tracking may prove to be more robust for
in vivo measurements as it is likely that it will be able to better tolerate motion
artifacts caused by patient and/or operator motion.
Several other related techniques have been proposed for measuring
deformation in OCE. In one technique, based on joint spectral- and time-
domain (STd) OCT, dynamic loading applied to a tissue sample results in
an amplitude spectrum of frequency tones described by Bessel functions.119
In this technique, as well as the standard Fourier transform used to obtain
depth-resolved information in Fourier domain OCT, a second Fourier
transform is performed in the time domain, i.e., across multiple A-scans.
The result is a Doppler spectrum containing frequency tones, which can
be used to extract the vibration amplitude of the tissue. It has been shown
that this technique is more resilient to noise that phase-sensitive detection.
Another method, closely related to speckle tracking, measures strain
following a compressive load directly from the speckle decorrelation.120
This is an attractive technique as it bypasses displacement estimation
inherent in other techniques. The principle of this technique is that for
softer tissue, higher speckle decorrelation results following compressive
loading than for stiffer tissue. Importantly, speckle decorrelation can result
from both displacement and strain. In order to isolate decorrelation caused
by strain, it is important to first translate the deformed speckle pattern
relative to the initial speckle pattern to maximize correlation between the
two speckle patterns, thus accounting for the decorrelation caused by
displacement.
9.13.1 Variants of OCE

A number of OCE techniques have been proposed and may be classified
according to the method used to load the sample as follows:116
• Compression/indentation techniques, in which a quasi-static external
load is applied to a portion or all of a region being imaged.121,122
548 Chapter 9
• Surface wave/shear wave techniques, in which a pulsed or periodic load

generates surface/shear waves that are detected using OCT.123
• Magnetomotive techniques, which employ magnetic nanoparticles
distributed in the tissue that are actuated using an external magnetic
field to produce localized displacements.124
• Acoustic radiation force techniques, in which a force is generated
by an ultrasound beam, the force being localized by the ultrasound
focus.125
• Swept-frequency loading techniques, which apply a frequency sweep of
an external load in dynamic OCE, thereby probing the temporal
response of samples.126
It is anticipated that no one technique will dominate future development of
the field and that OCE techniques will evolve according to their suitability for
particular applications. For example, noncontact methods based on the
propagation of surface/shear waves are likely to be more suited to delicate
tissues such as the eye, whereas compression techniques may be more suited to
rapid acquisition over large fields of view in applications such as
intraoperative tumor margin assessment. Currently, compression and elastic
wave techniques are the most prominent in OCE. In the next two sections, we
will describe these two techniques in more detail.
9.13.1.1 Compression OCE

Compression OCE was the first technique proposed and is the most
straightforward to implement experimentally.121 Typically, a step change in
an external compressive mechanical load is applied over the whole sample
between acquisitions (of either OCT A-scans or B-scans). The resulting tissue
displacement is measured using either speckle tracking or phase-sensitive
detection and the local axial strain (i.e., the strain, εl, measured over a small
depth range) is then estimated by measuring the change in displacement, Dd,
over an axial depth range, Dz,
Dd
εl ¼ : (9.91)
Dz
The elastogram is a map of this local strain, which is a relative measure

of a sample’s mechanical properties. Such relative measures have proven
effective in other forms of compression elastography, particularly in
ultrasound elastography.127 The lateral resolution in compression OCE
matches that of the underlying OCT system. However, the axial
resolution is determined by Dz in Eq. 9.91 and is typically 5–10 times
lower than the underlying OCT axial resolution. Calculating the Young’s
modulus from the local strain requires knowledge of the corresponding
local stress. Local stress can be measured at the sample surface, for
example, using optical palpation,128 an OCT-based tactile imaging tech-

nique that estimates surface stress by measuring the deformation in a layer
of known stiffness placed on the tissue surface between the tissue and the
compression plate. However, stress has not been directly measured within a
sample. Indirect approaches, which attempt to match the measured local
strain to a prediction based on an assumed stress distribution (so-called
inverse methods), could be attempted. Such methods have been extensively
studied in ultrasound elastography.129 It is expected that these methods will
be translated to OCE.
9.13.1.2 Surface wave/shear wave OCE

Assessment of mechanical properties through the measurement of
propagating mechanical waves is commonly performed in both seismol-
ogy130 and materials characterization131 and has been implemented with
success in both ultrasound elastography and magnetic resonance elasto-
graphy.132 In OCE, measurement of both surface acoustic waves133 and
shear waves134 has been proposed to quantify tissue stiffness. These
techniques are attractive as they provide, under assumptions of linear
elasticity, mechanical homogeneity and isotropy, a straightforward means
to directly estimate tissue stiffness. In the case of shear wave OCE, the
propagating wave has been initiated using a pulse of air, an acoustic
radiation force, or using pulsed laser irradiation via the photothermal
effect. The speed of the propagating wave, cs, can be calculated using an
OCT system off-axis to the mechanical excitation using
vDr
cs ¼ , (9.92)
Dw
where v is the acoustic angular frequency and Df is the phase shift of the
shear wave over a distance Dr. Assuming that soft tissue is incompressible
(Poisson’s ratio of 0.5), this allows the Young’s modulus, E, to be estimated
from the shear wave speed as
E ¼ 3rcs 2 : (9.93)
The speed of shear waves in soft tissue is typically in the range 1–10 m/s. It
has been evaluated from phase-sensitive OCT measurements, although in
principle could also be calculated using speckle tracking. To measure the shear
wave phase shift, an M-B scanning acquisition method has been developed
whereby a number of A-scans are acquired at each lateral position.135
Surface wave OCE is similar to shear OCE wave, with the key distinction
being that the wave propagates largely along the surface rather than in the
bulk of the tissue. Assuming the surface wave propagates as a Rayleigh
550 Chapter 9
wave, the Young’s modulus, E, can be calculated from the surface wave
velocity, csw, as136,137
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0.87 þ 1.12y E
csw ¼ : (9.94)
1þy 2rð1 þ yÞ
9.13.2 OCE probes

As described above, a number of OCE techniques have been demonstrated,
and a number of applications have been proposed. For OCE to successfully
transition to clinical applications, the development of probes tailored for
specific applications is a key requirement. For example, preliminary mea-
surements on excised tissues have shown the potential of OCE for assessing
atherosclerotic plaques.138 However, before this application can be explored
in vivo, a requirement is the development of intravascular OCE probes. In
addition to the usual challenges in developing small footprint optical imaging
probes, OCE is faced with the additional challenge of incorporating
mechanical loading into the probe. In one solution, the OCT probe has been
miniaturized and inserted into a hypodermic needle.139 In this forward-facing
probe, the needle acts as its own loading (indenting) mechanism, and the
tissue deformation ahead of the needle is measured by the probe during
insertion. Incorporation into a needle holds promise for measuring micro-
scale mechanical properties of tissues at depths far greater than those
achievable using external probes.
9.14 Acoustically Modulated Speckle Imaging

Jacques and Kirkpatrick78 presented a video laser speckle imaging technique
that uses the mechanical properties of the tissue as an image contrast
mechanism. In their system, called acoustically modulated speckle imaging
(amSI), the contrast within the tissue images is based upon the scale of the
fluctuations of laser speckle patterns in response to a mechanical driving
force swept through a range of frequencies, typically 0–30 Hz. As the tissue
is vibrated, the tissue movement alters the speckle pattern, causing the
intensity observed in any particular pixel on the video CCD to fluctuate. If
there is significant fluctuation during the integration time of the camera,
each pixel will then yield an intensity that is the average of the intensities of
all the speckles that passed over the pixel. The limit is reached when each
pixel yields a mean gray image. Without fluctuation, and by carefully
arranging the experiment so that the mean speckle size is equal to, or
slightly larger than, twice the pixel size, each pixel will record either a bright
or dark speckle intensity. The concept is straightforward: when acoustically
vibrated, less-stiff tissues that readily displace, should present a more nearly
Figure 9.25 (a) Static speckle image of skin sample and (b) the same sample subject to a
30 Hz acoustic vibration.
uniform gray image, whereas stiffer tissues should present a more strongly
speckled image. That is, stiffer tissues should have a higher speckle contrast
value.
Illustrating the concept is Fig. 9.25(a)–(b). Here, a skin sample was first
static [Fig. 9.25(a)] and then vibrated at 30 Hz [Fig. 9.25(b)], while
illuminated by an expanded He-Ne laser (633 nm) beam. A gradual
graying (i.e., loss of speckle contrast) is seen as the modulation frequency
is increased. The camera integration time was 0.0167 s. Concurrent with
this increase in grayness (decrease in speckle contrast) is a change in the
intensity probability distribution. As the image grays, the intensity
probability distribution takes on a more Gaussian shape, centered about
552 Chapter 9
0.9500
0.9000 Native
0.8500
0.8000
Treated
G
0.7500
0.7000
0.6500
0.6000
0 5 10 15 20 25 30 35
Driving Frequency (Hz)
Figure 9.26 Grayness values of speckle contrast for a chemically stiffened and native
(nontreated) portions of a tendon.
the mean pixel intensity.78 By defining speckle contrast76 as sI ∕Ī , a

grayness value, Gi, can be defined as
G i ¼ 1 sI ∕Ī : (9.95)
Thus, as a tissue becomes perfectly compliant, Gi ! 1.0. Likewise, and

assuming a fully developed speckle pattern (which is unlikely to be the case),
an infinitely stiff tissue should have a Gi ! 0.0. Real tissues should be
somewhere in between. Figure 9.26 displays the effect of stiffening one half of
a tendon by soaking it in a glutaraldehyde solution. The native (nonstiffened)
half has a higher Gi than does the stiffened half. Kirkpatrick et al.140 presented
a related method for assessing tissue mechanics by evaluating the spatial
variations in the temporal contrast of a sequence of translating laser speckle
patterns.
It is worth noting that both approaches described above are closely related
to the concept of laser speckle contrast analysis (LASCA) as first described by
Fercher and Briers.141 LASCA and some variations on this concept are
described more fully in Chapter 6 of this volume. Many developments aimed
at making LASCA a truly quantitative tool, as opposed to a purely correlative
one, have since been published.142–144
9.15 Conclusions
The optical elastography methods presented here characterize the mechanical
properties of tissue on a length scale intermediate between cell-based methods
and medical imaging methods. Because these methods lend themselves to in
vivo applications, they hold the promise of assisting in the diagnosis of diseases,
with early results in the eye, skin, and breast pointing the way to future clinical
tools. It is also evident that these methods will make important future
contributions to biomechanics, biomaterials, and tissue engineering as well.
References
1. J. Ophir, I. Cespedes, H. Ponnekanti, Y. Yazdi, and X. Li,
“Elastography: A quantitative method for imaging the elasticity of
biological tissues,” Ultrason. Imaging 13, 111–134 (1991).
2. R. Jones and C. Wykes, Holographic and Speckle Interferometry.
A Discussion of the Theory, Practice and Applications of the Techniques,
Cambridge University Press, Cambridge (1983).
3. K. J. Gåsvik, Optical Metrology Second Edition, John Wiley & Sons,
Chichester (1995).
4. G. Cloud, Optical Methods of Engineering Analysis, Cambridge
University Press, Cambridge (1995).
5. P. K. Rastogi (ed.), Optical Measurement Techniques and Applications,
Artech House, Inc., Boston (1997).
6. A. Oulamara, G. Tribillon, and J. Duvernoy, “Biological activity
measurement on botanical specimen surfaces using a temporal decorr-
elation effect of laser speckle,” J. Mod. Opt. 36, 165–179 (1989).
7. L. Gao, K. J. Parker, R. M. Lerner, and S. F. Levinson, “Imaging of the
elastic properties of tissue – a review,” Ultrasound in Med. & Biol. 22(8),
959–977 (1996).
8. R. M. Lerner, S. R. Huang, and K. J. Parker, “Sonoelasticity images
derived from ultrasound signals in mechanically vibrated tissues,”
Ultrasound in Med. & Biol. 16(3), 231–239 (1990).
9. H. E. Talhami, L. S. Wilson, and M. L. Neale, “Spectral tissue strain:
a new technique for imaging tissue strain using intravascular ultra-
sound,” Ultrasound in Med. & Biol. 20(8), 759–772 (1994).
10. R. Muthupillai, P. J. Rossman, D. J. Lomas, J. F. Greenleaf, S. J.
Riederer, and R. L. Ehman, “Magnetic resonance imaging of transverse
acoustic waves,” Magn. Reson. Med. 36, 266–274 (1996).
11. E. Konofagou and J. Ophir, “A new elastographic method for
estimation and imaging of lateral displacements, lateral strains, corrected
axial strains and Poisson’s ratios in tissues,” Ultrasound in Med. & Biol.
24(8), 1183–1199 (1998).
554 Chapter 9
12. S. J. Kirkpatrick and M. J. Cipolla, “High resolution imaged laser

speckle strain gauge for vascular applications,” J. Biomed. Optics 5(1),
62–71 (2000).
13. A. Chau, R. Chan, M. Shishkov, B. MacNeill, N. Iftimia, G. Tearney,
R. Kamm, B. Bouma, and M. Kaazempur-Mofrad, “Mechanical
analysis of atherosclerotic plaques based on optical coherence tomogra-
phy,” Ann. Biomed. Eng. v32(11), 1494–1503 (2004).
14. K. M. Kennedy, B. F. Kennedy, R. A. McLaughlin, and D. D.
Sampson, “Needle optical coherence elastography for tissue boundary
detection,” Opt. Lett. 37(12), 2310–2312 (2012).
15. S. J. Kirkpatrick, R. K. Wang, D. D. Duncan, M. Kulesz-Martin, and
K. Lee “Imaging the mechanics of skin lesions by in vivo acousto-optical
elastography,” Opt. Exp. 14, 9771–9779 (2006).
16. F. M. Hendriks, D. Brokken, C. W. Oomens, D. L. Bader, and F. P.
Baaijens, “The relative contributions of different skin layers to the
mechanical behavior of human skin in vivo using suction experiments,”
Med. Eng. Phys. 28(3), 259–266 (2006).
17. Y. Zhang, R. T. Brodell, E. N. Mostow, C. J. Vinyard, and H. Marie,
“In vivo skin elastography with high-definition optical videos,” Skin Res.
Technol. 15(15), 271–282 (2009).
18. X. Liang and S. A. Boppart, “Biomechanical properties of in vivo human
skin from dynamic optical coherence elastography,” IEEE Trans.
Biomed. Eng. 57(4), 953–959 (2010).
19. B. F. Kennedy, X. Liang, S. G. Adie, D. K. Gerstmann, B. C. Quirk,
S. A. Boppart, and D. D. Sampson, “In vivo three-dimensional optical
coherence elastography,” Opt. Exp. 19(7), 6623–6634 (2011).
20. C. Cacou, J. M. Anderson, and I. F. K. Muir, “Measurement of closing
force of surgical wounds and relation to the appearance of resultant
scars,” Med. Biolog. Eng. Comput. 32, 638–642 (1994).
21. E. E. Peacock, Jr. and I. K. Cohen, “Wound Healing,” in Plastic Surgery
Volume 1: General Principles, J. G. McCarthy, ed., Saunders Press,
Philadelphia (1990).
22. S. Es’haghian, K. M. Kennedy, P. Gong, D. D. Sampson, Robert A.
McLaughlin, and B. F. Kennedy, “Optical palpation in vivo: imaging
human skin lesions using mechanical contrast,” J. Biomed. Opt. 20(1),
016013 (2015).
23. X. Liang, A. L. Oldenburg, V. Crecea, E. J. Chaney, and S. A. Boppart,
“Optical micro-scale mapping of dynamic biomechanical tissue proper-
ties,” Opt. Exp. 16(15), 11052–11065 (2008).
24. A. Srivastava, Y. Verma, K. D. Rao, and P. K. Gupta, “Determina-
tion of elastic properties of resected human breast tissue samples using
optical coherence tomographic elastography,” Strain 47(1), 75–87
(2011).
25. B. F. Kennedy, R. A. McLaughlin, K. M. Kennedy, L. Chin,

A. Curatolo, A. Tien, B. Latham, C. M. Saunders, and D. D. Sampson,
“Optical coherence micro-elastography: mechanical-contrast imaging of
tissue microstructure,” Biomed. Opt. Exp. 5(7), 2113–2124 (2014).
26. S. Li, K. D. Mohan, W. W. Sanders, and A. L. Oldenburg, “Toward soft
tissue elastography using digital holography to monitor surface acoustic
waves,” J. Biomed. Opt. 16(11), 116005 (2011).
27. J. F. Waljee, E. S. Hu, L. A. Newman, and A. K. Alderman, “Predictors
of re-excision among women undergoing breast-conserving surgery for
cancer,” Ann. Surg. Oncol. 15, 1297–303 (2008).
28. C. Li, G. Guan, Y. Ling, Y-T. Hsu, S. Song, J. T-J. Huang, S. Lang,
R. K. Wang, Z. Huang, and G. Nabi, “Detection and characterisation of
biopsy tissue using quantitative optical coherence elastography (OCE) in
men with suspected prostate cancer,” Cancer Lett. 357, 121-8 (2015).
29. M. R. Ford, J. W. J. Dupps, A. M. Rollins, A. S. Roy, and Z. Hu,
“Method for optical coherence elastography of the cornea,” J. Biomed.
Opt. 16(1), 016005 (2011).
30. D. Alonso-Caneiro, K. Karnowski, B. J. Kaluzny, A. Kowalczyk, and
M. Wojtkowski, “Assessment of corneal dynamics with high-speed swept
source optical coherence tomography combined with an air puff system,”
Opt. Exp. 19(15), 14188–14199 (2011).
31. C. Li, G. Guan, Z. Huang, M. Johnstone, and R. K. Wang,
“Noncontact all-optical measurement of corneal elasticity,” Opt. Lett.
37(10), 1625–1627 (2012).
32. C. Dorronsoro, D. Pascual, P. Pérez-Merino, S. Kling, and
S. Marcos, “Dynamic OCT measurement of corneal deformation by
an air puff in normal and cross-linked corneas,” Biomed. Opt. Exp. 3(3),
473–487 (2012).
33. M. D. Twa, J. Li, S. Vantipalli, M. Singh, S. Aglyamov, S. Emilianov,
and K. V. Larin, “Spatial characterization of corneal biomechanical
properties with optical coherence elastography after UV cross-linking,”
Biomed. Opt. Express 5 (5), 1419–1427 (2014).
34. G. Scarcelli, S. Kling, E. Quijano, R. Pineda, S. Marcos, and S. H. Yun,
“Brillouin microscopy of collagen crosslinking: noncontact depth-
dependent analysis of corneal elastic modulus,” Invest. Ophthalmol.
Vis. Sci. 54(2), 1418–1425 (2013).
35. G. Scarcelli, P. Kim, and H. Y. Yun, “In vivo measurement of age-
related stiffening in the crystalline lens by brillouin optical microscopy,”
Biophysical J. 101(6), 1539–1545 (2011).
36. C. L. de Korte, G. Pasterkamp, A. F. W. van der Steen, H. A.
Woutman, and N. Bom, “Characterization of plaque components with
intravascular ultrasound elastography in human femoral and coronary
arteries in vitro,” Circulation 102(6), 617–623 (2000).
556 Chapter 9

phy,” Ann. Biomed. Eng. 32(11), 1494–1503 (2004).
38. J. Rogowska, N. Patel, J. Fujimoto, and M. Brezinski, “Optical
coherence tomographic elastography technique for measuring deforma-
tion and strain of atherosclerotic tissues,” Heart 90, no.(5), pp. 556–562
(2004).
39. W. Qi, R. Chen, L. Chou, G. Liu, J. Zhang, Q. Zhou, and Z. Chen,
“Phase-resolved acoustic radiation force optical coherence elastogra-
phy,” J. Biomed. Opt. 17(11), 110505 (2012).
40. S. K. Nadkarni, B. E. Bouma, T. Helg, R. Chan, E. Halpern, A. Chau,
M. S. Minsky, J. T. Motz, S. L. Houser, and G. J. Tearney,
“Characterization of atherosclerotic plaques by laser speckle imaging,”
Circulation 112(6), 885–892 (2005).
41. J. Yin, H.-C. Yang, X. Li, J. Zhang, Q. Zhou, C. Hu, K. K. Shung, and
Z. Chen, “Integrated intravascular optical coherence tomography ultra-
sound imaging system,” J. Biomed. Opt. 15(1), 010512 (2010).
42. S. J. Kirkpatrick and D. D. Duncan, “Direct measurement of strain rates
in biological tissues,” Proc. SPIE 3914, 630–638 (2000).
43. H. J. Ko, W. Tan, R. Stack, and S. A. Boppart, “Optical coherence
elastography of engineered and developing tissue,” Tissue Eng. 12(1),
63–73 (2006).
44. S. J. Kirkpatrick, M. T. Hinds, and D. D. Duncan, “Acousto-optical
characterization of the viscoelastic nature of a nuchal elastin tissue
scaffold,” Tissue Eng. 9(4), 645–656 (2003).
45. B. D. Boyer, C. H. Lohmann, J. Romero, and Z. Schwartz, “Bone and
cartilage tissue engineering,” Clin. Plast. Surg. 26(4), 629–645 (1999).
46. G. C. Niedeeraurer, M. A. Slivka, N. C. Leatherbury, D. L. Korvick,
and K. Kieswetter, “Multiphase implants with varying stiffness affect
osteochondral cartilage repair,” Trans. Soc. Biomat. 22, 261 (1999).
47. C. H. Liu, M. N. Skryabina, J. Li, M. Singh, E. N. Sobol, and K. V.
Larin, “Measurement of the temperature dependence of Young’s
modulus of cartilage by phase-sensitive optical coherence elastography,”
Quantum Electronics 44(8), 751–756 (2014).
48. P. A. Janmey and C. A. McCulloch, “Cell mechanics: integrating cell
responses to mechanical stimuli,” Annual Review of Biomedical
Engineering 9, 1–34 (2007).
49. Y. C. Fung, A First Course in Continuum Mechanics, 2nd Edition,
Prentice-Hall, Englewood Cliffs, N. J., USA (1977).
50. T. T. Tanaka and Y. C. Fung, “Elastic and inelastic properties of the
canine aorta and their variation along the aortic tree,” J. Biomech. 7,
357–370 (1974).
51. Y. C. Fung, Biomechanics: Mechanical Properties of Living Tissues,

Springer-Verlag, New York (1981).
52. Y. C. Fung, “Structure and stress-strain relationship of soft tissue,”
Amer. Zool. 24, 13–22 (1984).
53. Y. Lanir, “Skin Mechanics,” in Handbook of Bioengineering, R. Skalak
and S. Chien, eds., McGraw-Hill, New York (1987).
54. D. N. Bingham and P. H. Dehoff, “A constitutive equation for the
canine anterior cruciate ligament,” J. Biomech. Eng. 101, 15–24 (1979).
55. R. S. Lakes and J. L. Katz, “Interrelationships among the viscoelastic
function for anisotropic solids: application to calcified tissues and related
systems,” J. Biomech. 7, 259–267 (1974).
56. R. S. Lakes and J. L. Katz, “Viscoelastic properties and behavior of
cortical bone part II: relaxation mechanisms,” J. Biomech. 12, 679–688
(1979).
57. R. S. Lakes and J. L. Katz, “Viscoelastic properties of wet cortical bone
part III: A nonlinear constitutive equation,” J. Biomech. 12, 689–697
(1979).
58. S. C. Cowin, Bone Mechanics, CRC Press, Boca Raton, FL, USA
(1989).
59. K. Hayashi, “Experimental approaches on measuring the mechanical
properties and constitutive laws of arterial walls,” J. Biomech. Eng. 115,
481–488 (1993).
60. G. A. Holzapfel and R. W. Ogden “Constitutive modelling of arteries,”
Proc. R. Soc. A 466, 1551–1597 (2010).
61. R. N. Vaishnav, J. T. Young, and D. J. Patel, “Distribution of stresses
and strain-energy density through the wall thickness in a canine aortic
segment,” Circulation Research 32, 577–583 (1973).
62. R. N. Vaishnav and J. Vossoughi, “Incremental formulations in vascular
mechanics,” J. Biomech. Eng. 106, 105–111 (1984).
63. D. J. Patel and R. N. Vaishnav, Basic Hemodynamics and its Role in
Disease Process, University Park Press, Baltimore, MD, USA (1981).
64. Y. C. Fung, K. Fronek, and P. Patitucci, “Pseudoelasticity of arteries
and the choice of its mathematical expression,” Am. J. Physiol. 237,
H620–H631 (1979).
65. K. Takamizawa and K. Hayashi, “Strain energy density function and
uniform strain hypothesis for arterial mechanics,” J. Biomech. 20, 1–17
(1974).
66. D. J. Schneck, Mechanics of Muscle, 2nd Edition, New York University
Press, N.Y. (1992).
67. D. J. Schneck and M. C. Simanowith, “Power-law modeling of the
passive, nonlinear, time dependent constitutive behavior of physiologic
tissue,” in Biomedical Engineering/Recent Developments, J. Vossoughi, ed.,
University of the District of Columbia Press, Washington, D.C. (1994).
558 Chapter 9
68. F. J. Walburn and D. J. Schneck, “A constitutive equation for whole

human blood,” J. Biomech. 13, 201–212 (1976).
69. Y. C. Fung, S. Q. Liu, and J. B. Zhou, “Remodeling of the constitutive
equation while a blood vessel remodels itself under stress,” J. Biomech.
Eng. 115, 453–461 (1993).
70. G. A. Holzapfel and R. W. Ogden, “Constitutive modeling of arteries,”
Proc. R. Soc. A. 466, 1551–1597 (2010).
71. S. A. Wainwright, W. D. Biggs, J. D. Currey, and J. M. Gosline,
Mechanical Design in Organisms, Princeton University Press, Princeton,
N.J. (1976).
72. P. Wijesinghe, R. A. McLaughlin, D. D. Sampson, and B. F. Kennedy,
“Parametric imaging of viscoelasticity using optical coherence elasto-
graphy,” Physics in Medicine and Biology 60(6), 2293–2307 (2015).
73. S. J. Kirkpatrick and B. W. Brooks, “Micromechanical behavior of
cortical bone as inferred from laser speckle data,” J. Biomed. Mat. Res.
39, 373–379 (1998).
for Medical Diagnosis, second edition, PM 166, SPIE Press, Bellingham,
WA, (2007).
75. D. D. Duncan and S. J. Kirkpatrick, “Processing techniques for laser
speckle derived from biological tissues,” Proc. SPIE 3914, 639–647
(2000).
76. J. W. Goodman, “Statistical Properties of Laser Speckle Patterns,” in
Laser Speckle and Related Phenomena, J. C. Dainty, ed., Springer-
Verlag, N.Y. (1984).
77. A. E. Ennos, “Speckle Interferometry” in Laser Speckle and Related
Phenomena, J. C. Dainty, ed., Springer-Verlag, N.Y. (1984).
78. S. L. Jacques and S. J. Kirkpatrick, “Acoustically modulated speckle
imaging of biological tissues,” Opt. Lett. 23(11), 879–881 (1998).
79. L. Chin, A. Curatolo, B. F. Kennedy, B. J. Doyle, P. R. T. Munro, R. A.
McLaughlin, and D. D. Sampson, “Analysis of image formation in
optical coherence elastography using a multiphysics approach,” Biomed.
Optics Express 5(9), 2913–2930 (2014).
80. I. Yamaguchi, “Speckle displacement and decorrelation in the diffrac-
tion and image fields for small object deformation,” Opt. Acta 28, 1359–
1376 (1981).
81. I. Yamaguchi, “Simplified laser speckle strain gauge,” Opt. Eng. 21,
436–440 (1982).
82. I. Yamaguchi, “Advances in the laser speckle strain gauge,” Opt. Eng.
27, 214–218 (1988).
83. D. D. Duncan, F. F. Mark, and L. W. Hunter, “A new speckle technique
for noncontact measurement of small creep rates,” Opt. Eng. 31, 1583–
1589 (1992).
84. D. D. Duncan, S. J. Kirkpatrick, F. F. Mark, and L. W. Hunter,

“Transform method of processing for speckle strain-rate measurements,”
Appl. Opt. 33(22), 5177–5186 (1994).
85. J. P. Fitch, Synthetic Aperture Radar, Springer-Verlag, New York (1988).
86. D. D. Duncan, S. J. Kirkpatrick, F. F. Mark, and L. W. Hunter,
“Measurement of strain rates in reinforcement fibers,” Meas. Sci.
Technol. 6, 355–364 (1995).
87. S. J. Kirkpatrick, D. A. Covey, and B. W. Brooks, “Direct measurement
of strain rate variation in the vicinity of a round hole in mandibular
cortical bone,” Proc. 21st Am. Soc. Biomech., 296–297 (1997).
88. H. Kasprzak and H. Podbielska, “Speckle photography in biomechani-
cal testing,” Proc. SPIE 2083, 268–279 (1993).
89. N. Mo and J. C. Shelton, “Laser speckle photography to measure
surface displacements on cortical bone– verification of an automated
analysis system,” Med. Eng. & Physics 20, 594–601 (1998).
90. B. Ovyrn, “Whole-field displacement and strain measurements: applica-
tions to biomechanics,” Proc. SPIE 1889, 134–148 (1993).
91. J. C. Shelton and D. M. Katz, “ Laser interferometric techniques in
orthopaedics,” J. Med. Eng. & Technol. 15(6), 209–221 (1991).
92. S. J. Kirkpatrick and D. D. Duncan, “Noncontact microstrain
measurements in orthodontic wires,” J. Biomed Mater Res 29, 1437–
1442 (1995).
93. D. D. Duncan and S. J. Kirkpatrick, “Measurement of strain rate
in reinforcement fibers,” Proc. Soc. Exper. Mech., 742–746 (1994).
94. T. Takemori, K. Fujita, and I. Yamaguchi, “Resolution improvement
in speckle displacement and strain sensors by correlation interpolation,”
in Laser Interferometry IV: Computer-Aided Interferometry, R. J.
Pryputniewicz, ed., Proc. SPIE 1553, 137–148 (1991).
95. A. Nahas, M. Bauer, S. Roux, and A. C. Boccara, “3D static
elastography at the micrometer scale using Full Field OCT,” Biomed.
Optics Express 4(10), 2138–2149 (2013).
96. V. Y. Zaitsev, L. A. Matveev, G. V. Gelikonov, A. L. Matveyev, and
V. M. Gelikonov, “A correlation-stability approach to elasticity
mapping in optical coherence tomography,” Laser Physics Letters
10(6), 065601 (2013).
97. A. Papoulis, Probability, Random Variables, and Stochastic Processes,
3rd edition, McGraw-Hill, New York (1991).
98. J. S. Walker, Fast Fourier Transforms, CRC Press, Boca Raton (1991).
99. M. C. Roggemann and B. Welsh, Imaging Through Turbulence, CRC
Press, Boca Raton (1996).
100. A. D. Belegundu and T. R. Chandrupatla, Optimization Concepts and
Applications in Engineering, Prentice Hall, Upper Saddle River, NJ (1999).
560 Chapter 9
101. R. J. Shilling and S. L. Harris, Applied Numerical Methods for Engineers

Using MATLAB and C, Brooks/Cole Pacific Grove (2000).
102. B. Jähne, Practical Handbook on Image Processing for Scientific
Applications, CRC Press, Boca Raton (1997).
103. B. R. Frieden, Probability, Statistical Optics, and Data Testing: A
Problem Solving Approach, Second Edition, Springer-Verlag, Berlin
(1991).
104. H. L. Van Trees, Detection, Estimation, and Modulation Theory, Part I,
John Wiley and Sons, New York (1968).
105. J. M. Schmitt, “OCT elastography: imaging microscopic deformation
and strain of tissue,” Opt. Exp. 3(6), 199–209 (1998).
106. J. Rogowska, N. Patel, J. Fujimoto, and M. Brezinski, “Optical
coherence tomographic elastography technique for measuring defor-
mation and strain of atherosclerotic tissues,” Heart 90(5), 556–562
(2004).
phy,” Ann. Biomed. Eng. 32(11), 1494–1503 (2004).
108. R. Chan, A. Chau, W. Karl, S. Nadkarni, A. Khalil, N. Iftimia,
M. Shishkov, G. Tearney, M. Kaazempur-Mofrad, and B. Bouma,
“OCT based arterial elastography: Robust estimation exploiting tissue
biomechanics,” Opt. Exp. 12(19), 4558–4572 (2004).
109. S. J. Kirkpatrick, R. K. Wang, and D. D. Duncan, “OCT-based
elastography for large and small deformations,” Opt. Exp. 14(24),
11585–11597 (2006).
110. R. K. Wang, Z. Ma, and S. J. Kirkpatrick, “Tissue Doppler optical
coherence elastography for real time strain rate and strain mapping of
soft tissue,” Appl. Phys. Lett. 89(14), 144103 (2006).
111. R. Leitgeb, L. Schmetterer, W. Drexler, A. Fercher, R. Zawadzki, and
T. Bajraszewski, “Real-time assessment of retinal blood flow with
ultrafast acquisition by color Doppler Fourier domain optical coherence
tomography,” Opt. Exp. 11(23), 3116–3121 (2003).
112. B. Park, M. C. Pierce, B. Cense, S. -H. Yun, M. Mujat, G. Tearney,
B. Bouma, and J. de Boer, “Real-time fiber-based multi-functional
spectral-domain optical coherence tomography at 1.3 mm,” Opt. Exp.
13(11), 3931–3944 (2005).
113. S. J. Kirkpatrick, R. K. Wang, and D. D. Duncan, “OCT-based
elastography for large and small deformations,” Opt. Exp. 14(24),
11585–11597 (2006).
114. R. K. Wang, S. J. Kirkpatrick, and M. T. Hinds, “Phase-sensitive optical
coherence elastography for mapping tissue microstrains in real time,”
Appl. Physics Lett. 90, 164105 (2007).
115. B. F. Kennedy, F. G. Malheiro, L. Chin, and D. D. Sampson, “Three-

dimensional optical coherence elastography by phase sensitive compari-
son of C-scans,” J. Biomed. Opt. 19, 076006 (2014).
116. B. F. Kennedy, K. M. Kennedy, and D. D. Sampson, “A review of
optical coherence elastography: fundamentals, techniques and pro-
spects,” IEEE J. Sel. Top. Quantum Electron. 20, 7101217 (2014).
117. B. F. Kennedy, R. A. McLaughlin, K. M. Kennedy, L. Chin,
A. Curatolo, A. Tien, B. Latham, C. M. Saunders, and D. D.
Sampson, “Optical coherence micro-elastography: mechanical-contrast
imaging of tissue microstructure,” Biomed. Opt. Express 5, 2113–2124
(2014).
118. C. Blatter, B. Grajciar, L. Schmetterer, and R. A Leitgeb, “Angle
independent flow assessment with bidirectional Doppler optical coher-
ence tomography,” Opt. Lett. 38(21), 4433–4436 (2013).
119. B. F. Kennedy, M. Wojtkowski, M. Szkulmowski, K. M. Kennedy,
K. Karnowski, and D. D. Sampson, “Improved measurement of
vibration amplitude in dynamic optical coherence elastography,”
Biomed. Opt. Exp. 3(12), 3138–3152 (2012).
120. V. Y. Zaitsev, L. A. Matveev, A. L. Matveyev, G. V. Gelikoniv, and
V. M. Gelikonov, “Elastographic mapping in optical coherence
tomography using an unconventional approach based on correlation
stability,” J. Biomed. Opt. 19(2), 021107 (2014).
121. J. M. Schmitt, “OCT elastography: Imaging microscopic deformation
and strain of tissue,” Opt. Exp. 3(6), 199–211 (1998).
122. Y. Huang, Y. Zheng, S. Wang, Z. Chen, Q. Huang, and Y. He, “An
optical coherence tomography (OCT)-based air jet indentation system
for measuring the mechanical properties of soft tissues,” Meas. Sci.
Technol. 20(1), 1–11 (2009).
Biomed. Eng. 57(4), 953–959 (2010).
124. V. Crecea, A. L. Oldenburg, X. Liang, T. S. Ralston, and S. A. Boppart,
“Magnetomotive nanoparticle transducers for optical rheology of
viscoelastic materials,” Opt. Exp. 17(25), 23114–23122 (2009).
125. X. Liang, M. Orescanin, K. S. Toohey, M. F. Insana, and S. A. Boppart,
“Acoustomotive optical coherence elastography for measuring material
mechanical properties,” Opt. Lett. 34(19), 2894–2896 (2009).
126. S. G. Adie, X. Liang, B. F. Kennedy, R. John, D. D. Sampson, and S. A.
Boppart, “Spectroscopic optical coherence elastography,” Opt. Exp.
18(25), 25519–25534 (2010).
127. J. Ophir, I. Cespedes, H. Ponnekanti, Y. Yazdi, and X. Li,
“Elastography: A quantitative method for imaging the elasticity of
biological tissues,” Ultrasonic Imag. 13(2), 111–134 (1991).
562 Chapter 9
128. K. M. Kennedy, S. Es’haghian, L. Chin, R. A. McLaughlin, D. D.

Sampson, and B. F. Kennedy, “Optical palpation: Optical coherence
tomography-based tactile imaging using a compliant sensor,” Opt. Lett.
39(10), 3014–3017 (2014).
129. M. M. Doyley, “Model-based elastography: A survey of approaches to
the inverse elasticity problem,” Phys. Med. Biol. 57(3), R35–R73 (2012).
130. F. C. Lin, M. P. Moschetti, and M. H. Ritzwoller, “Surface wave
tomography of the western United States from ambient seismic noise:
Rayleigh and Love wave phase velocity maps,” Geophys. J. Int. 173(1),
281–298 (2008).
131. C. Glorieux, W. Gao, S. E. Kruger, K. van de Rostyne, W. Lauriks, and
J. Thoen, “Surface acoustic wave depth profiling of elastically
inhomogeneous materials,” J. Appl. Phys. 88(7), 4394–4400 (2000).
132. J. Parker, M. M. Doyley, and D. J. Rubens, “Imaging the elastic
properties of tissue: The 20 year perspective,” Phys. Med. Biol. 56(1),
R1–R29 (2011).
Biomed. Eng. 57(4), 953–959 (2010).
134. R. K. Manapuram, S. Aglyamov, F. M. Menodiado, M. Mashiatulla,
S. Wang, S. A. Baranov, J. Li, S. Emelianov, and K. Larin, “Estimation
of shear wave velocity in gelatin phantoms utilizing PhS-SSOCT,” Laser
Phys. 22(9), 1439–1444 (2012).
135. S. Song, Z. Huang, and R. K. Wang, “Tracking mechanical wave
propagation within tissue using phase-sensitive optical coherence
tomography: Motion artifact and its compensation,” J. Biomed. Opt.
18(12), 121505 (2013).
136. S. J. Kirkpatrick, D. D. Duncan, and L. Fang, “Low-frequency surface
wave propagation and the viscoelastic behavior of porcine skin,”
J. Biomed. Opt. 9(6), 1311–1319 (2004).
137. C. Li, G. Guan, X. Cheng, Z. Huang, and R. K. Wang, “Quantitative
elastography provided by surface acoustic waves measured by phase-
sensitive optical coherence tomograhpy,” Opt. Lett. 37(4), 722–724
(2012).
138. W. Qi, R. Chen, L. Chou, G. Liu, J. Zhang, Q. Zhou, and Z. Chen,
“Phase-resolved acoustic radiation force optical coherence elastogra-
phy,” J. Biomed. Opt. 17(11), 110505 (2012).
139. K. M. Kennedy, B. F. Kennedy, R. A. McLaughlin, and D. D.
Sampson, “Needle optical coherence elastography for tissue boundary
detection,” Opt. Lett. 37(12), 2310–2312 (2012).
140. S. J. Kirkpatrick, D. D. Duncan, R. K. Wang, and M. T. Hinds,
“Quantitative temporal speckle contrast imaging for tissue mechanics,”
J. Opt. Soc. Am. A. 25, 231–238 (2007).
141. A. F. Fercher and J. D. Briers, “Flow visualization by means of single-

exposure speckle photography,” Opt. Commun. 37, 326–329 (1981).
142. D. D. Duncan and S. J. Kirkpatrick, “Can laser speckle flowmetry be
made a quantitative tool?,” J. Opt. Soc. Am A 25, 2088–2094 (2008).
143. S. J. Kirkpatrick, D. D. Duncan, and E. M. Wells-Gray, “Detrimental
effects of speckle-pixel size matching in laser speckle contrast imaging,”
Opt. Lett. 33, 2886–2888 (2008).
144. D. Briers, D. D. Duncan, E. Hirst, S. J. Kirkpatrick, M. Larsson,
W. Steenbergen, T. Stromberg, and O. B. Thompson, “Laser speckle
contrast imaging: theoretical and practical limitations,” J. Biomed. Opt.
18, 066018 (2013).
Sean J. Kirkpatrick is a professor and department chair of

Biomedical Engineering at the Michigan Technological
University in Houghton, MI, USA. He is a Fellow of SPIE,
The International Society for Optics and Photonics and of
AIMBE, the American Institute of Medical and Biological
Engineers. His research interests are in coherent biomedical
optics, in particular, the theory an application of laser
speckle techniques to address issues in biomedicine including
fluid flows, tissue mechanics, and complex dynamics.
Donald D. Duncan is a research professor and associate chair

of the Electrical and Computer Engineering Department at
Portland State University in Portland, Oregon. His research
interests include speckle, computational Fourier optics, and
tissue optics. He is a member of SPIE, OSA, and SΞ.
Brendan F. Kennedy is a Senior Research Fellow in the

School of Electrical, Electronic and Computer Engineering
at the University of Western Australia. His principle research
interests include optical elastography and the measurement
of tissue mechanics, and the development of optical imaging
probes. He received his PhD from Dublin City University,
Ireland, in 2006. From 2006–2007, he was employed as a
Lecturer in the Electrical Engineering Department at the
University of Santiago in Chile. He has worked at The University of Western
Australia since 2008. Brendan is a member of both OSA and SPIE and in 2014
was awarded a Western Australia Tall Poppy Award for outstanding
achievements in the area of science and communication.
564 Chapter 9
David D. Sampson heads the Optical and Biomedical

Engineering Laboratory and is Director of the Centre for
Microscopy, Characterisation & Analysis, at The University
of Western Australia. He directs the Western Australian
nodes of the Australian Microscopy, Microanalysis Research
Facility and the National Imaging Facility (Australia). He is
a fellow of OSA, SPIE, and a senior member of the Institute
of Electrical & Electronics Engineers (IEEE). His research
interests are in the science and applications of light in medicine and biology
and research is focussed on the invention, study, and translation of
microscopy techniques to imaging in the living body – medical microscopy.
He is a recipient of the IEEE Photonics Society’s Distinguished Lecturer
Award.
Chapter 10
Optical Clearing of Tissues:
Benefits for Biology, Medical
Diagnostics, and Phototherapy
E. A. Genina, A. N. Bashkatov, and Yuri P. Sinichkin
I. Yu. Yanina
Saratov National Research State Medical University, Saratov, Russia
V. V. Tuchin
Saratov, Russia
10.1 Fundamentals of Optical Clearing (OC) of Tissues

and Cells
During the last 25 years, interest in the development and application of optical
methods for clinical functional imaging of physiological conditions has been
centered around diagnostics and therapy of cancer and other diseases and is
growing tremendously.1–4 It is initiated by the unique information, relative
simplicity, safety, and sufficiently low cost of optical instruments, as
compared, e.g., to x-ray computer tomography or magnetic resonance
tomography (MRT). However, the main limitation of optical diagnostic
methods, including optical diffusion tomography, optical coherence tomog-
raphy (OCT), confocal microscopy, reflection spectroscopy, etc., is the strong
scattering of light in tissues and blood that significantly reduces the contrast,
spatial resolution, and probing depth.1–4
565
566 Chapter 10
The scattering coefficient (ms) and the scattering anisotropy factor (g)
mainly depend upon the refractive index mismatch between the components
of the cells, such as cell membranes, mitochondria, nucleus, other organelles,
and cytoplasm and extracellular fluid. In fibrous tissues (stroma of eye sclera
and cornea, dermis, dura mater, connective tissue of vascular walls, fibrous
components of muscle tissue and mammary gland, cartilage, tendon, etc.)
the scattering is caused by the refractive index difference between the
interstitial fluid or cytoplasm and the extensive chains of scleroproteins
(collagen, elastin, and reticulin fibrils).5–7 The refractive index values for
nuclei and cytoplasm organelles of animal cells, containing nearly the same
amount of proteins and nucleic acids, lie within the relatively narrow interval
from 1.38 to 1.41.1,7 In particular, for the nucleus, the refractive index is
nnc ¼ 1.39,1,8 and for the cytoplasm n0 ¼ 1.35–1.37.1 The scattering particles
(organelles, protein fibrils, membranes, and globules) have a greater density of
proteins and lipids, and thus a higher refractive index (ns ¼ 1.39–1.47) in
comparison with the ground substance of the cytoplasm.1 The refractive
index magnitudes for the connective tissue fibrils lie in the range 1.41–1.53
and depend upon the degree of hydration of their major component, the
collagen.9 The refractive index of the interstitial fluid and the blood plasma
amounts to nearly 1.33–1.35, depending on the wavelength.1,10 The main
scatterers in the blood are the red blood cells (RBC) (erythrocytes), which are
acaryocytes, containing 70% water, 25% hemoglobin, and 5% lipids, sugars,
salts, enzymes, and proteins (see Chapter 2 of Vol. 1). The refractive index
values of dehydrated RBC at the wavelength of 550 nm fall within the
range of 1.61–1.66.11 The refractive index of the hemoglobin solution with a
concentration of 32 g/dL, which is a typical concentration of hemoglobin in
the erythrocyte, amounts to nearly 1.42.12,13 For human blood the refractive
index is 1.36–1.40, depending on the wavelength.1
Numerous methods have been developed to increase the tissue probing
depth.14 For example, for a long time the limit probing depth in multiphoton
microscopy did not exceed 100 mm;15 however, the combination of
multiphoton excitation of fluorescence with very high efficiency of light
collection increased the imaging depth in a scattering medium to 2 mm.16,17
The OCT allows one to study the internal structure of tissues to a depth up to
3 mm with a spatial resolution of 5–20 mm without disturbing the tissue
integrity.18 The probing depth up to a few centimeters is provided by the
multimodal imaging method, combining light absorption with acoustic
detection (referred to as photoacoustic tomography).19
One of the simple and efficient methods for solving the problem of
increasing the depth and quality of intratissular structure imaging, as well as
for increasing the precision of spectroscopic information from the deep
tissue layers and the blood, is the temporary reduction of tissue light
Optical Clearing of Tissues 567
scattering.1,6,18,20–22 In nonlinear spectroscopy and imaging, as well as in high-

precision laser surgery (particularly in cell nanosurgery), the reduction of
scattering in the tissue facilitates a decrease of the strongly focused laser beam
divergence and provides precise focusing onto the object, which essentially
improves the efficiency of the procedure and allows less radiation energy to be
delivered for tissue imaging or precise cutting.1,21–27
According to Web of Science, PubMed, and other sources, interest in the
optical clearing methods is growing tremendously, which is caused by the
progress of optical and laser technologies for application in biology and
medicine (see Fig. 10.1).
Numerous examples of physical and chemical impacts that allow one to
control the scattering properties of tissues are reported. They include
compression,28,29 stretching,30 dehydration,31,32 coagulation,33 impregnation
by biocompatible chemical agents,1,18,21–27,34–36 as well as photochemical37,38
and photothermal38 clearing.
The majority of methods reducing tissue scattering are based on the
matching of the refractive indices of the tissue components, either due to
replacing the interstitial fluid with the immersion agent that possesses a higher
refractive index, or due to increasing the concentration of proteins and
mucopolysaccharides in the interstitial fluid as a result of water diffusion from
the tissue caused by osmosis, making the refractive index of the interstitial
fluid closer to that of the fibrils. In addition, the optical homogeneity can
increase due to the packing (ordering) of scattering centers (e.g., collagen
50
40
Number of publications
various sources
Web of Science
30 PubMed
20
10
0
1960 1970 1980 1990 2000 2010
Years
Figure 10.1 Approximate evaluation of the number of publications related to the optical
clearing of tissues from 1955 to May 2015. The dependences are plotted using the
databases Web of Science, PubMed, and other available sources6HHFRORUSODWHV
568 Chapter 10
fibers) cause by both pushing out the interstitial fluid from the affected
volume and the dehydration at the expense of the agent impact or water
evaporation.1,18,21,22,25,26,28–32,39 The particular mechanisms of optical clear-
ing depend on the tissue type and the method used.
In this chapter, we discuss the physical and molecular mechanisms of
optical clearing (OC) methods based on agent immersion, tissue compression,
and photodynamic/photothermal action. The OC efficiency is demonstrated
for a few typical fibrous and cellular tissues. The beneficial applications of OC
methods for increasing the probing depth and resolution of a number of
prospective optical diagnostic methods, such as OCT, projection and
photoacoustic tomography, fluorescence imaging, nonlinear and Raman
microscopy, and terahertz imaging, are shown.
10.2 Immersion OC
Following the development of immersion refractometry applied to cells, in
1955, Barer et al.40 first proposed the optical clearing (OC) of a cell suspension
by means of a protein solution having the same refractive index as the cell
cytoplasm. In the late 1980s, the method of immersion optical clearing was
first applied to the eye sclera and cornea.41
Afterwards, a few research groups started intense studies of the specific
features and mechanisms of the OC phenomenon and demonstrated the
capabilities of the method to increase the probing depth or image contrast of
optical inhomogeneities inside a scattering medium.1,6,18,21–27,31,32,34–36,39,42–87
The beneficial results of OC obtained for many optical imaging techniques,
such as laser speckle-contrast imaging,34,70,74,76,78 OCT,20,29,35,59–60,64,68,79,81
microscopy,23,24,82,83 ultra-microscopy,84–86 etc., demonstrated the high
potential of their mutual use not only for obtaining high-resolution structural
and functional tissue images in vitro,1,21,22,39,54,55,71,72,82–86 but also
in vivo.1,22,25,34–36,39,45,50,54,55,60,64,69,70,74–76,78–80 Table 10.1 shows the increase
of the light penetration depth in tissues caused by application of optical
clearing agents (OCAs) to some imaging and spectroscopy techniques.
The study of tissue OC kinetics at penetration of an OCA allows one to
evaluate the OCA diffusion and permeability coefficients.1,22,35,43,47,87–96
Using these techniques, based mostly on collimated light transmittance
measurements for in vitro studies and OCT for in vivo, the diffusion rates of
glucose, some drugs, and OCAs were determined in eye tissues,43,91–95
muscle,96 skin,45,97,98 dura mater,87 arterial99 and lung100 tissues. The
monitoring of OCA diffusion with high temporal and depth resolution allows
one to differentiate healthy from pathologically modified tissues.101–105
The immersion OC, as already mentioned above, is based on the
impregnation (immersion) of the tissue by a biocompatible agent, possessing a
sufficiently high refractive index to match the refractive indices of the
Optical Clearing of Tissues
Table 10.1 Examples of the increase of light penetration depth at applications of OCAs to different imaging and spectroscopic techniques.
Light penetration
Technique Tissue OCA, refractive index Spectral range depth increase Reference
Confocal microscopy Dehydrated murine adipose Murray’s Clear, 1.55 334–799.3 nm ~35 fold 250
tissue, muscle tissue,
myocardium, brain
Two-photon microscopy Human skin ex vivo Anhydrous glycerol, 1.47 Excitation – 700–1000 nm, 2 fold 23
detection – 370–670 nm
Confocal/two-photon Cellulose membrane FocusClearTM, 1.46 Excitation – 820 nm, 2 fold 251
microscopy detection – 390–465 nm
3-D second-harmonic Muscle tissue of a mouse 50%-glycerol solution, 1.395 Excitation – 890 nm, 2.5 fold 24
microscopy (direct wave mode) detection – 445 nm
Raman microspectroscopy Porcine skin in vitro 80%-glycerol solution, 1.46 785 nm 4 fold 297
THz spectroscopy Bovine muscle tissue Anhydrous glycerol, 1.47 0.25–1.6 THz 45% 305
OCT Mouse embryo 50%-glycerol solution, 1.395 1310 nm 4.5 fold 59
Human skin in vivo Pre-polymer mixture of propylene 1305 nm 1.2 fold 70
glycol and polyethylene glycol, 1.47
Whole blood, diluted 6.5%-glycerol solution, 1.34 820 nm 2.4 fold 128
two times with saline
569
570 Chapter 10
scatterers and the surrounding medium, penetrating into the interstitial

space of the tissue. Commonly, the OCA has hyperosmotic proper-
ties.21,22 Many substances used in the compositions of different cosmetic
lotions and ointments satisfy these conditions. The OCAs can be roughly
divided into polyatomic alcohols [glycerol, polyethylene glycol (PEG),
polypropylene glycol, combined mixtures with a base of polypropylene
glycols and polyethylene glycols, mannitol, sorbitol, xylitol,
etc],56,57,75,83,87,103,105–108 sugars (glucose, dextrose, fructose, ribose, sac-
charose, etc.),81,87–91,94,96,97,99,100,104,109–111 organic acids (oleic and
linoleic acids),105,112,113 other organic solvents [dimethyl sulphoxide
(DMSO)],105,113–116 and x-ray contrast agents (verografin, trasograph,
hypaque, and omnipaque)1,22,42–44
At present, several physical and chemical mechanisms of light scattering
reduction under the action of OCA are proposed and thoroughly
described:29,31,32,39,43,46,47,55,56,58,62,66,71,75,77,83,87,106–117,118–122 dehydration of
tissue components, partial replacement of interstitial fluid with the immersion
agent, and structural modification or dissociation of collagen. The first
mechanism is related to the hyperosmotic properties of the OCA.31,36,39,56,120
The contact of hyperosmotic OCA with the tissue surface causes water
diffusion from the tissue. These processes produce a fast and considerable OC
effect, since first, the concentration of salts and proteins dissolved in the
interstitial fluid increases, therefore, the refractive index of the interstitial fluid
becomes closer to that of the scattering fibrils, and second, the weight and the
thickness of the tissue decrease, the tissue becomes denser, and the ordering of
the scattering components increases.36,39
Rylander et al.31 compared the optical transmission of rat skin under the
application of DMSO and glycerol with that under dehydration by
evaporation. It was shown that in the course of natural water evaporation,
the light transmission through the skin sample increased in the same way as
under dehydration caused by the OCA application but over a longer time.
In order to quantitatively analyze the water loss in the process of tissue
OC, the authors of Ref. 67 used the optical spectroscopy in the NIR.
Measuring the skin reflection coefficient at the wavelengths l1 ¼ 1100 nm and
l2 ¼ 1936 nm (the water absorption band), Xu and Wang used this double
wavelength analysis to assess the water content in the skin. The authors
demonstrated that the change of the water content in the skin correlates
with the OC efficiency during the first minutes, namely under the action of
80%-solutions of glycerol and propylene glycol. The water content rapidly
decreases during the first minutes of the OCA application, then the
dehydration process becomes slower with some rehydration. A similar
behavior was found for glucose and ethylene glycol with their action on the
rat muscle tissue.96
0.7 0.7
(2)
0.6 (2)
0.6
(1)
Dehydration degree
Dehydration degree
0.5 0.5
0.4 0.4
(1)
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
0 200 400 600 800 1000 0 50 100 150 200 250 300 350
Time, hrs Time, hrs
(a) (b)
Figure 10.2 Dehydration kinetics of human skin samples with intact (a) and perforated
(b) epidermis under the action of 88% glycerol solution (1) and in the course of interstitial
water evaporation in air (2). The symbols correspond to the averaged experimental data; the
curves show the result of approximation. The vertical lines present the standard deviation
(from the data of Ref. 56).
The kinetics of skin dehydration degree under the action of 88%-aqueous

solution of glycerol and air, studied by Genina et al.56 based on the weight
measurements of human skin samples [see Fig. 10.2(a)], allowed us to assess
the characteristic time t of the process, i.e., the time during which the
dehydration degree increases by e times. Under the action of OCA, t amounts
to 193.4 12.3 hrs, while under the conditions of natural evaporation of
interstitial water this parameter equals 1408.9 36.3 hrs.
At the same time, it is seen from Fig. 10.2 that, in contrast to the natural
drying in air, under the action of the OCA no full dehydration occurs, since a
part of the OCA penetrates into the interstitial space and is mixed with the
interstitial fluid. It is known that most of the OCAs are hygroscopic and in the
humid environment they adsorb water molecules until the saturation is
achieved. Thus, for glycerol, the saturation level amounts to 55% and for
propylene glycol 32% in volume.123 Therefore, OCAs inhibit complete
dehydration by keeping a portion of water in the tissue.
For fibrous tissues, such as sclera, dura mater, dermis, etc., both processes,
namely, the water loss and the diffusion of the hyperosmotic agent into the
skin, occur simultaneously, but the mechanism of the interstitial fluid
replacement with the OCA solution prevails for all the agents used since their
molecular size is much smaller than the mean separation between the fibrils.3,43
In the case of the transepidermal delivery of hydrophilic OCAs into the
skin, the diffusion is inhibited by the existence of the epidermis lipid barrier.
The human skin dehydration at the delivery of the hyperosmotic OCA
through the perforated stratum corneum (the perforation may be implemen-
ted, e.g., by fractional microablation) is presented in Fig. 10.2(b). The
epidermis permeability improvement resulted in the accelerated diffusion of
572 Chapter 10
both water and the 88%-glycerol solution. Figure 10.2 shows that when the
samples with an intact and perforated horny layer are dried under identical
conditions, their final dehydration degree should be practically similar.
However, from the analysis of the dehydration characteristic time (for the
perforated sample t amounts to 71.2 1.9 hrs.) it follows that the dehydra-
tion process in this case occurs by more than 20-fold faster than in the
intact epidermis (t ¼ 1408.9 36.3 hrs.). The value of t for the sample
with a perforated epidermis under the action of the OCA was only 36.4
0.7 hrs, which is almost by 5.5-fold smaller than for the intact sample
(t ¼ 193.4 12.3 hrs).56
In Fig. 10.2(b), it is well seen that with the perforated epidermis the
maximal dehydration degree under the evaporation of water from the tissue is
essentially higher than that under osmotic action. At the same time, the rate of
skin dehydration due to evaporation is nearly two times slower than that
under osmotic action.
The study of the water evaporation process allowed for the separation of
the dehydration mechanism of OC from the other mechanisms (i.e., the
matching of refractive indices due to the penetration of OCA into the tissue
and the collagen structure modification). For glycerol action, the degree of
dehydration in the samples with perforated stratum corneum is smaller than
that in the intact samples. Glycerol permeates through perforated skin much
more easily. Due to its hygroscopic properties, penetrating glycerol retains
water inside skin. From the practical point of view, the topical delivery of
OCA into the skin through the perforated channels hampers the formation of
a strong concentration gradient between the interstitial fluid and the OCA
solution, which reduces tissue dehydration and can be considered as a positive
factor for OC of skin in vivo.
Mao et al.107 focused their attention on the study of the efficiency of
six polyatomic alcohols with hydroxyl groups (1-butanol, 1,4-butanediol,
1,3-propanediol, PEG-200, PEG-400, and glycerol) with molecular weights
from 74 to 420 Da, and refractive index values from 1.40 to 1.47, affecting
the degree of skin OC. For better in vitro modeling of the topical OCA
application in vivo, the agents were applied to the pig skin epidermis topically,
while the dermis was moistened with saline. The relative transmission of the
samples, measured using the integrating sphere, was used to assess the
efficiency of OC.107 Authors found that glycerol, which is a triatomic alcohol,
provides the maximal OC effect, while monoatomic alcohol 1-butanol
provides the smallest effect, and four different diatomic alcohols provide an
intermediate one. After analysis of the interrelation between the OC efficiency
and the refractive index or molecular weight of the OCA under study, the
authors concluded that the OC effect of alcohols should be related to
the number of hydroxyl groups rather than to the refractive index or the
molecular weight of the OCA.
OCAs that use sugars and polyatomic alcohols can induce a reversible
solubility of tissue collagen and lead to the additional reduction of tissue
scattering due to a decrease of the basic scatterer size.121 The collagen fibers
have a complex self-organization structure and are the main scattering centers
in many tissues, such as skin dermis and eye sclera.1,117 The hydrogen bond is
the main bonding force between the triple collagen helices. The OCAs with
multiple hydroxyl groups possess a greater negative charge that destabilizes
the highly-ordered collagen structure until its dissociation. Since the hydrogen
bonds in the triple collagen helices belong to noncovalent interactions, the
OCA-induced collagen dissociation is reversible. Hirshburg et al.109,110,117,121,122
observed the glucose-induced dissociation of collagen fibers in tissues in vitro
and the corresponding reduction of the scattering coefficient. The subsequent
replacement of glycerol by phosphate-buffer solution provided the restoration
of the collagen structure and scattering properties.
Hirshburg et al.110 performed mathematical modeling of molecular
dynamics in order to clarify the formation of hydrogen bonds between the
alcohol (glycerol, xylitol, and sorbitol) and collagen molecules. They
divided the hydrogen bond bridges into different types depending on the
position of hydroxyl groups involved in the interaction with respect to
carbon atoms. They associated the type with the number of carbon atoms
in the hydrogen bond bridge. It was found that the bridges with a large
number, built in between the collagen molecules in the helix, can break the
collagen–collagen and collagen–hydroxyl bonds more efficiently than the
bridges with smaller numbers. Thus, the alcohols having pairs of hydroxyl
groups with a longer carbon chain between them should be more efficient
for OC than the alcohols having adjacent hydroxyl groups. The results of
the modeling explain the fact that 1,3-propanediol demonstrated an OC
potential twice as great as that of 1,2-propanediol, although both possessed
similar molecular weights (76.10 Da), close values of the refractive index
(1.44 and 1.43), and osmolality (8.3 and 8.7 Osm/kg). The reason is that
1,2-propanediol can only form bridges of the type I, while 1,3-propanediol
forms bridges of the type II.110 Since the modeling of molecular dynamics
can clarify the specific features of interaction between the OCA and the
structural elements of the tissue at the microscopic (molecular) level, it can
be a powerful tool in the study of OC mechanisms, particularly for finding
new, highly efficient OCAs.
OC of hard tissues, such as bone, cartilage and tendon, is of great practical
interest. The reduction of scattering in these tissues offers the possibility to
develop minimally invasive methods of laser diagnostics and therapy of brain
and other deeply located organs. The possibility of optical monitoring of the
cerebral blood flow (CBF) through intact mouse skull at OC was
demonstrated by Wang et al.80 The experiments with laboratory mice
in vivo have shown that the minimal diameter of microvessels that could be
574 Chapter 10
resolved at optical imaging through the optically cleared skull amounts to

14.4 0.8 mm.
The optical clearing of the human skull in vitro was studied by Genina
et al.57 Authors have shown that the exposure of the 5 mm thick skull
sample to glycerol reduces the transport scattering coefficient by 25% in the
spectral range from 1400 to 2000 nm. The main role in the clearing process is
played by the replacement of water in the interstitial space by OCA due to the
specific structure of the bone tissue that possesses a higher porosity than soft
tissues.
The transparency increase was studied in the samples of cartilage tissue
of the human nasal septum and knee joint in in vitro under the impact of
trasograph124 and in the rodent tail cartilage ex vivo under the impact of
glycerol.117,121 The authors demonstrate the increase of the contrast of the
image of the tissue by 2–2.5 fold during 1 min for the transverse section and
by 1.6 fold during 8 min for the longitudinal one124 and the reversible
dissociation of collagen fibrils under the action of OCA.117,121
OC of the mouse tail tendon in vitro exposed to the glycerol solutions with
different concentrations was demonstrated in the papers by LaComb et al.,71
Nadiarnykh and Campagnola,72 and Rylander et al.31 The authors managed
to increase the probing depth in the second harmonic generation microscopy
up to 450 mm at the expense of considerable reduction of the scattering
coefficient (up to 130 fold at the wavelength 890 nm during 5 hrs) and
increased the packaging density of fibrils (according to the data of electron
microscopy, the volume fraction of fibrils increased from 0.65 to 0.9).31
Blood vessels and capillaries transpierce all tissues. As the blood is a
strongly scattering medium, its OC is an important problem for successful
application of optical technologies to both interstitial and intravessel
diagnostics and therapy/surgery. The difference between the refractive indices
of RBC cytoplasm and blood plasma, as well as the specific size and structure
of blood corpuscles, define blood scattering properties.3,10,11 The refractive
index of RBC cytoplasm depends on the concentration of hemoglobin.11–13,125
The volume and the shape of an individual RBC are functions of the
osmolarity of the blood plasma.125,126 The blood scattering is also dependent
on the ability of the RBC to aggregate or disaggregate.127
When an OCA is injected into blood, the refractive index of the blood
plasma increases and becomes comparable with the refractive index of the
RBC. For example, the total attenuation coefficient of whole blood, diluted
by two folds with saline, under the addition of 6.5% glycerol solution
decreased from 42 to 20 cm–1, and, correspondingly, the optical probing depth
at the wavelength of 820 nm increased by 117%. For other OCAs tested
(glucose, dextran, propylene glycol, and trasograph), the increase of the
probing depth was from 20 to 150%.22,128 It was also shown that the
introduction of OCAs is able to change the size and aggregation properties of

RBCs, which allows one to also control the blood optical properties.128
Bashkatov et al.53 have shown that the minimal light scattering is
observed at the concentration of glucose in blood equal to 0.65 g/ml. In this
case, the blood appears to be completely immersed, but due to the difference
in the spectral dependence of refractive indices of the glucose solution and
RBCs the residual scattering is still present. Obviously, such a high
concentration of glucose can be only applied topically and during a short
period of time to prevent the destruction of blood and vascular wall cells.
Nevertheless, the use of endoscopic optical imaging systems (OCT or confocal
microscopy) and the controlled injection of small glucose volumes into the
vascular lumen near the region of imaging may essentially facilitate obtaining
a high-contrast image of an atherosclerotic plaque through the layer of blood.
The same group of authors showed the possibility of using a small volume
of hemoglobin as an OCA. This hemoglobin can be obtained as a result of
local hemolysis of RBC in a blood vessel near the site of optical endoscopic
probe.51 Blood scattering coefficient reduction by 30–40% was theoretically
demonstrated in the spectral region 400–1000 nm under the increase of the
local hemolysis by up to 20% in the immediate proximity of the optical probe.
See also Section 1.10 in Vol. 1.
Impregnation of blood-saturated tissues, such as the liver, with solutions
having different osmolarity also leads to the matching of refractive indices and
scattering coefficient reduction, but the effect is not as pronounced as in
fibrous tissues.22
The optical clearing of tissues in vivo is, generally, more complex,
since in this case a significant role begins to be played by the additional
factors; on one hand, the physiological temperature that accelerates the
diffusion of OCA, and on the other hand, the metabolic reaction of
the living tissue washing the OCA out. These factors can essentially
modify the kinetic characteristics and the magnitude of the clearing
effect.1,25,34,45,59,60,67,69,75,76,78,79,81,97,98,108,112,118,119 In living tissues, the
refractive index is a function of the physiological or pathological
condition of the tissue. Depending on the specific features of the tissue
condition, the refractive index of scatterers and/or the base substance can
change (increase or decrease), and correspondingly the light scattering
will change.1,22 In addition, the introduction of some concentrated
OCAs, in particular, glycerol and glucose, into a tissue affects the
condition of microcirculation in the tissue, causing a transient stasis of
the microvessels.50,64,70,76,108
One of the most important problems in optical diagnostics and treatment
is the reduction of skin scattering aimed at imaging the inhomogeneities
hidden inside the skin or under it. However, it is rather difficult to obtain a
satisfactory OC of skin using noninvasive or minimally invasive means, since
576 Chapter 10
the stratum corneum of the skin epidermis is a natural barrier that impedes the
OCA penetration into the dermis.129 For in vivo applications of the OC
method, one has to use direct exposure of the skin dermis or intradermal OCA
injection. In addition, the low concentrations of OCAs do not provide
sufficient OC, whereas at high concentrations they can induce edema,
chemical burn, partial necrosis and scarring.45,64,108 To develop an efficient
and safe way of breaking the integrity of the epidermis stratum corneum and
accelerating the penetration of OCA into the dermis, various physical
methods,56,118,130–140 chemical enhancers of permeability,113,118,141–147 and
their combinations118,148,149 have been proposed.
OCA diffusion enhancement is implemented by using the agents that serve
as enhancers of tissue permeability in medicine and cosmetology.36,39 In
particular, it is shown that Azone,141 oleic acid,142 DMSO,118,143–146
ethanol,150 propylene glycol,147,151 Thiazone149 considerably enhance tissue
permeability for OCAs and increase the efficiency of OC.
The possible mechanism of diffusion enhancement using Azone on
application to skin could be an increase of the fluidity of hydrophobic regions
of the stratum corneum and the corresponding reduction of resistance to the
penetration of OCAs.141 For example, after 60-min exposure of in vitro skin to
the mixture of 40%-glyercol Azone, the absorption of light at the wavelength
1276 nm increased by 41.1%, and the diffuse reflection at the wavelength
1066 nm decreased by 29.3%, which considerably exceeded the values of
these parameters under the application of a more-concentrated 80%-aqueous
solution of glycerol.141
Oleic acid is a monounsaturated fatty acid that is widely used as a safe
transdermal enhancer for drug delivery. In the region of skin clearing, the
synergetic effect of oleic acid as a promoter of OCA skin penetration was
studied.142
DMSO is a well-known agent widely used for improving the
transdermal delivery of drugs. It is a polar aprotic solvent that is effective
at application to lipids.143 DMSO enhances the permeability of the stratum
corneum for both hydrophilic and lipophilic agents.152,153 In addition, it
possesses a high refractive index and fast permeability, and thus may
serve as an OCA as well.144 It also interacts with the highly organized
structure of collagen fibers, changing the interfibrillar space at the sub-
micrometer scale,145 which is of great importance for the skin OC.
However, the results on the DMSO OC potential and safety are rather
contradictory.106,144,154,155 In Ref. 106, it is shown that the optical clearing
potential of DMSO is smaller than that of glycerol, propylene glycol,
ethylene glycol, and some other OCAs. The application of high-
concentration DMSO to the skin surface causes irritation, accompanied
by epidermal spongiosis.154 On the other hand, it was noticed that DMSO
is a high-efficiency OCA at topical application, and no side effects under
the 20-min exposure of the rat skin surface in vivo were reported.144 In
spite of the controversial evaluation of the DMSO efficiency and safety,
this substance is widely used for OC in mixtures with different
OCAs.63,118,143–146,152,153,156
Ethanol is also a solvent that modifies the skin barrier properties. Under
sufficiently high concentration (40%), ethanol facilitates the formation of
pores and essentially increases the transport of agents through them due to
increasing the size and/or the density of pores in the epidermal membrane.150
The mechanism of the stratum corneum permeability enhancement under
the action of propylene glycol is the solvation of keratin in the process of
water replacement in the binding hydrogen groups and the inclusion of
propylene glycol into the polar heads of the lipid bilayer.151 It was shown that
the mixtures of different OCAs with propylene glycol increase the efficiency of
the OC, however, the clearing effect induced by propylene glycol itself is
weaker than for these mixed OCAs.147
Thiazone is an innovative agent that increases the skin permeability nearly
three-fold more than Azon.36 It is also more efficient as compared to
propylene glycol.36,147
To increase the tissue permeability for OCAs, one can use combinations
of the abovementioned agents. For example, in Ref. 157 for OC of skin, a
combined OCA was used, including glycerol, PEG-300, ethanol and DMSO.
Beside the chemical agents, a number of physical methods were proposed
to overcome the skin barrier, including low- and high-intensity irradia-
tion,130,158 fractional lamp131,133 and laser132 microablation, mechanical
microperforation,137,138 ultrasonic (US) irradiation,134,135,159 electrophoresis,160
needleless injection,161 mechanical removal of the surface layer by means of
abrasive paper,136 epidermal tape stripping,162 and microdermabrasion.163
Different radiation sources (e.g., CO2 and Nd:YAG lasers, operating at
the wavelengths 532 and 1064 nm, respectively, broad-band sources of intense
pulsed light, operating in the ranges 650–1200, 525–1200, and 470–1400 nm)
have been used to irradiate the skin in vivo before OCA application with
different doses and exposures. The measurements of reflection spectra before
and after the exposure have shown that the radiation of an Nd:YAG laser in
the Q-switch and long pulse modes can improve a transepidermal OCA
penetration by 8–9 fold as compared to the intact skin.130 In the other study of
Stumpp et al.,158 a 980-nm diode laser was used for the rodent skin irradiation
using absorbing substrates on the skin surface providing heating of the
stratum corneum and causing a failure of the protective barrier function. After
the laser action and removal of absorbing substrates, the skin surface was
subjected to glycerol application. The OCT study of skin in the exposed
regions has shown the increase of the light-penetration depth up to 42%.
A similar principle was used to create regions of microablation under the
action of a broadband flash lamp. The transparent mask with a set of
578 Chapter 10
Figure 10.3 Images of tattooed skin surface: the tattooed sample before the glycerol
action (a); the tattooed sample after the surface microperforation and glycerol action during
24 hrs (b).133
absorbing carbon centers was applied to the skin surface in order to create
regions of epidermis microdefects under the light absorption.131,133
An increase of the skin transparency and image clearness of the tattoo,
located under the skin at depth of 300–400 mm, was demonstrated by Genina
et al.133 and Bashkatov et al.164 Figure 10.3 presents the tattooed skin images
before and after fractional epidermis thermal ablation and the action of the
88%-glycerol solution.
Fractional laser microablation of the skin surface by means of the erbium
laser (2940 nm) with a pulse energy from 0.5 to 3.0 J is also an efficient tool of
overcoming the barrier for hydrophilic and hydrophobic OCAs.132
The integrity of the epidermis stratum corneum can be disturbed mech-
anically using a roller with multiple needles, often applied in cosmetology.
Yoon et al.137 used this device to create transdermal microchannels in
samples of porcine skin ex vivo with the aim of improving the glycerol
penetration into the depth of the skin. The combination of multiple needle
perforation with a low-frequency ultrasonic treatment allowed one to
increase a delivery rate of the 70%-glycerol solution into the skin by 2.3-
fold in comparison with using the multiple-needle roller alone.138
The low-frequency US (sonophoresis) is one of the noninvasive methods
for improvement of OC, and increases the depth and rate of OCA delivery
into tissues.134,135,165–168 The combined use of OCA and US provided
considerable increase of the depth and contrast of the OCT images of porcine
skin in vitro and human skin in vivo.166 Xu et al.134 used the 1 MHz-US to
enhance the penetration of 60%-glycerol and 60%-PEG-200 solutions into
the porcine skin in vitro. OCT study has shown an increase of the imaging
depth by 40% and 93%, respectively, as compared to using these OCAs
without US.
Cavitation is the main mechanism of the sonophoresis that explains the

tissue permeability increase.159,167 The theoretical analysis of surface
interaction of the cavitation bubbles with lipid bilayers of the stratum
corneum considers three interaction modes, namely, the shock wave
propagation, the penetration of microflux into the stratum corneum, and the
impact of the microflux on the stratum corneum.159 At bubble growing they
can merge by forming larger bubbles that continue growing and creating
channels in the stratum corneum. These channels become larger with time in
the direction towards the boundaries, which finally leads to the appearance of
the inner transport pathways.167 As a result, temporal weakening of the skin
barrier function and enhanced permeation of OCAs and drugs are
expected.168
The transdermal electrophoresis is a well-known and widely used
technique of controlled delivery of drugs through the skin aided by a low-
intensity electric current. The specific features of the technique are: 1) the
molecular transport can be significantly intensified as compared to the
passive diffusion; 2) the delivery rate may be actively controlled by
modulating the electric current density, which allows for individual
dosages.169
The electrophoresis can be applied not only to deliver polar electrically
charged agents. The dependence of electroosmotic diffusion upon the pH of
the solution and the content of NaCl ions in it was studied in Ref. 160. In
these measurements, the solution containing 0.07 M of NaCl and 0.13 M of
D-mannitol in 5мM of citric-acid buffer with pH ¼ 6 was used. Compared to
the passive diffusion, the volume flux of the agent gradually increased
approximately up to 20-fold after switching the current on. After switching
the current off, the volume flux slowly decreased to values exceeding those
before the beginning of the electrophoresis.160
Stumpp et al.161 proposed to use a pistol for needleless injection of
glycerol with concentrations of 100, 50, and 25% into the porcine skin ex vivo.
The removal of the skin surface layer by slight rubbing with an abrasive
paper also helps to enhance the OCA penetration through the stratum corneum
and to improve OC. The abrasive paper with the gritness 220 was used by
Stumpp et al.136 for smooth rubbing of glycerol or dextrose solution into the
depilated hamster skin in vivo. After 2–4 min of rubbing, excluding any visible
damage, the skin gradually became more transparent, allowing for the
observation of subcutaneous blood vessels. The quantitative analysis of the
OCT signals has shown that the light penetration depth increased by 36–43%.
The mechanical removal of outer cell layers of the epidermis can be also
implemented via a sequence of surface epidermal strips. The procedure of
gluing strips of adhesive tape to the skin surface and tearing them off repeated
up to 30 times increases the permeability of skin without any essential
damage.162 In Ref. 139, authors used slides with cyanoacrylate adhesive. The
580 Chapter 10
slide is applied to the skin surface with moderate pressure, kept there
nearly 3 min, and then left for 2 min more without compression. When
removing the slide, the outer layer of the skin epidermis was also removed.
The procedure was repeated 3–6 times until the skin became shiny. Then an
OCA was applied to the processed skin surface under slight pressure, which
also improved the agent penetration into the skin.
The combined use of chemical and physical OCA diffusion enhancers
facilitates further increase of the OC efficiency.118,148,149 Xu et al.148 showed
that the sodium lauryl sulphate, a surface-active agent often used as an
enhancer of penetration of pharmaceutic and cosmetic products into skin, in
combination with the US demonstrates a synergetic enhancement effect on the
penetration of 60%-glycerol solution into the skin. As a result, the optical
transmission and OCT probing depth increased and the time needed for
effective OC decreased in comparison with the US used alone.
The US processing was also used in combination with Thiazone to
improve the penetration of PEG-400 into the skin.149 It was shown that after
the complex processing with Thiazone, PEG-400, and US the diffuse
reflection coefficient decreased by 33.7-fold as compared to the control
measurements (without OC). Using only PEG-400 or PEG-400 with
Thiazone, the reduction of the reflection coefficient was of 2.7- and 3.3-fold,
respectively. The probing depth increased by 41.3% in comparison with the
control samples.
The comparison of the total attenuation coefficients measured for the rat
skin in vivo at application of physical and chemical penetration enhancers was
carried out in Ref. 118. Figure 10.4 presents the result of calculation the value
m mtðcontrolÞ
Dmt ¼ tðtreatedÞ
mtðcontrolÞ 100% of the relative attenuation coefficient change as a
result of the multimodal skin processing with respect to the control
measurements. In this case, the control group consisted of animals not
subjected to the treatment. For this group Dmt ¼ 0. The results allow for
assessment of the efficiency of different OCA diffusion enhancers (the OCA
was a mixture of equal amounts of glycerol and PEG-400). It is seen that
the OCA bar has a minimal height, which means the minimal efficiency of
the 20-min OCA action on the surface of intact skin (OC is very small). The
20-min application of OCA with 9% DMSO appeared to be less efficient in
OC than the 4-min sonophoresis. However, the maximal efficiency of
OC was observed under the combined application of US-DMSO-OCA
during 4 min.
The combined use of surface epidermis stripping with chemical enhancers
of diffusion, such as Thiazone, Azon, and propylene glycol, was studied for
the rat skin in vivo.170 The mixture of an enhancer with PEG-400 was applied
to the skin area that was preliminarily processed by tape stripping. As a result,
the skin diffuse reflection coefficient was reduced, the greatest change being
observed for Thiazone-OCA, then, in descending order, Azon-OCA,
Figure 10.4 The difference between the averaged coefficients of total light attenuation by
the intact rat skin in vivo, calculated from the OCT data. The bar “control” corresponds to the
data, obtained before the OCA action. “OCA” is a 20-min topical action of OCA (glycerol-
PEG-400 mixture in equal proportions); “DMSO-OCA” is a 20-min action of DMSO
(9%)-OCA mixture; “US-OCA” is a 4-min action of low-frequency ultrasound and OCA;
“US-DMSO-OCA”is a combined 4-min action of ultrasound and the DMSO (9%)-OCA (from
the data of Ref. 118).
propylene glycol-OCA, and pure OCA. These results differ from the data of
in vitro studies for porcine skin, for which the optimal combination of OCA
and enhancer was propylene glycol-OCA, and the minimal reduction of the
reflection coefficient was demonstrated by Azon-OCA.147 The found
differences for in vitro and in vivo studies underline the importance of the
tissue physiological response on OC.
The influence of microdermabrasion on the skin permeability for
hydrophilic and lipophilic agents was studied in Ref. 163. Microdermabrasion
is a partial ablation and homogenization of the stratum corneum under the
action of a high-pressure flow of microparticles. Depending on the pressure
and the time of exposure, the skin permeability for hydrophilic agents
increased by 8–24 fold, as compared to the intact skin.
10.3 Compression OC
The in vivo study of external mechanical compression on the optical properties
of tissues is interesting for several reasons. The specific features of light
propagation in a tissue vary depending on its morphologic, biochemical, and
physiological characteristics, therefore, at compression, the spectral reflectiv-
ity of a tissue that carries information on its morphologic and functional
condition can be changed (see Chapter 3).171,172 When the mechanical
compression is applied to a relatively large area of the tissue surface (of the
582 Chapter 10
order of a few square centimeters), the change of its optical properties is due to
better packing density of the scatterers and leakage from the compression site
of absorbers, such as blood and water.
At local mechanical compression, produced, for example, by a fiber tip of
an optical probe with an area of a few square millimeters, a gradient of the RI
is induced at the compression site. This volume of the tissue plays the role of a
lens for the probing radiation, propagating through the tissue. At the point-
wise compression, besides tissue lensing, scatterer ordering and blood/water
displacement also take place.
Since the external mechanical compression can essentially and reversibly
improve optical properties of the tissue, this method can be considered as an
alternative to a widely used immersion OC. As discussed above, the main
mechanisms of the immersion OC are water transport from the tissue and
partial replacement of intercellular or intracellular water with an OCA. Better
scatterer packing and RI matching lead to OC as a result. Analogous
processes partially occur under the compression, since the removal of water
from the interstitial space and the increase of protein concentration in it as a
result of the external compression should also lead to the reduction of
scattering.
The effect of increasing the penetration depth of laser radiation into a
tissue under the application of the external local mechanical pressure was
demonstrated more than 30 years ago.173 Since that time, many papers related
to use of external mechanical compression to control tissues optical properties
(absorption and scattering) have been published.
The first studies were carried out with the tissue samples ex vivo.174,175 In
Ref. 174, the growth of the transmission and the reduction of the diffuse
reflectance in the spectral range from 400 to 1800 nm were found. It was
concluded that at compression, tissue absorption and scattering coefficients
increases, and the possible mechanisms of the OC are the reduction of the
sample thickness and its dehydration. Similar results were also obtained by the
authors of Ref. 175. The compression of porcine skin samples have shown
that it causes the increase of the light transmission through the sample, and
the effect is inertial.
The mechanical compression also changes the optical properties of tissues
in vivo, which is accompanied by the spectral changes both of the diffuse
reflection176–186 and of the fluorescence.185,187–189 The authors of Refs. 176
and 187 were the first who found the impact of the compression on the diffuse
reflection and autofluorescence spectra of the human skin in vivo. It was
reported that the pressure applied to the skin reduces the depth of the dip in
the green region of the reflectance spectrum, which indicates a lesser blood
content in the tissue. The reflection coefficient of skin in the green region
increases, while in the red and yellow regions it decreases. In the reflection
spectrum, an isobestic point was found for which the reflection coefficient
does not change at compression.176 The external compression also caused an

increase of the skin autofluorescence intensity in the short-wave range of the
visible spectrum.187
The reduction of diffuse reflection in the range 1100–1700 nm was
reported for human skin in vivo.178 It was shown that under the increase of the
pressure of the fiber tip of the optical probe on the skin surface, the diffuse
reflection is decreased. The authors related such behavior with the change of
the internal structure of the skin tissue. A similar fact was also found in the
visible range of the spectrum.184
In Ref. 185, the influence of a short-time (less than 2 s) and a long-time
(more than 30 s) mechanical compression on the diffuse reflection and
autofluorescence spectra of human skin in vivo was studied. It was shown that
at high pressure, the significant spectral changes occur both for short-time and
long-time compression—the probe pressure affects not only the optical but
also the physiological parameters of the skin.
The influence of external compression on the physiological parameters of
breast was studied by the authors of Ref. 180. Under the compression of the
tissue, the light scattering was reduced, and the total blood content and the
saturation of the tissue with oxygen were also decreased. After the removal of
compression, the effect of hyperaemia was observed. The changes in the blood
circulation system (compression of blood vessels and reduction of blood
oxygenation), in the opinion of the authors of Ref. 181, can enhance the
sensitivity and specificity of early diagnostics of cancer.
At present, external mechanical compression is used as a method that
allows one to increase image resolution and contrast of OCT and
microscopy.28,190–193 The method of tissue-compression OC (COC) was
implemented instrumentally (tissue optical clearing device, TOCD).29,31,194–196
The mechanical compression of skin allows for assessment of the content
of tissue chromophores, whose absorption under the normal conditions is
masked by the absorption hemoglobin or water. Thus, the extrusion of blood
from the region of compression makes it possible to evaluate the content of
carotenoids in the skin from the spectra of diffuse reflection,197 and the
content of melanin in the skin from the fluorescence spectra.171,172
Although the number of publications related to the mechanical
compression of tissues is relatively large, the results are often somewhat
controversial, which is mainly due to the difference in the conditions of the
compression force application (locally or distributed) and the detection of the
reflected light (a fiber-optical or an open detecting system).198 Besides, it is
necessary to take the inertial character of the tissue response to the external
action into account.
The COC method has a number of potential advantages in comparison
with the immersion method. For example, moderate mechanical compression
saves the barrier functions of the stratum corneum and the living epidermis.
584 Chapter 10
In the visible region, the absorption coefficient of water is insignificant as

compared to the NIR, where the absorption peaks caused by the combina-
tions and overtones of fundamental vibrations are located. In the skin the
absorption spectral bands of water bound with the proteins of the interstitial
matrix are shifted with respect to the bands of free water towards the long-
wave region of the spectrum. Thus, the manifestations of bands in the
reflection spectrum of skin at 1160 nm and 1220 nm were attributed to the
absorption bands of free and bound water.199 The absorption bands at
1879 nm and 1890 nm were attributed to the absorption of free water, and the
bands at 1909 nm and 1927 nm to the bound water.200
The effect of water transport under compression was studied by Rylander
et al.196 In the skin, the natural gradient of water content depends on the
depth. In the stratum corneum the water content depends upon the air
humidity and can be at the level of 15%. The water content grows with depth,
reaching 70% in dermis. A considerable part of this amount is free water, the
molecules of which due to their high permeability percolate from the region of
external compression into the adjacent regions. The water migration from the
skin regions subjected to compression leads to the matching of the RIs of the
tissue components,31 which reduces the scattering coefficient.
The water transport in the skin tissue can be analyzed on the basis of a
two-phase nonlinear mixture,201 according to which the skin is presented as a
solid elastic matrix, formed mainly by elastic fibrils and cells and filled with
water. This model is sufficient to present the composition and mechanical
properties of the skin well enough.196,201 Based on this model, the authors of
Ref. 193 studied the influence of mechanical compression on the optical
properties of the skin. The samples of porcine skin ex vivo and human skin
in vivo were studied using OCT at the mean wavelength l ¼ 1310 nm. It was
found that for the porcine skin samples ex vivo, the compression reduces the
water volume fraction in the tissue by more than 3-fold (from 0.66 to 0.2).
The detailed analysis of possible mechanisms affecting the skin optical
properties was carried out in Ref. 199, where the influence of human skin
compression on the variation of its scattering properties and the free and
bound water content were studied. The changes in the diffuse reflection
spectra were correlated with the effects of tissue deformation within the
framework of a two phase nonlinear mixture model. The optical properties
of the skin were considered, based on the three-layer medium (epidermis,
dermis, subcutaneous fat), and the scattering properties of the tissue were
determined as the combination of scattering by large (Mie scattering) and
small (Rayleigh scattering) particles. The expressions for the scattering
coefficients of each of the skin layers were taken from Refs. 202, 203, and
the skin absorption in the NIR was determined as the superposition of
absorption by water and lipids. The water contribution into the absorption
amounted to 20% for epidermis, 70% for dermis, and 60% for subcutaneous
tissue (fat and muscle).203,204
Experimental diffuse reflection spectra were analyzed for two absorption
lines of water in the NIR region, 1160 nm and 1220 nm, attributed by the
authors to the absorption of free and bound water, respectively, basing on the
comparative analysis of the reflection spectra of skin and pure water.
The measurement of water content in the skin in vitro has shown that
under compression, the intensity of the band at 1220 nm significantly grows,
while the intensity of the band at 1160 nm decreases. When the pressure is
increased to 376 kPa, the absorption peak at 1160 nm almost vanishes. This
indicates the fact that under the compression of skin, the free water leaves the
compressed region, while the bound water that forms complexes with proteins
stays. When the pressure reaches 400 kPa, the volume of free water decreases
to 30% of its initial value. This is the main cause of the tissue deformation.
Under long-time compression, the absorption peak of free water decreases
with time, and the peak of bound water increases. At the early stages of
the deformation due to the large stress in the tissue, the free water leaves the
compressed region with a sufficiently high rate. As the stress relaxes, the
deformation of the tissue and the rate of free water migration gradually
decrease. After nearly 6 min, the tissue deformation and the water transport
stop. The duration of this process is also related to the compression
magnitude.
Thus, for the analysis of the structural and optical changes occurring in
the skin in vivo under the conditions of external compression, the above model
of the tissue can be considered, according to which the skin consists of a solid
matrix formed by the collagen fibers and the interstitial fluid, the major part
of which is water. When the skin is compressed, the solid matrix is deformed
and the interstitial fluid leaves the region subject to compression with the rate
determined by the density of packing of the matrix fibers and the viscosity of
the fluid. This hypothesis is in good agreement with the experimental data.
The relation of skin deformation and relaxation time under the conditions of
compression depends upon the elastic properties of the fibrils and the density
of their packing, as well as upon the amount and viscosity of the fluid in the
tissue. Since the bound water is integrated in the tissue solid matrix,
the measurements of free and bound water content variations can reflect the
deformation of the solid matrix and the transport of free water in the tissue
under compression.
10.4 Photochemical, Thermal, and Photothermal OC

For a number of applications, particularly in the case of laser therapeutic or
surgical interventions, one can control the optical properties of the tissues by
means of photochemical or photothermal action. The heating and coagulation
586 Chapter 10
of a tissue change its optical properties, and this should be taken into account
in the calculation of the radiation dose in the process of treatment.3,22 The
method of controlling the optical properties of fat tissue, which is usually a
strongly scattering tissue, was proposed in Refs. 205 and 206. The method is
based on the combined photochemical and photothermal effects2,207 induced
in the fat tissue cells.
The selective photothermal action on the fat tissue can be implemented
by the choice of the laser radiation wavelength corresponding to the
absorption band of the endogenous tissue chromophores. The lipids
contained in the fat droplet of adipocytes exposed to light with a wavelength
of 1210 nm serve as tissue heaters.208 However, with exogenous chromo-
phores, e.g., the indocyanine green (ICG), the efficiency of light interaction
with tissue may be essentially higher and more selective.209–213 The intensity
and shift of ICG absorption bands depend upon the solvents used and
interaction with biological molecules.214–216 The solutions with complex
compositions, such as water-alcohol-glycerol, stabilize the ICG absorption
bands.215 Besides that, the alcohol component of the solution makes the cell
membrane much more permeable for the dye.217–219 The next step of the
interaction between the dye and the cell is related to the light action. The
light causes photochemical reactions; for ICG they simultaneously follow
two scenarios: 1) photodynamic reactions; 2) reactions yielding toxic
products.216 Depending on the intensity, the biological response of the cell
may lead to reversible or irreversible injury of the cell membrane. Reversible
membrane injury is associated with the creation of new pores or enlargement
of the already existing ones, which facilitates efficient exchange between the
cell content and the environment. In fat cells, the presence of pores increases
the lipolysis, the result of which is that the intercellular space becomes
filled up with the cell content and decay products (triglycerides, fatty acids,
water and glycerol).37,220–223 The appearance of such immersion fluid in
the intercellular space facilitates the process of optical clearing of the
adipose tissue.37
The probability of triglyceride leakage through the cell membrane pores
is rather small due to the relation between the pore diameter (0.1–2 nm) and
the triglyceride molecule size (nearly 1–2 nm).221–223 The hydrolysis of
triglycerides occurs in two stages. At the first stage, the hydrolysis of
external complex ether bonds occurs; this process is catalyzed by the lipase
enzyme. The hydrolysis of triglycerides is referred to as lipolysis. The
monoglyceride produced at the first stage of the triglyceride decay is further
hydrolyzed by the nonspecific esterase, producing glycerol and three
molecules of higher fatty acids.221–223 The hydrolysis products can easily
leave the cell through the pores, since their molecular size is small compared
to the pore size. In this process, the refractive index of the intercellular
fluid (nм ¼ 1.36)45 approaches the refractive index inside the adipose cell
Figure 10.5 Transmitted light images of a subcutaneous fat tissue sample stained with
brilliant green before (a) and after (b) – (e) light exposure of 15 min of a two-wavelength band
diode lamp (442 nm and 597 nm) with a total power density of 75 mW/cm2. The time interval
between the image recording and the end of light exposure is 8 min (b), 19 min (c), 39 min
(d), and 120 min (e). The sample temperature was 33°C.
itself (nв ¼ 1.44), the medium becomes more optically homogeneous and, as a
result, more transparent.22
The enhanced porosity of the adipose cell membrane at photodynamic
action (PDA) on cells in adipose tissue sensitized by brilliant green (BG) dye
was studied in vitro.37,220 As a result of induced cell porosity, their lipolysis
was observed. The hypothesis of PDA induced pore formation is in good
agreement with the experimental results.37,206,220,224
Figure 10.5 presents the transmitted light images of the subcutaneous fat
tissue sample, stained with BG, before (a) and after (b)–(e) the irradiation with
a dental diode lamp Ultra Lume Led5 with wavelengths of 442 and 597 nm
and total power density of 75 mW/cm2 during 15 min at the constant
temperature 33°C. The cell lipolysis is seen to stimulate the leakage of a part
of the intracellular fluid into the intercellular space and the tissue gradually
undergoes immersion clearing.
10.5 Applications of Optical Clearing

10.5.1 Optical coherence tomography
OCT is a noninvasive method of imaging that allows one to study tissue
internal microstructure with a high resolution and without disturbing its
integrity.18 However, the multiple scattering essentially worsens the OCT
imaging properties, namely, the image resolution, probing depth, and the
precision of localization.21 In this connection, the progress of optical
clearing techniques is of huge potential for noninvasive OCT medical
diagnostics.28,29,35,36,39,190,192,196
The efficiency of mechanical compression for better differentiation of
chronic inflammation and carcinoma in the OCT images was studied by
Agrba et al.28 They also determined the threshold impact that allows the
detection of a difference between these cases. Using the endoscopic probe of
588 Chapter 10
unique construction that provided control of the compression force applied to

the tissue, the possibility of differentiating the pathologic changes of the
rectum ex vivo with inflammation and carcinoma was demonstrated. The
same research team has shown that the application of compression to human
skin causes various changes of the optical properties of skin layers depending
on their elasticity.192 These changes increase the contrast of the OCT-imaged
boundaries between the layers.
The efficiency of the mechanical device for tissue optical clearing,
designed by Rylander et al.196 was determined using OCT.29 The authors have
shown that the depth of light penetration into human skin in vivo at the
wavelengths 850 and 1310 nm increase by three-fold in the area of com-
pression. The contrast of the OCT image of the epidermis-dermis junction was
also increased. OCT allowed the specification of the compression clearing
mechanism. Thus, the OCT M-scans recorded during the pressure application
have shown that the optical penetration depth monotonically increases,
undergoing a jump at the initial time period (5–10 s), during which the water
content in the tissue abruptly decreases and the group refractive index
increases in the local area of pressure application.
The authors of Ref. 191 proposed to use OCT with compression for
continuous noninvasive monitoring of blood glucose content. The results of
in vivo skin studies at different pressures have shown that the pressure of the
OCT probe on the skin reduces the artifacts related to the object motion, but
may shift the OCT signal slope that is an informative characteristic for blood
glucose monitoring. Thus, the controlled external pressure ,1 kPa signifi-
cantly improved the precision and reproducibility of the glucose OCT
monitoring.
The potentialities of immersion optical clearing are mainly related to
the fact that OCT allows the imaging of relatively thin layers of tissues
that can be rapidly impregnated with OCA applied to the surface. In
Refs. 21 and 22, it was demonstrated that the internal tissues, such as the
walls of blood vessels, esophagus, ventricle, large intestine, and other
organs, can be imaged to the depth of 1–2 mm. The application of
immersion liquids increases the probing depth by nearly 3.5-fold and
significantly improves the image contrast of the internal inhomogenei-
ties.225,226 The possibility of diagnostics of malignant melanoma, observa-
tion of subepidermal cavity, and control of the skin scattering properties
under the topical application of glycerol and propylene glycol are examples
of OCT imaging of human skin in vitro and in vivo.22 Moreover, the
variation rate of the OCT A-scan slope in the course of the optical clearing
allows one an additional differentiation of the healthy and tumor tissue of
the human esophagus and mammary gland.103–105,227,228
OCT offers a unique possibility to assess the variation of the tissue light
scattering coefficient with sufficient depth resolution. The total light
attenuation coefficient mt of the tissue within a scanned line, which is a sum of

the absorption coefficient ma and the scattering coefficient ms: mt ¼ ma þ ms,
may be found by fitting the parameters of the approximating curve, calculated
within the framework of the appropriate model, using the local A-scan slope
of the OCT signal.18,118,232,229
The single scattering model is based on the assumption that only the light
undergoing single scattering (ballistic photons) preserves the coherence
properties and contributes to the OCT signal. In this case the OCT signal
i(z) is determined as18,230,231 (〈i2(z)〉)1/2 (〈i2〉0)1/2(exp(2mtz))1/2, where z is
the depth from which the reflected signal comes.
It is known that OCT provides measurement of the reflected signal
intensity from a sample, R(z) ∝ (〈i2(z)〉)1/2, upon the depth z. The OCT signal
intensity depends on the reflectance a(z) of the tissue at the given depth and
the total attenuation coefficient mt of the tissue. In accordance with the single-
scattering model that is valid both for weakly scattering tissues and for the
surface layers of strongly scattering tissues,231 the reflected power is
proportional to exp(mtz),18 i.e., can be approximated with the expression:118
R(z) ¼ Aexp(mtz) þ B, where A is the coefficient equal to P0a(z), with P0
being the optical power of the beam incident on the tissue surface, a(z) being
determined by the local capability of the tissue to scatter the light in the
backward direction (to reflect), which is largely dependent upon the local
value of the refractive index, and B being the background signal.
Since the absorption coefficient is essentially smaller than the scattering
coefficient for many tissues in the NIR, the exponential decay of the ballistic
photons is mainly determined by the scattering coefficient.18 The scattering
coefficient depends on the mismatch of refractive indices in the tissue volume,
so that the increase of the refractive index of the interstitial fluid and the
corresponding reduction of scattering is recorded as the change (decrease) of
the slope of the OCT signal amplitude as a function of the probing
depth.18,35,89,90,93,95–100
This method allows for measurement of the permeability coefficient of
tissues both for OCAs and for the medicinal preparations, possessing the
properties of OCA, using two approaches: measuring the OCT signal slope
(OCTS) and the OCT signal amplitude (OCTA) [Fig. 10.6(a)].35,95,232 In the
OCTS approach, by analyzing the variation of the OCT signal slope due to
the OCA diffusion, one can calculate the mean permeability coefficient of a
certain region in the tissue. In this case, two-dimensional OCT images are
averaged over the lateral coordinate (x axis) to get a one-dimensional depth
distribution of the OCT signal intensity. In the tissue it is necessary to choose
the region for which the signal is linear and its fluctuations are minimal, where
the tissue thickness (zregion) must be measured. Then the monitoring of the
agent diffusion in the chosen region is performed and the time of diffusion
(tregion) is recorded. The mean permeability coefficient ðPÞ can be calculated as
590 Chapter 10
70 1.0
OCT Signal Slope, arb. un.

Cortexin diffusion
OCT Signal, arb. un.
60 0.8
OCTA Cortexin added

0.6
50
averaged signal
0.4
40
OCTS
0.2
30
0 100 200 300 400 500 600 700
0 5 10 15 20 25 30
Depth, μ m Time, min
(a) (b)
Figure 10.6 The region and the depth, indicated in the averaged OCT signal, used in
OCTS and OCTA approaches (a), and the time dependence of the OCT signal slope on
the process of interaction of the sclera with the cortexin solution (b) (from the data of
Ref. 93).
the thickness of the region divided by the time, during which the diffusion of
zregion
the agent through the chosen region occurs:232 ðP ¼ tregion Þ.
The OCTA approach can be used to calculate the permeability coefficient
at a definite depth in the tissue: P(z) ¼ zi/tzi, where zi is the measurement depth
and tzi is the time of agent diffusion to the given depth. The value of tzi should
be calculated from the time of the agent application to the time of the
beginning of the OCT signal amplitude variation, caused by the OCA.232
These approaches allowed the assessment of the following permeability
coefficients: rabbit cornea for mannitol (8.99 1.43) 10–6 cm/s [95], cm/s,95
rabbit sclera for mannitol and 20%-glucose solution (6.18 1.08) 10–6 and
(8.64 1.12) 10–6 cm/s, respectively,95 human sclera for cortexin (2.40
0.32) 10–5 cm/s [Fig. 10.6(b)],93 porcine aorta for 20%-glucose solution
(1.43 0.24) 10–5 cm/s,99 epidermis and dermis for 40%-glucose solution
(6.01 0.37) 10–6 cm/s and (2.84 0.68) 10–5 cm/s, respectively,97 lung
tissue for 30%-glucose solution (1.35 0.13) 10–5 cm/s in the norm and for
different malignant neoplasms.100
The depth-resolved analysis of the tissue optical properties using OCT
allows one to reconstruct the two-dimensional diffusion maps. The visual
representation of the molecular diffusion front was first demonstrated by
Ghosn et al.97 in the course of noninvasive determination of the penetration
rate of 20%-glucose solution into the rhesus macaque skin in vivo.
Using the OCT methods, one can reveal the variations of the effective
refractive index of the adipose tissue as a result of the light photodynamic
action, which arise both directly after the exposure, and in the process of the
systematic biological response.38,224 The observed changes of effective
refractive index can be interpreted as the reduction of the relative refractive
index of the scatterers, which can be related to the immersion optical clearing.
Figure 10.7 OCT images of adipose tissue stained with BG before the irradiation (a),
immediately after 15-min irradiation (b), at 60 min (c), and at 120 min (d) after the
irradiation. The source was the dental diode lamp Ultra Lume Led5 (442 and 597 nm,
75 mW/cm2). The dye concentration was 6 mg/ml. The sample temperature was 37 °C.
The curves correspond to A-scans of the OCT image, averaged over the whole B-scan
region.224
Figure 10.7 shows the series of OCT images of adipose tissue stained with
BG before (a) and after (b-d) the irradiation with the light of a diode lamp
during 15 min and under the heating of the samples to the physiological
temperature. Figure 10.7(b) shows the image of the adipose tissue immediately
after the irradiation, while the OCT images in Figs. 10.7(c) and (d) correspond
to relatively long observation times, 60 and 120 min, respectively. The mean
thickness of the samples was 237 10 mm.
The visual comparison of the OCT tomograms of the adipose tissue shows
that after the irradiation, the tissue structure considerably varies in time. One
can easily see the changes of the outer layer of the cellular structure
that becomes more homogeneous, which can be associated with partial
destruction of the cells and formation of an immersion layer consisting of
intracellular content and the products of the adipose component hydrolysis. In
Fig. 10.7(b)–(d), this layer is 80 to 90 mm in thickness. The mean size of
adipocytes in the tissue amounts to 60 to 70 mm.
If the geometric thickness of the studied layer is known, the effective
refractive index n is calculated as n ¼ z/l, where l is the true sample thickness in
mm, and z is the layer thickness observed in the OCT, i.e., the optical path
length in mm. The geometric thickness was measured using a micrometer
gauge. The optical path length was found from the OCT signal as the
difference between the depths of two peaks, corresponding to the sample
592 Chapter 10
boundaries (see Fig. 10.7). For better determination of the boundaries, the
A-scans were averaged over the lateral region of 2 mm. This operation
smooths out the stochastic noise and the disordered cell structure of the tissue,
while the peaks, corresponding to the sample boundaries, become more
distinct. The calculation of the refractive index showed its monotonic decrease
with the increase of observation time.
The observed changes in the OCT images can be interpreted as the result
of lipolysis and cell destruction at the surface of the sample due to the
photochemical effect. The intracellular fluid leaks from the cells and fills the
intercellular space, as a rule, containing the intercellular fluid, thus producing
the cleared (more homogeneous) outer layer and giving rise to optical clearing
inside the tissue by matching the refractive indices of the cells and intercellular
medium.
In Ref. 233 the chemically induced lipolysis was assayed by measuring the
amount of glycerol released from the cells into the surrounding medium. In
addition, using frame-by-frame CARS (coherent anti-Stokes Raman spec-
troscopy) processing of a single living cell, the authors could control the
morphological changes of lipid droplets in the cells. They found that the
microscopic lipid droplets appeared gradually in the course of the experiment,
achieving a size of nearly 1 mm 60 min after the impact. All these observations
allow for the assumption that analogous processes may occur in the case of
light-induced lipolysis.
The results of OCT measurements of the refractive index of adipose tissue
after the exposure to light can be explained based on the idea of forming two
layers over the course of time. One of these layers is cleared (the released
intracellular fluid) and the other one is scattering (cells). The refractive index
m of the scattering medium can be expressed as128
n2 n2
m ≡ n0 þ Dm ¼ n þ Qðl∕lÞ, (10.1)
n
where

n Vs
n ¼ n0 1 þ 1 , (10.2)
n0 V0
where Vs is the total volume of scattering particles, V0 is the volume of the

scattering medium, n2 is the mean square value of the refractive index
fluctuations, Q(l/l) depends upon the shape and aggregation of scatterers, and
l is the correlation length of randomly distributed fluctuations of the refractive
index. Q ¼ 1.17 for the limit case of large correlation length, l ≫ l (large
particles). In the process of the adipose tissue degradation the two-phase
system appears, consisting of cytoplasm of the adipose cells (the cleared outer
layer in Fig. 10.7(b) and the adipose cells themselves. Each phase of the
medium has its own thickness and refractive index. Denoting the time-
dependent thickness of the layer of adipose cells as H(t), and the thickness of
the outer layer of transparent cytoplasm as [L H(t)], the mean refractive
index of the layer with the thickness L, containing two layers with different
refractive indices, can be written as
½L HðtÞ HðtÞ
mT ¼ nþ m, (10.3)
L L
where m and n are defined by Eqs. (10.1) and (10.2). Since m is always
greater than n due to the effect of scattering and tissue degradation, this
leads to the decrease of H(t), and the summary refractive index should
decrease with time.
In the process of adipose tissue degradation and formation of a greater
amount of lipolysis products, the conditions of refractive index matching can
be implemented, which will lead to a further reduction of light scattering, i.e.,
the part of the refractive index m, depending on these processes. Both of these
mechanisms, the tissue degradation and the refractive index matching, explain
the experimentally observed reduction of the mean refractive index of the
sample layer, i.e., the reduction of its scattering capability over the course of
time.
10.5.2. Optical projection tomography

Optical projection tomography (OPT) is a new approach to 3-D imaging of
small biological objects. It fills the gap between the MRI and confocal
microscopy (CM), being the most suitable for studying objects with the
size from 1 to 10 mm.234 For example, the possibility to analyze
the organization of a tissue in 3-D is inestimable for understanding the
embryonal development, in which the tissues undergo a complex sequence
of transformations.235 Since the OPT allows for registration of both the
absorption and emission profiles, this method offers an opportunity to use
different staining techniques developed to observe the spatial distribution of
gene activity.234,235 OPT also helps to analyze normal and abnormal tissue
morphology and to determine the position of labelled cells within a
tissue.235
Very often, OPT imaging implies the suspension of the object in the
immersion liquid in order to decrease the refractive index nonuniformity all
over the sample and, therefore, to reduce the surface and internal light
scattering. This means that the light passes through the sample along nearly
straight trajectories (ballistic photons), and the standard algorithm of back
projection can generate images with high resolution.234 Alanentalo et al.236,237
have shown that the OPT allows for successful construction of 3D images of
specially labelled structures inside the optically cleared organs of a mouse
594 Chapter 10
Figure 10.8 Single image from the OPT scan datasets with the selected regions of interest
(insert) (a) and volume reconstruction of the pancreas based on the background
autofluorescence (dark gray) and the signal from insulin-specific antibodies (white islets)
with the regions of interest (insert) (b).237
(see, Fig. 10.8). As an OCA «Murray’s clear» (1 part of benzyl alcohol and 2
parts of benzyl benzoate (BABB)) with the refractive index 1.55 was used.
10.5.3 Fluorescence imaging

The methods of fluorescence spectroscopy and microscopy belong to
noninvasive methods of biomedical diagnostics. Fluorescence spectra often
provide detailed information about the fluorescent molecules, their con-
formations, sites of bonding, and interaction with cells and tissues.238 Under
excitation by ultraviolet light (l ≤ 300 nm), one can observe the autofluor-
escence of both the proteins and the nucleic acids. The porphyrin molecules
are also endogenic phosphors, the content of which in cells considerably
increases under some pathologic conditions. In addition, some bacteria, which
cause, e.g., the inflammation diseases of skin239 and gums,240 and inhabit the
caries tooth lesions,241 can accumulate considerable amounts of porphyrins.
Combining the measured autofluorescence spectra with those of reflection and
absorption, OCT and other optical methods allows for diagnostics of caries
and tooth tissue demineralisation,241–245 skin cancer,246,247 and precancer
changes of cervical tissue,248 mammary gland,249 and other diseases.
Thanks to the growing number of fluorescent dyes specifically staining
biological molecules (nucleic acids, proteins) or cell organelles, the methods of
fluorescence microscopy acquire greater and greater significance.238 At
present in different biomedical studies, confocal microscopy (CM) is widely
used, which allows for imaging of the internal structure of tissues at the
cellular and subcellular levels.250,251 The CM illuminates a small volume
inside the object and detects the scattered fluorescence light from the same
small volume. The main CM advantage as compared to the methods of
common microscopy is the capability to obtain the optical cross sections of

thick samples.251 CM allows for creation of high-quality (with micron spatial
resolution) images of cell layers. The high contrast and spatial resolution of
CM images are due to probing of a small volume of the medium, limited by
the size of the central focal spot formed by the focusing optical system. The
main limitation of CM in the skin studies is the considerable scattering that
spoils the quality of cell images. The increase of the transparency of the outer
layers of skin improves the light penetration depth, the contrast, and the
spatial resolution of the CM.
Meglinskii et al.,48,49 by using the Monte Carlo modeling, proved the
possibility to considerably increase the human skin probing depth for
reflectance CM by the reduction of spatial fluctuations of the refractive index
of the outer skin layers by optical clearing. Later Dickie et al.,82 described the
experimental method that allowed for imaging of microvessels of mouse
tissues using the CM at the depth up to 1500 mm below the sample surface due
to the optical clearing of thick slices of the tissue. As an OCA for CM
imaging, the commercially available multicomponent agent FocusClear™
was used,252 especially developed for both fluorescent and non-fluorescent
light microscopy. FocusClear™ does not give rise to the tissue dehydration
and allows the improvement of the obtained image quality. The refractive
index of this agent is 1.46. FocusClear™ was successfully used for CM
imaging of murine rectum and ileum,253,254 the wall of the human ileum,255
the murine pancreas,256 and the brain of the insect (Diploptera punctata)257 in
vitro. The optical clearing facilitated the identification of spatial and temporal
changes of crypt morphology of the murine large intestine with colitis, as well
as the detection of transgene fluorescent proteins, expressed in the large
intestine and ileum.253 The increase of resolution and imaging depth of the
microstructure and vascular system of the ileum was observed under the
combined use of blood vessel staining and optical clearing.254 The authors of
Ref. 255 declared that by using OCA, the thickness of the studied samples can
be increased by 80-fold, which allows for better reproduction of the structure
and the vascular system of the tissue. The use of OCA allowed for 3D
reconstruction of the microvascular architecture of the insect brain with sub
micron resolution to the depth of 1500 mm.257
However, the imaging depth of 500–1500 mm is not sufficient for 3D
reconstruction of neuron networks of the murine brain as a whole. Hama
et al.84 developed an OCA, called Scale, which not only increases the
transparency of a sample, but also does not reduce the intensity of
fluorescence signals from the cleared structures.
Dodt et al.85 used the idea of illuminating the object with a so called light
sheet, which allows for observation of macroscopic objects, such as the brain
as a whole, with micrometer resolution. The sample is illuminated from two
sides with the laser radiation in the blue part of the spectrum, forming a thin
596 Chapter 10
planar light beam. The fluorescence signal is, therefore, emitted only from a
thin optical section and is collected by the objective lenses. The declined light
is blocked by a filter, and the image is projected through a cylindrical lens
onto the screen of a digital camera. Since all sample parts below or above the
light sheet are not illuminated, the emission is not excited beyond the limits of
the focus and, therefore, is not to be eliminated from the useful signal. This
approach is referred to as ultra-microscopy (UM). In comparison with CM or
two-photon microscopy, UM can rapidly create optical sectioning of
macroscopic samples, since the objectives with small focal power and small
numerical aperture can provide a large field of view. However, the use of UM
is completely dependent upon the optical transparency of biological objects,
so that the improvement of optical clearing efficiency, including that due to
the development of new OCAs, remains an urgent problem.85,86,258–262 In
Ref. 258, it was shown that the use of optical clearing allowed for cellular
resolution in sectioning of the fixed mouse brain and detecting individual
neurons, labelled with green fluorescent protein, in the removed murine
hippocampus. In the isolated hippocampus the 3D images of dendrite
structures and processes of neuron populations were obtained. The use of UM
includes the imaging of cleared murine organs and total embryo specimens,
adult insects Drosophila, and other fixed tissues with the size of a few
millimeters.259
Ertürk et al.86 found that tetrahydrofurane (THF) in combination with
BABB can completely clear the spinal cord, preserving its fluorescence.
Recently Ertürk et al.262 developed a new protocol of optical clearing using
the dibenzyl ether instead of BABB in combination with THF to get three-
dimensional images of neurons of the whole mouse brain using the UM. This
method was called 3-D imaging of solvent-cleared organs (3DISCO). In this
study, the clearing of the tissue took only three hours, and the imaging was
performed during 45 min.
Yang et al.263 presented techniques for tissue clearing in which whole
organs and bodies were rendered macromolecule permeable and optically
transparent, thereby exposing their cellular structure with intact connectiv-
ity. A passive clarity technique (PACT) and corresponding protocols for
quicker passive lipid extraction and immunostaining of intact organs were
described. To image PACT-cleared thick tissue, a refractive index matching
solution (RIMS) with an outcome similar to FocusClear™ was designed.
For whole body clearing and immunolabeling, a methodology based on the
method of perfusion assisted agent release in situ (PARS) was proposed to
facilitate fast, whole-brain and whole-body clearing using systemic or
cerebrospinal circulation to directly deliver clarifying agents. These
methods are applicable for high-resolution, high-content mapping, and
phenotyping of normal and pathological elements within intact organs and
bodies. Figure 10.9 illustrates how PARS enables whole-brain mapping
Figure 10.9 PACT/PARS/RIMS whole-brain mapping of widespread and sparse geneti-

cally encoded fluorescent signals with subcellular resolution.263 Whole-brain image (z ¼
6 mm) of adult Thy1-eYFP mouse after clearing for 10 days. The boxes on the right show
high magnification images of indicated areas6HHFRORUSODWHV
of widespread and sparse genetically encoded fluorescent signals with

subcellular resolution.
These in vitro studies showed that probing depth, image contrast, and
resolution of many optical imaging methods can be significantly improved,
even a whole-organ or animal body can be imaged using comprehensive
optical clearing protocols and corresponding optical imaging facili-
ties.84–86,254–268 Despite recent achievements, which are extremely valuable
for biology, especially for neural science, such clearing protocols are not
suitable for in vivo tissue examination. For in vivo applications, the immersion
optical clearing method must provide a rapid treatment, during seconds or
minutes, sufficient transparency of tissue layers, and safety for animals and
humans. All those make the design of fluorescence optical clearing technology
for in vivo use a challenging and innovative problem, the solution of which
has a great potential for the enhancement of contrast and resolution of
fluorescence optical imaging.1,22,48,49,269
10.5.4 Photoacoustic imaging

Immersion optical clearing is also beneficial for improvement of absorption-
based optical imaging technologies, such as photoacoustic (PA) micro-
scopy.1,114 Optothermal effects occur in a tissue due to its interaction with
pulsed or intensity-modulated optical radiation. Such interaction gives rise to
a number of thermoelastic effects in the tissue, in particular, the generation of
598 Chapter 10
acoustic waves. Detection of the acoustic waves is the basis of the

photoacoustic (PA) method.1,3 The main fields of PA applications are
imaging of strongly absorbing objects, such as blood supply or melanin
pigmented tumors270 and blood vessels.271,272 This method can also be used
for imaging of subcutaneous structures273 and mapping of blood oxygen-
ation.3,271 The quality of the PA signal depends upon the penetration depth of
the optical signal and the damping of the acoustic signal.
The optical clearing technologies allow one to increase tissue transpar-
ency, which facilitates the reduction of light scattering and increases the depth
of the detected inhomogeneities. On the other hand, the OCA flow into and
water out of the tissue may change the composition and structure of the tissue,
thus change the acoustic impedance and absorption and, therefore, the
acoustic wave damping. Recently it was demonstrated that the optical
resolution of PA microscopy can be significantly increased using the aqueous
solutions of glycerol as an OCA.274 Liu et al.114 studied several OCAs in vitro
and in vivo, and showed that PEG-400 in combination with Thiazon improves
the images of deep blood vessels of the skin, while glycerol is better for
imaging small branching blood vessels. In this case, the immersion time
should be controlled to prevent the reduction of the signal amplification.
Based on the amplification of the PA signal at the expense of a reduction
of scattering, this method can be also efficiently used for noninvasive
monitoring of glucose content in blood and tissues, which was demonstrated
for phantoms with in vitro and in vivo studies.275,276 By means of PA
spectroscopy, glucose can be detected by measuring the interval between the
peaks of the pressure waves induced by the laser radiation.276 The maximal
relative change in the PA response was observed in the region of the second
C–H overtone at wavelength 1126 nm with the additional peak in the region
of the second O–H overtone at wavelength 939 nm.275 In addition, the
analysis of the generated pulsed temporal profile of the PA signal allows for
glucose detection due to tissue scattering change, which decreases with the
growth of glucose concentration.22,277
Clinical applications of PA flow cytometry (PAFC) for detection of
circulating tumor cells in deep blood vessels are hindered by laser beam
scattering that results in loss of PAFC sensitivity and resolution.278,279 The
rapid and efficient optical clearing of skin to minimize light scattering and
thus, to increase optical resolution and sensitivity of PAFC was recently
demonstrated (Figs. 10.10 and 10.11).280 The optical clearing effect was
achieved in 20 min using the following protocol: advanced skin cleaning,
microdermabrasion, and then glycerol topical application for which delivery
into skin was enhanced by massage and sonophoresis. For in vitro phantom
consisting of 0.8-mm mouse-skin layer over a blood vessel, 1.6-fold decrease
in laser spot blurring accompanied by a 1.6-fold increase in PA signal
amplitude against blood background were found. As a result, the peak rate for
Figure 10.10 Improvement of PAFC detection sensitivity for B16F10 melanoma cells
demonstrated for in vitro phantom (0.8-mm mouse-skin layer over a blood vessel) at skin
optical clearing with glycerol. Typical PAFC traces for melanoma cells in flow before (a) and
after optical clearing (b). Dependence of peak rate of melanoma cell counting (cells/min) on
the laser pulse energy before and after optical clearing (c).280
Figure 10.11 Optical clearing of human skin: visual contrast of the vein (a); typical
changes in PA signal waveform before/after optical clearing (b); PAFC traces for a
vein in human hand before/after optical clearing (c); US-imaging of the selected
vein (d).280
600 Chapter 10
B16F10 melanoma cell counting in blood flow increased 1.7-fold (Fig. 10.10).
By using optical clearing, the feasibility of PA contrast improvement for
human hand veins was also demonstrated (Fig. 10.11).
10.5.5 Nonlinear and Raman microscopy

Multiphoton excitation of molecules is a nonlinear process involving the
absorption of two or more photons, the total energy of which is sufficient to
induce a transition of the molecule to the excited state. The two-photon
technique uses the photons with the wavelength of the second harmonic of the
incident radiation coming exactly from the focal region of the incident
beam.1–3,281 The unique advantage of the multiphoton microscopy consists in
the possibility to study three-dimensional distributions of chromophores,
excited by the ultraviolet radiation in thick samples. Such an investigation
becomes possible because the chromophores can be excited by the radiation of
the second harmonic (e.g., at the wavelength of 350 nm), providing a high
quantum yield under the irradiation of the tissue, e.g., with the laser light
having the wavelength of 700 nm, where the tissues possess high transparency.
In this case, the long-wave radiation can reach deeper layers with less tissue
damage.281–284
The use of OCA can appear particularly important for improving the two
photon microscopy,23,24,285 since it was shown that the light-scattering effect
strongly reduces the light-penetration depth to values smaller that the depth of
the single photon fluorescence, whereas the resolution remains generally
unchanged.251 This happens mainly because of the exciting beam defocusing
in the scattering medium. On the other hand, this method is useful for better
understanding of molecular mechanisms of the optical clearing of the tissue
under immersion and dehydration.23,24,83
The improvement of the deep tissue two-photon signal by means of the
optical immersion method using hyperosmotic agents such as glycerol and
propylene glycol in the dehydrated form and the glucose solution, was first
demonstrated by Cicchi et al.23 in experiments with human skin ex vivo. The
optical clearing of deeper layers of the tissue occurs mainly at the expense of
the cumulative effect of reducing the scattering in the near-surface layers of
the tissue sample, which provides smaller damping of the incident and
detected radiation of fluorescence. Better focusing (smaller distortion of the
focused beam) is achieved in a medium with weaker scattering.23
Tseng et al.251 have shown that the combined confocal/two photon
microscopy provides a noninvasive method of microscopic studying of the
scaffold structure, which can be a valuable tool for complex investigation of
biomaterials and their interaction with the molecules/cells of interest inside the
scaffold. The integration of optical clearing with FocusClearTM, confocal/two
photon microscopy, and a three-dimensional presentation is an efficient
approach to the microscopic study of the scaffold structure.
Additional morphological information is provided by combining the

second harmonic generation (SHG) microscopy with the microscopy based on
two-photon excitation.282 At present, SHG is one of the new high-resolution
nonlinear-optical imaging methods for the study of intact tissues and cell
structures.283–292 Similar to multiphoton excited fluorescence, the SHG
method has a few important advantages compared to the methods of linear
spectroscopy and imaging, since the exciting near-infrared light is weakly
scattered by the tissue, while SHG occurs inside the tissue, in the small focal
region of the focused laser beam. This provides high spatial resolution,
acceptable probing depth, and decoupling of exciting and detected signals.
Strongly focused laser beams with high power density and very short pulse
duration of tens and hundreds of femtoseconds allow for harmonic generation
in a living tissue practically without any damage, because of the small
interaction time and the total energy that is too small for molecular
ionization.292
SHG is a nonlinear optical process of the second order that can occur only
in the media possessing no central symmetry. The method can be used for
imaging of highly ordered structure proteins without any exogenous labelling,
as well as for probing of biological membranes with high specificity to the
membrane type.284 In turn, third harmonic generation is very sensitive to the
refractive index variations and can be used to image interface regions in cells
and tissues.293
Due to its coherent nature, the SHG wave is mainly radiated in the
forward direction. For thicker and strongly scattering samples, a part of the
SHG signal is scattered. The major part of the SHG radiation still propagates
in the forward direction. A great part of the structural information encoded in
the forward signal is lost in the backward signal due to multiple scattering
events. Nevertheless, the SHG backscattering mode allows for study of upper
layers of intact skin or whole animals that could not be implemented in the
transmission mode.24,286,292,294
The SHG in skin is provided, mainly, by the dermis due to its major
component, the collagen, that possesses considerable nonlinear susceptibil-
ity.288–291 Obviously, the reduction of scattering due to optical clearing for the
incident long-wave light and, particularly, for the short-wave second-harmonic
detected light can improve the SHG images of the collagen structures of the
dermis in the forward direction and worsen the signal in the reflection mode.
Wen et al.75 injected glycerol solutions into the rat skin in vivo, and after
10 min the samples were cut for SHG imaging. For a 75%-glycerol solution,
the dehydration of dermis collagen expressed as a fiber diameter reduction
was observed.
Figure 10.12 illustrates the change of packing density of the dermis
collagen fibers under its interaction with 50%-PEG-400 solution. The images
were obtained using SHG microscopy.
602 Chapter 10
Figure 10.12 The packing density change of collagen fibers in the rat skin ex vivo under
the action of 50%-aqueous solution of PEG-400: the initial state (a); immediately after the
OCA application (b), and 15 min after the application of OCA (c). The images were obtained
by means of SHG microscopy. The incident light wavelength was 790 nm, the detected
wavelength 395 nm6HHFRORUSODWHV
Yeh et al.121 and Wen et al.75 demonstrated that for the collagen-based
connective tissues (samples of tendon and model tissues), the optical clearing
process using the high-concentration glycerol (13 M) manifested itself in
considerable reduction of the backscattered signal intensity and the increase of
the transmitted signal.
For deeper probing of the murine skeletal muscle and tendon samples
using three-dimensional SHG microscopy in the transmission mode, the
authors of Ref. 24 used the 50%-glycerol solution. They obtained an increase
of SHG imaging depth by 2.5 times in the muscle tissue. The amplification of
the signal was also due to better packing of the fibrils. It was also shown that
the axial attenuation of the direct SHG signal is reduced with the increase of
the glycerol concentration (25, 50, and 75%).71 The backward SHG signal
considerably decreased in the process of optical clearing due to the scattering
reduction and the change of the local density of dipoles, producing the second
harmonic.71
The Raman nonlinear spectroscopy also finds its application in studying
the mechanisms of optical clearing. Thus, e.g., in Ref. 295 the signal of
coherent anti-Stokes Raman scattering (CARS) served as the reference in the
study of the DMSO impact on the collagen structure and, therefore, the loss of
the SHG signal and the reduction of light scattering by human skin.
Hirshburg et al.110 found the correlation between the rate of the water loss by
the rodent skin ex vivo under the effects of different OCAs and the potential of
optical clearing, which was defined as m0s (before clearing)/m0s (after clearing).
Raman spectroscopy is a potentially noninvasive method for studying the
bone development and for diagnostics of bone diseases.296 Essential
amplification of commonly undetectable Raman signals from internal tissues
can be implemented by means of optical clearing of near-surface layers of the
tissue that shade the object of study.296,297 In this way, the transcutaneous
spectroscopy of the rat tibial bone in vivo was performed under the skin
optical clearing with glycerol.296 The process of porcine skin optical clearing
was studied in vitro by means of Raman microspectroscopy.297 The intensity
of Raman peaks at the depth of 400 mm was increased by 2-4 times. In
addition, the shifts of the peaks in the Raman spectrum of skin at different
concentrations of glycerol were observed.297
The confocal Raman spectroscopy can also be used for in vivo monitoring
of the penetration of clearing agents and permeability enhancers (particularly
DMSO) into the skin stratum corneum.298
10.5.6 Terahertz spectroscopy

The terahertz (THz) frequency range is intermediate between the IR and
microwave frequency ranges (the frequency n ¼ 1 THz corresponds to the
free-space wavelength l ¼ 300 mm). Recent progress in generating and
detecting the THz radiation and its usage offer its promising applications in
biology and medicine.299 Since the energy of THz photons is too small to
ionize molecules in biological systems, the THz spectroscopy is expected to be
a promising method of study in biology and medicine. Many vibrational
transitions of biomolecules lie just in this frequency range, and the light
scattering is not so strong as in the visible and NIR, so that the excitation by
ultrashort pulses can allow one to study a wide frequency range in a single
measurement with high temporal resolution.1,299–302 Terahertz spectroscopy
allows for determination of the medium complex refractive index in a single
shot, which is important for creating a functional THz tomography with high
sensitivity to the change of metabolite concentrations and precise mapping of
the pathologic process boundaries.299
Keeping in mind that the frequencies of the THz range mainly coincide
with the frequencies of rotational and vibrational transitions of the organic
molecules, the study of this range may be used for spectral identification of
molecules.303 This circumstance allows one to use THz technologies in
addition to the optical spectroscopic techniques. In this connection, the THz
region has attracted great interest in recent years, especially as an additional
channel in multimodal systems in combination with optical methods
providing the detection and functional imaging of metabolic and pathologic
processes.304
However the major problem of THz application for studies of soft tissues
is a great content of water in them, up to 75–80%. With extremely low
scattering for THz radiation, soft tissues have a very high absorption
coefficient on the level of 300 cm–1 and an index of refraction of 2–3 because
of water. As we demonstrated, the immersion optical clearing solves many
problems of applying optical methods to tissue study due to reversible
reduction of light scattering. For many OCAs, optical clearing is accompanied
by tissue dehydration that facilitates greater permeability for THz radiation
due to temporal reduction of water absorption. This concept was directly
604 Chapter 10
proved for in vitro measurements of porcine muscle sample dehydration under

its immersion in glycerol during 30 min.300 A significant reduction of
absorption and index of refraction of THz radiation in the region 0.1–1.5 THz
was found. This opens the way for more efficient combined application of
optical and terahertz imaging using the same OCA. A high sensitivity of the
THz radiation to the concentration of water in tissues can also serve as one of
the criteria for pathology recognition.304
Obviously, the implementation of in vivo dehydration requires temporal
and reversible changes in the tissues under study. It was proposed to perform
the dehydration of muscle tissue using the hyperosmotic agents.305 As clearing
agents, the authors used PEG-600, glycerol, and propylene glycol. Depending
on the frequency of radiation within the studied range 0.25–1.6 THz, the
absorption coefficient reduction approached 60%, and the necessary
application time was 8–10 min, which is acceptable for using this technology
in medical practice.
The dehydration of tissue by freezing or heating is also possible, but
application of these techniques is limited to in vitro studies. To improve
transmittance of different rat organs (kidney, diaphragm, liver, rectum,
stomach) in the range 0.4–2.2 THz, freezing (lyophilisation) was used.302
10.6 Determination of OCA and Drug Diffusion

Coefficients in Tissues
In spite of multiple studies related to the control of optical properties of
tissues, the problem of determining diffusion coefficients for hyperosmotic
fluids in tissues still remains insufficiently studied. Due to the complex
multicomponent structure of tissues and nonlinear character of diffusion
processes the measurement of the diffusion coefficients of hyperosmotic fluids
in tissues is a complex research problem. According to the present day
conceptions, the diffusion of different substances in tissues occurs in several
stages. At the first stage, the penetration of diffusing agents into the
intercellular and interfibrillar space and their interactions with cell
membranes and interstitial matrix of tissues occur. At the next stage, the
transmembrane diffusion of the agent into cells accompanied by the change of
the intracellular osmotic pressure is expected. One should not exclude the
possibility of transformation of the membranes themselves, related to the
change of the intercellular fluid properties. Under sufficient saturation of
interfibrillar space of a connective tissue with the OCA its interaction with the
tissue fibrils will obviously take place.
However, the use of sufficiently simple optical and diffusion models
allows for solution of the problem without significant loss of accuracy, and the
natural dispersion of optical and structural properties of tissues makes this
simplification acceptable. The resulting values of diffusion coefficients can be
successfully used in mathematical models, describing the processes of

interaction between hyperosmotic immersion fluids and tissues, which, in
particular is rather urgent in the study of processes related to transcutaneous
delivery of medicinal preparations and metabolic agents.
At present, a number of biophysical methods and techniques for
determining the diffusion coefficients of different agents exists.306–319
However, only a small part of the developed methods can be successfully
applied to assess the diffusion coefficients in tissues. The available methods
are mainly based on using the fluorescent measurement techniques or
radioactive labels for recording the flow of diffusing substance. The use of a
fluorescent measurement technique is impossible for the assessment of
diffusion coefficients of nonfluorescent substances (e.g., aqueous solutions
of glucose or glycerol), and the use of radioactive substances can be
undesirable in the case of in vivo measurements. Moreover, the high
sensitivity of fluorescence and radioactive measurements makes their use
difficult because of the natural metabolic activity of the organism, which
introduces additional errors into measurements. The method that allows
for assessment of diffusion coefficients of nonfluorescent hyperosmotic
fluids in tissues was proposed in Ref. 43 in 1997 and later developed by
the authors of Refs. 47 and 90. This method is based on measuring the
time dependence of the tissue scattering characteristics under the action of
hyperosmotic fluids and can be successfully used for both in vitro and in vivo
measurements.
The process of immersion fluid transport in fibrous tissues can be
described using the free diffusion model. In this case, the following
assumptions related to the transport process are commonly made: 1) only
the concentration diffusion takes place, i.e., the exchange flux of the
immersion fluid or drug into the tissue and of water from the tissue at a
given point is proportional to the gradient of concentration of the agent at this
point; 2) the diffusion coefficient is constant at all points in the studied tissue
sample.
Geometrically, the tissue sample is presented as a plane-parallel slab.
Since the area of the upper and lower surfaces of the slab is much larger than
the side area, it is possible to neglect the edge effects and solve the one
dimensional diffusion problem described by the equation90
∂Cðx,tÞ ∂2 Cðx,tÞ
¼D , (10.4)
∂t ∂x2
presenting the second Fick law, where C(x,t) is the concentration of the
immersion agent in the tissue (g/ml); D is the diffusion coefficient (cm2/s); t is
the time, during which the diffusion process occurs (s); x is the spatial
coordinate along the sample thickness direction (cm). Since in the experiments
606 Chapter 10
the volume of the immersion agent, as a rule, significantly exceeds the volume
of the tissue sample, the appropriate boundary conditions have the form90
∂Cðl,tÞ
Cð0,tÞ ¼ C 0 and ¼ 0, (10.5)
∂x
where C0 is the agent concentration in the solution (g/ml); l is the tissue sample
thickness (cm). The second boundary condition reflects the fact that the
diffusion of the agent into the tissue slab occurs only from one side of the
sample. In the case, when the agent diffuses from both sides of the tissue slab,
the boundary conditions have the following form:
Cð0,tÞ ¼ C 0 and Cðl,tÞ ¼ C 0 : (10.6)
The initial conditions express the absence of the agent in all internal points
of the sample before its incubation into the solution, i.e.,
Cðx,0Þ ¼ 0: (10.7)
The solution of the diffusion equation (10.4) with the initial (10.7) and
boundary (10.5) conditions taken into account has the following form:
X̀
4 ð2i þ 1Þpx ð2i þ 1Þ2 Dp2 t
Cðx,tÞ ¼ C 0 1 sin exp :
i¼0
pð2i þ 1Þ 2l 4l 2
The mean concentration of the immersion solution in the tissue sample

C(t) as a function of time is determined by the expression

8 X̀ 1 2 p
2
CðtÞ ¼ C 0 1 2 exp ð2i þ 1Þ t D∕l 2
: (10.8)
p i¼0 ð2i þ 1Þ2 4
In the case of double-side diffusion, the solution of the diffusion

equation (10.4) with the initial (10.7) and boundary (10.6) conditions has
the form
X̀
4 ð2i þ 1Þpx ð2i þ 1Þ2 Dp2 t
Cðx,tÞ ¼ C 0 1 sin exp :
i¼0
pð2i þ 1Þ l l2
In this case, the mean concentration of the immersion solution in the

tissue sample C(t) as a function of time is determined by the expression

8 X̀ 1
CðtÞ ¼ C 0 1 2 expðð2i þ 1Þ t p D∕l Þ :
2 2 2
(10.9)
p i¼0 ð2i þ 1Þ2
In the first approximation, Eq. (10.9) can be rewritten as
CðtÞ C 0 ð1 expðtpD∕l 2 ÞÞ, (10.10)

and Eq. (10.9) as
CðtÞ C 0 ð1 expðtp2 D∕l 2 ÞÞ: (10.11)

Note that Eqs. (10.10) and (10.11) correspond to the equation describing
the diffusion through a thin penetrable membrane.
As the agent penetrates into the tissue, the increase of the refractive index
of the interstitial fluid (ISF) will be observed. The assessment of the refractive
index of the ISF depending on the diffusion time can be performed based on
the Gladstone-Dale law, according to which the refractive index of a mixture
of noninteracting liquids is a sum of the refractive indices of the individual
components multiplied by the volume fractions of these components.320
Mathematically, the Gladstone-Dale law is written as
X X
nS ¼ ni C i , where C i ¼ 1: (10.12)
i i
Here, nS is the refractive index of the multicomponent mixture of non-

interacting liquids; ni and Ci are the refractive indices and volume fractions of
each component. In the case of two-component solutions (ISF and agent), the
Gladstone-Dale law has the form
nI ðtÞ ¼ ð1 CðtÞÞnbase þ CðtÞnosm , (10.13)

where nbase is the refractive index of the ISF at the initial moment of time, and
nosm is the refractive index of the hyperosmotic agent. For the aqueous
solution of glucose, the value of the refractive index (ngl) can be found from
the equation321
ngl ðlÞ ¼ nH 2 O ðlÞ þ 0.1515 C gl , (10.14)
where nH2O(l) is the spectrally dependent refractive index of water ([l] ¼ nm),
and Cgl is the glucose concentration in the solution(g/ml). The spectral
dependence of the water refractive index has the following form:322
6.878 103 1.132 109 1.11 1014

nH 2 O ðlÞ ¼ 1.3199 þ þ : (10.15)
l2 l4 l6
The optical model of a tissue can be presented as a layer having the
thickness l, containing the scatterers in the form of infinitely long thin
dielectric cylinders, parallel to the tissue surface. In the first approximation
one can assume that in the process of interaction of the tissue with the agent
the scatterer size does not change. For densely packed systems, i.e., the tissues
608 Chapter 10
with sufficiently large volume fraction of scatterers, it is important to allow for

the interference effects that arise because each scatterer is located in the near-
field zone of radiation, scattered by the other scatterers, described by
introducing the packing factor.1,3,172 For a system of thin dielectric
cylinders arranged parallel to each other, the packing factor has the form
(1 w)3/(1 + w),172 where w is the volume fraction of the scatterers. Thus, the
expression describing the tissue scattering coefficient can be written in the
form:90

w px3 2 2 ð1 wÞ3
ms ¼ ðm 1Þ 1 þ 2
2
: (10.16)
a 8 ðm þ 1Þ2 1þw
The time dependence of the collimated transmittance of the tissue slab
immersed in the solution of a hyperosmotic agent or medicinal preparation
has the following form:
T c ðtÞ ≃ expððma þ ms ðtÞÞ lÞ, (10.17)
where ma is the absorption coefficient of the tissue.
Equations (10.4)–(10.16) determine the dependence of the collimated
transmittance upon the concentration of the immersion agent in the sample,
i.e., formulate the direct problem. The inverse problem is to reconstruct the
diffusion coefficient from the time dependence of the collimated transmit-
tance. This problem is solved by minimization of the target functional
X
Nt
f ðDÞ ¼ ðT c ðD,ti Þ T c ðti ÞÞ2 (10.18)
i¼1
where Nt is the total number of experimental points, obtained by recording the

time dependence of collimated transmittance at the fixed wavelength: Tc(D,t)
is the transmittance, calculated using the Eq. (10.17) at the time moment t for
the given values of D; T c ðtÞ is the experimentally measured transmittance at
the time moment t.
The flowchart of the program implementing the present algorithm is
shown in Fig. 10.13. The main steps of the algorithm include
1. Specification of the initial parameters for the chosen wavelength (the
absorption coefficient (ma, the refractive index of the interstitial fluid ns,
the refractive index of the agent nI, the thickness of the tissue sample l)
obtained either from independent measurements, or from literature data.
2. Specification of the initial value of the diffusion coefficient D. The initial
value of D can be obtained from the analysis of variation P of the sample
n
t2
collimated transmittance based on the equation t ¼ Pn j¼1 j
ðtj ln yj Þ
, where tj
j¼1
is the time moment of measuring of the collimated transmittance Tc,

Figure 10.13 Flowchart of the program for calculating the diffusion coefficients of agents
based on the kinetics of collimated transmittance of tissue samples.
y ¼ 1 Tc/A; A is the maximal value of the collimated transmittance; t is

l2
the diffusion time constant (s); and t ¼ pD for a single-side diffusion and
l2
t ¼ p2 D for a double-side diffusion.
3. Calculation of the kinetics of collimated transmittance versus time for
given D and tj using the Bouguer’s law [Eq. (10.17)].
4. Comparison of the calculated and measured values of collimated
transmittance. If the difference does not exceed the prescribed error, the
value of D is found and the process is terminated, otherwise the value of D
is changed using the minimization simplex method,323 and the procedure
is repeated until the prescribed accuracy is achieved.
As an example, Table 10.2 presents diffusion coefficients of glucose
aqueous solutions in tissues, measured using the above method at the room
temperature (20°C) that were extrapolated to physiological temperature of
37°C:324
T 2 hðT 1 Þ
DðT 2 Þ ¼ DðT 1 Þ ,
T 1 hðT 2 Þ
Table 10.2 Glucose diffusion coefficients in tissues.47

Aqueous solution Diffusion coefficient, Diffusion coefficient,
Tissue of glucose × 106, cm2/s, 20°C × 106, cm2/s, 37°C
Eye sclera 0.2 g/ml 0.57 0.09 0.91 0.09
Eye sclera 0.3 g/ml 1.47 0.36 2.34 0.36
Eye sclera 0.4 g/ml 1.52 0.05 2.42 0.05
Pachymeninx 0.2 g/ml 1.63 0.29 2.59 0.29
Skin 0.4 g/ml 1.10 0.16 1.75 0.16
Skin (in vivo) 0.4 g/ml — 2.56 0.13
610 Chapter 10
where h(T) is the viscosity of the medium, where the diffusion occurs (e.g.,
water) at the given temperature T, T1 ¼ 20°C and T2 ¼ 37°C.
From Table 10.2 it is well seen that the temperature increase from room to
physiological, i.e., by 17°C, causes the increase of the diffusion coefficient by
nearly 1.5 times.
10.7 Conclusion
We have reviewed the specific features and methods of optical clearing and
related interactions of light with tissues. The impact of the OCA on a tissue
allows for efficient control of the optical properties, particularly, the reduction
of the tissue scattering coefficient, which facilitates the increase of efficiency of
optical imaging (optical biopsy) in medical applications. Both immersion and
compression techniques of optical clearing possess considerable potential for
many diagnostic, therapeutic, and surgical methods, in which the laser impact
on the target site hidden in the tissue is used.
Acknowledgments
The work was supported by grant 14-15-00186 of the Russian Science
Foundation.
This chapter is written on the basis of the recently published review paper
of the authors, E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, I. Yu.
Yanina, V. V. Tuchin, Optical clearing of biological tissues: prospects of
application in medical diagnostics and phototherapy [Review], J. of
Biomedical Photonics & Eng., 1(1), 22–58, 2015.
References
for Medical Diagnosis, 3rd edition, PM 254, SPIE Press, Bellingham,
WA, (2015).
2. T. Vo-Dinh (Ed.), Biomedical Photonics Handbook, CRC Press, Boca
Raton, FL (2003); second edition (2014).
3. L. V. Wang and H.-I. Wu, Biomedical Optics: Principles and Imaging,
Wiley-Interscience, Hoboken, New Jersey (2007).
4. R. K. Wang and V. V. Tuchin, eds., Advanced Biophotonics: Tissue
Optical Sectioning, CRC Press, Taylor & Francis Group, London
(2013).
5. V. V. Tuchin, Tissue Optics and Photonics: Biological Tissue Structures
[Review], J. of Biomedical Photonics & Eng. 1(1), 3–21, (2015).
6. V. V. Tuchin, L. V. Wang, and D. A. Zimnyakov, Optical Polarization
in Biomedical Applications, Springer-Verlag, NY (2006).
7. R. Drezek, A. Dunn, and R. Richards-Kortum, “Light scattering from

cells: finite-difference time-domain simulations and goniometric mea-
surements,” Appl. Opt. 38(16), 3651–3661 (1999).
8. K. Sokolov, R. Drezek, K. Gossagee, and R. Richards-Kortum,
“Reflectance spectroscopy with polarized light: is it sensitive to cellular
and nuclear morphology,” Opt. Express 5, 302–317 (1999).
9. D. W. Leonard and K. M. Meek, “Refractive indices of the collagen
fibrils and extrafibrillar material of the corneal stroma,” Biophysical
J. 72, 1382–1387 (1997).
10. A. G. Borovoi, E. I. Naats, and U. G. Oppel, “Scattering of light by a
red blood cell,” J. Biomed. Opt. 3, 364–372 (1998).
11. G. Mazarevica, T. Freivalds, and A. Jurka, “Properties of erythrocyte
light refraction in diabetic patients,” J. Biomed. Opt. 7, 244–247 (2002).
12. O. Sydoruk, O. Zhernovaya, V. Tuchin, and A. Douplik, “Refractive
index of solutions of human hemoglobin from the near infrared to the
ultraviolet range: Kramers–Kronig analysis,” J. Biomed. Opt. 17(11),
115002 (2012).
13. M. Friebel and M. Meinke, “Model function to calculate the refractive
index of native hemoglobin in the wavelength range of 250–1100 nm
dependent on concentration,” Appl. Opt. 45(12), 2838–2842 (2006).
14. E. Gratton, “Deeper tissue imaging with total detection,” Science 331,
1016–1017 (2011).
15. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning
fluorescence microscopy,” Science 248, 73–76 (1990).
16. C. A. Combs, A. Smirnov, D. Chess, D. B. Mcgavern, J. L. Schroeder,
J. Riley, S. S. Kang, M. Lugar-Hammer, A. Gandjbakhche, J. R. Knutson,
and R. S. Balaban, “Optimizing multiphoton fluorescence microscopy light
collection from living tissue by noncontact total emission detection
(epiTED),” J. Microsc. 241(2), 153–161 (2011).
17. C. A. Combs, A. V. Smirnov, J. D. Riley, A. H. Gandjbakhche, J. R.
Knutson, and R. S. Balaban, “Optimization of multiphoton excitation
microscopy by total emission detection using a parabolic light reflector,”
J. Microsc. 228(3), 330–337 (2007).
18. R. K. Wang and V. V. Tuchin, “Optical coherence tomography. Light
scattering and imaging enhancement,” Chap. 16 in Handbook of
Coherent-Domain Optical Methods. Biomedical Diagnostics, Environ-
mental Monitoring, and Material Science, 2nd ed., V. V. Tuchin, Ed.,
New York, Heidelberg, Dordrecht, London: Springer, 665–742 (2013).
19. L. V. Wang and S. Hu, “Photoacoustic tomography: in vivo imaging
from organelles to organs,” Science 335, 1458–1462 (2012).
20. Y. M. Liew, R. A. McLaughlin, F. M. Wood, and D. D. Sampson,
“Reduction of image artifacts in three-dimensional optical coherence
tomography of skin in vivo,” J. Biomed. Opt. 16(11), 116018 (2011).
612 Chapter 10
21. V. V. Tuchin, “A clear vision for laser diagnostics (Review),” IEEE J.

Sel. Top. Quantum Electron. 13, 1621–1628 (2007).
22. V. V. Tuchin, Optical Clearing of Tissues and Blood, SPIE Press,
23. R. Cicchi, F. S. Pavone, D. Massi, and D. D. Sampson, “Contrast and
depth enhancement in two-photon microscopy of human skin ex vivo by
use of optical clearing agents,” Opt. Exp. 13, 2337–2344 (2005).
24. S. Plotnikov, V. Juneja, A. B. Isaacson, W. A. Mohler, and P. J.
Campagnola, “Optical clearing for improved contrast in second
harmonic generation imaging of skeletal muscle,” Biophys. J. 90, 328–
339 (2006).
25. E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, and V. V. Tuchin,
“Optical clearing of the eye sclera in vivo caused by glucose,” Quantum
Electronics 36(12), 1119–1124 (2006).
26. G. Vargas, J. K. Barton, and A. J. Welch, “Use of hyperosmotic
chemical agent to improve the laser treatment of cutaneous vascular
lesions,” J. Biomed. Opt. 13(2), 021114 (2008).
27. M. H. Khan, S. Chess, B. Choi, K. M. Kelly, and J. S. Nelson, “Can
topically applied optical clearing agents increase the epidermal damage
threshold and enhance therapeutic efficacy?” Lasers Surg. Med. 35,
93–95 (2004).
28. P. D. Agrba, M. Yu. Kirillin, A. I. Abelevich, E. V. Zagaynova, and
V. A. Kamensky, “Compression as a method for increasing the
informativity of optical coherence tomography of biotissue,” Optics
and Spectroscopy 107(6), 853–858 (2009).
29. C. Drew, T. E. Milner, and C. G. Rylander, “Mechanical tissue optical
clearing devices: evaluation of enhanced light penetration in skin using
optical coherence tomography,” J. Biomed. Opt. 14(6), 064019 (2009).
30. N. Guzelsu, J. F. Federici, H. C. Lim, H. R. Chauhdry, A. B. Ritter, and
T. Findley, “Measurement of skin strech via light reflection,” J. Biomed.
Opt. 8, 80–86 (2003).
31. C. G. Rylander, O. F. Stumpp, T. E. Milner, N. J. Kemp, J. M.
Mendenhall, K. R. Diller, and A. J. Welch, “Dehydration mechanism of
optical clearing in tissue,” J. Biomed. Opt. 11, 041117 (2006).
32. T. Yu, X. Wen, V. V. Tuchin, Q. Luo, and D. Zhu, “Quantitative
analysis of dehydration in porcine skin for assessing mechanism of
optical clearing,” J. Biomed. Opt. 16, 095002 (2011).
33. W.-C. Lin, M. Motamedi, and A. J. Welch, “Dynamics of tissue optics
during laser heating of turbid media,” Appl. Opt. 35(19), 3413–3420
(1996).
34. D. Zhu, J. Wang, Z. Zhi, X. Wen, and Q. Luo, “Imaging dermal blood
flow through the intact rat skin with an optical clearing method,”
J. Biomed. Opt. 15, 026008 (2010).
35. K. V. Larin, M. G. Ghosn, A. N. Bashkatov, E. A. Genina, N. A.

Trunina, and V. V. Tuchin, “Optical clearing for OCT image
enhancement and in-depth monitoring of molecular diffusion,” IEEE
J. Select. Tops. Quantum Electron. 18, 1244–1259 (2012).
36. D. Zhu, K. Larin, Q. Luo, and V. V. Tuchin, “Recent progress in tissue
optical clearing,” Laser & Photonics Reviews 7(5), 732–757 (2013).
37. V. A. Doubrovskii, I. Yu. Yanina, and V. V. Tuchin, “Kinetics of
changes in the coefficient of transmission of the adipose tissue in vitro as
a result of photodynamic action,” Biophysics 57(1), 94–98 (2012).
38. I. Yu. Yanina, N. A. Trunina, and V. V. Tuchin, “Optical coherence
tomography of adipose tissue at photodynamic/photothermal treatment
in vitro,” J. Innovative Optical Health Sciences 6(2), 1350010 (2013).
39. E. A. Genina, A. N. Bashkatov, and V. V. Tuchin, “Tissue optical
immersion clearing,” Expert Rev. Med. Devices 7, 825–842 (2010).
40. R. Barer, “Spectrophotometry of clarified cell suspensions,” Science 121,
709–715 (1955).
41. V. V. Bakutkin, I. L. Maksimova, P. I. Saprykin, V. V. Tuchin, and L. P.
Shubochkin, “Light scattering by the human eye sclera,” J. Appl.
Spectrosc. (USSR) 46(1), 104–107 (1987).
controlled optical properties,” Proc. SPIE 2925, 118–142 (1996).
controlled optical properties,” J. Biomed. Opt. 2, 401–417 (1997).
44. A. N. Bashkatov, E. A. Genina, V. I. Kochubey, V. V. Tuchin, and Yu.
P. Sinichkin, “The influence of osmotically active chemical agents on the
transport of light in the scleral tissue,” Proc. SPIE 3726, 403–409 (1998).
45. V. V. Tuchin, A. N. Bashkatov, E. A. Genina, Yu. P. Sinichkin, and
N. A. Lakodina, “In vivo investigation of the immersion-liquid-induced
human skin clearing dynamics,” Technical Physics Letters 27(6), 489–
490 (2001).
46. E. A. Genina, Investigation of optical immersion and staining of
biological tissues in vivo for optical diagnostics and laser therapy, Ph.D.
thesis, Saratov State University, Saratov, Russia (2002) (in Russian).
47. A. N. Bashkatov, Control of tissue optical properties by means of
osmotically active immersion liquids, Ph.D. thesis, Saratov State
University, Saratov, Russia (2002) (in Russian).
48. I. V. Meglinskii, A. N. Bashkatov, E. A. Genina, D. Yu. Churmakov,
and V. V. Tuchin, “Study of the possibility of increasing the probing
depth by the method of reflection confocal microscopy upon immersion
clearing of near-surface human skin layers,” Quantum Electronics 32(10),
875–882 (2002).
614 Chapter 10
49. I. V. Meglinski, A. N. Bashkatov, E. A. Genina, D. Y. Churmakov, and

V. V. Tuchin, “The enhancement of confocal images of tissues at bulk
optical immersion,” Laser Physics 13(1), 65–69 (2003).
50. E. I. Galanzha, V. V. Tuchin, A. V. Solovieva, T. V. Stepanova, Q. Luo,
H., “Skin backreflectance and microvascular system functioning at the
action of osmotic agents,” Phys. D: Appl. Phys. 36, 1739–1746 (2003).
51. V. V. Tuchin, D. M. Zhestkov, A. N. Bashkatov, and E. A. Genina,
“Theoretical study of immersion optical clearing of blood in vessels at
local hemolysis,” Optics Express 12(13), 2966–2971 (2004).
52. E. A. Genina, A. N. Bashkatov, V. I. Kochubey, and V. V. Tuchin,
“Optical clearing of human dura mater,” Optics and Spectroscopy 98(3),
470–476 (2005).
53. A. N. Bashkatov, D. M. Zhestkov, E. A. Genina, and V. V. Tuchin,
“Immersion clearing of human blood in the visible and near-infrared
spectral regions,” Optics and Spectroscopy 98(4), 638–646 (2005).
54. V. V. Tuchin, “Optical clearing of tissues and blood using the immersion
55. V. V. Tuchin, “Optical immersion as a new tool for controlling the
optical properties of tissues and blood,” Laser Phys. 15, 1109–1136
(2005).
56. E. A. Genina, A. N. Bashkatov, A. A. Korobko, E. A. Zubkova, V. V.
Tuchin, I. V. Yaroslavsky, and G. B. Altshuler, “Optical clearing of
human skin: comparative study of permeability and dehydration of
intact and photothermally perforated skin,” J. Biomed. Opt. 13, 021102
(2008).
57. E. A. Genina, A. N. Bashkatov, and V. V. Tuchin, “Optical clearing of
cranial bone,” Advanced Optical Technologies, 267867 (2008).
58. E. A. Genina, A. N. Bashkatov, K. V. Larin, and V. V. Tuchin, “Light-
tissue interaction at optical clearing,” Chap. 7 in Laser Imaging and
Manipulation in Cell Biology, F. S. Pavone, Ed., 115–164, Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim (2010).
59. I. V. Larina, E. F. Carbajal, V. V. Tuchin, M. E. Dickinson, and K. V.
Larin, “Enhanced OCT imaging of embryonic tissue with optical
clearing,” Laser Phys. Lett. 5, 476–479 (2008).
60. M. Bonesi, S. G. Proskurin, and I. V. Meglinski, “Imaging of
subcutaneous blood vessels and flow velocity profiles by optical
coherence tomography,” Laser Phys. 20, 891–899 (2010).
61. G. Vargas, E. K. Chan, J. K. Barton, H. G. Rylander, and A. J. Welch,
“Use of an agent to reduce scattering in skin,” Lasers Surg. Med. 24(2),
133–141 (1999).
62. G. Vargas, “Reduction of light scattering in biological tissue: implica-
tions for optical diagnostics and therapeutics,” PhD Thesis, The
University of Texas, USA (2001).
63. G. Vargas, K. F. Chan, S. L. Thomsen, and A. J. Welch, “Use of

osmotically active agents to alter optical properties of tissue: effects on
the detected fluorescence signal measured through skin,” Lasers Surg.
Med. 29(3), 213–220 (2001).
64. G. Vargas, A. Readinger, S. S. Dosier, and A. J. Welch, “Morphological
changes in blood vessels produced by hyperosmotic agents and measured
by optical coherence tomography,” Photochem. Photobiol. 77, 541–549
(2003).
65. R. K. Wang, X. Xu, V. V. Tuchin, and J. B. Elder, “Concurrent
enhancement of imaging depth and contrast for optical coherence
tomography by hyperosmotic agents,” J. Opt. Soc. Aim. B 18, 948–953
(2001).
66. X. Xu and R. K. Wang, “The role of water desorption on optical
clearing of biotissue: studied with near infrared reflectance spectros-
copy,” Med. Phys. 30, 1246–1253 (2003).
67. Y. He, R. K. Wang, and D. Xing, “Enhanced sensitivity and spatial
resolution for in vivo imaging with low-level light-emitting probes by use
of biocompatible chemical agents,” Opt. Lett. 28, 2076–2078 (2003).
68. Y. He and R. K. Wang, “Dynamic optical clearing effect of tissue
impregnated with hyperosmotic agents and studied with optical
coherence tomography,” J. Biomed. Opt. 9, 200–206 (2004).
69. M. H. Khan, B. Choi, S. Chess, K. M. Kelly, J. McCullough, and J. S.
Nelson, “Optical clearing of in vivo human skin: implications for light-
based diagnostic imaging and therapeutics,” Lasers Surg. Med. 34,
83–85 (2004).
70. H. Cheng, Q. Luo, S. Zeng, S. Chen, W. Luo, and H. Gong,
“Hyperosmotic chemical agent’s effect on in vivo cerebral blood flow
revealed by laser speckle,” Appl. Opt. 43, 5772–5777 (2004).
71. R. LaComb, O. Nadiarnykh, S. Carey, and P. J. Campagnola,
“Quantitative second harmonic generation imaging and modeling of
the optical clearing mechanism in striated muscle and tendon,”
J. Biomed. Opt. 13, 021109 (2008).
72. O. Nadiarnykh and P. J. Campagnola, “Retention of polarization
signatures in SHG microscopy of scattering tissues through optical
clearing,” Opt. Exp. 17, 5794–5806 (2009).
73. D. Zhu, Q. Luo, and J. Cen, “Effects of dehydration on the optical
properties of in vitro porcine liver,” Lasers Surg. Med. 33, 226–231 (2003).
74. D. Zhu, J. Zhang, H. Cui, Z. Mao, P. Li, and Q. Luo, “Short-term and
long-term effects of optical clearing agents on blood vessels in chick
chorioallantoic membrane,” J. Biomed. Opt. 13(2), 021106 (2008).
75. X. Wen, Z. Mao, Z. Han, V. V. Tuchin, and D. Zhu, “In vivo skin
optical clearing by glycerol solutions: mechanism,” J. Biophotonics
3(1-2), 44–52 (2010).
616 Chapter 10
76. J. Wang, D. Zhu, M. Chen, and X. Liu, “Assessment of optical clearing

induced improvement of laser speckle contrast imaging,” J. Innovat. Opt.
Health Sci. 3, 159–167 (2010).
77. R. Shi, M. Chen, V. V. Tuchin, and D. Zhu, “Accessing to arteriovenous
blood flow dynamics response using combined laser speckle contrast
imaging and skin optical clearing,” Biomedical Optics Express 6 (6),
1977–1989 (2015).
78. J. Wang, Y. Zhang, T. Xu, Q. Luo, and D. Zhu, “An innovative
transparent cranial window based on skull optical clearing,” Laser Phys.
Lett. 9, 469–473 (2012).
79. X. Wen, S. L. Jacques, V. V. Tuchin, and D. Zhu, “Enhanced optical
clearing of skin in vivo and optical coherence tomography in-depth
imaging,” J. Biomed. Opt. 17, 066022 (2012).
80. J. Wang, Y. Zhang, P. Li, Q. Luo, and D. Zhu, “Review: tissue optical
clearing window for blood flow monitoring,” IEEE J. Sel. Topics in
Quantum Electron. 20(2), 6801112 (2014).
81. M. Kinnunen, R. Myllylä, and S. Vainio, “Detecting glucose-induced
changes in in vitro and in vivo experiments with optical coherence
tomography,” J. Biomed. Opt. 13, 021111 (2008).
82. R. Dickie, R. M. Bachoo, M. A. Rupnick, S. M. Dallabrida, G. M.
DeLoid, J. Lai, R. A. DePinho, and R. A. Rogers, “Three-dimensional
visualization of microvessel architecture of whole-mount tissue by
confocal microscopy,” Microvasc. Res. 72, 20–26 (2006).
83. V. Hovhannisyan, P.-S. Hu, S.-J. Chen, C.-S. Kim, and C.-Y. Dong,
“Elucidation of the mechanisms of optical clearing in collagen tissue
with multiphoton imaging,” J. Biomed. Opt. 18(4), 046004 (2013).
84. H. Hama, H. Kurokawa, H. Kawano, R. Ando, T. Shimogori, H. Noda,
K. Fukami, A. Sakaue-Sawano, and A. Miyawaki, “Scale: a chemical
approach for fluorescence imaging and reconstruction of transparent
mouse brain,” Nature Neurosci. 14, 1481–1488 (2011).
85. H. U. Dodt, U. Leischner, A. Schierloh, N. Jährling, C. P. Mauch,
K. Deininger, J. M. Deussing, M. Eder, W. Zieglgnsberger, and
K. Becker, “Ultramicroscopy: three-dimensional visualization of
neuronal networks in the whole mouse brain,” Nature Methods 4(4),
331–336 (2007).
86. A. Ertürk, C. P. Mauch, F. Hellal, F. Förstner, T. Keck, K. Becker,
N. Jährling, H. Steffens, M. Richter, M. Hübener, E. Kramer,
F. Kirchhoff, H. U. Dodt, and F. Bradke, “Three-dimensional imaging
of the unsectioned adult spinal cord to assess axon regeneration and glial
responses after injury,” Nature Med. 18, 166–171 (2012).
87. A. N. Bashkatov, E. A. Genina, Yu. P. Sinichkin, V. I. Kochubey, N. A.
Lakodina, and V. V. Tuchin, “Glucose and mannitol diffusion in human
dura mater,” Biophys. J. 85(5), 3310–3318 (2003).
88. S. G. Proskurin and I. V. Meglinski, “Optical coherence tomography

imaging depth enhancement by superficial skin optical clearing,” Laser
Phys. Lett. 4, 824–826 (2007).
89. M. G. Ghosn, V. V. Tuchin, and K. V. Larin, “Depth-resolved
monitoring of glucose diffusion in tissues by using optical coherence
tomography,” Opt. Lett. 31, 2314–2316 (2006).
90. A. N. Bashkatov, E. A. Genina, and V. V. Tuchin, “Measurement of
glucose diffusion coefficients in human tissues,” Chap. 19 in Handbook
of Optical Sensing of Glucose in Biological Fluids and Tissues, V. V.
Tuchin, Ed., Taylor & Francis Group LLC, CRC Press, 587–621
(2009).
91. A. N. Bashkatov, E. A. Genina, Yu. P. Sinichkin, V. I. Kochubei, N. A.
Lakodina, and V. V. Tuchin, “Estimation of the glucose diffusion
coefficient in human eye sclera,” Biophysics 48(2), 292–296 (2003).
92. E. A. Genina, A. N. Bashkatov, E. A. Zubkova, T. G. Kamenskikh, and
V. V. Tuchin, “Measurements of Retinalamin diffusion coefficient in
human sclera by optical spectroscopy,” Optics and Lasers in Engineering
46, 915–920 (2008).
93. E. A. Genina, A. N. Bashkatov, V. V. Tuchin, M. G. Ghosn, K. V.
Larin, and T. G. Kamenskikh, “Cortexin diffusion in human eye sclera,”
94. M. G. Ghosn, V. V. Tuchin, and K. V. Larin, “Nondestructive
quantification of analyte diffusion in cornea and sclera using optical
coherence tomography,” Invest. Ophthal. Visual Sci. 48, 2726–2733
(2007).
95. M. G. Ghosn, E. F. Carbajal, N. Befrui, V. V. Tuchin, and K. V. Larin,
“Differential permeability rate and percent clearing of glucose in
different regions in rabbit sclera,” J. Biomed. Opt. 13, 021110 (2008).
96. Luís M. Oliveira, M. Inês Carvalho, Elisabete M. Nogueira, and Valery
V. Tuchin, “Diffusion characteristics of ethylene glycol in skeletal
muscle,” J. Biomed. Opt. 20(5), 051019-1–10 (2015)
97. M. G. Ghosn, N. Sudheendran, M. Wendt, A. Glasser, V. V. Tuchin,
and K. V. Larin, “Monitoring of glucose permeability in monkey skin in
vivo using optical coherence tomography,” J. Biophotonics 3, 25–33
(2010).
98. X. Guo, Z. Guo, H. Wei, H. Yang, Y. He, S. Xie, G. Wu, X. Deng,
Q. Zhao, and L. Li, “In vivo comparison of the optical clearing efficacy
of optical clearing agents in human skin by quantifying permeability
using optical coherence tomography,” Photochem. Photobiol. 87,
734–740 (2011).
99. K. V. Larin, M. G. Ghosn, S. N. Ivers, A. Tellez, and J. F. Granada,
“Quantification of glucose diffusion in arterial tissues by using optical
coherence tomography,” Laser Physics Letters 4(4), 312–317 (2007).
618 Chapter 10
100. X. Guo, G. Wu, H. Wei, X. Deng, H. Yang, Y. Ji, Y. He, Z. Guo,

S. Xie, H. Zhong, Q. Zhao, and Z. Zhu, “Quantification of glucose
diffusion in human lung tissues by using Fourier domain optical
coherence tomography,” Photochem. Photobiol. 88, 311–316 (2012).
101. S. Tanev, V. V. Tuchin, and P. Paddon, “Cell membrane and gold
nanoparticles effects on optical immersion experiments with noncancer-
ous and cancerous cells: finite-difference time-domain modelling,”
J. Biomed. Opt. 11, 064037 (2006).
102. M. G. Ghosn, E. F. Carbajal, N. A. Befrui, A. Tellez, J. F. Granada,
and K. V. Larin, “Permeability of hyperosmotic agent in normal and
atherosclerotic vascular tissues,” J. Biomed. Opt. 13, 010505 (2008).
103. H. Q. Zhong, Z. Y. Guo, H. J. Wei, C. C. Zeng, H. L. Xiong, Y. H. He,
and S. H. Liu, “Quantification of glycerol diffusion in human normal
and cancer breast tissues in vitro with optical coherence tomography,”
Laser Phys. Lett. 7, 315–320 (2010).
104. Q. L. Zhao, J. L. Si, Z. Y. Guo, H. J. Wei, H. Q. Yang, G. Y. Wu, S. S.
Xie, X. Y. Li, X. Guo, H. Q. Zhong, and L. Q. Li, “Quantifying glucose
permeability and enhanced light penetration in ex vivo human normal
and cancerous oesophagus tissues with optical coherence tomography,”
Laser Phys. Lett. 8, 71–77 (2011).
105. Z. Zhu, G. Wu, H. Wei, H. Yang, Y. He, S. Xie, Q. Zhao, and X. Guo,
“Investigation of the permeability and optical clearing ability of different
analytes in human normal and cancerous breast tissues by spectral
domain OCT,” J. Biophoton. 5, 1–8 (2012).
106. B. Choi, L. Tsu, E. Chen, T. S. Ishak, S. M. Iskandar, S. Chess, and J. S.
Nelson, “Determination of chemical agent optical clearing potential
using in vitro human skin,” Lasers Surg. Med. 36, 72–75 (2005).
107. Z. Mao, D. Zhu, Y. Hu, X. Wen, and Z. Han, “Influence of alcohols on
the optical clearing effect of skin in vitro,” J. Biomed. Opt. 13, 021104
(2008).
108. E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, and V. V. Tuchin,
“Optical clearing of skin under action of glycerol: ex vivo and in vivo
investigations,” Optics and Spectroscopy 109(2), 225–231 (2010).
109. J. Hirshburg, B. Choi, J. S. Nelson, and A. T. Yeh, “Correlation between
collagen solubility and skin optical clearing using sugars,” Lasers Surg.
Med. 39, 140–144 (2007).
110. J. M. Hirshburg, K. M. Ravikumar, W. Hwang, and A. T. Yeh,
“Molecular basis for optical clearing of collagenous tissues,” J. Biomed.
Opt. 15(5), 055002 (2010).
111. J. Wang, N. Ma, R. Shi, Y. Zhang, T. Yu, and D. Zhu, “Sugar-induced
skin optical clearing: from molecular dynamics simulation to experimen-
tal demonstration,” IEEE J. Selected Topics in Quantum Electronics
20(2), 7101007 (2014).
112. Y. Ding, J. Wang, Z. Fan, D. Wei, R. Shi, Q. Luo, D. Zhu, and X. Wei,
“Signal and depth enhancement for in vivo flow cytometer measurement
of ear skin by optical clearing agents,” Biomed. Opt. Exp. 4(11), 2518–
2526 (2013).
113. J. Jiang and R. K. Wang, “Comparing the synergistic effects of oleic acid
and dimethyl sulfoxide as vehicles for optical clearing of skin tissue in
vitro,” Phys. Med. Biol. 49, 5283–5294 (2004).
114. Y. Liu, X. Yang, D. Zhu, and Q. Luo, “Optical clearing agents improve
photoacoustic imaging in the optical diffusive regime,” Optics Letters
38(20), 4236–4239 (2013).
115. X. Xu and R. K. Wang, “Synergistic effect of hyperosmotic agents of
dimethyl sulfoxide and glycerol on optical clearing of gastric tissue
studied with near infrared spectroscopy,” Phys. Med. Biol. 49, 457–468
(2004).
116. J. Jiang, M. Boese, P. Turner, and R. K. Wang, “Penetration kinetics of
dimethyl sulphoxide and glycerol in dynamic optical clearing of porcine
skin tissue in vitro studied by Fourier transform infrared spectroscopic
imaging,” J. Biomed. Opt. 13(2), 021105 (2008).
117. A. T. Yeh and J. Hirshburg, “Molecular interactions of exogenous
chemical agents with collagen-implications for tissue optical clearing,”
J. Biomed. Opt. 11(1), 014003 (2006).
118. E. A. Genina, A. N. Bashkatov, E. A. Kolesnikova, M. V. Basco, G. S.
Terentyuk, and V. V. Tuchin, “Optical coherence tomography monitor-
ing of enhanced skin optical clearing in rats in vivo,” J. Biomed. Opt.
19(2), 021109 (2014).
119. A. N. Bashkatov, A. N. Korolevich, V. V. Tuchin, Y. P. Sinichkin, E. A.
Genina, M. M. Stolnitz, N. S. Dubina, S. I. Vecherinski, and M. S.
Belsley, “In vivo investigation of human skin optical clearing and blood
microcirculation under the action of glucose solution,” Asian J. Physics
15(1), 1–14 (2006).
120. E. A. Genina, A. N. Bashkatov, and V. V. Tuchin, “Glucose-induced
optical clearing effects in tissues and blood,” Chap. 21 in Handbook of
Optical Sensing of Glucose in Biological Fluids and Tissues, V. V.
Tuchin, Ed., Taylor & Francis Group LLC, CRC Press, 657–692
(2009).
121. A. T. Yeh, B. Choi, J. S. Nelson, and B. J. Tromberg, “Reversible disso-
ciation of collagen in tissues,” J. Invest. Dermatol. 121, 1332–1335 (2003).
122. J. Hirshburg, B. Choi, J. S. Nelson, and A. T. Yeh, “Collagen
solubility correlates with skin optical clearing,” J. Biomed. Opt. 11,
040501 (2006).
123. J. W. Wiechers, J. C. Dederen, and A. V. Rawlings, “Moisturization
mechanisms: internal occlusion by orthorhombic lipid phase stabilizers - a
novel mechanism of action of skin moisturization,” Chap. 9 in Skin
620 Chapter 10
Moisturization, 2nd ed., A. V. Rawlings and J. J. Leyden, Eds.,

Informa Healthcare, Taylor & Francis Group, New York, 309–321
(2009).
124. A. V. Papaev, G. V. Simonenko, V. V. Tuchin, and T. P. Denisova,
“Optical anisotropy of a biological tissue under conditions of immersion
clearing and without them,” Optics and Spectroscopy 101(1), 46–53
(2006).
125. A. Roggan, M. Friebel, K. Dorschel, A. Hahn, and G. Mueller, “Optical
properties of circulating human blood in the wavelength range 400–
2500 nm,” J. Biomed. Opt. 4, 36–46 (1999).
126. M. Friebel, J. Helfmann, and M. Meinke, “Influence of osmolarity on
the optical properties of human erythrocytes,” J. Biomed. Opt. 15(5),
055005 (2010).
127. L. D. Shvartsman and I. Fine, “Optical transmission of blood: Effect of
erythrocyte aggregation,” IEEE Trans. Biomed. Eng. 50, 1026–1033
(2003).
128. V. V. Tuchin, X. Xu, and R. K. Wang, “Dynamic optical coherence
tomography in studies of optical clearing, sedimentation, and aggrega-
tion of immersed blood,” Appl. Opt. 41(1), 258–271 (2002).
129. H. Schaefer and T. E. Redelmeier, Skin Barrier, Karger, Basel (1996).
130. C. Liu, Z. Zhi, V. V. Tuchin, Q. Luo, and D. Zhu, “Enhancement of skin
optical clearing efficacy using photo-irradiation,” Lasers Surg. Med. 42,
132–140 (2010).
131. V. V. Tuchin, G. B. Altshuler, A. A. Gavrilova, A. B. Pravdin,
D. Tabatadze, J. Childs, and I. V. Yaroslavsky, “Optical clearing of skin
using flashlamp-induced enhancement of epidermal permeability,”
Lasers Surg. Med. 38, 824–836 (2006).
132. E. A. Kolesnikova, A. S. Kolesnikov, E. A. Genina, L. E. Dolotov,
D. K. Tuchina, A. N. Bashkatov, and V. V. Tuchin, “Use of fractional
laser microablation of skin for improvement of its immersion clearing,”
Proc. SPIE 8699, 86990B (2013).
133. E. A. Genina, A. N. Bashkatov, V. V. Tuchin, G. B. Altshuler, and I. V.
Yaroslavski, “Possibility of increasing the efficiency of laser-induced
tattoo removal by optical skin clearing,” Quantum Electronics 38(6),
580–587 (2008).
134. X. Xu, Q. Zhu, and C. Sun, “Assessment of the effects of ultrasound-
mediated alcohols on skin optical clearing,” J. Biomed. Opt. 14, 034042
(2009).
135. H. Zhong, Z. Guo, H. Wei, C. Zeng, H. Xiong, Y. He, and S. Liu,
“In vitro study of ultrasound and different-concentration glycerol-
induced changes in human skin optical attenuation assessed with
optical coherence tomography,” J. Biomed. Opt. 15, 036012
(2010).
136. O. Stumpp, B. Chen, and B. Welch, “Using sandpaper for noninvasive

transepidermal optical skin clearing agent delivery,” J. Biomed. Opt. 11,
041118 (2006).
137. J. Yoon, T. Son, E. Choi, B. Choi, J. S. Nelson, and B. Jung,
“Enhancement of optical clearing efficacy using a microneedle roller,”
J. Biomed. Opt. 13(2), 021103 (2008).
138. J. Yoon, D. Park, T. Son, J. Seo, J. S. Nelson, and B. Jung, “A physical
method to enhance transdermal delivery of a tissue optical clearing
agent: combination of microneedling and sonophoresis,” Lasers Surg.
Med. 42, 412–417 (2010).
139. M. A. Fox, D. G. Diven, K. Sra, A. Boretsky, T. Poonawalla, A.
Readinger, M. Motamedi, and R. J. Nichols, “Dermal scatter reduction
in human skin: a method using controlled application of glycerol,”
Lasers Surg. Med. 41, 251–255 (2009).
140. C. G. Rylander, T. E. Milner, S. A. Baranov, and J. S. Nelson,
“Mechanical tissue optical clearing devices: enhancement of light
penetration in ex vivo porcine skin and adipose tissue,” Lasers Surg.
Med. 40, 688–694 (2008).
141. X. Xu and Q. Zhu, “Evaluation of skin optical clearing enhancement
with Azone as a penetration enhancer,” Optics Communications 279,
223–228 (2007).
142. J. Jiang and R. K. Wang, “How different molarities of oleic acid as
enhancer exert its effect on optical clearing of skin tissue in vitro,”
J. X-Ray Science and Technology 13, 149–159 (2005).
143. J. Jiang, M. Boese, P. Turner, and R. K. Wang, “Penetration kinetics of
dimethyl sulphoxide and glycerol in dynamic optical clearing of porcine
skin tissue in vitro studied by Fourier transform infrared spectroscopic
imaging,” J. Biomed. Opt. 13(2), 021105 (2008).
144. A. K. Bui, R. A. McClure, J. Chang, C. Stoianovici, J. Hirshburg, A. T.
Yeh, and B. Choi, “Revisiting optical clearing with dimethyl sulfoxide
(DMSO),” Lasers Surg. Med. 41, 142–148 (2009).
145. M. Zimmerley, R. A. McClure, B. Choi, and E. O. Potma, “Following
dimethyl sulfoxide skin optical clearing dynamics with quantitative
nonlinear multimodal microscopy,” Appl. Opt. 48(10), D79–D87 (2009).
146. S. Karma, J. Homan, C. Stoianovic, and B. Choi, “Enhanced
fluorescence imaging with DMSO-mediated optical clearing,”
J. Innovative Optical Health Sciences 3(3), 153–158 (2010).
147. Z. Zhi, Z. Han, Q. Luo, and D. Zhu, “Improve optical clearing of skin in
vitro with propylene glycol as a penetration enhancer,” J. Innovative
Optical Health Sciences 2(3), 269–278 (2009).
148. X. Xu, Q. Zhu, and C. Sun, “Combined effect of ultrasound-SLS on skin
optical clearing,” IEEE Photonic Technol. Lett. 20(24), 2117–2119
(2008).
622 Chapter 10
149. H. Zhong, Z. Guo, H. Wei, L. Guo, C. Wang, Y. He, H. Xiong, and

S. Liu, “Synergistic effect of ultrasound and thiazone–PEG 400 on
human skin optical clearing in vivo,” Photochem. Photobiol. 86(3),
732–737 (2010).
150. T. Kurihara-Bergstrom, K. Knutson, L. J. de Noble, and C. Y. Goates,
“Percutaneous absorption enhancement of an ionic molecule by ethanol–
water system in human skin,” Pharm. Res. 7, 762–766 (1990).
151. A. C. Williams and B. W. Barry, “Penetration enhancers,” Adv. Drug
Deliv. Rev. 56, 603–618 (2004).
152. J.-M. Andanson, K. L. A. Chan, and S. G. Kazarian, “High-throughput
spectroscopic imaging applied to permeation through the skin,” Appl.
Spectrosc. 63(5), 512–517 (2009).
153. A. P. Funke, R. Schiller, H. W. Motzkus, C. Gunther, R. H. Muller, and
R. Lipp, “Transdermal delivery of highly lipophilic drugs: in vitro fluxes
of antiestrogens, permeation enhancers, and solvents from liquid
formulations,” Pharm. Res. 19(5), 661–668 (2002).
154. A. Pagnoni, A. Knuettel, P. Welker, M. Rist, T. Stoudemayer, L. Kolbe,
I. Sadiq, and A. M. Kligman, “Optical coherence tomography in
dermatology,” Skin Res. Technol. 5(2), 83–87 (1999).
155. R. Samatham, K. G. Phillips, and S. L. Jacques, “Assessment of optical
clearing agents using reflectance-mode confocal scanning laser micros-
copy,” J. Innovative Opt. Health Sci. 3(3), 183–188 (2010).
156. P. Liu, Y. Huang, Z. Guo, J. Wang, Z. Zhuang, and S. Liu,
“Discrimination of dimethyl sulphoxide diffusion coefficient in the
process of optical clearing by confocal micro-Raman spectroscopy,”
J. Biomed. Opt. 18(2), 020507 (2013).
157. G. Terentyuk, E. Panfilova, V. Khanadeev, D. Chumakov, E. Genina,
A. Bashkatov, V. Tuchin, N. Khlebtsov, and B. Khlebtsov, “Gold
nanorods with hematoporphyrin-loaded silica shell for dual-modality
photodynamic and photothermal treatment of tumors in vivo,”
NanoResearch 7(3), 325–337 (2014).
158. O. Stumpp, A. J. Welch, and J. Neev, “Enhancement of transdermal skin
clearing agent delivery using a 980-nm diode laser,” Lasers Surg. Med.
37, 278–285 (2005).
159. A. Tezel and S. Mitragotri, “Interaction of inertial cavitation bubbles
with stratum corneum lipid bilayers during low-frequency sonophor-
esis,” Biophys. J. 85, 3502–3512 (2003).
160. A. K. Nugroho, G. L. Li, M. Danhof, and J. A. Bouwstra,
“Transdermal iontophoresis of rotigotine across human stratum
corneum in vitro: influence of pH and NaCl concentration,” Pharm.
Res. 21(5), 844–850 (2004).
161. O. Stumpp and A. J. Welch, “Injection of glycerol into porcine skin for
optical skin clearing with needle-free injection gun and determination of
agent distribution using OCT and fluorescence microscopy,” Proc. SPIE

4949, 44–50 (2003).
162. H.-J. Weigmann, J. Lademann, S. Schanzer, U. Lindemann, R. von
Pelchrzim, H. Schaefer, W. Sterry, and V. Shah, “Correlation of the
local distribution of topically applied substances inside the stratum
corneum determined by tape stripping to differences in bioavailability,”
Skin Pharmacol. Appl. Skin Physiol. 14, 93–103 (2001).
163. W. R. Lee, R. Y. Tsai, C. L. Fang, C. J. Liu, C. H. Hu, and J. Y. Fang,
Microdermabrasion as a novel tool to enhance drug delivery via the skin:
an animal study,” J. Dermatol. Surg. 32, 1013–1022 (2006).
164. A. N. Bashkatov, E. A. Genina, V. V. Tuchin, and G. B. Altshuler,
“Skin optical clearing for improvement of laser tattoo removal,” Laser
Physics 19(6), 1312–1322 (2009).
165. X. Xu and Q. Zhu, “Feasibility of sonophoretic delivery for effective
skin optical clearing,” IEEE Trans. Biomed. Eng. 55(4), 1432–1437
(2008).
166. X. Xu and Q. Zhu, “Sonophoretic delivery for contrast and depth
improvement in skin optical coherence tomography,” IEEE J. Sel. Top.
Quantum Electron. 14(1), 56–61 (2008).
167. I. Lavon, N. Grossman, J. Kost, E. Kimmel, and G. Enden, “Bubble
growth within the skin by rectified diffusion might play a significant role
in sonophoresis,” J. Controlled Release 117(2), 246–255 (2007).
168. X. Xu and C. Sun, “Ultrasound enhanced skin optical clearing:
microstructural changes,” J. Innovative Opt. Health Sci. 3(3), 189–194
(2010).
169. A. K. Nugroho, O. Della Pasqua, M. Danhof, and J. A. Bouwstra,
“Compartmental modeling of transdermal iontophoretic transport: I.
in vitro model derivation and application,” Pharmaceutical Research
21(11), 1974–1984 (2004).
170. J. Wang, X. Zhou, S. Duan, Z. Chen, and D. Zhu, “Improvement of
in vivo rat skin optical clearing with chemical penetration enhancers,”
Proc. SPIE 7883, 78830Y (2011).
171. Yu. P. Sinichkin and S. R. Utz, In vivo reflectance and fluorescence
spectroscopy of human skin, Saratov: Saratov University Press (2001) (in
Russian).
172. J. M. Schmitt and G. Kumar, “Optical scattering properties of soft
tissue: a discrete particle model,” Appl. Opt. 37(13), 2788–2797 (1998).
173. G. A. Askar’yan, “Enhancement of transmission of laser and other
radiation by soft turbid physical and biological media,” Sov. J. Quantum
Electron. 12(7), 877–880 (1982).
174. E. K. Chan, B. Sorg, D. Protsenko, M. O’Neil, M. Motamedi, and A. J.
Welch, “Effects of compression on soft-tissue optical properties,” IEEE
J. Sel. Topics in Quantum Electr. 2(4), 943–950 (1996).
624 Chapter 10
175. H. Shangguan, S. A. Prahl, S. L. Jacques, and L. W. Casperson,

“Pressure effects on soft tissues monitored by changes in tissue optical
properties,” Proc. SPIE 3254, 366–371 (1998).
176. Yu. P. Sinichkin, S. R. Uts, and E. A. Pilipenko, “Spectroscopy of
human skin in vivo: 1. Reflection spectra,” Optics and Spectroscopy
80(2), C. 228–234 (1996).
177. B. W. Murphy, R. J. Webster, B. A. Turlach, C. J. Quirk, C. D. Clay,
P. J. Heenan, and D. D. Sampson, “Toward the discrimination of early
melanoma from common and dysplastic nevus using fiber optic diffuse
reflectance spectroscopy,” J Biomed. Opt. 10(6), 064020 (2005).
178. W. Chen, R. Liu, K. Xu, and R. K. Wang, “Influence of contact state on
NIR diffuse reflectance spectroscopy in vivo,” J. Phys. D: Appl. Phys. 38,
2691–2695 (2005).
179. L. L. Randeberg, “Diagnostic applications of diffuse reflectance
spectroscopy,” PhD thesis, Norwegian University of Science and
Technology, Trondheim, Norway (2005).
180. S. A. Carp, T. Kauffman, Q. Fang, E. Rafferty, R. Moore, D. Kopans,
and D. Boas, “Compression-induced changes in the physiological state
of the breast as observed through frequency domain photon migration
measurements,” J. Biomed. Opt. 11(6), 064016 (2006).
181. R. Reif, M. S. Amorosino, K. W. Calabro, O. A’Amar, S. K. Singh, and
I. J. Bigio, “Analysis of changes in reflectance measurements on
biological tissues subjected to different probe pressures,” J. Biomed.
Opt. 13(1), 010502 (2008).
182. Y. Ti and W. C. Lin, “Effects of probe contact pressure on in vivo optical
spectroscopy,” Opt. Express 16(6), 4250–4262 (2008).
183. A. Cerussi, S. Siavoshi, A. Durkin, C. Chen, W. Tanamai, D. Hsiang,
and B. J. Tromberg, “Effect of contact force on breast tissue optical
property measurements using a broadband diffuse optical spectroscopy
handheld probe,” Appl. Opt. 48, 4270–4277 (2009).
184. J. A. Delgado Atencio, E. E. Orozco Guillén, S. Vázquezy Montiel,
M. Cunill Rodríguez, J. Castro Ramos, J. L. Gutiérrez, and F. Martínez,
“Influence of probe pressure on human skin diffuse reflectance
spectroscopy measurements,” Optical Memory & Neural Networks
(Information Optics) 18(1), 6–14 (2009).
185. L. Lim, B. Nichols, N. Rajaram, and J. W. Tunnell, “Probe pressure
effects on human skin diffuse reflectance and fluorescence spectroscopy
measurements,” J. Biomed. Opt. 16(1), 011012 (2011).
186. S. Ruderman, A. J. Gomes, V. Stoyneva, J. D. Rogers, A. J. Fought,
B. D. Jovanovic, and V. Backman, “Analysis of pressure, angle and
temporal effects on tissue optical properties from polarization-gated
spectroscopic probe measurements,” Biomedical Optics Express 1(2),
489–499 (2010).
187. Yu. P. Sinichkin, S. R. Uts, I. V. Meglinskii, and E. A. Pilipenko,

“Spectroscopy of human skin in vivo: II. Fluorescence spectra,” Optics
and Spectroscopy 80(3). C. 383–389 (1996).
188. A. Nath, K. Rivoire, S. Chang, D. Cox, E. N. Atkinson, M. Follen, and
R. Richards-Kortum, “Effect of probe pressure on cervical
fluorescence spectroscopy measurements,” J. Biomed. Opt. 9(3), 523–533
(2004).
189. K. Rivoire, A. Nath, D. Cox, E. N. Atkinson, R. Richards-Kortum, and
M. Follen, “The effects of repeated spectroscopic pressure measurements
on fluorescence intensity in the cervix,” Am. J. Obstet. Gynecol. 191(5),
1606–1617 (2004).
190. A. Izquierdo-Roman, W. C. Vogt, L. Hyacinth, and C. G. Rylander,
“Mechanical tissue optical clearing technique increases imaging resolu-
tion and contrast through ex vivo porcine skin,” Lasers Surg. Med. 43,
814–823 (2011).
191. V. V. Sapozhnikova, R. V. Kuranov, I. Cicenaite, R. O. Esenaliev, and
D. S. Prough, “Effect on blood glucose monitoring of skin pressure
exerted by an optical coherence tomography probe,” J. Biomed. Opt.
13(2), 021112 (2008).
192. M. Y. Kirillin, P. D. Agrba, and V. A. Kamensky, “In vivo study of the
effect of mechanical compression on formation of OCT images of human
skin,” J. Biophotonics 3(12), 752–758 (2010).
193. A. A. Gurjarpadhye, W. C. Vogt, Y. Liu, and C. G. Rylander, “Effect of
localized mechanical indentation on skin water content evaluated using
OCT,” Int. J. Biomed. Imag. 2011, 817250 (2011).
194. M. H. Khan, B. Choi, S. Chess, K. M. Kelly, J. McCullought, and J. S.
Nelson, “Optical clearing of in vivo human skin: Implications for light-
based diagnostic imaging and therapeutics,” Lasers Surg. Med. 34(2),
83–85 (2004).
195. H. Kang, T. Son, J. Yoon, K. Kwon, J. S. Nelson, and B. Jung,
“Evaluation of laser beam profile in soft tissue due to compression,
glycerol, and micro-needling,” Laser Surg. Med. 40(8), 570–575
(2008).
196. C. G. Rylander, T. E. Milner, S. A. Baranov, and J. S. Nelson,
“Mechanical tissue optical clearing devices: enhancement of light
penetration in ex vivo porcine skin and adipose tissue,” Lasers Surg.
Med. 40(10), 688–694 (2008).
197. I. V. Ermakov and W. Gellermann, “Dermal carotenoid measurements
via pressure mediated reflection spectroscopy,” J. Biophotonics 5(7),
559–570 (2012).
198. L. E. Dolotov and Yu. P. Sinichkin, “Features of applying fiber-optic
sensors in spectral measurements of biological tissues,” Optics and
Spectroscopy 115(2), 187–192 (2013).
626 Chapter 10
199. C. Li, J. Jiang, and K. Xu, “The variations of water in human tissue
under certain compression: studied with diffuse reflectance spectros-
copy,” J. Innov. Opt. Health Sci. 6(1), 1350005 (2013).
200. K. A. Martin, “Direct measurement of moisture in skin by NIR
spectroscopy,” J. Soc. Cosm. Chem. 44, 249–261 (1993).
201. C. W. J. Oomens, D. H. Vancampen, and H. J. Grootenboer,
“A mixture approach to the mechanics of skin,” J. Biomech. 20(9),
877–885 (1987).
202. A. Hidenobu and E. Mariko, “Non-contact skin moisture measurement
based on near-infrared spectroscopy,” Appl. Spectrosc. 58, 1439–1446
(2004).
203. A. N. Bashkatov, E. A. Genina, and V. V. Tuchin, “Optical properties of
skin, subcutaneous and muscle tissues, a review,” J. Innovative Opt.
Health Sci. 14(1), 9–38 (2011).
204. T. L. Troy and S. N. Thennadil, “Optical properties of human skin in the
NIR wavelength range of 1000–2200 nm,” J. Biomed. Opt. 6, 167–176
(2001).
205. I. Yu. Yanina, G. V. Simonenko, and V. V. Tuchin, “Destructive fat
tissue engineering using photodynamic and selective photothermal
effects,” Proc SPIE 7179, 71790C (2009).
206. V. A. Doubrovsky, I. Yu. Yanina, and V. V. Tuchin, “Inhomogeneity of
photo-induced fat cell lipolysis,” Proc. SPIE 7999, 79990M (2011).
207. B. C. Wilson, “Photodynamic therapy/diagnostics: Principles, practice and
advances,” in Handbook of Photonics for Biomedical Science, V. V. Tuchin,
Ed., CRC Press, Taylor & Francis Group, London, 649–686 (2010).
208. M. Wanner, M. Avram, D. Gagnon, M. C. Mihm, Jr., D. Zurakowski,
K. Watanabe, Z. Tannous, R. R. Anderson, and D. Manstein, “Effects
of noninvasive, 1210 nm laser exposure on adipose tissue: results of a
human pilot study,” Lasers Surg. Med. 41, 401–407 (2009).
209. W. R. Chen, R. L. Adams, S. Heaton, D. T. Dickey, K. E. Bartels, and
R. E. Nordquist, “Chromophore-enhanced laser-tumor tissue photo-
thermal interaction using an 808-nm diode laser,” Cancer Lett. 88, 15–19
(1995).
210. W. R. Chen, R. L. Adams, A. K. Higgins, K. E. Bartels, and R. E.
Nordquist, “Photothermal effects on murine mammary tumors using
indocyanine green and an 808-nm diode laser: an in vivo efficacy,”
Cancer Lett. 98, 169–173 (1996).
211. S. Fickweiler, R. M. Szeimies, W. Baumler, P. Steinbach, S. Karrer,
A. E. Goetz, C. Abels, F. Hofstadter, and M. Landthaler, “Indocyanine
green: intracellular uptake and phototherapeutic effects in vitro,”
J. Photochem. Photobiol. B 38, 178–183 (1997).
212. V. V. Tuchin, E. A. Genina, A. N. Bashkatov, G. V. Simonenko, O. D.
Odoevskaya, and G. B. Altshuler, “A pilot study of ICG laser therapy of
acne vulgaris: photodynamic and photothermolysis treatment,” Lasers

Surg. Med. 33(5), 296–310 (2003).
213. E. A. Genina, A. N. Bashkatov, G. V. Simonenko, O. D. Odoevskaya,
V. V. Tuchin, and G. B. Altshuler, “Low-intensity Indocyanine green -
laser phototherapy of acne vulgaris: Pilot study,” J. Biomed. Opt. 9(4),
828–834 (2004).
214. V. I. Kochubey, T. V. Kulyabina, V. V. Tuchin, and G. B. Altshuler,
“Spectral characteristics of indocyanine green upon its interaction with
biological tissues,” Optics and Spectroscopy 99(4), 560–566 (2005).
215. E. A. Genina, A. N. Bashkatov, Yu. P. Sinichkin, V. I. Kochubey, N. A.
Lakodina, G. B. Altshuler, and V. V. Tuchin, “In vitro and in vivo study
of dye diffusion into the human skin and hair follicles,” J. Biomed. Opt.
7(3), 471–477 (2002).
216. E. Engel, R. Schraml, T. Maisch, K. Kobuch, B. König, R.-M. Szeimies,
J. Hillenkamp, W. Bäumler, and R. Vasold, “Light-induced decomposi-
tion of indocyanine green,” Invest. Ophthalmol. Vis. Sci. 49, 1777–1783
(2008).
217. M. Sznitowska, “The influence of ethanol on permeation behavior of
the porous pathway in the stratum corneum,” Int. J. Pharmacol. 137,
137–140 (1996).
218. A. K. Levang, K. Zhao, and J. Singh, “Effect of ethanol/propylene
glycol on the in vitro percutaneous absorption of aspirin, biophysical
changes and macroscopic barrier properties of the skin,” Int. J. Pharm.
181, 255–263 (1999).
219. C. A. Squier, M. J. Kremer, and P. W. Wertz, “Effect of ethanol on lipid
metabolism and epidermal permeability barrier of skin and oral mucosa
in the rat,” J. Oral Pathol. Med. 32, 595–599 (2003).
220. V. A. Dubrovskii, B. A. Dvorkin, I. Yu. Yanina, and V. V. Tuchin,
“Photoaction upon adipose tissue cells in vitro,” Cell and Tissue Biology
5(5), 520–529 (2011).
221. B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, and J. D. Watson,
Molecular Biology of the Cell, 2nd edition, Vol. 1, Garland Publishing.
Inc. New York London (1989).
222. A. L. Lehninger, Principles of Biochemistry, Vol. 2, Worth Publisher,
Inc. (1982).
223. L. Stryer, Biochemistry, W.H. Freeman and Company, San Francisco
(1981).
224. I. Yu. Yanina, N. A. Trunina, and V. V. Tuchin, “Photoinduced cell
morphology alterations quantified within adipose tissues by spectral
optical coherence tomography,” J. Biomed. Opt. 18(11), 111407 (2013).
225. R. K. Wang and V. V. Tuchin, “Enhance light penetration in tissue for
high-resolution optical imaging techniques by the use of biocompatible
chemical agents,” J. X-Ray Science and Technol. 10, 167–176 (2002).
628 Chapter 10
226. R. K. Wang and J. B. Elder, “Propylene glycol as a contrasting agent for

optical coherence tomography to image gastrointestinal tissues,” Lasers
Surg. Med. 30(3), 201–208 (2002).
227. H. Xiong, Z. Guo, C. Zeng, L. Wang, Y. He, and S. Liu, “Application of
hyperosmotic agent to determine gastric cancer with optical coherence
tomography ex vivo in mice,” J. Biomed. Opt. 14(2), 024029 (2009).
228. H. Q. Zhong, Z. Y. Guo, H. J. Wei, J. L. Si, L. Guo, Q. L. Zhao, C. C.
Zeng, H. L. Xiong, Y. H. He, and S. H. Liu, “Enhancement of
permeability of glycerol with ultrasound in human normal and cancer
breast tissues in vitro using optical coherence tomography,” Laser
Physics Letters 7(5), 388–395 (2010).
229. Z. Zhu, H. Wei, G. Wu, H. Yang, Y. He, and S. Xie, “Synergistic effect
of hyperosmotic agents and sonophoresis on breast tissue optical
properties and permeability studied with spectral domain optical
coherence tomography,” J. Biomed. Opt. 17(8), 086002 (2012).
230. D. J. Faber, F. J. van der Meer, M. C. G. Aalders, and T. G. van
Leeuwen, “Quantitative measurement of attenuation coefficients of
weakly scattering media using optical coherence tomography,” Optics
Express 12(19), 4353–4365 (2004).
231. P. Lee, W. Gao, and X. Zhang, “Performance of single-scattering model
versus multiple-scattering model in the determination of optical
properties of biological tissue with optical coherence tomography,”
Appl. Opt. 49(18), 3538–3544 (2010).
232. K. V. Larin and V. V. Tuchin, “Monitoring of glucose diffusion in
epithelial tissues with optical coherence tomography,” Chap. 20 in
Handbook of Optical Sensing of Glucose in Biological Fluids and Tissues,
V. V. Tuchin, Ed., Taylor & Francis Group LLC, CRC Press, 635–668
(2009).
233. T. Yamaguchi, N. Omatsu, E. Morimoto, H. Nakashima, K. Ueno,
T. Tanaka, K. Satouchi, F. Hirose, and T. Osumi, “CGI-58 facilitates
lipolysis on lipid droplets but is not involved in the vesiculation of lipid
droplets caused by hormonal stimulation,” J. Lipid Res. 48(5), 1078–
1089 (2007).
234. J. Sharpe, “Optical projection tomography,” Annual Review of
Biomedical Engineering 6, 209–228 (2004).
235. J. Sharpe, U. Ahlgren, P. Perry, B. Hill, A. Ross, J. Hesksher-Sorensen,
R. Baldock, and D. Davidson, “Optical projection tomography as a tool
for 3D microscopy and gene expression studies,” Science 296, 541–545
(2002).
236. T. Alanentalo, A. Asayesh, H. Morrison, C. E. Loren, D. Holmberg,
J. Sharpe, and U. Ahlgren, “Tomographic molecular imaging and 3D
quantification within adult mouse organs,” Nature Methods 4(1), 31–33
(2007).
237. T. Alanentalo, C. E. Loren, A. Larefalk, J. Sharpe, D. Holmberg, and

U. Ahlgren, “High-resolution three-dimensional imaging of islet-infiltrate
interactions based on optical projection tomography assessments of the
intact adult mouse pancreas,” J. Biomed. Opt. 13(5), 054070 (2008).
238. H. Schneckenburger, P. Weber, T. Bruns, and M. Wagner, Advances in
Fluorescence Spectroscopy and Imaging, in Handbook of Photonics for
Medical Science, V. V. Tuchin (Ed.), CRC Press, Taylor & Francis
Group, London, 119–136 (2010).
239. H. Ashkenazi, Z. Malik, Y. Harth, and Y. Nitzan, “Eradication of
Propionibacterium acnes by its endogenic porphyrins after illumination
with high intensity blue light,” FEMS Immunol. and Med. Microbiol. 35,
17–24 (2003).
240. Y. Kotoku, J. Kato, G. Akashi, Y. Hirai, and K. Ishihara, “Bactericidal
effect of a 405-nm diode laser on Porphyromonas gingivalis,” Laser
Physics Letters 6(5), 388–392 (2009).
241. K. König, G. Flemming, and R. Hibst, “Laser-induced autofluores-
cence spectroscopy of dental caries lesion,” Cell. Mol. Biol. 44, 1293–
1300 (1998).
242. E. G. Borisova, T. T. Uzunov, and L. A. Avramov, “Early differentia-
tion between caries and tooth demineralization using laser-induced
autofluorescence spectroscopy,” Lasers Surg. Med. 34, 249–253 (2004).
243. E. Borisova, T. Uzunov, and L. Avramov, “Laser-induced autofluor-
escence study of caries model in vitro,” Lasers Med. Sci. 21(1), 34–41
(2006).
244. R. R. Alfano, W. Lam, H. J. Zarrabi, M. A. Alfano, J. Cordero, D. B.
Tata, and C. E. Swenberg, “Human teeth with and without caries
studied by laser scattering, fluorescence, and absorption spectroscopy,”
IEEE J. Quantum Electronics QE 20(12), 1512–1516 (1984).
245. K. Konig, H. Schneckenburger, and Hibst, “Time-gated in vivo
autofluorescence imaging of dental caries,” Cell. Mol. Biol. 45, 233–
239 (1999).
246. E. Borisova, P. Troyanova, P. Pavlova, and L. Avramov, “Diagnostics
of pigmented skin tumors based on laser-induced autofluorescence and
diffuse reflectance spectroscopy,” Quantum Electronics 38(6), 597–605
(2008).
247. E. Borisova, E. Carstea, L. Cristescu, E. Pavlova, N. Hadjiolov,
P. Troyanova, and L. Avramov, “Light-induced fluorescence spectros-
copy and optical coherence tomography of basal cell carcinoma,”
Journal of Innovative Optical Health Sciences 2(3), 261–268 (2009).
248. S. K. Chang, Y. N. Mirabal, E. N. Atkinson, D. Cox, A. Malpica,
M. Follen, and R. R. Richards-Kortum, “Combined reflectance and
fluorescence spectroscopy for in vivo detection of cervical pre-cancer,”
J. Biomed. Opt. 10(2), 024031 (2005).
630 Chapter 10
249. A. Alimova, A. Katz, V. Sriramoju, Y. Budansky, A. A. Bykov,

R. Zeylikovich, and R. R. Alfano, “Hybrid phosphorescence and
fluorescence native spectroscopy for breast cancer detection,” J. Biomed.
Opt. 12(1), 014004 (2007).
250. A. Gerger, S. Koller, T. Kern, C. Massone, K. Steiger, E. Richtig,
H. Kerl, and J. Smolle, “Diagnostic applicability of in vivo confocal laser
scanning microscopy in melanocytic skin tumors,” J. Invest. Dermatol.
124, 493–498 (2005).
251. S.-J. Tseng, Y.-H. Lee, Z.-H. Chen, H.-H. Lin, C.-Y. Lin, and S.-C.
Tang, “Integration of optical clearing and optical sectioning microscopy
for three-dimensional imaging of natural biomaterial scaffolds in thin
sections,” J. Biomed. Opt. 14(4), 044004 (2009).
252. A. S. Chiang, “Aqueous tissue clearing solution,” US Patent,
US6472216 (2002).
253. Y.-Y. Fu, C.-W. Lin, G. Enikolopov, E. Sibley, A.-S. Chiang, and S.-C.
Tang, “Microtome-free 3-dimensional confocal imaging method for
visualization of mouse intestine with subcellular-level resolution,”
Gastroenterology 137(2), 453–465 (2009).
254. Y.-Y. Fu and S.-C. Tang, “Optical clearing facilitates integrated 3D
visualization of mouse ileal microstructure and vascular network with
high definition,” Microvascular Res. 80, 512–521 (2010).
255. Y.-A. Liu, Y. Chen, A.-S. Chiang, S.-J. Peng, P. J. Pasricha, and S.-C.
Tang, “Optical clearing improves the imaging depth and signal-to-noise
ratio for digital analysis and three-dimensional projection of the human
enteric nervous system,” Neurogastroenterology & Motility 23, e446–
e457 (2011).
256. Y.-Y. Fu, C.-H. Lu, C.-W. Lin, J.-H. Juang, G. Enikolopov, E. Sibley,
A.-S. Chiang, and S.-C. Tang, “Three-dimensional optical method for
integrated visualization of mouse islet microstructure and vascular
network with subcellular-level resolution,” J. Biomed. Opt. 15(4), 046018
(2010).
257. A.-S. Chiang, Y.-C. Liu, S.-L. Chiu, S.-H. Hu, C.-Y. Huang, and
C.-H. Hsieh, “Three dimensional mapping of brain neuropils in the
cockroach Diploptera punctate,” J. Comp. Neurol. 440, 1–11 (2001).
258. J. Huisken and D. Y. R. Stainier, “Selective plane illumination
microscopy techniques in developmental biology,” Development 136,
1963–1975 (2009).
259. K. Becker, N. Jährling, E. R. Kramer, F. Schnorrer, and H.-U. Dodt,
“Ultramicroscopy: 3D reconstruction of large microscopical specimens,”
J. Biophotonics 1, 36–42 (2008).
260. K. Becker, N. Jährling, S. Saghafi, R. Weiler, and H. U. Dodt,
“Chemical clearing and dehydration of GFP expressing mouse brains,”
PLOS One 7(3), e33916 (2012).
261. O. I. Efimova and K. V. Anokhin, “Increase of optical permeability of

isolated adult mouse brain structures,” Bull. Exp. Biol. Med. 147(1), 4–7
(2009).
262. A. Ertürk, K. Becker, N. Jährling, C. P. Mauch, C. D. Hojer, J. G.
Egen, F. Hellal, F. Bradke, M. Sheng, and H. U. Dodt, “Three-
dimensional imaging of solvent-cleared organs using 3DISCO,” Nature
Protocols 7(11), 1983–1995 (2012).
263. B. Yang, J. B. Treweek, R. P. Kulkarni, B. E. Deverman, C.-K. Chen,
E. Lubeck, S. Shah, L. Cai, and V. Gradinaru, “Single-cell phenotyping
within transparent intact tissue through whole-body clearing,” Cell, 158
(4), 945–958 (2014).
264. K. Becker, N. Jährling, S. Saghafi, and H. U. Dodt, “Dehydration and
clearing of whole mouse brains and dissected hippocampi for
ultramicroscopy,” Cold Spring Harb. Protoc 7 DOI: 10.1101/pdb.
prot075820 (2013).
265. C. Leahy, H. Radhakrishnan, and V. J. Srinivasan, “Volumetric imaging
and quantification of cytoarchitecture and myeloarchitecture with
intrinsic scattering contrast,” Biomed. Opt. Express 4(10), 1978–1990
(2013).
266. K. Chung, J. Wallace, S.-Y. Kim, S. Kalyanasundaram, A. S.
Andalman, T. J. Davidson, J. J. Mirzabekov, K. A. Zalocusky,
J. Mattis, A. K. Denisin, S. Pak, H. Bernstein, C. Ramakrishnan,
L. Grosenick, V. Gradinaru, and K. Deisseroth, “Structural and
molecular interrogation of intact biological systems,” Nature 497,
332–337 (2013).
267. L. Silvestri, A. L. Allegra Mascaro, J. Lotti, L. Sacconi, and F. S.
Pavone, “Advanced optical techniques to explore brain structure and
function,” JIOHS 6 (1) 1230002 (2013).
268. V. Marx, “Microscopy: seeing through tissue,” Nature Methods 11(12),
1209–1214 (2014).
269. E. V. Migacheva, A. B. Pravdin, and V. V. Tuchin, Alterations in
autofluorescence signal from rat skin ex vivo under optical immersion
clearing, JIOHS 3 (3) 147–152 (2010).
270. A. A. Oraevsky, “Optoacoustic Tomography: from fundamentals to
diagnostic imaging of breast cancer,” in Biomedical Photonics Hand-
book, Tuan Vo-Dinh (Ed.), Taylor & Francis Group, LLC, Boca Raton,
FL, CRC Press Inc., 715–757 (2014).
271. R. O. Esenaliev, I. V. Larina, K. V. Larin, D. J. Deyo, M. Motamedi,
and D. S. Prough, “Optoacoustic technique for noninvasive monitoring
of blood oxygenation: a feasibility study,” Appl. Opt. 41, 4722–4731
(2002).
272. R. G. M. Kolkman, J. H. G. M. Klaessens, E. Hondebrink, J. C. W.
Hopman, F. F. M. de Mul, W. Steenbergen, J. M. Thijssen, and T. G.
632 Chapter 10
van Leeuwen, “Photoacoustic determination of blood vessel diameter,”

Phys. Med. Biol. 49, 4745–4756 (2004).
273. H. F. Zhang, K. Maslov, and L. H. Wang, “In vivo imaging of
subcutaneous structures using functional photoacoustic microscopy,”
Nature Protocols 2(4), 797–804 (2007).
274. Y. Zhou, J. J. Yao, and L. H. Wang, “Optical clearing-aided
photoacoustic microscopy with enhanced resolution and imaging depth,”
Opt. Lett. 38(14), 2592–2595 (2013).
275. H. A. MacKenzie, H. S. Ashton, S. Spiers, Y. Shen, S. S. Freeborn,
J. Hannigan, J. Lindberg, and P. Rae, “Advances in photoacoustic
noninvasive glucose testing,” Clin. Chem. 45, 1587–1595 (1999).
276. M. Kinnunen and R. Myllyla, “Application of optical coherence
tomography, pulsed photoacoustic technique, and time-of-flight tech-
nique to detect changes in the scattering properties of a tissue-simulating
phantom,” J. Biomed. Opt. 13(2), 024005 (2008).
277. O. S. Khalil, “Non-invasive glucose measurement technologies: an
update from 1999 to the dawn of the new millennium,” Diabetes
Technol. Ther. 6, 660–697 (2004).
278. V. V. Tuchin, A. Tárnok, and V. P. Zharov, “In vivo Flow Cytometry: A
Horizon of Opportunities,” Cytometry A, 79A, N10, 737–745 (2011).
279. V. V. Tuchin, “In vivo optical flow cytometry and cell imaging,” Rivista
Del Nuovo Cimento, 37(7), 375–416 (2014).
280. Y. A. Menyaev, D. A. Nedosekin, M. Sarimollaoglu, M. A. Juratli, E. I.
Galanzha, V. V. Tuchin, and V. P. Zharov, “Skin optical clearing for
in vivo photoacoustic flow cytometry,” Biomed. Opt. Express 4 (12),
3030–3041 (2013).
281. F. S. Pavone and P. J. Campagnola (eds.), Second Harmonic Generation
Imaging (CRC Press, Taylor & Francis Group, Boca Raton, London,
NY, (2014).
282. R. Cicchi, S. Sestini, V. De Giorgi, D. Massi, T. Lotti, and F. S. Pavone,
“Nonlinear laser imaging of skin lesions,” J. Biophotonics 1(1), 62–73
(2008).
283. P. J. Campagnola, H. A. Clark, W. A. Mohler, A. Lewis, and L. M.
Loew, “Second-harmonic imaging microscopy of living cells,”
J. Biomed. Opt. 6(3), 277–286 (2001).
284. P. J. Campagnola, A. C. Millard, M. Terasaki, P. E. Hoppe, C. J.
Malone, and W. A. Mohler, “Three-dimensional high-resolution second-
harmonic generation imaging of endogenous structural proteins in
biological tissues,” Biophys. J. 82, 493–508 (2002).
285. O. Nadiarnykh and P. J. Campagnola, “SHG and optical clearing,” in
Second Harmonic Generation Imaging, F. S. Pavone and P. J.
Campagnola (eds.), CRC Press, Taylor & Francis Group, Boca Raton,
London, NY, 169–189 (2014).
286. P. Bianchini and A. Diaspro, “Three-dimensional (3D) backward and

forward second harmonic generation (SHG) microscopy of biological
tissues,” J. Biophotonics 1(6), 443–450 (2008).
287. V. Ajeti, O. Nadiarnykh, S. M. Ponik, P. J. Keely, K. W. Eliceiri, and
P. J. Campagnola, “Structural changes in mixed Col I/Col V collagen
gels probed by SHG microscopy: implications for probing stromal
alterations in human breast cancer,” Biomedical Optics Express 2(8),
2307–2316 (2011).
288. T. Yasui, Y. Tohno, and T. Araki, “Characterization of collagen
orientation in human dermis by two-dimensional second-harmonic-
generation polarimetry,” J. Biomed. Opt. 9(2), 259–264 (2004).
289. H. G. Breunig, M. Weinigel, R. Buckle, M. Kellner-Hofer,
J. Lademann, M. E. Darvin, W. Sterry, and K. Konig, “Clinical
coherent anti-Stokes Raman scattering and multiphoton tomography of
human skin with a femtosecond laser and photonic crystal fiber,” Laser
Physics Letters 10, 025604 (2013).
290. G. Deka, W.-W. Wu, and F.-J. Kao, “In vivo wound healing diagnosis
with second harmonic and fluorescence lifetime imaging,” J. Biomed.
Opt. 18(6), 061222 (2013).
291. A. Ghazaryan, H. F. Tsai, G. Hayrapetyan, W.-L. Chen, Y.-F. Chen,
M. Y. Jeong, C.-S. Kim, S.-J. Chen, and C.-Y. Dong, “Analysis
of collagen fiber domain organization by Fourier second
harmonic generation microscopy,” J. Biomed. Opt. 18(3), 031105
(2013).
292. R. Cicchi, L. Sacconi, and F. Pavone “Nonlinear Imaging of Tissues,”
Chap. 20 in Handbook of Photonics for Biomedical Science, V. V. Tuchin
(Ed.), CRC Press, Taylor & Francis Group, Boca Raton, 509–545
(2010).
293. M. Muller, J. A. Squier, T. Wilson, and G. Brakenhoff, “3D microscopy
of transparent objects using third-harmonic generation,” J. Microsc. 191,
266–272 (1998).
294. G. Hall, K. W. Eliceiri, and P. J. Campagnola, “Simultaneous
determination of the second-harmonic generation emission directionality
and reduced scattering coefficient from three-dimensional imaging of
thick tissues,” J. Biomed. Opt. 18(1), 116008 (2013).
295. M. Zimmerley, R. A. McClure, B. Choi, and E. O. Potma,
“Following dimethyl sulfoxide skin optical clearing dynamics with
quantitative nonlinear multimodal microscopy,” Appl. Opt. 48(10),
D79–D87 (2009).
296. M. V. Schulmerich, J. H. Cole, K. A. Dooley, M. D. Morris, J. M.
Kreider, and S. A. Goldstein, “Optical clearing in transcutaneous
Raman spectroscopy of murine cortical bone tissue,” J. Biomed. Opt.
13(2), 021108 (2008).
634 Chapter 10
297. D. Huang, W. Zhang, H. Zhong, H. Xiong, X. Guo, and Z. Guo,

“Optical clearing of porcine skin tissue in vitro studied by Raman
microspectroscopy,” J. Biomed. Opt. 17(1), 015004 (2012).
298. P. J. Caspers, A. C. Williams, E. A. Carter, H. G. M. Edwards,
B. W. Barry, H. A. Bruining, and G. J. Puppels, “Monitoring the
penetration enhancer dimethyl sulfoxide in human stratum corneum
in vivo by confocal Raman spectroscopy,” Pharm. Res. 19(10), 1577–
1580 (2002).
299. A. A. Angeluts, A. V. Balakin, M. G. Evdokimov, M. N. Esaulkov,
M. M. Nazarov, I. A. Ozheredov, D. A. Sapozhnikov, P. M. Solyankin,
O. P. Cherkasova, and A. P. Shkurinov, “Characteristic responses of
biological and nanoscale systems in the terahertz frequency range,”
300. M. Nazarov, A. Shkurinov, V. V. Tuchin, and X.-C. Zhang, “Terahertz
tissue spectroscopy and imaging,” Chapter 17 in Handbook of Photonics
for Biomedical Science, V. V. Tuchin, Ed., CRC Press, London, Taylor
& Francis Group, 592–617 (2010).
301. M. M. Nazarov, A. P. Shkurinov, E. A. Kuleshov, and V. V. Tuchin,
“Terahertz time-domain spectroscopy of biological tissues,” Quantum
Electronics 38(7), 647–654 (2008).
302. G. M. Png, J. W. Choi, B. W.-H. Ng, S. P. Mickan, D. Abbott, and
X.-C. Zhang, “The impact of hydration changes in fresh bio-tissue on
THz spectroscopic measurements,” Phys. Med. Biol. 53, 3501–3517
(2008).
303. A. V. Borodin, V. Ya. Gayvoronsky, O. D. Kachkovsky,
Ya. A. Prostota, A. V. Kargovskii, M. M. Nazarov, D. A. Sapozhnikov,
Yu. L. Slominskii, I. N. Smirnova, and A. P. Shkurinov, “Structure
sensitive changes in the terahertz absorption spectra of merocyanine dye
derivatives,” Optics and Spectroscopy 107(4), 505–514 (2009).
304. C. S. Joseph, R. Patel, V. A. Neel, R. H. Giles, and A. N.
Yaroslavsky, “Imaging of ex vivo nonmelanoma skin cancers in the
optical and terahertz spectral regions,” J. Biophotonics 7(5), 295–303
(2014).
305. A. S. Kolesnikov, E. A. Kolesnikova, A. P. Popov, M. M. Nazarov,
A. P. Shkurinov, and V. V. Tuchin, “In vitro terahertz monitoring of
muscle tissue dehydration under the action of hyperosmotic agents,”
306. R. E. Beck and J. S. Schultz, “Hindrance of solute diffusion within
membranes as measured with microporous membranes of known pore
geometry,” Biochem. Biophys. Acta 255, 272–303 (1972).
307. A. Kotyk and K. Janacek, Membrane Transport: An Interdisciplinary
Approach, Plenum Press, New York (1977).
308. I. H. Blank, J. Moloney, A. G. Emslie, I. Simon, and C. Apt, “The

diffusion of water across the stratum corneum as a function of its water
content,” J. Invest. Dermatol. 82, 188–194 (1984).
309. B. Sennhenn, K. Giese, K. Plamann, N. Harendt, and K. Kolmel, “In vivo
evaluation of the penetration of topically applied drugs into human
skin by spectroscopic methods,” Skin Pharmacol. 6, 152–160 (1993).
310. K. D. Peck, A.-H. Ghanem, and W. I. Higuchi, “Hindered diffusion of
polar molecules through and effective pore radii estimates of intact and
ethanol treated human epidermal membrane,” Pharmaceutical Research
11(9), 1306–1314 (1994).
311. T. Inamori, A.-H. Ghanem, W. I. Higuchi, and V. Srinivasan,
“Macromolecule transport in and effective pore size of ethanol
pretreated human epidermal membrane,” Int. J. Pharmaceutics 105,
113–123 (1994).
312. R. Bertram and M. Pernarowski, “Glucose diffusion in pancreatic islets
of Langerhans,” Biophys. J. 74, 1722–1731 (1998).
313. P. Gribbon and T. E. Hardingham, “Macromolecular diffusion of
biological polymers measured by confocal fluorescence recovery after
photobleaching,” Biophys. J. 75, 1032–1039 (1998).
314. A. N. Yaroslavskaya, I. V. Yaroslavsky, C. Otto, G. J. Puppels,
H. Guindam, G. F. J. M. Vrensen, J. Greve, and V. V. Tuchin, “Water
exchange in human eye lens monitored by confocal Raman microspec-
troscopy,” Biophysics 43(1), 109–114 (1998).
315. E. E. Alanis, G. G. Romero, and C. C. Martinez, “Interferometric
measurement of diffusion coefficients through a scanning laser beam,”
Opt. Eng. 39(3), 744–750 (2000).
316. S. Papadopoulos, K. D. Jurgens, and G. Gros, “Protein diffusion in
living skeletal muscle fibers: dependence on protein size, fiber type, and
contraction,” Biophys. J. 79, 2084–2094 (2000).
317. A. E. Kamholz, E. A. Schilling, and P. Yager, “Optical measurement of
transverse molecular diffusion in a microchannel,” Biophys. J. 80, 1967–
1972 (2001).
318. S. S. Olmsted, J. L. Padgett, A. I. Yudin, K. J. Whaley, T. R.
Moench, and R. A. Cone, “Diffusion of macromolecules and virus-
like particles in human cervical mucus,” Biophys. J. 81, 1930–1937
(2001).
319. J. Kusba, L. Li, I. Gryczynski, G. Piszczek, M. Johnson, and J. R.
Lakowicz, “Lateral diffusion coefficients in membranes measured by
resonance energy transfer and a new algorithm for diffusion in two
dimensions,” Biophys. J. 82, 1358–1372 (2002).
320. W. Heller, “Remarks on refractive index mixture rules,” J. Phys. Chem.
69(4), 1123–1129 (1965).
636 Chapter 10
321. J. S. Maier, S. A. Walker, S. Fantini, M. A. Franceschini, and E.

Gratton, “Possible correlation between blood glucose concentration and
the reduced scattering coefficient of tissues in the near infrared,” Opt.
Lett. 19(24), 2062–2064 (1994).
322. M. Kohl, M. Esseupreis, and M. Cope, “The influence of glucose
concentration upon the transport of light in tissue-simulating phan-
toms,” Phys. Med. Biol. 40, 1267–1287 (1995).
323. W. H. Press, S. A. Tuekolsky, W. T. Vettering, and B. P. Flannery,
Numerical recipes in C: the art of scientific computing, Cambridge:
Cambridge University Press (1992).
324. I. S. Grigoriev and E. Z. Meylikhov (Eds.), Physical Values: Handbook,
Moscow: EnergoAtomIzdat (1991).
Elina A. Genina is an associate professor in the Departments

of Optics and Biophotonics of Saratov National Research
State University. She received a PhD in biophysics from
Saratov National Research State University, Saratov,
Russia, in 2002. She has authored more than 200 peer-
reviewed publications and 7 book chapters on Biomedical
Optics. E.A. Genina is a reviewer and a guest-editor of peer-
reviewed journals, such as J. Biomed. Opt., J. Innovative
Optical Health Sciences, Quantum Electronics, Optics and Spectroscopy, etc.
She is a scientific secretary of the International School for Junior Scientists
and Students on Optics, Laser Physics and Biophotonics (Saratov Fall
Meeting). Her research interests include biomedical optics, laser medicine,
and the development of methods for controlling tissue absorption and
scattering properties for medical optical diagnostics and therapy.
Alexey N. Bashkatov received his PhD in biophysics from

Saratov State University (SSU), Saratov, Russia, in 2002. He
has authored 200 invited and contributed publications, 7
book chapters, and over 250 invited and contributed
conference presentations. From 2002 to present he is
Associate Professor of the Departments of Optics and
Biophotonics of Saratov National Research State University.
He is the Head of Laboratory of Biomedical Optics of
Research-Educational Institute of Optics and Biophotonics at SSU and Senior
Researcher of Interdisciplinary Laboratory of Biophotonics at Tomsk State
University. His current research interests include study of optical properties of
tissues; spectroscopy; development of methods for control of tissue absorption
and scattering properties for medical optical diagnostics and therapy; and
modeling of light propagation in turbid media.
Yuri P. Sinichkin is a full professor of Chair of Optics and

He is head of a laboratory of Optical Medical Diagnostics of
Research-Educational Institute of Optics and Biophotonics
of SSU and senior researcher of Interdisciplinary Laboratory
of Biophotonics at Tomsk State University. His research
interests include biophotonics, tissue optics, spectroscopy,
medical optical diagnostics, and laser medicine. He is a
member of SPIE and IEEE.
Irina Yu. Yanina received her PhD degree in biophysics from

Saratov State University (SSU), Saratov, Russia, in 2013.
She has authored 17 papers in peer-reviewed journals and 18
papers in conference proceedings. From 2010 to present, she
is an Associate Fellow Researcher of Laboratory of
Biomedical Optics of Research-Educational Institute of
Optics and Biophotonics at SSU. Her research interests
include the development of optical methods of fat tissue
destructive engineering, photodynamic/photothermal therapy, biomedical
optics, drug delivery, spectroscopy and imaging in biomedicine, and optical
and laser measurements.

He is also the head of laboratory, Institute of Precision
and OSA, and has been awarded Honored Science Worker of the Russia, SPIE
Educator Award, FiDiPro (Finland), Chime Bell Prize of Hubei Province
(China), and Joseph W. Goodman Book Writing Award (OSA/SPIE).
Index
3-D imaging of solvent-cleared Cauchy infinitesimal strain tensor,
organs (3DISCO), 602 542
cell morphology, 40
A cerebral blood flow (CBF), 579
absorption, 247 cerebral metabolic rate of oxygen
acoustically modulated speckle (CMRO2), 326
imaging (amSI), 556 chromaticity coordinates, 135
adaptive computation of contrast for chromophore, 104
laser speckle contrast analysis circulating lymphatic cells, 359
(adLASCA), 470 circulating tumor cell (CTC), 346
aggregation force, 26 coherence condition, 321
aging, 112 coherent anti-Stokes Raman
angiography, 345 spectroscopy (CARS), 598
atomic force microscopy (AFM), collagen, 111
21, 511 collagen dissociation, 579
attenuated total reflectance Fourier collagen fibers, 591
transform infrared (ATR FTIR) color analysis, 134
spectroscopy, 166, 200 color imaging, 139
autofluorescence (AF), 108 compression OCE, 554
autoregressive (AR) spectral confocal laser scanning microscopy
estimator, 526 (CLSM), 281, 286
Azone, 582 confocal light absorption and
scattering spectroscopic (CLASS)
B microscopy, 91
ballistic photons, 599 confocal microscopy, 170
benzyl benzoate (BABB), 600 confocal Raman spectroscopy, 609
biospeckle, 316, 322 constitutive constants, 512
blood content, 112 constitutive equations, 511
blood flow measurements in Couette chamber, 12
humans, 325 cross-polarization, 142
blood vessel, 104
boundary conditions, 612 D
brightness change constraint dehydration, 574
equation (BCCE), 539 dermatology, 507
Brillouin microscopy, 510 dermis, 104
diffusing-wave spectroscopy
C (DWS), 436
carcinogenesis, 48 diffusion approximation, 446
carcinoma in situ (CIS), 49 diffusion coefficient, 160, 611
cardiology, 406, 510 dimethyl sulfoxide (DMSO), 162,
carotenoids, 229, 233 216
639
640 Index
discrete dipole approximation frequency-domain OCT, 395

(DDA), 53 Fresnel’s reflectance, 118
dissociation, 576
Doppler frequency shift (DFS), 317 G
Doppler OCT, 550 Gladstone and Dale law, 159, 613
Doppler optical coherence green fluorescent protein (GFP),
tomography (DOCT), 333, 402 254
dysplasia, 38 Green–Lagrange strain tensor, 542
E H
elastin, 111 hematocrit, 19
elastography, 505 human skin, 103
electrophoresis, 585 hydroxyl groups, 579
endoscopic polarized scanning hyperosmotic agent, 171
spectroscopy (EPSS), 86 hyperosmotic properties, 576
energy transfer spectroscopy, 274
epidermal stripping, 161 I
epidermis, 104 indocyanine green (ICG), 353, 592
epidermis–dermis junction, 594 infrared (IR), 105
epithelium, 45, 46 infrared spectroscopy, 197
erythema, 112 in-plane strain, 523
erythema index, 126 intravital microscopy (IM), 343
erythema–melanin meter, 144
excitation-emission map (EEM), J
108 Jablonski diagram, 249
extinction coefficient, 119
K
F keratin, 111, 222
fibrous tissue, 577 keratinocyte, 160
Fick law, 611
finite difference time domain L
(FDTD) method, 58 laser Doppler anemometry (LDA),
flow cytometry (FC), 346 333
fluorescence, 248 laser heating, 173
fluorescence imaging, 259 laser speckle, 518
fluorescence quantum yield, 121 laser speckle contrast analysis
Förster resonance energy transfer (LASCA), 436, 558
(FRET), 278 laser tweezers (LT), 21
Fourier transform, 524 laser-induced thermal effects, 358
Fourier-domain OCT (FdOCT), 550 lattice of islets of damage (LID),
fractional microablation, 577 164
free diffusion, 158 lazer tweezers (LT), 22
frequency-domain measurements, light scattering, 51
169
frequency-domain fluorometry, 263
Index 641
light sheet fluorescence microscopy optical coherence elastography
(LSFM), 280 (OCE), 507, 550
light transport, 66 optical coherence tomography
lipid bilayers, 585 (OCT), 78, 169, 391
LSS imaging, 85 optical density (OD), 119
lycopene, 234 optical projection tomography
lymph flow cytometry (LFC), 349 (OPT), 599
lymph flow profile, 353 oximetry, 129
lymphedema, 356 oxygen saturation, 130
lymphography, 345 oxyhemoglobin, 104
M P
mechanical compression, 119, 588 palpation, 506
melanin, 105 passive clarity technique (PACT),
metastatic breast cancer, 366 602
microparticle image velocimetry, penetration depth, 211
318 perfusion assisted agent release in
Mie scattering, 590 situ (PARS), 602
minimum mean square error permeability coefficient, 595
(MMSE), 533 photoacoustic flow cytometry
minimum speckle size, 520 (PAFC), 604
Monte Carlo modeling, 460, 601 photoacoustic imaging, 603
mucosa, 44 photodynamic action (PDA), 593
mucosal tissues, 39 photothermal action, 592
pigmentation index, 141
N Piola–Kirchhoff stress tensor, 542
natural moisturizing factor (NMF), polarization, 142
221 polarization degree, 146
near-infrared (NIR), 107 polarization-sensitive spectroscopy,
nicotine intoxication, 357 143
nitric oxide (NO), 358 polyatomic alcohols, 578
numerical error, 527 polyethylene glycol (PEG), 576
propylene glycol (PG), 162
O pseudo-elastic potential, 513
OCE probes, 556
oleic acid, 162 R
oncology, 412, 507 Raman nonlinear spectroscopy, 608
ophthalmology, 404 Raman spectroscopy, 197
optical clearing, 574 Raman spectrum, 223
optical biopsy, 616 Rayleigh scattering, 53, 590
optical clearing, 153, 346 Rayleigh–Gans approximation, 54
optical clearing agent (OCA), 576 reciprocal spatial difference (RSD),
optical clearing potential (OCP), 324
162 red blood cell (RBC) aggregation, 5,
6, 10
642 Index
red, green, and blue (RGB), 140

reflectance spectrum, 107 V
refractive index matching solution van de Hulst theory, 55, 57
(RIMS), 602 video dimensional analysis (VDA),
retardation spectrum, 515 531
viscoelasticity, 512
S volume fraction, 614
second harmonic generation (SHG),
172 W
sedimentation measurements, 20 water spectral bands, 166
skin barrier properties, 583
skin color, 141 X
sonophoresis, 584 x-ray contrasts, 165
Soret band, 104
speckle fluctuations, 316, 321, 322
speckle interferometer, 331
speckle motion, 528
spectral broadening, 316
spectrometer, 399
square blur rate (SBR), 324
stimulated emission depletion
(STED) microscopy, 290
strain energy density, 513
stratum corneum, 104, 221
sun protection factor (SPF), 150
sunscreen, 150
surface wave/shear wave OCE, 555
T
terahertz spectroscopy, 609
tetrahydrofurane (THF), 602
Thiazone, 582
three-dimensional SHG microscopy,
608
time-domain OCT, 394
tissue engineering, 510
transdermal microchannels, 584
triglyceride, 592
tryptophan, 105
tyrosine, 105
U
ultra-microscopy (UM), 602
ultrasonic treatment, 584
ultraviolet A (UVA), 106
HANDBOOK OF
OPTICAL
BIOMEDICAL
DIAGNOSTICS
SECOND EDITION
Volume 2: Methods
Valery V. Tuchin
EDITOR
Since the publication of the first edition of the Handbook in 2002, optical methods for
biomedical diagnostics have developed in many well-established directions, and new
trends have also appeared. The text has been updated and expanded into two volumes.
Volume 1: Light–Tissue Interaction features eleven chapters, five of which focus on

the fundamental physics of light propagation in turbid media such as biological tissues.
The six following chapters introduce near-infrared techniques for the optical study of
tissues and provide a snapshot of current applications and developments in this
dynamic and exciting field. Topics include the scattering of light in disperse systems, the
optics of blood, tissue phantoms, a comparison between time-resolved and
continuous-wave methods, and optoacoustics.
Volume 2: Methods begins by describing the basic principles and diagnostic

applications of optical techniques based on detecting and processing the scattering,
fluorescence, FT IR, and Raman spectroscopic signals from various tissues, with an
emphasis on blood, epithelial tissues, and human skin. The second half of the volume
discusses specific imaging technologies, such as Doppler, laser speckle, optical
coherence tomography (OCT), and fluorescence and photoacoustic imaging.
P.O. Box 10
Bellingham, WA 98227-0010
ISBN: 9781628419139
SPIE Vol. No.: PM263

Handbook of Optical Biomedical Diagnostics, Vol.2 Methods, 2nd Edition

Uploaded by

Copyright:

Available Formats

Handbook of Optical Biomedical Diagnostics, Vol.2 Methods, 2nd Edition

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Handbook of Optical Biomedical Diagnostics, Vol.2 Methods, 2nd Edition

Uploaded by

Copyright:

Available Formats

HANDBOOK OF

Names: Tuchin, V. V. (Valerii Viktorovich), editor.

Copyright © 2016 Society of Photo-Optical Instrumentation Engineers (SPIE)

All rights reserved. No part of this publication may be reproduced or distributed in

Printed in the United States of America.

III Scattering, Fluorescence, Infrared, and Raman Spectroscopy

1.7.2.1 Measurement of the disaggregation force 24

2 Light Scattering Spectroscopy of Epithelial Tissues: Principles and

3 Reflectance and Fluorescence Spectroscopy of Human Skin in vivo 99

3.4 Combined Reflectance and Fluorescence Spectroscopy

4.2.3.1 Water bending mode and low-wave-number region 199

5.2.1 Intrinsic fluorophores 245

IV Coherent-Domain Methods for Biological Flows and Tissue

6 Laser Speckles, Doppler, and Imaging Techniques for Blood and

6.2.1 Laser Doppler technique 314

7 Real-Time Imaging of Microstructure and Function Using Optical

8 Speckle Technologies for Monitoring and Imaging Tissues and

8.6 Imaging Using Contrast Measurements of Partially Developed

9 Optical Assessment of Tissue Mechanics 497

9.13.1 Variants of OCE 547

on laser–tissue interactions, and the most interesting clinical applications—are

• Part III presents various spectroscopic techniques of tissues based on

6. G. Müller, B. Chance, R. Alfano, et al. (Eds.), Medical Optical

41. F. S. Pavone and P. J. Campagnola (Eds.), Second Harmonic Generation

Yoshihisa Aizu Ekaterina I. Galanzha

Vadim Backman E. A. Genina

David J. Briers Brendan F. Kennedy

Izumi Nishidate Wolfgang S. L. Strauss

Herbert Schneckenburger Naomichi Yokoi

Yuri P. Sinichkin Vladimir P. Zharov

Karl Stock George I. Zonios

accelerating the penetration of the index-matching compounds by enhancing

1.1 Introduction. Microrheological Structure of Blood:

Microscopic observations of RBC motion in in vitro conditions show that

volumes, the RBCs have a higher probability of collisions and, consequently,

Figure 1.1 Number distributions of RBCs in aggregates of different sizes at initial

The morphological studies of RBC aggregates in stasis or under shear

Figure 1.2 Microphotographs showing typical shapes of RBC aggregates at normal

1.2 Importance of Quantitative Measurement of Red Blood Cell

In acute experiments carried out on dogs, the effect of RBC aggregation

It is worth noting here that optical transmission measurements with blood

1.3 Arrangement of a Couette-Chamber-Based Laser

Figure 1.4 Schematic layout of a rotational Couette cell-based erythro-aggregometer.

1.3.1 Measurement procedure

membranes of individual cells. At this phase, the detectors yield an output

1.4 Kinetics of the Aggregation and Disaggregation Process

1.4.1 Determination of the characteristic parameters of the

where Ia is the full amplitude of aggregation; T1, T2 are the characteristic

In analogy with the aggregation process, the kinetics of disaggregation can

1.5 Parameters Influencing the Aggregation and Disaggregation

1.5.2 Effect of blood sample oxygenation

Figure 1.8 Effect of blood sample oxygenation on remission intensity.53

1.5.3 Effect of sedimentation

1.5.4 Effect of hematocrit

unnecessary if the device-specific dependence of the measuring characteristic

1.6 Comparison of Aggregation and Disaggregation

determination of pathological states such as in the case of sedimentation

1.7 Laser Tweezers as a New Tool for Studying RBC

obtaining of detailed kinetics of RBC interaction and provided the grounds

1.7.1 Laser tweezers operation principle and experimental

Figure 2.1 0DMRU RUJDQHOOHV DQG LQFOXVLRQV RI WKH FHOO6HHFRORUSODWHV