MP 13301

Estimating a size-speciﬁc dose for helical head CT examinations using Monte
Carlo simulation methods

Anthony J. Hardya) and Maryam Bostani
Department of Radiology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA 90024, USA
Physics and Biology in Medicine Graduate Program, David Geffen School of Medicine, University of California, Los Angeles,
Los Angeles, CA 90024, USA
Andrew M. Hernandez
Departments of Radiology and Biomedical Engineering, Biomedical Engineering Graduate Group, University of California Davis,
Sacramento, CA 95817, USA
Maria Zankl
Helmholtz Zentrum M€unchen, German Research Center for Environmental Health (GmbH), Institute of Radiation Protection,
Ingolstaedter Landstrasse 1, Neuherberg 85764, Germany
Cynthia McCollough
Department of Radiology, Mayo Clinic, Rochester, MN 55905, USA
Chris Cagnon
John M. Boone
Departments of Radiology and Biomedical Engineering, Biomedical Engineering Graduate Group, University of California Davis,
Sacramento, CA 95817, USA
Michael McNitt-Gray
(Received 10 May 2018; revised 22 October 2018; accepted for publication 23 October 2018;
published xx xxxx xxxx)
Purpose: Size-specific dose estimates (SSDE) conversion factors have been determined by AAPM
Report 204 to adjust CTDIvol to account for patient size but were limited to body CT examinations.
The purpose of this work was to determine conversion factors that could be used for an SSDE for
helical, head CT examinations for patients of different sizes.
Methods: Validated Monte Carlo (MC) simulation methods were used to estimate dose to the center
of the scan volume from a routine, helical head examination for a group of patient models represent-
ing a range of ages and sizes. Ten GSF/ICRP voxelized phantom models and five pediatric voxelized
patient models created from CT image data were used in this study. CT scans were simulated using a
Siemens multidetector row CT equivalent source model. Scan parameters were taken from the AAPM
Routine Head protocols for a fixed tube current (FTC), helical protocol, and scan lengths were
adapted to the anatomy of each patient model. MC simulations were performed using mesh tallies to
produce voxelized dose distributions for the entire scan volume of each model. Three tally regions
were investigated: (1) a small 0.6 cc volume at the center of the scan volume, (2) 0.8–1.0 cm axial
slab at the center of the scan volume, and (3) the entire scan volume. Mean dose to brain parenchyma
for all three regions was calculated. Mean bone dose and a mass-weighted average dose, consisting
of brain parenchyma and bone, were also calculated for the slab in the central plane and the entire
scan volume. All dose measures were then normalized by CTDIvol for the 16 cm phantom
(CTDIvol,16). Conversion factors were determined by calculating the relationship between normalized
doses and water equivalent diameter (Dw).
Results: CTDIvol,16-normalized mean brain parenchyma dose values within the 0.6 cc volume, 0.8–
1.0 cm central axial slab, and the entire scan volume, when parameterized by Dw, had an exponential
relationship with a coefficient of determination (R2) of 0.86, 0.84, and 0.88, respectively. There was
no statistically significant difference between the conversion factors resulting from these three differ-
ent tally regions. Exponential relationships between CTDIvol,16-normalized mean bone doses had R2
values of 0.83 and 0.87 for the central slab and for the entire scan volume, respectively. CTDIvol,16-
normalized mass-weighted average doses had R2 values of 0.39 and 0.51 for the central slab and for
the entire scan volume, respectively.
1 Med. Phys. 0 (0), xxxx 0094-2405/xxxx/0(0)/1/xx © 2018 American Association of Physicists in Medicine 1
2 Hardy et al.: An SSDE for head CT using Monte Carlo methods 2
Conclusions: Conversion factors that describe the exponential relationship between CTDIvol,16-nor-
malized mean brain dose and a size metric (Dw) for helical head CT examinations have been reported
for two different interpretations of the center of the scan volume. These dose descriptors have been
extended to describe the dose to bone in the center of the scan volume as well as a mass-weighted
average dose to brain and bone. These may be used, when combined with other efforts, to develop an
SSDE dose coefficients for routine, helical head CT examinations. © 2018 American Association of
Physicists in Medicine [https://doi.org/10.1002/mp.13301]
Key words: head CT, Monte Carlo dose simulations, size-specific dose estimate
1. INTRODUCTION doses rather than dose to the center of the scan volume, the
latter being consistent with SSDE as defined in AAPM
A recent study conducted by the University of California Report 204.
Dose Optimization and Standardization Endeavor (UC Dose) Therefore, the purpose of this current study was to estimate
summarizing CT doses across 12 University of California dose to the “center of the scan volume” for helical head CT
medical centers found that head scans comprised 16% of all examinations that can be used to help determine conversion
adult CT examinations.1 The same study also found that the factors for an SSDE of the head. AAPM Report 204 states that
most frequent area imaged in pediatric patients was the head, the size-specific dose estimate (SSDE) gives an estimate of the
accounting for 33% of the total procedures administered.1 dose at “the center of the scanned region (along z) in the
The fact that radiation exposure from head CT examinations patient.”5 However, for helical head CT examinations, the defi-
is a large contributor to the total medical radiation exposures nition of the “center of the scanned region (along z)” is open
underscores the need for accurate patient dose assessments to several interpretations, which are explored in this investiga-
from head CT procedures, particularly for younger patients. tion to determine if there is difference in results based on these
The radiation dose metric commonly reported on most interpretations. As such, this work employed voxelized patient
scanners is the volume computed tomography dose index models along with MC simulation techniques with mesh tallies
(CTDIvol).2,3 This metric, however, is a measure of dose to a of the entire head to produce voxelized dose distributions
reference phantom, not a measure of patient dose.2,3 Turner wherein two different interpretations of “center of the scan vol-
et al. showed that utilizing CTDIvol as normalization metric ume” were investigated: (1) a small central region within the
for Monte Carlo (MC)-simulated organ doses from abdomi- brain parenchyma and (2) a central slab comprising both brain
nal CT scans compensated both for the differences among parenchyma and bone of the cranium. Additionally, the entire
scanner manufacturers and reduced the variation in organ scan volume was also investigated for sake of comparison and
doses across scanners from 31.5% down to 5.2%.4 Subse- completeness. In the case of the central slab, as well for the
quently, AAPM Report 204 developed the size-specific dose entire scan volume, doses both to the brain parenchyma and
estimate (SSDE) quantity to adjust CTDIvol using a set of bone were also estimated separately. To account for the dose
CTDIvol-to-patient-dose conversion coefficients from either deposited in both the brain parenchyma and bone in the head,
the 32 cm CTDI reference phantom (CTDIvol,32) or the a mass-weighted average dose comprising both brain parench-
16 cm phantom (CTDIvol,16) to account for patient size in yma and bone was devised to account for the presence of both
adult and pediatric body CT examinations, respectively.5 brain parenchyma and bone within the slab tally region as well
SSDE “. . .provides an estimate of the dose at the center of as for the entire scan volume. In a manner similar to that used
the scanned region (along z) in the patient” and is defined as in AAPM Report 204, all doses were normalized by CTDIvol,16
the patient dose estimate that takes into account corrections and were parameterized in an exponential fashion with Dw.
based on patient size by AAPM Report 204.5 Although
SSDE has been shown to be a good substitute for organ dose
in the context of abdominal scans,6 the work of AAPM 2. MATERIALS AND METHODS
Report 204 was limited only to body CT examinations.
2.A. Patient models
The work of McMillan et al. in 2014 sought to extend the
approach developed by Turner et al. and used in AAPM Ten voxelized phantom models from the GSF
Report 204 for the body to investigate organs of interest in (Gesellschaft f€ur Strahlen-und Umweltforschung; National
the head, including the brain and the lens of eye, for routine Research Centre for Environment and Health — current
helical and axial acquisitions.7 In that study, strong predictive name: Helmholtz Zentrum M€unchen, German Research Cen-
exponential correlations were observed when MC-simulated ter for Environment Health, Institute of Radiation Protection,
organ doses from detailed voxelized phantoms were normal- Neuherberg, Germany) family9 and the ICRP (International
ized by CTDIvol,16 and were parameterized by water equiva- Commission Radiological Protection, Ottawa, ON, Canada)
lent diameter (Dw) as a metric of patient size,8 yielding voxelized reference male and female10,11 were used in this
coefficients of determination (R2) of 0.93 for whole brain study. The GSF/ICRP phantom models had all radiosensitive
dose for helical scans.7 While predictive correlations were organs identified. The eight GSF voxel-based models were
determined by McMillan et al.5, that work focused on organ created from CT images with up to 131 organs and anatomic
Medical Physics, 0 (0), xxxx

structures segmented. The two ICRP reference male and voxelized models centered within the gantry and with the
female voxelized models were each based on modifications patient table removed. The scan range was defined from the
of two corresponding male and female GSF models of similar top of the C1 lamina through the top of the calvarium.14 The
external dimensions. The GSF/ICRP voxelized models used widest nominal collimation setting of 28.8 mm (measured
in this study had the in-plane resolution subsampled from the beam width of 32.2 mm) on the Siemens scanner was used in
original to decrease computation time.9–11 the simulations because it is the most dose efficient collima-
Additionally, to supplement the pediatric size range pro- tion setting.7 The AAPM’s Routine Head CT protocol recom-
vided by the GSF/ICRP models, five voxelized patient mends either the gantry or head be tilted to reduce the dose to
models were created from anonymized head CT volume data- the lens of the eye;14 however, for the scanner being modeled,
sets of pediatric patients. These datasets were obtained from helical scans are not performed with gantry tilt, so no tilt
clinically indicated scans under IRB approval. Figure 1 con- angle was used in these simulations.
tains an example of an axial slice of a 23-month-old pediatric
head CT scan and the corresponding voxelized representation
2.C. Size metrics
utilized in the MC simulations. All scans were acquired on a
Siemens Sensation 64 multidetector row CT (MDCT) and Dw is a size metric referenced in AAPM Report 220 as the
were performed in the supine position. To create voxelized “x-ray attenuation of a patient in terms of a water cylinder
models of each patient’s anatomy from the image data, voxels having the same x-ray absorption” and was used in this study
within each image series were modeled as either fat, water, as a measure of patient size.8 For the five pediatric patients,
muscle, bone, or air and were subdivided into one of 17 den- Dw was estimated at the center of the scan volume directly
sity levels depending on its CT number.12 The density varia- from the CT numbers (in Hounsfield units, HU) in their
tions for the six different material designations are based on image data. For the GSF/ICRP models, it is not possible to
the CT scanner calibration curve and the linear relationship directly calculate Dw since they are constructed with pixel
between mass density and electron density. This number of data containing tissue identification numbers, not CT num-
bins has been shown to be sufficient for CT dosimetry and bers. The Dw estimates for GSF/ICRP voxelized phantoms
has been validated in previous studies.12,13 Individual organs were instead obtained indirectly from a correlation between
were not segmented for these patient models, but brain par- effective diameter and Dw.7
enchyma tissue was semiautomatically contoured and explic-
itly identified. The MCNPX model characteristics for all
2.D. Monte Carlo simulations
voxelized models used in this study are summarized in
Table I. Detailed descriptions of scan length determination All CT dose simulations for this investigation were con-
and patient size metrics in terms of Dw can be found in Sec- ducted using a modified version of the radiation transport
tions 2.B and 2.C, respectively. software package MCNPX (Monte Carlo N-Particle
eXtended version 2.7.a).13,15,16 Specifically, the source code
was modified to model a MDCT scanner geometry and x-ray
2.B. CT scanner and scanning protocol
source trajectory. All simulations were conducted in photon
The scanning protocol used in this investigation was taken transport mode with a 1 keV low-energy cutoff. This trans-
from the AAPM Adult Routine Head CT protocol for a Sie- port mode does not transport secondary electrons and instead
mens Sensation 64 MDCT.14 Table II contains the CT scan- assumes their energy to be deposited at the interaction site.
ning protocol used in this investigation. All simulations were All MC CT dosimetry for helical head scans were performed
performed as fixed tube current (FTC) helical scans with the using an equivalent source model of the Siemens Sensation
Brain parenchyma
Bone
Cloth
Fat
Muscle
Air
W/L: 400/40
FOV: 180 mm
FIG. 1. (Left) Head CT image of a pediatric patient (Peds5) who underwent a routine CT head examination with window/level settings and reconstruction field of
view (FOV). (Right) Monte Carlo representation of the patient produced using a CT number-based lookup table. The image on the right is color coded for the
material designations for each voxel.

TABLE I. MCNPX model resolution characteristics, scan lengths, and Dw for GSF/ICRP phantom models and five patient models used in this investigation.
In-plane Image Lateral Anterior–posterior Slice thickness Scan

Name Age Gender resolution slices width (mm) width (mm) (mm) length (cm) Dw (cm)
Peds2 a
7 days Male 128 9 128 30 3.5 3.5 4.8 14.3 12.6
Peds1a 7 weeks Male 128 9 128 24 3.5 3.5 4.8 11.6 10.6
Baby 8 weeks Female 67 9 69 142 3.4 1.7 4.0 10.2 11.1
Peds3a 21 months Female 128 9 128 36 3.9 3.9 4.8 16.7 15.6
Peds5a 23 months Male 128 9 128 30 3.5 3.5 4.8 14.8 17.1
Peds4a 2 yr Male 128 9 128 30 3.5 3.5 4.8 14.5 15.7
Child 7 yr Female 64 9 64 144 6.2 6.2 8.0 14.8 17.2
Helga 28 yr Female 64 9 64 114 7.8 7.8 10.0 14.5 18.2
Irene 32 yr Female 66 9 66 348 7.5 3.8 5.0 15.8 17.1
Golem 38 yr Male 64 9 64 220 8.3 8.3 8.0 15.6 18.3
Visible human 38 yr Male 64 9 64 250 8.6 4.3 5.0 15.3 19.6
Regina 38 yr Female 75 9 69 348 7.1 3.6 4.8 17.1 19.9
Donna 40 yr Female 64 9 64 179 7.5 7.5 10.0 16.5 18.7
Rex 43 yr Male 64 9 64 222 8.6 4.3 8.0 16.0 20.2
Frank 48 yr Male 64 9 64 193 5.9 5.9 5.0 21.8 19.2
a
Indicates a voxelized patient model created from image data obtained from clinically indicated scans.
TABLE II. Routine helical head FTC scanning protocol and associated
Normalization factors are necessary to convert dose per
CTDIvol,16 per mAs for the scanner used in this investigation.
simulated source particle (mGy/source particle) to absolute
Parameter Setting dose per tube current time product (mGy/mAs). To achieve
this, all MCNPX tally results were multiplied by a scanner,
kV 120 collimation, and beam energy-specific normalization factor.19
Rotation time (s) 0.5 Each simulation was performed with 108 photons to ensure a
Helical pitch 0.55 statistical uncertainty less than 2% for each individual mesh
Nominal collimation (mm) 28.8 element. As mesh tallies were used to investigate dose distri-
Bowtie filter Standard butions, the computation time was on the order of 10–15 h
Central half value layer 8.9 mm Al per voxelized model, depending on the resolution of the
CTDIvol,16/mAs (mGy/mAs) 0.24 phantom. This study used computational and storage services
associated with the Hoffman2 Shared Cluster provided by
64 MDCT scanner.17 The equivalent source model, as previ- UCLA Institute for Digital Research and Education’s
ously described by Turner et al.17, generates and incorporates Research Technology Group.
scanner-specific x-ray spectra and bowtie filter profiles.
Concerning the voxelized models mentioned in Sec- 2.E. CTDIvol measurements
tion 2.A, incorporation into MCNPX simulations required
that each model be represented as a three-dimensional matrix Since CT head scans performed in this study were all
of organ or nonanatomic material. Integer identification num- simulated scans, estimates of CTDIvol were needed for nor-
bers were allocated for material descriptions based on ele- malization purposes. Conventional CTDI100 exposure mea-
mental compositions of tissue substitutes and their densities surements were taken at the center and peripheral position of
as defined in ICRU Report 44.10,12,18 Three-dimensional dose a CTDIvol,16 phantom with the scan parameters outlined in
distributions of the entire head of each voxelized model were Section 2.B. Exposure measurements in milliroentgen (mR)
produced using the track-averaged rectangular mesh tally were made with a standard 100-mm pencil ionization cham-
configuration (RMESH) in MCNPX. This tally configuration ber (model: 10X6-3CT, Radcal, Monrovia, CA; calibrated by
tracks particles through a mesh grid that is independent of the Radcal) coupled with a calibrated electrometer (MDH 1015,
regular transport problem.15 The mesh tally grid was defined Radcal, Monrovia, CA; calibrated by Radcal) and converted
to match the matrix size and resolution of each individual to units of air kerma (mGy) using the conversion factor
voxelized model to ensure that the dose on a per voxel basis 1 mR = 0.00876 mGy. The 100-mm pencil ionization cham-
was accurately estimated. The average energy deposition ber and electrometer were calibrated for diagnostic energies
within each voxel was tallied in units of MeV/cm3/source par- with an energy dependence of 5% for 3–20 mm AL HVL.
ticle.15 The resulting voxel-wise energy deposition maps were The air kerma was then normalized by the tube current-rota-
then divided by a density map to get units of MeV/g/source tion time product (mAs) used to take the initial measure-
particle. ments. CTDIvol was then calculated from the CTDI100

measurements at the central and peripheral locations and was brain tissue and bone was calculated using Eq. (1),
recorded on an air kerma per tube current-rotation time pro- Dbone M bone þ Dbrain M brain
duct basis (mGy/mAs). Dwt-avg ¼ (1)
M bone þ M brain
where Dbone and Dbrain are the mean dose contributions from
2.F. Dose analyses
bone and brain parenchyma, respectively, and Mbone and
All dose values for each voxel in the patient models were Mbrain represent the mass contributions from bone and brain
obtained using mesh tallies as outlined in Section 2.D. Three parenchyma, respectively. Similarly, the mean of the dose
regions were investigated in this study: (1) a small 0.6 cc vol- voxels of both brain parenchyma and bone within the entire
ume at the center of the scan volume, (2) a 0.8–1.0 cm axial scan volume was calculated as well as a mean mass-weighted
slab at the center of the scan volume, and (3) the entire scan average dose. The standard deviations and coefficients of
volume. A representation of each tally region is shown in variation for brain parenchyma and bone doses within the
Fig. 2. Tally regions #1 and #2 were investigated as separate entire scan volume were also recorded. In this study, mean
interpretations of “center of the volume.” For tally region #1, doses are designated using the nomenclature Dtissue,tally region
a 0.6 cc volume was positioned at the center of the scan vol- where tissue represents the tissue type and tally region repre-
ume and the mean brain parenchyma dose values within all sents one of the three tally regions. The tissue contents and
voxels in this small volume were averaged. Tally #1 was used doses calculated within each tally region are summarized in
to estimate the dose in a volume that is comparable to a Table III.
Farmer chamber located in the center of the head. This con- All dose values were normalized by CTDIvol,16. Consistent
figuration allows for a simplified comparison against empiri- with AAPM Report 204, normalized dose values were
cal measurements wherein a Farmer chamber that is placed in parameterized as a function Dw using the following exponen-
the center hole of physical head CT dose phantoms (e.g., tial relationship:
CIRS head CT phantoms20). The coefficient of variations Dtissue;tally region
(CV) was also recorded. ¼ A eBDw (2)
CTDIvol
For tally region #2, dose values within a slab parallel to an
axial plane at the center of the scan volume were identified. where A and B (units of cm1) are regression constants for a
The thickness of the slab ranged from 0.8 to 1.0 cm, depend- given tissue classification. The coefficient of determination
ing on the slice thickness of the voxelized phantom, as
detailed in Table I in Section 2.A. Dose estimations within
this slab consist of dose to the brain parenchyma and the bone TABLE III. Summary of tally regions, tissue contents within each tally region,
surrounding it. Under this configuration, the mean of the and mean dose estimates measured.
dose voxels to both brain parenchyma and bone within the
Tally region Tissue(s) in tally region Doses calculated
slab was calculated. The standard deviation and coefficient of
variation for both brain parenchyma and bone dose within the #1: 0.6 cc volume Brain parenchyma Dbrain,1
slab were also calculated. Additionally, to consider the pres- #2: central slab Brain parenchyma, bone Dbrain,2, Dbone,2, Dwt-avg,2
ence of both brain parenchyma and bone in tally #2 and tally #3: entire scan volume Brain parenchyma, bone Dbrain,3, Dbone,3, Dwt-avg,3
#3, a mass-weighted average of dose contributions from both
Tally Region #2
Tally Region #3
Tally Region #1
Brain parenchyma Skin Nasal passage

Cortical bone Muscle tissue Air
FIG. 2. MCNPX voxelized representation of ICRP male “Rex” depicting (1) the 0.6 cc volume positioned at the center of scan volume (tally region #1), (2) the
0.8–1.0 cm axial slab positioned at the center of the scan volume (tally region #2), and (3) the entire scan volume (tally region #3) as specified by the AAPM
Routine Head CT14 protocols with corresponding color-coded material designation for each voxel.

scan direction
bottom center top
Peds3
0.50
0.375
Dose (mGy/mAs)
Peds4 0.25
0.125
0.0
Rex
FIG. 3. Axial view of voxelized dose distribution maps for Peds3 (a), Peds4 (b), and Rex (c), respectively, at the top, center, and bottom of the scan volume. The
red arrow at the top of the figure indicates the direction of the scan range.
(R2) was used to assess the ability of the correlations to uniformity of the dose distribution within the brain parench-
explain the proportion of variation explained by Dw. yma. Each of the following sections below describes the
Dose matrix analysis was performed using Matlab scripts doses for each tissue group: brain parenchyma dose, bone
(R2014b, TheMathWorks, Inc., Natick, MA, USA). Brain par- dose, and the mass-weighted average of brain parenchyma
enchyma dose voxels from all three tally regions were com- and bone dose.
pared using the one-way analysis of variance (ANOVA).
ANOVA was also performed to compare conversion factors
3.A.1. Brain parenchyma doses
from AAPM Report 204 for a CTDIvol,16 phantom with nor-
malized brain parenchyma dose voxels from the three tally Dbrain,1, Dbrain,2, and Dbrain,3, for each voxelized model,
regions. Bone doses for tally regions #2 and #3 were compared can be seen in Table IV with values ranging from 0.188 to
using a paired t-test. Additionally, differences between Dbone,2 0.292 mGy/mAs, 0.185 to 0.286 mGy/mAs, and 0.178 to
and Dbone,3 were also tabulated and were defined as follows: 0.284 mGy/mAs, respectively. This table also shows that the
CV was below 2.6%, 6.5%, and 9.4% within tally regions #1,
jDDj
100% (3) #2, and #3, respectively, across all voxelized models and
ðD2 þ D3 Þ=2 below 3.9% across all tally regions within each voxelized
where D2 and D3 are the doses corresponding to tally regions model. ANOVA analysis with multiple comparison showed
#2 and #3, respectively, and ΔD is the difference between D2 that Dbrain,1, Dbrain,2, and Dbrain,3 were not significantly differ-
and D3. Similarly, the mass-weighted average doses for tally ent from each other [F(2, 42) = 0.07, P = 0.93].
regions #2 and #3 were also compared using a paired t-test
and the differences between Dwt-avg,2 and Dwt-avg,3 were cal- 3.A.2. Bone doses
culated using Eq. (3) and were tabulated. ANOVA was also
performed to compare conversion factors from AAPM Report Dbone,2 and Dbone,3, for each voxelized model, ranged from
204 with mass-weighted average doses from tally regions #2 0.664 to 1.040 mGy/mAs and 0.604 to 0.957 mGy/mAs,
and #3. Statistical analyses were performed using GraphPad respectively, as indicated in Table V. The CV for Dbone,2 and
Prism 6.00 for Mac OS X (GraphPad Software, La Jolla, CA, Dbone,3 was less than 27% and 29%, respectively, within the
USA, www.graphpad.com). individual patient models and differences of less than 12%
were observed across all models. The differences between
Dbone,2 and Dbone,3 were found to be statistically significant
3. RESULTS (P < 0.0001, using paired t-test).
3.A. Mesh tally results
3.A.3. Mass-weighted average
Dose distribution maps from the mesh tally simulations of
three different voxelized models are shown in Fig. 3. These Dwt-avg,2 and Dwt-avg,3 ranged from 0.306 to 0.397 mGy/
mesh tally results provide a graphical representation of the mAs and 0.380 to 0.472 mGy/mAs, respectively, across all

TABLE IV. Mean brain doses by tally region type with coefficients of variation within each tally region, and the coefficient of variation (CV) across tally regions
for each patient.
0.6 cc volume (#1) Slab (#2) Entire scan volume (#3)

Across regions
Name Dbrain,1 (mGy/mAs) CV (%) Dbrain,2 (mGy/mAs) CV (%) Dbrain,3 (mGy/mAs) CV (%) CV (%)
Peds2 0.257 2.6 0.254 4.2 0.273 5.0 3.9

Peds1 0.290 1.7 0.286 5.3 0.284 4.8 1.1
Baby 0.292 1.4 0.286 3.4 0.283 5.6 1.6
Peds3 0.230 2.5 0.226 4.0 0.238 7.3 2.6
Peds5 0.200 2.6 0.197 4.8 0.197 5.9 0.9
Peds4 0.217 2.0 0.215 4.8 0.216 6.7 0.5
Child 0.229 1.9 0.227 2.9 0.221 4.9 1.8
Helga 0.204 1.3 0.207 4.8 0.198 7.1 2.3
Irene 0.212 1.2 0.210 5.6 0.204 6.4 2.0
Golem 0.217 0.8 0.211 5.0 0.208 5.9 2.2
Visible human 0.188 1.7 0.187 6.5 0.180 9.4 2.4
Regina 0.216 2.5 0.215 5.3 0.207 7.8 2.3
Donna 0.210 2.8 0.214 4.4 0.203 7.1 2.7
Rex 0.197 0.8 0.195 3.9 0.189 5.9 2.1
Frank 0.190 1.4 0.185 5.0 0.178 9.2 3.3
voxelized models as indicated in Table VI. Differences of Dbone,2 and Dbone,3 were 0.83 and 0.87, respectively. Results
less than 27% were observed between Dwt-avg,2 and Dwt-avg,3 of the regression analysis are tabulated in Table VIII.
for each of the individual patient models. In addition, Dwt-
avg,3 was consistently higher than Dwt-avg,2 across all patient
3.B.3. Normalized mass-weighted average doses
models. These differences were statistically significant
and comparisons to AAPM report 204 values
(P < 0.0001, using paired t-test).
Figure 6 shows normalized Dwt-avg,2 and Dwt-avg,3 parame-
terized as functions of Dw. The R2 values for normalized Dwt-
3.B. Size-speciﬁc, scanner-independent dose avg,2 and Dwt-avg,3 were 0.39 and 0.51, respectively. ANOVA
estimates revealed a statistically significant difference between AAPM
Report 204 conversion factors and normalized Dwt-avg,2 and
3.B.1. Normalized brain parenchyma doses
estimates and comparison with AAPM report 204
values
TABLE V. Mean bone doses by tally region type with coefficients of variation
Figure 4 shows normalized brain parenchyma dose esti- within each tally region and differences between the means of each region.
mates for the three tally regions (Dbrain,1, Dbrain,2, and Dbrain,3)
Slab (#2) Entire scan volume (#3)
parameterized as functions of Dw. For comparison, AAPM
Report 204 conversion coefficients for the 16 cm pediatric Dbone,2 CV Dbone,3 CV Difference
body phantom as a function of Dw (Fig. 6 from that report) are Name (mGy/mAs) (%) (mGy/mAs) (%) (%)
included in the same figure. The R2 values for normalized Peds2 0.929 7.5 0.916 11 1.4
Dbrain,1, Dbrain,2, and Dbrain,3 dose estimations were 0.86, 0.84, Peds1 0.917 27 0.894 29 2.5
and 0.88, respectively. Results from the regression analysis are Baby 1.040 6.1 0.957 14 8.3
summarized in Table VII. ANOVA showed that there was no Peds3 0.839 16 0.768 21 8.8
statistically significant difference between Dbrain,1, Dbrain,2, and Peds5 0.857 14 0.768 19 11
Dbrain,3 when compared to the AAPM Report 204 conversion Peds4 0.759 18 0.731 24 3.8
factors based on CTDIvol,16 [F(3, 56) = 0.70, P = 0.56]. How- Child 0.792 5.0 0.733 15 7.8
ever, it should be noted that estimates using AAPM report 204 Helga 0.734 10 0.651 16 12
conversion factors were consistently higher by 5–10% than Irene 0.730 9.0 0.697 12 4.7
those obtained using the regression equations for Dbrain,1, Golem 0.723 8.1 0.688 15 11
Dbrain,2, and Dbrain,3 from Table VII. Visible human 0.673 13 0.603 18 5.3
Regina 0.730 9.4 0.693 11 5.0
3.B.2. Normalized bone doses estimates Donna 0.750 8.8 0.680 13 9.7
Rex 0.661 8.3 0.636 11 3.9
Figure 5 contains normalized Dbone,2 and Dbone,3 parame- Frank 0.664 10 0.604 17 9.4
terized as functions of Dw. The R2 values for normalized

TABLE VI. The mass-weighted average of brain and bone dose for tally
regions #2 and #3 (Dwt-avg,2 and Dwt-avg,3) and the differences between the
means of each region.
Slab (#2) Entire scan volume (#3) Difference

Name Dwt-avg,2 (mGy/mAs) Dwt-avg,3 (mGy/mAs) (%)
Peds2 0.338 0.412 20

Peds1 0.366 0.436 18
Baby 0.397 0.472 17
Peds3 0.359 0.399 11
Peds5 0.326 0.408 22
Peds4 0.326 0.395 19
Child 0.324 0.397 20
Helga 0.311 0.398 25
Irene 0.328 0.417 24 FIG. 5. Normalized Dbone,2 and Dbone,3 with associated regression fits.
Golem 0.351 0.411 16
Visible human 0.332 0.380 13
Regina 0.306 0.402 27
TABLE VIII. Regression analysis for normalized Dbone,2 and Dbone,3.
Donna 0.350 0.427 20
Rex 0.317 0.379 18 Normalized dose A B (cm1) R2
Frank 0.361 0.401 11
Dbone,2 6.17 0.039 0.83
Dbone,3 6.17 0.043 0.88
dose estimations as indicated in Fig. 6. The fit coefficients

for AAPM Report 204 are the same as those in Table VII.
4. DISCUSSION AND CONCLUSIONS

In this study, MC simulation methods were performed to
obtain estimates of brain parenchyma and bone dose from
patients of different sizes with different tally configurations
that could be used as a basis for determining SSDE conver-
sion coefficients for routine, helical head CT examinations.
Two different tally configurations were considered as possible
candidates for the condition that the measured dose be in the
FIG. 4. Brain parenchyma dose estimations for the three tally regions “center scan volume” as described in AAPM Report 204.5 In
(Dbrain,1, Dbrain,2, and Dbrain,3) normalized by CTDIvol,16 are plotted as a func- addition, a third tally configuration estimated the dose to the
tion of Dw along with the associated regression fits. AAPM Report 204 con- entire scan volume of each patient for comparison. A mass-
version factors based on CTDIvol,16 are also plotted for comparison. weighted average dose quantity was used to consider the
presence of bone in the central slab tally configuration, as
well for the entirety of the scan volume. Lastly, normalized
TABLE VII. Regression analysis results for Dbrain,1, Dbrain,2, and Dbrain,3, brain parenchyma dose estimations under all the three tally
along with AAPM report 204 regression curve coefficients. configurations and normalized mass-weighted average dose
quantity for the both slab and the entire scan volume were
Normalized dose A B (cm1) R2
compared with conversion coefficients from AAPM Report
Dbrain,1 1.80 0.041 0.86 204 based on CTDIvol,16.
Dbrain,2 1.74 0.041 0.84 Normalized Dbrain,1, Dbrain,2, and Dbrain,3 had R2 values of
Dbrain,3 1.93 0.046 0.88 0.86, 0.84, and 0.88, respectively. This indicates that Dw pro-
AAPM report 204 1.87 0.039 – vides good correlative function for the normalized brain par-
enchyma dose using the three tally configurations
investigated in this study. Unlike the study conducted by
Dwt-avg,3 dose estimations [F(2,42) = 168.1, P < 0.0001]. McMillan et al., which only investigated normalized organ
Results from the regression analysis are summarized in doses,7 the current study employed meshed tallies to map
Table IX. It should be noted here that the AAPM Report 204 dose distributions on a per voxel basis. Using this approach,
values are consistently lower than the Dwt-avg,2 and Dwt-avg,3 Dbrain,1, Dbrain,2, and Dbrain,3 were found to be homogeneous

either for a central slab (i.e., tally region #2) or for entirety of
the head (i.e., tally region #3). The motivation for investigat-
ing dose to bone as a function of patient size comes from the
fact that, within the cranium of pediatric patients, there is a
fair amount of red bone marrow (RBM). The amount of
RBM in the head (relative to the entire body) is 12% for chil-
dren 10 yr of age and up to 29% for infants.22,23 The cranium
is composed of the inner and outer layers of cortical bone that
enclose bone spongiosa, wherein RBM, yellow bone marrow
(YBM), and trabecular bone are found.23 RBM is the primary
tissue of interest for the radiogenic risk of leukemia and is
considered highly radiosensitive, as reflected by the high tis-
sue weighting designation in ICRP 103 (wT = 0.12).24 In this
FIG. 6. Normalized mass-weighted average dose to the brain parenchyma study, RBM and YBM were not modeled. The cranial micro-
and cortical bone for tally regions #2 and #3 (Dwt-avg,2 and Dwt-avg,3) along dosimetry necessary to accurately assess RBM dose is
with the associated regression fits. AAPM Report 204 conversion factors beyond the scope of this study, as is assessing the leukemia
based on CTDIvol,16 are also plotted for comparison. risk associated with head CT procedures. RBM and YBM
doses could, in principle, be calculated using fluence-to-dose
response functions or dose enhancement factors (e.g., the
TABLE IX. Regression analysis for normalized Dwt-avg,2 and Dwt-avg,3. Annex on Skeletal Dosimetry in ICRP Publication 11625).
This would, however, require age- and bone-specific dose
Normalized dose A B (cm1) R2 response functions or enhancement factors, which were not
Dwt-avg,2 1.76 0.014 0.39
available for this current study. SSDE was only intended to
Dwt-avg,3 2.08 0.013 0.51
estimate patient dose using metrics of radiation output dis-
played by scanners and was not intended to assess cancer risk
from CT procedures.5 In routine head CT examinations,
with CVs below 10% across all voxelized models and below although the cortical bone would provide some shielding for
4% across all tally regions within each voxelized model. The the spongiosa containing RBM, RBM within the cranium
implications of these results are twofold. The first is that if would still receive some appreciable amount of radiation
“center of the scan volume” is defined as a small, central vol- dose. The potential effects of RBM dose should be taken into
ume (Dbrain,1) or a central slab within the brain parenchyma consideration as a consequence of the scanning techniques
(Dbrain,2), then normalized doses within these regions should used in routine head CT examinations, particularly for pedi-
yield similar results for a head SSDE. The second is that atric patients.24,26
since the dose to the brain is fairly uniform, a dose measure In accordance with the second interpretation of “center of
from either of these two tally regions would be similar to scan volume,” this study also investigated dose to a central
dose to the entire brain (Dbrain,3). slab of the head, which consists of both bone and brain par-
Dbone,2 and Dbone,3 dose estimations had CVs as high as enchyma. A mass-weighted average of the dose contributions
29% and 27%, respectively. Variations in surface dose as from both bone and brain parenchyma was devised to take
high as 30% for helical scans were previously observed by into consideration the presence of both tissue types. The R2
Zhang et al.21 as a consequence of wider beam collimations values for normalized Dwt-avg,2 and Dwt-avg,3 were 0.39 and
and tube start angle. When investigating the surface dose pro- 0.51, respectively. The loss of an exponential relationship
file of a CTDIvol,32 phantom using MC, for example, Zhang effects with respect to the normalized mass-weighted average
et al. reported substantial dose peaks when utilizing a pitch dose as a function of patient size can be explained by consid-
of less than one and when the simulated beam width was ering the relationship between bone mass (and tissue mass)
wider than the nominal beam width.21 A similar effect was fraction of the head as a function of patient size. The mass of
seen when investigating the variability in surface dose in bone increases with age which competes with the decreasing
anthropomorphic phantoms in the abdominal and thoracic exponential of normalized dose vs patient size. Weighting
regions, whereby a pitch of 0.75 was shown to result in a normalized doses of brain parenchyma and bone by their
37% increase in surface dose.21 The results of this study indi- respective masses accounts for the effects of size of the
cate that the dose variations observed within voxels of the patients in effect, making the relationship of normalized
bone could be due to surface dose variations, particularly weighted average dose more linear with respect to patient
given the use of the low pitch and wide beam collimations size. Additionally, the statistically significant difference
recommended in the AAPM’s Adult Routine Head CT observed between normalized Dwt-avg,2 and Dwt-avg,3 and
Protocol.14 AAPM Report 204 can be attributed to the fact that, as men-
The R2 values for normalized Dbone,2 and Dbone,3 were tioned in Section 3.B.3, the conversion coefficients in
0.83 and 0.87, respectively, indicating that Dw is a satisfac- AAPM Report 204 were devised for the abdomen region,
tory size metric for parameterization of normalized bone dose which contains a small amount of bone relative to the

percentage of soft tissue. The values reported in AAPM Additionally, the patient table was not considered in these
Report 204 were consistently lower than normalized Dwt-avg,2 simulation, the omission of which could lead to an overesti-
and Dwt-avg,3 due to taking the dose contributions of bone mation of patient dose.27 However, this overestimation is
into consideration. expected to be under 10% relative to doses considering the
In several places in this study dose estimates were com- inclusion of the table.27
pared with AAPM Report 204 coefficients of SSDE to assess In summary, this study developed conversion coefficients
the generalizability of those coefficients to helical head scans for routine helical head CT procedures using MC methods
when estimating dose to the center of the scan volume. When and voxelized patient models for two interpretations of “cen-
estimating brain dose, Fig. 4 and Table VII showed the ter of the scan volume” that may be used in a manner similar
Dbrain,1, Dbrain,2, and Dbrain,3 estimates and the regression fits to those described in AAPM Report 204.5 In addition, nor-
for each of these estimates as well as the SSDE-based esti- malized dose coefficients were estimated as a function of
mates and coefficients. A one-way ANOVA showed no statisti- patient size for both bone and a mass-weighted average of
cally significant difference between the estimates obtained brain and bone, all in the middle of the scan volume. These
from the regression fits and those obtained using AAPM may contribute to the efforts to report size-specific doses aris-
Report 204 conversion factors; however, the conversion factors ing from CT examinations of the head.
from AAPM Report 204 were consistently higher than those
provided by the regression fits of Dbrain,1, Dbrain,2, and Dbrain,3.
The differences observed between AAPM Report 204 and nor- ACKNOWLEDGMENTS
malized Dbrain,1, Dbrain,2, and Dbrain,3 can be attributed to the
This work was supported in part by a grant from the
fact that the AAPM Report 204 conversion factors were origi-
NIBIB (R01-EB017095) and by a grant from the NIH (T32-
nally devised to estimate dose to the center of the scan volume
EB002101). J.M.B is supported in part by NIH grants (R01-
for the abdomen, which is a homogenous region comprised of
CA181081, R01-EB025829, and R01-CA214515). A.M.H is
soft tissue. The head, in contrast, is comprised of the soft-tis-
supported in part by NIH grant R01-CA18108. C.H.M.
sue brain parenchyma encased in a layer of bone. The presence
receives research support from Siemens Healthcare. M.M.G’s
of the bone provides an inherent source of shielding for the
department has a master research agreement with Siemens
brain parenchyma, which decreases the normalized dose of the
Healthcare.
brain parenchyma relative to the normalized dose to the center
of the scan volume for the abdomen.
When estimating the mass-weighted average dose to brain a)
Author to whom correspondence should be addressed. Electronic mail:
parenchyma and cortical bone, Fig. 6 and Table IX showed [email protected]; Telephone: (310) 481-7558.
the Dwt-avg,2 and Dwt-avg,3 estimates and the regression fits for
each of these estimates, as well as the conversion factors from
AAPM Report 204. This time, the one-way ANOVA analysis REFERENCES
did show a statistically significant difference between the esti- 1. Smith-bindman R, Moghadassi M, Wilson N, et al. Radiation doses in
mates obtained from the regression fits and those obtained consecutive CT examinations from five University of California Medical
Centers. Radiology. 2015;277:134–141.
using AAPM Report 204 conversion factors. In addition, the 2. McCollough CH. CT dose: how to measure, how to reduce. Health
conversion factors from AAPM Report 204 were shown to be Phys. 2008;95:508–517.
consistently lower by 32–50% than those provided by the 3. McNitt-Gray MF. AAPM/RSNA physics tutorial for residents: topics in
regression fits of Dwt-avg,2 and Dwt-avg,3. Thus, the conversion CT radiation dose in CT1. Radiographics. 2002;22:1541–1553.
4. Turner AC, Zhang D, Khatonabadi M, et al. The feasibility of patient
factors found in AAPM Report 204 do not appear to be appli- size-corrected, scanner-independent organ dose estimates for abdominal
cable when the dose to bone is included. CT exams. Med Phys. 2011;38:820–829.
There are some limitations to this study. This study only 5. AAPM Task Group 204. Size-Specific Dose Estimates (SSDE) in Pedi-
investigated dose distributions from voxelized phantom mod- atric and Adult Body CT Examinations. College Park, MD: American
Association of Physicists in Medicine; 2011.
els and voxelized patient data from a single scanner model. In 6. Moore BM, Brady SL, Mirro AE, Kaufman RA. Size-specific dose esti-
order to devise an official SSDE for head CT examinations, mate (SSDE) provides a simple method to calculate organ dose for pedi-
the data presented in this study may be combined with other atric CT examinations. Med Phys. 2014;41:1–10.
7. McMillan K, Bostani M, Cagnon C, Zankl M, Sepahdari AR, McNitt-
physical air kerma measurements of head-sized phantoms
Gray M. Size-specific, scanner-independent organ dose estimates in con-
and MC simulations from different scanner models as was tiguous axial and helical head CT examinations. Med Phys.
done for AAPM Report 204.5 Moreover, the voxelized phan- 2014;41:121909.
tom models used in this investigation were of low spatial res- 8. AAPM Task Group 220. Use of Water Equivalent Diameter for Calcu-
lating Patient Size and Size-Specific Dose Estimates (SSDE) in CT. Col-
olution. Despite the low resolution of the phantoms, dose
lege Park, MD: American Association of Physicists in Medicine; 2014.
distribution within the brain parenchyma was still observed to 9. Petoussi-Henss N, Zankl M, Fill U, Regulla D, Zanki M. The GSF fam-
be fairly uniform. This dose uniformity was observed even ily of voxel phantoms. Phys Med Biol. 2002;47:89–106.
for the voxelized patient models, which have much higher 10. International Commission on Radiological Protection. Adult reference
computational phantoms. ICRP Publ 110 Ann ICRP 39; 2009:(2).
resolution. However, the low resolution could affect the accu- 11. Zankl M, Eckerman KF, Bolch WE. Voxel-based models representing
racy of surface bone dose (and mass-weighted average dose) the male and female ICRP reference adult – the skeleton. Radiat Prot
due to volume averaging used to make coarser voxels. Dosimetry. 2007;127:174–186.

12. Demarco JJ, Solberg TD, Smathers JB. A CT-based Monte Carlo simu- cylindrical and anthropomorphic phantoms. Phys Med Biol.
lation tool for dosimetry planning and analysis. Med Phys. 1998;25:1– 2005;50:3989–4004.
11. 20. Computerized Imaging Reference Systems Inc. Tissue Equivalent CT
13. Bostani M, Mueller JW, McMillan K, et al. Accuracy of Monte Carlo Dose Phantoms [Internet]; 2013 [cited 2018 Sep 20]. Available
simulations compared to in-vivo MDCT dosimetry. Med Phys. from: http://www.cirsinc.com/file/Products/007TE/007TE DS 063017(1).
2015;42:1080. pdf
14. American Association of Physicists in Medicine. Adult Routine Head 21. Zhang D, Savandi AS, Demarco JJ, et al. Variability of surface and cen-
CT Protocols Version 2.0 [Internet]; 2016:1–20. Available from: https:// ter position radiation dose in MDCT: Monte Carlo simulations using
www.aapm.org/pubs/CTProtocols/documents/AdultRoutineHeadCT.pdf CTDI and anthropomorphic phantoms. Med Phys. 2009;36:1025–1038.
15. Pelowitz DB. MCNPX User’s Manual. Version 2.7.0, Los Alamos 22. ICRP 70. Basic anatomical & physiological data for use in radiological
National Laboratory, LA-CP-11-00438. Los Alamos, New Mexico: protection - the skeleton. Ann ICRP. 1995;25:1–80.
LANL; 2011. 23. Cristy M. Active bone marrow distribution as a function of age in
16. AAPM Task Group 195. Monte Carlo Reference Data Sets for Imaging humans. Phys Med Biol. 1981;26:389–400.
Research. College Park, MD: American Association of Physicists in 24. ICRP. The 2007 Recommendations of the International Commission on
Medicine; 2015. Radiological Protection. ICRP Publ 103 Ann ICRP 37; 2007.
17. Turner AC, Zhang D, Kim HJ, et al. A method to generate equivalent 25. Petoussi-Henss N, Bolch WE, Eckerman KF, et al. Conversion coeffi-
energy spectra and filtration models based on measurement for multide- cients for radiological protection quantities for external radiation expo-
tector CT Monte Carlo dosimetry simulations. Med Phys. 2009;36:2154– sures. Ann ICRP [Internet] 2010;40:1–257.
2164. 26. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT
18. The International Commission on Radiation Units and Measurements. scans in childhood and subsequent risk of leukaemia and brain tumours:
Tissue substitutes in radiation dosimetry and measurement. ICRU Report a retrospective cohort study. Lancet. 2012;380:499–505.
No. 44, ICRU, Bethesda, MD; 1989. 27. Nowik P, Bujila R, Kull L, Andersson J, Poludniowski G. The dosimet-
19. DeMarco JJ, Cagnon CH, Cody DD, et al. A Monte Carlo based ric impact of including the patient table in CT dose estimates. Phys Med
method to estimate radiation dose from multidetector CT (MDCT): Biol [Internet]. 2017;62:N538–N547.

MP 13301

Uploaded by

MP 13301

Uploaded by

Estimating a size-speciﬁc dose for helical head CT examinations using Monte

Carlo simulation methods

Medical Physics, 0 (0), xxxx

Medical Physics, 0 (0), xxxx

In-plane Image Lateral Anterior–posterior Slice thickness Scan

Medical Physics, 0 (0), xxxx

Brain parenchyma Skin Nasal passage

Medical Physics, 0 (0), xxxx

bottom center top

Medical Physics, 0 (0), xxxx

0.6 cc volume (#1) Slab (#2) Entire scan volume (#3)

Peds2 0.257 2.6 0.254 4.2 0.273 5.0 3.9

Medical Physics, 0 (0), xxxx

Slab (#2) Entire scan volume (#3) Difference

Peds2 0.338 0.412 20

dose estimations as indicated in Fig. 6. The fit coefficients

4. DISCUSSION AND CONCLUSIONS

Medical Physics, 0 (0), xxxx

Medical Physics, 0 (0), xxxx

Medical Physics, 0 (0), xxxx

Medical Physics, 0 (0), xxxx

You might also like