Insights Imaging (2013) 4:539–553

DOI 10.1007/s13244-013-0269-1

REVIEW

Quality assurance and quality control in mammography:

a review of available guidance worldwide
Cláudia Reis & Ana Pascoal & Taxiarchis Sakellaris &
Manthos Koutalonis

Received: 18 February 2013 / Revised: 20 June 2013 / Accepted: 24 June 2013 / Published online: 4 August 2013
# The Author(s) 2013. This article is published with open access at Springerlink.com

Abstract systems associated with mammographic equipment. All

Objectives Review available guidance for quality assurance guidance reviewed highlighted the importance of dose as-
(QA) in mammography and discuss its contribution to har- sessment and testing the Automatic Exposure Control (AEC)
monise practices worldwide. system. Recommended tests for assessment of IQ showed
Methods Literature search was performed on different variations in the proposed methodologies. Recommended
sources to identify guidance documents for QA in mammog- testing focused on assessment of low-contrast detection,
raphy available worldwide in international bodies, healthcare spatial resolution and noise. QC of image display is recom-
providers, professional/scientific associations. The guidance mended following the American Association of Physicists in
documents identified were reviewed and a selection was com- Medicine guidelines.
pared for type of guidance (clinical/technical), technology and Conclusions The existing QA guidance for mammography
proposed QA methodologies focusing on dose and image is derived from key documents (American College of
quality (IQ) performance assessment. Radiology and European Union guidelines) and proposes
Results Fourteen protocols (targeted at conventional and similar tests despite the variations in detail and methodolo-
digital mammography) were reviewed. All included recom- gies. Studies reported on QA data should provide detail on
mendations for testing acquisition, processing and display experimental technique to allow robust data comparison.
Countries aiming to implement a mammography/QA pro-
gram may select/prioritise the tests depending on available
C. Reis technology and resources.
Department of Science and Technologies of Radiation and Main messages
Biosignals in Health, Lisbon School of Health • An effective QA program should be practical to imple-
Technology-ESTeSL, Lisbon, Portugal
ment in a clinical setting.
C. Reis (*) : T. Sakellaris • QA should address the various stages of the imaging
Faculty of Engineering, Catholic University of Portugal, Lisbon, chain: acquisition, processing and display.
Portugal • AEC system QC testing is simple to implement and
e-mail: [email protected]
provides information on equipment performance.
C. Reis
e-mail: [email protected] Keywords Mammography . Quality control . Quality
T. Sakellaris assurance . Dose . Image quality
e-mail: [email protected]

A. Pascoal
Introduction
KITEC-King’s Imaging Technology Evaluation Centre for NICE,
King’s College Hospital, London, UK
e-mail: [email protected] To ensure the key goals of mammography are achieved,
quality standards should be adopted. Ideally, these should
M. Koutalonis
be wide in scope and address the various aspects with impact
Medical Physics Department, Colchester Hospital University NHS
Foundation Trust, Colchester, UK on the mammography imaging process (e.g. technical, clin-
e-mail: [email protected] ical and training).
540 Insights Imaging (2013) 4:539–553

A systematic approach for assessing critical performance French were considered for comparability issues, as other lan-
indicators can be achieved through the implementation of a guages were not mastered by the team.
quality assurance (QA) program. QA provides a framework The guidance documents identified were reviewed and
for constant improvement through a feedback mechanism. It compared for structure, editorial details, target staff profiles,
allows the identification of deviations from optimum perfor- technologies addressed and type of guidance (technical and
mance of mammographic equipment, suboptimal clinical clinical). Comparative tables are presented summarising the
practice and training needs [1–3]. most relevant findings.
An effective QA program should be practical to imple-
ment in a clinical setting. Adequate test equipment is neces-
sary as well as standard methodology that provides ability to Results
obtain the relevant objective, and subjective metrics of qual-
ity. Also, an effective QA program should be implementable Guidance documents for QA and quality control (QC)
at a low or moderate cost [4]. in mammography
The testing of equipment should address the various crit-
ical stages of the imaging chain (acquisition, processing and Fourteen guidance documents for QA and QC in mammog-
display) and be implemented in a multidisciplinary team raphy published between 1991 and 2011 were identified
approach by trained staff (radiographer, medical physicist, (Table 1). Two are recommended by European bodies
radiologist) [3, 5]. (European Reference Organisation for Quality Assured
In the past 20 years, several guidance documents have Breast Screening and Diagnostic Services [EUREF] and
been developed nationally and internationally to promote European Commission [EC]), three are internationally
quality in mammography. The scope of the guidance docu- proposed by the IAEA and ten have national or regional
ments varies with some focused on technical aspects [4, scope (United States of America [USA], Canada, Australia,
6–10], whereas others include also clinical aspects (e.g. United Kingdom [UK], Ireland, Nordic) by governmental
epidemiology, interventional, pathology, surgery) [7, 11]. bodies, professional and/or scientific organisations.
The developments in digital mammography over the last
10 years have resulted in developments in QA programmes Guidance documents for QA and QC in mammography—
and promoted the recommendation of new tests and proce- scope and professional groups targeted
dures for quality control [12].
This study aimed to identify, analyse and compare select- Four documents address both conventional and digital mam-
ed protocols currently available for QA in mammography, mography. All documents are primarily focused on provid-
and to discuss their contribution to harmonise practices in ing technical guidance. Three documents include both tech-
mammography worldwide. nical and clinical guidance.
This review aims to provide useful guidance to countries Thirteen documents are targeted at medical physicists and
aiming to implement (or further develop) a QA program in nine also include guidance for radiographers and radiolo-
mammography. gists. One protocol is specifically targeted at radiographers
(Table 2).
The EC protocol, Australian and Irish protocols are
Methods broader in scope and include guidance to epidemiologists,
nurses, oncologists and surgeons.
An extensive search was performed to identify guidance
documents and protocols for QA in mammography. Performance testing of mammographic systems and breast
Sources used included scientific databases, organisations dose assessment
of national healthcare systems (hospitals, regulatory bod-
ies, etc.), international agencies (e.g. International Atomic Most documents (exceptions are the European Protocol [EP]
Energy Agency [IAEA], International Commission on and IAEA-D protocols) recommend performance testing of
Radiological Protection [ICRP]), professional colleges the three main stages of the mammography imaging chain
(e.g. American College of Radiology [ACR], Royal (Tables 3, 4, 5, 6 and 7):
College of Radiologists [RCR]) and scientific associations
(e.g. Institution of Physics and Engineering in Medicine 1. Image acquisition (the stage with more intensive
[IPEM], American Association of Physicists in Medicine testing)
[AAPM]). The search returned various documents pub- 2. Image processing (following the manufacturers’
lished in English, French, Portuguese, Spanish, German, Italian, recommendations)
Swedish and Dutch. Only documents published in English or 3. Image display (includes monitor and printer testing)
Table 1 Guidance documents for quality assurance and quality control in mammography

Edition details

Edition Publisher Country Title Short titlea Status Scale Referenceb

Insights Imaging (2013) 4:539–553

2011 IAEA Various Quality assurance programme for digital mammography IAEA-DM In use Worldwide [4]
2009 IAEA Various Quality assurance programme for screen film mammography IAEA-SF In use Worldwide [10]
2007 IAEA Various Dosimetry in diagnostic radiology: an international IAEA-D In use Worldwide [13]
code of practice
2006 European Commission/ Various European guidelines for quality assurance in breast EC In use/update in Europe [14]
EUREF cancer screening and diagnostic 4th edition progress
1996 EP European Various European protocol on dosimetry in mammography EP In use Worldwide [6]
Commission
1991/1994 Swedish Radiation Denmark, Finland, Report on Nordic radiation protection Co-operation– Nordic Supersede by the Nordic [17]
Protection Institute Iceland, Norway, Number 1-mammography European Countries
Sweden Protocol
2009 NHSBSP UK Commissioning and routine testing of full field digital NHSBSP/UK In use National [18]
mammography systems
2009 RANZCR Australia and New Mammography quality assurance program: guidelines for RANZCR In use National [19]
Zealand quality control testing for digital (CR & DR) mammography
2008 The National Cancer Ireland Guidelines for quality assurance in mammography screening Irish Protocol In use National [11]
Screening Service
Board
2008 NQMCBSA Australia Breast screen Australia Quality improvement program Australian Protocol In use National [20]
2005 IPEM UK The commissioning and routine testing of mammographic IPEM/UK In use/update in National [8]
X-ray systems progress
1999 ACR USA Mammography quality control manual for radiologists, ACR In use/update in National [9]
medical physicists and technologists progress
2006 Ministère de la Canada (Quebéc) Manuel de contrôle de la qualité pour la mammographie Canadian In use Regional [21]
Santé-Québec et la biopsie guidée par stéréotaxie. Volume 2–physicien Protocol
biomédical
2001 Ministère de la Canada (Quebéc) Manuel de contrôle de la qualité. Volume 1-technologue en radiologie Canadian In use Regional [22]
Santé-Québec Protocol
a
Short title used as reference in this manuscript
b
Listed on the reference list
541
542 Insights Imaging (2013) 4:539–553

Table 2 Guidance documents for QA and QC in mammography: target staff profiles, technologies and guidance type (clinical and/or technical)

Target profile Technologies Type of guidance

Short title Radiographers/ Medical Other healthcare profiles Screen film Digital Technical Clinical Tolerances or
radiologists physicists (epidemiologists, nurses; recommended
oncologists, surgeons) levels
IAEA-DM Y Y NA NA Y Y N Y
IAEA-SF Y Y NA Y NA Y Y Y
IAEA-D N Y N Y Y Y N Y
EC Y Y Y Y Y Y Y Y
EP NA Y N Y Y Y N Y
Nordic Protocol NA Y NA Y N Y NA Y
NHSBSP/UK NA Y NA NA Y Y N Y
RANZCR NA Y NA NA Y Y N Y
Irish Protocol Y Y Y Y Y Y Y Y
NQMCBSA Y Y Y Y Y Y N Y
ACR Y Y N Y N Y Y Y
IPEM/UK Y Y N Y Small Field Y N Y
Canadian Protocol N Y N Y Y Y N Y
Canadian Protocol N Y N Y Y Y N Y

Y yes, N no, NA not applicable/not available

Testing the image acquisition system The ESAK is required to calculate MGD and can be mea-
sured with a calibrated ionisation chamber (IC), semicon-
X-ray production system All documents recommend testing ductor dosimeter or TLD material (Table 6). If measurements
the generator and X-ray source, the Automatic Exposure include the effect of backscatter (e.g. TLDs), an appropriate
Control (AEC) and the breast compression systems. correction factor should be applied [6]. The recommended
Recommended tests include (1) alignment of X-ray field/light phantoms to perform dosimetry testing vary between the
field/image receptor area, (2) repeatability and accuracy of protocols (Table 6).
tube output exposure, (3) half-value layer (HVL), (4) AEC Also, the various protocols propose different methodolo-
response versus breast thickness and tube voltage compensa- gies to measure the required data for MGD calculation (e.g.
tion and (5) alignment of the compression plate. the ACR, Canadian and UK/IPEM propose measurements to
be performed at 40 mm from the chest wall edge, whereas the
Breast dose Table 6 reviews the guidance for dosimetry EP recommends 60 mm).
testing. All guidance documents provide recommendations Since the conversion factors used to estimate the MGD
for assessment of breast dose and two (i.e. EP; IAEA-D) are from the incident air kerma depend on the X-ray beam
dedicated to this topic and include detailed methodology. quality, it is necessary to keep track of the target/filter (T/F)
The mean glandular dose (MGD) is the recommended combination and tube voltage used in the experimental pro-
parameter for assessing the risk of radiation-induced cancer cedure, as well as the half-value layer (HVL) of the X-ray
in mammography. Proposed methodologies for MGD as- beam.
sessment (reviewed in Table 6) include: The EC protocol proposes conversion factors by Dance
et al. (1990) and Dance et al. (2000), whereas the ACR uses
& Measurements on patients using a thermoluminescent factors by Dance et al. (1990); Wu et al. (1991) and Sobol
dosimeter (TLD) (a minimum of ten patients is et al. (1997) [9], the Canadian protocol uses Stanton et al.
recommended) (1984) and Wu et al. (1991) [13] and the Nordic protocol
& Dose estimation from clinical exposure data (10–60 pa- propose conversion of Rosenstein et al. (1985) [14].
tients recommended)
& Dose estimation using test objects/phantoms (the en- Image receptor The most frequently recommended tests for
trance surface air kerma without backscatter (ESAK) digital mammography include (1) the system’s response
should be measured and multiplied by a conversion function, (2) image noise, (3) missed tissue at chest wall
factor, which compensates for X-ray beam quality, breast edge, (4) signal homogeneity and (5) image artefacts
thickness and composition (percentage glandularity) (Table 3).
Table 3 Recommended tests for QC for image detection and acquisition in mammographic systems (include testing the x-ray generation and image receptor)

International National Regional Frequency

X-ray Test type Target parameter to assess EC (2006) IAEA-SF IAEADM ACR UK/IPEM Ireland Australia RANZCR UK/NHSBSP Canada (Quebec) Total (10)a
generation (2009) (2011) (1999) (2005) (2008) (2008) (2009) (2009) (2001/2006)
X-ray source Focal spot size Y N N Y Y Y N N N Y 5
Source-to-image distance Y N N N Y Y N N Y Y 5
Alignment of X-ray Y Y Y Y Y Y Y N Y Y 9
field/image receptor
Radiation leakage Y Y N N Y Y Y N Y N 7
Insights Imaging (2013) 4:539–553

Tube output Y N Y Y Y Y Y N Y Y 8
Tube voltage and Reproducibility and Y Y Y Y Y Y Y N Y Y 9
beam quality accuracy
Half value layer (HVL) Y Y Y Y Y Y Y N Y Y 9
AEC system Optical density control setting: Y Y N Y Y N Y N N Y 7
performance central value and difference
per step
Back-up timer and security Y N N N Y Y Y N Y N 5
cut-off
Short term reproducibility Y N N Y Y Y Y Y Y R 8
Long term reproducibility Y N Y N N N Y N N R 4
Object thickness and tube Y Y Y Y Y Y Y Y Y Y 10
voltage compensation
Correspondence between Y N N N N N Y Y Y Y 5
AEC sensors
Compression Compression force Y N Y Y Y N Y Y N Y 7
Compression plate alignment Y Y Y Y N Y Y Y N Y 8
Bucky and Anti-scatter grid Grid system factor Y N N N Y Y N N N Y 4
image Grid imaging Y N N N Y Y N N N Y 4
receptor
Screen-film Inter cassette sensitivity and Y Y N Y Y N Y N N Y 6
attenuation variation and
optical density range
Screen-film contact Y Y N Y Y N Y N N Y 6
Image receptor Response function Y N Y N N Y Y N Y R 6
response (digital) Noise Y N Y N R Y Y N Y Y 7
Missed tissue at Y N Y N N N Y N Y Y 5
chest wall
edge (digital
Image receptor Image receptor homogeneity Y N Y N Y Y Y Y Y Y 8
homogeneity and Detector element failure Y N N N N Y Y Y N N 4
stability (digital) (DR systems)
Uncorrected defective detector Y N N N N Y N N N N 2
elements (DR systems)
Inter-plate sensitivity Y N Y N N N Y Y N Y 5
variation (CR)
Other sources of Y N N N N N N N N Y 2
radiation (CR)
Fading of latent Y N N N N N N N Y Y 4
543

image (CR)
544 Insights Imaging (2013) 4:539–553

Frequency
Some protocols propose specific tests for CR systems,
namely (1) inter-plate sensitivity variations, (2) image

10
9
9

9
5
7

7
8

6
artefacts, (3) evaluation of the influence of secondary
sources of radiation and (4) fading of the latent image
signal. Guidance is also included for testing the scanning
Regional

mechanism of the CR plate and the efficiency of the

erasure cycle. Specific tests for SFM are proposed in the
Y
Y
Y
Y

Y
Y

Y
R
older protocols (EC protocol, UK/IPEM, Canada, IAEA-
SF) (Table 3).

Quality of the acquired image Table 7 summarises eight

Y
Y
Y
Y
N
Y

Y
Y

Y
groups of tests for assessment of IQ recommended in the
guidance documents reviewed. The tests address technical
and clinical IQ criteria using test objects and phantoms.
N
N
Y
Y
N
N

Y
Y

Phantoms and test objects The recommended phantoms to

produce the images for low contrast IQ assessment vary
Y
Y
Y
Y
Y
Y

Y
Y

between the protocols. CDMAM is frequently recommended

in Europe (EC PROTOCOL, UK/IPEM, UK/NHSBSP and
Ireland) whereas the ACR phantom is the standard in use in
Y
Y
Y
Y
Y
Y

N
Y

the US and Canada.

IAEA does not recommend a particular phantom but
highlights the importance of using a phantom that contains
Y
Y

Y
Y

Y
N

N
R
R

structures able to mimic those typically found in the breast.

National

For high-contrast IQ assessment the MTF is the key

recommended parameter. The MTF bar pattern method is
Y
Y
Y
Y
N
N

Y
Y

more straightforward to implement than the calculation of

the MTF using the edge phantom.
Y
Y
Y
N
N
Y

N
Y

Image processing Image quality is affected by the process-

ing stage. For SF systems the guidance reviewed recom-
Total refers to the total number of guidance documents that recommend the test
Y
Y
Y
Y
N
N

N
N

mends testing the performance of the chemical processor

International

Y yes/exists, N no/not provided, R referred without detail on the methodology

(e.g. time, temperature, base and fog levels). The EC guide-

lines highlight the importance of testing image processing.
For digital mammography systems, the manufacturer’s guid-
Y
Y
Y
Y
Y
Y

Y
Y

ance should be followed because image-processing algo-

Ghost image/erasure thoroughness
Geometric distortion and artefacts

rithms are manufacturer-specific.

and noise power spectrum
Modulation transfer function
Threshold contrast visibility

Artefacts Artefact analysis is an important test recommend-

ed in all guidelines reviewed. For SFM it focuses on artefacts
Spatial resolution
Image contrast

Exposure time

resulting from the chemical processing or from the degrada-

evaluation
(optional)

tion of the screen-film detector characteristics. In digital

Artefacts

systems, artefact analysis is focused on investigating prob-

lems originating in the image acquisition system and during
plate handling and processing (CR systems). Testing in-
cludes assessment originated by printing devices (e.g. laser
Image quality

printers). A clinical evaluation protocol (type testing) is

Dosimetry
Table 3 (continued)

available in the EUREF website (www.euref.org) and repeat-

ed/rejected analysis is recommended on the IAEA-DM
protocol.
properties
System

Image display QA guidelines for testing image display sys-

tems (Table 5) refer to the AAPM report Task Group 18 [15]
a
Insights Imaging (2013) 4:539–553 545

Australia RANZCR UK/NHSBSP Canada (Quebec) Total (10)a

Frequency
for testing electronic monitors and printers. The testing of
light boxes is included in the QC guidance for SFM systems
[11, 12, 16].

5
6
6
6
4
(2001/2006)

Test frequency and reference (or limiting) values

Regional

All guidance documents provide recommendations on

Y

Y
Y
Y
Y
Y
the frequency of the tests (Table 1). A number of tests
are recommended at acceptance only. Others should be
(2009)

performed periodically (yearly, 6-monthly, monthly,

weekly or daily). Intermediate testing should be performed
N

N
N
N
N
N
when necessary (e.g. following major equipment repair).
(2009)

The guidance documents also provide reference values

and pass/fail criteria. These originate from manufacturer
N

N
N
N
N
N

recommendations, expert knowledge, survey QC data,

baseline values and national policies (e.g. existing dose
(2008)

reference levels). A critical aspect is to ascertain when

the measured (including uncertainties) is substantially
Y

Y
Y
Y
Y
N

lower than the reference/limiting value. As an example,

(2008)
UK/IPEM Ireland

UK/IPEM guidance recommends that measured values

for the relevant performance indicators not exceed one-
N

N
N
N
N
N

third of the range proposed for the limiting or remedial

(2005)

values.
Y

Y
Y
Y
Y
Y
(1999)
National

Discussion
EC (2006) IAEA-SF IAEA-DM ACR

N
Y
Y
Y
N

The study showed that in the last 20 years comprehensive

Table 4 Recommended test for QC: image processing stage in mammographic systems

guidance documents have been developed worldwide to

(2011)

support the implementation of QA in mammography.

Total refers to the total number of guidance documents that recommend the test
N

N
N
N
N
N

Target technology
(2009)

The IAEA-DM protocol (edited 2011) is the most up-to-

Y

Y
Y
Y
Y
Y
International

date guidance and is dedicated to digital mammography.

The UK/IPEM, EC, IAEA-SF and ACR protocols are
well-established documents originally developed for
Y

Y
Y
Y
Y
Y

SFM that have been adopted in many countries world-

Daily performance

wide. The EC guidelines were updated and an addendum

Target parameter

Processing time
and baseline

on digital mammography was included [1, 17]. At the

verification

Light leakage
Sensitometry
Temperature
to assess

date of submission of this paper, an updated version of

Artefacts

the ACR protocol is known to be in progress to include

guidance specifically targeted at digital mammography.
Also, as per information available on the EUREF website, a
Y yes/exists, N no/not provided
performance

revised edition (5th) of the EC Guidelines is in development

processor
Darkroom
Processor
Test type

[18].
Film and

As new techniques in digital mammography are becom-

ing widespread, it is expected that revised versions of the
existing protocols will be produced, including guidance for
processing

testing the capabilities of state of the art technology (e.g.

tomosynthesis, dual-energy contrast-enhanced digital sub-
Film

traction mammography).
a
 546

Table 5 Recommended test for QC: image display stage in mammographic systems

International National Regional Frequency

Test type Target parameter/ EC (2006) IAEA-SF IAEA-DM ACR UK/IPEM Ireland Australia RANZCR UK/NHSBSP Canada (Quebec) Total (10)a
characteristic to (2009) (2011) (1999) (2005) (2008) (2008) (2009) (2009) (2001/2006)
assess
Viewing Viewing box Luminance Y Y Y Y Y N Y Y N Y 8
condition Homogeneity Y Y Y Y Y N Y Y N Y 8
Ambient light – Y Y Y Y Y Y Y N N Y 8
Monitors Ambient light (CRT Y N Y N R Y Y N Y Y 7
displays)
Geometrical distortion Y N Y N R Y Y Y Y Y 8
(CRT displays)
Contrast visibility Y N Y N R Y Y Y Y Y 8
Resolution Y N Y N R Y Y Y Y Y 8
Display artefacts Y N Y N R Y Y N Y Y 7
Luminance range Y N Y N R Y Y Y Y Y 8
Printers Greyscale display Y N Y N R Y Y N Y Y 7
function
Luminance uniformity Y N Y N R Y Y Y Y Y 8
Geometrical distortion Y Y Y N R N Y Y Y Y 8
Contrast visibility Y Y Y N R N Y Y Y Y 8
Resolution Y Y N N R N Y Y Y Y 7
Printer artefacts Y Y Y N R N Y Y Y Y 8
Optical density range Y Y Y N R N Y N Y Y 7
(optional)
Greyscale display Y Y Y N R N Y Y Y Y 8
function
Density uniformity Y Y Y N R N Y N Y Y 7
Other Electrical tests – N N Y Y Y Y N N N R 5
Mechanical tests – N Y Y Y Y Y Y Y N N 7
Repeat image – N Y Y Y Y Y Y Y N Y 8
analysis

Y yes/exists, N no/not provided, R referred without detail on the methodology

a
Total refers to the total number of guidance documents that recommend the test
Insights Imaging (2013) 4:539–553
Table 6 Recommended tests for dosimetry in mammography: technical aspects, dose estimation conversion factors and reference dose per projection

Dosimetry Dose

Protocol ID Technical (with test Clinical (with patient data) Test equipment Quantities and units AGD estimation Conversion factors Reference dose per
objects/phantom) (dosimeters) projection
Nordic Standard breast model NA IC ESAK (mGy) and AGD = ESD × Rosenstein (1985) ≤0.8 mGy without grid
45 mm PMMA AGD or MGD conversion ≤2 with grid (OD = 1)
equivalent to average (mGy) factors
breast (50 % adipose
+ 50 % glandular)
EP Standard breast model 10 patients with a compressed TLD or other ESD (mGy), ESAK AGD = ESAK × gPB Dance (1990) 2.3 mGy for a standard
Insights Imaging (2013) 4:539–553

45±5 mm PMMA breast thickness between 40 dosimeter (mGy) and AGD or phantom
to 60 mm for dose with a dynamic MGD (mGy
measurements on patients range 0.5–100
with TLD mGy
ACR (1) Blocks of PMMA NA IC Entrance Exposure estimated for various Dance (1990); ≤3 mGy (42 mm
(20, 40, 60 and 80 mm) estimated thickness Wu (1991) compressed
(2) Standard breast Model from technical and Sobol breast thickness)
40-mm PMMA factors recorded (1997)
equivalent to 42 mm and tube output
50/50 mixture (mR/mAs) and
MGD (mGy)
IPEM (1) Blocks of PMMA AGD for a series of breast IC and electrometer ESAK(K), AGD D = Kgcs Dance (2000) for two 2 mGy (40 mm
(20-80 mm) examinations on or MGD (mGy) age ranges 40– compressed
(2) Standard breast model each mammography 49 and 50-64 breast thickness)
45-mm PMMA, equivalent system periodically.
breast thickness 55 mm Data collection:
with 30 % glandularity breast thickness;
kVp; mAs. Accuracy:
±2 mm
EC (1) Blocks of PMMA AGD for a series of – ESAK(K); AGD or D = Kgcs Dance (2000) for two 2.5 mGy for 45 mm
(20-80 mm) breast examinations MGD (mGy) age ranges 40–
(2) Standard breast model on each mammography 49 and 50-64
45-mm PMMA, system. Data collection:
equivalent breast breast thickness; kVp;
thickness 53 mm mAs. Accuracy: ±2 mm
PQDCS (1) Standard breast model NA IC and electrometer Entrance Surface AGD = ESD Stanton (1984); ≤3 mGy for breast
40-mm PMMA, equivalent Dose (ESD) × conversion Wu (1991, 1994) thickness of 42 mm
to 42 mm 50/50 mixture (mR); AGD or factors
MGD (mrad/R)
IAEA-D 45-mm thick PMMA A range of 10–50 IC or semiconductor Incident air kerma, DG = CDG50,Ki,P Dance (2000) –
phantom equivalent to patients. Reference Dosimeter or TLD (mGy); Entrance MMA sKi
‘standard’ breast of requires that the surface air kerma (standard breast)
thickness 50 mm and compressed breast is (mGy); AGD or DG = cDG50,Ki.
glandularity 50 % between 40 and 60 mm MGD (mGy) cDg,DG50.s
thick, with a mean value Ki (patient studies)
of 50 ± 5 mm
BC NBSP Blocks of PMMA Based on IPSM89 (2005) IPSM89 (2005) and ESAK(K); AGD D = Kgcs Dance (2000) for two 2.5 mGy for 45 mm
(20–70 mm) and the European the European or MGD (mGy) age ranges 40–49
Protocol in Dosimetry Protocol in and 50–64
in Mammography (1996) Dosimetry in
Mammography
(1996)
547
Table 6 (continued)
548

Dosimetry Dose

NQMCBSA Standard phantom – – – – – ≤2.0 mGy for exposures

42 mm 50 % made using typical
adipose, 50 % clinical settings
glandular breast
(i.e. ACR accreditation
phantom)
NHSBSP (1) Blocks of PMMA 50 patients recommended – ESAK(K); AGD or D = Kgcs Dance (2000) 1 mGy for 20 mm
(20–70 mm) with a compressed breast MGD (mGy) PMMA;
(2) Standard breast thickness of 55±5 mm; 2.5 mGy for 45 mm
model 45-mm PMMA, 10 patients should be PMMA;
equivalent breast included in the dose sample
6.5 mGy for 70 mm
thickness 53 mm PMMA
IAEA-SF Standard breast model NA IC ESAK(K); AGD D = Kgcs Dance (2000) Achievable: 2.0 mGy;
45-mm PMMA, or MGD (mGy) Acceptable: 2.5 mGy
equivalent breast
thickness 53 mm
IAEA-DM Blocks of PMMA NA Calibrated detector Incident air kerma; D = Kgcs Dance (2000) 1 mGy for 20 mm
(20, 45, 70 mm) at appropriated AGD or MGD PMMA; 2.5
mammographic (mGy) mGy for 45 mm
energies PMMA; 6.5
mGy for 70 mm
PMMA

NA not applicable, – not available/not accessible

g Mo/Mo is the conversion factor of incident air KERMA (K) to MGD for Mo/0.030 mm Mo at 28 kVp, c is a factor that corrects for glandularity different from 50 % and s corrects for any anode/filter
material combination, other than the Mo/Mo at 28 kVp only. cDG50,Ki,PMMA is the conversion coefficient to calculate the MGD for a 50 mm standard 50 % glandular breast from the air kerma for a
45 mm PMMA phantom. The coefficient cDGg,DG50 converts MGD for a 50 % glandular breast to the MGD for a breast of glandularity, g, and of the same thickness
Insights Imaging (2013) 4:539–553
Table 7 Overview of recommended tests for image quality assessment in mammography (for further detail on the methodology please refer to the original guidance documents)

Test Guideline Materials Comments and reference values

Positioning ACR NA Subjective evaluation using clinical criteria

EC Ireland,
Australia,
IAEA-SF
Compression ACR NA Subjective evaluation using clinical criteria
EC Compression force device, foam rubber, tape Display force = measured force ± 20 N. Max motorised force = 130–200 N
measure
Insights Imaging (2013) 4:539–553

Ireland Subjective evaluation using clinical criteria, Max motorised force = 200 N
Scales, compressible material Display force = measured force ± 20 N. Max misalignment <5 mm
for symmetric load.
Australia Force measuring device (e.g. analogue bathroom Maximum motorised force between 150–200 N
scales)
IAEA-DM Scales (e.g. analogue bathroom scales), foam, Test motorised and manual compression. Max motorised force = 150 N-200 N.
PMMA slabs Max manual force = 300 N. Display thickness = slab thickness ± 8 mm
Contrast resolution and ACR, Canada, Australia, ACR accreditation phantom. Canada (alternative CANADA: provides reference values for SFM and DR
visualisation of breast RANZCR phantoms are RMI 156 or NA 18–220 or ACR, Australia: RANZCR provides minimum threshold for visible details
lesions (in phantom) CIRS 015)
EC CDMAM ACR, Australia: RANZCR provides minimum threshold for visible details
UK/IPEM, UK/NHSBSP TOR (MAM) or CDMAM IPEM: recommend remedial values for low contrast detail detectability
NHSBSP: recommend acceptable and achievable values
Ireland CDMAM, PMMA blocks, CDCOM software, Recommends achievable and minimum limits for automated analysis,
TORMAM following EC guidance
IAEA-DM Not specified (phantom should mimicking IQ assessed for digital systems should be as good as, or better than, that
breast structures) expected with high quality SF mammography
Spatial resolution UK/IPEM TOR(MAX) Recommends remedial value
EC MTF test tool, software to calculate MTF Recommends considering the acceptance value as reference
Ireland MTF edge phantom, ImageJ software Recommended considering the acceptance value as reference
Canada 2 test patterns; PMMA Minimum threshold for broad
ACR Bar pattern Recommended values for perpendicular and parallel MTF
IAEA-DM MTF test tool-metal foil with straight edges Acceptable values are presented for all available manufactures.
(e.g. copper, stainless steel, brass, etc.)
PMMA to support the MTF test tool and MTF
software
UK/NHSBSP Bar pattern Recommended values presented according manufactures (should be at least
<70 % of Nyquist frequency of the detector)
UK/IPEM TORMAX Remedial and suspension values provided
Noise EC NPS phantom (standard test block), optional software Manufacturer’s specifications
to calculate NPS
Ireland Standard PMMA test block dosimeter Consider acceptance values as reference.
Canada PMMA blocks with various thickness and 0.1 mm Acceptable values are presented for all types of breast tissue.
of aluminium
ACR NA Subjective evaluation using clinical criteria
549
550 Insights Imaging (2013) 4:539–553

Patient and facility name, projection view, side, cassette number, mammography
Professional targets

unit used and the initials of technologists who performed the examination.
The EC and Irish protocols are wider in scope and may be
useful to a broader range of healthcare professionals. Other
Acceptable values are presented for all available manufactures.
protocols focus on dosimetry and IQ assessment and are
targeted at medical physicists, radiographers and breast ra-
diologists. Hendrick et al. [5] showed that the profile of staff
performing QA testing differs between countries. Often,
radiographers are in charge of the most frequent tests (daily,
weekly), whereas medical physicists perform in-depth tech-
Recommended minimum SNR variation

nical performance assessment (e.g. collimation, X-ray tube

Subjective using established criteria

output, and AEC testing). In Japan, radiographers perform

Comments and reference values

all QC testing, whereas in Finland, Iceland and Hungary the

service engineers tend to be in charge of the QC tasks. As
highlighted in the IAEA-DM protocol a critical aspect is that
QC testing is delegated to staff holding appropriate expertise
and training [4].

QA testing of mammographic systems and breast dose

assessment

Image detection and acquisition system

All protocols reviewed recommend testing the X-ray source

(tube voltage and HVL) and the AEC system. AEC testing is
one of the most important procedures due to its direct impact
on IQ and breast dose [19]. It should consider the effects of
PMMA blocks or aluminium plate

variations in object/attenuator thickness and radiation beam

PMMA blocks and aluminium

quality. Hendrick et al. [5] compared QC practices in 22

(or PMMA contrast object)

countries (affiliated with the International Breast Cancer

Screening Network) and concluded that this test that was
performed in all countries.
PMMA blocks

Breast dose The recommended methodologies for breast

Materials

dose estimation vary (Table 6). Measurements using test

NA

objects and breast phantoms are frequently recommended

and more practical to implement than measurements based
on TLD techniques.
Dose assessment with a standard test object/phantom
facilitates the comparison of different mammographic tech-
ACR and IAEA-SF

niques and the investigation of the impact of technical set-

UK/NHSBSP

All Protocols

tings on breast dose [20, 21]. Clinical dose assessment (using

IAEA-DM
Guideline

clinical exposure data) provide valuable information on the

clinical practice and takes into account the influence of
breast thickness and composition on dose [6].
NA not available/not applicable

Variations in dosimetry techniques in mammography may

prevent a robust comparison of breast dose in mammography
Table 7 (continued)

between countries and between radiology departments

[22–24].
Signal-to-noise
ratio (SNR)

Dance et al. [16] also highlighted that national protocols

adopt different phantoms, optical densities, measurement
Labelling
Artefacts

points and conversion factors, which make it difficult to com-

Test

pare the doses estimated with different protocols.

Insights Imaging (2013) 4:539–553 551

Hemdal et al. [23] measured the impact of variations in consideration the technology to be tested (screen-film of
experimental technique (e.g. positioning of the dosimeter, digital). Huda et al. [30] examined the effectiveness of the
compression plate in or out of the beam) on MGD values and ACR phantom to assess image quality in digital mammog-
found noticeable variations. raphy and concluded that it is unsatisfactory due to an
When the European protocol was used, the value of the inappropriate range and sensitivity to characterise simulated
MGD increased by 5±2 % (total variation 0–9 %) at clinical breast lesions.
settings and by 9±3 % (4–17 %) compared with the use of Variations in recommended test objects originate differ-
the Nordic protocol [21]. The same authors also compared ences in reference/tolerance values (Table 7). The number
measurements with different dosimeters (ionisation chambers and type of recommended IQ tests varied (between 1 and 9)
vs solid-state detectors) [23]. They concluded that HVL mea- as well as the recommended methodologies. Examples of
surements can be performed accurately with a sensitive solid- methods found in the guidance for rating IQ include absolute,
state detector and a collimated radiation field, correcting for or relative, scales (e.g. five-step scale, 1 (worst) to 5 (best);
energy dependence. two-step scale with 1 (criterion was fulfilled) and 0 (criterion
This review showed variations in the conversion factors was not fulfilled); four-step scale as designed by PGMI scale
used in the estimation of breast dose (to account for X-ray (perfect, good, moderate and inadequate).
spectrum characteristics and breast composition) amongst The guidance documents reviewed do not include recom-
the guidance documents. mendations on observer training for IQ assessment. This
Zoetelief and Jansen [25] compared protocols for dosim- could be useful to reduce inter-observer variability in the
etry in mammography and concluded that the use of different assessment of IQ.
radiation transport codes and different spectra could cause Also, breast compression force is influenced by breast
differences in the conversion factor g by up to about 7 %. They thickness and composition. However, no recommendations
also showed that inclusion of the compression plate in the are provided to promote the optimisation of compression force
beam results in a 4.5±1.5 % smaller g value for the same according to individual characteristics of the breast (compress-
HVL. Also, when breast thickness increases from 2 cm to ibility, composition and thickness) [31, 32]. Maximum values
8 cm, the g value decreases by a factor of 4. for compression in mammography are recommended [7, 11,
Tsai et al. [15] showed that the MGD calculated using 33, 34].
Dance’s method is 9–21 % higher than that using Wu’s The composition of breast tissue is an important issue
method. Jamal et al. [24] also compared MGD per film because increased breast density is known as a risk factor
considering eight different studies using different protocols for developing breast cancer [35]. Nevertheless, in the
and conversion factors and found MGD values with notice- reviewed QA guidance for IQ assessment breast density was
able variations for a same breast thickness. not used as a standard.
The MGD critically depends on the X-ray spectrum gen- In 2011, an addendum to the EC protocol, containing
erated by the TF combination and tube voltage used. Modern guidance for clinical evaluation of mammographic images,
digital mammography systems offer innovative TF combi- was published promoting harmonisation in image quality
nations (e.g. W/Ag, W/Al) and new conversion factors have analysis. Clinical IQ assessment conducted by experienced
been developed [24, 26, 27]. The protocols reviewed do not radiologists is important because it takes into account the
yet include the most recent published data. effects of image processing which may directly affect the
visibility of relevant features and the subsequent diagnostic
Quality of the acquired image All guidance reviewed rec- outcome [36].
ommends performing low-contrast threshold detection test-
ing, breast lesion visualisation (e.g. simulated in phan- Image display/presentation and processing
toms) and artefact analysis. Compression force, image
noise and spatial resolution testing are also recommend- All protocols including guidance for digital mammography
ed with variations in the proposed methods and test recommend testing monitor displays and printers (Table 2).
materials. No recommendations are provided regarding the format for
The EC protocol recommends assessment of image qual- delivering mammography examinations/images to the pa-
ity of digital mammographic systems using images produced tient and practices vary—some healthcare institutions deliv-
with a specific low-contrast-detail test object (CDMAM) er the examination in hardcopy (paper or film), whereas
[28, 29], which is a costly tool not readily available in all others provide digital images on CD.
imaging departments. The UK/IPEM and ACR protocols Despite the potential critical impact of image processing
recommend alternative test objects to CDMAM, namely in the quality of the final image the testing of image process-
TOR (MAM) and the ACR accreditation phantom, respec- ing tools in still at early states (compared with testing of
tively. The choice of a suitable IQ phantom should take into hardware). Most protocols for testing digital mammography
552 Insights Imaging (2013) 4:539–553

systems recommend testing based on raw image data and do Acknowledgments The authors express their gratitude to Edward
Hendrick, Christopher Lawinski, João Alves and Mário Oliveira for
not include recommendations for testing post-processing
their valuable feedback for improving scientific content and English
algorithms used in clinical images. Establishing testing pro- writing is this manuscript.
tocols for post-processing tools in digital mammography is a
challenging task as processing tends to be manufacturer- Conflict of interest The authors declare no conflicts of interest. No
specific and frequently manufacturers are reluctant to reveal funding was received for this work.
details of the post-processing algorithms incorporated in
their systems. A recent publication [37] addresses briefly Open Access This article is distributed under the terms of the Creative
Commons Attribution License which permits any use, distribution, and
the challenges of testing post-processing in tomosynthesis.
reproduction in any medium, provided the original author(s) and the
Work is in process in collaboration with the manufacturers of source are credited.
digital breast tomosynthesis imaging systems to identify a
method for technical evaluation incorporating the clinically
used tomosynthesis reconstruction technique. Testing post- References
processing tools in DM is a topic that requires further
research. 1. Joy JE, Penhoet EE, Petitti DB (2005) Saving Women’s Lives-
Strategies for Improving Breast Cancer Detection and Disgnosis,
The National Academies
2. Klabunde C, Bouchard F, Taplin S, Scharpantgen A, Ballard-
Conclusion
Barbash R (2001) Quality assurance for screening mammography:
an international comparison. J Epidemiol Community Health
In this study the published guidance for QA in mammogra- 55(3):204–212
phy was reviewed. The recommended performance tests for 3. Li Y, Poulos A, Mclean D, Rickard M (2010) A review of methods
of clinical image quality evaluation in mammography. Eur J Radiol
image acquisition, processing and display systems were
74:122–131
discussed and compared. Noticeable variations exist in the 4. International Atomic Energy Agency (2011) Quality assurance
proposed methods, test objects and phantoms. Also, refer- programme for digital mammography. International Atomic
ence values and acceptability criteria vary between proto- Energy Agency, Vienna
5. Hendrick RE, Klabunde C, Grivegnee A, Pou G, Ballard-Barbash R
cols, which raises the question of whether it would be pos-
(2002) Technical quality control practices in mammography
sible to have a mammography system complying with a test screening programs in 22 countries. Int J Qual Health Care
procedure and acceptability criteria, whereas using another 14(3):219–226
test procedure the system would fail. 6. European Comission (1996) European protocol on Dosimetry in
mammography. European Commission, Luxembourg
Harmonisation and best practices in mammography
7. European Commission (2006) European communities/European
would benefit from more detailed guidance on the experi- reference organisation for quality assured breast screening and diagnos-
mental methods for QC testing and recommendations of tic services European guidelines for quality assurance in breast cancer
more affordable test equipment and materials that could be screening and diagnosis, 4th edn. European Communities, Brussels
8. National Breast Screening Quality Assurance (2005) The
acquired by the majority of X-ray departments.
Commissioning and Routine Testing of Mammographic X-
When a recommended protocol cannot be implemented in Ray Systems, York, Institute of Physics and Engineering in
full, a selection of tests may be adequate. Selection criteria Medicine
should take into consideration resources and expertise avail- 9. American College of Radiology (1999) Mammography Quality
able and the relevance of the tests to local practices. It should Control Manual. American College of Radiology, Reston
10. International Atomic Energy Agency (2009) Quality assurance pro-
be noted to highlight the value of testing the AEC system, gramme for screen film mammography, internatio. International
which is a simple procedure to implement that provides Atomic Energy Agency, Vienna
valuable information on the overall performance of the mam- 11. The National Cancer Screening Service (2008) Guidelines for
mography system. Quality Assurance in Mammography Screening, 3rd edn.
Members of the Quality Assurance Committee/The National
A key factor to promote the success of a QA program Cancer Screening Service Board, Dublin
for mammography is teamwork and the collaboration of 12. Marshall NW, Mackenzie AHI (2011) Quality control measure-
all key staff (e.g. radiographers, radiologists, medical ments for digital x-ray detectors. Phys Med Biol 56:979–999
physicists and healthcare managers). Training and contin- 13. Francine N, Richard T (2006) Manuel de contrôle de la qualité pour
la mammographie et la biopsie guidée par stéréotaxie. Ministère de
uous feedback mechanisms are essential to improve the la Santé et des Services Sociaux, Quebec
testing procedures and strengthen the outcomes of the 14. Nordisk Rapportserie om Stralskyddsfragor (1994) Mammography_
program. Nordic_protocol_Selected_pages.pdf, Stockolm
Also, the use of professional networks and special interest 15. Tsai HY, Chong NS, Ho YJ, Tyan YS (2010) Evaluation of depth
dose and glandular dose for digital mammography. Radiat Meas
groups to exchange experiences with colleagues worldwide 45(3–6):726–728
can be of great value in the initial phases of implementation 16. Dance DR, Young KC, Van Engen RE (2011) Estimation of mean
of a mammography QA program. glandular dose for breast tomosynthesis: factors for use with the
Insights Imaging (2013) 4:539–553 553

UK, European and IAEA breast dosimetry protocols. Phys Med for the estimation of average glandular dose in contact and magni-
Biol 56(2):453–471 fication mammography. Phys Med Biol 51(21):5539–5548
17. International Cancer Screening Network (2011) Other Characteristics 28. Thierens H, Bosmans H, Buls N, De Hauwere A, Bacher K, Jacobs
of Breast Cancer Screening Programs in 27 ICSN Countries, 2007– J, Clerinx P (2009) Typetesting of physical characteristics of digital
2008. National Cancer Institute, Bethesda mammography systems for screening within the Flemish breast
18. European Communities/European Reference Organisation for cancer screening programme. Eur J Radiol 70(3):539–548
Quality Assured Breast Screening and Diagnostic Services, 29. Marshall NW (2006) A comparison between objective and subjec-
European Commission (2006) European guidelines for quality tive image quality measurements for a full field digital mammog-
assurance in breast cancer screening and diagnosis. Http://www. raphy system. Phys Med Biol 51(10):2441–2463
euref.org/european-guidelines 19(4):1–432 30. Huda W, Sajewicz AM, Ogden KM, Scalzetti EM, Dance D (2002)
19. Meeson S, Young KC, Hollaway PB, Wallis MG (2001) Procedure How good is the ACR accreditation phantom for assessing image
for quantitatively assessing automatic exposure control in mammog- quality in digital mammography? Acad Radiol 9:764–772
raphy: a study of the GE Senographe 600 TS. Br J Radiol 31. Poulos A, McLean D (2004) The application of breast compression
74(883):615–620 in mammography: a new perspective. Radiography 10(2):131–137
20. Zoetelief J, Fitzgerald M, Leitz W, Sabel M (1998) Dosimetric methods 32. O’Leary D, Teape A, Hammond J, Rainford L, Grant T (2011)
for and influence of exposure parameters on the establishment of refer- Compression force recommendations in mammography must be
ence doses in mammography. Radiat Prot Dosim 80:175–180 linked to image quality. Proceedings of the European Congress of
21. Hemdal B, Bengtsson G, Leitz W, Andersson I, Mattson S (2000) Radiology 2011, Vienna, pp 1–19
Comparison of the european and nordic protocols on dosimetry in 33. National Quality Management Committee (2008) Australia B
mammography involving a standard phantom. Radiat Prot Dosim BreastScreen Australia Quality National Accreditation Standards-
90:149–154 Quality Improvement Program, Australia, National Quality
22. Hemdal B (2009) Evaluation of absorbed dose and image quality in Management Committee of BreastScreen Australia
mammography. Lund University, Malmö 34. Royal Australian and New Zealand College of Radiologists
23. Hemdal B, Herrnsdorf L, Andersson I, Bengtsson G, Heddson B, Mammography Quality Assurance Program (2009) Guidelines for
Olsson M (2005) Average glandular dose in routine mammography quality control testing fpr digital (CR & DR) mammography. Royal
screening using a sectra MicroDose mammography unit. Radiat Australian and New Zealand College of Radiologists, Sidney
Prot Dosim 114(1–3):436–443 35. Vachon CM, Pankratz VS, Scott CG, Maloney SD, Ghosh K,
24. Jamal N, K-N NG, Mclean D (2003) A study of mean glandular Brandt KR, Milanese T, Carston MJ, Sellers TA (2007)
dose during diagnostic mammography in Malaysia and some of the Longitudinal trends in mammographic percent density and breast
factors affecting it. Br J Radiol 76(904):238–245 cancer risk. Cancer Epidemiol Biomark Prev/Am Assoc Cancer
25. Zoetelief J, Jasen JTM (1995) Calculation of Air kerma to average Res 16(5):921–928
glandular tissue dose conversion factors for mammography. Radiat 36. Bosmans H, van Engen R, Heid P, Lazzari B, Schopphoven S,
Prot Dosim 57(1–4):397–400 Thijssen M, Young K (2011) EUREF type test protocol. European
26. Dance DR, Skinner CL, Young KC, Beckett JR, Kotre CJ (2000) Reference Organisation for Quality Assured Breast Screening and
Additional factors for the estimation of mean glandular breast dose Diagnostic Services, Nijmegen
using the UK mammography dosimetry protocol. Phys Med Biol 37. Engen R van, Bosmans H, Bouwman R, Dance D, Heid P, Lazzari B,
45(11):3225–3240 Marshall N, Schopphoven S, Strudley C et al (2013) Protocol for the
27. Koutalonis M, Delis H, Spyrou G, Costaridou L, Tzanakos G, Quality Control of the Physical and Technical Aspects of Digital Breast
Panayiotakis G (2006) Monte Carlo generated conversion factors Tomosynthesis Systems—Draft Version 0.10. EUREF, Nijmegen