guidelines for quality control testing

for digital (cr DR) mammography

THE ROYAL AUSTRALIAN AND NEW ZEALAND COLLEGE OF RADIOLOGISTS®

The Royal Australian and New Zealand College of Radiologists Eligibility to Attempt the Radiology Part 1 Examination & Successful Completion of the Part 1 Examination

Name of document and version:

Guidelines for Quality Control Testing for Digital (CR DR) Mammography v3

Approved by:
College Council

Date of approval:
3 August 2012

ABN 37 000 029 863

Copyright for this publication rests with The Royal Australian and New Zealand College of Radiologists ®

The Royal Australian and New Zealand College of Radiologists

Level 9, 51 Druitt Street
Sydney NSW 2000, Australia

Email: [email protected]
Website: www.ranzcr.edu.au
Telephone: + 61 2 9268 9777
Facsimile: + 61 2 9268 9799

Disclaimer: The information provided in this document is of a general nature only and is not intended as a substitute
for medical or legal advice. It is designed to support, not replace, the relationship that exists between a patient and
his/her doctor.

2
TABLE OF CONTENTS

1. INTRODUCTION ................................................................................................................................. 5
1.1 College Mission ......................................................................................................................... 6
1.2 Revision Rationale .................................................................................................................... 6
1.3 Purpose and scope ................................................................................................................... 6
1.4 Acknowledgements ................................................................................................................... 6
1.5 Guidelines Development ........................................................................................................... 7
1.6 Feedback .................................................................................................................................. 7
2. Tests to be Performed Before Each Examination .......................................................................... 8
2.1 Infection Control of Breast Imaging Equipment ........................................................................ 9
2.2 Stereotactic Accuracy Confirmation ........................................................................................10
3. Daily Tests........................................................................................................................................11
3.1 Image Plate Erasure – CR Only..............................................................................................12
3.2 Viewing Conditions ..................................................................................................................13
3.3 Monitor and View-Box Inspection ...........................................................................................14
3.4 Printer QC (if applicable) – Wet Laser ....................................................................................15
3.4a Record Sheet – Printer QC (if applicable) – Wet Laser ..........................................................16
3.5 Full Field Artefact Evaluation & System Check (DR Systems only) .......................................17
3.5a Record Sheet – Full Field Artefact Evaluation (DR Systems only) .........................................20
4. Weekly Tests ....................................................................................................................................21
4.1 Detector Calibration – Flat Field Test (DR Systems only) ......................................................22
4.2 Image Quality Evaluation with Phantom .................................................................................26
4.2a Record Sheet – Image Quality Evaluation with Phantom .......................................................28
4.3 Contrast to Noise Ratio (CNR), (DR Systems only) ...............................................................30
4.3a Record Sheet – Contrast to Noise Ratio (CNR) (DR Systems only) ......................................34
4.4 Monitor QC ..............................................................................................................................35
4.4a Record Sheet – Monitor QC ....................................................................................................36
4.5 Monitor and View Box Cleaning ..............................................................................................37
5. Monthly Tests ..................................................................................................................................38
5.1 Mechanical Inspection & Breast Thickness Indicator .............................................................39
5.1a Record Sheet – Mechanical Inspection & Breast Thickness Indicator ...................................40
5.2 Image Plate Cleaning – CR Systems only ..............................................................................41
5.3 Printer QC (if applicable) – Dry Laser .............................................................................42
5.3a Record Sheet – Printer QC (if applicable) – Dry Laser ...........................................................43
5.4 Film Digitiser QC (if applicable) ..............................................................................................44
5.4a Record Sheet – Film Digitiser QC ...........................................................................................45

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 3 of 75
6. Quarterly Tests ................................................................................................................................46
6.1 AEC Calibration, CR ...............................................................................................................47
6.1a Record Sheet - AEC Calibration, CR ................................................................................48
6.2 AEC Calibration, DR ...............................................................................................................49
6.2a Record Sheet - AEC Calibration, DR ......................................................................................51
6.3 Image Receptor Homogeneity, CR .........................................................................................52
6.3a Record Sheet – Image Receptor Homogeneity, CR ...............................................................54
6.4 Image Receptor Homogeneity, DR .........................................................................................55
6.4a Record Sheet – Image Receptor Homogeneity, DR (except GE) ..........................................57
6.4b Record Sheet – Image Receptor Homogeneity, GE DR .........................................................58
6.5 Repeat Analysis ......................................................................................................................59
7. Six-Monthly Tests ............................................................................................................................60
7.1 Cassette Image Plate Condition & Interplate Sensitivity Variation, CR only ..........................61
7.1a Record Sheet – Cassette/Image Plate Condition & Interplate Sensitivity Variation
CR only....................................................................................................................................62
7.2 Compression ...........................................................................................................................63
7.2a Record Sheet – Compression Test .........................................................................................64
8. Testing As Required .......................................................................................................................65
8.1 Maintenance & Fault Logging .................................................................................................66
9. Summary Forms ..............................................................................................................................67
9a Mammography Quality Control Checklist: Monthly or less frequently, CR only .....................68
9b Mammography Quality Control Checklist: Monthly or less frequently, DR only .....................69
9c Mammography Quality Control Checklist: Daily or Weekly, CR only .....................................70
9d Mammography Quality Control Checklist: Daily or Weekly, DR only .....................................71

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 4 of 75
1. INTRODUCTION

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 5 of 75
1.1 College Mission

As a fellowship based organisation, The Royal Australian and New Zealand College of
Radiologists sets, promotes and continuously improves the standards of training and
practice in radiology and radiation oncology, for the betterment of the people of Australia
and New Zealand.

1.2 Revision Rationale

As CR and laser printers are mature technologies, the sections of Version 2.1 relating to
these have only required minor upgrading. The main driver for this revision has been the
need to produce a set of digital mammography (DM) quality control tests that reflect the
technical advances that have been made to DM equipment since 2009. The tests in
Version 3.0 match the sophistication of DM equipment available in 2012.

1.3 Purpose and scope

(a) This document is intended to assist radiologists, radiographers, and mammography

equipment assessors in addressing quality control requirements relating to the
Mammography Quality Assurance Program. This document sets out a standardised
process and instructions to complete quality control processes across a range of
digital mammography machine types.

(b) Please note that this document contains guidelines which are subject to review.
Please refer to individual manufacturers’ guidelines for any additional Quality
Assurance requirements that may be applicable for specific systems.

1.4 Acknowledgements

The Royal Australian and New Zealand College of Radiologists gratefully acknowledges
the extensive work undertaken by Dr John Heggie, the Australasian College of Physical
1 2
Sciences and Engineering in Medicine , and BreastScreen Victoria for providing the
original documents from which the guidelines have been derived, in relation to Quality
Assurance for Digital (CR and DR) Radiography Mammography systems in Australia.
3
The College would also like to acknowledge Ms Jennifer Diffey of NSW BreastScreen for
her contribution of equipment specific ‘recipes’ that are now an integral part of the
document.

The RANZCR has drawn on this work in order to develop this CR/DR Quality Assurance
Guide which is designed to support Diagnostic Mammography facilities and which
underpins the Mammography Quality Assurance Program (MQAP) Digital Mammography
Accreditation.

1
I. D. McLean et al, Interim recommendations for a digital mammography quality assurance
program, Australasian Physical & Engineering Sciences in Medicine, Vol 30, 2, 2007
2
BreastScreen Victoria – enquiries should be directed to [email protected]
3
BreastScreen NSW, Mammography Quality Assurance Program. Guidelines for Routine Quality
Control Testing for Digital Mammography (CR&DR) Version 1 June 2010
Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 6 of 75
 The RANZCR would also like to thank mammography practitioners and equipment vendors
for providing the College with their feedback on the document which has informed this
revised version.

1.5 Guidelines Development

As with any set of Guidelines or Standards in a Quality Program, they are never complete
and are always being improved and upgraded. The development of these Guidelines is
ongoing.

Consultation between the author - Dr John Heggie, the Australasian College of Physical
Sciences and Engineering in Medicine, and feedback received from other stakeholders is
taken into consideration to ensure the Guidelines remain pertinent to the Mammography
Quality Assurance Program and Radiography Mammography in Australia.

1.6 Feedback

Feedback on or suggestions for improvements to this document are welcomed. Please

provide such advice in writing to:

Mail: Director, Standards of Practice and Accreditation

RANZCR
Level 9, 51 Druitt Street
SYDNEY NSW 2000

Email: [email protected]

Fax: 02 9268 9799

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 7 of 75
2. TESTS TO BE PERFORMED BEFORE EACH EXAMINATION

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 8 of 75
2.1 Infection Control of Breast Imaging Equipment

Frequency: Before EACH examination

Before each examination, the breast support or Bucky should be cleaned using alcohol wipes, or
as per manufacturer’s recommendations and/or suitable infection control advice.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 9 of 75
2.2 Stereotactic Accuracy Confirmation

Frequency: Prior to patient use on each day that the biopsy unit is used

Localisation accuracy confirmation should be performed on the stereotactic unit.

4
Calibration in air is recommended as per the ACR manual unless the manufacturer specifies an
alternative technique using a suitable localisation phantom.

Record

 Date test was performed

 Person performing test
 Any action taken/required
Limits

 The required accuracy is that the indicated needle tip coordinates be within ±1mm of the
actual preset needle position in each direction (horizontal, vertical, and depth)

4
ACR, Stereotactic breast biopsy quality control manual”, American College of
Radiology, Reston, VA, 1999
Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 10 of 75
3. DAILY TESTS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 11 of 75
3.1 Image Plate Erasure – CR Only

Frequency: Daily/Weekly Test

CR image plates are sensitive to scattered and naturally occurring radiation sources and if left
unused for long periods of time will store energy absorbed from these sources. It is
recommended that all CR image plates be subjected to erasure procedures on a daily basis as
per manufacturers’ instructions. Some manufacturers refer to this as a “secondary” erasure and
also require a “primary” erasure to be performed on a weekly basis.

DAILY - All manufacturers

WEEKLY - If recommended by manufacturer

Task should be noted as being completed on daily / weekly check list.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 12 of 75
3.2 Viewing Conditions

Frequency: Daily Test

Procedure

 Visually inspect ambient lighting conditions to ensure conformance with acceptable viewing
condition configuration.
• Check that ambient lighting is acceptably low (note that the room should not
be pitch-black)
• Check for reflections on monitors (e.g. room lights, other monitors, viewboxes
and self)
• Check that light from doorways and windows is eliminated
Record

 Date inspection was performed

 Person performing test
 Outcome on daily checklist

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 13 of 75
3.3 Monitor and View-Box Inspection

Frequency: Daily Test

Monitor screens (and view-boxes if applicable) must be free of dust, fingerprints and other marks
that might interfere with image interpretation.

Procedure

 Inspect and, if necessary, clean all monitor screens (and view-boxes if applicable) gently
with a lint-free cloth as per manufacturer’s instructions
Record

 Date inspection was performed

 Person performing test
 Outcome on daily checklist

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 14 of 75
3.4 Printer QC (if applicable) – Wet Laser

Frequency: Daily Test, or prior to printing if used less frequently

Please note: Although many practices will not be printing films for primary reporting, it will
be important to make sure that printed images are of a relatively high standard if being
sent as the medical record.

Test should be performed at installation to establish a baseline, then regularly for QC

purposes.

Procedure

 Print the TG18-QC test pattern (see Appendix 1). This should be installed on the printer. If
not, it is acceptable to send the TG18-QC test pattern stored on the acquisition workstation
to the printer
 Check visibility and distortion of several items used for evaluating the quality of the image
 Check for disturbing artefacts
 Measure mid density (MD), density difference (DD), maximum density (Dmax) and
Base+Fog (B+F)
 Complete daily check list and Record Sheet 2.4a

Record

 Date test was performed

 Person performing test

 Printer identification
 Test results

Limits

 Borders are visible

 Lines are straight

 Squares appear square
 All corner patches are visible
 Squares of different shades from black to white are distinct
 The finest horizontal and vertical line pairs are visible in all four corners and in the centre
 The 5% and 95% gray squares are clearly visible within the 0% and 100% backgrounds
 At least 11 letters (i.e. QUALITY CONT) are visible in the phrase “QUALITY CONTROL” for
each of the dark, mid-grey and light renditions
 No disturbing artefacts are visible
 MD , and DD are within ±0.15 OD of their baseline values
 B+F, and Dmax are within ± 0.03 OD and 0.1 OD of their baseline values. respectively.
Ideally, B+F should be ≤ 0.25 and Dmax should be > 3.4.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 15 of 75
3.4a Record Sheet – Printer QC (if applicable) – Wet Laser

Frequency: Daily Test

Printer manufacturer: Printer Model:

BASELINE DATA Date:

Baseline Lower limit Upper limit Tolerance

B+F ±0.03
Dmax ±0.1
MD ±0.15
DD = D1 – D2 ±0.15

Date
Initials
Checkmark () = Pass/Adequate: = Fail:
General Image quality
No smearing
No artefacts
Ramps continuous
General distortion
Lines straight
Pattern greyed
Boxes square
Luminance
Patches distinct
0-5% visible
95-100% visible
Finest Resolution elements visible in 4 corners and in centre
Horizontal line pairs
Vertical line pairs
Number of letters visible (at least 11 or “QUALITY CONT”)
Dark
Mid-grey
Light
Optical density measurements
B+F
MD
DD = D1 – D2
Dmax
Comments

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 16 of 75
3.5 Full Field Artefact Evaluation & System Check (DR Systems only)

Frequency: Daily Test or before patient use if equipment is used less

frequently.

It is important that the image is free from artefacts caused by the detector as these could be
detrimental to image quality, or mistaken for clinical abnormalities. Examples include faulty pixels
(single, clusters or lines), stitching / registration artefacts, ghosting or areas of unusually low
noise (where the background mottle appears blurred or smoother than other areas of the image).

This test will detect changes in the performance of the entire imaging chain including the X-ray
system and the detector.

Procedure (all DR systems except Sectra)

 Use a rigid compression paddle if one has been provided

 Position 4 cm Perspex on the Bucky. Ideally, the blocks should completely cover the
detector. This prevents possible ghosting artefacts and provides an image for the full-field
artefact evaluation. Note: Siemens provide a Perspex block for this test that conveniently
slides onto collimator.
 Compress to 50N (= 5kg / 5dN)
 Acquire an unprocessed image e.g. QC RAW (see Appendix 2) with the clinically used
mode (e.g. GE: Std Auto, Hologic: Autofilter, Siemens: Opdose).
 Record kV, target / filter, AEC mode and post-exposure mAs
2
 Draw a Region of Interest (ROI) of approximate size 4cm , positioned centrally along the
midline of the image and 6cm back from the chest wall.
 Measure the mean pixel value (MPV) within this ROI
 View the image on the acquisition monitor and using a narrow window and appropriate
level examine image for significant artefacts and look for any variations in noise etc. Zoom
and roam may be used to look for dead detector elements (dels).
 Print the image and examine it if interpretation is performed using hard copy

Procedure (Sectra)

Sectra L30 Software Version 8.3:

 The standard rigid or high edge compression paddle must be used

 Select the Daily quality control menu option in the application (or press the Quality Control
shortcut key on the keypad). Daily quality control examination automatically opens
 Place the daily quality phantom on the patient support. The phantom has a metal guiding
edge. Slide the phantom on the patient support so that the guiding edge of the phantom
reaches the front end of the patient support. The contrast markers should be on the chest
wall side
 Lower the compression paddle to so that it barely touches the phantom
 Perform an exposure with the default settings
 Right-click on the image and select Relabel  Raw
 You will get a message: WARNING: This operation will modify image settings. Do you want
to proceed? Click ‘Yes’

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 17 of 75
  Right-click on the image and select Measurements  CNR measurement
 Left-click on the image at the point where you wish to position the ROI: centrally along the
midline of the image and 6cm back from the chest wall. Make sure that this does not
overlap the contrast discs.
 Left-click somewhere else in the image. The location is unimportant but you must have 2
ROIs before you can view the ROI information.
 The first ROI will be labelled as ‘S’ [S]ignal. This is the one you are interested in. Record
the Mean from the box.
 Using a narrow window and appropriate level examine image for significant artefacts and
look for any variations in noise etc.
 Print the image and examine it if interpretation is performed using hard copy

Sectra D40:
 The standard rigid or high edge compression paddle must be used
 Select the Daily QA Check from the Edit menu option in the application. Daily quality
control examination automatically opens
 Place the daily quality phantom on the patient support centered laterally with the long side
parallel to the chest wall. The contrast markers should be on the chest wall side
 Lower the compression paddle to so that it barely touches the phantom
 Perform an exposure with the default AEC settings
 Using a narrow window and appropriate level examine image for significant artefacts and
look for any variations in noise etc.
 Print the image and examine it if interpretation is performed using hard copy
 Note: The extraction of the MPV from a ROI is problematical with the D40 so it is
suggested that this part of test be waived.
Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, AEC mode, and post exposure mAs
 The MPV
Limits

 mAs and MPV within baseline ± 10% (provided kV and filter are the same as for the
baseline measurement)
 There must be no clinically significant artefacts (e.g.):
 Structures that are more conspicuous than the objects in the phantom used for weekly
testing
 Blotches or regions of altered noise appearance
 Observable grid lines or table top structures
 Bright or dark pixels
 Stitching or registration artefacts


If artefacts are seen it is important to determine whether they are due to the display or the
detector. Take the following steps:

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 18 of 75
 1) To eliminate the possibility of display artefacts, rotate or pan the image. If the artefact
moves with the image it is due to the imaging system; if it stays in the same place, it is due
to the display system (monitor).

2) To eliminate the possibility of artefacts (dust or scratches) on the Perspex block, rotate the
block and repeat the test. If the artefact stays in the same place, it is due to the detector; if
it moves, it may be attributable to dust on the block. Clean the block and repeat the test.

3) If there is an artefact, it may be possible to remove it by carrying out a flatfield calibration.

Please note that Sectra also have a ‘quick calibration’ for this purpose. Do this before
calling an engineer.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 19 of 75
3.5a Record Sheet – Full Field Artefact Evaluation (DR Systems only)

Frequency: Daily Test or before patient use if equipment is used less frequently

X-ray Machine ID
AEC detector position (if applicable)
AEC mode (e.g. Auto STD, “Opdose”)

BASELINE DATA

Date: _____ Baseline Lower limit Upper limit Tolerance

mAs ± 10%

MPV ± 10%

Date kVp Target / mAs MPV Comments Initials

filter

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 20 of 75
4. WEEKLY TESTS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 21 of 75
4.1 Detector Calibration – Flat Field Test (DR Systems only)

Frequency: Weekly Test or as recommended by manufacturer (see below). This test

should also be performed if artefacts are seen on images

GE: Weekly (must be completed before image quality evaluation)

Hologic: At least bi-weekly

Sectra: Weekly

Fuji: As required (a message appears on the AWS when it is due)

Siemens Novation: At least every two weeks

Siemens Inspiration: At least every 3 months

Notes:

1) It is preferable to carry out this test on Monday morning before any other QC
measurements.
2) Please ensure the test block and breast support platform are clean and free from dust and
scratches.
3) Use the light field to ensure that you are using the largest field of view

This test ensures that the detector is properly calibrated, the image is uniform over the entire field
of view, and that a high and consistent level of image quality is maintained.

Procedure

Sectra L30 users: (see Sectra QC manual ‘Full calibration’)

 The standard or high edge compression paddle must be used
 Select the menu option Full calibration in the application (or press the Full calibration
shortcut key on the keypad)
 Instructions appear on the screen describing how to place the Calibration phantom on the
patient support. The calibration phantom is the stepwedge. The phantom has a metal
guiding edge. Slide the phantom on the patient support so that the guiding edge of the
phantom reaches the front end of the patient support
 Follow the displayed instructions and lower the compression paddle to a point just above
the phantom i.e. do NOT compress
 Make an exposure
 Wait for the system to process the data
 Follow the instructions on the screen and repeat scanning five times. The system will
change kV-values automatically for each measurement
 When the calibration is complete the dialog box will close

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 22 of 75
Sectra D40 users: (see Sectra QC manual ‘Complete calibration’)
 Select the menu option Service Tool in the application and select Complete Calibration
from the Calibration dialog box
 Follow instructions on the screen to perform Asic Calibration and Idle auto-refinement.
 Click Wake Up
 Instructions appear on the screen describing how to place the Calibration phantom on the
patient support. The calibration phantom is the stepwedge. The phantom has a metal
guiding edge. Slide the phantom on the patient support so that the guiding edge of the
phantom reaches the front end of the patient support. Click OK in the dialog box
 Follow the displayed instructions and lower the compression paddle to a point just above
the phantom i.e. do NOT compress
 Make an exposure
 Wait for the system to process the data
 Follow the instructions on the screen and repeat scanning as instructed.
 When the calibration is complete a dialog box with the text Calibration Completed appears.
Click OK.

Hologic users: (see Hologic QA manual ‘Detector Flat-Field calibration’)

 Make sure the detector has been switched on for at least 40 minutes
 Select the Detector Flat Field Calibration from the Quality Control (Technologist) list on the
AWS
 Position the 4cm acrylic block on the breast support platform. The compression paddle is
not required
 Follow the directions on the screen. You will be asked to acquire a number of images
 Ensure that each image is artefact-free; a dark band at the nipple edge is normal but there
should be no other signs of collimators

Fuji users: (see Fuji Amulet QA manual ‘Field calibration’)

 Select Calibration menu button
 Select ‘Field Calibration’
 If an exposure is necessary a message requesting X-ray irradiation appears
 A confirmation message will appear when the calibration is complete

GE users: (see GE QA manual ‘Flat-Field Test’)

 Click on the QAP button. Select Flat Field
 Follow the on–screen instructions
 Remove the compression paddle and Bucky
 Place the large acrylic block directly on the image receptor
 Exposure parameters are selected automatically. Make two exposures.

Siemens Novation users: (see Siemens operator manual ‘Calibrating the detector’)
 Slide the 40 mm PMMA block onto the collimator and install the simulator compression
plate
 Select the anode / filter combination that is used most commonly
 Select 28 kVp and 250 mAs
 Ensure automatic data transmission to PACS is disabled if applicable
 Select Patient > Calibration from the main menu of the Exam task card. The Service
Patient is automatically registered and the Calibration dialogue box is displayed

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 23 of 75
  Click on Gain button and confirm old calibration data is to be overwritten with a YES
 In the Images subtask card double click on the upper left button to enable operating
console
 Take an exposure and examine quality of image. If acceptable press Accept button: if not
acceptable press Reject button.
 Repeat until 8 acceptable images have been obtained.
 Click on the relevant button to close patient in the Close patient task card
 Reset automatic data transmission to PACS if applicable and remove simulator
compression plate and PMMA block

Siemens Inspiration users: (see Siemens operator manual ‘Calibrating the detector’)
 Slide the 40 mm PMMA block onto the collimator and remove compression paddle
 Ensure automatic data transmission to PACS is disabled if applicable
 The calibration will be performed for all three possible anode / filter combinations (Mo/Mo,
Mo/Rh and W/Rh) automatically
 Select Patient > Detector Calibration from the main menu of the Examination task card.
The Service Patient is automatically registered and the Calibration subtask card is
displayed
 Select the type of calibration to be performed (Contact or Magnification). The exposure
parameters will be automatically selected for the anode / filter combination.
 Click on Start Calibration and confirm with a YES
 In the Images subtask card double click on the upper left button to enable operating
console
 Take an exposure and examine quality of image. If acceptable press Accept button: if not
acceptable press Reject button
 Repeat until approximately 5 acceptable images have been obtained with each anode /
filter combination
 Confirm with a YES to start calibration calculation
 After successful calibration confirm the message window and calibration will be closed
automatically
 Reset automatic data transmission to PACS if applicable and remove PMMA block.

Record

 Date test was performed

 Person performing test
 This is a pass / fail test. The task should be noted as being completed on daily/weekly
checklist.

Note (GE users only): A number of test results will be displayed. The only measurement required
is the brightness non-uniformity, which is equivalent to the image receptor homogeneity test (see
Section 5.4) and the outcome should be recorded on Form 5.4b.

Limits
Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 24 of 75
  The test must pass

If the test fails, you should check the following and then repeat the test:

1) Confirm that the detector has warmed up. It may need to be powered for an hour; this is
manufacturer-dependent.
2) Are the test block and breast support platform clean and free from scratches, dirt, scrapes
or significant marks etc?
3) Does the test block cover the entire field of view?
4) Have you followed the manufacturer’s instructions exactly e.g. correct exposure factors,
correct number of exposures, compression paddle removed etc.

If the test continues to fail, contact the service engineer. The unit should be taken out of
clinical use immediately.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 25 of 75
4.2 Image Quality Evaluation with Phantom

An image quality test object or phantom contains features which attempt to mimic anatomical and
pathological structures in the breast. The phantom can provide quality control checks on the
performance of the imaging system and also allow comparison between systems. Detailed
information on the use of the phantom should be obtained from its accompanying instruction
manual. This test provides an overall assessment of the image quality produced by the imaging
system.

Frequency: Weekly Test

Note: Use the image processing algorithm recommended by the manufacturer as outlined in
Appendix 2 and observe manufacturer specific instructions noted in procedure below if
applicable.

Procedure

 Use the ACR accreditation phantom (e.g. the RMI 156 or equivalent). You will NOT be
granted accreditation under RANZCR if you use an alternative image quality design
phantom (e.g. CIRS XII). If hardcopy images are used for reporting or if this image is to be
used for a measure of contrast to noise ratio (see Section 3.3), you must also use the
acrylic disc.
 Position the phantom flush with the chest wall edge.
 Compress the phantom to 50 N (5 kg or 5 dN).
 Place the acrylic disc that comes with the phantom on the compression paddle in the usual
location. (Placing it on the paddle avoids damaging the paddle when applying
compression).
 For CR choose a designated “test” cassette and imaging plate that is in routine clinical use
 Use the AEC mode of operation used clinically for a breast of comparable thickness to the
phantom, a consistent AEC detector position where manually selected, and the density
control setting in current clinical use (if applicable)

o Sectra DR users: Make an exposure using ‘Smart AEC’

o Hologic DR users: Select an ‘ACR Phantom’ view from the acquisition
workstation. This means that the system will set factors appropriate for a
phantom of 4.2cm thickness. Then make an exposure using ‘Autofilter’.
o GE DR users: This test should only be carried out after successful completion of
the flat-field test. Use the clinically relevant AEC mode (e.g. Auto STD). Note
that the Phantom IQ test on the AWS should NOT be run as the GE tolerance
levels are not as strict as those specified by RANZCR and should not be applied.
o Siemens DR users: Ensure the study is opened as a ‘QC Processed’ one. Then
make an exposure using the clinically used AEC mode (e.g.’Opdose’). For the
Inspiration make sure ‘Auto-segmentation’ is OFF.
o Fuji DR users: From the EDR Menu choose ‘ACR phantom’. Make an exposure
using the usual clinical AEC mode.

 For CR process plate after a fixed delay (e.g. 30 seconds or 1 minute) to avoid image
fading issues Use the processing algorithm recommended by the manufacturer (see
Appendix 2), for example, Fuji recommend the ‘ACR phantom’ EDR.
Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 26 of 75
  Score the number of fibres, masses and speck groups visible (these should be scored as
for film/screen systems). Use appropriate ambient lighting conditions. It is acceptable to
use magnification and adjust the window width and level in order to fully visualise the speck
groups and fibres.
Note: Ideally, image quality should be scored on the modality used for reading clinical
images i.e. the reporting monitors or the printed copy if hardcopy is used for reporting.
However, this may not always be practicable, especially if images are sent to a separate
site for reading. In this case, it is acceptable to score the phantom on the AWS but it is best
practice if the image is also scored on a reporting monitor at least once a month to check
that PACS causes no image deterioration. Furthermore, there is variation in the resolution
of acquisition monitors supplied by each vendor. If the ACR phantom score (particularly
speck groups) is not acceptable on the acquisition monitor, it should be verified that it is
satisfactory on the reporting monitors.

Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, AEC detector position, density control setting, AEC mode, post exposure
mAs and exposure indicator (CR systems only)
 The monitor or printer used for evaluation
 The number of fibres, masses and speck groups visible
 The optical density in a reproducible part of the phantom image, on and next to the contrast
disk, if reporting is performed from hard copy.

Limits

 Visual scoring of phantom image with consistent viewing conditions (note that zoom and
roam may be needed to fully visualise the number of speck groups) should allow
visualisation of:
o At least 5 fibres
o At least 3.5 speck groups
o At least 4 masses
 The mAs, should be within ±10% of baseline values. Note that this requirement may be
waived for GE DR units as the use of AEC may result in quite variable technique factors
being chosen and, in particular, the mAs not appearing to be well controlled. The important
issue is that the phantom image quality score should remain acceptable.
 The exposure index for CR systems:
o Fuji, Philips and Konica, S# within ±10% of baseline
o Carestream Health (Kodak), EI within ±40 of baseline
o Agfa, SAL within ±5% of baseline, or SAL log within ± 430 of baseline or PVI
log16 within ± 580 of baseline (NIP plates)
 The OD in the centre of printed phantom image must be ≥ 1.6 and ≤ 2.0
 The difference in OD (contrast) between disk and area next to it must be ≥ 0.4.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 27 of 75
4.2a Record Sheet – Image Quality Evaluation with Phantom

Frequency: Weekly Test

X-ray Machine ID
Monitor /printer used for scoring
AEC detector position (if applicable)
Density control position (if applicable)
AEC Mode

BASELINE DATA Date:

Baseline Lower limit Upper limit Tolerance

mAs ±10%
Exposure
±10%*
Indicator (CR)
* Use appropriate limit for the CR manufacturer (see above)

Date

kVp

mAs

Target / filter

Exposure indicator (if applicable)

Fibres

Specks

Masses

OD in hard copy (if applicable)

OD next to disk (OD1)

OD in disk (OD2)

Contrast (OD1 – OD2)

Comments/Actions

Initials

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 28 of 75
Limits

 Visual scoring of phantom image with consistent viewing conditions (note that zoom and
roam may be needed to fully visualise the number of speck groups) should allow
visualisation of:
• At least 5 fibres
• At least 3.5 speck groups
• At least 4 masses
 The mAs, should be within ±10% of baseline value. Note that this requirement may be
waived for GE DR units as the use of AEC may result in quite variable technique factors
being chosen and, in particular, the mAs not appearing to be well controlled. The important
issue is that the phantom image quality score should remain acceptable.
 The exposure index for CR systems:
 Fuji, Philips and Konica, S# within ±10% of baseline
 Carestream Health (Kodak), EI within ±40 of baseline
 Agfa, SAL within ±5% of baseline or SAL log within ± 430 of baseline or
PVI log16 within ± 580 of baseline (NIP plates)
 The OD in the centre of phantom image must be ≥ 1.6 and ≤ 2.0.
 The difference in OD (contrast) between disk and area next to it must be ≥ 0.4.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 29 of 75
4.3 Contrast to Noise Ratio (CNR), (DR Systems only)

When screen-film was used, one of the important parameters for image quality was contrast.
However, digital detectors have a much wider dynamic range and therefore wider exposure
latitude. Combined with image processing and the ability to adjust the contrast and brightness of
the image, this means that the important parameter is not simply contrast but a new parameter
called contrast to noise ratio (CNR). It is thought that the way to optimise a digital mammography
system is to achieve established minimum target CNR values as a function of breast thickness.
Equipment assessors will carry out tests to assess if this is the case. An important routine test is
to ensure that CNR for a single phantom thickness remains approximately constant over time and
ideally this measurement should be undertaken on those systems where it is performed semi-
automatically by the system software.

Frequency: Weekly Test

Procedure

Each DR system has its own specific procedure outlined below.

Figure 1: Selection of ROIs on the Sectra L30 unit using Software version 8.3

 Position the Daily Quality Phantom on the breast support platform, so that the guiding edge
of the phantom reaches the front end of the patient support and the contrast markers are
on the chest wall side

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 30 of 75
  Choose “Daily quality control” from the menu in the application or press the shortcut key on
the keypad
 Lower the compression paddle so that it barely touches the phantom
 Acquire an image
 Right-click on the image and select Relabel  Raw
 You will get a message: WARNING: This operation will modify image settings. Do you want
to proceed? Click ‘Yes’
 Right-click on the image and select Measurements  CNR measurement
 Left-click on the image where you wish to position the ROI for the [S]ignal (this should be in
the top left contrast disc). Note that the default ROI size is the same size as the contrast
disc
 Left-click on the image where you wish to position the ROI for the [B]ackground (this
should be next to the first ROI, in line with the chest wall edge). ROI positions are shown in
Figure 1 above
 Record the displayed CNR value

Sectra D40:
A. Using the AEC to find the relevant exposure factors
 Select the Daily QA Check from the Edit menu option in the application. Daily quality
control examination automatically opens
 Place the daily quality phantom on the patient support centred laterally with the long side
parallel to the chest wall. The contrast markers should be on the chest wall side
 Lower the compression paddle to so that it barely touches the phantom
 Initiate an exposure with the default AEC settings and note the kVp and mAs used in the
exposure as these will be required for subsequent testing.
 Close the examination
B. Acquiring image and measuring CNR
 Start the Service Tool from the File menu in the application.
 Press the Advanced button and select Data Acquisition from the Tools menu
 Select Normal image
 Select the kVp used in the previous image from the list. In the next menu select the scan
time so the displayed mAs is as close as possible to the actual mAs used in the previous
image.
 Press OK and then press Scan
 Lower the compression paddle to so that it barely touches the phantom
 Initiate an exposure with the previously noted kVp and mAs
 Once image appears scroll until contrast objects are visible.
 Click Window WL from Min/Max in Region and select Calculate CNR from drop down list
 Draw a ROI inside left contrast object (see Figure 2, ROI 1A) by clicking and dragging in
the image. Create a second ROI in position shown by ROI 2.
 A number of values are shown near the Save cord/point history button. Note the CNR
value.
 Press Cancel to close the image and then close the service tool.
 Note the method described above differs slightly from the method in the Sectra manuals
but is considered sufficient for a routine test.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 31 of 75
 Figure 2: Selection of ROIs on the Sectra D40 unit

Hologic & Siemens users:

 Use the image of the ACR accreditation phantom with acrylic disc acquired in the Image
Quality – Phantom Test (see Section 3.2)
 Draw two small Regions of Interest (ROIs); ROI1 next to the contrast disc, parallel with the
chest wall and ROI2 in the contrast disc
 Measure the MPV in ROI1 and ROI2 and the noise or standard deviation (SD or σ) in the
background ROI (ROI1)
 Calculate the CNR using the formula:
 CNR = (MPV1 – MPV2) / SD1.

Fuji users:
 Follow the method for Hologic and Siemens users. Alternatively, it may be possible to
obtain a direct measurement of the CNR using the ‘One-Shot’ phantom if this has been
purchased.

GE users: (See GE QA Manual ‘CNR and MTF Measurement’)

 Click on the QAP button on the right column of the Browser window. Select the CNR and
MTF test
 Enter or verify the reference of the IQST device (Serial Number or SN, written on the side
of the device) on the AWS screen, then click Start
 Position the IQST device on top of the Bucky. The compression paddle is not required
 The following parameters are selected automatically: Rh/Rh/30kV/56mAs
 Make one exposure; the results of the tests are displayed
 Record the value for CNR.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 32 of 75
Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, AEC detector position (if applicable), density control setting (if applicable),
AEC mode, and post exposure mAs
 MPV1 and MPV2 in ROI1 and ROI2 respectively and SD2
 GE, Sectra and Fuji One-Shot users: record automatically calculated CNR

Limits

 The CNR should be within ±20% of baseline value

Note: CNR is highly dependent on exposure factors, particularly kV. If the CNR is out of
tolerance, check that it is not due to a change in exposure factors. Repeat the measurement by
setting manual factors which are close to those used in the previous week. Additionally, it should
be noted that the CNR may be dependent on software version. If software is updated then it may
be that a new baseline CNR value needs to be established.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 33 of 75
4.3a Record Sheet – Contrast to Noise Ratio (CNR) (DR Systems only)

Frequency: Weekly Test

X-ray Machine ID
AEC detector position (if applicable)
AEC Mode

BASELINE DATA Date:

Lower Upper
Date Baseline Tolerance
limit limit

CNR ±20%

Date

kVp

Target / filter

mAs

Density control setting

MPV1 (in background)

MPV2 (in disc)

SD1 (noise in background)

CNR = (MPV1 – MPV2) / SD1

Comments/Actions

Initials

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 34 of 75
4.4 Monitor QC

Frequency: Weekly Test

The monitor in the mammography, or biopsy, room, attached to the operating console, is referred
to as the acquisition monitor. The high resolution monitors used for reporting the mammograms
are known as the reporting or diagnostic review monitors. It is necessary to carry out this test on
all acquisition and reporting monitors (left and right).

Procedure

 Display the appropriate version of the TG18-QC test pattern (see Appendix 1)
 If window width (WW) and window level (WL) are displayed, these should be set to 4096
and 2048 respectively. Please note that these will not always be displayed, or be
adjustable. If the pattern has been installed correctly, these will be the default settings
 Ensure viewing conditions are acceptable
 Examine image paying particular attention to the following:
o Visibility and distortion of items used for the evaluation of the quality of the image
o Disturbing artefacts
o Additionally, for reporting monitors, the luminance and contrast of the left and
right monitor should be matched i.e. white, dark and grey squares should appear
to have the same grayscale (please be aware that this test is subjective!)
 Complete weekly check list

Record

 Date test was performed

 Person performing test
 Monitor identification
 Test results

Limits

 Borders are visible

 Lines are straight
 Squares appear square
 No smearing at black-white transitions
 All corner patches are visible
 Squares of different shades from black to white are distinct
 The finest horizontal and vertical line pairs are visible in all four corners and in the centre
 The 5% and 95% pixel value squares are clearly visible within the 0% and 100%
backgrounds
 The pattern is centred in the active area
 No disturbing artefacts are visible
 At least 11 letters (i.e. QUALITY CONT) are visible in the phrase “QUALITY CONTROL” for
the dark, mid-grey and light renditions.
 Additionally for reporting monitors: the monitors are matched subjectively in terms of
luminance and contrast.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 35 of 75
4.4a Record Sheet – Monitor QC

Frequency: Weekly Test

Date Initials:
Monitor Acquisition Radiologist 1 Radiologist 2 Radiologist 3
Left Right Left Right Left Right
Window level
Window width
Checkmark () = Pass/Adequate:  = Fail:
General image quality
No smearing
No artefacts
Ramps continuous
Geometric distortion
Lines straight
Pattern greyed
Boxes square
Luminance
Patches distinct
0-5% visible
95-100% visible
Finest Resolution elements visible in 4 corners and in centre
Horizontal line pairs
Vertical line pairs
Number of letters visible (at least 11 or “QUALITY CONT”)
Dark
Mid-grey
Light
Monitor matching (Subjectively by eye)
Dark areas on two
NA
monitors matched
White areas on two
NA
monitors matched
Contrast on two
NA
monitors matched
Comments

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 36 of 75
4.5 Monitor and View Box Cleaning

Frequency: Weekly Test

Monitor screens (and view-boxes if applicable) must be free of dust, fingerprints and other marks
that might interfere with image interpretation.

Procedure

 Clean all monitor screens (and view-boxes if applicable) gently with a lint-free cloth as per
manufacturer’s instructions

Record
(on daily/weekly checklist)

 Date test was performed

 Person performing task

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 37 of 75
5. MONTHLY TESTS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 38 of 75
5.1 Mechanical Inspection & Breast Thickness Indicator

Frequency: Monthly

As in screen/film mammography the facility staff must perform an overall mechanical inspection of
the digital mammography system and associated components. The inspection should be carried
out monthly to ensure that there are no hazardous, inoperative, out of alignment or improperly
operating items on the system.

Procedure

Visual inspection of the system to ensure safe and optimum operation. The checklist includes
confirmation that the breast thickness indicator is accurate. Ideally, this test should be performed
using a rigid compression paddle as tilting paddles are more likely to give inaccurate results.
Some manufacturers provide tilting paddles for clinical use and rigid paddles for QA tests.

Perform the test as follows:

 Place 4cm Perspex on the breast support platform

 Lower paddle to contact Perspex lightly (up to 50N = 5kg = 5dN compression)
 Record the digital readout of breast thickness

 Sectra DR users: because the Bucky and compression paddle are curved, you will get an
inaccurate result for rigid Perspex. Two alternatives are proposed:
o Measure the air gap between the Bucky and the Perspex with a ruler and
subtract this from the indicated readout, or
o Compress a non-rigid material (e.g. foam, pillow, soft ball) and measure the
compressed breast thickness using a ruler. Either method is acceptable but
please ensure consistency.

Record
(on monthly checklist)

 Date inspection performed

 Inspection results
 Person performing test

Limits

 Mechanical and safety functions operating correctly

5
 Breast thickness indicator accurate to ± 5mm

5
This is particularly important because many digital mammography units depend on this
measurement to determine the technique factors.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 39 of 75
5.1a Record Sheet – Mechanical Inspection & Breast Thickness Indicator

Frequency: Monthly Test

X-ray Equipment ID:

Date Initials
Pass = 
Item Description
Fail = 

Measure the temperature in the mammography acquisition room and

1
confirm it is in acceptable range for equipment.

Visually inspect the unit for loose parts, cracks in the compression
2
paddles, compressor, Bucky cleanliness and overall integrity.

Check that all hoses and cables are free from breaks, crimps, or knots.
3
Hoses and cables should not be under other heavy equipment.

4 Verify that the angulation indicator is working correctly

5 Verify that the interlocks are working correctly

6 Ensure that the gantry moves smoothly

7 Ensure that panel switches, indicator lights and meters are functioning

8 Ensure that the field light is functioning

9 Ensure that the current technique chart is posted

On the interpretation workstation, display a recent clinical image, and

10 verify that the time and date as well as the facility identification are
correct in the image annotation
1
Check that the breast thickness indicator is accurate to ± 5 mm .
11
Record result for 40 mm: _______________

12 Confirm face guard is present or not damaged.

Confirm automatic compression release and manual compression

13
release in the advent of power failure both work

14 Confirm the operator shield integrity

Ensure that the cleaning solution for the breast support plate and
15
compressor is available.

Verify any other functions that are specified for monthly monitoring by
16
the equipment manufacturer
1
This is particularly important since many digital mammography units may depend on this
measurement to determine the technique factors.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 40 of 75
5.2 Image Plate Cleaning – CR Systems only

Frequency: Monthly Test or as recommended by manufacturer

Images should be inspected on a daily basis to ensure that they are free from excessive artefacts
attributable to dust on the imaging plates or in the readout system, or defects on the imaging
plates. If artefacts are detected at a level that could interfere with the diagnostic quality of the
images, then inspection of the plates for dust or defects should be undertaken. If the plates are
dusty, then they should be cleaned according to the manufacturer’s protocol. Plates containing
significant defects should be replaced.

If it is found that plates must be cleaned frequently (certainly more than weekly) this may indicate
problems with dust or cleanliness of the imaging environment. If this is the case, attention should
be given to ventilation (possible need for improved air filtration in the room and/or the CR plate
reader) and room cleaning protocols.

Note:

Excessive frequency of plate cleaning may lead to premature wear of the plates.

Procedure

Clean plates according to manufacturer’s cleaning protocol.

Record

(on monthly checklist)

 Date inspection and cleaning performed

 Person performing test

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 41 of 75
5.3 Printer QC (if applicable) – Dry Laser

Frequency: Monthly Test

Please note: Although many practices will not be printing films for primary reporting, it will
be important to make sure that printed images are of a relatively high standard if being sent
as the medical record.

Test should be performed at installation to establish a baseline, then regularly for QC

purposes.

Procedure
 Print the TG18-QC test pattern (see Appendix 1). This should be installed on the printer. If
not, it is acceptable to send the TG18-QC test pattern stored on the acquisition workstation
to the printer
 Check visibility and distortion of several items used for evaluating the quality of the image
 Check for disturbing artefacts
 Measure mid density (MD), density difference (DD), maximum density (Dmax) and
Base+Fog (B+F
 Complete monthly check list

Record

 Date test was performed

 Person performing test
 Printer identification
 Test results

Limits

 Borders are visible

 Lines are straight
 Squares appear square
 All corner patches are visible
 Squares of different shades from black to white are distinct
 The finest horizontal and vertical line pairs are visible in all four corners and in the centre
 The 5% and 95% gray squares are clearly visible within the 0% and 100% backgrounds
 At least 11 letters (i.e. QUALITY CONT) are visible in the phrase “QUALITY CONTROL” for
each of the dark, mid-grey and light renditions
 No disturbing artefacts are visible
 MD , and DD are within ±0.15 OD of their baseline values
 B+F, and Dmax are within ± 0.03 OD and 0.1 OD of their baseline values, respectively.
Ideally, B+F should be ≤ 0.25 and Dmax should be > 3.4.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 42 of 75
5.3a Record Sheet – Printer QC (if applicable) – Dry Laser

Frequency: Monthly Test

Printer manufacturer: Printer Model:

BASELINE DATA Date:

Baseline Lower limit Upper limit Tolerance
B+F ±0.03
Dmax ±0.1
MD ±0.15
DD = D1 – D2 ±0.15

Date
Initials
Checkmark () = Pass/Adequate: = Fail:
General Image quality
No smearing
No artefacts
Ramps continuous
General distortion
Lines straight
Pattern greyed
Boxes square
Luminance
Patches distinct
0-5% visible
95-100% visible
Finest Resolution elements visible in 4 corners and in centre
Horizontal line pairs
Vertical line pairs
Number of letters visible (at least 11 or “QUALITY CONT”)
Dark
Mid-grey
Light
Optical density measurements
B+F
MD
DD = D1 – D2
Dmax
Comments

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 43 of 75
5.4 Film Digitiser QC (if applicable)

Frequency: Monthly Test

Please note: This test is performed to ensure that the film digitiser is producing
consistently accurate digital reproductions of a hard copy image.

Test should be performed at installation to establish a baseline, then regularly for QC

purposes.

Procedure

 Produce an analogue film with a step wedge pattern. This may be generated automatically
by a laser printer, for example.
 Measure the Optical Densities (OD) for each step
 Digitise the image
 Measure the mean pixel value (MPV) at a convenient place in each step
 Plot the MPV versus OD using an EXCEL spreadsheet. This should be linear if the correct
look up table (LUT) is used.
 Note: A Look-up-table (LUT) is part of the soft ware in the digitiser or workstation it is
attached to.
2
 Obtain the value of the correlation coefficient (R ) using EXCEL for this plot.
 Complete monthly checklist.

Record

 Date test was performed

 Person performing test
2
 The R value for the plot of MPV versus OD.

Limits
2
 The plot of MPV versus OD should be linear over full range of OD from B+F to Dmax with R
>0.95.
 Note: If the plot of MPV versus OD is not linear, it could be due to the wrong LUT being
selected by the equipment. This should be rectified by service personnel.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 44 of 75
5.4a Record Sheet – Film Digitiser QC

Frequency: Monthly Test

Digitiser ID:

2
Date R Comment Initials

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 45 of 75
6. QUARTERLY TESTS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 46 of 75
6.1 AEC Calibration, CR

Frequency: Quarterly Test

It is important to ensure that the AEC compensates correctly for changes in breast thickness and
exposure factors (kV, target / filter etc.) and also that the operation of the AEC remains stable
over time. The average dose to the detector should remain approximately constant; this is
assessed using the Exposure Indicator.

Procedure

 Use Perspex (PMMA) blocks of thickness 2 cm, 4 cm and 6 cm.

 Position the blocks in a consistent manner (e.g. flush with the chest wall edge of the
detector)
 Choose a designated 18 cm x 24 cm “test” cassette and imaging plate that is in routine
clinical use for all exposures
 Expose each thickness of the blocks using factors that would be used in a clinical setting.
(i.e. kVp, target/filter, density control setting and AEC mode used for corresponding
thickness of breast)
 Process plate after a fixed delay (e.g. 30 seconds or 1 minute) to avoid image fading
issues. See Appendix 2 for the relevant reader mode to use.
 Record the kVp, target/filter, post exposure mAs and exposure indicator
 Repeat for 24 x 30 cm “test cassette”

Record
 Date test was performed
 Person performing test
 X-ray system identification
 kVp, target/filter, AEC detector position, density control setting, exposure indicator and post
exposure mAs

Limits
 For CR units, the basic requirement is that the average dose to the plate for each of the
three thicknesses of Perspex be within ±10% of the baseline value for that thickness.
Because the relationship between Exposure Indicator and dose to the detector is not linear
for CR systems we need to apply the following:
 Fuji, Philips and Konica S# for each of the 2, 4, and 6 cm blocks should
be within ±10% of the respective baseline value
 Carestream Health (Kodak), the EI value for each of the 2, 4, and 6 cm
blocks should be within ±40 of the respective baseline value
 Agfa, the SAL value for each of the 2, 4, and 6 cm blocks should be
within ±5% of the respective baseline value: or SAL log within ± 430 of
respective baseline; or PVI log16 within ± 580 of respective baseline
(NIP plates)

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 47 of 75
6.1a Record Sheet - AEC Calibration, CR

Frequency: Quarterly Test

X-ray machine ID: CR Plate size:

AEC position: AEC Mode:

BASELINE DATA - Specify Exposure Indicator: e.g. S#, SAL : ___________

Date: ______ Baseline Lower limit Upper limit Tolerance*
EI for 2 cm ±10%*
EI for 4 cm ±10%*
EI for 6 cm ±10%*
* Use appropriate tolerance as listed in limits below.

Date: ______________ Initials: ________________________

Phantom kVp Target & mAs AEC Exposure EI
thickness filter density indicator acceptable
setting (EI)

2 cm
4 cm
6 cm

Date: ______________ Initials: ________________________

Phantom kVp Target & mAs AEC Exposure EI
thickness filter density indicator acceptable
setting (EI)

2 cm
4 cm
6 cm

Limits

 Fuji, Philips and Konica S# for each of the 2, 4, and 6 cm blocks should be within ±10% of
the respective baseline value
 Carestream Health (Kodak), the EI value for each of the 2, 4, and 6 cm blocks should be
within ±40 of the respective baseline value
 Agfa, the SAL value for each of the 2, 4, and 6 cm blocks should be within ±5% of the
respective baseline value; or SAL log within ± 430 of respective baseline; or PVI log16
within ± 580 of respective baseline (NIP plates) .

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 48 of 75
6.2 AEC Calibration, DR

Frequency: Quarterly Test

It is important to ensure that the AEC compensates correctly for changes in breast thickness and
exposure factors (kV, target / filter etc.) and also that the operation of the AEC remains stable
over time. The average dose to the detector should remain approximately constant; this is
assessed using the Mean Pixel Value in a standard ROI since the MPV is proportional to detector
dose for integrated digital systems.

The procedure calls for the use of the unprocessed image e.g. QC RAW (see Appendix 2).

Procedure

See special notes below for GE and Sectra users.

 Use a rigid compression paddle if one has been provided

 Position Perspex (PMMA) blocks of thickness 2 cm, 4 cm and 6 cm in turn on the breast
support platform. Ideally the Perspex blocks should cover the full detector or be at least
18x24 cm
 Position the blocks in a consistent manner (e.g. flush with the chest wall edge of the
detector)
 Compress to 50N (5kg or 5dN)
 Make an exposure in the clinically used mode (e.g. Sectra: Smart AEC, GE: Std Auto,
Hologic: Autofilter, Siemens : Opdose with Auto segmentation OFF)
 Record the kV, target/filter and post exposure mAs
2
 Draw a Region of Interest (ROI) of approximate size 4cm , positioned centrally along the
midline of the image and 6cm back from the chest wall
 Record mean pixel value (MPV) in this ROI

 Note for GE users: It is acceptable to use the triangular blocks provided. This test is
similar in principle to the AOP test but it is necessary to draw a ROI on the image and
measure the MPV. Additionally, it is felt that the GE tolerance levels for mAs variation are
too broad. It is therefore recommended that the procedure in this manual is followed
instead.

 Note for Sectra L30 users: The curved compression paddle means that the indicated
breast thickness will not represent the thickness of Perspex and this will result in an
inappropriate kV selection. Before making an exposure:
o Select the image type Phantom and
o Select the PMMA 20 (PMMA 40, PMMA 60 as appropriate) option in the drop-
down menu.

 Note for Sectra D40 users: This test is not required because of the complexity required to
obtain “unprocessed” image data and the MPV.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 49 of 75
Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, AEC mode and post exposure mAs
 MPV

Limits

• Mean pixel value (MPV) for each of 2, 4, and 6 cm Perspex should be within 10% of
baseline values

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 50 of 75
6.2a Record Sheet - AEC Calibration, DR

Frequency: Quarterly Test

X-ray machine ID:

AEC position: AEC Mode:

BASELINE DATA
Date: ______ Baseline Lower limit Upper limit Tolerance
MPV for 2 cm ±10%
MPV for 4 cm ±10%
MPV for 6 cm ±10%

Date: _______________ Initials: ________________

Phantom thickness kVp Target & filter mAs MPV
2 cm
4 cm
6 cm

Date: _______________ Initials: ________________

Phantom thickness kVp Target & filter mAs MPV
2 cm
4 cm
6 cm
Date: _______________ Initials: ________________

Phantom thickness kVp Target & filter mAs MPV

2 cm
4 cm
6 cm

Date: _______________ Initials: ________________

Phantom thickness kVp Target & filter mAs MPV
2 cm
4 cm
6 cm

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 51 of 75
6.3 Image Receptor Homogeneity, CR

Frequency: Quarterly Test (OR more frequently if recommended by the manufacturer)

This test is performed to ensure that the image receptor is uniform (that is, each part of the image
receptor provides the same response. The X-ray intensity will decrease as you move from the
chest wall to nipple edge due to the anode heel effect and the response will appear to be position
dependent accordingly. For this reason, the Regions of Interests (ROIs) used in the analysis are
drawn in a line parallel to the chest wall.

Unfortunately, not all CR systems will allow the extraction of pixel values within ROIs. To facilitate
this task it is necessary for the acquisition workstation (AWS) to have tools which allow
measurement of the mean pixel value (MPV) in a region of interest (ROI). However, if the image
can be sent to PACS then the analysis can be conducted on a reporting workstation. If that is not
possible then a visual evaluation of the image using a narrow window and suitably adjusted level
is all that is possible.

Procedure

 Note: The image obtained from the AEC Calibration test using the 18x24cm field size with
4 cm Perspex (section 5.1) may be used in lieu of procedure outlined below.
 Image a standard test block using the manufacturers recommended QC settings, (see QC
manual for your equipment). If no such settings are recommended use technical factors
appropriate for imaging a 4 cm breast and an acquisition algorithm that has minimal image
processing (see Appendix 2).
 The test block must cover the entire image receptor
2
 If the acquisition or other workstation allows place a ROI of size 1 cm in three specified
places on the image: one centrally on the chest wall edge of the image and the other two
placed near the chest wall corners of the image. The edge of each ROI should be no
closer to the image margins than 20 mm (see Figure 3 below). Note: one way to ensure
consistency in positioning of the ROIs is to create a ‘mask’ (e.g. a piece of cardboard with
holes cut out) which you can lay over the image
 Record the mean pixel values from each ROI.

Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, AEC detector position, density control setting, AEC Mode, post exposure
mAs and Exposure indicator
 Test results (MPV for each ROI)

Limits
Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 52 of 75
  Maximum deviation in the mean pixel value (MPV) between any two of the ROIs, <±10%.
 Maximum variation of the MPV in the central ROI between successive quarterly images <
±10%

C = Centre
L = Left
R = Right

20 mm ROIC 20 mm

ROIL ROIR

20 mm

Chest wall

Figure 3. Position of ROIs for CR uniformity Measurement

If the results lie outside the limiting values, an attempt should be made to confirm that this is due
to non-uniformity of the image plate and not inhomogeneities associated with the cassette, grid /
grid movement or the foils used as beam filters. It can be difficult to determine the exact cause of
the non-uniformity but it may be possible to eliminate some variables by repeating the test with a
different cassette and image plate combination. To rule out inhomogeneities associated with the
Perspex itself, rotate the Perspex 180° and repeat the test.

* If limitations in the software available for some CR units prevent a quantitative assessment, a
visual inspection of the image using a narrow “window” with the “level” suitably adjusted may be
all that can be done.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 53 of 75
6.3a Record Sheet – Image Receptor Homogeneity, CR

Frequency: Quarterly Test

X-ray machine ID:

AEC detector position: AEC Mode:

Date

kVp

Density control setting

Target/filter

mAs

Exposure Indicator

MPV for ROIC (MPVC)

MPV for ROIL (MPVL)

MPV for ROIR (MPVR)

MPVC - MPVL

MPVC – MPVR

MPVL – MPVR

Change in MPVC between

successive measurements

Initials

Limits

 Maximum deviation in the MPV between any two of the ROIs, <±10%.
 Maximum variation of the MPV in the central ROI between successive quarterly images <
±10%

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 54 of 75
6.4 Image Receptor Homogeneity, DR

Frequency: Quarterly Test (OR more frequently if recommended by the manufacturer)

This test is performed to ensure that the image receptor is uniform (that is, each part of the image
receptor provides the same response). For DR systems, a ‘flat-field’ correction is usually applied,
even to the ‘unprocessed’ image. This corrects for the variation in intensity due to the radiation
field, such as anode-heel and inverse square law effects. For this reason, the Regions of Interest
(ROI) are drawn in the centre and four corners of the image.

GE users: This test is included as part of the weekly flat-field calibration (see section
3.1). There is no need to follow the protocol below. It is acceptable to record the
brightness non-uniformity result on Form 5.4b.

The procedure calls for the use of the unprocessed image e.g. RAW (see Appendix 2).

Procedure

 Note: The image obtained for the “Full Field Artefact and System Check” (see Section 2.5)
may be used for this analysis in lieu of procedure outlined below.
 The standard, rigid or high edge compression paddle must be used unless the
manufacturer’s QC Manual says otherwise
 Image a standard test block (e.g. 4 cm Perspex) using the manufacturers recommended
QC settings, (see QC manual for your equipment). If no such settings are recommended
use technical factors appropriate for imaging a 4 cm breast and an acquisition algorithm
that has minimal image processing (e.g. “raw”).
 The test block must cover the entire image receptor
 Place a 1 cm square ROI in five specified places on the image: one centrally located and
the other four placed near the corners of the image. The edge of each of these ROIs
should be no closer to the image margins than 2 cm (see Figure 4 below). Note: one way
to ensure consistency in positioning of the ROIs is to create a ‘mask’ (e.g. a piece of
cardboard with holes cut out) which you can lay over the image
 Record the mean pixel values (MPVs) from each ROI.

Record

 Date test was performed

 Person performing test
 X-ray system identification
 kVp, target/filter, density control setting (if applicable), AEC Mode and post exposure mAs
 MPV test results
 GE Users only (use form 5.4b)

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 55 of 75
 20 mm 20 mm

ROILN ROIRN
20 mm 20 mm
C = Centre
ROIC
LCW = Left chest wall
LN = Left nipple
RCW = Right chest wall
RN = Right nipple
20 mm
20 mm
ROILCW ROIRCW

20 mm 20 mm

Chest wall

Figure 4: Position of ROIs for DR Uniformity measurement

Limits

 Maximum deviation in the mean pixel value (MPV) in each of the corner ROIs <±15% of the
MPV in the central ROI
 Maximum variation of the MPV in the central ROI between successive quarterly images <
±10%
 GE users: The brightness non-uniformity should be < ±10%.

Note: meeting the specification may be problematical if the Perspex block is not uniform. To
exclude failure due to such non uniformities a second image should be obtained with the block
rotated 180 degrees between exposures. The average of the MPVs in each of the comparable
ROIs is then used in the analysis. Note that this should be established at baseline. If the Perspex
block is found to be uniform at baseline then subsequent non-uniformity will be due to other
factors which should be investigated.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 56 of 75
6.4a Record Sheet – Image Receptor Homogeneity, DR (except GE)

Frequency: Quarterly Test

X-ray Machine ID ______________________

Date

kVp

AEC Density setting or detector

position (if relevant)

AEC Mode

Target/filter combination

mAs

MPV for ROIC (MPVC)

MPV for ROILCW (MPVLCW)

MPV for ROIRCW (MPVRCW)

MPV for ROILN (MPVLN)

MPV for ROIRN (MPVRN)

Max deviation in corner MPV from MPVC

(%)

Change in MPVC between successive

measurements

Initials

Limits

 Maximum deviation in the MPV between any corner ROI and the central ROI <±15%.
 Maximum variation of the MPV in the central ROI between successive quarterly images <
±10%

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 57 of 75
6.4b Record Sheet – Image Receptor Homogeneity, GE DR

Frequency: Weekly Test

X-ray Machine ID AEC Mode

Date Initials Brightness non- Date Initials Brightness non-

uniformity uniformity

Limits

 Brightness non-uniformity <±10%.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 58 of 75
6.5 Repeat Analysis

Frequency: Quarterly Test

This procedure remains as per existing mammography recommendations.

However, some new categories for repeat causes may need to be created to reflect the digital
environment, (e.g. software failures, blank images, non appearance of images on the acquisition
workstation, although an exposure was made etc.).

It is recommended that the PACS or digital system has some way of recording the number and
reason for rejected images and it may be possible to use this for reject / repeat analysis.
However, an independent record (paper or electronic) should still be kept.

Record

 The percentage of repeated images should be recorded on quarterly checklist

Limits

 Repeat rate <3%

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 59 of 75
7. SIX-MONTHLY TESTS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 60 of 75
7.1 Cassette Image Plate Condition & Interplate Sensitivity Variation, CR only

Frequency: Six Monthly test

The test is time consuming and it is also carried out by the equipment assessor on an annual
basis. It is therefore considered acceptable to do this test on an annual basis, in the interim 6
months, providing that you keep the assessor’s results on file.

This test is analogous to the uniformity of screen speed test of screen-film mammography. It is
necessary to ensure that all cassettes and image plates are comparable in their attenuation and
response respectively. If exposures are performed under AEC, we would expect approximately
constant mAs and Exposure Indicator.

Procedure
 Visually examine condition of cassettes and image plates
 Secondary erase image plates as per manufacturer’s recommendations
 Position 4cm Perspex on the breast support platform. Ideally the Perspex blocks should
cover the full detector or be at least 18x24cm. The edge of the Perspex should be flush
with the chest wall edge of the Bucky
 Compress to approximately 50 N (equals 5dN or 5 kg)
 Use a consistent AEC detector position (e.g. chest wall)
 Use the currently selected density control setting.
 Select the AEC mode used clinically (e.g. Auto STD for GE, H P2 for Siemens etc.)
 Image the QC “test” cassette/plate on three separate occasions to confirm repeatability of
the X-ray tube output acquiring the image with minimal processing as recommended in
Appendix 2
 Record kV, target / filter, post-exposure mAs and Exposure Indicator
 Using the same exposure conditions perform a single exposure with all other image plates
 Visually examine images for presence of artefacts.
 Repeat for 24x30 cm image plates if applicable

Record
 Date test was performed
 Person performing test
 X-ray system identification
 kVp, target/filter, AEC mode, density control setting
 Exposure indicator and post exposure mAs for each plate

Limits
 Confirm cassettes are in good condition and image plates are free of dust
 Fuji, Philips and Konica units: the S# for any image plate should differ from the mean S#
for that size by less than ±5%
 Carestream Health (Kodak) units: the EI value for any image plate should differ from the
mean EI value for that size by less than ±20
 Agfa units: the SAL value for any image plate should differ from the mean SAL value for
that size by less than ±2.5%; or SAL log within ± 220; or PVI log16 within ± 290 (NIP
plates)
 Confirm no major inhomogeneities or artefacts on the images.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 61 of 75
7.1a Record Sheet – Cassette/Image Plate Condition & Interplate Sensitivity Variation CR
only

Frequency: Six Monthly Test

Date X-ray Machine ID

kVp Target / Filter

AEC density setting AEC detector position

AEC Mode Initials

Field size 18 x 24 cm
Cassette mAs Exposure Deviation from Artefact
number* indicator (EI) mean EI (%) (Y/N?)
QC test
cassette*

Mean EI

Field size 24 x 30 cm
Cassette mAs Exposure Deviation from Artefact
number* indicator (EI) mean EI (%) (Y/N?)
QC test
cassette*

Mean EI
* PLEASE NOTE: The QC “test” cassette/plate should be irradiated three times to confirm
repeatability of the X-ray output.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 62 of 75
7.2 Compression

Frequency: Six Monthly Test

Whilst adequate compression is required for optimal mammography it is important the unit does
not allow application of unnecessarily high levels of force in either manual or power driven
modes. This also protects the compression paddle from undue pressure leading to possible
damage.

For QC purposes, no direct confirmation of the accuracy of the displayed force is expected. All
that is required is that the displayed value does not exceed the maximum allowed levels for
motorized and manual application of the compression force.

Procedure

 Place a rolled up towel or other suitable compressible material placed on the breast
support platform.
 Apply motorised compression and note the maximum force
 Apply compression manually and note the maximum force
 Sectra DR users: it is not possible to apply manual compression so measure the
maximum motorised force only.
 Record the values

Record

 Date test was performed

 Person performing test
 X-ray system identification
 Manual and motorised maximum forces

Limits

 Maximum compression force under manual compression ≤300 N (equals 30 kg or 30 dN)

 Maximum compression force under motorised compression ≥150 N (equals 15 kg or 15
dN) and ≤200 N (equals 20 kg or 20 dN).

Action

It may be possible to adjust maximum motorised force in the ‘set-up menu’ on the system. If this
is done, the test should be repeated to confirm the adjustments. If this is not possible, an
engineer should be called.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 63 of 75
7.2a Record Sheet – Compression Test

Frequency: Six Monthly Test

X-ray Machine ID
Initials

Date Displayed Force Displayed Force Pass / Fail Action Initials

Manual (N) Motorised (N)

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Y N

Limits

 Maximum compression force under manual compression ≤300 N (equals 30 kg or 30 dN)

 Maximum compression force under motorised compression ≥150 N (equals 15 kg or 15
dN) and ≤200 N (equals 20 kg or 20 dN).

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 64 of 75
8. TESTING AS REQUIRED

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 65 of 75
8.1 Maintenance & Fault Logging

Frequency: As Required

Separate logbooks for each imaging system, including diagnostic monitors, & film printer if
relevant should be kept. It is important to ‘close the loop’ on fault reports.

This means that in addition to recording the fault, you should note any engineer action taken and
any QA you, the medical physicist or the engineer have carried out to confirm that the problem
has been resolved. It is useful to keep the engineer’s service reports, either alongside the fault
log or in a separate folder.

Record

 Date maintenance or fault noted

 Description of fault encountered and/or maintenance performed
 Name of person making entry.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 66 of 75
9. SUMMARY FORMS

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 67 of 75
 9a Mammography Quality Control Checklist: Monthly or less frequently, CR only

For tests that should be performed monthly or less frequently

Site:

Machine ID:

Year

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Repeat analysis -enter %
(quarterly)
Image Receptor Homogeneity
(quarterly)
AEC Calibration test (quarterly)
Mechanical inspection (Monthly)
Image plate cleaning (Monthly)
Printer QC (dry) (Monthly)
Film digitiser QC (Monthly)
Compression (half yearly)
Cassette/image plate condition and inter plate
sensitivity variation
(half yearly)
Physicist’s check
(yearly)
Note: QC radiographer to initial relevant box when check completed.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 68 of 75
 9b Mammography Quality Control Checklist: Monthly or less frequently, DR only

For tests that should be performed monthly or less frequently

Site:

Machine ID:

Year

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Repeat analysis – enter %

(quarterly)
Image Receptor Homogeneity
(quarterly)
AEC Calibration test
(quarterly)

Printer QC (Dry) (Monthly)

Film digitiser QC (Monthly)

Mechanical inspection
(Monthly)
Compression
(half yearly)
Physicist’s check
(yearly)
Note: QC radiographer to initial relevant box when check completed.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 69 of 75
9c Mammography Quality Control Checklist: Daily or Weekly, CR only

For tests that should be performed daily or weekly

Site:
MQAP ID:
Year: Month:

Date 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Initials
DAILY

Monitor & view-box

inspection
Viewing conditions
Image plate erasure
(Secondary)
Printer QC (wet)
WEEKLY

Monitor & view-box

cleaning
Image Quality
Evaluation
Monitor QC
Image plate erasure
(Primary)

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 70 of 75
9d Mammography Quality Control Checklist: Daily or Weekly, DR only

For tests that should be performed daily or weekly

Site:
MQAP ID:
Year: Month:
Date 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Initials

DAILY

Monitor & view-box

inspection

Viewing conditions

Printer QC (wet)

Artefact evaluation

WEEKLY

Detector calibration –
Flat Field Test
Image Quality
Evaluation
Monitor & view-box
cleaning
Contrast to Noise Ratio
(CNR)

Monitor QC

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 71 of 75
Appendix 1

TG18-QC Test Pattern

Display devices (monitors and printers) form part of the imaging chain and it is imperative that QC is
carried out to ensure that:

 they meet the required standards at baseline, and

 their performance does not deteriorate over time.

In order to do this, the American Association of Physicists in Medicine (AAPM) has developed a
series of test patterns known as the Task Group 18 (TG18) series. These are not physical test objects
like the ACR phantom, but exist as digital files which can be loaded on to systems and displayed on
the monitor or printed. When new equipment is installed, it is important to know that these test
patterns have been loaded and how to access them. Sometimes they are listed as a patient but this
will not always be the case.
The TG18-QC test pattern is shown overleaf. This should be displayed at full resolution (one display
pixel for each pixel in the digital image). A visual inspection of the image should be carried out
including tests such as:

 Are the squares from black (100%) to white (0%) distinct?

 Can you see the 95% square inside the 100% square?
 Can you see the 5% square inside the 0% square?
 Can you resolve the high contrast line pair groups in all four corners and the centre?
 Can you see at least 11 letters in the phrase QUALITY CONTROL in all three rectangular
boxes?
 Are the grayscale ramps on the left and right of the image continuous (i.e. does the black blend
smoothly into the white)?

Further details are given in the relevant test protocols (2.4, 3.3 & 4.3).

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 72 of 75
Appendix 1

D1 MD D2

95% square 5% square

Dmax B+F

Figure A1. The TG18-QC test pattern with measurement points for optical densities and 5% and 95%
squares noted

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 73 of 75
Appendix 2

Image processing for CR and DR systems

For the majority of tests described in this protocol, it is necessary to acquire and view an image with
minimal image processing. In order to do this, we must use specific processing algorithms for CR, or
select a ‘RAW’ image for DR systems. Details of how to achieve this are shown below in Table A1
(CR) and Table A2 (DR).

An exception to the above requirement is the image quality test in Section 3.1. Because the ACR
phantom contains structures intended to mimic those found in the breast, it should be processed
using the ACR Phantom algorithm where available, (Fuji CR/DR and Hologic systems) or in other
instances an algorithm used for a clinical breast image (Siemens). For Carestream, Konica and Agfa
CR systems the minimally processed image should be used as advised by these manufacturers. For
Sectra and GE the immediately acquired and automatically displayed processed image should be
used for image quality scoring.

CR systems have a parameter called the ‘Exposure Indicator’ which is related to the dose received by
the detector. For screen-film systems we set a target Optical Density in range 1.6- 2.0 and we aim to
achieve this value on all films, regardless of breast density or thickness. We cannot measure OD for
softcopy images so instead we aim for a target Exposure Indicator, defined by the manufacturer, and
check for consistency. Further details are shown in Table A1.

Table A1: Exposure Indicator and Processing Algorithms for CR Systems

Manufacturer Exposure Indicator Relevant processing algorithms

Konica S-value Mammo test phantom for all tests including

ACR phantom image

Fuji S-value QC test/Sensitivity or equivalent. (minimal

image processing)
ACR Phantom (for ACR phantom image):

Philips S-value Service Test Sens High Matrix (minimal

image processing)
Unique (for ACR phantom image):

Carestream EI Pattern Raw (minimal image processing) for

all tests, including ACR phantom image

Agfa SAL, SAL Log, PVI log, Flat field for all tests including ACR phantom
EI image

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 74 of 75
Appendix 2

Table A2: Relevant image types for DR Systems

Manufacturer Raw or pre-processed image ACR Phantom Image

GE Raw - after making the exposure, close the Processed - use the
study. Go back to the Browser window and immediately acquired and
click on your ‘patient’. You will see that you automatically displayed
can choose the RAW or PROC images. processed image for image
Choose RAW for minimal image processing. quality scoring.

Hologic Flatfield - when adding additional views to ACR Phantom - add the
the patient, add Flatfield (minimal image ACR Phantom view to the
processing). patient study.

Sectra Pre-processed Processed - The ACR

phantom image should be
L30 Software v. 8.3 – Right-click on the
scored using the
image. Go to Re-label – Raw. Click ‘Yes’ on
immediately acquired and
the subsequent warning message.
displayed processed image.
D40: Not applicable for routine radiographer
QC

Fuji QC Test - Non-linearised raw (minimal ACR Phantom - when

image processing) - when adding additional adding additional views to
views to the patient, simply add QC Test – the patient, simply add ACR
Non linearised raw from the EDR menu. Phantom from the EDR
menu

Siemens QC Raw - the displayed image is matched to QC Processed - a separate

the selection of QC raw (or QC processed if study must be created if you
applicable) decided on at Registration of wish to image the ACR
Study. phantom for scoring image
quality purposes as QC Raw
and QC processed images
cannot be mixed in one
study.

Guidelines for Quality Control Testing for Digital (CR & DR) Mammography Version 3.0
© The Royal Australian and New Zealand College of Radiologists®
August 2012
Page 75 of 75
THE ROYAL AUSTRALIAN AND NEW ZEALAND COLLEGE OF RADIOLOGISTS®