Position paper: recommendations for a digital mammography quality assurance program V4.0

Abstract

In 2001 the ACPSEM published a position paper on quality assurance in screen film mammography which was subsequently adopted as a basis for the quality assurance programs of both the Royal Australian and New Zealand College of Radiologists (RANZCR) and of BreastScreen Australia. Since then the clinical implementation of digital mammography has been realised and it has become evident that existing screen-film protocols were not appropriate to assure the required image quality needed for reliable diagnosis or to address the new dose implications resulting from digital technology. In addition, the advantages and responsibilities inherent in teleradiology are most critical in mammography and also need to be addressed. The current document is the result of a review of current overseas practice and local experience in these areas. At this time the technology of digital imaging is undergoing significant development and there is still a lack of full international consensus about some of the detailed quality control (QC) tests that should be included in quality assurance (QA) programs. This document describes the current status in digital mammography QA and recommends test procedures that may be suitable in the Australasian environment. For completeness, this document also includes a review of the QA programs required for the various types of digital biopsy units used in mammography. In the future, international harmonisation of digital quality assurance in mammography and changes in the technology may require a review of this document. Version 2.0 represented the first of these updates and key changes related to image quality evaluation, ghost image evaluation and interpretation of signal to noise ratio measurements. In Version 3.0 some significant changes, made in light of further experience gained in testing digital mammography equipment were introduced. In Version 4.0, further changes have been made, most notably digital breast tomosynthesis (DBT) testing and QC have been addressed. Some additional testing for conventional projection imaging has been added in order that sites may have the capability to undertake dose surveys to confirm compliance with diagnostic reference levels (DRLs) that may be established at the National or State level. A key recommendation is that dosimetry calculations are now to be undertaken using the methodology of Dance et al. Some minor changes to existing facility QC tests have been made to ensure the suggested procedures align with those most recently adopted by the Royal Australian and New Zealand College of Radiologists and BreastScreen Australia. Future updates of this document may be provided as deemed necessary in electronic format on the ACPSEM’s website (https://www.acpsem.org.au/whatacpsemdoes/standards-position-papers and see also http://www.ranzcr.edu.au/quality-a-safety/radiology/practice-quality-activities/mqap).

Appendices

Appendix 1: Summary of recommendations for facility QC procedures for DR units in 2D mode

Procedure	Recommended control-limits/requirements	Minimum frequency	Key procedure elements	Recommendations for record keepinga
Viewing conditions	Appropriate viewing conditions All viewbox lamps must be operational and appropriate masking available	Daily	Visual inspection of ambient lighting conditions to ensure conformance with acceptable viewing condition configuration (see text for detail) Visual inspection of viewboxes for uniformity of brightness Confirmation of presence and operation of masking for viewboxes	Checklist/logbook entry showing  Date performed  Person performing task
Full field artefact evaluation	mAs = baseline ± 10% Mean pixel value in image = baseline ± 10% There must be no evidence of  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 breast support structures  Bright or dark pixels  Dust artefacts mimicking calcifications  Significant stitching or registration artefacts	Daily	Expose a uniform thickness of PMMA using clinically relevant technique factors  Image must be acquired in “processed” or “for presentation” form  Measure mean pixel value in 4 cm2 ROI positioned centrally along long axis of image and 6 cm from chest wall  View image on acquisition monitor using zoom and roam to check for possible detector faults  Print image if interpretation performed using hard copy	Records showing  Date test was performed  Person performing test  Test results  kVp, target/filter and mAs
Monitor QC (Monitors used for interpretation and attached to the acquisition workstation)	Borders must be visible, lines must be straight, squares must appear square, the ramp bars should appear continuous without any contour lines, there must be no smearing or bleeding at black-white transitions, all corner patches must be visible, squares of different shades from black to white must be distinct, all high contrast resolution patterns and two low contrast patterns must be visible in all four corners and in the centre, the 5% and 95% pixel value squares must be clearly visible, pattern must be centred in the active area and no disturbing artefacts must be visible on the displayed TG18-QC test pattern The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Weekly	Display TG18-QC test pattern Ensure viewing conditions are acceptable Use window-width set to maximum and window-level set to half of maximum	Records showing  Date test was performed  Person performing test  Monitor identification  Monitor settings  Test results
Monitor cleaning	Monitor screens must be free of dust, fingerprints and other marks that might interfere with image interpretation	Weekly	Clean all monitor screens gently with lint-free cloth as per manufacturer’s instructions	Checklist/logbook entry showing  Date performed  Person performing task
Printer area cleanliness (if applicable)	Clean and dust free environment	Weekly	Wet cleaning of printer area floor and open shelves. Inspect and clean air intake filters on the film printer	Checklist/logbook entry showing  Date performed  Person performing task
Image quality evaluation	mAs = baseline ± 10% For hard copy reporting optical density = baseline ±20% and must be in the range of 1.60–2.0 The ability to clearly visualise 5 fibres, 3.5 speck groups (4 is desirable) and 4 masses in an image of an ACR Accreditation phantom or Alternatively: the ability to clearly visualise 4 fibres, 3 speck groups and 4 masses in an image of the new ACR DM phantom	Weekly	Obtaining the phantom image  Use of ACR Accreditation phantom or new ACR DM phantom  Use of a consistent AEC detector position where this is manually selected  Light contact between the compression paddle and the phantom surface  Consistent positioning of the phantom  Consistent selection of clinically relevant kVp and target/filter combinations  Selection of the density setting in current clinical use (if applicable) Evaluating the phantom image (preferably on reading workstation or on printed copy if hardcopy reporting used)  Use “for presentation” image with zoom and modest adjustment of window/level functions to score fibres and specks  Use consistent (baseline) viewing conditions that reflect those used to read actual mammograms  Image quality scoring by the same person, if possible  Measure optical density in reproducible part of phantom image if hardcopy reporting  Use of a control chart to record results	Record numerical mAs values and image quality scores Control chart showing  Plots of mAs, image quality score/s, and OD if applicable ≥25 results  Clearly marked control limits  Baseline values  Radiographic settings (kVp, target/filter combination, density setting and SID)  Remarks e.g. corrective action Phantom images identifying  Date  The X-ray system  The technique factors
Detector calibration—flat field test	Pass or fail	Weekly or as per manufacturer’s requirements	Follow manufacturer’s specific procedure	Checklist/logbook entry showing  Date performed  Person performing task
Signal difference to noise ratio (SDNR)	SDNR = baseline ± 20%	Weekly	Preferably follow manufacturer procedure. Alternatively  Use the “for presentation” image obtained with either the ACR Accreditation or ACR DM phantom for image quality purposes but with PMMA disc on paddle (if using the ACR DM phantom there is a negative contrast disc in the phantom)  Measure the mean pixel value (MPV1) and SD in a small ROI next to PMMA disc (or negative contrast disc in the ACR DM phantom)  Measure mean pixel value (MPV2) in ROI centred in disc  Calculate SDNR = (MPV1 − MPV2)/SD	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, AEC mode and mAs  Test results
Printer QC (if applicable)	Borders must be visible, lines must be straight, all corner patches must be visible, squares of different shades from black to white must be distinct, all high contrast resolution patterns must be visible in all four corners and the centre, the 5% and 95% pixel value squares must be clearly visible, and no disturbing artefacts must be visible on the printed TG18-QC test pattern The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11.The mid density (MD) and density difference (DD) = baseline ± 0.15 Base + fog (B+F) = baseline ± 0.03 and ≤0.25 Dmax = baseline ± 0.10 and ≥3.4	Monthly for dry lasers and daily or as used for wet lasers	Print the TG18-QC test pattern Check visibility and distortion of several items used for evaluating the quality of the image Check for disturbing artefacts Measure MD, DD, B+F and Dmax	Control charts and records showing  Date test was performed  Person performing test  Printer identification  Test results
Mechanical inspection	Indicated breast thickness accurate to ±5 mm No hazardous, inoperative, out of alignment or improperly operating items on the system All items listed on the visual check list have received a pass	Monthly	Confirm accuracy of thickness indication Visual inspection of the system to ensure safe and optimum operation	Checklist/logbook entry showing  Date inspection performed  Inspection results  Person performing test
Repeat analysis	Repeat rate <3% (<2% preferred) [3]	Quarterly	Inclusion of images from at least 250 consecutive client examinations The ability to determine repeat rates attributable to a range of equipment faults and positioning errors	Worksheet/logbook entries showing all results/calculations
Image receptor homogeneity	Maximum deviation in mean pixel value in ROI <±10% of mean pixel value in central ROI Maximum variation of the mean pixel value in central ROI between successive quarterly images <±10%	Quarterly or more frequently if recommended by the manufacturer	Use manufacturer’s protocol and test block if available; otherwise Image a standard test block at clinical settings On the “for processing” image, draw 100 mm2 square or circular ROIs in the centre and four corners If the mean pixel value of a ROI deviates by more than 15% from the mean pixel value in the central ROI, the detector gain map may require re-calibration If required, to exclude failure due to non-uniformities in the standard test block, rotate latter by 180° and repeat measurement	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, density setting and mAs  Test results
AEC calibration test	Mean pixel value for each of 2, 4 and 6 cm PMMA within 10% of baseline values	Quarterly	Assess for both contact and magnification modes Use PMMA thickness between 2 and 6 cm covering complete image receptor Use clinical AEC settings (kVp, target/filter and mode) Measure mean pixel value in 4 cm2 ROI positioned centrally along axis and 6 cm from chest wall Examine image for clinically significant artefacts	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, AEC mode and mAs  Test results
Compression	Maximum motorised compression force in range 150–200 N	6 monthly	Confirm machine indicated compression force meets requirements	Checklist/logbook entry showing  Date test performed  Test results  Person performing test
Test equipment quality control  Densitometer calibration check	Optical density measurement accurate to within  ±0.03 (0–3.0 OD)  ±3% (3.0–4.0 OD)	6 monthly	Verification of accuracy using an optical density calibration strip traceable to an accepted standard	Checklist/logbook entry showing  Date test performed  Test results  Person performing test
Maintenance and fault logging	Separate logbooks for each imaging system including diagnostic monitors and film printer if relevant	As required	Dated entries describing fault encountered and/or maintenance performed	Logbooks with dated and initialled entries
Infection control of breast imaging equipment	Clean equipment	Before each examination	Cleaning using alcohol swipes, or as per manufacturer’s recommendations and/or suitable infection control advice	Nil

1. ^aAll written/electronic QC records should be retained for a minimum of 1 year unless otherwise indicated by local Regulatory requirements. Images used to assess image quality with the ACR Accreditation or ACR DM phantom should be retained for a minimum of 1 month

Appendix 2: Summary of recommendations for facility qc procedures for CR units

Procedure	Recommended control-limits/requirements	Minimum frequency	Key procedure elements	Recommendations for record keepinga
Viewing conditions	Appropriate viewing conditions All viewbox lamps must be operational and appropriate masking available	Daily	Visual inspection of ambient lighting conditions to ensure conformance with acceptable viewing condition configuration (see text for detail) Visual inspection of viewboxes for uniformity of brightness Confirmation of presence and operation of masking for viewboxes	Checklist/logbook entry showing  Date performed  Person performing task
Image plate erasure	Erasure of energy absorbed from scattered radiation or naturally occurring radiation by CR image plates before they are used	Daily/weekly	On a daily basis or if left unused for more than 8 hours, all CR image plates must be subjected to an erasure (following manufacturer’s instructions) On a weekly basis all Fuji CR image plates must be subjected to a primary erasure	Logbooks with dated and initialled entries
Monitor/viewboxes cleaning	Monitor screens and viewboxes must be free of dust, fingerprints and other marks that might interfere with image interpretation	Weekly	Clean all monitor screens and viewboxes gently with lint-free cloth as per manufacturer’s instructions	Checklist/logbook entry showing  Date performed  Person performing task
Monitor QC (monitors used for interpretation and attached to the acquisition workstation)	Borders must be visible, lines must be straight, squares must appear square, the ramp bars should appear continuous without any contour lines, there should be no smearing or bleeding at black-white transitions, all corner patches must be visible, squares of different shades from black to white must be distinct, all high contrast resolution patterns and two low contrast patterns must be visible in all four corners and the centre, the 5 and 95% pixel value squares must be clearly visible, pattern must be centred in the active area and no disturbing artefacts must be visible on the displayed TG18-QC test pattern The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Weekly	Display TG18-QC test pattern Ensure viewing conditions are acceptable Use window-width set to maximum and window-level set to half of maximum	Records showing  Date test was performed  Person performing test  Monitor identification  Monitor settings  Test results
Monitor cleaning	Monitor screens must be free of dust, fingerprints and other marks that might interfere with image interpretation	Weekly	Clean all monitor screens gently with lint-free cloth as per manufacturer’s instructions	Checklist/logbook entry showing  Date performed  Person performing task
Printer area cleanliness (if applicable)	Clean and dust free environment	Weekly	Wet cleaning of printer area floor and open shelves. Inspect and clean air intake filters on the film printer	Checklist/logbook entry showing  Date performed  Person performing task
Image quality evaluation	mAs = baseline ± 10% Dose to plate = baseline ±10% Exposure indicator (see Appendix 6 for manufacturer dependent tolerances) For hard copy reporting optical density = baseline ±20% and must be in the range of 1.60–2.0 The ability to clearly visualise 5 fibres, 3.5 speck groups (4 is desirable) and 4 masses in an image of an ACR Accreditation phantom or The ability to clearly visualise 4 fibres, 3 speck groups and 3 masses in an image of an ACR DM phantom	Weekly	Obtaining the phantom image  Use an ACR Accreditation phantom or the new ACR DM phantom  Use of a designated test cassette and imaging plate that is in routine clinical use  Use of a consistent AEC detector position where this is manually selected  Light contact between the compression paddle and the phantom surface  Consistent positioning of the phantom  Consistent selection of clinically relevant kVp and target/filter combinations  Selection of the density setting in current clinical use  Consistent time delay between plate irradiation and readout Evaluating the phantom image (preferably on reading workstation or on printed copy if hardcopy reporting used)  Use “for presentation” image with zoom and modest adjustment of window/level functions to score fibres and specks  Use of consistent viewing conditions that reflect those used to read actual mammograms. This applies to both soft and hard copy  Image quality scoring by the same person, if possible  Measure optical density in reproducible part of phantom image if hardcopy reporting  Use of a control chart to record results	Record numerical mAs values and image quality scores Control chart showing  Plots of mAs, exposure indicator, image quality score/s and OD if applicable ≥25 results  Clearly marked control limits  Baseline values  Radiographic settings (kVp, target/filter combination, density setting and SID)  Remarks e.g. corrective action Phantom images identifying  Date  The X-ray system  The technique factors
Printer QC	Borders must be visible, lines must be straight, all corner patches must be visible, squares of different shades from black to white must be distinct, all high contrast resolution patterns must be visible in all four corners and the centre, the 5% and 95% pixel value squares must be clearly visible, and no disturbing artefacts must be visible on the printed TG18-QC test pattern The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11 The mid density (MD) and density difference (DD) = baseline ± 0.15 Base + fog (B+F) = baseline ± 0.03 and ≤0.25 Dmax = baseline ± 0.10 and ≥3.4	Monthly for dry lasers and daily or as used for wet lasers	Print the TG18-QC test pattern Check visibility and distortion of several items used for evaluating the quality of the image Check for disturbing artefacts Measure MD, DD, B+F and Dmax	Control charts and records showing  Date test was performed  Person performing test  Printer identification  Test results
Mechanical inspection	Indicated breast thickness accurate to ±5 mm No hazardous, inoperative, out of alignment or improperly operating items on the system All items listed on the visual check list have received a pass	Monthly	Confirm accuracy of thickness indication Visual inspection of the system to ensure safe and optimum operation	Checklist/logbook entry showing  Date performed  Inspection results  Person performing task
Repeat analysis	Repeat rate <3% (<2% preferred) [3]	Quarterly	Inclusion of images from at least 250 consecutive client examinations The ability to determine repeat rates attributable to a range of equipment faults and positioning errors	Worksheet/logbook entries showing all results/calculations
Image receptor homogeneity	Maximum difference in mean pixel value between any two ROIs <±10% Maximum variation of the mean pixel value in central ROI between successive QC images <±10%	Quarterly or more frequently if recommended by the manufacturer)	Use manufacturer’s protocol and test block if available; otherwise Image a standard test block at clinical settings Use ‘test’ cassette Perform measurements on the “for processing” (unprocessed) image, if possible, using a 100 mm2 square or circular ROI. Three ROIs are placed at the left, right and centre on a line 20 mm back from chest wall If the mean pixel value of any two ROIs differ by more than 10% from each other, the CR unit’s shading correction may require re-calibration or imaging plate(s) may require replacement If ROI analysis is not possible, do a visual inspection at narrow window width If required, to exclude failure due to non-uniformities in the standard test block, rotate by 180° and repeat measurement	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, density setting and mAs  Test results
AEC calibration test	Dose to plate for each of 2, 4 and 6 cm PMMA = baseline ±10% See Appendix 6 for equivalent manufacturer specific exposure index requirements	Quarterly	Assess for both contact and magnification modes Use PMMA thickness between 2 and 6 cm covering complete cassette Use clinical AEC settings (kVp, target/filter and mode including density setting) Use a designated test cassette and imaging plate that is in routine clinical use Use a consistent AEC detector position where this is manually selected Consistent positioning of the PMMA Consistent time delay between plate irradiation and readout	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, AEC mode and mAs  Test results
Compression	Maximum motorised compression force in range 150–200 N	6 monthly	Confirm machine indicated compression force meets requirements	Checklist/logbook entry showing  Date test performed  Test results  Person performing test
Test equipment quality control  Densitometer calibration check	Optical density measurement accurate to within ±0.03 (0–3.0 OD) ±3% (3.0–4.0 OD)	6 monthly	Verification of accuracy using an optical density calibration strip traceable to an accepted standard	Checklist/logbook entry showing  Date test performed  Test results  Person performing test
Cassette/image plate condition and inter plate sensitivity variation	Clean and dust free cassettes and image plates No major inhomogeneities on the images See Appendix 6 for manufacturer specific tolerances on interplate variations	6 monthly	Cassette/image plate cleaning as per manufacturer’s recommendations Image a standard test block at clinical settings Pre-processing should be turned off as much as possible and no post processing must be applied Evaluate for artefact on both film (if applicable) and monitor	Records showing  Date test was performed  Person performing test  kVp, target/filter, AEC mode  Exposure indicator and mAs for each plate
Maintenance and fault logging	Separate logbooks for each imaging system, including diagnostic monitors, and film printer if relevant.	As required	Dated entries describing fault encountered and/or maintenance performed	Logbooks with dated and initialled entries
Infection control of breast imaging equipment	Clean equipment	Before each examination	Cleaning using alcohol wipes, or as per manufacturer’s recommendations and/or suitable infection control advice	Nil

1. ^aAll written/electronic QC records should be retained for a minimum of 1 year unless otherwise indicated by local Regulatory requirements. Images used to assess image quality with the ACR Accreditation or ACR DM phantom should be retained for a minimum of 1 month

Appendix 3: Summary of recommendations for medical physics testing only at acceptance or equipment upgrade in 2D mode

Procedure	Performance requirements/guidelines	Routine testing guidelines	Key procedure elements
Focal spot	≥11 lp/mm for line-pair bars perpendicular to anode–cathode axis and ≥13 lp/mm for line-pair bars parallel to anode–cathode axis OR complies with IEC 60336 [38] for 0.3 and 0.1 mm focal spot sizes	Not required unless tube has been changed	As per “Focal spot size” section or IEC 60336 [38]
Leakage radiation	≤1 mGy/h at 1m from focus and ≤0.01 mGy/100 mAs @ 30 kVp and 30 cm from focus	Not required unless tube has been changed or system relocated	As per AS/NZS IEC 60601.1.30.1.3 [39]
Transmission through breast support	≤0.001 mGy @ max kVp and mAs	Not required unless change made to image receptor system	As per AS/NZS IEC 60601.1.3 [39]
Missed tissue @ chest wall	Width of missed tissue at chest wall ≤5 mm in contact mode and ≤7 mm in magnification mode	Not required unless tube has been changed or change made to image receptor system or system relocated
Plate fogging (CR only)	Image of coin must not be visible	Not required unless changes in storage of cassettes have occurred	Monitor during acceptance testing
MTF	Bench mark testing, compare to manufacturer’s specification	Not required unless tube has been changed or change made to image receptor system	As per IEC 62220-1-2 [48]
Threshold contrast visibility		Not required unless tube has been changed or change made to image receptor system	Use CDMAM phantom
Spatial linearity and geometric distortion		Not required unless change made to image receptor system	Use wire mesh tool
Distance calliper accuracy	Measured dimensions of ruler in image must be within 2% of true dimensions in plane specified by manufacturer Check both contact and magnification modes	Not required unless change made to image receptor system	Use steel rulers. Determine dimensions in image, ideally using reporting workstation

1. This test allows digital systems to be benchmarked against European standards [9]

Appendix 4: Summary of recommendations for medical physics annual testing of DR units in 2D mode

Procedure	Performance requirements/guidelines	Routine testing guidelines	Acceptance and additional tests
Mammography unit assembly evaluation	Correct and safe function of system components. Thickness display accuracy within ±5 mm, note: Flexi paddles will not comply (manufacturer recommendation varies ~11–12 mm for flexi paddles). Reproducible to 2 mm. Verify DICOM image header for correct display of parameters	Confirm function of all motorised components, warning lights, displays etc. Evaluate system for any miscellaneous safety risks etc. DICOM verification required after software upgrades	As per routine tests
Collimation and alignment assessment
X-ray field/Image receptor alignment	The X-ray field shall irradiate the image receptor fully but not extend beyond the breast support on the chest wall edge of the image receptor by more 2 mm	Assess alignment for each target/geometry combination	As per routine tests
Paddle/Image alignment	The chest wall edge of the compression paddle shall be aligned just beyond the chest wall edge of the image receptor such that it does not appear in the image. In addition, the compression paddle shall not extend beyond the chest wall edge of the image by more than 1% of the SID	Assess alignment for all clinically relevant Bucky/paddle/target/geometry combinations	As per routine tests
System resolution/MTF	Compare to baseline values, variation less than 10%	Measure MTF using system software if possible. Otherwise measure limiting resolution  Use a 4cm PMMA block or equivalent  Place resolution pattern on PMMA  Measure both parallel and perpendicular to chest wall  Repeat for Magnification mode if applicable	Establish base line values
AEC system performance assessment
Reproducibility	Coefficient of variation (COV) for both absorbed dose and mAs for at least three phototimed exposures of a test object shall be better than or equal to 0.05	Use a 4cm PMMA block or equivalent Assess COV for each AEC detector at a typical clinical kVp	As per routine tests
Compensation and SDNR system performance assessment	Compare SDNR values to baseline and to the minimum acceptable values for 4 cm PMMA (SDNRaccept)  SDNR2cm > 1.1 × SDNRaccept  SDNR4cm > SDNRaccept  SDNR6cm > 0.9 × SDNRaccept	Assess the most commonly used AEC modes for contact and magnification geometry Use 0.2 mm Al foil as contrast test tool and measure SDNR for 2, 4 and 6 cm PMMA (also see section on glandular dose). Note: measurements are to be undertaken on “for processing” (unprocessed) images	Establish baseline values. Assess all available AEC modes for contact and magnification geometries
Density control (if applicable)	The density control must be capable of changing the mAs from the value used normally by −25 to +50%	Assess change in mAs for at least two density settings either side of the usual clinical setting using 4 cm of PMMA	Assess change in mAs across full range of density settings
Back-up timer/security cut-out	Security cut-out mechanisms shall be present and terminate the exposure within 50 ms or within 5 mAs, otherwise the back-up timer should terminate the exposure at ≤500 mAs and must terminate the exposure at ≤800 mAs	Use lead sheet or other heavily attenuating material to intercept beam and confirm that the back-up timer security cut-out functions within specified limits	Confirm that the back-up timer/security cut-out functions within specified limits
Image uniformity and artefact	Max. deviation of mean pixel value <±15% of mean pixel value for central ROI Max. deviation in SNR as a function of time is ±10% There must be no evidence of blotches or regions of altered noise appearance, observable grid lines or breast support structures, bright or dark pixels	Assess for 40 mm PMMA covering complete detector Use five ROIs (one central, with the other 4 approximately 20 mm from any edge) each of 100 mm2 Measurements performed on unprocessed image Exclude phantom non uniformity by rotating block 180° and repeating Repeat in magnification mode if applicable	Assess also at 20 and 60 mm
Detector element failure	Limits currently not established. Must monitor independent of manufacturer Inspect bad pixel map	A mammographic screen-film mesh can be used to determine if correction for bad columns successful	Bad pixel map must be available at any time, independent of manufacturer
Image quality evaluation	The ability to clearly visualise 5 fibres, 3.5 speck groups (4 is desirable) and 4 masses in an image of an ACR Accreditation phantom or the ability to clearly visualise 4 fibres, 3 speck groups and 3 masses in an image of the ACR DM phantom Additionally, with the ACR DM phantom the SDNR with contrast object ≥2.0	Use typical clinical settings Measure MPVs and SDs in relevant ROIs of ACR DM phantom so that SDNR may be calculated.	As per routine testing
Ghost image evaluation	“Ghost image” factor <2.0	Assess using 40 mm PMMA (see “Ghost image evaluation” section for testing guidelines)	As per routine testing
System linearity and noise analysis	Linearity plot versus ESAK: R2 > 0.99 SD2 plot versus MPV: R2 > 0.99 Noise parameters: compare to baseline results	Use standard test block (e.g. 4 cm PMMA) at typical clinical beam settings Measure ESAK at 6 cm from chest wall Measure mean pixel value and SD in ROI placed 6 cm from chest wall Plot mean pixel value as a function of ESAK Plot SD2 as a function of MPV corrected for any pixel offset	Baseline measurements at clinical kVp, also at max and min clinical kVps for all target filter combinations
Generator performance
kVp, reproducibility	COV ≤ 0.02 for a minimum of three exposures	Assess kVp reproducibility at a typical clinical kVp value	As per routine testing
kVp accuracy	Measured kVp shall be within ±5% of the specified value over the clinically relevant range	Assess kVp accuracy over the clinically relevant range in, at most, 2 kVp increments Note: The kVp need only be verified for one target filter combination per kVp, however the kVp meter must be calibrated for that particular target/filter combination	Assess kVp accuracy over clinically relevant range in 1 kVp increments
Beam quality	$\left( {{\text{kVp}}/100} \right)+0.03 \leq {\text{HVL}}<\left( {{\text{kVp}}/100} \right)+{\text{C}}$ $\begin{aligned} {\text{where C }}&=&0.{\text{12 mm Al for Mo}}/{\text{Mo}} \hfill \\&=&0.{\text{19 mm Al for Mo}}/{\text{Rh}} \hfill \\&=&0.{\text{22 mm Al for Rh}}/{\text{Rh}} \hfill \\&=&0.{\text{23 mm Al for Rh}}/{\text{Ag}} \hfill \\&=&0.{\text{3}}0{\text{ mm Al for W}}/{\text{Rh}} \hfill \\&=&0.{\text{32 mm Al for W}}/{\text{Ag}} \hfill \\&=&0.{\text{25 mm Al for W}}/{\text{Al}}. \hfill \\ \end{aligned}$	Measure the HVL required for mean glandular dose calculations and for establishing compliance with DRLs (see “Mean glandular dose” section for details)	As per routine tests plus measure HVL at 28 kVp for all target/filter combinations, with the compression paddle removed if unit used for biopsy purposes with open paddle
Mean glandular dose	≤2.0 mGy for a 4.2 cm 50% adipose, 50% glandular breast (i.e. ACR Accreditation phantom or ACR DM phantom) <1 mGy for 2.0 cm PMMA (2.3 cm 50% adipose, 50% glandular breast) <4.5 mGy for 6.0 cm PMMA, (6.5 cm 50% adipose, 50% glandular breast) the displayed MGD values must agree with calculated values to ≤25%	Assess for an AEC controlled exposure using typical clinical settings using ACR Accreditation phantom (or ACR DM phantom) and also for 20 mm and 60 mm PMMA Additional dose (and HVL) measurements may be necessary to confirm compliance with DRLs (see “Mean glandular dose”)	As per routine tests
Exposure time	For all clinically relevant SID settings the maximum exposure time when irradiating 6 cm PMMA must be less than 3.5 and 2 s for fine and broad focus, respectively	Assess for both contact and magnification modes Use 6 cm of PMMA Use clinically relevant technique factors for this PMMA thickness consistent with SDNR and MGD measurements Record mAs and infer the exposure time from tube rating or measure directly using a manual exposure matched to mAs needed for AEC initiated exposure	As per routine tests
Viewbox luminance and room illuminance (hardcopy only)	Viewing area illuminance ≤50 lx Viewbox luminance ≥3000 cd/m2	Assess viewing conditions for all viewers	As per routine tests
Monitor luminance and viewing conditions	Image interpretation must not be done on a monitor of less than 4.2 mega pixels Luminance ratio approximately 350:1 Maximum luminance >450 cd/m2 and maximum luminance of paired monitors matched to ≤5% Minimum luminance preferably not less than 1 cd/m2 Ambient light <20 lx In PACS situations images must be stored with lossless compression	Measure luminance ratio under clinical lighting conditions Confirm luminance uniformity Confirm no cross-talk and pixel defects	As per routine testing with the additional requirement of checking GSDF Monitor or workstation may have comprehensive QC program which needs to be validated
Monitor performance	No smearing artefact, ramps without terracing Lines straight, boxes square, active display centred, borders complete Squares of different shades from black to white must be distinct and small squares in corners of each clearly discernible Free from artefact The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Test patterns to be displayed at full resolution Test under clinical lighting conditions Use TG18-QC test pattern	As per routine testing. Monitor or workstation may have comprehensive QA program
Printer (hardcopy only)	B + F = baseline ± 0.03 and ≤0.25 OD Dmax = baseline ± 0.10 and ≥3.4 OD The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Print TG18-QC test pattern as per weekly printer QC test	As per routine tests

Appendix 5: Summary of recommendations for medical physics annual testing of CR units

Procedure	Performance requirements/guidelines	Routine testing guidelines	Acceptance and additional tests
Mammography unit assembly evaluation	Correct and safe function of system components. Thickness display accuracy within ±5 mm, reproducible to 2 mm. Verify DICOM image header for correct display of parameters	Confirm function of all motorised components, warning lights, displays etc. Evaluate system for any miscellaneous safety risks etc. DICOM verification required after software upgrades	As per routine tests
Collimation and alignment assessment
X-ray field/image/breast-support alignment	The X-ray field shall irradiate the image receptor fully but not extend beyond the breast support on the chest wall edge of the image receptor by more than 2 mm	Assess alignment for largest collimator in clinical use for each Bucky/target combination. For magnification geometry only assess chest wall alignment	As per routine tests
Paddle/image alignment	The chest wall edge of the compression paddle shall be aligned just beyond the chest wall edge of the image receptor such that the chest wall compression paddle does not appear in the image. In addition the compression paddle shall not extend beyond the chest wall edge of the image receptor by more than 1% of the SID	Assess alignment for all clinically relevant Bucky/paddle/geometry combinations	As per routine tests
System resolution/MTF	Compare to baseline values, variation less than 10%	Measure MTF using system software if possible. Otherwise measure limiting resolution  Use a 4cm PMMA block or equivalent  Place resolution pattern on PMMA  Measure both parallel and perpendicular to chest wall  Repeat for Magnification mode if applicable	Establish base line values
AEC system performance assessment
Reproducibility	Coefficient of variation (COV) for both absorbed dose and mAs for at least three phototimed exposures of a test object shall be better than or equal to 0.05	Use a 4cm PMMA block or equivalent Assess COV for each AEC detector at a typical clinical kVp	As per routine tests
Compensation and SDNR system performance assessment	Compare SDNR values to baseline and to the minimum acceptable values for 4 cm PMMA (SDNRaccept)  SDNR2cm > 1.1 × SDNRaccept  SDNR4cm > SDNRaccept  SDNR6cm > 0.9 × SDNRaccept Note: for magnification mode this last requirement is relaxed to  SDNR6 cm > 0.65 × SDNRaccept	Assess the most commonly used AEC modes for contact and magnification geometry Use clinical AEC settings (kVp, target/filter and mode including density setting) Use a designated test cassette and imaging plate that is in routine clinical use Use a consistent AEC detector position where this is manually selected Use 0.2 mm Al foil as contrast test tool and measure SDNR for 2, 4 and 6 cm PMMA (also see section on glandular dose). Note: measurements are to be undertaken on “for processing” (unprocessed) image Consistent time delay between plate irradiation and readout Record exposure indicator for each PMMA thickness Measure film density for each image if applicable	Establish baseline values. Assess all available AEC modes for contact and magnification geometries. Assess both 18 × 24 cm2 and 24 × 30 cm2 Buckys
Density control	The density control must be capable of changing the mAs from the value used normally by −25 to +50%	Assess change in mAs for at least two density settings either side of the usual clinical setting using 4 cm of PMMA	Assess change in mAs across full range of density settings
Back-up timer/security cut-out	Security cut-out mechanisms shall be present and terminate the exposure within 50 ms or within 5 mAs, otherwise the back-up timer should terminate the exposure at ≤500 mAs and must terminate the exposure at ≤800 mAs	Use lead sheet or other heavily attenuating material to intercept beam and confirm that the back-up timer/security cut-out functions within specified limits	Confirm that the back-up timer/security cut-out functions within specified limits
Image uniformity and artefact	Max. deviation of mean pixel value <±10% of mean pixel value for central ROI Max. deviation in SNR of central ROI as a function of time is ±10%. No major inhomogeneities on the images	Assess for 40 mm PMMA covering complete CR plate Use three ROIs each of ~100 mm2 placed on a line parallel to and approximately 20 mm from chest wall	Assess also at 20 and 60 mm
Uniformity of cassette/image plate response	Maximum mAs variation <±5% between all plates of one size Maximum mAs variation <±20% between plates of different sizes See Appendix 6 for manufacturer dependent allowed tolerances on the exposure indicator	Assess for 40 mm PMMA covering complete CR plate	As per routine testing
Image quality evaluation	The ability to clearly visualise 5 fibres, 3.5 speck groups (4 is desirable) and 4 masses in an image of an ACR Accreditation phantom or the ability to clearly visualise 4 fibres, 3 speck groups and 3 masses in an image of the ACR DM phantom Additionally, with the ACR DM phantom the SDNR with contrast object ≥2.0	Use typical clinical settings Measure MPVs and SDs in relevant ROIs of ACR DM phantom so that SDNR may be calculated	As per routine testing
Ghost image evaluation	“Ghost image” factor <2.0	Assess using 40 mm PMMA (see “Ghost image evaluation” section for testing guidelines)	As per routine testing
System linearity and noise analysis	Compare to baseline results and note requirement for linearity (see text and Appendix 6) has R2 > 0.99 Noise analysis remains optional	Use standard test block (e.g. 4 cm PMMA) at typical clinical beam settings Use the same cassette/image plate for all exposures Record exposure indicator Plot exposure indicator as a function of ESAK (see Appendix 6)	Baseline measurements at clinical kVp, also at max and min clinical kVps for all target filter combinations
Generator performance
kVp, reproducibility	COV ≤ 0.02 for a minimum of three exposures	Assess kVp reproducibility at a clinical kVp value	As per routine testing
kVp accuracy	Measured kVp shall be within ±5% of the specified value over the clinically relevant range	Assess kVp accuracy over the clinically relevant range in, at most, 2 kVp increments	Assess kVp accuracy over clinically relevant range in 1 kVp increments
Beam quality	$\left[ {\left( {{\text{kVp}}/100} \right)+0.03} \right] \leq {\text{HVL}}<\left[ {\left( {{\text{kVp}}/100} \right)+{\text{C}}} \right]$ $\begin{aligned} {\text{where C }}&=&0.{\text{12 mm Al for Mo}}/{\text{Mo}}\\ &=&0.{\text{19 mm Al for Mo}}/{\text{Rh}}\\&=&0.{\text{22 mm Al for Rh}}/{\text{Rh}}\\&=&0.{\text{23 mm Al for Rh}}/{\text{Ag}}\\&=&0.{\text{3}}0{\text{ mm Al for W}}/{\text{Rh}} \\ &=&0.{\text{32 mm Al for W}}/{\text{Ag}} \\ &=&0.{\text{25 mm Al for W}}/{\text{Al}}. \\ \end{aligned}$	Measure the HVL required for Mean Glandular Dose calculations and for establishing compliance with DRLs (see “Mean glandular dose” for details)	As per routine tests plus measure HVL at 28 kVp for all target/filter combinations, with the compression paddle removed if unit used for biopsy purposes with open paddle
Mean glandular dose	≤2.0 mGy for a 4.2 cm 50% adipose, 50% glandular breast (i.e. ACR Accreditation phantom or ACR DM phantom) <1 mGy for 2.0 cm PMMA (2.3 cm 50% adipose, 50% glandular breast) <4.5 mGy for 6.0 cm PMMA, (6.5 cm 50% adipose, 50% glandular breast)	Assess for an AEC controlled exposure using typical clinical settings using ACR Accreditation phantom (or ACR DM phantom) and also for 20 mm and 60 mm PMMA Additional dose measurements may be necessary to confirm compliance with DRLs (see “Mean glandular dose”)	As per routine tests
Exposure time	For all clinically relevant SID settings the maximum exposure time when irradiating 6 cm PMMA must be less than 3.5 and 2 s for fine and broad focus, respectively	Assess for both contact and magnification modes Use 6 cm of PMMA Use clinically relevant technique factors for this PMMA thickness consistent with SDNR and MGD measurements Record mAs and infer the exposure time from tube rating or measure directly using a manual exposure matched to mAs needed for AEC initiated exposure	As per routine tests
Viewbox luminance and room illuminance (hardcopy only)	Viewing area illuminance ≤50 lx Viewbox luminance ≥3000 nit	Assess viewing conditions for all viewers	As per routine tests
Monitor luminance and viewing conditions	Image interpretation must not be done on a monitor of less than 4.2 mega pixels Luminance ratio approximately 350:1 Maximum luminance >450 cd/m2 and maximum luminance of paired monitors matched to ≤5% Minimum luminance preferably not less than 1 cd/m2 Ambient light <20 lx In PACS situations images must be stored with lossless compression	Measure luminance ratio under clinical lighting conditions Confirm luminance uniformity Confirm no cross-talk and pixel defects	As per routine testing with the additional requirement of checking GSDF As per routine testing. Monitor or workstation may have comprehensive QC program which needs to be validated
Monitor performance	No smearing artefact, ramps without terracing Lines straight, boxes square, active display centred, borders complete Squares of different shades from black to white must be distinct and small squares in corners of each clearly discernible Free from artefact The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Test patterns to be displayed at full resolution Test under clinical lighting conditions Use TG18-QC test pattern	As per routine testing. Monitor or workstation may have comprehensive QA program
Printer (hardcopy only)	B + F = baseline ± 0.03 and ≤0.25 OD Dmax = baseline ± 0.10 and ≥3.4 OD The number of letters visible in the phrase “Quality Control” for the dark, mid-gray and light renditions must be ≥11	Print TG18-QC test pattern as per weekly printer QC test	As per routine tests

Appendix 6: Summary of recommendations for facility QC for biopsy units

Procedure	Minimum frequency	Fully integrated digital/biopsy unit	Digital mammography with add on image system	Stand alone biopsy system
Viewing conditions	Weekly	Previously covereda	Previously covered	See “Viewing conditions” section
Monitor QC	Weekly	Previously covered	ssab	See “Monitor QC” section
Monitor cleaning	Weekly	Previously covered	ssa	See “Monitor/viewbox cleaning” section
Image quality evaluation	Weekly	ssa	ssa	See “Image quality evaluation” section Note may use ACR ‘mini’ digital stereotactic phantom—see text
Printer QC (if applicable)	Weekly	Previously covered	ssa	See “Printer QC” section
Mechanical inspection	Monthly	ssa	ssa	See “Mechanical inspection and breast thickness indication” section. Note additionally image receptor and compression plate/ biopsy window must be free of wobble; Vernier drive and needle guide rigid and wobble free, localisation system zeros and biopsy device properly immobilised—see text
Repeat analysis	Quarterly	ssa	ssa	See “Repeat analysis” sectin
Image receptor homogeneity	Quarterlyc	Previously covered	ssa	See “Image receptor homogeneity” section. Note; procedure should be modified as seen in text
AEC calibration test	Quarterly	ssa	ssa	See “AEC calibration test” section. Procedure may vary for different types of units—see text
Compression	6 monthly	ssa	ssa	See “Compression” section
Test equipment quality control Densitometer calibration check	6 monthly	Previously covered	Previously covered	See “Test equipment calibration” section
Maintenance and fault logging	As required	ssa	ssa	See “Maintenance and fault logging” section
Infection control of breast imaging equipment	Before each examination	Previously covered	Previously covered	See “Infection control of breast imaging equipment” section
Stereotactic Accuracy confirmation	Prior to first use on day of procedures	ssa	ssa	Localisation within ±1 mm. Procedure as per manufacturer’s recommendations; Checklist/logbook entry showing  Date test performed  Test results  Person performing test

1. ^aTest previously completed as part of mammography tests
2. ^b Ssa see stand alone biopsy units
3. ^cOr more frequently if recommended by the manufacturer

Appendix 7: Summary of recommendations for medical physics testing for biopsy units

Procedure	Frequency	Fully integrated digital/biopsy unit	Digital mammography with add on image system	Stand alone biopsy system
Focal spot	Acceptance	Previously covereda	Previously covered	See “Focal spot size” section
Leakage radiation	Acceptance	Previously covered	Previously covered	See “Leakage radiation” section
MTF	Acceptance	Previously covered	ssab	See “Modulation transfer function” section
Spatial linearity and geometric distortion	Acceptance	Previously covered	ssa	See “Spatial linearity and geometric distortion” section
Distance calliper accuracy	Acceptance	Previously covered	ssa	See “Distance calliper accuracy” section
Mammography unit assembly evaluation	Annual	ssa	ssa	See “Mammography unit assembly evaluation” section. Note additionally ensure X-ray tube angular locations positively locked; image receptor and compression plate/ biopsy window free of wobble; Vernier drive and needle guide rigid and wobble free, localisation system zeros; biopsy device properly immobilised and AEC chart displayed—see text
Collimation assessment	Annual	ssa	ssa	FOV defined by biopsy window and is aligned centrally with digital image receptor, with tolerances of ±5 mm—see text
System resolution	Annual	Previously covered	ssa	See “System resolution/MTF” section
AEC/SDNR	Annual	Previously covered	ssa	See “Automatic exposure control system performance assessment/signal difference to noise ratio” section. Note: technique charts should be consulted for correct factor settings. Minimum PMMA thickness of 2 cm used for SDNR see text
Image uniformity and artefact evaluation	Annual	Previously covered	ssa	See “Image uniformity and artefact evaluation” section .Note—ROIs to be in corners of image, 10 mm from edge
Image quality evaluation	Annual	Previously covered	ssa	See “Image quality evaluation” section .Note may use ACR ‘mini’ digital stereotactic phantom—see text for revised scoring
Ghost image evaluation	Annual	Previously covered	ssa	See “Ghost image evaluation” section
System linearity and noise analysis	Annual	Previously covered	Previously covered	See “System linearity and noise analysis” section
kVp performance	Annual	Previously covered	Previously covered	See “Generator performance” section
HVL	Annual	Previously covered	Previously covered	See “Beam quality or half value layer” section
Mean glandular dose	Annual	ssa	ssa	See “Mean glandular dose” section—note; see technique chart for factors used in dose calculations
Exposure time	Annual	Previously covered	Previously covered	See “Exposure time” section
Viewbox and room luminance	Annual	ssa	ssa	See “Viewbox luminance and room illuminance (hardcopy only)” section
Monitor performance	Annual	Previously covered	ssa	See “Monitor luminance and viewing conditions” section
Printer (hardcopy)	Annual	Previously covered	ssa	See “Printer (hardcopy)” section
Localisation accuracy test	Annual	ssa	ssa

1. ^aTest previously completed as part of mammography tests
2. ^bSee stand alone biopsy units

Appendix 8: Summary of criteria in terms of CR exposure indicators

A number of companies currently manufacture CR units for use in mammography and they have developed unique exposure indicators. Reviews of these indicators, with a comparison between the different manufacturers, have been reported in the literature [9, 87]. The table below can be used to indicate the test criteria that should be applied in terms of the current CR exposure indicators.

Test	Test criteria	Tolerance in terms of CR exposure indicator
		Fuji, Philips and Konica	Kodak (Carestream)	Agfaa
Image quality evaluation	Air kerma (dose) to the plate must not change by greater than ±10%	±10% in S# of baseline	±40 units in EI of baseline	±5% in SAL or ±430 in SAL log or ±580 in PVL log16 of baseline
AEC calibration test	Air kerma (dose) to the plate for each of the three thicknesses of PMMA be within ±10% of the baseline value for each thickness	±10% in S# of baseline for each thickness	±40 units in EI of baseline for each thickness	±5% in the SAL, or ±430 in SAL log or ±580 in PVI log16 of baseline for each thickness
Cassette image plate condition and interplate sensitivity variation (also “Image uniformity and artefact evaluation”	Air kerma (dose) to individual plate must differ from mean for that size by less than ±5% Difference in mean air kerma (dose) to plates of different sizes <20%	S# for individual plates must be within ±5% of mean for same size S# difference for two different plate sizes <20%	EI for individual plate must be within ±20 units of mean for same size EI difference for two different plate sizes <100 units	SAL for individual plates must be within ±2.5% or SAL log must be within ±220 or PVI log16 must be within ±290 of mean for same size SAL difference <10% or SAL log difference <1000, or PVI log16 <1300 for two different plate sizes
System linearity and noise analysis	R2 value of appropriate plot of exposure indicator versus ESAK must be >0.99	Plot S# versus reciprocal of ESAK	Plot EI versus log (ESAK)	Plot SAL versus SQRT(ESAK) or SAL log versus log(ESAK) or PVI log 16 versus log(ESAK)
Exposure indicator calibration and image fading	Under specified conditions (see Table 4) Exposure Indicator must meet criteria outlined in columns to right	S# = 120 ± 20	EI = 2300 ± 100	SAL = 1130 ± 100 SAL log = 21,600 ± 1000 PVI log 16 = 41,100 ± 1300

1. ^aAgfa have indicated that the preferred exposure indices to use with mammography plates are SAL log (sometimes called PVI log15) or PVI log16, the actual choice being dictated by software version and plate type. SAL may be used in some old software versions and LgM should not be used

Appendix 9: Summary of recommendations for facility QC for DBT units

Procedure	Recommended control-limits/requirements	Minimum frequency	Key procedure elements	Recommendations for record keepinga
Full field artefact evaluation	mAs = baseline ± 10% There must be no evidence of  Clinically significant 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 breast support structures  Bright or dark pixels  Dust artefacts mimicking calcifications  Significant stitching or registration artefacts	Daily	Expose a uniform thickness of PMMA using clinically relevant technique factors under AEC Central projection image and central reconstructed image should be inspected closely for potential artefacts View images on acquisition monitor using zoom and roam to check for possible detector faults. The magnification should be sufficient to achieve at least 1:1 resolution	Records showing  Date test was performed  Person performing test  Test results  kVp, target/filter and mAs
Image quality evaluation	mAs = baseline ± 10% The ability to clearly visualise 4 fibres, 3 speck groups and 3 masses in an image of an ACR accreditation phantom OR The ability to clearly visualise 2 fibres, 1 speck group and 2 masses in an image of an ACR DM phantom The position of the reconstructed slice used for scoring the phantom must not change by more than ±1 mm	Weekly	Obtaining the phantom image  Use an ACR accreditation phantom or the new ACR DM phantom  Light contact between the compression paddle and the phantom surface  Consistent positioning of the phantom  Consistent selection of clinically relevant kVp and target/filter combinations Evaluating the phantom image  Scroll through the reconstructed images until the slice displaying the speck details most clearly is reached. Use zoom and modest adjustment of window/level functions to score fibres and specks  Use of consistent viewing conditions that reflect those used to read actual mammograms  Image quality scoring by the same person, if possible  Use of a control chart to record results	Record radiographic settings (kVp, target/filter combination, mAs values) and image quality scores, position of slice used for scoring Control chart showing  Plots of mAs, image quality scores, slice position  ≥25 results  Clearly marked control limits  Baseline values  Remarks e.g. corrective action Phantom images identifying  Date  The X-ray system  The technique factors
Detector calibration—flat field test	Pass or fail	Weekly or as per manufacturer’s requirements	Follow manufacturer’s specific procedure	Checklist/logbook entry showing  Date performed  Person performing task
AEC calibration test	mAs = baseline ± 10% for same target/filter combination for each thickness of PMMA	Quarterly	Use PMMA thicknesses of 2, 4 and 6 cm covering complete image receptor Use clinical AEC settings (kVp, target/filter and mode)	Records showing  Date test was performed  Person performing test  X-ray system identification  kVp, target/filter, AEC mode and mAs
Compressed breast thickness	For units using special Bucky’s for DBT indicated breast thickness accurate to ±5 mm	Monthly	Confirm accuracy of thickness indication under conditions as indicated by the manufacturer	Checklist/logbook entry showing  Date performed  Inspection results  Person performing task

1. ^aAll written/electronic QC records should be retained for 1 year unless otherwise indicated by local Regulatory requirements. Images used to assess image quality with the ACR Accreditation or ACR DM phantom should be retained for a minimum of 1 month

Appendix 10: Summary of recommendations for medical physics testing for DBT units

Procedure	Performance requirements/guidelines	Routine testing guidelines	Acceptance and additional tests
Collimation and alignment assessment
X-ray field/Image receptor alignment	The X-ray field shall irradiate the image receptor fully but not extend beyond breast support on the chest wall edge of the image receptor by more than 2 mm or beyond the Bucky support on the other three margins	Assess alignment for each target/geometry combination Note: When a special Bucky is used for DBT that may also be used to acquire normal 2D projection images the requirements outlined in “Collimation and alignment assessment” section must be met	As per routine tests
Paddle/Image alignment	The chest wall edge of the compression paddle shall be aligned just beyond the chest wall edge of the image receptor such that it does not appear in the image. In addition, the compression paddle shall not extend beyond the chest wall edge of the image by more than 1% of the SID		As per routine tests
Compressed breast thickness	For units using special Bucky’s for DBT, indicated breast thickness accurate to ±5 mm	Confirm accuracy of thickness indication under conditions as indicated by the manufacturer	As per routine tests
Missed tissue	The missing tissue must be ≤5 mm Full thickness of breast tissue must be imaged	Confirm in DBT mode of acquisition even if previously confirmed in projection imaging mode	As per routine tests
Distance calliper accuracy	Measured dimensions of object within reconstructed image plane must be within 2% of true dimensions	Use the ACR accreditation (or ACR DM) phantom as per image quality test below Select the reconstructed slice which best displays the speck details for image scoring and perform in plane distance measurements The outside location of the detail insert is useful for this purpose Compare with actual dimensions	Confirm accuracy of measurements at reporting workstation if possible and also confirm accuracy in more than one slice
AEC system performance assessment	mAs = baseline ± 10% for same target/filter combination for each thickness of PMMA	Use PMMA thicknesses of 2, 4 and 6 cm covering complete image receptor Use clinical AEC settings (kVp, target/filter and mode)	As per routine tests
Image uniformity and artefact evaluation	mAs = baseline ± 10% There must be no evidence of  Clinically significant 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 breast support structures  Bright or dark pixels  Dust artefacts mimicking calcifications  Significant stitching or registration artefacts	Assess for 40 mm PMMA covering complete detector using clinically relevant technique factors under AEC Central projection image and central reconstructed image should be inspected closely for potential artefacts View images on acquisition monitor using zoom and roam to check for possible detector faults. The magnification should be sufficient to achieve at least 1:1 resolution	Assess all projection and reconstructed images in the acquisition for all available target/filter combinations
Image quality evaluation	mAs = baseline ± 10% for same target/filter combination Slice used for scoring should be 37 ± 2 mm (ACR accreditation phantom) or 34 ± 2 mm (ACR DM phantom) above breast support and must not change by more than ±1 mm from previous measurement The ability to clearly visualise 4 fibres, 3 speck groups and 3 masses in an image of an ACR accreditation phantom or the ability to clearly visualise 2 fibres, 1 speck group and 2 masses in an image of the ACR DM phantom	Use typical clinical acquisition parameters selected under AEC. Note that the acquisition may be combined with FFDM mode using “Combo mode” Select the reconstructed slice which best displays the speck details for image scoring. This is typically 37 ± 2 mm or 34 ± 2 mm above the breast support with the ACR accreditation phantom and ACR DM phantom, respectively	As per routine testing
Beam quality	$\left( {{\text{kVp}}/100} \right)+0.03 \leq {\text{HVL}}<\left( {{\text{kVp}}/100} \right)+{\text{C}}$ $\begin{aligned} {\text{where C }}&=&0.{\text{12 mm Al for Mo}}/{\text{Mo}}\\ &=&0.{\text{19 mm Al for Mo}}/{\text{Rh}}\\&=&0.{\text{22 mm Al for Rh}}/{\text{Rh}}\\&=&0.{\text{23 mm Al for Rh}}/{\text{Ag}}\\&=&0.{\text{3}}0{\text{ mm Al for W}}/{\text{Rh}} \\ &=&0.{\text{32 mm Al for W}}/{\text{Ag}} \\ &=&0.{\text{31 mm Al for W}}/{\text{Al}}. \\ \end{aligned}$	Measure the HVL required for mean glandular dose calculations	As per routine tests
Mean glandular dose	≤2.0 mGy for a 4.2 cm 50% adipose, 50% glandular breast (i.e. ACR accreditation phantom or ACR DM phantom) <1.2 mGy for 2.0 cm PMMA (2.3 cm 50% adipose, 50% glandular breast) <4.5 mGy for 6.0 cm PMMA, (6.5 cm 50% adipose, 50% glandular breast) Displayed MGD values must agree with calculated values within ±25%	Assess for an AEC controlled exposure using typical clinical settings using ACR phantom (or ACR DM phantom) and also for 20 mm and 60 mm PMMA Confirm displayed and calculated MGDs agree to ± 25%	As per routine tests