OVERVIEW
QC for FFDM: What You Must
Do and What Really Matters
Eric A. Berns, Ph.D.
Northwestern University Medical School
Lynn Sage Comprehensive Breast Center
Chicago, IL
Important PrePre-Survey Events
• Obtain proper training & CE credits (8 hours)
– HandsHands-on training on actual unit:
• Mechanics
• Software
• Artifacts
• Learn vendor specific tests and tricks
• Important prepre-survey events
• Manufacturer’
Manufacturer’s tests and equipment
• Performing the survey
• Summary of important points and what really
matters
Important PrePre-Survey Events
• ACR Accreditation - www.acr.org
– GE Senographe 2000D
– Fischer Senoscan
– Lorad Selenia
• FDA Accreditation - www.fda.gov/cdrh/mammography/
www.fda.gov/cdrh/mammography/
– GE Senographe DS
– Siemens, Sectra,
Sectra, Fuji, etc., when approved
1
Important PrePre-Survey Events
Important PrePre-Survey Events
•ACR Accreditation – Equipment Evaluation Forms
• ACR Accreditation
– Before clinical use
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
Full-Field Digital – GE Medical Systems
Site
• Medical Physicist equipment evaluation and indicate it passes
Report Date
Survey Date
Model
GE Medical Systems
X-Ray Unit Manufacturer
Date of Installation
Medical Physicist
Fischer
X-Ray Unit Manufacturer
Date of Installation
Medical Physicist
Medical Physicist's QC Tests
? 2277390-100 Rev 1, 2000
? 2371472-100 Rev 0, 2003
• New unit application
PASS/FAIL/NA
Flat Field (as described in radiologic technologist tests)
Image Quality (Phantom) (as described in radiologic technologist tests)
CNR
(required)
Change in CNR =0.2 (NA for Equipment Evaluations)
Fibers Specks Masses
Phantom IQ Test on AWS
Phantom IQ Test on RWS – Left*
(* NA for units with Seno
Phantom IQ Test on RWS – Right*
Advantage workstations)
Phantom IQ Test on Printer
3. MTF Measurement (as described in radiologic technologist tests)
4. AOP Mode and Signal-to-Noise (SNR) (as described in radiologic technologist tests)
5. Collimation Assessment
Deviation between X-ray field and light field is less than 2% of SID
X-ray field does not extend beyond any side of the IR by more than 2% of SID
Chest wall edge of compression paddle doesn't extend beyond IR by more than 1% of SID
6. Evaluation of Focal Spot Performance
Measured performance within acceptable limits for large focal spot
Measured performance within acceptable limits for small focal spot
7. Breast Entrance Exposure, Average Glandular Dose and Reproducibility
Average glandular dose for average breast is below 3 mGy (300 mrad)
Average glandular dose to a 4.2-cm-thick breast on your unit is
mrad
Exposure reproducibility (CV) for R and mAs must be less than 0.05
8. Artifact Evaluation and Flat Field Uniformity
9. Viewing Condition Check and Setting
10. Monitor Calibration
11. Image Quality – SMPTE Pattern
12. Analysis of Review Work Station (RWS) or Seno Advantage Screen Uniformity
• However, no reimbursement without FDA receiving ACR app.
• Approximately 3 days for accreditation approval from ACR
(required for Equipment Evaluations and as necessary)
13. kVp Accuracy and Reproducibility
Measured average kVp within ±5% of indicated kVp
kVp coefficient of variation =0.02
14. Beam Quality Assessment (Half-Value Layer Measurement)
15. Radiation Output
Radiation output is =800 mR/s
16. Mammographic Unit Assembly Evaluation
Important PrePre-Survey Events
(not required for Equipment Evaluations)
Frequency
(Fischer, continued)
(Lorad, continued)
Evaluation of Site's Technologist QC Program
Evaluation of Site's Technologist QC Program
PASS/FAIL
MQSA Regs
1. Monitor cleaning
Daily
Daily
2. Viewing conditions for Review Work Station (RWS) or Seno Advantage
Daily, if applicable
3. Darkroom cleanliness
4. Processor QC
Daily, if applicable
5. Mobile unit quality control
After every move, if applicable
6. Flat field
Weekly
7. Image quality (phantom)
Weekly
Weekly
8. Viewbox and viewing conditions
Monthly
9. MTF measurement
10. AOP mode and signal-to-noise (SNR)
Monthly
11. Visual checklist
Monthly
12. Monitor calibration check
Monthly
Quarterly
13. Repeat analysis - performed, records maintained, radiologist reviewed
Quarterly, if applicable
14. Analysis of fixer retention
15. Compression force (pressure)
Semi-annually
16. Darkroom fog
Semi-annually, if applicable
17. Laser film printer QC (according to printer manufacturer recommendations)
Medical Physicist's Recommendations for Quality Improvement
(not required for Equipment Evaluations )
(not required for Equipment Evaluations)
PASS/FAIL
Frequency
1.
2.
3.
4.
5.
6.
7.
8.
9.
Laser Imager Quality Test (if applicable)
Image Display Monitor Test (Review Station)
Phantom Image Acquisition Test
Phantom Image Quality Test
Detector Calibration and Flat Field Test
System Resolution (Detector Alignment/Scan Speed Uniformity)
System Operation
Reject/Repeat Analysis
Compression Force Test
MQSA Regs
Daily
Daily
Weekly
Weekly
Weekly
Monthly
Monthly
Quarterly
Semi-annually
Medical Physicist's Recommendations for Quality Improvement
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Darkroom Cleanliness (if applicable)
Processor Quality Control (if applicable)
Laser Printer Quality Control
Viewboxes and Viewing Conditions
Softcopy Workstation QC
Artifact Evaluation
Signal-To-Noise and Contrast-To-Noise Measurements
Phantom Image
Detector Flat-Field Calibration
Compression Thickness Indicator
Visual Checklist
Analysis of Fixer Retention in Film (if applicable)
Repeat Analysis
Darkroom Fog (if applicable)
Compression
mR/s
Fischer QC manual version
at facility (check one) :
? P-55943-OM Issue 1, Rev. 000 (2002)
? P-55943-OM Issue 1, Rev. 3 (July 2003)
? P-55943-OM Issue 1, Rev. 1 (April 2003)
? P-55943-OM Issue 1, Rev. 7 (March 2004)
PASS/FAIL/NA
X-Ray Field Size Alignment and Chest Wall Missed Tissue Checks
X-ray field size aligns with field area indicated on breast support within =2% SID
Chest wall edge of the digital image to the ruler reference mark of breast support is =8.5 mm
Compression Paddle Alignment
Compression paddle edges not visible within field of view
Chest wall edge of compression paddle extends beyond image by =1% SID
3. kVp Accuracy Test
Measured average kVp within ±5% of indicated kVp
kVp coefficient of variation =0.02
4. Linearity, Reproducibility and Accuracy
Linearity <0.08
Reproducibility <0.035 for each technique
5. Half-Value Layer and Output
HVL =0.33 mm Al at 30 kVp, 100 mA
6. Dosimetry – Average Glandular Dose and Output
Average glandular dose for average breast is below 3 mGy (300 mrad)
mrad
Average glandular dose to a 4.2-cm-thick breast on your unit is
7. Phantom Image Acquisition Test
No obvious artifacts
(ADU values required for both QC and Equipment Evaluations; P/F results are NA for Equipment Evaluations)
within +50/-0 ADU counts of baseline
Background StDev
within ±100 ADU counts of baseline
Background mean
within ±300 ADU counts of baseline
ADU level difference
8. Image Quality
Fibers Specks Masses
Largest 4 fibers, 3 speck groups and 3 masses
Phantom IQ Test on Review Work Station
9. System Resolution/Scan Speed Uniformity
Standard imaging mode @ 7 lp/mm: =5 transitions, =10% modulation
High resolution imaging mode @ 11.1 lp/mm: =5 transitions, =5% modulation
10. Flat Field Test
Flat field test
Deviations between corner ROIs and center within ±20%
11. Geometric Distortion and Resolution Uniformity
12. Automatic Decompression Control
13. System Artifacts
14. Image Display Monitor(s) Check (review daily log)
15. Image Viewing Room Illuminance Test (=50 lux)
1.
2.
Lorad
X-Ray Unit Manufacturer
Date of Installation
Medical Physicist
Report Date
Survey Date
Model
Room ID
Signature
Selenia
Medical Physicist's QC Tests
? June 11, 2002
? 9-500-0285, Rev. 004 (2004)
Lorad QC manual version at facility (check one) :
? 9-500-0285, Rev. 003 (2003)
? 9-500-0285, Rev. 002 (2003)
PASS/FAIL
1.
Mammographic Unit Assembly Evaluation
Autodecompression can be overidden to maintain compression (& status maintained)
Manual emergency compression release can be activated in the event of power failure
Collimation Assessment
Deviation between X-ray field and light field is less than 2% of SID
X-ray field does not extend beyond any side of the IR by more than 2% of SID
Chest wall edge of compression paddle doesn't extend beyond IR by more than 1% of SID
Artifact Evaluation
Artifacts were not apparent or not significant
4. kVp Accuracy and Reproducibility
Measured average kVp within ±5% of indicated kVp
kVp coefficient of variation =0.02
5. Beam Quality Assessment - HVL Measurement
Half-value layer is within acceptable lower and upper limits at all kVp values tested
6. Evaluation of System Resolution
Measured performance within acceptable limits
7. Breast Entrance Exposure and Average Glandular Dose
Average glandular dose for average breast is below 3 mGy (300 mrad)
Average glandular dose to a 4.2-cm-thick breast on your unit is
mrad
8. Radiation Output Rate
Radiation output rate is greater than 800 mR/sec
mR/sec
9. Phantom Image Quality Evaluation
5 largest fibers, 4 largest speck groups and 4 largest masses are visible
Fibers
Background Optical Density
Phantom image quality scores:
Specks
Disk Optical Density
Disk Contrast
Masses
Hard copy background density must be =1.20 (with operating level =1.40)
Hard copy density difference over acrylic disk must be =0.35 (with operating level =0.40)
10. Signal-To-Noise Ratio and Contrast-To-Noise Ratio Measurement
Signal-To-Noise Ratio should be equal or greater to 40
SNR
Contrast-To-Noise Ratio should not vary by more than ±15%
CNR
11. Viewbox Luminance and Room Illuminance
Mammographic viewbox is capable of a luminance of at least 3000 cd/sq m (nit)
Room illuminance (viewbox surface as seen by observer) is 50 lux or less
Room illuminance (monitor surface) is 20 lux or less
12. Softcopy Workstation QC
White level performance
Black level performance
Quality level performance
Uniformity performance
2.
3.
• FDA Forms
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
(GE Medical Systems, continued)
Evaluation of Site's Technologist QC Program
Site
SenoScan
Important PrePre-Survey Events
•ACR Accreditation – Equipment Evaluation Forms
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
? 2277390-100 Rev 3, 2001
? 2277390-100 Rev 3 + Addendum 2354312-100, 2003
? Mobile 2371698-100 Rev 0, 2003 ? Seno Advantage 2391082-100 Rev 1, 2003
1.
2.
– Not required to wait for ACR response
Report Date
Survey Date
Model
Room ID
Signature
Medical Physicist's QC Tests
GE QC manual version(s) at facility (check all that apply):
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
Full-Field Digital – Lorad
MEDICAL PHYSICIST'S MAMMOGRAPHY QC TEST SUMMARY
Full-Field Digital - Fischer
Site
Senographe 2000D
Room ID
Signature
Frequency
Daily
Daily
Weekly*
Weekly
Weekly
Weekly
Weekly
Weekly
Bi-weekly
Bi-weekly
Monthly
Quarterly
Quarterly
Semi-annually
Semi-annually
PASS/FAIL
MQSA Regs
* Dry laser printer (daily if wet processor used)
Medical Physicist's Recommendations for Quality Improvement
– MQSA Certification Extension for GE Senographe DS
• Site must already be screenscreen-film certified by FDA
• Three things to submit to FDA:
– Application of qualifications
– Medical Physicist Survey
– Printed ACR Phantom
• Must have approval letter from FDA before clinical imaging
can be performed
2
Important PrePre-Survey Events
• Contact the site – things to confirm:
– Site is aware of ACR or FDA application process
– FFDM unit is operable
– Review workstation is operable
• Images can be transmitted
– Laser printer works, can print mammo images, and hooked up
to all RWS’
RWS’s
– Discuss QC issues
• Many QC failures result in stopping clinical imaging
– ACR Phantom – do they have one onon-site?
Important PrePre-Survey Events
• Gather forms
– Copy of ACR or FDA forms
– Physics test forms
• Ensure you have tests that are required by manufacturer
• Gather test tools
– Check required tests in manufacturer’
manufacturer’s manual
– Artifact test tool – 1 or 2 inches of acrylic
Important PrePre-Survey Events
• Contact Manufacturer’
Manufacturer’s Service Engineers
– Complete rere-calibration
– Can they be present?
• If not, how can they be contacted?
– Is the system working properly?
– Can the laser printer service engineer be present?
Performing the Survey
• Must perform manufacturer’
manufacturer’s tests
• Turn off auto push and/or auto print
– Remember to turn them back on
• Order of tests is important
• Use “Raw”
Raw” of “Processed”
Processed” images for testing
– Lead sheet
3
Performing the Survey
GE 2000D
http://www.gemedicalsystems.com/services/repl_parts/documentation.html
http://www.gemedicalsystems.com/services/repl_parts/documentation.html
Raw Image
Processed Image
GE 2000D - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– Mammography unit evaluation *
– Flat field uniformity
– Artifact evaluation *
– AOP Mode and SNR Check
– ACR Phantom and Contrastto-Noise Ratio (CNR) Check
Contrast-to– MTF measurement
– Collimation Assessment *
– Evaluation of Focal Spot *
GE Review Workstation
*30 Days to Repair
4
GE 2000D - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– Breast entrance exposure, average glandular dose, and
reproducibility
GE 2000D - Performing the Survey
• FlatFlat-field uniformity
– Test to ensure detector performance acceptable
• Measures detector uniformity (signal & noise)
• Measures bad pixels
– Beam quality (HVL) *
– kVp accuracy and reproducibility *
– System automatically calculates pass/fail
– Radiation output *
– Viewing conditions check and setting
– Monitor calibration *
– Image quality – SMPTE pattern
QAP
– Analysis of RWS screen uniformity *
*30 Days to Repair
GE 2000D - Performing the Survey
• Artifact evaluation
– 1 inch acrylic phantom
– Use clinical techniques
– Image at each target/filter
No grid or compression paddle
GE 2000D - Performing the Survey
• AOP Mode and SNR Check
– Variable thicknesses of acrylic - 2.5, 4, 6 cm
– Std, Auto
– Evaluate:
• Correct techniques?
• Adequate SNR?
Acrylic
Thickness
2.5 cm
TargetFilter
Mo-Mo
Selected
kVp
27 kVp
Selected
mAs
20-60
– Review images at window width ~ 400 to 450
4.0 cm
Mo-Rh
28 kVp
35-90
– Review artifact images on RWS, AWS, and printed film
6.0 cm
Rh-Rh
32 kVp
35-90
– Image at each magnification mode
2.5 cm Acrylic
4.0 cm Acrylic
Each “raw”
raw” image must have a measured SNR of at least 50
6.0 cm Acrylic
5
GE 2000D - Performing the Survey
GE 2000D - Performing the Survey
• ContrastContrast-toto-Noise Test (CNR)
• ACR Phantom Imaging
– To examine consistency of CNR ratio
– Manual technique & 3 auto
measured over time
– Manual technique
modes
– Use the raw image
– Score the processed image
– A control level is established
– Acquisition workstation
over 5 days
– + 20% of baseline
– Each monitor of the RWS
Background ROI
Mass ROI
CNR = (Mean
(Meanbackground - Meanmass)/SDbackground
– Laser imager
GE 2000D - Performing the Survey
• MTF Measurement
GE 2000D - Performing the Survey
• Collimation
– Mo/Mo
– Rh/Rh
MTF (%) = (Std. Dev.) x 222 / (Mean Dark ROI – Mean Light ROI)
MTF (%) @ 2 lp/mm
lp/mm > 58%
MTF (%) @ 4 lp/mm
lp/mm > 25%
6
GE 2000D - Performing the Survey
• Focal Spot Evaluation
GE 2000D - Performing the Survey
• Beam Quality (HVL)
•Remove Compression Paddle
•Mo/Mo
•Rh/Rh
•Large & Small Spots (1.5 mag)
mag)
Note the lead sheet
GE 2000D - Performing the Survey
• Average Glandular Dose
GE 2000D - Performing the Survey
• kVp Accuracy
– Measure using AOP system, or
– Measure entrance exposure and calculate AGD
7
GE 2000D - Performing the Survey
• Monitor Quality Control
– Image Quality - SMPTE Pattern
– Luminance Levels
– RWS Screen Uniformity
– Viewing Conditions Check
High Contrast Line Pair
Patterns
1 pixel/line
Vert Horiz
RWS SMPTE Patterns: 5% & 95% contrast boxes
10% contrast boxes
High contrast
contrast lineline-pair resolution
Monitor Calibration
Measure luminance at:
10
60
2 pixel/line Vert Horiz
120
180
255
4 pixels/line Vert Horiz
Compare to baseline – must be
within action limits
8
Analysis of Screen Uniformity
Viewing Conditions Check
14. Viewing Conditions Map
•Set monitor luminance to
completely white (255) via menu
• To ensure optimal viewing conditions
angles to search for defects in
glass or other nonnon-uniformities
• Configure room for optimal
viewing conditions
surface of monitors - < 20 lux
Monitor position:
Room Lights:
Desk Lights:
Other Viewboxes:
Other Alternators:
Doors:
Other:
Left
Monitor
Right
Monitor
Ambient room illuminance (lux):
Room Layout:
• Record data on form
• Post in reading room
• Update if there is a room
reconfiguration
Fischer
Date:
Unit:
Senographe 2000D
ACR Site # or MAP#:
Room Description:
• Measure ambient room light at
•Look at monitor at various
Site:
Room #:
Action Limit:
The ambient light levels must not exceed 50 lux. The measured value must
stable over one minute, with a tolerance of + 5 lux over the measuring time with
regards to the nominal lighting values. It the system fails the test, the problem
must be fixed before further reviews are performed.
Fischer Senoscan
Some slides courtesy of Fischer
Imaging and Idris Elbakri,
Elbakri, Ph.D.
9
Fischer Senoscan
Fischer Senoscan - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– X-ray field size and Chest wall missed tissue
– Compression paddle alignment
– kVp accuracy
– Linearity, reproducibility, and accuracy
– Beam Quality (HVL)
– Dosimetry – average glandular dose
– Phantom image acquisition
Fischer Senoscan - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– Image quality
– System resolution/scan speed uniformity
– Flat field test
– Geometric distortion and resolution uniformity
– Automatic decompression control
Fischer Senoscan - Performing the Survey
• Flat Field and Artifact Evaluation
–
–
–
–
–
–
–
Use 4 cm acrylic
Exposure resulting in 1000 ADU’
ADU’s
Display “raw”
raw” image
No Artifacts at WW > 800
Compute deviation between corner and center means
Automatic in new software release
Must be within + 20% of center ROI
– System artifacts
– Image display monitor(s) check – Tech Review
– Image viewing room illuminance
10
Fischer Senoscan - Performing the Survey
• Collimation
Fischer Senoscan - Performing the Survey
• Compression Paddle Alignment
Fischer Senoscan - Performing the Survey
• Collimation
Fischer Senoscan - Performing the Survey
• Collimation
– Error between fieldfield-size markers and image receptor
must be less than 2% of SID
– Missed chest wall tissue less than 8.5 mm
– Compression paddle: distance between image receptor
at chest wall and inside of edge of paddle must
be < 8.5 mm
11
Fischer Senoscan - Performing the Survey
• kVp Accuracy
Fischer Senoscan - Performing the Survey
• Phantom Image Acquisition Test and Image Quality
– Select Techniques to give 1000 ADU
– Performed invasively through special BNC connectors
– Compare background mean, StdDev,
StdDev, and ADU level difference
to baseline values
– NonNon-invasive method under evaluation
– Score : 4 fibers, 3 speck groups, and 3 masses
Fischer Senoscan - Performing the Survey
• Mean Glandular Dose
– 30 kVp
– mA between 150 to 190 to result in 1000 ADU
– mRad/R
mRad/R conversion table provided in manual
Fischer Senoscan - Performing the Survey
• HalfHalf-Value Layer
– Ion chamber affixed 5 cm to the right of center and close to chest
chest wall
– Position paddle 4.2 cm above chest wall
– kVp = 30 mA = 100
– Criterion: HVL > 0.33 mm
– Watch collimation!
12
Fischer Senoscan - Performing the Survey
Fischer Senoscan - Performing the Survey
• Exposure Linearity and Reproducibility
• System Resolution/Scan Speed Uniformity
– Procedure:
– Purpose: Ensure correct detector/beam alignment and constant
• Center ion chamber on breast support near chest
scanning speed
wall with paddle in the beam
• Measure exposure at specified techniques
– Equipment: line pair phantom
• Calculate:
– Perform imaging with scan lines parallel and perpendicular to
•
Output = Average exposure reading/mA
reading/mA
•
Linearity = (A(A-B)/(A+B) < 0.08
•
Reproducibility = StdDev/Avg < 0.035 for each technique
scan direction
– Normal and high resolution modes
Fischer Senoscan - Performing the Survey
Normal Mode:
7 lp/mm:
lp/mm: > 10%
modulation
Fischer Senoscan - Performing the Survey
•Geometric Distortion
–Use a 40 mesh Cu screen
HighHigh-Res Mode
–Acquire image at low technique
11.1 lp/mm:
lp/mm: > 5%
modulation
–Visually inspect for
distortions or blurring
13
Fischer Senoscan - Performing the Survey
• Viewing Room Luminance
– Need a photometer
Lorad
Selenia
– Monitors off
– Measure luminance at the monitor screen and 50 cm away
– Illuminance must be < 50 lux
Lorad Selenia
Some slides courtesy of Lorad and
Nikolaos A. Gkanatsios,
Gkanatsios, Ph.D.
Lorad Selenia
14
Lorad Selenia - Performing the Survey
Lorad Selenia - Performing the Survey
• Manufacturer’
Manufacturer’s tests
• Manufacturer’
Manufacturer’s tests
– Unit assembly evaluation *
– kVp accuracy and reproducibility *
– Artifact evaluation *
– Beam quality—
quality— HVL *
– Phantom image quality
– Breast Entrance exposure and average glandular dose
– Evaluation of system resolution
– Radiation output rate *
– SignalSignal-toto-Noise and ContrastContrast-toto-Noise Measurements
– Viewbox luminance and room illuminance *
– Collimation assessment *
– Softcopy Workstation QC
*30 Days to Repair
*30 Days to Repair
Lorad Selenia - Performing the Survey
• Collimation Assessment
– Use coin techniques as described in ACR Manual
• Test 24x29 cm detector mode
• Test 18x24 cm detector mode
– X-Ray field to light field coincidence
– X-Ray field to image receptor alignment
Lorad Selenia - Performing the Survey
• Artifact Evaluation
– 4 cm acrylic block
– Mo/Mo
– Mo/Rh
Mo/Rh
– Large & Small Spot
– Evaluate for artifacts at WW ~ 250
– Print films – check printer
– Compression Paddle to Image Receptor Alignment
15
Lorad Selenia - Performing the Survey
• kVp –
– Described in the 1999 ACR QC Manual
• HVL –
– Described in the 1999 ACR QC Manual
• Phantom Image Quality
Lorad Selenia - Performing the Survey
• Breast Entrance Exposure and Average Glandular Dose
– Cover detector for protection – lead sheet
– Technique set to clinically image average breast
• 28 kVp,
kVp, 65 mAs,
mAs, Mo/Mo
– Calculate dose - < 3.0 mGy
– 28 kVp,
kVp, 65 mAs,
mAs, Mo/Mo – Until AEC system implemented
– Print film – measure OD and Contrast
– Score on each SCW (Soft Copy Workstation)
• 5 fibers
• 4 speck groups
• 4 masses
Lorad Selenia - Performing the Survey
• Radiation Output Rate
– Cover detector for protection – lead sheet
– Technique set to clinically image average breast
• 28 kVp,
kVp, 320 mAs,
mAs, Mo/Mo
– Output Exposure - > 800 mR/sec
mR/sec
Lorad Selenia - Performing the Survey
• Evaluation of System Resolution
– 5-15 lp/mm
lp/mm Test Pattern
– 4-cm Attenuation Block
– Pattern at 4545-Degree Angle
to the Detector
• The system limiting spatial
resolution must be > 7 lp/mm
lp/mm
16
Lorad Selenia - Performing the Survey
• SNR and CNR Measurements
Lorad Selenia - Performing the Survey
• Viewbox Luminance and Room Illuminance
– Desribed in the 1999 ACR QC Manual
– SNR at least equal or greater than 40
• SNR = (Mean
(MeanBkgd-DCoffset)/SDBkgd DCoffset = 50
– Establish CNR during acceptance testing
• SoftCopy Workstation QC
– Use supplied photometer and run monitor QC software
• White level = 300 cd/m2 for Barco MG521 & 400 cd/m2 for Barco MG521M
– Warning level = + 3%
Tolerance level = + 6%
Recalibrate
Tolerance level = 1.0%
Recalibrate
• Black level = 0 cd/m2
• CNR = (Mean
(MeanBkgd-MeanDisk)/SDBkgd
– Warning level = 0.5%
• Quality Level Performance – checks full monitor calibration automatically
– CNR should stay within ±15% of measurement obtained during
– Warning level = 5%
Tolerance level = 10%
Recalibrate
• Uniformity Performance – minimize nonnon-uniformities away from center of display
acceptance testing
GE Seno DS
– Warning level = 10%
Tolerance level = 15%
Recalibrate
GE Seno DS
Some slides courtesy of GE
and Vince Polkus and Marcia Hill
17
GE Senographe DS
Seno Advantage RWS
GE Senographe DS
Quality Assurance Plan
•
•
•
•
•
•
Automated Evaluative Procedure
Patented Technology
Tracks IQ Over Time
• Spatial Resolution
• Small Signal Contrast
• Dynamic Range
• Resolution Uniformity
• Distortion
• Other
Run by Technologist
Pass/Fail Result
In-Site Interactive
– Remote Corrections
– Automatic Service Dispatch
GE Image Quality Signature
Test (IQST) Phantom
QAP Not a calibration but a process that maximizes Senographe
digital image quality consistency
Fuji CR
GE Senographe DS
QA Phantom Overview
Uniform area for noise power
spectrum measurement
Step wedge for contrast
measurement
Mesh for resolution
uniformity measurement
Edge object for MTF
measurement
CR Reader
Fuji FCR 5000 MA
CR QC
Workstation
CR Cassettes
Rulers for measuring
distance of detector from
chest wall edge of Bucky
Tabs to position IQST with
respect to chest wall edge of
image receptor.
Rails to position the IQST leftleft-toto-right
24 x 30 cm
18 x 24 cm
18
Fuji CR
Fuji FCR 5000 MA
Performing the Survey
• Manufacturer’
Manufacturer’s tests – Fuji
– Physics tests unique to Fuji CR FFDM
• CR reader sensitivity (“
(“S” number)
• CR reader shading correction
• Imaging plate fogging test
• Verification of AEC with CR cassettes on each unit that is tested
Siemens
Mammomat NovationDR
Siemens - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– AWS Monitor and Viewing Conditions
– Chest Wall and Missed Tissue
– Collimator
– Compression Plate Position
– Spatial Resolution
– Phantom Image Quality
– Mechanical Inspection
– Detector Uniformity
– Radiation Safety – (Optional in U.S)
19
Siemens - Performing the Survey
Siemens - Performing the Survey
• Manufacturer’
Manufacturer’s tests
– kVp Accuracy and Reproducibility
– HVL
– Mean Glandular Dose
• Manufacturer’
Manufacturer’s tests
– Film Printer
– AEC Stability, Reproducibility & SNR
– Viewing Conditions
– Ghost Image
– Illuminance
– SNR and CNR
– AEC Thickness Tracking
– Monitor Constancy
– Pixel Correction
– Detector Calibration
Sectra
Artifacts
20
Performing the Survey
Performing the Survey
• Artifact evaluation - windowing
• Artifact evaluation – Contact Mode GE
Mo/Mo
Performing the Survey
Mo/Rh
Rh/Rh
Performing the Survey
• Artifact evaluation – Mag mode 1.5
Mo/Mo
Mo/Rh
• Artifact evaluation – Mag mode 1.8
Rh/Rh
Mo/Mo
Mo/Rh
Rh/Rh
21
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
22
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
23
Performing the Survey
• Artifact evaluation
Performing the Survey
• Artifact evaluation
24
Performing the Survey
• Artifact evaluation
Artifact at edge of image receptor
Cause: hole in the graphite cover of the digital detector
25
There Are Currently Seven FDAFDA-Approved
Laser Imagers for Digital Mammography
OD Requirements for
HiHi-Resolution Laser Imagers
• Agfa LR5200 Laser Imager (Wet Chemistry)
• Agfa DS4500M
Dmax > 3.5 OD
• Kodak 8600 Laser Imager
• Kodak 8610 Laser Imager
• Kodak 8900M
MidMid-density > 1.5 OD
• Fuji Drypix 7000
• Fuji Drypix FMFM-DP L
Laser Processor QC
Kodak daily sensitometry
Base + Fog
• Does the background match?
Density Difference – OD closest to 2.20 minus OD closest
but not less than 0.45
MidMid-density – step closest to but not less than 1.20
Action Limits:
MD & DD + 0.15 OD
Dmax
RWS Clinical Image Check
B+Fog = 0.03
• Is the background dark enough?
• Does the dense tissue area match?
• Is the dense tissue light enough?
• Is the contrast adequate?
Dmax + 0.25
26
RWS Clinical Image Check
RWS Clinical Image Check
RWS Clinical Image Check
RWS Clinical Image Check
27
Key TakeTake-home Points
• Obtain proper handshands-on training
Key TakeTake-home Points
• Review workstation monitors – look at the clinical images!
– Do they match?
• ACR & FDA applications and forms
• Turn off and on auto print and/or auto push
• Artifacts – most problems can be seen on this test
• Lead sheet protecting detector for Focal Spot, HVL & kVp
– Appropriate dark and light levels
• Do all work on correct images – raw vs. processed
• Can be archiving physics test images on one workstation while
working on another
• Laser Printer
– Dmax at least 3.5 OD
• Take your time and use your professional judgement
– MidMid-density about 1.5 OD
Thank You
28