Objectives

Film-Screen
Mammography QA:
What You Need to
Know

For those new to mammography:

Overview of what you need to do as a
medical physicist
Specific details of how to perform tests will
not be reviewed here best learned in a
hands-on environment

Beth Schueler
Mayo Clinic Rochester

For those who are experienced:

Medical Physicist
Responsibilities

Annual Physics Survey

Annual physics survey

Equipment Performance Evaluation

CFR Part 900 Quality Mammography

Standards
900.12(e)(5) Annual quality control tests
900.12(e)(2) Image quality evaluation

Required for
New mammography units or processors
Disassembled and reassembled mammography
units and processors
After mammography units or processors have
undergone a major repair

tips for improving efficiency

focusing the testing process on items that are
most likely to fail

Includes annual QC tests and Medical

Physicists Checklist

Can be found at the FDA website

http://www.fda.gov/cdrh/mammography/index.html

Includes rules and guidance documents

Annual Physics Survey

ACR Mammography
QC Manual 1999
provides
recommendations for
performance testing
that complies with
MQSA

AAPM 2008: FS Mammography QA

ACR and MQSA tests

and action limits differ
in some cases

ACR QC Manual

ACR QC Manual

MQSA 900.12(e)

1.Unit Assembly

1.Unit Assembly

(5)(i) AEC Performance

2.Collimation

2.Collimation

(5)(ii) kVp Accuracy/Reprod.

3.System Resolution

3.System Resolution

(5)(iii) Focal Spot Condition

4.AEC Performance

4.AEC Performance

(5)(iv) Beam Quality (HVL)

5.Uniformity Screen Speed

5.Uniformity Screen Speed (5)(v) Entrance Air Kerma/Reprod.

6.Artifact Evaluation

6.Artifact Evaluation

7.Image Quality Evaluation

7.Image Quality Evaluation (5)(vii) Collimation

(5)(vi) Dosimetry

8.kVp Accuracy/Reprod.

8.kVp Accuracy/Reprod.

(5)(viii) Uniformity Screen Speed

9.Beam Quality (HVL)

9.Beam Quality (HVL)

(5)(ix) Artifact Evaluation

10.Breast Dosimetry

10.Breast Dosimetry

(5)(x) Radiation Output

11.Viewbox Luminance

11.Viewbox Luminance

(5)(xi) Decompression
(2) Image Quality Evaluation

ACR QC Manual

MQSA 900.12(e)

1.Unit Assembly

(5)(xi) Decompression

2.Collimation

(5)(vii) Collimation

3.System Resolution

(5)(iii) Focal Spot Condition

4.AEC Performance

(5)(i) AEC Performance

5.Uniformity Screen Speed

(5)(viii) Uniformity Screen Speed

6.Artifact Evaluation

(5)(ix) Artifact Evaluation

7.Image Quality Evaluation

(2) Image Quality Evaluation

8.kVp Accuracy/Reprod.

(5)(ii) kVp Accuracy/Reprod.

9.Beam Quality (HVL)

(5)(iv) Beam Quality (HVL)

10.Breast Dosimetry

(5)(vi) Dosimetry, (5)(v)

Entrance Air Kerma/Reprod.,
(5)(x) Radiation Output

11.Viewbox Luminance

1. Unit Assembly
Unit is mechanically stable
Cassettes do not slip when gantry is angled
Compressed breast thickness scale is
accurate to within 5 mm
Test with a phantom at 2, 4, and 6 cm thickness

MQSA (5)(xi) Decompression:

10

Automatic decompression can be overridden

Status is displayed
Compression can be released manually if
power fails

2. Collimation

2. Collimation

X-ray to light field

alignment
X-ray to image
receptor alignment
Compression
paddle chest wall
alignment to image
receptor

Repeat for all routinely used collimator /

bucky / paddle / target material
combinations

11

AAPM 2008: FS Mammography QA

MQSA: Only those combinations used for

producing full-field clinical images in the
contact mode

Recommend that magnification mode

collimation testing be included
12

2. Collimation

3. System Resolution

Action Limits:

Evaluated with a high

contrast resolution
pattern

X-ray to light:
L + R or anterior + chest deviations > 2% SID

X-ray to IR:

within 1 cm of chest wall

centered L-R
4.5 cm above breast
support surface

clipping is visible at chest wall

clipping R or L > 2% SID (ACR only)
clipping anterior > 4% SID (ACR only)
x-ray extends beyond IR > 2% SID on any side

Compression paddle:
14

Clinical Technique for a

Standard Breast

3. System Resolution
Repeat for all target
materials, focal spots,
film-screen combinations
View pattern on the film
with a 7-10X magnifier
Action limits:

Target, filter and kVp used for 4.2 cm thick

50% glandular breast
Same as used for Image Quality Evaluation

What if this is not known for sure for a

new installation?
Use best guess
MQSA guidance recognize it could change

15

Technique: AEC with

kVp used for standard
breast

16

< 11 lp/mm perpendicular

to the A-C axis
< 13 lp/mm parallel to the
A-C axis

ANTERIOR

13

edge visible in image

projects beyond chest wall > 1% SID

CHEST WALL

ANODE-CATHODE
AXIS

4. AEC System Performance

4. AEC System Performance

Thickness / kVp tracking

Action limit:

Homogeneous phantom 2, 4, 6, 8 cm thick

representative of typical breast size
Appropriate kVp for each thickness

Thickness / kVp tracking:

2-6 cm thickness film OD > 0.15 OD of the mean

Image mode tracking:

Overall film OD > 0.3 OD of the mean

Image mode tracking

includes image modes and 8 cm thickness

4 cm thickness for small bucky, large bucky,

magnification stand

MQSA: required for Equipment Performance

Evaluation only

Density control function

Recommend testing all image modes annually

Film optical density < 1.20 OD

17

AAPM 2008: FS Mammography QA

18

Multiple AEC Detectors

AEC Calibration Methods

If the system has a different AEC detector

for each bucky:

Some systems have a separate AEC

calibration for each kVp
Automatic kVp selection may result in
variable results
Recommend testing all kVp values in the
clinically used range for these specific
systems

Test thickness / kVp tracking on each one

If the system has multiple AEC detectors

in a single bucky that are individually
selectable

19

Test thickness / kVp tracking on one detector

and 4 cm thickness on all others
Must be within 0.3 OD of the mean

20

Example 1: Thickness/kVp
Tracking

Example 1: Thickness/kVp
Tracking
2
Film Optical Density

Film Optical Density

2
1.9
1.8
1.7
1.6
1.5
1.4
2

1.7
1.6
1.5

Phantom Thickness (cm)

21

Phantom Thickness (cm)

22

Example 1: Thickness/kVp
Tracking

Example 2: Thickness/kVp
Tracking
2

1.9

24 kVp

1.8

25 kVp

25 kVp

Film Optical Density

2
Film Optical Density

1.8

1.4
1

28 kVp

1.7
1.6
1.5

26 kVp

1.4
1

23

1.9

Phantom Thickness (cm)

AAPM 2008: FS Mammography QA

1.9
1.8
1.7
1.6
1.5
1.4

24

Phantom Thickness (cm)

Example 2: Thickness/kVp
Tracking

4. AEC System Performance

Density control function (ACR only)

Film Optical Density

2
1.9

4 cm thickness at clinically-used density

settings (typically -2 to +2)

25 kVp 25 kVp

1.8

Action limit:

1.7
1.6

24 kVp

1.5

28 kVp

1.4
1

25

Steps should be approximately 0.15 OD

change (12%-15% mAs change)

26 kVp

Phantom Thickness (cm)

26

Mammographic DCF Test Tool

Mammographic DCF Test Tool

Allows measurement of multiple AEC

exposure ODs on a single film

Available from Gammex-RMI

http://www.gammex.com/

Eliminates variations in film emulsion and

processing

27

5. Uniformity of Screen Speed

5. Uniformity of Screen Speed

Expose all cassettes using AEC

Phantom: uniform 4 cm thick PMMA
that covers the entire image receptor
Image a control cassette multiple times
to ensure repeatability of AEC and
processor
Review all images for presence of
screen artifacts

Action limit:

29

AAPM 2008: FS Mammography QA

Difference between maximum and minimum

film OD > 0.3 OD

If groups of cassettes with different size

or speed class are outside this range:
Appropriate compensation should be
specified on the technique chart
Groups can be evaluated separately
30

6. Artifact Evaluation

6. Artifact Evaluation

Phantom: uniform 4cm thick PMMA that

covers the entire
image receptor
Make 2 identical
exposures on the
same cassette and
process lengthwise
and widthwise

Repeat for

31

Both image receptor sizes

Magnification stand and small focal spot
Each filter
Use large bucky for filter evaluations

Review films for artifacts

32

6. Artifact Evaluation

6. Artifact Evaluation

Processor artifacts
will be oriented on
both films in the
same direction
relative to the
direction of film
feed

X-ray unit artifacts

will be oriented on
both films in the
same direction
relative to the long
axis of the film

33

34

Artifact Problem-solving

Artifact Problem-solving

To isolate filter or processor artifacts

To isolate grid artifacts

Eliminate grid/bucky cover artifacts by placing

cassette and phantom on top of the breast
support tray

35

AAPM 2008: FS Mammography QA

Image the grid in a stationary position

Imaging method will vary depending on the
system design
Some options include:

36

If system allows exposure without a bucky, detach

bucky from the gantry and image the grid with a
cassette in the bucky
If system requires a bucky to be in place, install
one and place the other on top of it and image the
grid with a cassette inside

6. Artifact Evaluation

7. Image Quality Evaluation

For units with multiple targets:

ACR Mammography
Accreditation Phantom
with contrast disc
AEC exposure using
clinical technique factors
for a 4.2-cm thick breast

All focal spot / target / filter combinations

used clinically must be tested for Equipment
Performance Evaluation
e.g. Mo/Mo/LFS, Mo/Rh/LFS, Rh/Rh/LFS,
Mo/Mo/SFS, Mo/Rh/SFS, Rh/Rh/SFS 6 combos

An alternative standard allows for testing

each focal spot, target and filter (not each
combination) for annual physics survey only
37

e.g. Mo/Mo/LFS, Mo/Rh/LFS, Rh/Rh/LFS,

Mo/Mo/SFS, Mo/Rh/SFS, Rh/Rh/SFS 3 combos

38

7. Image Quality Evaluation

7. Image Quality Evaluation

Action limit:

Additional recommendations:

Background density
< 1.4 OD

Background density
1.8 OD or higher
within 0.20 OD of facilitys target

(MQSA < 1.2 OD)

Change by > 0.20 OD

Density difference

Density difference
< 0.40

0.55 or higher
within 0.05 of facilitys target

Change by > 0.05

39

41

Verify with technologist

(kVp, target, filter, density
control setting, AEC cell
location)

40

All units producing films that are read at a

specific location should have the same target
values

Phantom Scoring: Fibers

Phantom Scoring: Specks

Use optimal viewing

conditions and a
magnifying lens
Count from thickest to
thinnest
Count as 0.5 if not all but
more than half is visible
Deduct for a fiber-like
artifacts

Count from largest to

smallest group
Count as 1

from the last whole or half

fiber counted only

AAPM 2008: FS Mammography QA

if 4 or more specks in the

group are visible

Count as 0.5
if 2 or 3 specks visible

Deduct for a speck-like

artifacts 1 for 1 speck
42

from the last whole or half

group counted only

Phantom Scoring: Masses

7. Image Quality Evaluation

Count from largest to

smallest
Count as 1

Action limit:
< 4 fibers, < 3 speck
groups, < 3 masses
visible
Change in score of >
0.5 from previous
measurement

if > 75% perimeter is visible

and generally circular

Count as 0.5
if visible and not generally
circular

Deduct for a mass-like

artifacts
43

from the last whole or half

mass counted only

44

8. kVp Accuracy and

Reproducibility

8. kVp Accuracy and

Reproducibility

kVp accuracy

Action limits:

Measure at these settings:

Measured kVp > 5% from set kVp

Coefficient of variation (standard deviation /
mean) > 0.02

Lowest clinical kVp that the test device can

measure
Most commonly used kVp (typically the same as
the phantom kVp)
Highest available clinical kVp

kVp reproducibility
45

Make 4 exposures at the most commonly

used kVp

46

9. Beam Quality Assessment

(HVL Measurement)

10. Breast Dosimetry

Include compression paddle

Breast entrance
exposure

Use type 1145 aluminum sheets (99.9%

purity)
Repeat for all target / filter combinations

Measure for ACR

mammography
phantom using clinical
technique factors

Action limit:
MQSA: HVL < kVp/100 (mm Al)
ACR: HVL < kVp/100 + 0.3 (mm Al) or

47

HVL > kVp/100 + 0.12 (Mo/Mo),

HVL > kVp/100 + 0.19 (Mo/Rh),
HVL > kVp/100 + 0.22 (Rh/Rh)

AAPM 2008: FS Mammography QA

AEC Reproducibility
Repeat 4 times

Radiation output rate

48

10. Breast Dosimetry

Average Glandular Dose (Dg)

Action limit:

Dg = DgN x XESE

Coefficient of variation > 0.05 for either

exposure or mAs

where Dg = average glandular dose,

DgN = glandular dose to exposure factor,
and XESE = entrance skin exposure
DgN is a function of filter/target, kVp and
HVL

Multiple AEC Detectors

If different AEC detector for each bucky:
Test AEC reproducibility on each one

Tables 1-3 in ACR Quality Control Manual, 1999

If multiple AEC detectors in a single bucky

that are individually selectable

Action limit:
Dg > 3 mGy

Test AEC reproducibility on one only

49

50

Radiation Output Rate

11. Viewbox Luminance and

Room Illuminance

Measure 4.5 cm above breast support tray

Photometers

At 28 kVp, Mo target, Mo filter, > 3 sec

manual exposure time
Maximum SID
With compression paddle in place

Vendors include:
UDT Instruments
Quantum Instruments
Various display monitor
luminance measuring
units (MSfit ACT, )

Action limit:
Radiation output rate < 7.0 mGy/sec

51

52

30 min warm-up time for

viewboxes needed for
stable measurements

11. Viewbox Luminance and

Room Illuminance

Report

Action limits:

Preliminary report recommended

Let the facility know about any repairs
needed at the time of the survey if possible

kcd/m2

Viewbox luminance < 3

Illuminance on viewbox surface and seen by
observer > 50 lux

Final report required

Returned within 30 days of the survey date
Dated with report date
Signed

Most commercially available rollo-scopes

for mammography can achieve > 6 kcd/m2
Change in luminance of 1-2 kcd/m2 generally
indicates a bulb or ballast needs to be
replaced
53

AAPM 2008: FS Mammography QA

Include a review of technologists QC

tests
54

Corrective Action

Test Equipment Calibration

For Equipment Performance Evaluation

Air kerma measurement device must be

calibrated at least once every 2 years and
after a repair

All test failures must be corrected prior to

patient use

For annual QC survey

NIST traceable calibration laboratory

Record calibration date on survey report

Keep letter or certification from calibration
laboratory for possible review

Image quality evaluation or Average glandular

dose:
Failures must be corrected prior to patient use

All other tests:

55

Failures must be corrected within 30 days of survey

date

56

Practical Tips

Practical Tips: QC Test Order

Talk with the technologist before you start

kVp first

Phantom technique

Must pass for most other test results to be

valid

kVp, typical mAs, AEC mode, AEC cell location,

density control function setting, target OD and DD

Dosimeter tests

Technique chart use

HVL, Output, Dose and Image Quality

Use last Dose image for IQ evaluation

What techniques do they actually use?

Does the chart need changes?

57

What density function control settings do they

use in practice?
Any specific equipment issues or problems?

58

Practical Tips: QC Test Order

Practical Tips: QC Test Order

Small bucky

Large bucky, magnification stand

Collimation, Resolution
AEC

Complete applicable tests

Reading room

Test compressed breast thickness scale readout

with different phantom thicknesses

Score IQ phantom image

Viewbox luminance and room illuminance

Screen Speed Uniformity, Artifact

Use repeat Screen Speed Uniformity images for
Artifact Evaluation

59

AAPM 2008: FS Mammography QA

60

10

Common Test Failures

QC Test
Artifact Evaluation:
Processor (27)
X-ray unit (5)
Cassette (3)
Collimation:
Light to x-ray (12)
X-ray to IR (10)
Compression paddle chest (12)
61

Common Test Failures

# of Failures
Found *
35

QC Test
Unit Assembly: Thickness readout
AEC: Thickness/kVp tracking
Uniformity Screen Speed
System Resolution
Image Quality Evaluation
kVp Accuracy/Reprod.
Beam Quality (HVL)
Breast Dosimetry

34

62

* Out of 110 surveys

Mammography Equipment
Evaluation

Defined in Policy Guidance Help System

Under Index: Medical Physics Survey: Medical
Physics Annual Survey

New mammography units or processors

Disassembled and reassembled
mammography units and processors
After mammography units or processors have
undergone a major repair

Major repairs require medical physicist

conducts evaluation on-site
For other repairs, FDA recommends

What is considered a major repair?

63

64

AEC sensor replacement

X-ray tube replacement

* Out of 110 surveys

Major Repairs

Required for

Item

# of Failures
Found *
31
24
5
0
0
0
0
0

Major
MP
Repair Involvement
Y
On-site testing
Y
On-site testing

Filter replacement

On-site testing

Collimator/blade replacement
Collimator adjustment

Y
N

On-site testing
Oversight

Density control internal

adjustment
Bucky replacement (no AEC
sensor)
Processor roller replacement

Oversight

Oversight

Optional

Medical physicist oversight (review of tests

performed by the technologist or service
engineer)
Medical physicist involvement optional

Medical Physicists Checklist

Items found in 900.12(b)
Includes items related to x-ray equipment:
Magnification between 1.4 and 2.0
Focal spot / target material displayed to user
Hands-free power-driven compression from
both sides of the patient

Items related to processing and viewing:

Hot light and film-masking devices available

AAPM 2008: FS Mammography QA

66

11

Medical Physicists Checklist

Field Light Illumination

Can be
downloaded
as an Excel
file from ACR
website
www.acr.org

900.12(b)(5)
Illumination of not
less than 160 lux at
the maximum SID
Difference between
readings with and
without field light on
Photometer

67

68

Compression Paddle Deflection

Compression Paddle Deflection

900.12(b)(8)(ii)(B) The compression

paddle shall be flat and parallel to the
breast support table and shall not deflect
from parallel by more than 1.0 cm at any
point on the surface of the compression
paddle when compression is applied.

Compress foam or
tennis balls to 111 N
(25 lbs)
Measure the height
of each corner of
the paddle above
the breast tray
Maximum deviation
< 1.0 cm

Measurement procedure found in Policy

Guidance Help System
69

Under Index: Compression Device: Compression

paddle

70

Compression Paddle Deflection

Compression Paddle Deflection

900.12(b)(8)(ii)(C) Equipment intended by

the manufacturers design to not be flat
and parallel to the breast support table
during compression shall meet the
manufacturers design specifications and
maintenance requirements.

Example: Hologic Fully Automatic Selfadjusting Tilt (FAST) Paddle

71

AAPM 2008: FS Mammography QA

Designed to flex to conform to tissue

Manufacturer specification:
18x24 cm: 1.5 cm deflection at 107 N (24 lbs)
24x30 cm: 2.25 cm deflection at 138 N (31 lbs)

72

12

Information Resources

Information Resources

FDA Mammography website

Physics QC Test data forms in Excel with

formulas (Doug J. Simpkin, Ph.D.)

http://www.fda.gov/cdrh/mammography/index.html

CFR Part 900.12 Quality Mammography

Standards
Policy Guidance Help System

http://www.geocities.com/djsimpkin/

ACR Mammography QC Manual (1999)

ACR website
73

http://www.acr.org/
Test data forms in Excel

AAPM 2008: FS Mammography QA

74

13