October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Stereotactic Breast Biopsy
Timely Tips for Medical Physicists
Acknowledgements
Thanks to
ACR SBBAP Physics Sub-Committee, especially
Maynard High, Ph.D.
Allen Hrejsa, Ph.D.
Andrew Maidment, Ph.D.
For contributing valuable slides to this presentation
Robert J. Pizzutiello, Jr., F.A.C.M.P.
Upstate Medical Physics, Inc.
Side View
Lorad
Focal Spot
Side View
Front View
Focal Spot
X-ray
Tube
X-ray
Tube
Small area
Collimator
Small area
Collimator
Compressed
Breast
Compressed
Breast
Phosphor
Fiber Optic
CCD
Mirror
CCD
CCD
Lens
Front View
Fischer
Phosphor
A/D
Optical coupling/mirror system
Light reflection from phosphor
DMA
A/D
2:1 fiberoptic taper demagnification
Light transmission through phosphor
Digital Image Quality
‹ Contrast
‹ Blur
‹ Noise
‹ Artifacts
‹ Dose
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
DMA
Noise
‹ Noise
decreases (improves) with
increasing mAs
‹ Images may be produced using any
mAs technique (from 10 - 500 mAs)
‹ Window and level controls can be
used to make the image “appear”
properly exposed
‹ System noise will change
11
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
113 mrad
58 mrad
Factors Affecting Breast Dose
‹ kVp,
mAs
thickness
‹ breast composition (dense or fatty)
‹ multiple exposures
‹ breast
28 kVp
‹ digital
image processing does NOT affect
dose
‹ optical density of film (if hardcopy is used)
does NOT affect dose
‹ Let’s
take a look at kVp dependence…
339 mrad
227 mrad
30 kVp
28 kVp
To Minimize Breast Dose
Develop and maintain a good technique chart
Obtain manufacturer’s suggested techniques
‹ Evaluate image quality at different mAs values
(Technologist and Medical Physicist)
‹ Moderately higher mAs will reduce image
noise, but increase dose
‹ Insufficient mAs will produce a noisy (grainy)
image, but can be made to appear “well
exposed” with window/level control
‹ Excessive mAs images may also appear “OK”
with window/level adjustment
‹ Minimize retakes
‹
‹
120 mrad
32 kVp
34 kVp
QC Tests Unique to SBB
Minimum Testing Frequencies
Zero Alignment Test
Before each patient
(only on some units)
Localization Accuracy Test (in Air)
Daily
Phantom Image Quality Test
Weekly
Hardcopy Output Quality
Monthly
(if hard copy is produced
from digital data)
Visual Equipment Check
Repeat Analysis
Compression Force Test
Zero Alignment Test
‹ Perform
before each patient
that zero coordinate is
accurate
‹ Assures that stereotactic unit is not
improperly installed
‹ Verify
Monthly
Semi-annually
Semi-annually
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
B
RT
22
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Setting Z zero on Lorad:
Phantom Image
Quality Evaluation
Nuclear
Associates Digital
Mini Phantom
Maynard High, Ph.D.
Fibers
Specks
Masses
ACR
Accreditation
NA Digital
1.56
1.12
0.8
0.75
0.54
0.54
0.4
0.32
0.24
0.16
2
1
0.75
0.5
0.25
x
x
0.93
0.74
0.54
0.54
x
0.32
0.24
0.2
x
1
0.75
0.5
0.25
Mammography
Accreditation
Phantom
RT
Minimum Passing
Phantom Image Scores
Fibers
Specks
Masses
ACR-MAP
Accreditation
Phantom
MiniPhantom
Screen/film
Digital
Digital
4.0
3.0
3.0
5.0
4.0
3.5
3.0
3.0
2.5
Be sure to use only an approved phantom
Phantom Imaging:
a common avoidable failure
‹ NAD
Digital Mini Phantom
1st image (image quality)
2nd image (TLD)
‹ Mammo
Accreditation Phantom
4 images for image quality
5th image for TLD
‹ OK
to window/level digital images
grid (or not) per clinical technique
‹ Use
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
W
33
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Hardcopy
Output Quality
Measure optical density
M
Record or plot data
‹ Laser
or multiformat camera
SMPTE Test Pattern, if
available
‹ Record window width, level
‹ Produce hardcopy
‹ Measure OD at a consistent location
on each image
‹ Record and monitor for consistency
Image Jan Feb Mar
‹ Evaluate
RT
Repeat Analysis
S
1.62 1.62 1.63
3
1.59 1.57 1.61
4
1.61 1.60 1.58
5
1.63 1.64 1.62
6
1.60 1.61 1.62
SBB Annual Medical Physics Survey
‹
‹
Count repeated and rejected film by
category and tabulate
‹ Use a log of images repeated
‹ Document analysis and corrective action even if your repeat rate is low
‹ Repeat rate will typically be higher than
for mammography
‹
‹
‹
‹
‹
‹
‹
‹
MP
LOOK at
THIS
BEFORE
YOU
START !
1.63 1.61 1.60
2
Set limits
(e.g.) 1.60 + 0.05
‹
RT
1
‹
SBB Unit Assembly Evaluation
Collimation Assessment
Focal Spot Performance and System Limiting
Resolution
kVp Accuracy and Reproducibility
Beam Quality Assessment (HVL)
Automatic Exposure Control System
Performance
Uniformity of Screen Speed or Digital Field
Breast ESE, AGD, AEC Reproducibility
Image Quality Evaluation (phantom)
Artifact Evaluation
Localization Accuracy
Getting Started
If you have an upright
stereo, LET TRAINED
TECHNOLOGIST MAKE
ALL CONNECTIONS
! NEVER remove or install
cables when power is on.
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
44
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Assembly Evaluation
‹
Free-standing unit is mechanically stable
‹
All moving parts move smoothly, without
obstructions to motion
‹
All locks and detents work properly
‹
Image receptor holder is free from vibrations
‹
Image receptor is held securely by assembly in
any orientation
MP
‹
Image receptor slides smoothly into holder
assembly
‹
Compressed breast thickness scale is accurate
to ± 0.5 cm, reproducible to ± 2 mm
‹
Patient or operator is not exposed to sharp or
rough edges or other hazards
‹
Operator technique charts are posted
‹
Operator protected by adequate radiation
shielding
MP
Collimation
‰
Assembly Evaluation
X-Ray Field should extend beyond Image Receptor
on all 4 sides
‰
X-ray Field should extend < 5 mm on any side (in
plane of image receptor)
‰
Note: X-rays beyond the digital image receptor will not
be seen on the monitor
‰
Does the biopsy window align with the image field of
view?
Lorad Collimation Assessment:
Film cassette behind compression paddle
BIOPSY
WINDOW in
compression
paddle
MP
Collimation Assessment:
Digital Image
‹
Measure visible
diameter of coin with
TOOLS/CALIPERS
(Lorad).
‹
Anterior missing image
is 19.0 – 17.6 = 1.4 mm
‹
Should be <5 mm
BUT… How about the X-Ray Field ?
CHEST WALL
Incorrect or misadjusted
collimator aperture
17.6 mm
19.0 mm
Anterior
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
Film behind compression paddle (does not
show full extent of x-ray field)
55
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Focal Spot Size Performance System Limiting Resolution
Line Pair Test Pattern
Use film (x-ray machine)
‹ Use CRT image (“system”)
‹ Technique, clinical kVp
‹ Scoring the image
‹
‹
Film - Lines distinct over 1/2
length
CRT - Lines distinct, correct #
over any part of pattern
‹
Need to measure in
BOTH:
512 matrix
1024 matrix
‹
Need to measure
BOTH:
Must ZOOM Image:
Parallel to A-C axis
Perpendicular to A-C
Typical with this
Test object:
kVp Accuracy Reproducibility
5.6 lp/mm (512 mode)
7.1 lp/mm (1024 mode)
Beam Quality (HVL)
Thickness of aluminum to
reduce radiation exposure
by one-half
‹ Affects contrast and dose
‹ Used in dose calculation
‹ minimum = kVp/100
‹ No compression paddle
lucite in the beam
‹
‹
Verify that actual kVp’s are the same as the
indicated kVp’s
‹
Range of clinical kVp values
‹
Accuracy within 5%
‹
Reproducible CV < 0.02
MP
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
MP
66
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
AEC System Performance
‹
‹
‹
‹
‹
‹
MP
Lorad: Matrix Size and Patient Dose
AEC available on some digital SBB units
Performance Capability
Record signal level as
function of thickness
and technique
Monitor exposure time
Performance Capability
(4,6,8 cm)
Performance Criteria:
a) 6, 8 cm DR# should be within 20% of 4 cm.
b) DR# should meet Lorad target of 4000 (512) / 6000
(1024).
Provide suggested
technique chart
‹ For
same kVp & mAs, and same
amplifier gain, DR# in 1024 will be 1/4
DR# in 512.
‹ Lorad boosts amplifier gain 2X in
1024, so DR# in 1024 is 1/2 DR# in
512 at same dose
‹ Thus, for equal DR#, mAs needs to
be doubled in 1024 mode.
‹ However…
Develop a Technique Chart
6. Automatic Exposure Control (AEC) System
or Manual Exposure Performance
‹ Lorad
recommends HIGHER target
DR# for 1024:
Performance Capability
DR# = 4000 in 512 mode
DR# = 6000 in 1024 mode
Thickness Compensation
Imaging mode
Digital 512 - Manual mode
Focal spot:
Large
Prepare for both 512 and
1024 modes (Lorad)
mA:
Phantom
thickness
‹ Therefore
input exposure must be
increased in 1024 mode by:
2 * (6000/4000) = 3 times !!
mAs
Target
28
56
Mo
Mo
4269
6 cm
30
140
Mo
Mo
4361
8 cm
34
175
Mo
Mo
4491
4 cm value
4269
Uniformity of Screen
Speed or Digital Field
‹ Image
a uniform phantom
‹ Screen Film systems
Each cassette produces the same
optical density under the same
conditions
‹ Digital
Systems
Digital detector produces uniform signal
values across the field of view
Mean Signal
Value
kVp
4 cm
Filter
E v a lua tio n
Density or Signal Range Allowable Range
4269 - 4491
3415.2 - 5122.8
Pass
Digital Receptor Uniformity:
Lorad Protocol
28 kVp
mAs for DR# =4000
‹ Measure SNR’s with
TOOLS/STATS at
specified locations.
‹ 32 x 32 pixel ROI –
set with trackball.
‹ Lorad spec +/-20%
of SNR(center).
‹
‹
SNR
SNR
UL
UR
(100, 100)
(400, 100)
SNR
Centr
(256, 256)
SNR
LR
(100, 400)
(400, 400)
MP
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
SNR
LL
77
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Lorad “Hidden” Statistics Functions:
NT System ROI Statistics:
S
Press “S”
displayed
1)
2)
key on keyboard when ROI is
to obtain:
STANDARD DEVIATION
SIGNAL / NOISE
NT System ROI Statistics:
Phantom Image Quality
Same procedure as for
technologists
‹ Medical Physicist
reviews scoring
procedure and checks
for consistency
‹ Uses technique factors
for dose determination
‹
MP
Breast Entrance Exposure,
AGD
Artifact Evaluation
Unwanted irregularity not caused by
structures of interest
Causes (Digital)
Digital Image Receptor
Data per technique chart
Measure ESE
‹ HVL determines DgN
‹ AGD = ESE * DgN
‹ AGD < 300 mrad
‹ Dose and Optical Density
‹
‹
‹
‹
‹
corrected by white-fielding
‹
MP
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
Dust (camera, screen, lens, mirror)
Pixel defects (dropouts)
Non-uniformities (light pipe stucture,
vignetting, linear shading)
Clipping (dose too high)
MP
88
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
On Lorad, “White-Field” is
a service procedure:
Dust Speck Artifact:
- Dust Moved after White-Fielding
‹ Can
correct some non-uniformities
‹ Consists of the average of 10
exposures on 4-cm lucite phantom
‹ Correction formula is:
(Image – Dark Field)
(White Field – Dark Field)
“Moved” Dust Artifacts
“Moved” Dust Artifacts
White Speck next
to Black Speck
White Speck next
to Black Speck
Targeting Accuracy
Gel Phantom Pre-Fire Images:
Correct placement for needle:
Performed annually by technologist
under supervision of medical physicist
‹ Position gel-type phantom
‹ Image, target and sample
‹ Result: was the lesion collected?
‹
MP
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
99
October
October 1999
1999
QC
R.
QC for
for Stereotactic
Stereotactic Breast
Breast Biopsy
Biopsy -- R.
Pizzutiello
Pizzutiello
Role of the Surgeon in QC
Understand the importance of QC in SBB
Assures that personnel remain qualified
‹ Support QC activities
Allow enough time for QC
Provide for QC training
Periodically check that QC is done as required
‹ Confer with medical physicist annually
‹ Assure that follow-up is done if the QC
program indicates corrective action is required
‹ Accreditation
‹
‹
copyright
copyright Upstate
Upstate Medical
Medical Phyics,
Phyics, Inc.
Inc.
(716)
0350
(716) 924
924--0350
10
10