11/21/19
Advanced QA/QC in
fluoroscopy
Ehsan Samei, PhD, DABR, FAAPM, FSPIE, FAIMBE, FIOMP
Duke University Health System
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 1
11/21/19
Fluoroscopy
• An adaptive modality
• One of the highest dose
procedures
• Most rigorous estimation of
dose and visual image quality
• Least availability of quantitative
measures of quality
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 2
11/21/19
Why QC?
• Because quality is a required necessity of
medical care
• Because the actual clinical performance of
an imaging system can never be fully
assured with a priori knowledge and
actions
10000
1000
Frequency
100
10
1
10
11
10.5
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
9.5
Air KERMA (Gy)
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 3
11/21/19
Fluoroscopy tests
• Exposure • Temporal performance
– Maximum exposure – Pulse-rate accuracy
– AKR Accuracy – Pulse width consistency
– Exposure reproducibility – Temporal resolution
– Leakage radiation • Image quality
• Spectrum – High contrast resolution
– kVp accuracy – Low contrast detectability
– kVp reproducibility – Scatter characterization
– HVL – Display monitor evaluation
• Geometry – Artifact evaluation
– X-ray field/imaging chain congruence – Task-based IQ
– X-ray field/imaging change alignment • Consistency
– Minimum field size – Intra-fleet exposure consistency
– Minimum source-to-skin distance
What is “Advanced” QC?
• QC informed by the purpose of the
fluoroscopy examination
– Providing the needed information for the
clinical task at the highest possible safety
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 4
11/21/19
Fluoroscopy 1.0 3.0
Focus of MP’s Equipment, focused on Patient, focused on consistency of care
attention individual systems and isolated across practice
performance
Image quality Visual and subjective, in and Quantitative, in and through phantoms and
evaluation through phantoms patient cases
Evaluation Standardized techniques Techniques most closely reflecting clinical
condition use and variation across cases
MP “tools of the Visibility of dots/holes, line MTF, NPS, detectability, quantitative
trade” for image pairs, wire meshes characterization of temporal performance
quality evaluation
Patient dosimetry DAP, Air KERMA, fluoro time, Standard dosimetry, organ dose, risk index
skin exposure
Image processing Often ignored Incorporated in quantitative evaluation and
optimization
System evaluation Focused on x-ray source and Focused on the system as a part of the fleet
detector alone
Applications Focused on 2D applications Expansion to 2.5-3D applications
Protocols Ad hoc Systematic analysis of dose/IQ patterns
informing refinements of procedures
Precision by inference
Technology assessment
Precision by prescription Precision by outcome
Prospective use definition Retrospective quality audit
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 5
11/21/19
Clinical physics support in
fluoroscopy
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
Outline
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 6
11/21/19
Outline
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
Operational Performance
Evaluation
• Quantitative metrology
– Moving from visual to quantitative
– Digital implies “measureable”
• Task-based metrology
– MTF, NPS
– Metrics relevant to a clinical task, d’ in in 2D, 2.5D, 3D, 4D
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 7
11/21/19
Quantitative performance
Quantitative performance
Low Contrast High Contrast,
Detectability Holes Resolution Inserts
Hole Depth (inches) Line Pairs / inch
0.0680 12
0.0490 16
0.0350 20
0.0250 24
0.0180 30
0.0126 40
0.0091 50
0.0063 60
0.0040 80
Low Contrast targets calibrated
for range of kVp
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 8
11/21/19
Quantitative performance
Qualitative -> Quantitative
& -, /01 = % &2345674
' +,
!"# = %& ( )*
N
*"89:;<=> − *"8@A<
&-* =
B@A<
Quantitative performance
Low Dose Mode – 85 kVp, 2.4 mAs Manual Mode – 85 kVp, 1.5 mAs
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 9
11/21/19
Outline
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
8. Last image hold
9. Cine and spot-film
acquisitions
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 10
11/21/19
Use factors
1. Inherent filtration
2. Patient positioning Softer beam
3. Dose modes
Energy deposited
Harder beam
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
8. Last image hold
9. Cine and spot-film Depth
acquisitions
Skin
Use factors
.
+,
1. Inherent filtration !"#$ = !&! ' ( )*$ '
+, + +.
2. Patient positioning .
+, + +.
3. Dose modes II !&! = !"#$ ' ( *$
'
+,
4. Pulse rate
5. Collimation
d2 Keep this as small as
6. Mag modes
possible
7. Beam on time
8. Last image hold Detector as close as
d1 possible to patient
9. Cine and spot-film
acquisitions Tube as far as possible
X-ray tube from patient
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 11
11/21/19
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
8. Last image hold
9. Cine and spot-film
acquisitions
Operating Modes
Continuous Mode
• Continuous on x-ray beam
• Video camera / FPD displays images at 30 frames/sec
• Any motion within the 33 msec window will blur the image
• Maximum entrance exposure rate is 10 R/min (regulation)
High Dose Rate Mode
• Used for obese patients (sparingly please!)
• Allows entrance exposure rates up to 20 R/min (regulation)
• Separate pedal / additional pressure and audible signal is required
Low Dose
• Feature on most c-arm systems
• Standard GI units often have size selections available which adds filtration
to the beam to lower radiation dose (to children) 24
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 12
11/21/19
Dose modes
• How does the system set appropriate
technique?
• Low Dose Mode
• High Quality Mode
• High Level Fluoro
Mode
*Adapted from Essential Physics of Medical Imaging by Bushberg, et al.
Dose modes
Low Dose Mode = ~½x Normal AKR
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 13
11/21/19
Dose modes
High Level Fluoro = ~2x Normal AKR
Road Mapping
• Allows operator to capture an image (usually with a little
contrast medium injection) for display next to the live
fluoroscopy image or overlay onto the live fluoroscopy
image.
28
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 14
11/21/19
Digital Subtraction Angiography (DSA)
• Allows for real-time subtraction of pre- and post-
contrast images
• Clinically used for vessel visualization
29
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
8. Last image hold
9. Cine and spot-film
acquisitions
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 15
11/21/19
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
8. Last image hold
9. Cine and spot-film
acquisitions
Pulse rate 33 ms = 30
Intensi fps
ty
Time
5 ms 66 ms
33 ms
= 15= 30
Intensi fps fps
ty
Time
Lowerframe
Same framerate
rate==Same
Lowerdose
dose,
with
butless
“choppier”
in-frameimages
blurring from
motion
sequences
• Pulsed fluoro can reduce the dose compared to continuous fluoro
(remember spoke phantom?)
• Pulsed mode can reduce in-frame motion artifacts compared to continuous
fluoro
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 16
11/21/19
Operating Modes
Frame Averaging
• Fluoroscopy images provide excellent temporal resolution, but can be
relatively noisy.
• sacrifices some temporal resolution for less noise.
• Real-time averaging by digitizing images in computer memory
• May allow lower dose imaging as well!
• May produce noticeable image lag
34
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
7. Beam on time
Wasted exposure
8. Last image hold
9. Cine and spot-film
acquisitions
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 17
11/21/19
Collimation vs mag
Mag mode
FOV1
&ℎ(#()* 2345
!"#$"#0(16788 ≈ 012345
) = 01
+,-+
Unused region
of detector
FOV2
:;<
:;< 2345
2345
Unused region 011 ≪
≈ 011
of detector A!B1 E
Higher Noise for
>?$(*",- ∝ same
A!B2
exposure
Mag mode increases skin exposure but also improves image quality
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 18
11/21/19
“Hidden” mag mode
In mobile C-Arms:
– 9” II AKR ≈ 12” II AKR
in Mag Mode
• Important to understand
differences in equipment
• Smaller II ≈ Mag Mode in Larger II
• Choose Smaller II for appropriate
reasons (Image Quality, not FOV)
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes Image from NCRP 168
7. Beam on time • Any fluoro time when the
8. Last image hold operator is not viewing
9. Cine and spot-film the images is
acquisitions unnecessary exposure
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 19
11/21/19
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes
• Use the LIH to reduce
7. Beam on time
the beam on time
8. Last image hold
9. Cine and spot-film
acquisitions
Use factors
1. Inherent filtration
2. Patient positioning
3. Dose modes
4. Pulse rate
5. Collimation
6. Mag modes • Only use Cine and spot-
7. Beam on time film modes when
8. Last image hold diagnostic quality images
9. Cine and spot-film are needed
acquisitions
• Newer systems allow the
LIH to be sent to PACS
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 20
11/21/19
II vs. FPD
Resolution:
• II system the resolution improves as magnification increases (geometric
magnification); limited by video output device
• FPD system the resolution does not change with FOV; pixelated detector size
determines resolution
Exposure rate:
• II system the exposure rate will increase when increasing magnification mode (as
result of ABC)
• FPD system no need to increase exposure rate for smaller FOVs, although often time
they will to reduce image noise
Geometric distortions:
• FPD will not suffer from pincushion and s-distortion
Low exposure rates:
• II systems currently outperform FPDs at low dose rates
• Due to brightness gain in the II chain
• FPDs can perform pixel binning; reduction in spatial resolution 42
Outline
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 21
11/21/19
Intra-fleet performance
• AKR Accuracy Insufficient
– System A: 25 mGy/min
– System B: 50 mGy/min
– Both systems can pass accuracy checks, but deliver significantly
different dose!
• Remove equipment choice from the equation
– Why should dose depend on choice of System A vs System B?
Intra-fleet Consistency
Normal Fluoro - AKR
50
45
Air Kerma Rate (mGy / min)
40
35
30
25
20
15
10
0
6" 8" 10" 12"
Acrylic
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 22
11/21/19
Inter-Fleet Consistency
Outline
1. Prospective system selection and assurance
– Intrinsic performance and specification evaluation
– Operational performance evaluation
2. Task-based system characterization and optimization
– Prospective use optimization
3. Assurance of system quality in practice
– System performance monitoring
– Patient-based performance monitoring
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 23
11/21/19
Stay tuned …
Take-home points
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 24
11/21/19
Take-home points
• Fluoroscopy is a complex modality with
inherent risk of deterministic radiation injuries
• Risks can be minimized by understanding
equipment capabilities and establishing
quantifiable performance standards
• Proper education and communication ensures
that patient safety with positive outcomes
remains the primary goal of our practice
Take-home points
• Modern clinical fluoroscopy physics requires
active clinical participation in
– Designing precise performance metrics
– Task-specific optimization
– Integrated, patient-centric optimization
– Retrospective audit towards operational quality
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 25
11/21/19
Take-home points
• Thinking physicists needed to address the
new needs of new technologies
– New temporal techniques
– 2D+ and cone beam systems
– Targeted systems: DSA, dental, MSK, Neuro
– II to flat panel (noise, resolution, mag,
temporal filtering implications)
Questions?
(c) Ehsan Samei, 2019. Use for non-personal
purposes by prior permission only. 26
