Fluoroscopic Imaging Equipment Fluoroscopy Evaluation · tenti on) Rec ursi fil ve t e r s ett i ng...

1

Fluoroscopic Imaging EquipmentGuidelines for Detector Input Dose Settings

and Image Optimization

Phil RauchHenry Ford Health Systems

Detroit, MI2009 AAPM Meeting-Phil Rauch

Entrance Air Kerma for 22 cm FOV(Measured at Interventional Ref Point)

0

20

40

60

80

100

120

140

46

810Nominal Phantom Thickness (inches)

Pa

tien

t E

ntr

ance

Air

Ker

ma

Ra

te (

mG

y/m

in)

No Grid-70 kV 10R AF 2kW 4ms/7msmicrofocusNote: Thi s one at 40 nGy/pulse and 10 pps100kV 10R A F 2kW smallfocus

No Grid-70kV 10R A F 2kW 4ms/7msmicrofocusNote: Thi s one at 23 nGy/pulse and 30 pps80kV 10R 0302 3kW smallfocusNote: Thi s one at 36 nGy/pulse and 15 ppsso meas. Dose was mul tiplied by 2Intervent ional IQ 2

80kV 10R 0302 2kW smallfocusNote: Thi s one at 36 nGy/pulse and 15 ppsso meas. Dose was mul tiplied by 270kV 10R A F 2kW 5ms/8ms microfocus

80kV 10R A F 2kW smallfocus

80kV 10R 02 3kW small focus

Intervent ional IQ 1 (no DHHS)

70kV 10R A F 2kW microfocus

70kV Service 02

70kV 10R A F 3kW 6ms/9ms small focus

70kV 10R A F 3kW 8ms/12ms smallfocus





70kV 20R A F 3kW 8ms/12ms smallfocus

70kV Service 00

RELATIVE EERD vs kVp

80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

210%

40 50 60 70 80 90 100 110

kVp

Det

ecto

r E

xpo

sure

Rat

e

No

rmal

ized

to

Min

imu

m

Shimadzu Bransist - 22 cm FOV Direct

Flat Panel Detector (Se/TFT)

Philips Xper - 22 cm FOV Indirect Flat

Panel Detector (CsI(Tl)/TFT)

Siemens Artis Zee - 42 cm FOV Indirect

Flat Panel Detector (CsI(Tl)/TFT)

Siemens Siregraph - 22 cm FOV X-ray

Image Intensifier Detector

2009 AAPM Meeting-Phil Rauch

Fluoroscopy Evaluation

�Patient Dose�Image Quality


Dose Reduction vs Image Quality

�Dose reduction depends on….….technology….proper equipment design….proper set up of equipment parameters….proper utilization of the equipment….knowledge and skill of the radiologist


� Image quality depends on….….technology….proper equipment design….proper set up of equipment parameters….proper utilization of the equipment….knowledge and skill of the radiologist

Dose Reduction vs Image Quality

2


What is the exam protocol?

What is the Patient Dose?


What is the Exam Protocol?



0

20

40

60

80

100

120

140

4 6 8 10

Nominal Phantom Thickness (inches)

Pat

ien

t E

ntra

nce

Air

Ker

ma

Rat

e (m

Gy/

min

)

No Grid-70 kV 10R AF 2kW 4ms/7msmicrofocusNote: This one at 40 nGy/pulse and 10 pps100kV 10R AF 2kW smallfocus

No Grid-70kV 10R AF 2kW 4ms/7msmicrofocusNote: This one at 23 nGy/pulse and 30 pps80kV 10R 0302 3kW smallfocusNote: This one at 36 nGy/pulse and 15 ppsso meas. Dose was multiplied by 2Interventional IQ 2

80kV 10R 0302 2kW smallfocusNote: This one at 36 nGy/pulse and 15 ppsso meas. Dose was multiplied by 270kV 10R AF 2kW 5ms/8ms microfocus

80kV 10R AF 2kW smallfocus

80kV 10R 02 3kW smallfocus

Interventional IQ 1 (no DHHS)

70kV 10R AF 2kW microfocus

70kV Service 02

70kV 10R AF 3kW 6ms/9ms smallfocus







70kV Service 00

FDA Limit

Patient dose (and image quality) are highly dependent on the exam protocol setting



0

20

40

60

80

100

120

140

4 6 8 10


Pat

ien

t E

ntra

nce

Air

Ker

ma

Rat

e (m

Gy/

min

)








70kV Service 02








70kV Service 00

Low doses for this protocol may be appropriate for pediatric imaging

FDA Limit

3



0

20

40

60

80

100

120

140

4 6 8 10


Pat

ien

t E

ntra

nce

Air

Ker

ma

Rat

e (m

Gy/

min

)








70kV Service 02








70kV Service 00

Settings optimized for low dose pediatric imaging may not be appropriate for adults

FDA Limit


Image Quality Assessment



Contrast

Noise

Sharpness

Saturation

Artifacts

Descriptors



Edge Detection

Pattern Recognition

Comparison

Assessment

Relevance

Conclusion

Evaluation(HVS)

4


Is Pulse Width Important?


Is Pulse Width Important?


Fluoroscopic Image Quality

Blur (Fram

e

integratio

n

Time)

Image Intensifier

Flat Panel


Exposure ON-time – Pulsed vs Cont

Pulsed Fluoro*(30 pulses per sec)

*Displayed at 7.5 fps

Continuous Fluoro*(30 video frames per sec)

*Displayed at 7.5 fps

5



Frame

integration

time

LAG(Signal retention)

Image Intensifier

Flat Panel



Frame

integration

time


Recursive filter setting

Image Intensifier

Flat Panel


Blur versus Recursive Filtering

Continuous Fluoroscopy Pulsed FluoroscopyWithout Recursive Filtering With Recursive Filtering


Blur versus Recursive Filtering

Continuous Fluoroscopy Pulsed FluoroscopyWithout Recursive Filtering With Recursive Filtering

6



Frame

integration

time


Recursive filter settingPerc

eptual

integrat

ion

Image Intensifier

Flat Panel


Where are the 3’s

Perceptual Integration



30 Frames/sec

Where are the 3’s



8 Frames/sec

Where are the 3’s

7

†EERD = Entrance Exposure Rate to Detector

Beam Alignment

Geometry

SID SSD

FDA Max

FDA 21CFR

Normal Level

Minimum Filtration

Spatial Beam

Shaping

FOVElectronic Zoom

Measurement

Iris (Optical)

Slow Scan

Normal Scan

Video Camera

Detector Characteristics

Display Characteristics

Spatial Resolution

Anti-scatter Grid

Contrast Resolution

Temporal Resolution

Noise

MTF

X-ray Tube

kV-mA Power Curves

X-ray Generator

Continuous Fluoroscopy

Pulsed Fluoroscopy

Fluoroscopy Waveform

Fluoroscopy ON-Time

Last Image Hold

High Level

Pulse Width

Collimation without

Radiation

Pulse Rate

Acquisition Type

Image Size

Wedge Filter

Scatter to Primary Ratio

Dynamic Range

DQE

Luminance

Contrast

Gamma

Iris (Lead)

Spectral Beam

Shaping

Grid Control

Dose Mode

SNR

Patient Size

Spectral Filtration

Patient Exposure

Image Quality

EERD†

kV


Generator Fluoroscopy Technique

Generator Fluoroscopy Power Curves

60

70

80

90

100

110

120

130

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

mA

kVp

Cont. Low - 30 (video) fps




60

70

80

90

100

110

120

130

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

mA

kVp Cont. Norm - 30 (video) fps





60

70

80

90

100

110

120

130

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

mA

kVp

Pulsed - 30 pps

Cont. Norm - 30 (video) fps


8




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

N o rm a l D o s e P uls e d - 3 .7 5 pp s / 0 .2 m m C u S pe c tra l F ilte r




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

Hig h D o s e P u ls e d - 3 .7 5 p ps / 0 .2 m m C u S p e c t ra l F ilte r

N o rm a l D o s e P uls e d - 3 .7 5 pp s / 0 .2 m m C u S pe c tra l F ilt e r




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

N o rm a l D o s e P u ls e d - 15 p ps / 0 .2 m m C u S p e c t ra l F ilte r




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

H ig h P uls e d - 15 pp s / 0 .2 m m C u S pe c t ra l F ilte r

N o rm a l D o s e P u ls e d - 15 p ps / 0 .2 m m C u S p e c t ra l F ilte r

9




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

Normal Dose Continuous / 0.2mm Cu Spectral Filter




50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45

mA

kVp

High Dose Continuous / 0.2mm Cu Spectral Filter)

Normal Dose Continuous / 0.2mm Cu Spectral Filter


Entrance Exposure Rate to Detector

Solid state detectors must be positioned outside the automatic dose rate measurement field


EERD Test Conditions


80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

210%

40 50 60 70 80 90 100 110

kVp

Det

ecto

r E

xpos

ure

Rat

e N

orm

aliz

ed to

Min

imum

Siemens Siregraph - 22 cm FOV X-rayImage Intensifier Detector

10




80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

210%

40 50 60 70 80 90 100 110

kVp

Det

ecto

r E

xpos

ure

Rat

e N

orm

aliz

ed to

Min

imum Siemens Artis Zee - 42 cm FOV Indirect

Flat Panel Detector (CsI(Tl)/TFT)





80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

210%

40 50 60 70 80 90 100 110

kVp

Det

ecto

r E

xpos

ure

Rat

e N

orm

aliz

ed to

Min

imum

Philips Xper - 22 cm FOV Indirect FlatPanel Detector (CsI(Tl)/TFT)

Siemens Artis Zee - 42 cm FOV IndirectFlat Panel Detector (CsI(Tl)/TFT)





80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

210%

40 50 60 70 80 90 100 110

kVp

Det

ecto

r E

xpos

ure

Rat

e N

orm

aliz

ed to

Min

imum

Shimadzu Bransist - 22 cm FOV DirectFlat Panel Detector (Se/TFT)

Philips Xper - 22 cm FOV Indirect FlatPanel Detector (CsI(Tl)/TFT)

Siemens Artis Zee - 42 cm FOV IndirectFlat Panel Detector (CsI(Tl)/TFT)




EERD for Different Attenuators

0

50

100

150

200

250

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0

Attenuator Thickness

EE

RD

(m

icro

-R p

er s

ec)

Acrylic (inches)

Aluminum (inches)

Copper (mm)

11


Image Quality: What is it?

� “Image quality depends only on intrinsic, objective physical characteristics of an imaging system, and can be measured independently of an observer”

Definitions courtesy Ralph Schaetzing, Agfa Corp.


Detector Metrics Image Intensifier

Flat Panel

Detector Input Dose Rates (FP)


Indirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPD

Image Detector


Pixel Pitch – Estimated Resolution

Pixel Pitch

Detector Mfg.PixelPitch

Nyquist (lp/mm)

Nyquist x 2^(.5)(lp/mm)

Mesh Lines per Inch at 45-degrees

Safire (a-Se, a-Si TFT) Shimadzu 0.150 3.33 4.71 120Philips, Siemens (CsI(Tl), a-Si TFT) - Angio Trixell 0.154 3.25 4.59 117

Philips, Siemens (CsI(Tl), a-Si TFT) - Cardiac Trixell 0.184 2.72 3.84 98GE, Revolution (CsI(Tl), a-Si TFT) GE 0.200 2.50 3.54 90

12


Actual Resolution

Archived 512 x 512 Archived 1024 x 1024



� Detroit Phantom





� FOV Coverage� Motion blur� Resolution� Contrast to noise� Comparison with other

fluoroscopy equipment� Rotational Angiography� 3D Reconstruction� Recursive Filter effects� Image processing effects� Pixel matrix effects

13



� FOV Coverage � Motion blur� Resolution� Contrast to noise� Comparison with other

fluoroscopy equipment� Rotational Angiography� 3D Reconstruction� Recursive Filter effects� Image processing effects� Pixel matrix effects



Rotational Angiography Digital Radiograph



Rotational Angiography Digital Radiograph


3-D Reconstruction


Digital Radiograph

14


Image Detector (Fluoro LIH)

� Can you match the fluoroscopy LIH image to the detector?

� A. Image Intensifier� B. Indirect Flat Panel

� C. Direct Flat Panel


Image Detector (Fluoro LIH)Indirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPD

?



Car Moving



Car Stationary

15


Resolution

Can you see a 50µ moving wire using a flat panel detector with a pixel pitch of 150µ? With a pixel pitch of 200µ?


Resolution

Direct Capture, 150µ pixel pitch Indirect Capture, 200µ pixel pitch

Wire size numbers are located below the corresponding wire


Image Detector (Fluoro LIH)Indirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPDIndirect Conversion FPD

Cannot rely on reported detector metrics alone. Must understand and appropriately utilize optimum technical settings and image processing


Detector Input Dose Rates (XRII)

Dose Rate Selection

Field of View Selection

16



40 cm 30 cm 22 cm 17 cm

Low

Dose Rate Selection


Norm High



30 cm

Low

Dose Rate Selection


Continuous Fluoroscopy: Input dose per frame of fluoroscopy?



30 cm

• Low Dose Mode• Cont. Fluoroscopy• 30 cm FOV

Detector Input Dose/Fr?

A. 0.37 µR (3.2 nGy)B. 0.75 µR (6.6 nGy)C. 1.5 µR (13.1 nGy)D. 3.0 µR (26.3 nGy)E. 6.0 µR (52.6 nGy)F. No Clue

Low

……values assume FDA minimum beam filtration



30 cm




Low


17







……can use 2x these values if minimum 0.2 mm Cubeam filtration is utilized




Detector Input Dose/Fr?1x 1.5x 3.0x




Detector Input Dose/Fr?1x 1.5x 3.0x

Detector dose scales inversely as either the ratio of the FOV, or as the ratio squared

40 cm 30 cm 22 cm 17 cm


Video Frame Rate

What is the video frame rate in the U.S.A.?

Ans: 30 video frames per second

18


� Rauch's 30-30-30 Rule:� For a nominal 30 cm image intensifier field of view (FOV),

Pulse rate of 30 pps, Set the EERD to 30 µR/sec

� For other FOV's, scale by either the ratio of the FOV (auto optical lens aperture) or the square of the ratio

� For spectral beam filtration…� 0.1 mm Cu equivalent, multiply by 1.41

� 0.2 mm Cu equivalent, multiply by 2


For flat panel, special consideration is required!!


Image Intensifier

Flat Panel


Detector Metrics(Fluoroscopy Doses)

Adapted from Koch, A., Macherel, J.M., Wirth, T., de Groot, P., Ducourant, T., Couder, D., Moy, J.P., & Calais, E. (2001). Detective quantum efficiency of an X-ray image intensifier chain as a benchmark for amorphous silicon flat panel detectors, Proc. SPIE, 4320, 115-20


Image Intensifier

Flat Panel


Detector Metrics(Fluoroscopy Doses)

“The minimum operating dose level is defined by the magnitude of the noise arising in the AM array and readout electronics. This typically sets the lower operating limit for detector dose at 20-50 nGy.”Cowen AR et al., Solid-state, flat-panel, digital radiography detectors and their physical imaging characteristics, Clin Radiol (2008)


Pixel Pitch – Air Kerma per Pixel

Pixel Pitch

For the transition from XRII to Flat Panel, the required EERD can be estimated by determining the air kerma required for the XRII when it is operating at the same pixel pitch at that for the flat panel.

Detector Mfg.PixelPitch

Safire (a-Se, a-Si TFT) Shimadzu 0.150Philips, Siemens (CsI(Tl), a-Si TFT) - Angio Trixell 0.154

Philips, Siemens (CsI(Tl), a-Si TFT) - Cardiac Trixell 0.184GE, Revolution (CsI(Tl), a-Si TFT) GE 0.200

19


Pixel Pitch vs Input Dose

*Per Rauch’s 30-30-30 rule

Pixel Pitch(mm)

EquivalentXRII Diam (cm)for 1024 Pixels Diam^2

Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*




Pixel Pitch(mm)


Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*




Pixel Pitch(mm)


Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*




Pixel Pitch(mm)


Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*

20




Pixel Pitch(mm)


Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*




Pixel Pitch(mm)


Ratio wrt 30 cm FOV

EERD(µR)

EERD(nGy)

XRII0.293 30.00 900.00 1.00 1.00 8.76

Flat Panel0.150 15.36 235.93 3.81 3.81 33.420.154 15.77 248.68 3.62 3.62 31.700.184 18.84 355.01 2.54 2.54 22.210.200 20.48 419.43 2.15 2.15 18.80

*


Detector Input Dose Rates

1x 1.5x 3.0x

What about pulsed fluoroscopy?


� For pulse rates lower than 30 pps, use Aufrichtig* perceptual pulsed fluoroscopy scale factor to adjust dose per frame for different frame rates

*Perceptual comparison of pulsed and continuous fluoroscopy, Richard Aufrichtig, Med. Phys. 21 (2), February 1994

Dose Per Pulse (<30 pps)

(Dose/Pulse)2 = (Dose/Pulse)1*SQRT[(Pulse Rate)1/(Pulse Rate)2]

Example: Change from 30pps to 7.5 pps

(Dose/Pulse)2 = (Dose/Pulse)1*SQRT[30/7.5]

= (Dose/Pulse)1*2

21


� For pulse rates lower than 30 pps, use Aufrichtig* perceptual pulsed fluoroscopy scale factor to adjust dose per frame for different frame rates

*Perceptual comparison of pulsed and continuous fluoroscopy, Richard Aufrichtig, Med. Phys. 21 (2), February 1994

Dose Per Pulse (<30 pps)

For pulse rate of 5 per second or lower, use the Aufrichtig calculated value for 5 pps

(The retina can only retain an image for 200 ms, so there is no integration of noise for pulse rates of 5 pps or lower)

C-arm Dose Rates


� Mobile C-arm Operational Settings

� XRII Imaging System (31cm FOV)� Filtration equiv to 0.1 mm Cu� Fluoro modes:

� Cont., 8, 4, 2, 1 pps� Ability to select “Low” dose� Two foot pedals:

– Left: Normal Fluoro– Right: Boost Fluoro

� Operator can store 0 to 30 frames/sec (maxed at selected pulse rate

� Acquisition Modes� 15, 30 pps� Ability to select “Low dose”� Ability to use the two foot pedals� All frames recorded

C-arm Dose Rates

314374Default dose Boost Pulsed

13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose

Non-boost Continuous

559Low dose

Boost Continuous

376Low dose


Max Dose Rate(mGy/min)

EERD 1(nGy/pulse)Operational Mode

1 EERD per pulse is the same for all pulse rates2 Data for 8pps & 4pps; scales with pulse rate below 4pps3 Data for 30pps; scales with pulse rate, except 8pps and 4pps have the same max rate due to change in pulse width2009 AAPM Meeting-Phil Rauch

� Mobile C-arm Operational Settings

� XRII Imaging System (31cm FOV)� Filtration equiv to 0.1 mm Cu� Fluoro modes:

� Cont., 8, 4, 2, 1 pps� Ability to select “Low” dose� Two foot pedals:

– Left: Normal Fluoro– Right: Boost Fluoro

� Operator can store 0 to 30 frames/sec (maxed at selected pulse rate

� Acquisition Modes� 15, 30 pps� Ability to select “Low dose”� Ability to use the two foot pedals� All frames recorded

� EERD and Max Dose Rate (FDA)

C-arm Dose Rates


13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose


559Low dose

Boost Continuous

376Low dose





� Dose Rate Analysis

� Are these values acceptable?

� Values exceed the FDA limits of 176 mGy/min for high dose fluoroscopy

� EERD dose not change with pulse rate, so the same detector dose is delivered at 1pps as at 30pps

� Appears to be pulsed fluoroscopy….

� ….but the 15pps and 30pps are labeled “Cine”, and all frames are recorded.

� The FDA limits do not apply


22




13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose


559Low dose

Boost Continuous

376Low dose





� EERD Analysis� Using Rauch 30-30-30 rule,

adjustment for filtration, and Aufrichtig scale, expect 11.6 nGy per pulse for 30 pps†

� Measured default non-boost values are in agreement for 30pps

� However, for pulsed mode the dose does not change as the pulse rate is reduced (Does not follow the Aufrichtig scale)

� Result is too low a dose for the lower pulse rates, which may discourage operators from using the lower rates

†8.76nGy*(30/31)2*SQRT(30/30)*1.41




13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose


559Low dose

Boost Continuous

376Low dose





� Analysis� Expect low dose mode to be about

1/2 to 2/3 of the default dose

� Measured non-boost values are in agreement




13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose


559Low dose

Boost Continuous

376Low dose





� Analysis� Expected values for default boost

pulsed would be 2x to 3x the non-boost values

� Measured default boost pulsed values are 6x larger than non-boost




13212Default dose

Non-boost Pulsed

283337Low dose

Boost Pulsed

826Low dose

Non-boost Pulsed

14822Default dose

Boost Continuous

6712Default dose


559Low dose

Boost Continuous

376Low dose





� Analysis� Using Rauch 30-30-30 rule,

adjustment for filtration, and Aufrichtig scale, expect 28.4 to 56.8 nGy per pulse for boost at 5 pps or lower†

� Measured default boost value is much larger than this estimate

� Furthermore, the dose does not change as the pulse rate is increased (Does not follow the Aufrichtig scale)

� Result is much larger than necessary doses for the higher frame rates

†8.76nGy*(30/31)2*SQRT(30/5)*2

23


Rauch’s 30-30-30 Rule

� Use the previous type analysis to determine how the vendor specified or measured EERD compares with the expected values

� Use the rule to compare measured values with the expected change in EERD whenever there are variations in the field of view, filtration, or pulse rate




� “Image quality is whatever the observer says it is (i.e., it is a subjective perception of the image, ‘in the eye of the beholder’)”



Human Visual Perception


Biliary endoscopic catheterization

Mobile C-arm

Gastroenterology clinic (No X-ray techs)

1024 x 1280 x 12 bit image

Secondary capture of fluoro frame

24

Observations

Mottled

Blurred, probably due to long pulse width

Low dynamic range

Lack of detail

Inappropriate anatomical setting

No x-ray technique or image processing information available in the DICOM metadata

Fluoro Frame Fluoro Frame8-Bit Grayscale 8-Bit Grayscale

Flat Detector XRII512 x 512 Matrix 768 x 576 Matrix

Modality Image downsized from 10242 Digital Video Screen Capture via DVR


Flat Detector XRII1024 x 1024 Matrix 768 x 576 Matrix

Modality Store Monitor Image Digital Video Screen Capture via DVR


1024 x 1024 Matrix 884 x 884 MatrixFlat Detector XRII

Modality Store Monitor Image Screen Capture via Paxport

25


Dynamic Range

Fluoroscopy – Vascular ModeTrixell Pixium 4700 (0.154 µ; 2480 x 1910 matrix

Digital RadiographyTrixell Pixium 4600 (0.143 µ; 3001 x 3001 matrix

Fluoroscopy – Vascular ModeTrixell Pixium 4700 (0.154 µ; 2480 x 1910 matrix

Fluoroscopy – Cardiac ModeTrixell Pixium 4700 (0.154 µ; 2480 x 1910 matrix


Video Bandwidth

•Three display monitors provide with installation(Live fluoroscopy, Roadmap, Reference)

•A single video graphics display card is used to simultaneously paint all three images and graphics

•Result is a loss of resolution when compared with the display of single images on a modality workstation or on PACS

26


Test Your Knowledge


Test Your Knowledge

Question: Name two ways the vessel contrast can appear as in this image

Answer:1. Start injection of

contrast well before starting the run

2. Change the mask to a frame after the contrast has filled some vessels


Test Your KnowledgeWhy are these image so different?


Test Your KnowledgeWhy are these image so different?

72 kVp, 250 mA, 8 ms1024 x 1024 x 8 bits

65 kVp, 15 mA, 13 ms512 x 512 x 8 bits

27


Image Quality Analysis

� Spectral filtration� Bolus filters� Generator power curves� Pulse width� Grid vs Non-grid� Collimation� Detector dose per

image� Anatomical program

selection� Recursive filter setting

� Edge enhancement� Harmonization (Dynamic

Range(� Pixel matrix and bit depth

for acquisition� Pixel matrix and bit depth

stored� Pixel matrix and bit depth

archived




� “Image quality is whatever the observer says it is (i.e., it is a subjective perception of the image, ‘in the eye of the beholder’)”

� “Image quality is defined by an observer's ability to achieve an acceptable level of performance for a specified task”



Image Interpretation

Successful interpretation depends on…� Image Attributes�Human Observer (education, experience)�Human Visual Perception�Pattern Recognition�Clarity of the imaging task�Minimal external distractions�Level of confidence


Conclusions

� Image quality is not just detector metrics

� Image quality is not just a subjective impression

� Image quality for fluoroscopy must be assessed with moving objects in order to get a complete understanding of performance

� Use pulsed fluoroscopy and the Aufrichtig scale to produce better images, not solely to reduce dose

28


Final Imaging Task

Why is there an extra square?

How can this be true?

Below the four parts

are re-ordered

The four parts are

exactly the same as

those used above

Thank youPhil Rauch

[email protected]

Date post:	12-Oct-2018
Category:	Documents
Upload:	lecong
View:	215 times
Download:	0 times

Fluoroscopic Imaging Equipment Fluoroscopy Evaluation · tenti on) Rec ursi fil ve t e r s ett i ng...

Documents