1 Image acquisition, display, & interpretation Image - - PDF document

SLIDE 1

1

Design and Performance Characteristics

• f Digital Radiographic Receptors

Design and Performance Characteristics Design and Performance Characteristics

• f Digital Radiographic Receptors
• f Digital Radiographic Receptors
• J. Anthony Seibert, Ph.D.

University of California, Davis Medical Center Sacramento, California

• J. Anthony Seibert, Ph.D.
• J. Anthony Seibert, Ph.D.

University of California, Davis Medical Center University of California, Davis Medical Center Sacramento, California Sacramento, California

Learning Objectives Learning Objectives Learning Objectives

• Describe digital detector technologies for

radiography and mammography

• Review functional attributes
• Compare detectors in terms of IQ and dose
• Summarize advantages/disadvantages
• Describe digital detector technologies for

Describe digital detector technologies for radiography and mammography radiography and mammography

• Review functional attributes

Review functional attributes

• Compare detectors in terms of IQ and dose

Compare detectors in terms of IQ and dose

• Summarize advantages/disadvantages

Summarize advantages/disadvantages

Presentation Outline Presentation Outline Presentation Outline

• Acquisition System Overview
• Digital Detector Attributes
• Digital Detector Technologies
• Factors affecting Image Quality & Dose
• Clinical Implementation and QC
• Acquisition System Overview

Acquisition System Overview

• Digital Detector Attributes

Digital Detector Attributes

• Digital Detector Technologies

Digital Detector Technologies

• Factors affecting Image Quality & Dose

Factors affecting Image Quality & Dose

• Clinical Implementation and QC

Clinical Implementation and QC

SLIDE 2

2

Detector

Efficiency Resolution Scatter grid DQE

Detector Detector

Efficiency Efficiency Resolution Resolution Scatter grid Scatter grid DQE DQE

Image acquisition, display, & interpretation Image acquisition, display, & interpretation Image acquisition, display, & interpretation

X-rays

kVp mAs Tube filtration Collimation

X X-

• rays

rays

kVp kVp mAs mAs Tube filtration Tube filtration Collimation Collimation

Patient

Size Restraints Exam type ESE, dose

Patient Patient

Size Size Restraints Restraints Exam type Exam type ESE, dose ESE, dose

Human

Radiologist Physician Experience Condition

Human Human

Radiologist Radiologist Physician Physician Experience Experience Condition Condition

Computer

Digitization Preprocessing Postprocessing Configuration

Computer Computer

Digitization Digitization Preprocessing Preprocessing Postprocessing Postprocessing Configuration Configuration

PACS

Data delivery Data display Data storage Workflow

PACS PACS

Data delivery Data delivery Data display Data display Data storage Data storage Workflow Workflow

Acquisition to Interpretation: Image Quality Acquisition to Interpretation: Image Quality Acquisition to Interpretation: Image Quality

• Image quality is an indicator of the relevance
• f information presented in the image to the

task we seek to accomplish using the image

• Considered in terms of portrayal of

– Normal anatomy – Depiction of potential pathology

• Not necessarily the “same” in all images
• A constraining factor is radiation dose
• Image quality is an indicator of the relevance

Image quality is an indicator of the relevance

• f information presented in the image to the
• f information presented in the image to the

task we seek to accomplish using the image task we seek to accomplish using the image

• Considered in terms of portrayal of

Considered in terms of portrayal of

– – Normal anatomy Normal anatomy – – Depiction of potential pathology Depiction of potential pathology

• Not necessarily the

Not necessarily the “ “same same” ” in all images in all images

• A constraining factor is radiation dose

A constraining factor is radiation dose

Image Quality Image Quality Image Quality

• Screen-film radiography

– IQ “built in” to the characteristics of the film – Film is acquisition, display and archive medium – Dose is determined by screen-film speed

• Digital radiography

– IQ dependent on Signal to Noise Ratio (SNR) – Separation of acquisition, display, and archive – Dose is variable and dependent on required SNR

• Screen

Screen-

• film radiography

film radiography

– – IQ IQ “ “built in built in” ” to the characteristics of the film to the characteristics of the film – – Film is acquisition, display and archive medium Film is acquisition, display and archive medium – – Dose is determined by screen Dose is determined by screen-

• film speed

film speed

• Digital radiography

Digital radiography

– – IQ dependent on Signal to Noise Ratio (SNR) IQ dependent on Signal to Noise Ratio (SNR) – – Separation of acquisition, display, and archive Separation of acquisition, display, and archive – – Dose is variable and dependent on required SNR Dose is variable and dependent on required SNR

SLIDE 3

3

Presentation Outline Presentation Outline Presentation Outline

• Acquisition System Overview
• Digital Detector Attributes
• Digital Detector Technologies
• Factors affecting Image Quality & Dose
• Clinical Implementation and QC
• Acquisition System Overview

Acquisition System Overview

• Digital Detector Attributes

Digital Detector Attributes

• Digital Detector Technologies

Digital Detector Technologies

• Factors affecting Image Quality & Dose

Factors affecting Image Quality & Dose

• Clinical Implementation and QC

Clinical Implementation and QC

Gray Scale Gray Scale encoded on encoded on film film Film processing: Film processing: light to optical density light to optical density

Log Relative Exposure Log Relative Exposure Optical Density Optical Density

Conventional screen/film detector Conventional screen/film detector

Film Film Intensifying Screens Intensifying Screens x x-

• rays

rays → → light light Transmitted x Transmitted x-

• rays

rays through patient through patient

• 1. Acquisition, Display, Archiving
• 1. Acquisition, Display, Archiving

Digital X Digital X-

• ray Detector

ray Detector

Transmitted x Transmitted x-

• rays

rays through patient through patient Charge Charge collection collection device device X X-

• ray converter

ray converter x x-

• rays

rays → → electrons electrons

Analog to Digital Analog to Digital Conversion Conversion

• 1. Acquisition
• 1. Acquisition
• 2. Display
• 2. Display
• 3. Archiving
• 3. Archiving

Digital Digital processing processing Digital to Analog Digital to Analog Conversion Conversion

Digital Pixel Digital Pixel Matrix Matrix

SLIDE 4

4

Exposure Latitude Exposure Latitude

Log relative exposure Log relative exposure Signal output Signal output

10000:1 10000:1 Digital Digital

Spatial Resolution Spatial Resolution

Analog versus Digital Analog versus Digital

MTF of pixel aperture (DEL)

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 2 3 4 5 6 7 8 9 10 11 Frequency (lp/mm) M odulation 1000 µm 200 µm 100 µm

Detector Detector Element, Element, “ “DEL DEL” ” Sampling Sampling Pitch Pitch

Film Film 100:1 100:1

Digital Detectors Digital Detectors Digital Detectors

• Separation of acquisition, display and archive
• Digital acquisition is not contrast limited

– Image processing

• Signal to Noise Ratio (SNR) and Contrast to

Noise Ratio (CNR) impacts “image quality”

• Detector DQE determines the exposure required

to achieve a required SNR

• Separation of acquisition, display and archive

Separation of acquisition, display and archive

• Digital acquisition is

Digital acquisition is not contrast limited not contrast limited

– – Image processing Image processing

• Signal to Noise Ratio (SNR) and Contrast to

Signal to Noise Ratio (SNR) and Contrast to Noise Ratio (CNR) impacts Noise Ratio (CNR) impacts “ “image quality image quality” ”

• Detector DQE determines the exposure required

Detector DQE determines the exposure required to achieve a required SNR to achieve a required SNR

Digital Detectors Digital Detectors Digital Detectors

• Sampling and quantization (new noise sources)
• Detector pre-processing (correct imperfections)
• Image post-processing (enhance image contrast)
• Sampling and quantization (new noise sources)

Sampling and quantization (new noise sources)

• Detector pre

Detector pre-

• processing (correct imperfections)

processing (correct imperfections)

• Image post

Image post-

• processing (enhance image contrast)

processing (enhance image contrast)

SLIDE 5

5

Presentation Outline Presentation Outline Presentation Outline

• Acquisition System Overview
• Digital Detector Attributes
• Digital Detector Technologies
• Factors affecting Image Quality & Dose
• Clinical Implementation and QC
• Acquisition System Overview

Acquisition System Overview

• Digital Detector Attributes

Digital Detector Attributes

• Digital Detector Technologies

Digital Detector Technologies

• Factors affecting Image Quality & Dose

Factors affecting Image Quality & Dose

• Clinical Implementation and QC

Clinical Implementation and QC

Digital Detector Technologies Digital Detector Technologies Digital Detector Technologies

• Photostimulable Storage Phosphor (PSP or CR)
• Charge Coupled Device (CCD)
• Complementary MetalOxide Semiconductor (CMOS)
• Thin-Film-Transistor array (TFT)
• Photon counters (not discussed)
• Photostimulable Storage Phosphor (PSP or CR)

Photostimulable Storage Phosphor (PSP or CR)

• Charge Coupled Device (CCD)

Charge Coupled Device (CCD)

• Complementary

Complementary MetalOxide MetalOxide Semiconductor (CMOS) Semiconductor (CMOS)

• Thin

Thin-

• Film

Film-

• Transistor array (TFT)

Transistor array (TFT)

• Photon counters (not discussed)

Photon counters (not discussed)

Computed Radiography (CR) Computed Radiography (CR) Computed Radiography (CR)

...is the generic term applied to an imaging system comprised of:

Photostimulable Storage Phosphor

to acquire the x-ray projection image

CR Reader

to extract the electronic latent image

Digital electronics

to convert the signals to digital form ...is the generic term applied to an imaging system ...is the generic term applied to an imaging system comprised of: comprised of:

Photostimulable Storage Phosphor Photostimulable Storage Phosphor

to acquire the x to acquire the x-

• ray projection image

ray projection image

CR Reader CR Reader

to extract the electronic latent image to extract the electronic latent image

Digital electronics Digital electronics

to convert the signals to digital form to convert the signals to digital form

SLIDE 6

6

CR Detector CR Detector CR Detector

• Photostimulable Storage Phosphor (PSP)
• Direct replacement for S/F; positioning flexibility
• Photostimulable Storage Phosphor (PSP)

Photostimulable Storage Phosphor (PSP)

• Direct replacement for S/F; positioning flexibility

Direct replacement for S/F; positioning flexibility Phosphor Plate Phosphor Plate Phosphor Plate Cassette Holder Cassette Holder Cassette Holder

CR Image Acquisition CR Image Acquisition CR Image Acquisition

Phosphor plate Phosphor plate X X-

• ray

ray system system

1.

• 1. X

X-

• ray Exposure

ray Exposure Image Image Scaling Scaling

3. 3.

Image Image Record Record

4. 4.

Patient Patient Computed Computed Radiograph Radiograph

5. 5.

unexposed unexposed

Image Image Reader Reader

2. 2.

exposed exposed

Display / Archive Display / Archive

Laser film printer Laser film printer DICOM / PACS DICOM / PACS

Image Acquisition Image Acquisition

Latent image produced Latent image produced

CR CR Reader Reader

Latent Latent image image extracted extracted

CR QC CR QC Workstation Workstation

Patient information

SLIDE 7

7

Photostimulated Luminescence Photostimulated Luminescence

Conduction band Conduction band Valence band Valence band PSL PSL 3.0 eV 3.0 eV

τ τ

Eu Eu

Eu Eu 2+

2+

Eu Eu 3+

3+ /

/

4f 4f 7

7

8.3 eV 8.3 eV

Laser Laser stimulation stimulation

2.0 eV 2.0 eV F/F F/F+

+

PSLC complexes (F centers) are PSLC complexes (F centers) are created in numbers proportional to created in numbers proportional to incident x incident x-

• ray intensity

ray intensity

e e-

• τ

τ

tunneling tunneling

τ τ recombination

recombination

4f 4f 6

6 5d

5d

phonon phonon

CR: How does it work? CR: How does it work?

Incident Incident x x-

• rays

rays

e e

Energy Energy Band BaFBr Band BaFBr

300 300 400 400 500 500 600 600 700 700 800 800

Relative intensity Relative intensity

0.0 0.0 0.5 0.5 1.0 1.0

λ λ (nm)

(nm) Energy (eV) Energy (eV) 3 3 4 4 2.5 2.5 2 2 1.5 1.5

Stimulation and Emission Spectra Stimulation and Emission Spectra

1.75 1.75 Emission Emission Stimulation Stimulation Diode Diode 680 nm 680 nm

BaFBr: Eu BaFBr: Eu2+

2+

Optical Optical Barrier Barrier

Photostimulated Luminescence Photostimulated Luminescence

Incident Laser Incident Laser Beam Beam

PMT PMT

Protective Layer Protective Layer Phosphor Layer Phosphor Layer Base Support Base Support

Light Light Scattering Scattering Laser Light Spread Laser Light Spread Photostimulated Photostimulated Luminescence Luminescence

"Effective" readout diameter "Effective" readout diameter

Exposed Exposed Imaging Imaging Plate Plate Light guide Light guide

PSL PSL Signal Signal

SLIDE 8

8

CR: Latent Image Readout CR: Latent Image Readout

PMT PMT Polygonal Polygonal Mirror Mirror Laser Laser Source Source Light channeling guide Light channeling guide

Output Signal Output Signal

Reference Reference detector detector Cylindrical mirror Cylindrical mirror f f-

• θ

θ lens lens ADC ADC

Laser beam: Laser beam: Scan direction Scan direction Plate translation: Plate translation: Sub Sub-

• scan direction

scan direction

To image To image processor processor ADC ADC

x= 1279 x= 1279 y= 1333 y= 1333 z= 500 z= 500

Scan Direction Scan Direction Sub Sub-

• scan Direction

scan Direction

Laser beam deflection Laser beam deflection Plate translation Plate translation Typical Typical CR resolution: CR resolution:

35 x 43 cm 35 x 43 cm --

• - 2.5 lp/mm (200

2.5 lp/mm (200 µ µm) m) 24 x 30 cm 24 x 30 cm --

• - 3.3 lp/mm (150

3.3 lp/mm (150 µ µm) m) 18 x 24 cm 18 x 24 cm --

• - 5.0 lp/mm (100

5.0 lp/mm (100 µ µm) m)

Screen/film resolution: Screen/film resolution:

7 7-

• 10 lp/mm (80

10 lp/mm (80 µ µm m -

• 25

25 µ µm) m)

Phosphor Plate Cycle Phosphor Plate Cycle

PSP PSP Base support Base support

reuse reuse

plate erasure: plate erasure: remove residual signal remove residual signal light erasure light erasure plate exposure: plate exposure: create latent image create latent image x x-

• ray exposure

ray exposure plate readout: plate readout: extract latent image extract latent image laser beam scan laser beam scan

SLIDE 9

9

CR Innovations CR Innovations CR Innovations

• High-speed line scan systems (<10 sec)
• Dual-side readout capabilities (increase DQE)
• Structured phosphors
• Mammography applications ????
• Low cost table-top CR readers
• High

High-

• speed line scan systems (<10 sec)

speed line scan systems (<10 sec)

• Dual

Dual-

• side readout capabilities (increase DQE)

side readout capabilities (increase DQE)

• Structured phosphors

Structured phosphors

• Mammography applications ????

Mammography applications ????

• Low cost table

Low cost table-

• top CR readers

top CR readers

5 sec scan

Laser Line Source Shaping Lens Linear CCD Array Lens Array Line excitation PSL

Sub-scan Direction

CR CR “ “line line-

• scan

scan” ”

Side View

Linear Laser Source Light Collection Lens Linear CCD Array Stationary IP

Charge Coupled Device: CCD Charge Coupled Device: CCD Charge Coupled Device: CCD

• X-rays on scintillator
• Light collection: optical or fiberoptical
• Emitted light to CCD photo-sensor
• Electronic charge created on silicon
• Charge transfer moves packets
• Charge to voltage conversion & amplification
• X

X-

• rays on scintillator

rays on scintillator

• Light collection: optical or

Light collection: optical or fiberoptical fiberoptical

• Emitted light to CCD photo

Emitted light to CCD photo-

• sensor

sensor

• Electronic charge created on silicon

Electronic charge created on silicon

• Charge transfer moves packets

Charge transfer moves packets

• Charge to voltage conversion & amplification

Charge to voltage conversion & amplification

SLIDE 10

10

CCD Charge Collection CCD Charge Collection CCD Charge Collection

Transparent Transparent polysilicon polysilicon “ “gate gate” ” electrode electrode Silicon dioxide Silicon dioxide Silicon substrate Silicon substrate Photo Photo-

• generated

generated electrons electrons

• V

Vo

• V

Vo

• +V

+Vo

• Potential

Potential Well Well Potential Potential Barrier Barrier Potential Potential Barrier Barrier Light photons Light photons e e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• e

e-

• CCD charge transfer

CCD charge transfer CCD charge transfer

• Voltage potential of gate electrode pushes

electrons towards collection amplifiers

• 24 volts bias for good transfer efficiency
• Larger pixel dimension has inefficient transfer
• Pixel dimensions:

– 8, 10, 15, 20 micron

• Form factor of CCD array very small
• Voltage potential of gate electrode pushes

Voltage potential of gate electrode pushes electrons towards collection amplifiers electrons towards collection amplifiers

• 24 volts bias for good transfer efficiency

24 volts bias for good transfer efficiency

• Larger pixel dimension has inefficient transfer

Larger pixel dimension has inefficient transfer

• Pixel dimensions:

Pixel dimensions:

– – 8, 10, 15, 20 micron 8, 10, 15, 20 micron

• Form factor of CCD array very small

Form factor of CCD array very small

Light emission & Optical coupling Light emission & Optical coupling Light emission & Optical coupling

Lens Scintillator

CCD CCD Detector Detector

Large loss of light!!! Demagnification >10:1

Optical coupling inefficiency Optical coupling inefficiency

X X-

• rays

rays Light Light

SLIDE 11

11

CCD detector CCD detector CCD detector

• Silicon chip with photosensitive layer
• Silicon chip with photosensitive layer

Silicon chip with photosensitive layer

35 cm x 43 cm

2.5 cm 2.5 cm

Optical de-magnification Lens efficiency? Secondary Quantum Sink High fill factor ~ 100 % Good light conversion efficiency (~85%)

5 cm 5 cm

Larger CCD Larger CCD arrays arrays Scintillator Scintillator

CCD acquisition and readout CCD acquisition and readout CCD acquisition and readout

Pixel Silicon dioxide Electrode CCD Horizontal Readout Register Masking

Parallel Parallel clocks clocks Serial clocks Serial clocks Serial Register Serial Register

CMOS

Complementary Metal Oxide Semiconductor

CMOS CMOS

Complementary Metal Oxide Semiconductor Complementary Metal Oxide Semiconductor

• “RAM” with photodiode converter
• Random access readout
• Low voltage operation (5V)
• ? NOISE ……
• Large FOV detector available (tiled CMOS)
• High sampling resolution possible
• “

“RAM RAM” ” with photodiode converter with photodiode converter

• Random access readout

Random access readout

• Low voltage operation (5V)

Low voltage operation (5V)

• ? NOISE

? NOISE …… ……

• Large FOV detector available (tiled CMOS)

Large FOV detector available (tiled CMOS)

• High sampling resolution possible

High sampling resolution possible

SLIDE 12

12

CMOS detector on a chip CMOS detector on a chip

LATCHES COUNTER DECODER ROW DRI VERS COLUMN SI GNAL CONDI TI ONI NG DECODER COUNTER LATCHES

TI MI NG AND CONTROL PI XEL ARRAY

CLK RUN DEFAULT LOAD ADDRESS DATA + 5V VS_OUT VR_OUT READ FRAME

“Tiled” matrix array of CMOS in large FOV Available:

Array of tiled CMOS sensors Array of tiled CMOS sensors Array of tiled CMOS sensors

Xrays Xrays Scintillator Scintillator Fiberoptic plate Fiberoptic plate Microlens Microlens optics

• ptics

CMOS sensors CMOS sensors Controller electronics Controller electronics 7000x7000 element array, 17 7000x7000 element array, 17” ” x 17 x 17” ” FOV for one implementation FOV for one implementation

Dead Zone

Amorphous Silicon Amorphous Silicon

TFT TFT active matrix active matrix array array

Thin Thin-

• Film

Film Transistor Transistor Storage Storage Capacitor Capacitor Charge Charge Collector Collector Electrode Electrode G1 G1 G2 G2 G3 G3 Gate Gate switches switches

D2 D2 D1 D1 D3 D3

Data lines Data lines

CR2 CR2 CR3 CR3 CR1 CR1 Charge Charge Amplifiers Amplifiers Analog to Analog to Digital Digital Converters Converters

Amplifiers – Signal out

Active Area

SLIDE 13

13

Amorphous Silicon Amorphous Silicon

TFT TFT active matrix active matrix array array

G1 G1 G2 G2 G3 G3

Expose to x Expose to x-

• rays

rays Store the charge Store the charge Active Readout Active Readout Activate gates Activate gates Amplify charge Amplify charge Convert to Digital Convert to Digital

Amplifiers – Signal out

Indirect detector: Indirect detector:

a a-

• Si

Si TFT/ CsI phosphor TFT/ CsI phosphor

G G S S D D Adjacent gate line Adjacent gate line

TFT TFT

Storage capacitor Storage capacitor

Photodiode Photodiode

Source Source Gate Gate Drain Drain Structured X Structured X-

• ray

ray phosphor (CsI) phosphor (CsI) X X-

• ray

ray Light Light + + Charge Charge

X X-

• rays to light to electrons to electronic signal

rays to light to electrons to electronic signal

Direct detector: Direct detector:

a a-

• Se / TFT array

Se / TFT array

Top electrode Top electrode Dielectric layer Dielectric layer Selenium photoconductor Selenium photoconductor Charge collection electrode Charge collection electrode (pixel size) (pixel size) Thin Thin-

• Film

Film-

• Transistor

Transistor Storage capacitor Storage capacitor Glass substrate Glass substrate High High voltage voltage

Incident x Incident x-

• rays

rays

+

• +

+ + + + + + +

• +

+

• +

+

Stored charge Stored charge

X X-

• rays to electrons to electronic signal

rays to electrons to electronic signal

+ + + + + + + +

SLIDE 14

14

Presentation Outline Presentation Outline Presentation Outline

• Acquisition System Overview
• Digital Detector Attributes
• Digital Detector Technologies
• Factors affecting Image Quality & Dose
• Clinical Implementation and QC
• Acquisition System Overview

Acquisition System Overview

• Digital Detector Attributes

Digital Detector Attributes

• Digital Detector Technologies

Digital Detector Technologies

• Factors affecting Image Quality & Dose

Factors affecting Image Quality & Dose

• Clinical Implementation and QC

Clinical Implementation and QC

Image acquisition and display Image acquisition and display Image acquisition and display

Uncorrected Uncorrected “ “ Raw Raw” ”

Acquisition Acquisition

Corrected Corrected “ “ Raw Raw” ”

Pre Pre-

• processing

processing

“ “ For display For display” ” enhanced enhanced

Post Post -

• processing

processing

Image Image comparisons comparisons

Display Display

X X-

• ray syst em

ray syst em

• S

pectrum S pectrum

• Detector

Detector Dead pixels Dead pixels Column/ line defects Column/ line defects S hading/ flat S hading/ flat -

• fielding

fielding S ignal amplification S ignal amplification Enhancement: Enhancement:

• Equalization

Equalization

• Contrast / Det ail

Contrast / Det ail Hard/ S

• ft Copy

Hard/ S

• ft Copy

Percept ual linearization Percept ual linearization VOI LUT VOI LUT --

• - DICOM GS

DF DICOM GS DF Hanging / Viewing Hanging / Viewing

Outside Outside images images

? ? Signal to Noise Ratio (SNR) Signal to Noise Ratio (SNR) Signal to Noise Ratio (SNR)

• Determines detectability of an object
• The signal is derived from the x-ray quanta
• The noise is from a variety of sources:

– X-ray quantum statistics – Electronic noise – Fixed pattern noise – Sampling noise (aliasing) – Anatomical noise

• System pre and post processing are crucial
• Determines detectability of an object

Determines detectability of an object

• The signal is derived from the x

The signal is derived from the x-

• ray quanta

ray quanta

• The noise is from a variety of sources:

The noise is from a variety of sources:

– – X X-

• ray quantum statistics

ray quantum statistics – – Electronic noise Electronic noise – – Fixed pattern noise Fixed pattern noise – – Sampling noise (aliasing) Sampling noise (aliasing) – – Anatomical noise Anatomical noise

• System pre and post processing are crucial

System pre and post processing are crucial

SLIDE 15

15

Pre-Processing Pre Pre-

• Processing

Processing

• Detector / x-ray system flaws

– Pixel defects – Sensitivity variations – Offset gain variations

• Wide detector dynamic range

– Identify image location – Scale image data – Optimize quantization levels for “post-processing”

• Detector / x

Detector / x-

• ray system flaws

ray system flaws

– – Pixel defects Pixel defects – – Sensitivity variations Sensitivity variations – – Offset gain variations Offset gain variations

• Wide detector dynamic range

Wide detector dynamic range

– – Identify image location Identify image location – – Scale image data Scale image data – – Optimize quantization levels for Optimize quantization levels for “ “post post-

• processing

processing” ”

Two major steps correct and adjust for: Two major steps correct and adjust for: Two major steps correct and adjust for:

Preprocessing, Step 1: correct flaws Preprocessing, Step 1: correct flaws Preprocessing, Step 1: correct flaws

• Detector origin

– stationary patterns (structured), fixed-point noise – thickness non-uniformities – drop-outs, dead pixels, dead columns – dark current variation

• Equipment origin

– heel effect – stationary patterns/artifacts (e.g., tube filter, grid)

• Detector origin

Detector origin

– – stationary patterns (structured), fixed stationary patterns (structured), fixed-

• point noise

point noise – – thickness non thickness non-

• uniformities

uniformities – – drop drop-

• outs, dead pixels, dead columns
• uts, dead pixels, dead columns

– – dark current variation dark current variation

• Equipment origin

Equipment origin

– – heel effect heel effect – – stationary patterns/artifacts (e.g., tube filter, grid) stationary patterns/artifacts (e.g., tube filter, grid)

Pre-processing schemes Pre Pre-

• processing schemes

processing schemes

• 1-D shading correction

– Computed Radiography (CR) – Linear CCD

• 2-D flat-field correction

– Area CCD, CMOS – TFT arrays

• 1

1-

• D

D shading shading correction correction

– – Computed Radiography (CR) Computed Radiography (CR) – – Linear CCD Linear CCD

• 2

2-

• D

D flat flat-

• field

field correction correction

– – Area CCD, CMOS Area CCD, CMOS – – TFT arrays TFT arrays

SLIDE 16

16

n 1, i ; O(x)

• I(x)

(x) I

i i O

= =

Shading correction techniques: 1 Shading correction techniques: 1-

• D data

D data

(x) I I Sh(x)

O x O

=

( )

x) Sh O(x)

• I(x)

C(x) ( × =

Apply offset correction to uniform exposures, n averages: Apply offset correction to uniform exposures, n averages: Create normalized shading correction array: Create normalized shading correction array: Implement shading correction (line by line): Implement shading correction (line by line):

1-D shading correction 1 1-

• D shading correction

D shading correction

Response Response Low noise, invert ed, normalized correction trace Low noise, invert ed, normalized correction trace

Pre Pre-

• processing

processing

Corrected response Corrected response Scan Direction Scan Direction

2-D flat-field correction 2 2-

• D flat

D flat-

• field correction

field correction

• Non-functioning components:

– Dead pixels in columns and/or rows

• Intensity variations:

– Uneven phosphor coating – Optical coupling (vignetting, barrel distortion) – Converter sensitivity

• Variation in offset and gain of sub-panels
• Variation in black-level correction
• Non

Non-

• functioning components:

functioning components:

– – Dead pixels in columns and/or rows Dead pixels in columns and/or rows

• Intensity variations:

Intensity variations:

– – Uneven phosphor coating Uneven phosphor coating – – Optical coupling (vignetting, barrel distortion) Optical coupling (vignetting, barrel distortion) – – Converter sensitivity Converter sensitivity

• Variation in offset and gain of sub

Variation in offset and gain of sub-

• panels

panels

• Variation in black

Variation in black-

• level correction

level correction

SLIDE 17

17

Uncorrected flat-panel image Uncorrected flat Uncorrected flat-

• panel image

panel image

+ , + , -

• column defects

column defects row defects row defects pixel defects pixel defects Sub Sub-

• panel offset gain variation

panel offset gain variation Background Background signal signal

n 1, i ; y) O(x,

• y)

I(x, y) (x, I

i i O

= =

Flat Flat-

• field techniques: 2D image

field techniques: 2D image

(fixed detector) (fixed detector)

y) (x, I I y) FF(x,

O O

=

( )

y) x, FF y) O(x,

• y)

I(x, y) C(x, ( × =

Apply offset and average of <n> uniform exposures: Apply offset and average of <n> uniform exposures: Create normalized flat Create normalized flat-

• field correction matrix:

field correction matrix: Implement flat Implement flat-

• field correction on acquired image:

field correction on acquired image:

2-D Flat-field correction 2 2-

• D Flat

D Flat-

• field correction

field correction

Raw, Raw Raw, Raw

Background variations Background variations Column, line defect s Column, line defect s “ “ Del Del” ” dropout s dropout s

Correction Correction “ “ mask mask” ”

Avg, invert ed background Avg, invert ed background Column, line, pixel repair Column, line, pixel repair Normalized values Normalized values

Raw Raw

Pre Pre-

• processed image

processed image Image pixel value t o Image pixel value t o exposure relat ionship? exposure relat ionship?

Processed Processed

Cont rast , resolut ion Cont rast , resolut ion enhancement ; propriet ary enhancement ; propriet ary processing processing

Pre Pre-

• processing

processing

SLIDE 18

18

Preprocessing, Step 2: find / scale image Preprocessing, Step 2: find / scale image Preprocessing, Step 2: find / scale image

Determine Collimation Determine Collimation

Collimation Collimation Border Border Shift and Subtract Shift and Subtract

Create / analyze Histogram Distribution Create / analyze Histogram Distribution Create / analyze Histogram Distribution

The shape is dependent on radiographic study, The shape is dependent on radiographic study, positioning and technique positioning and technique Anatomy Anatomy Frequency Frequency Useful signal Useful signal Collimated Collimated area area Direct Direct x x-

• ray

ray area area Pixel value Pixel value

Data conversion Data conversion

Exposure into digital number Exposure into digital number

200 200 600 600 1,000 1,000 200 200 400 400 600 600 800 800 1,000 1,000 Raw Input digital number Raw Input digital number Output digital number Output digital number

Grayscale transformation Grayscale transformation

Input to output digital number Input to output digital number 10 103

3

10 10-

• 1

1 10

100 10 102

2

10 101

1

Exposure input Exposure input

Relative PSL Relative PSL

10 101

1

10 10-

• 1

1

10 100 10 102

2

511 511 1023 1023 Raw Digital Output Raw Digital Output

Histogram Histogram

min min max max

Find the Find the signal signal 1. 1. Scale to Scale to range range 2. 2. Create film Create film look look-

• alike

alike 3. 3.

SLIDE 19

19

Histogram: pediatric image Histogram: pediatric image

200 400 600 800 400 600 800 1000 Digital value

Frequency

to 8323 to 9368

Useful image range for anatomy Useful image range for anatomy

200

Pre Pre-

• processed

processed “ “raw raw” ” image image Scaled and inverted: Scaled and inverted: “ “For Processing For Processing” ” image image

Data conversion for overexposure Data conversion for overexposure

Exposure into digital number Exposure into digital number

Exposure Exposure input input Relative PSL Relative PSL 511 511 1023 1023 Raw Digital Output Raw Digital Output (scaled and log amplified) (scaled and log amplified) Reduce overall gain Reduce overall gain 10 103

3

10 10-

• 1

1 10

100 10 102

2

10 101

1

10 101

1

10 10-

• 1

1

10 100 10 102

2

min min max max

• verexposure
• verexposure

SLIDE 20

20

Screen – Film Digital

Identical exposure

Data conversion for wide latitude Data conversion for wide latitude

Exposure into digital number: less kV dependence Exposure into digital number: less kV dependence

Relative PSL Relative PSL

10 101

1

10 10-

• 1

1

10 100 10 102

2

511 511 1023 1023 Raw Digital Output Raw Digital Output (scaled and log amplified) (scaled and log amplified) Exposure Exposure input input 10 103

3

10 10-

• 1

1 10

100 10 102

2

10 101

1

Change gradient Change gradient (auto mode) (auto mode)

low kVp low kVp

(broad histogram) (broad histogram) min min max max

Contrast Enhancement Contrast Enhancement Contrast Enhancement

• Optimize image contrast via non-linear

transformation curves

• Unprocessed images: “subject contrast”
• Proprietary processing:

– “Gradation processing” (Fuji) – “Tone scaling” (Kodak) – “MUSICA” (Agfa) – …….. And others by the various manufacturers

• Optimize image contrast via

Optimize image contrast via non non-

• linear

linear transformation curves transformation curves

• Unprocessed images:

Unprocessed images: “ “subject contrast subject contrast” ”

• Proprietary processing:

Proprietary processing:

– – “ “Gradation processing Gradation processing” ” (Fuji) (Fuji) – – “ “Tone scaling Tone scaling” ” (Kodak) (Kodak) – – “ “MUSICA MUSICA” ” (Agfa) (Agfa) – – …… …….. And others by the various manufacturers .. And others by the various manufacturers

SLIDE 21

21

Look-up-table transformation Look Look-

• up

up-

• table transformation

table transformation

Input digital number Input digital number Output digital number Output digital number

200 200 400 400 600 600 800 800 1,000 1,000 200 200 400 400 600 600 800 800 1,000 1,000 A A E E L L M M

Fuji System

Example LUTs

Fuji System Fuji System

Example LUTs Example LUTs

Raw Raw Unprocessed Unprocessed Contrast Contrast Enhanced Enhanced

“ “For processing For processing” ” CAD CAD VOI LUT VOI LUT “ “For presentation For presentation” ” Proprietary Proprietary Limited variability Limited variability “ “Pre Pre-

• processed

processed” ” No scaling No scaling No flatfield No flatfield

Types of image output: Types of image output: Types of image output: VOI LUT: a more flexible approach VOI LUT: a more flexible approach VOI LUT: a more flexible approach

• Value Of Interest Look-Up-Table (DICOM)
• Adjustment of contrast, brightness with non-linear

LUT adjustment

• Provides for manipulation of raw data

(“For Processing” images)

• Universal support (modalities, PACS) not available
• Future universal image processing workstation?
• Value Of Interest Look

Value Of Interest Look-

• Up

Up-

• Table (DICOM)

Table (DICOM)

• Adjustment of contrast, brightness with non

Adjustment of contrast, brightness with non-

• linear

linear LUT adjustment LUT adjustment

• Provides for manipulation of raw data

Provides for manipulation of raw data ( (“ “For Processing For Processing” ” images) images)

• Universal support (modalities,

Universal support (modalities, PACS) not available PACS) not available

• Future universal image processing workstation?

Future universal image processing workstation?

SLIDE 22

22

DICOM VOI LUT DICOM VOI LUT DICOM VOI LUT

• Configure CR/DX modality to send specific VOI LUT

– Eliminates “burned-in” LUT and potential information loss

• PACS must be able to use and vary VOI LUT
• Configure CR/DX modality to send specific VOI LUT

Configure CR/DX modality to send specific VOI LUT

– – Eliminates Eliminates “ “burned burned-

• in

in” ” LUT and potential information loss LUT and potential information loss

• PACS must be able to use and vary VOI LUT

PACS must be able to use and vary VOI LUT

Raw image histogram values Raw image histogram values P P -

• values

values 4095 4095 Adjustable VOI LUT

Adapted from Mike Flynn presentation Adapted from Mike Flynn presentation

Spatial Frequency Processing Spatial Frequency Processing Spatial Frequency Processing

“Edge Enhancement” “ “Edge Enhancement Edge Enhancement” ”

Original Original Blurred Blurred Difference Difference Edge enhanced Edge enhanced

Spatial frequency Spatial frequency

MTF: original response MTF: original response

Response Response

low low high high

Spatial frequency Spatial frequency

Solid: original response Solid: original response Dash: low pass filtered Dash: low pass filtered

Response Response

low low high high

Spatial frequency Spatial frequency

Original Original -

• filtered

filtered Difference: Difference: low low high high

Spatial frequency Spatial frequency Sum Sum

low low Difference + Original Difference + Original Edge Enhanced: Edge Enhanced: high high

Non Non-

• linear

linear weighting weighting Non Non-

• linear

linear weighting weighting Non Non-

• linear

linear weighting weighting Non Non-

• linear

linear weighting weighting

“ “Multi frequency Multi frequency” ” enhanced image enhanced image

Optimize sub Optimize sub-

• band weighting

band weighting

Multi-band Frequency Processing Multi Multi-

• band Frequency Processing

band Frequency Processing

SLIDE 23

23

Standard Processing Standard Processing Multi Multi-

• frequency Processing

frequency Processing

Compliments of Keith Strauss, Boston Compliments of Keith Strauss, Boston Childrens Childrens Hospital Hospital

SNR and CNR (dSNR) SNR and CNR ( SNR and CNR (dSNR dSNR) )

• SNR: Average value / Std Dev of background
• CNR: ∆ Attenuation / Std Dev of background

– Contrast: tissue differences, tissue/bone differences – Subject contrast: X-ray energy

• Detection: CNR of 3 to 5

– Size (diameter); image processing

• SNR: Average value / Std Dev of background

SNR: Average value / Std Dev of background

• CNR:

CNR: ∆ ∆ Attenuation / Std Dev of background Attenuation / Std Dev of background

– – Contrast: tissue differences, tissue/bone differences Contrast: tissue differences, tissue/bone differences – – Subject contrast: X Subject contrast: X-

• ray energy

ray energy

• Detection: CNR of 3 to 5

Detection: CNR of 3 to 5

– – Size (diameter); image processing Size (diameter); image processing

SNR and CNR SNR and CNR SNR and CNR

Background Background 420.3 420.3 ± ±3.3 3.3 Object Object 411.8 411.8 ± ±3.3 3.3 420.3 SNR = = 127.4 3.3 420.3- 411.8 CNR = = 2.6 3.3

SLIDE 24

24

Noise Sources Noise Sources Noise Sources

• Incomplete x-ray absorption: η
• Secondary quantum noise: quantum sink

– # secondary quanta ≤ incident q

• Spatial gain variation (flat-field)
• Aliasing (insufficient sampling)
• Swank Factor

– Different x-ray photons produce variable signal

• Lubberts Effect

– Different x-ray photons produce variable PSF’s

• Additive system noise

– Electronic, quantization, shot, etc.

• Incomplete x

Incomplete x-

• ray absorption:

ray absorption: η η

• Secondary quantum noise: quantum sink

Secondary quantum noise: quantum sink

– – # secondary quanta # secondary quanta ≤ ≤ incident incident q q

• Spatial gain variation (flat

Spatial gain variation (flat-

• field)

field)

• Aliasing (insufficient sampling)

Aliasing (insufficient sampling)

• Swank Factor

Swank Factor

– – Different x Different x-

• ray photons produce variable signal

ray photons produce variable signal

• Lubberts

Lubberts Effect Effect

– – Different x Different x-

• ray photons produce variable

ray photons produce variable PSF PSF’ ’s s

• Additive system noise

Additive system noise

– – Electronic, quantization, shot, etc. Electronic, quantization, shot, etc.

Visual Detection of Object Visual Detection of Object Visual Detection of Object

• SNR (CNR) is x-ray quanta dependent
• Detection is determined by CNR and object size
• k = SNR × d × C

C = contrast d = diameter k = 3 to 5 for detection

• SNR (CNR) is x

SNR (CNR) is x-

• ray quanta dependent

ray quanta dependent

• Detection is determined by CNR and object size

Detection is determined by CNR and object size

• k = SNR

k = SNR × × d d × × C C C = contrast C = contrast d = diameter d = diameter k = 3 to 5 for detection k = 3 to 5 for detection

Low Contrast Response: Leeds TO-16 Low Contrast Response: Leeds TO Low Contrast Response: Leeds TO-

• 16

16

0.5 mR 0.5 mR 0.5 mR 3.5 mR 3.5 mR 3.5 mR 70 kVp 70 kVp 70 kVp

SLIDE 25

25

What determines necessary dose? What determines What determines necessary necessary dose? dose?

• Patient thickness
• X-ray technique; GRID or NO GRID
• Detector absorption AND conversion efficiency
• Detector electronic and stationary noise
• Detector Detective Quantum Efficiency (DQE)
• Required SNR / CNR of examination
• Pre and post processing algorithms
• Display and viewing conditions
• Patient thickness

Patient thickness

• X

X-

• ray technique; GRID or NO GRID

ray technique; GRID or NO GRID

• Detector absorption

Detector absorption AND AND conversion efficiency conversion efficiency

• Detector electronic and stationary noise

Detector electronic and stationary noise

• Detector Detective Quantum Efficiency (DQE)

Detector Detective Quantum Efficiency (DQE)

• Required SNR / CNR of examination

Required SNR / CNR of examination

• Pre and post processing algorithms

Pre and post processing algorithms

• Display and viewing conditions

Display and viewing conditions

10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 110 120 130 140

Energy (keV) % Absorption Fraction

CsI: 175 mg/cm2 Gd2O2S: 120 mg/cm2 BaFBr: 100 mg/cm2

X-ray absorption Efficiency: CsI, BaFBr, Gd2O2S

Detective Quantum Efficiency (DQE) Detective Quantum Efficiency (DQE) Detective Quantum Efficiency (DQE)

• A measure of the information transfer

efficiency of a detector system

• Dependent on:

– Absorption efficiency – Conversion efficiency – Spatial resolution (MTF) – Conversion noise – Electronic noise – Detector non-uniformities / pattern noise

• A measure of the

A measure of the information transfer information transfer efficiency efficiency of a detector system

• f a detector system
• Dependent on:

Dependent on:

– – Absorption efficiency Absorption efficiency – – Conversion efficiency Conversion efficiency – – Spatial resolution (MTF) Spatial resolution (MTF) – – Conversion noise Conversion noise – – Electronic noise Electronic noise – – Detector non Detector non-

• uniformities / pattern noise

uniformities / pattern noise

2 2

• ut

2 in N

SNR MTF( ) DQE( ) = SNR NPS ( ) f f f q = ×

2 2

• ut

2 in N

SNR MTF( ) DQE( ) = SNR NPS ( ) f f f q = ×

SLIDE 26

26

“Pre-sampled” MTF “ “Pre Pre-

• sampled

sampled” ” MTF MTF

a a-

• Selenium: 0.13 mm

Selenium: 0.13 mm Screen Screen-

• film

film CsI CsI-

• TFT: 0.20 mm

TFT: 0.20 mm CR: 0.10 mm CR: 0.10 mm 0.0 0.0 0.1 0.1 0.2 0.2 0.3 0.3 0.4 0.4 0.5 0.5 0.6 0.6 0.7 0.7 0.8 0.8 0.9 0.9 1.0 1.0 0.0 0.0 0.5 0.5 1.5 1.5 2.5 2.5 3.5 3.5 4.5 4.5 1.0 1.0 2.0 2.0 3.0 3.0 4.0 4.0 5.0 5.0

Frequency (lp/mm) Frequency (lp/mm) Modulation Modulation

CR: 0.05 mm CR: 0.05 mm

Noise Power Spectrum Noise Power Spectrum Noise Power Spectrum

• Noise transfer characteristics of detector
• Analyze sub-images, Fourier Transform, average

– (IEC 62220-1 standard, AAPM Task Group on NPS)

• Output is the noise power estimate as a function
• f spatial frequency, NPS(f) in 2 dimensions
• Noise transfer characteristics of detector

Noise transfer characteristics of detector

• Analyze sub

Analyze sub-

• images, Fourier Transform, average

images, Fourier Transform, average

– – (IEC 62220 (IEC 62220-

• 1 standard, AAPM Task Group on NPS)

1 standard, AAPM Task Group on NPS)

• Output is the noise power estimate as a function

Output is the noise power estimate as a function

• f spatial frequency,
• f spatial frequency, NPS(

NPS(f f) in 2 dimensions ) in 2 dimensions

CR Image NPS CR Image NPS Scan direction Scan direction DQE( f ) Spatial Frequency (cycles/mm) 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.2 0.4 0.6 0.8 CsI - TFT Screen-film CR Conventional a-Se - TFT CR “dual-side”

Detective Quantum Efficiency, Radiography

CCD

SLIDE 27

27

Screen Screen-

• Film

Film aSi aSi/CsI Flat /CsI Flat-

• Panel

Panel CR CR

125 kVp 125 kVp 2 mAs 2 mAs

MDACC: Chris Shaw, et al MDACC: Chris Shaw, et al

Flat Flat-

• field pre

field pre-

• processing

processing

Low contrast Low contrast resolution resolution

Image quality depends on more than Image quality depends on more than quantum mottle! quantum mottle! Flat Flat-

• field, absorption efficiency, scatter

field, absorption efficiency, scatter … …

Shading, Flat-Field correction Shading, Flat Shading, Flat-

• Field correction

Field correction

• Reduce structured noise
• Eliminate variable background
• Increase DQE( f )
• Reduce structured noise

Reduce structured noise

• Eliminate variable background

Eliminate variable background

• Increase DQE(

Increase DQE( f f ) )

0.1 0.2 0.3 0.4 0.5 2 4 6 8 10 f (mm-1) DQE(f) Flatfield Raw

Digital Radiography: Radiation Exposure Digital Radiography: Radiation Exposure Digital Radiography: Radiation Exposure

• CR and DR tolerate poor radiographic technique
• Dose is dependent on DQE and “required” SNR
• Dose is roughly proportional to inverse of DQE
• CR and DR tolerate poor radiographic technique

CR and DR tolerate poor radiographic technique

• Dose is dependent on DQE and

Dose is dependent on DQE and “ “required required” ” SNR SNR

• Dose is roughly proportional to inverse of DQE

Dose is roughly proportional to inverse of DQE

SLIDE 28

28

Exposure issues Exposure issues Exposure issues

• Incident exposure can be “hidden”
• Low exposures have excessive image noise
• High exposures lead to saturation signal loss
• Technique complacency, instead of “just enough”
• Feedback is necessary!!

– S number, Exposure Index, LgM, f-number, other?

• Incident exposure can be

Incident exposure can be “ “hidden hidden” ”

• Low exposures have excessive image noise

Low exposures have excessive image noise

• High exposures lead to saturation signal loss

High exposures lead to saturation signal loss

• Technique complacency, instead of

Technique complacency, instead of “ “just enough just enough” ”

• Feedback is necessary!!

Feedback is necessary!!

– – S number, Exposure Index, S number, Exposure Index, LgM LgM, f , f-

• number, other?

number, other?

Digital Digital Film Film-

• screen

screen (400 speed) (400 speed) 0.01 0.1 1 10 100 1 10 100 1,000 10,000 Exposure, mR Relative intensity Film Optical Density 1 2 3 4 20000 2000 200 20 2 Sensitivity (S) Underexposed Underexposed Overexposed Overexposed Correctly exposed Correctly exposed

Characteristic Curve:

Response of screen/film vs. digital detectors

Useless Useless Useless Useless 5

How do manufacturers indicate estimated exposure? How do manufacturers indicate How do manufacturers indicate estimated exposure? estimated exposure?

• Fuji: “S” – sensitivity number
• S ≅ 200 / Exposure (mR)
• Kodak: “Exposure Index” – EI
• EI ≅ 1000 × log (Exposure [mR] ) + 2000
• Agfa: “lg M” – relative exposure database
• IDC: “f-number” – provides analogy to camera speed
• +1 = 2x exposure; +2 = 4x exposure
• DR: most systems currently do not have a feedback

signal… but use phototiming (AEC)

• Fuji:

Fuji: “ “S S” ” – – sensitivity number sensitivity number

• S

S ≅ ≅ 200 / Exposure (mR) 200 / Exposure (mR)

• Kodak:

Kodak: “ “Exposure Index Exposure Index” ” – – EI EI

• EI

EI ≅ ≅ 1000 1000 × × log (Exposure [mR] ) + 2000 log (Exposure [mR] ) + 2000

• Agfa:

Agfa: “ “lg lg M M” ” – – relative exposure database relative exposure database

• IDC:

IDC: “ “f f-

• number

number” ” – – provides analogy to camera speed provides analogy to camera speed

• +1 = 2x exposure; +2 = 4x exposure

+1 = 2x exposure; +2 = 4x exposure

• DR: most systems currently do not have a feedback

DR: most systems currently do not have a feedback signal signal… … but use phototiming (AEC) but use phototiming (AEC)

SLIDE 29

29

CR vs DR and dose efficiency CR CR vs vs DR and dose efficiency DR and dose efficiency

• CR ~ 2X more exposure than a 400 speed film

~200 equivalent speed

• DR DQE(0) values vary substantially (20 - 80%)
• Dose efficiency related to DQE for given SNR
• Slot-scan systems most efficient
• CR ~ 2X more exposure than a 400 speed film

CR ~ 2X more exposure than a 400 speed film ~200 ~200 equivalent equivalent speed speed

• DR DQE(0) values vary substantially (20

DR DQE(0) values vary substantially (20 -

• 80%)

80%)

• Dose efficiency related to DQE for given SNR

Dose efficiency related to DQE for given SNR

• Slot

Slot-

• scan systems most efficient

scan systems most efficient

Presentation Outline Presentation Outline Presentation Outline

• Acquisition System Overview
• Digital Detector Attributes
• Digital Detector Technologies
• Factors affecting Image Quality & Dose
• Clinical Implementation and QC
• Acquisition System Overview

Acquisition System Overview

• Digital Detector Attributes

Digital Detector Attributes

• Digital Detector Technologies

Digital Detector Technologies

• Factors affecting Image Quality & Dose

Factors affecting Image Quality & Dose

• Clinical Implementation and QC

Clinical Implementation and QC

• Modality interface: DICOM & HL-7

– PACS, RIS connections, Modality Worklist

• Image Size & Storage considerations
• 8 - 32 Mbytes Uncompressed

– 10 - 12 Pixels/mm – Up to 4000 x 4000 x 2 Bytes

• 3 - 13 Mbytes: ~2.5:1 Lossless Compression
• Network Transmission
• 100 Mbit/sec minimum
• Modality interface: DICOM & HL

Modality interface: DICOM & HL-

• 7

7

– – PACS, RIS connections, PACS, RIS connections, Modality Worklist Modality Worklist

• Image Size & Storage considerations

Image Size & Storage considerations

• 8

8 -

• 32 Mbytes Uncompressed

32 Mbytes Uncompressed

– – 10 10 -

• 12 Pixels/mm

12 Pixels/mm – – Up to 4000 x 4000 x 2 Bytes Up to 4000 x 4000 x 2 Bytes

• 3

3 -

• 13 Mbytes: ~2.5:1 Lossless Compression

13 Mbytes: ~2.5:1 Lossless Compression

• Network Transmission

Network Transmission

• 100

100 Mbit Mbit/sec minimum /sec minimum

CR/DR implementation CR/DR implementation CR/DR implementation

SLIDE 30

30

CR/DR implementation CR/DR implementation CR/DR implementation

• Uniformity for CR/DR images and Display

– Acceptance Testing

• Measurement of Performance
• Correction of Substandard Performance

– Calibration of CR/DR Response (presentation state) – Calibration of Monitors

• Maximum brightness
• Look-up-Tables, DICOM GSDF, Part 14

– Periodic Quality Control

• Evaluation of resolution, contrast, artifacts
• Monitor technologist performance, exposure indices
• Uniformity for CR/DR images

Uniformity for CR/DR images and and Display Display

– – Acceptance Testing Acceptance Testing

• Measurement of Performance

Measurement of Performance

• Correction of Substandard Performance

Correction of Substandard Performance

– – Calibration of CR/DR Response (presentation state) Calibration of CR/DR Response (presentation state) – – Calibration of Monitors Calibration of Monitors

• Maximum brightness

Maximum brightness

• Look

Look-

• up

up-

• Tables, DICOM GSDF, Part 14

Tables, DICOM GSDF, Part 14

– – Periodic Quality Control Periodic Quality Control

• Evaluation of resolution, contrast, artifacts

Evaluation of resolution, contrast, artifacts

• Monitor technologist performance, exposure indices

Monitor technologist performance, exposure indices

CR/DR implementation CR/DR implementation CR/DR implementation

• Image Processing optimization

– Establish Contrast Scale – Balance Edge Enhancement with perceived noise – Multi-frequency Enhancement parameter adjustments – Determine DC offset (brightness) for display monitors

• Provide processing “looks” to Radiologists
• Verify image display conditions

– Soft copy and hard copy

• Image Processing optimization

Image Processing optimization

– – Establish Contrast Scale Establish Contrast Scale – – Balance Edge Enhancement with perceived noise Balance Edge Enhancement with perceived noise – – Multi Multi-

• frequency Enhancement parameter

frequency Enhancement parameter adjustments adjustments – – Determine DC offset (brightness) for display Determine DC offset (brightness) for display monitors monitors

• Provide processing

Provide processing “ “looks looks” ” to Radiologists to Radiologists

• Verify image display conditions

Verify image display conditions

– – Soft copy and hard copy Soft copy and hard copy

What is emerging as the lead technology? What is emerging as the lead technology? What is emerging as the lead technology?

Attribute CR DR CCD Positioning flexibility **** ** ** Replacement for S/F **** ** ** DQE / dose efficiency ** *** ** Patient throughput * *** ** X-ray system integration ** **** **** PACS integration ** **** **** Cost per pat. throughput *** ** *** Technologist ease of use * *** ***

SLIDE 31

31

Digital Radiography Considerations Digital Radiography Considerations Digital Radiography Considerations

• Replacement of S/F, aging CR
• High throughput, ambulatory imaging
• Advanced image acquisition and processing

– Digital tomosynthesis and CT – Dual energy radiography – Replacement of image intensifiers

• Low dose screening devices with CAD

– Lung cancer screening with dual energy – Quantitative bone density analysis?

• Replacement of S/F, aging CR

Replacement of S/F, aging CR

• High throughput, ambulatory imaging

High throughput, ambulatory imaging

• Advanced image acquisition and processing

Advanced image acquisition and processing

– – Digital tomosynthesis and CT Digital tomosynthesis and CT – – Dual energy radiography Dual energy radiography – – Replacement of image intensifiers Replacement of image intensifiers

• Low dose screening devices with CAD

Low dose screening devices with CAD

– – Lung cancer screening with dual energy Lung cancer screening with dual energy – – Quantitative bone density analysis? Quantitative bone density analysis?

Conclusions Conclusions

• CR is the most

CR is the most flexible flexible and cost and cost-

• effective

effective technology technology

• Direct

Direct digital radiographic devices have advantages in digital radiographic devices have advantages in efficiency and throughput efficiency and throughput

• The distinction between CR and DR is blurring

The distinction between CR and DR is blurring

– – Portable versus integrated; active versus passive Portable versus integrated; active versus passive – – “ “Cassette Cassette” ” versus versus “ “Cassetteless Cassetteless” ”

• All technologies are becoming faster, better, cheaper

All technologies are becoming faster, better, cheaper

• The digital solution is best accomplished as a

The digital solution is best accomplished as a complementary mix of technologies complementary mix of technologies