Digital Image Processing in Radiography

Outline
Digital Image Processing in
Radiography
Xiaohui Wang, PhD
David H. Foos, MS
Health Group Research Laboratory
Eastman Kodak Company
• Display Processing
– Data preprocessing
– ROI segmentation and analysis
– Tonal rendering
– Signal equalization
– Edge restoration
– Noise suppression
– Collimation masking
– Display compensation
• Image Processing Features
– Stationary grid detection and suppression
– Long-length imaging
– Dual energy imaging
– Mammography
– Oncology processing
– Quality control testing
1
2
“Analog” Image Processing…
3
Screen/Film Radiography (S/F)
Display Processing
Density
2
1.0
1
0.8
0.6
0
4 5 6 7
Log Exposure
0.4
Modulation Transfer
Low Speed Screens
Film Speed & Contrast
Screen Speed & MTF
0.2 High Speed Screens
0
0 1 2 3 4 5
Spatial Frequency (mm -1 )
3
4
1

Digital Image Processing
Why Image Processing?
Digital Image Acquisition
Digital Image Processing
• Maintain the familiar characteristics of S/F
– Provide a similar tonal rendering
– Restore desired sharpness
• Beyond the familiar
– Automatically adjust for errors in exposure
– Automatically accommodate changes in latitude
– Increase the range of exposures visualized
– Enhance selected spatial frequencies
– Highlight regions of interest (ROI)
– Assist the radiologist to find features of interest
Digital Image Display
5
6
Display Processing
• Transform digital radiography raw data to display values for presentation
using a workstation or film printer, automatically, robustly, and
consistently.
• Components of Image Quality
– Latitude
– Contrast
– Brightness
– Sharpness
– Noise
Original Image
Tone Scale
Edge Restoration
Signal Equalization
Collimation Masking
PSP plate
Relative intensity of PSL
10,000
1,000
100
10
Film-screen
(400 speed)
Overexposed
Correctly exposed
Underexposed
4
3
2
1
Film Optical Density
7
1
0.01 0.1 1 10 100
Incident exposure, mR
0
8
2

Schematic Flow Chart of Display Processing
Original Image
Tone Scale
Image
Capture
Data
Preprocessing
ROI
Segmentation
& Analysis
Tone-Scale
Generation
Edge Restoration
Signal Equalization
Collimation Masking
Edge
Restoration
Signal
Equalization
Tonal
Rendering
Collimation
Masking
Display
Compensation
PACS
& Print
Noise
Suppression
Multi-Frequency Processing
9
10
Data Preprocessing
Data Preprocessing (cont.)
• Image reformation
– Composition (dual-side CR reading)
– Decomposition (dual energy)
– Resize
• Signal filtering
– Gain, offset, and bad pixel correction (DR)
– Noise reduction
– Stationary grid artifact suppression
• Data space conversion
– Linear to logarithmic (const. object contrast vs. pixel value)
– Linear to square root (const. quantum noise vs. pixel value)
Linear-to-log conversion
Shape of image histogram invariant to exposure
I 0
f (x)
I
I
I
0
e
f ( x)
1
1
f ( x)
log( I / I ) ~
0
log( I ) const.
Baseline
4x over exposure
4x under-ex posure
0 1000 2000 3000 4000 5000
11
12
3

ROI Segmentation & Analysis
• Extract diagnostically relevant ROIs
• Analyze ROI characteristics
• Derive the optimal display-rendering parameters
• Include four basic steps
– Detect collimation mask
Direct
– Detect direct exposure
Exposure
– Extract anatomy regions
– Calculate key image descriptors
Collimation
Segmentation – Collimation Mask
• Confine exposure regions
• Mask applied to collimated regions to reduce viewing flare
Anatomy
13
14
“Method for recognizing multiple radiation fields in computed radiography,” X. Wang, J. Luo, R. Senn, and D. Foos,
Proc SPIE 3661, 1625-36, 1999.
Segmentation – Collimation Mask (cont.)
• Collimation boundary pixels
– Edge profile analysis
– Transition segments classification
• Candidate collimation blades
– Edge delineation
– FOM analysis
• straightness
• connectedness…
• Candidate configurations
• Select “best” configuration
– Parallelism, convexity, orthogonality, etc.
Segmentation – Collimation Mask (cont.)
• Multiple radiation field masking
– Optimal individual image processing
– Exam workflow improvement
15
J. Luo and R. Senn, “Collimation detection for digital radiography," Proc. SPIE 3034, 74-85 (1997).
16
“Method for recognizing multiple radiation fields in computed radiography,” X. Wang, J. Luo, R. Senn, and D. Foos, Proc SPIE
3661, 1625-36, 1999.
4

Segmentation – Collimation Mask (cont.)
Segmentation - Direct Exposure
Detection
Aggressive
Failure
Conservative
Failure
Exclude non-anatomical regions
within the collimation.
• Compensate
– Radiation field non-uniformity
– X-ray scatter
– Multiple exposures
• Transition segment analysis
– Line profile analysis
– Background transitions characterized by slope and extent
– Background pixel histogram analysis
– Spatial correlation of exposure variations
L. Barski and R. Senn “Determination of direct x-ray exposure regions in digital medical imaging,” U.S. Patent 5,606,587 (1997).
17
18
Segmentation – Anatomy Extraction
ROI Analysis - Key Image
Descriptors
Lateral Lumbar Spine ROI Selection
1
0.9
Normalized Frequency
0.8
0.7
0.6
0.5
0.4
0.3
act hist
cv. hist
Anatomy
Anatomy
0.2
0.1
0
Left pt
Far left pt
Right pt
Far right pt
19
“Automatic and exam-type independent algorithm for the segmentation and extraction of foreground, background, and
anatomy regions in digital radiographic images,” X. Wang, H. Luo, Proc. SPIE 5370, 1427-1434, 2004.
20
5

4000
3500
3000
2500
2000
1500
1000
500
0
shift = -0.5
shift = 0
shift = 0.5
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Code Value In
Tone-Scale Generation
• Render image with proper brightness and contrast
– Calculate average exposure within ROI
– Automatically adjust for errors in exposure
• Sigmoid curve shape in general
– Curve shift (brightness adjustment)
– Curve rotation (contrast adjustment)
– Toe & shoulder adjustment
• Bear different names
– Kodak: PTS (Perceptual Tone Scale)
– Fuji: Gradation Processing
–…
Brightness Adjustment
Code Value Out
Effect of Density Shift Pa rameter
21
22
Contrast Adjustment
Code Value Out
4000
3500
3000
2500
2000
1500
1000
500
Effect of Contrast Parameter
c = 0.8
c = 1.3
c = 1.8
Toe & Shoulder
Adjustment
Code Value Out
4000
3500
3000
2500
2000
1500
1000
Effect of Toe and Shoulder Parameters
T=0, S=0
T=0.3, S=0.7
0
500
T=0.6, S=1.4
0 1000 2000 3000 4000 5000
Code Value In
0
0 1000 2000 3000 4000 5000
Code V alue In
c
23
24
6

Visually Optimized Tone Scale
Perceptual Linearity -
Visually Optimized Tone Scale (cont.)
Radiologists prefer S/F systems with perceptually linear
sensitometric response.
Render ROI such that… equal physical contrast being
perceived as equal brightness by the observer across the full
brightness range
X-rays
2
3
Kodak Insight Thoracic
Imaging System
Kodak Insight HC Thoracic
Imaging System
25
26
Perceptual Tone Scale
Perceptual Tone Scale (cont.)
Density – Luminance
(Physics)
dmax
flp
dmin
lp
rp
frp
Perceptually
Linearity
Tone Scale
Equal Log (E)
Daly’s Global Cone Model
B = B m L n
L n + L 0
n
Equal Brightness
B = perceived brightness
L = luminance of the image area
B m = scale factor
n = 0.7
Visual Perception
Model
Perception
Linearity
L 0 = 12.6*(0.2*L w ) 0.63 + 1.083*10 -5
L w = luminance of the reference white
H. Lee, S. Daly, and R. Van Metter, “Visual optimization of radiographic tone scale,” Proc. SPIE 3036, 118-129, 1996.
27
28
7

Tone Scale Failures
Signal Equalization
4000
Effect of Contrast Parameter
3500
3000
Balance between
contrast vs. latitude
Code Value Out
2500
2000
1500
1000
500
Bone
Soft
c = 0.8
c = 1.3
Tissue
c = 1.8
0
0 1000 2000 3000 4000 5000
Code Value In
29
Too Bright Too Dark Too Much
Contrast
Too Less
Contrast
30
Good Soft Tissue Contrast
Signal Equalization
Good Bone Contrast
31
Signal Equalization (cont.)
• Automatically accommodate changes in latitude
– Compress the image-signal dynamic range such that all
information within ROI can be rendered with optimal
contrast
• Increase the range of exposures visualized
• Reduce exposure re-take and improve workflow
• Signal equalization processing is
– 2D spatial processing
– Digital wedge filter
– Bearing different names:
• EVP - Enhanced Visualization Processing (Kodak)
• DRC - Dynamic Range Compression (Fuji)
• Latitude Reduction (AGFA)
• Tissue Equalization Processing (GE)
• … 32
Signal Equalization (Kodak EVP)
Input Image &
PTONE LUT
Original
Image
EVP
KERNEL SIZE
PTONE
LUT
Blurred
Image
EVP GAIN
NEW
PTONE LUT
EVP GAIN and EVP DENSITY
E’(i,j) = { E(i,j) K } + ( 1 - ) E mid + { E(i,j) - ( E(i,j) K ) }
D(i,j) = [ E’(i,j) ].
-
Output Image &
PTONE LUT
“Enhanced latitude for digital projection radiography,” R. Van Metter and D. Foos, Proc. SPIE 3658, 468-483, 1999.
8

Signal Equalization (cont.)
Signal Equalization (cont.)
Increasing
Contrast,
…Decreasing
Latitude
11-11
8-8
5-5
Increasing
Contrast,
…Constant
Latitude
11-11
8-11
5-11
33
34
RSNA 2001 Education Exhibit
Observer “rendering preferences” established
through collaborative studies with university hospitals
(…to set default parameters for automatic processing)
High
Contrast
Low
Detail Latitude
Contrast LogE
D/ LogE => 0.35 0.47 0.56 0.78 0.92 1.09 1.46 1.75 2.06
1.4 1.8 2.2 3.0 3.6 4.2 5.6 6.8 7.9
6.75
5.75 1.5 1.9 2.6 3.0 3.6 4.8 5.8 6.8
5 1.0 1.3 1.6 2.2 2.6 3.1 4.2 5.0 5.9
3.75 1.0 1.2 1.7 2.0 2.3 3.1 3.8 4.4
3.1 1.0 1.4 1.6 1.9 2.6 3.1 3.6
2.25 1.0 1.2 1.4 1.9 2.3 2.6
1.9 1.0 1.2 1.6 1.9 2.2
1.6 1.0 1.3 1.6 1.9
1.2 1.0 1.2 1.4
1 1.0 1.2
0.85 1.0
Latitude
rel. to
Ref. 0.38 0.51 0.61 0.84 1.00 1.19 1.58 1.90 2.24
Narrow
Latitude
Wide
“Optimal display processing for digital radiography,” M. Flynn, M. Couwenhoven, W. Eyler, B. Whiting, E. Samei, D. Foos, R. Slone,
and E. Marom, Proc. SPIE 4319-36, 2001.
35
36
9

Signal Equalization (cont.)
Optimal PA
View
Kodak T-Mat G
Film detail
contrast with 2X
extended latitude
Equalization Processing Artifact
Properly Processed
Halo
Artifact
Overprocessed
37
38
Edge Restoration
Edge Restoration (cont.)
MTF
1
0.8
0.6
0.4
0.2
0
Modulation Transfer Function
CR Hi-Res Screen
CR Std Screen
CsI DR
Selenium DR
Selectively
boost high
frequency
0 1 2 3 4 5
Spatial Frequency (cycles/mm)
High-frequency
signal is
suppressed by
system MTF.
• Selectively boost high-frequency signals based on
– Exam type
– Brightness
– Exposure
– Diagnostic features
– Capture device characteristics
• Multi-frequency processing (2D spatial)
– Kodak: EVP & USM (Enhanced Visualization Processing & UnSharp Mask)
– AGFA: MUSICA (MUlti-Scale Image Contrast Amplification)
– Konica: Hybrid (Mutil-Resolution Hybrid Processing)
– Fuji: USM & MFP (Multi-Objective Frequency Processing)
– Philips: UNIQUE (UNified Image QUality Enhancement)
39
40
10

Edge Restoration (cont.)
Edge Restoration (cont.)
Unsharp Mask Processing
USM Gain
1 .2
1
M o d u la tio n T ra n s fe r F u n c tio n
K = 5, g = 1.2
B a e li n e
U S M 1
U S M 2
Original
Image
-
+
High-Freq.
Image
Output
Image
MTF
0 .8
0 .6
0 .4
0 .2
K = 3, g = 1.5
USM
Kernel Size
Blurred
Image
0
0 1 2 3 4 5
S p a tia l F re q u e n c y (c y c le s /m m )
41
42
Edge Restoration (cont.)
Multi-Frequency
Processing
1
Edge Restoration (cont.)
Processing Artifact
Halo
Artifact
2
3
+
Original
Image
…
n
Edge-Enhanced
Image
n+1
Properly Processed
Overprocessed
43
44
11

Edge Restoration (cont.)
Noise Suppression
Processing Artifact
Properly Processed
Overprocessed
Halo
Artifact
• It is desired to drive toward lower x-ray exposures to reduce
patient dose
• The appearance of noise increases as exposure level is decreased
• A predominant source of noise in digital radiography is generally
the quantum noise.
• Noise suppression should be signal dependent, it should be
applied only to areas of the image that have a low SNR.
• A noise suppression algorithm needs to reduce the appearance of
noise while preserving diagnostic detail.
“Observer study of a noise suppression algorithm for computed radiography images,” M. Couwenhoven et al, Proc. SPIE 5749, 318-327, 2005.
45
46
Noise Suppression (cont.)
Suppress the noise in low
signal areas and phase out
suppression in high signal
areas
High Signal Areas / Less
Dense Anatomy
Low Signal Areas / More
Dense Anatomy
Noise Suppression (cont.)
Noise Suppression
• 2D spatial processing
• Applies to high freq. signals
• Signal dependent
• Balance between sharpness &
noise
47
48
12

Display Compensation
Display Compensation (cont.)
• Image pixel values can be mapped for different output devices
– Film printer (monochrome 1)
– Softcopy display (monochrome 2)
• Both capture and output devices need to be configured properly
• Output device calibration is critical to optimal image display
– Different dynamic range and response
– CRT vs. flat-panel
– Images from multi-vendors viewed at same PACS workstation
– Archived images
• DICOM Part 14 specifies grayscale display standard function
(GSDF)
• AAPM TG-18 specifies display QA & QC testing
Luminance
Workstation Calibration Check
100.00
10.00
1.00
Aim Luminance
Measured Luminance
0.10
0% 20% 40% 60% 80% 100%
SMPTE Patch
Image looks darker
throughout the dark regions
49
50
Reference
Non-calibrated Display
Display Compensation (cont.)
Monitor Responses
Monitor Response Examples
1000
Luminance
100
10
CRT
Flat-Panel
1
0 1024 2048 3072 4096
Image displayed at CRT
looks brighter
Image Processing Features
0.1
DDL
51
Reference (flat-panel)
CRT
52
13

Stationary Grid Suppression
Automatic Grid Line Detection & Suppression
Moiré Pattern
A B C
(A) CR chest image minified by a factor of 0.2 showing a grid-caused Moiré pattern
(B) Filtered image fragment without Moiré pattern
(C) Difference image
“Antiscatter stationary grid artifacts automated detection and removal in projection radiography imagery,”
I. Belykh and C. Cornelius, Proc. SPIE 2000.
53
54
Long-Length Imaging
Screen Film Systems
35cm x130cm
and
35cm x 90cm
Cassettes
Computed Radiography
Currently Limited to
35cm x 43cm
Long-Length Imaging with CR
• Multiple 35cm x 43cm CR screens arranged in
alternating and partially overlapping fashion for patient
imaging
• Storage phosphor screens scanned individually
• Image processing software used to automatically
– Determine CR screen sequence and orientation
– Correct for magnification, translation, and rotation
among individual screens
– Remove redundant image data in the overlap region
– Construct (stitch) a large geometrically accurate
composite
– Eliminate the seam lines in the composite image
• Composite image stored to PACS for interpretation
55
56
14

Scan Individually
Storage Phosphor
Screens
Long-Length Imaging (cont.)
Image Processing
to Construct
Large Composite
57
58
“Fully automatic and reference-marker-free image stitching method for
full-spine and full-leg imaging with computed radiography,” X. Wang, D.
H. Foos, J. Doran, and M. K. Rogers, Proc. SPIE 5368, p361-369, 2004.
Long-Length Imaging (cont.)
• Measurements from CR equivalent to screen-film
– Angular
– Absolute distance
• Visual quality of CR superior to S/F
– Wide exposure latitude of CR
– Equalization processing
• 35% retake rate with S/F reduced to 0% retake rate with CR
– 43 cm width
– Equalization processing
Dual-Energy Imaging
High kVp Radiograph Tissue-Only Bone-Only
Images acquired at high and low energies are processed
to selectively cancel overlying tissues
59
60
Images courtesy of Dr. Jeff Siewerdsen at the Ontario Cancer Institute, Princess Margaret Hospital
and University of Toronto.”
15

Mammography Processing
S/F: A large dynamic range is needed to detect and display all
parts of the breast with good contrast
Mammography Processing (cont.)
Recent advances in s/f mammography
Min-R EV 150 system vs. Min -R 2000 system
Min-R EV 150 vs Min -R 2000 System
6.5
6.0
5.5
5.0
4.5
4.0 Higher contrast
2.5
Better 3.5
2.0 visualization Density of
breast 3.0
1.5 parenchyma
Sharp toe
High Dmax and shoulder contrast
Results in greater overexposure latitude due to
higher upper scale contrast
1.0
Results in “whiter whites”, more “sparkle”,
improved visibility of microcalci fications
0.5
0.0
0.0 0.5 1.0 1.5 2.0 2.5
Log E
X-ray Sensitometry
61
62
KODAK MIN -R EV/
EV 150 Screen
KODAK MIN -R
2000/2000 Screen
Mammography Processing
Oncology Processing
Digital Mammography
• Wide dynamic range (> 1000:1) captures all the image information
• Edge restoration enhances image details of different sizes
• Equalization processing compresses image latitude while
maintaining contrast
Simulation
Localization
Verification
Black surround
Signal Equalization
S/F look
Edge
Enhanced
Equalized
look
“Method for contrast-enhancement of digital portal images,” S. Young, W. E. Moore, D. H. Foos, US Patent US6836570 B2
63
64
16

Quality Control Testing
Test Phantom for Kodak DIRECTVIEW
Total Quality Tool
0.5 mm Cu
1.0 mm Al
180 cm
10.0 0.2 mR
@ 80 kVp
18 x 24 cm
24 x 30 cm
35 x 43 cm
65
66
KODAK TQT User Interface
Test Result Details
MTF (slow scan)
29.5%
67
68
17

Quality Control Testing (cont.)
Automated Image Quality Control Tool
Precise and accurate quality control testing
Highly reproducible quantitative results
Detects sub-visible changes in CR image quality
performance to initiate timely preventive
maintenance
Avoids hours of tedious and labor-intensive effort
with a highly automated procedure
Full data reporting in Excel format
Summary
• Tone scale processing establishes the overall image brightness and
global contrast
• Edge restoration enhances detail contrast
• Signal equalization extends the latitude that can be visualized while
maintaining detail contrast
• Edge restoration, signal equalization, and noise suppression are 2D
spatial processing
• Multi-frequency has been widely adopted, yet users should be aware
of processing artifacts
• Display processing are becoming easier and more intuitive to use
• Image processing provides many new features unique to digital
capture
• Image processing can provide many automations to improve work
efficiency
69
70
Acknowledgements
• Richard VanMetter
• Lori L. Barski
• William J. Sehnert
• Lynn M. Fletcher-Heath
71
18