0% found this document useful (0 votes)
35 views37 pages

DR-Lab 08b

The document discusses the differences between digital radiology (DR) displays and traditional film, focusing on the technology of CRTs and LCDs. It highlights the importance of display quality, calibration, and the impact of viewing conditions on image interpretation. Additionally, it outlines the potential consequences of degraded display performance and the need for regular quality control assessments to ensure accurate medical imaging.

Uploaded by

srinibme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
35 views37 pages

DR-Lab 08b

The document discusses the differences between digital radiology (DR) displays and traditional film, focusing on the technology of CRTs and LCDs. It highlights the importance of display quality, calibration, and the impact of viewing conditions on image interpretation. Additionally, it outlines the potential consequences of degraded display performance and the need for regular quality control assessments to ensure accurate medical imaging.

Uploaded by

srinibme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd

Radiation Protection in Digital

Radiology

Optimising DR Displays
L08

IAEA
International Atomic Energy Agency
Educational Objectives

• List three major differences between DR


displays and transilluminated films
• Explain how CRTs and LCDs differ with
respect to the display of medical images
• Appreciate the differences between medical and
commercial grade flat panels
• Give an example of how differences between
a technologist’s display and a radiologist’s
display can contribute to unnecessary
radiation exposure.
IAEA 2
Radiation Protection in Digital Radiology L08 Optimising DR Displays
The Cathode Ray Tube (CRT) is a fifty-
year old device for displaying electronic
images
• Electrons produced in a
vacuum tube strike a
luminescent screen
• The path of the electron
beam is deflected by a coil
• The amount of light
produced in any position is
related to the intensity of
the electron beam at that
time
• Color can be produced by
means of a shadow mask
or aperture grill

Bushberg et al. 2002 The Essential Physics of Medical Imaging 2nd


IAEA 3
Radiation Protection in Digital Radiology L08 Optimising DR Displays
The CRT provides a dynamic display
of even stationary images
• The electron beam scans
across the face of the display in
a raster fashion
• The standard video frame rate
is 30 fps (SMPTE)
• Historical lowest rate to avoid
perception of flicker
• Convenient: ½ of 60 Hz
• Alternate frame rates, such as
24 fps for motion pictures
• Interleaved display would use 2
frames for one image – higher
spatial resolution
• A picture element (pixel) is an
arbitrary segment of a scan line

Bushberg et al. 2002 The Essential Physics of Medical Imaging 2nd


IAEA 4
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Active Matrix Liquid Crystal Displays
(AMLCD) present electronic images by a
different method
• The LCD controls the transmission
of a uniform backlight “Flat Panel”
• The transmission of light through a
given LC cell can be considered
binary (on/off)
• Actually much more complex
• A pixel is composed of six
components arranged in a chevron
pattern
• Two domains
• Three colors
• “Active” refers to control of each
pixel independently via TFT array

Bushberg et al. 2002 The Essential Physics of Medical Imaging 2nd


IAEA 5
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Light transmitted by a flat panel is a
composite of pixel component
states
• Un-calibrated response is irregular
• Display controller needs 10-12 bits for
medical applications

IAEA 6
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Ability to produce color, limits
performance of commercial flat
panels
• Color filter allows only 3-5% transmittance of
backlight vs. 8-15% for monochrome
• Combination of sub-pixel intensities to yield
true white is additional complication

IAEA 7
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Off-axis viewing is problematic with
flat panels

• CRT emissive luminance is Lambertian, the


intensity appears the same from all viewing
angles
• Flat panel transmissive luminance is non-
Lambertian, the intensity appears different from
any viewing angle other than normal (rounds?)
• This is not a problem for single viewer, unless
the viewer must move (interventional?,
surgery?)
• Even if radiographer has same display as
radiologist, off-axis viewing differences can
cause discrepancy in rendering the image
IAEA 8
Radiation Protection in Digital Radiology L08 Optimising DR Displays
“How do you know that it’s okay for the
physician to view images on that
display?”
• Capabilities of the display technology
• Characteristics of the images to be viewed
• Idiosyncrasies of human visual system
• Configuration of the display device
• Calibration of the display device
• Local viewing environment
• Workstation software and controls
• Viewing task to be performed
• Active maintenance of display quality
• Ambient lighting condition

IAEA 9
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Digital imaging is often“display-
limited”

• The information contained in the image


cannot be presented by the display in a
single rendering
• Spatial resolution
• Contrast resolution
• Dynamic range
• Workstations address this problem by
software tools to display a portion of the
image at full resolution
• Zoom and Pan
• Window-width and window-level
IAEA 10
Radiation Protection in Digital Radiology L08 Optimising DR Displays
The smallest feature that can be
displayed is limited by the number of
pixels
CRT Beam spot size for 300 X 400 mm Field
Pixels Array size Spot size
(mm)
1MP 900 X 1100 0.35
"1K X 1K”
2 MP 1200 X 0.25
1600
“1K X 1.5K”
3 MP 1500 X 0.20?
2000
“1.5K X 2K”
5 MP 2000 X 0.15
2500
IAEA “2K XMJ2K”
Flynn 2004 11
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Corduroy” artefact

• Interference
pattern
between fixed
grid lines and
down-sampling
rate for display
• Disappeared
on zoom
• Bad choices
• Display default
magnification
factor
• Line rate of
grid

IAEA 12
Radiation Protection in Digital Radiology L08 Optimising DR Displays
This is one place where size doesn’t
matter
• The physical size of the display affects only viewing
distance
• The pixel matrix is what matters
• Large displays are useful for interventional, surgical, and
multiple simultaneous observers
• Projection Displays
• Plasma Displays
• DLP displays
• Small displays may be useful for reference or
image navigation
• PDA
• Cell Phone
• LEP (light emitting polymers)
IAEA 13
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Spatial resolution of the display is
limited by blur

• Blur is a major factor in CRT displays because of


the dynamic way the pixel is produced
• Blur is much improved in flat panels because of the
stationary structure of the pixel
• a 3MP flat panel performs as well as a 5MP CRT display

IAEA 14
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Electronic displays are limited with
respect to maximum luminance

• Typical light box luminance is >500 fL (1700 cd/m 2)


• Typical medical CRT is 70-90 fL (240-300 cd/m 2 )
• minimum ACR is 50 fL (171 cd/m2 ) for primary
interpretation
• Typical general purpose CRT is 30 fL (100 cd/m 2 )
• Medical monochrome LCD is 200 fL (700 cd/m2 )
• Typical consumer electronics color LCD display is
60 fL (200 cd/m2 )
• Paper SSA20-06 Visser M et al. RSNA 2005
describes prototype backlight up to 2000 cd/m 2

IAEA 15
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Because of their limited luminance, the
viewing environment for electronic
displays is critical
• Ambient illumination limits the contrast that can be
appreciated from an electronic display
• The higher the ambient illumination, the higher the
maximum luminance the display will need.
• The more reflective the display, the lower the allowable
level of ambient illumination.
• Big problem with CRT
• Changes in ambient illumination strongly affect contrast in
the dark areas of the display, so one strategy is to raise the
minimum luminance.
• Convenient for flat panels with poor black levels

IAEA 16
Radiation Protection in Digital Radiology L08 Optimising DR Displays
The luminance function of electronic
displays is not appropriate for viewing
digital images
• The luminance function is modified by a software look-up-
table (LUT) in an attempt to elicit equal human visual
response for equal changes in grayscale value.
• The mathematics of this transformation are defined in
DICOM Part 14 Grayscale Standard Display Function
(GSDF).
• The result is that a graph of luminance expressed in units of
“just noticeable differences” (JND) is linear with respect to
grayscale value.

IAEA 17
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Luminance (Cd/m^2) versus PV
(s erial num ber A2I-05975)

1000

Luminance (Cd/m^2)
100

Luminance of Measured (Nominal Lmax=600)

properly 10
GSDF

calibrated
display is curved
1
0 2000 4000 6000 8000 10000 12000 14000 16000 18000

function of
Pixel Value

greyscale value JND Index versus Input Pixel Value


(serial number A2I-05975)
600

Luminance 500

translated into 400


JND Index

JND is linear
300
R^2 = 0.99984

function of
200

100
greyscale Series1
Least Squares Fit
0
value 0 2000 4000 6000 8000 10000 12000 14000 16000 18000
Pixel Value
IAEA 18
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Viewer software can affect display
function
Dell Model E771P Color Monitor

600
y = 4.3844x + 60.704
500 R2 = 0.9706

400
Stentor
JND Index

W ebb1000
300
Linear (Stentor)
Linear (W ebb1000)
200

100 y = 4.888x + 36.444


R2 = 0.9856

0
0 20 40 60 80 100
SMPTE %

IAEA 19
Radiation Protection in Digital Radiology L08 Optimising DR Displays
“The best electronic image,
improperly displayed is terrible.”
• CRT monitors degrade over time. LCDs last longer.
• The wrong display Look-up-table (LUT) can spoil a
great electronic image (DICOM Part 14 GSDF)
• Test patterns, notably the SMPTE, can make display
problems obvious.

IAEA 20
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Just because you ran the calibration routine
does not mean the display is DICOM Part
14 compliant
Dome Flatpanel post cal

900
800 y = 5.8183x + 202.51
700 R2 = 0.9721
600
JND index

500
400
300
200
100
0
0 20 40 60 80 100 120
SMPTE%

IAEA 21
Radiation Protection in Digital Radiology L08 Optimising DR Displays
For GSDF conformance, consumer color flat
panels require control of downloadable color
ramps
HPDome
Color Flatpanel post cal

700
600 y = 4.286x + 150.2
500 R2 = 0.9717
JND index

400
300
200
100
0
0 20 40 60 80 100 120
SMPTE%

IAEA 22
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Display problems affect the radiologist’s
ability to practice digital radiology

• Potential errors (hard or soft copy)


• Incorrect GSDF calibration
• Inadequate matrix
• Moire’ (interference) patterns
• Inadequate spatial resolution
• Incorrect or missing demographics or
annotations
• Inadequate viewing conditions
• QC => Radiologist “Film” critique

IAEA 23
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Quantifiable Consequences of
Degraded Performance

• Loss of Contrast Sensitivity


• Loss of Sharpness/Spatial Resolution
• Loss of Dynamic Range
• Increase in Noise
• Decrease in System Speed
• Geometric Distortion
• Artefacts

IAEA 24
Radiation Protection in Digital Radiology L08 Optimising DR Displays
AAPM Task Group 18 has developed
procedures for assessing display
quality
• GSDF
• Luminance Uniformity
• Noise (Low contrast performance)
• Resolution and resolution uniformity (CRT only)
• Veiling Glare (CRT only)
• Geometric Distortion (CRT only)
• Bandwidth Artifacts (CRT only)
• Dead Pixel Count (LCD only)

IAEA 25
Radiation Protection in Digital Radiology L08 Optimising DR Displays
AAPM Task Group 18 report on assessment of
display performance for medical imaging
systems
• Recommended tests
and frequency
• Useful test patterns

http://aapm.org/pubs/reports/OR_03.pdf
IAEA 26
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Some aspects of display
performance can change over time

 GSDF
 Luminance Uniformity
 Noise (Low contrast performance)
 Resolution and resolution uniformity (CRT only)
X Veiling Glare (CRT only)
 Geometric Distortion (CRT only)
 Bandwidth Artifacts (CRT only)
 Dead Pixel Count (LCD only)

IAEA 27
Radiation Protection in Digital Radiology L08 Optimising DR Displays
To assure display quality, you
are going to have to measure it

• Will need photometer


• May need a chromaticity attachment
• Will need test patterns
• MANY available from TG18
• Will need to measure more stuff, more
frequently (monthly) with CRT
• Useful lifetime of CRTs is limited compared to
flat panels

IAEA 28
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Good news is that remote QC
technology is available

• Automatic GSDF calibration


• Automatic monitoring and compensation for
changes in ambient lighting and maximum
luminance
• Remote monitoring, reporting, and adjustment via
SNMP client.
• Luminance level, drive level, system temperature,
etc

Ref: Raimond Pohlman and Jeff Shepard, UT MDACC


IAEA 29
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Difference in appearance on two
GSDF calibrated displays

Acquisition Station PACS Display


Even with proper calibration, viewer interpretation
IAEA of greyscales can differ … 30
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Discrepancy between the DX VOI
LUT and the PACS Linear LUT

• PACS viewer applied


linear LUT to
greyscales intended WW=2747, WL=4897

to have sigmoidal 16384

LUT
• Consequence: clipped 12288

light and dark regions Output Grayscale

DX VOI LUT
8192
PACS Linear LUT

4096

0
0 4096 8192 12288 16384
Input Grayscale

IAEA 31
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Teleradiology – the forgotten
display

• With remote viewing, one can no longer control


what display is going to be used to display the
image.
• ACR Standard calls for transmission and
assessment of SMPTE test pattern weekly.
• Only workable approach is to provide the physician
with an assessment tool at session log-in where he
must affirm that he can see features.

IAEA 32
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Field Effect Display (FED) may
challenge AMLCD

• Can be built as thin as LCD


• Emissive display: no backlight

IAEA 33
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Conclusions:

• Active Matrix Liquid Crystal Displays will


continue to displace Cathode Ray Tube Displays
for medical imaging
• Displays for medical imaging require special
calibration according to DICOM GSDF
• Increasing use of pseudo-color in digital imaging
imposes special demands on displays
• Novel display technologies are likely to find use
in specific limited applications, except possibly
FEDS

IAEA 34
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Answer True or False

• The spatial resolution in flat panel


monitors are better than CRT
• There can be artefacts arising from
display screens
• The display systems can be used in
any kind of environment

IAEA 35
Radiation Protection in Digital Radiology L08 Optimising DR Displays
Answer True or False

• True. Blur in flat panel monitors is less


than the CRT monitors because of the
stationary structure of the pixel.
• True. Corduroy artefact. It is the
interference pattern between fixed
grid lines and down-sampling rate for
display.
• False. Ambient illumination limits the
contrast that can be appreciated from
an electronic display
IAEA 36
Radiation Protection in Digital Radiology L08 Optimising DR Displays
References:
• Bushberg JT, Seibert JA, Leidholdt EM Jr, and Boone JM. The
Essential Physics of Medical Imaging 2nd Ed. Lippincott Williams
and Wilkins. Philadelphia. (2002) 933).
• Flynn MJ. Softcopy Display: Technology, Performance, and
Quality. In Specifications, Performance Evaluations and Quality
Assurance of Radiographic and Fluoroscopic Systems in the
Digital Era. Goldman LW and Yester MV eds. AAPM Monograph
No. 30.Medical Physics Publishing. Madison. (2004) 335-351.
• Baldano A. Principles of Cathode-Ray Tube and Liquid Crystal
Display Devices. In Advances in Digital Radiography: RSNA
Categorical Course in Diagnostic Radiology Physics. (2003) 91-
102.
• Samei E, Badano A, Chakraborty D et al. Assessment of
display performance for medical imaging systems: Executive
summary of AAPM TG18 report. Medical Physics 32(4)
(2005)1205-1225.

IAEA 37
Radiation Protection in Digital Radiology L08 Optimising DR Displays

You might also like