38
Image Quality, T.T. Liu, Spring 2006 Image Quality Lecture 1 Resident Physics Course March 27, 2006
Image Quality, T.T. Liu, Spring 2006
Image QualityLecture 1
Resident Physics CourseMarch 27, 2006
Image Quality, T.T. Liu, Spring 2006
Topics
Image Contrast
Resolution
Noise
Image Quality, T.T. Liu, Spring 2006
Contrast
Contrast is the difference in image gray scalebetween closely adjacent regions in an image.
Types of Contrast are:
Subject Contrast
Detector Contrast
Radiographic Contrast
Digital Image Contrast
Displayed Contrast
Image Quality, T.T. Liu, Spring 2006
Contrast
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Contrast
Image Quality, T.T. Liu, Spring 2006
!
A = N0exp("µx)
B= N0exp("µ(x + z))
CS
=A"B
A
=N0exp("µx) "N
0exp("µ(x + z))
N0exp("µx)
=1"exp("µz)
Subject Contrast
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006!
CS
=1"exp("µz)
As attenuation increases, so does subject contrast
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Characteristic Curves
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Detector Contrast = Slopeof Characteristic Curve
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Display Contrast
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Intuition: the PSF is the response of a system toan input of infinitesimal width and unit area.
Point Spread Function (PSF)
Blurred ImageOriginalImage
Image Quality, T.T. Liu, Spring 2006
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Stationary PSF
Nonstationary PSF
For convenience, usually assume PSF is stationary.
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Full Width Half Maximum (FWHM) is a measure of resolution.
Prince and Link 2005
Image Quality, T.T. Liu, Spring 2006
Convolution
Image Quality, T.T. Liu, Spring 2006
Convolution with PSF blurs object
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Blurring Mechanisms
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Image Quality, T.T. Liu, Spring 2006
Image Quality, T.T. Liu, Spring 2006
Magnification
!
Mobject
=A+B
A
=Source to Image Distance (SID)
Source to Object Distance (SOD)
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Blurring due to magnification of finite source
!
Msource
=B
A
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006 Prince and Link 2005
Image Quality, T.T. Liu, Spring 2006
Image Quality, T.T. Liu, Spring 2006
Line Spread Function
Edge Spread Function
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Spatial Frequency
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Spatial Frequencies
Prince and Link 2005
Image Quality, T.T. Liu, Spring 2006
Modulation Transfer Function (MTF) orFrequency Response
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Modulation Transfer Function
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
System MTF = Product of MTFs of Components
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
MTF = Fourier Transform (LTF)
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Useful Approximation
!
FWHMSystem
= FWHM1
2+FWHM
2
2+LFWHM
N
2
Example
FWHM1
=1mm
FWHM2
= 2mm
FWHMsystem
= 5 = 2.24mm
Image Quality, T.T. Liu, Spring 2006
Image Quality, T.T. Liu, Spring 2006
Answer: B; step wedges are used to create a gray scale which can be used toevaluate the contrast of the image receptor system
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Noise and Image Quality
Prince and Links 2005
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
What is Noise?Fluctuations in either the imaging system or the objectbeing imaged.
Quantization Noise: Due to conversion from analogwaveform to digital number.
Quantum Noise: Random fluctuation in the number ofphotons emittted and recorded.
Thermal Noise: Random fluctuations present in allelectronic systems. Also, sample noise in MRI
Other types: flicker, burst, avalanche - observed insemiconductor devices.
Structured Noise: physiological sources, interference
Image Quality, T.T. Liu, Spring 2006
Histograms and Distributions
3rd grade heights 6th grade heightsBushberg et al 2001
Image Quality, T.T. Liu, Spring 2006
Gaussian Distribution
1, 2, and 3 standard deviation intervals correspond to 68%,95%, and 99% of the observations
Bushberg et al 2001
Image Quality, T.T. Liu, Spring 2006!
Poisson Distribution describes x - ray counting statistics.
Gaussian distribution is good approximation to Poisson when " = X
Bushberg et al 2001
