Parameter estimation of finite mixtures using the EM algorithm and information criteria with application to medical image processing

A method for parameter estimation in image classification or segmentation is studied within the statistical frame of finite mixture distributions. The method models an image as a finite mixture. Each mixture component corresponds to an image class. Each image class is characterized by parameters, su...

Full description

Saved in:
Bibliographic Details
Published inIEEE Transactions on Nuclear Science (Institute of Electrical and Electronics Engineers); (United States) Vol. 39; no. 4; pp. 1126 - 1133
Main Authors Liang, Z., Jaszczak, R.J., Coleman, R.E.
Format Journal Article Conference Proceeding
LanguageEnglish
Published United States IEEE 01.08.1992
Subjects
663620 > Physics of Radiations Other Than Neutrons > (1992-)
990200 > Mathematics & Computers
ALGORITHMS
Biomedical imaging
BIOMEDICAL RADIOGRAPHY
CALCULATION METHODS
Computational modeling
DIAGNOSTIC TECHNIQUES
GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
Image classification
IMAGE PROCESSING
Image reconstruction
Image segmentation
Imaging phantoms
MATHEMATICAL LOGIC
Maximum likelihood estimation
MAXIMUM-LIKELIHOOD FIT
MEDICINE
MOCKUP
NUCLEAR MEDICINE
NUCLEAR PHYSICS AND RADIATION PHYSICS
NUMERICAL SOLUTION
Parameter estimation
PARAMETRIC ANALYSIS
PHANTOMS
Pixel
PROCESSING
RADIOLOGY
RADIOLOGY AND NUCLEAR MEDICINE
STRUCTURAL MODELS 550602 > Medicine > External Radiation in Diagnostics > (1980-)
Online AccessGet full text
ISSN0018-9499
1558-1578
DOI10.1109/23.159772

Cover

More Information
Summary:A method for parameter estimation in image classification or segmentation is studied within the statistical frame of finite mixture distributions. The method models an image as a finite mixture. Each mixture component corresponds to an image class. Each image class is characterized by parameters, such as the intensity mean, the standard deviation, and the number of image pixels in that class. The method uses a maximum likelihood (ML) approach to estimate the parameters of each class and employs information criteria of Akaike (AIC) and/or Schwarz and Rissanen (MDL) to determine the number of classes in the image. In computing the ML solution of the mixture, the method adopts the expectation maximization (EM) algorithm. The initial estimation and convergence of the ML-EM algorithm were studied. The accuracy in determining the number of image classes using AIC and MDL is compared. The MDL criterion performed better than the AIC criterion. A modified MDL showed further improvement.< >
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-2
ObjectType-Feature-1
content type line 23
SourceType-Conference Papers & Proceedings-1
ObjectType-Conference-3
ObjectType-Feature-2
ObjectType-Conference Paper-1
ObjectType-Article-3
SourceType-Scholarly Journals-1
CONF-911106--
None
ISSN:0018-9499
1558-1578
DOI:10.1109/23.159772