By Topic

Fuzzy expert system for edema segmentation

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
S. Loncaric ; Fac. of Electr. Eng., Zagreb Univ., Croatia ; D. Kovacevic ; D. Cosic

A method for segmentation of edema region in a computed tomography (CT) image of the brain is presented in this paper. The edema region is a dark region surrounding the intracerebral brain hemorrhage (ICH) region. Edema segmentation, in general, is difficult because of very subtle grayscale variations. The proposed edema segmentation procedure consists of two major parts. Fuzzy feature extraction is performed in the first part of the procedure. Extracted features include local contrast and distance from ICH region. The fuzzy features are then input into the rule-based expert system as fuzzy facts. In the second part, a fuzzy expert system is used to perform the actual segmentation. A set of rules is formulated to represent the knowledge about the possible image structure. The set of rules is constant and does not depend on the input image. The fuzzy facts extracted in the first part of the procedure are derived from the particular image to be segmented. Fuzzy-CLIPS software package has been used to derive a conclusion from the facts based on the rules. The conclusion represents the segmented edema region. Experiments have shown that the method can be used for edema segmentation

Published in:

Electrotechnical Conference, 1998. MELECON 98., 9th Mediterranean  (Volume:2 )

Date of Conference:

18-20 May 1998