Automated brain tissue segmentation and MS lesion detection using fuzzy and evidential reasoning

This paper presents a fuzzy and evidential reasoning approach for segmenting main brain tissues: white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF), as well as detecting multiple sclerosis lesions (MS) based on multi-modality MR images. The method performs intensity based tissue segm...

Full description

Saved in:
Bibliographic Details
Main Authors: Hongwei Zhu, Basir, O.
Format: Conference Proceeding
Language:English
Subjects:
Biological neural networks
Brain
Degenerative diseases
Fuzzy reasoning
Fuzzy systems
Image analysis
Image databases
Image segmentation
Lesions
Multiple sclerosis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a fuzzy and evidential reasoning approach for segmenting main brain tissues: white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF), as well as detecting multiple sclerosis lesions (MS) based on multi-modality MR images. The method performs intensity based tissue segmentation using a fuzzy Dempster-Shafer evidential reasoning data fusion scheme while MS lesions are detected by means of a fuzzy inferencing scheme. The approach is fully automated and unsupervised. Experiments have been carried out for segmenting 15 axial slices of multi-modality MR images obtained from the Simulated Brain Database (SBD). The average overall accuracy is 96.77% for segmenting tissues CSF, GM, and WM. The average sensitivity is 84.34%, and the average similarity index is 81.94% in MS detection in terms of ground truth images.
DOI:10.1109/ICECS.2003.1301695