An Robust Algorithm for Segmenting Multiple Sclerosis from Magnetic Resonance Images

EasyChair Preprint 7534

Abstract

Multiple Sclerosis (MS) is an autoimmune inflammatory disease that leads to lesions in the central nervous system. Magnetic Resonance Imaging (MRI) provide sufficient imaging contrast to visualize and detect MS lesions, particularly those in the white matter (WM). Medical image segmentation is an essential step for most consequent image analysis tasks. The proposed segmentation algorithm is composed of three stages: segmentation of the brain into regions using Fuzzy Particle Swarm Optimization (FPSO) in order to obtain the characterization of the different healthy tissues (White matter, grey matter and cerebrospinal fluid (CSF)). After the extraction of WM, atypical data (outliers) is eliminated using Fuzzy C-means algorithm, and finally, we introduce a Mamdani-type fuzzy model to extract MS lesions among all the absurd data. Although the FCM algorithm yields good results for segmenting noise free images, it fails to segment images corrupted by noise, atypical data (outliers) and other imaging artifact. The purpose of this study is to segment high dimensional data of WM lesions using Fuzzy Possibilistic C-means (FPCM). This approach is a generalized version of FCM algorithm.  The objective of the work presented in this paper is to obtain an improved accuracy in segmentation of WM. Comparison results to the method of FPSOFCM showed that the defuzzification of the atypical data of the segmentation was 56.79 showing that the proposed FPSOFPCM outperformed the other method (FPSOFCM).

Keyphrases: FPCM, FPSO, MRI, Mamdani., Multiple Sclerosis, Segmentation

@booklet{EasyChair:7534,
  author    = {Hakima Zouaoui and Abdelouahab Moussaoui},
  title     = {An Robust Algorithm for Segmenting Multiple Sclerosis from Magnetic Resonance Images},
  howpublished = {EasyChair Preprint 7534},
  year      = {EasyChair, 2022}}