Segmentation of multiple sclerosis lesions from MR brain images using the principles of fuzzy-connectedness and artificial neuron networks

Abstract

Segmentation is an important step for the diagnosis of multiple sclerosis. In this paper, a method for segmentation of multiple sclerosis lesions from Magnetic Resonance (MR) brain image is proposed. The proposed method combines the strengths of two existing techniques: fuzzy connectedness and artificial neural networks. From the input MR brain image, the fuzzy connectedness algorithm is used to extract segments which are parts of Cerebrospinal Fluid (CSF), White Matter (WM) or Gray Matter (GM). Segments of the MRI image which are not extracted as part of CSF, WM or GM are processed morphologically, and features are computed for each of them. Then these computed features are fed to a trained artificial neural network, which decides whether a segment is a part of a lesion or not. The results of our method show 90% correlation with the expert's manual work.

Segmentation of multiple sclerosis lesions in MR images: a review

“…Admasu et al [84] suggested an improvement to the Udupa's approach by using an ANN, instead of user, to make decision about the void pixels. This method is not dependant on the used MR modality.…”

“…Admasu et al [84] Using fuzzy connectedness as mentioned in Udupa et al [82] and reducing false classified segments by means of ANN FN=2.7? FP=7% Boudraa et al [85] The first pass of FCM on PD-w images to discard MS/CSF and the second pass for detecting MS lesions, rejection of small lesions FN=0 SI=0.65…”

“…Shiee et al [99] used statistical and topological atlases to detect WMLs. The fuzzy connectedness performance was enhanced using ANN by Admasu et al [84].…”

Computer-Aided Diagnosis System for Classification of Multi- Class Brain Tumor in MRI Images using Hybrid Texton Structure Descriptor and Particle Swam Optimized Neural Network

“…In medical image processing, Magnetic Resonance Imaging (MRI) plays an essential role that provides the detailed anatomical information of any part of the body. It is an important diagnostic tool for tumor, cancer, and other dangerous diseases (Fu, 1981), (Admasu, 2003).…”

Medical Image Segmentation using Modified Morphological Reconstruction

“…Unsupervised algorithms are fully automatic and partition the regions in feature space with high density [8]. The different unsupervised algorithms are Feature-Space based Techniques, Clustering (K-means algorithm, C-means algorithm, E-means algorithm), Histogram thresholding, Image-Domain or Region Based Techniques (Split-and-merge techniques, Region growing techniques [9], Neural-network based techniques, Edge Detection Technique), Fuzzy Techniques [10], Hybrid techniques, etc. In recent years, mathematical morphology is a well-known technique used in image processing and computer vision [11] [12] [13].…”