Fuzzy C-means clustering algorithm with level set for MRI cerebral tissue segmentation

Abstract

The brain is the most complex organ in the human body, and it consists of four regions namely, gray matter, white matter, cerebrospinal fluid and background. It is widely accepted as an imaging modality for detecting a variety of conditions of the brain such as tumours, bleeding, swelling, infections, or problems associated with blood vessels. Brain images mostly contain noise, inhomogeneity and sometimes deviation. Therefore, accurate segmentation of brain images is a very difficult task. This thesis presents a new approach of Magnetic Resonance Imaging (MRI) brain tissue segmentation, which consists of three main phases: (1) Noise removal using median filter, (2) Tissue clustering based on the fuzzy c-means, and (3) Tissue segmentation using the fuzzy level set method, which finally separates white matter from gray matter. The results show that the segmentation’s accuracy rates of 98% is achieved when tested on 100 samples of MRI brain images atlas dataset.