A novel fuzzy approach for segmentation of brain MRI

In this paper, an unsupervised fuzzy technique for segmentation of brain magnetic resonance (MR) images is presented, which combines fuzzy clustering algorithm with maximum a posteriori (MAP) criterion. As fuzzy C-means (FCM) tends to balance the number of points in each cluster, cluster centers of smaller clusters are drawn to larger adjacent clusters. In order to overcome this problem occurred in the fuzzy segmentation of MR images, the technique is done in two steps. In the first step, FCM algorithm is used to segment the brain into four major classes of white matter, gray matter, cerebrospinal fluid (CSF) and background. In the second step, the results are refined by a new MAP criterion, which is improved by fuzzy technique. Experimental results show that our approach is effective and can get higher segmentation accuracy than that of the conventional FCM segmentation.