A Modified Fuzzy C-Means Algorithm for Segmentation of Magnetic Resonance Images.

Abstract

This paper presented a new approach for robust segmenta- tion of Magnetic Resonance images that have been corrupted by intensity inhomogeneities and noise. The algorithm is formulated by modifying the objective function of the standard fuzzy C-means (FCM) method to com- pensate for intensity inhomogeneities. A additional term is injected into the objective function to constrain the behavior of membership func- tions with the neighborhood effect. And an adaptive K-means clustering algorithm that initializes the centroids is described. The efficacy of the algorithm is demonstrated on both simulated and real Magnetic Reso- nance images.