An adaptive fuzzy clustering algorithm for medical image segmentation

Abstract

An adaptive fuzzy clustering algorithm is presented for the fuzzy segmentation of medical images. By using a novel dissimilarity index in the cost functional of the fuzzy clustering algorithm our algorithm is capable of utilising contextual information in a 3/spl times/3 neighborhood to impose local spatial homogeneity, as well as the usual feature space homogeneity. This has the effects of smoothing out random noise and resolving classification ambiguities. By introducing a multiplicative bias field into the cost functional, artifacts due to smooth, non-uniform intensity variation can also be corrected. The bias field is regularized by a Laplacian term which forces the bias field to resist bending and to be smooth. To solve for the bias field, the full multigrid algorithm is employed. Experimental results on a synthetic image and a simulated MRI brain image with noise and non-uniform intensity variation have illustrated the effectiveness of the proposed algorithm.