Abstract
In this paper, we present alternative Kernelized FCM algorithms (KFCM) that could improve magnetic resonance imaging (MRI) segmentation. Then we implement the proposed KFCM method with considering some spatial constraints on the objective function. The algorithms incorporate spatial information into the membership function and the validity procedure for clustering. We use the intra-cluster distance measure, which is simply the median distance between a point and its cluster centre. The number of the cluster increases automatically according the value of intra-cluster, for example when a cluster is obtained, it uses this cluster to evaluate intra-cluster of the next cluster as input to the KFCM and so on, stop only when intra-cluster is smaller than a prescribe value. The most important aspect of the proposed algorithms is actually to work automatically. Alterative is to improve automatic image segmentation.
These methods are applied on two different sets: reference images, for objective evaluation based on estimation of segmentation accuracy and time, and non reference images, for objective evaluation based on combined judgment of opinions of specialists.