Abnormality Segmentation in Brain Images using Adaptive K-FCM Approach

Segmentation of tumors on medical images is not only of high interest in serial treatment monitoring of ”disease burden” in oncologic imaging, but also gaining popularity with the advance of image guided surgical approaches. Adaptive k- Fuzzy C-Means (Adaptive k-FCM) is a unsupervised clustering technique that has been extensively used in image segmentation. Conservative FCM algorithm has an inadequacy; it is not consider spatial information into the account. This causes the exisiting FCM algorithm to work only on definite images with stumpy level of noise. In this paper an improvement to fuzzy clustering is describe. To detect the abnormalities of Brain MRI images using a new spatial Adaptive k-FCM and compare the results with k -means and FCM techniques. To estimate the intensity inhomogeneity, the global intensity is introduced into the coherent local intensity clustering algorithm and takes the local and global intensity information into account. The proposed method has been successfully applied to recorded MR images with desirable results. Our results show that the proposed Adaptive k-FCM algorithm can effectively segment the test images and MR images. Comparisons with other FCM approaches based on number of iterations and time complexity demonstrate the superior performance of the proposed algorithm.