Fast FCM Algorithm for brain MR Image Segmentation

In medical applications it is very important for a physician to be informed of patient situation as soon as possible especially in emergency circumstances. Therefore all efficient agents in patient health must be fast even medical algorithms such as clustering ones. Among clustering methods Fuzzy C-Means (FCM) clustering has been frequently used for segmentation of medical images. In this paper an optimized method is presented to decrease execution time of standard FCM. Applying the new method simultaneously decreases convergence time and iteration numbers of FCM. Experimental results show that the proposed Fast FCM (FFCM) spend moderately half time of standard FCM and number of its iterations is decreased significantly; Quantitative assessment using conventional fuzzy validation functions shows similar performances of FCM and FFCM.