Abstrato

IMPROVED FUZZY POSSIBLISTIC C-MEANS (IFPCM) ALGORITHMS FOR MAGNETIC RESONANCE IMAGES SEGMENTATION

Hussian AbouSora, Said Ghoniemy, Sayed A. Banwan, E.A.Zanaty, Ashraf Afifi

In this paper, we propose a new method called “improved fuzzy possiblistic c-means (IFPCM)” which could improve medical image segmentation. The proposed method combines the fuzzy c-means (FCM) and possiblistic c-means (PCM) functions without considering any spatial constraints on the objective function. It is realized by modifying the objective function of the conventional PCM algorithm with Gaussian exponent weights to produce memberships and possibilities simultaneously, along with the usual point prototypes or cluster centers for each cluster. IFPCM avoids various problems of existing fuzzy clustering methods solves the noise sensitivity defect of FCM and overcomes the coincident clusters problem of PCM. The proposed algorithm is evaluated and compared with the most popular modified possibilistic c-means techniques via application to simulated MRI brain images corrupted with noise. The quantitative results suggest that the proposed algorithm yields better segmentation results than the others for all tested images.