Medical Image Segmentation using Fuzzy Expectation Maximization based Clustering

In modern world medical imaging plays a major role for diagnosing diagnose of many neurodegenerative and psychiatric diseases for analyzing purpose. Recent developments made a rapid change to detect the image with high quality and provide exact report. While processing the image, segmentation is an important aspect because it is used to remove the noise from original image. Hence this work contributes the segmentation process with Fuzzy clustering process. Fuzzy sets rough sets and combination of fuzzy forms the best role in diagnosis.  Segmentation deals with the segmentation of tumour in Computer Tomography images for improving quality in diagnosis. Here several fuzzy C clustering algorithms such as Fuzzy C Means, Intuitionistic Fuzzy C-Means (IFCM) and are compared and semi-automated Fuzzy Expectation Maximization (FEM) based techniques among that best Fuzzy C clustering is considered as proposed algorithm. The proposed design compared made with other methods in the following parameters such as mean square error and misclassification error. The merit of this proposed design is to remove the noisy spots and less sensitive to noise.