Tumor Detection using Particle Swarm Optimization to Initialize Fuzzy C-means

Image processing techniques are extensively used in different medical fields for earlier detection and treatment stages, where time factor is more important to find abnormality issues in target images in several tumors. Now a days tumor is discovered at advanced stages with the help of Magnetic Resonance Imaging (MRI).This paper proposes an approach which combines the Particle Swarm Optimization (PSO) techniques and Fuzzy C-Means (FCM) algorithm to perform image segmentation on Magnetic Resonance Imaging (MRI).The FCM algorithm has some limitations that it requires initialization of cluster centroids and the number of cluster. In this work, the PSO techniques is applied to the MRI medical images for the purpose of initialization of cluster centroids, which could overcome the requirement of manual initialization of FCM algorithm. Natural Computing (NC) is a novel approach to solve real life problems inspired in the life itself. A diversity of algorithms had been proposed such as evolutionary techniques and particle swarm optimization (PSO). This approach, together with fuzzy c means (FCM), give powerful tools in a diversity of problems of optimization, classification, data analysis and clustering.