Automatic Detection Of Brain Tumor Through Magnetic Resonance Image

Abstract: Magnetic Resonance Imaging (MRI) plays an important role in Brain Tumor diagnisation in advanced stages. It is a form of medical imaging using nuclear magnetic resonance of protons in the body. Segmentation process to extract suspicious region from complex medical images is very important. Brain image segmentation is a complex and challenging part in the Medical Image Processing.This research deals with new approaches for brain Tumor detection using Meta Heuristic Algorithm. It aims to develop and effective algorithm for the segmentation of Brain MRI images.Pre-processing, Enhancement and Segmentation are deeply analyzed in this work. In the Pre- processing system film artifacts and unwanted portions of MRI brain image are removed. In the Enhancement process the noise and high frequency components are removed using median filters.The Segmentation process it has three different approaches like block based (non algorithmic), PSO and HPACO algorithm segmentation. MRI helps to diagnose accurate condition of the Brain Tumor and to give necessary proper treatment.

Keywords: MRI Brain Image segmentation, Particle Swarm Optimization, Hybrid Parallel Ant Colony Optimization, FCM Algorithm