Automated Detection and Grading of Brain Tumors using Particle Swarm Optimization

Abstract

There has been great impact on the field of medical imaging by the advancements in the computer technology as new and improved techniques of data acquisition newline, analysis, processing and visualization have evolved. Magnetic resonance images (MRI) provides information about potential abnormal tissues necessary for medical newlinefollow up. Brain MRI gets additional importance in medical science as it is the only preliminary method of diagnosing a brain tumor. This thesis addresses the newlineproblem of automated detection and grading of brain tumors. The important points in this research have been to develop and implement a robust algorithm for newlinethe classification of the Brain MRI images as normal or abnormal, to determine the grade of tumor present and to study the applicability of Particle Swarm Intelligence newlinein the field of medical segmentation and classification. Two traditional and two new methods are proposed and implemented for the detection of tumor from human newlinebrain MRI. The new methods were based on Particle Swarm Optimization. It was found experimentally that the PSO based method yielded better results compared to the newlinestandard methods.PSO based method was also applied for grading of the tumors as per the WHO standard. newline