Brain Tumor Identification Using Hybrid Genetic Algorithm with OTSU Segmentation

Abstract

A brain tumour is an abnormal mass of tissue in which cells grow and multiply uncontrollably,  seemingly unchecked by the mechanisms that control normal cells. The symptoms of a brain tumour rely upontumoursize, type, and location. Image processing (IP) is a technique to convert an image in to digital form and do some operations in order to get an improved image or to extract some useful data from it. Various effective and efficient techniques are employed for automatic segmentation ofanimage for better performance of segmented image in the early detection and diagnosis of tumour part. This brain tumour dataset contains pre-processed contrast-enhanced MRI images with five kinds of brain tumour. Acoustic Neuroma, Chordoma, CNS Lymphoma, Craniopharyngioma, Pituitary Tumour. The superiorities of the proposed methods can be observed in terms of both visualperceptionandobjective metrics. The MRImagesHybrid Genetic Algorithm with Otsu proposed in this thesis willbe a better contributiontothemedicalimage segmentation forimprovingthe qualityof medical imagesand forpre-clinical and clinical activities.