Review of Brain Tumor Segmentation and Classification

Brain tumor segmentation is important task in medical image processing. Manual segmentation of brain tumor for cancer diagnosis, from large amount of MR images generated in clinical routine, is a difficult and time consuming task. There is a need for automatic brain tumor image segmentation. This paper does the review of different literatures of brain tumor segmentation. For segmentation, mostly used clustering algorithm like fuzzy c-means, k-means some researchers used convolution neural network approach, Particle Swarm Optimization (PSO). The purpose of every segmentation algorithm is to achieve accurate and efficient system developed, so that to find tumor in minimum time with maximum accuracy. Mostly support vector machine (SVM) classification technique is used for finding the accuracy of the developed system some different technique like K-NN, NSC, SRC and k-means are also used and different researchers compared these techniques for better accuracy.