Pixel Classification based Brain MR Image Segmentation

Brain image segmentation is challenging task for proper clinical diagnosis. Automatic segmentation of the brain into four classes namely background, cerebro spinal fluid, grey and white matter is presented in this work. Accurate segmentation of the tumor in the brain is also achieved using the proposed method. Classification of the pixels in different classes is achieved by comparing their inter class distances. The proposed method ensures average Jaccard index and Dice coefficient as 0.8173 and 0.8952 respectively.