An Improved Brain Tumor Segmentation Method from MRI Brain Images

Abstract

The objective of the paper is to segment the tumor regions from Magnetics Resonance Imaging (MRI image) brain images. The method incorporates the following steps during the tumor localization process namely, smoothing, skull stripping, filtering, image enhancement, followed by defining the region of interest and segmenting the identified tumor region from the input MRI brain image. Experimental results concluded that the proposed method has better performance regarding segmentation accuracy and execution time karma when tested on 15 live brain images with significant tumor regions.