A Soft Computing Framework for Brain Tumor Detection through MRI Images

Abstract

Brain Tumor is one of the deadly diseases that has taken the lives of many people. A tumor can be benign or malignant. Benign tumors are curable if detected at the early stage. In today’s modern era of medical technology, MRI (Magnetic Resonance Imaging) has proved to be an efficient method of detecting the presence of brain tumor in the patient. Proper detection of brain tumor is necessary for further treatment of the patient which is possible through accurate segmentation of the brain. Brain segmentation plays a vital role in brain tumor detection. Over the years many researchers have proposed different methods for brain tumor detection but use of soft computing tool is much more preferred as far as human error is concerned. Here, a method of classification of images with and without tumor is dictated using Artificial Neural Network (ANN). The ANN has been configured to detect the presence of tumor by using various parameters of Gray-Level Co-occurrence Matrix (GLCM).

Keywords:Brain tumor, MRI, pre-processing, soft computing, neural network