Brain Tumour Segmentation and Stage Classification Using SVM Classifier and Notification Using GSM Module

Abstract

A tumour is any unwanted growth of tissues in any part of the body. About 50% of the people affected by primary brain tumour die each year. The foremost aim of this paper is to put forth a method that not only identifies the tumour affected area in brain, but also indicates its stage and sends notification by means of messages and mails to the concerned people (relatives, guardian, friends or even the patient itself, if needed). Here MRI (Medical Resonance Imaging) images are used to detect the tumour affected brain region by SVM (Support Vector Machine) algorithm. The software output has been simulated using MATLAB software. Furthermore GSM (Global System for Mobile communication) is used to send messages and mails to the concerned people. The final extracted image will be sent through mail. The medications to be taken by the patient will be sent through messages (SMS). The entire procedure includes, pre-processing (RGB to gray conversion and noise removal) of the input image, feature extraction, Localization of tumour, Morphological operations (dilation and erosion), segmentation and finally information sharing using UART and GSM module