Analyzing the Performance of MRI-Based Brain Tumor Detection and Segmentation with Deep Convolution Neural Networks

Main Article Content

S. Santhana Prabha , D.Shanthi

Abstract

A brain tumour is an uncontrolled development of irregular cells formed in various parts of the brain. The image processing based approach is one of the promising solutions for accurate identification of tumour’s in a brain like Magnetic Resonance Imaging (MRI) and Computer Tomography (CT).  Manual segmentation of tumour’s lead to errors in detection and is more time-consuming. Recently, many segmentation approaches have been developed for brain tumour segmentation and classification, among them, deep learning (DL) methods have a good impact and outperform other machine learning algorithms. In this work, a complete summary of MRI-based brain tumour segmentation methods is explained. Primarily, basic steps of image processing steps are explained. Then, segmentation methods proposed by various researchers using deep learning algorithms are summarized. Finally, performance parameters used for segmentation is presented.

Article Details

Section
Articles