High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Brain Tumor Segmentation and Classification using Neural Network and SVM for MRI Images

Author(s):

Navdeep Kaur , Department of Computer Engineering Yadavindra College of Engineering Punjabi University Guru Kashi Campus Talwandi Sabo, Bathinda; A. P Gianetan Sekhon, Department of Computer Engineering Yadavindra College of Engineering Punjabi University Guru Kashi Campus Talwandi Sabo, Bathinda

Keywords:

MRI, Pre-Processing, Segmentation, Classification, Feature Extraction

Abstract

Medical imaging techniques gains its importance with the increase in the need of automated & efficient diagnosis of brain tumor. MRI brain tumor segmentation is difficult procedure to segment tumor area from MRI images because of inconsistency of anomalous tissues. As there are number of techniques are already presented for effectively segmentation of brain tumor but still it is not achieved the required level of accuracy, abnormalities classification is not predictable. The whole process incorporates the steps- preprocessing, segmentation with Lab color space & then feature extraction using GLCM matrix like energy, entropy, contrast etc. These features are trained and tested using SVM & Neural Network to classify the images into normal and abnormal & then abnormal into different tumor grades.

Other Details

Paper ID: IJSRDV4I90311
Published in: Volume : 4, Issue : 9
Publication Date: 01/12/2016
Page(s): 963-965

Article Preview

Download Article