Study of support vector machine for classification of brain tumours

A Brain tumour is the growth of abnormal cells in the Brain. MRI Brain image is used to show the abnormal cells in the brain. This brain tumour affects normal brain activity. Normally brain tumour is the group or cluster of cells that are accumulated in any part of the brain. Based on the size or shape of clustering of cells it is divided into two types Benign and Malignant. The term Benign means normal in which the shape is equal or regular in size and whereas the term Malignant means abnormal, in which the shape is irregular in size. This research is focused on detecting and classifying the type of Benign or Malignant tumour from the brain image. Based on the size of the tumour it is further classified as stage I, II, and III in Benign or Malignant. Gray level co-occurrence matrix(GLCM) plays an important role in finding the types and stages of tumours. For feature extraction, Gray level co-occurrence matrix(GLCM) techniques are used. For classifying the tumours into benign and malignant Support Vector Machine(SVM) is experimented. By this proposed technique the results were comparatively better than the Discrete wavelet transform or continuous wavelet transform used for Feature Extraction.