MRI brain cancer classification using hybrid classifier (SVM-KNN)

Authors:
Publisher website
Google Scholar
Abstract
This paper proposes an intellectual classification system to recognize normal and abnormal MRI brain images. Nowadays, decision and treatment of brain tumors is based on symptoms and radiological appearance. Magnetic resonance imaging (MRI) is a most important controlled tool for the anatomical judgment of tumors in brain. In the present investigation, various techniques were used for the classification of brain cancer. Under these techniques, image preprocessing, image feature extraction and subsequent classification of brain cancer is successfully performed. When different machine learning techniques: Support Vector Machine (SVM), K- Nearest Neighbor (KNN) and Hybrid Classifier (SVM-KNN) is used to classify 50 images, it is observed from the results that the Hybrid classifier SVM-KNN demonstrated the highest classification accuracy rate of 98% among others. The main goal of this paper is to give an excellent outcome of MRI brain cancer classification rate using SVM-KNN.