Brain MRI Image Segmentation and Classification using PCA-SVM and CNN
Senthil Pandi S, S. Vishnu, Zubair Ali. L, V Sudharsan, T Kumaragurubaran, Sooraj Nikam. P
Abstract
Brain tumor is a malicious growth on surface of the brain or within the brain tissues, it is a life-threatening disease which can spread to other vital parts with slow pace, if not detected early it can cause various effects on your body. These abnormal growths can be identified from images on medical reports generated from Magnetic Resonance Imaging (MRI). Categorizing different kinds of brain tumors can be made easier by using machine learning and deep learning techniques. These techniques could prove advantageous in identifying the existence of a tumor during its early phases. In order to classify MRI images, this study uses Convolutional Neural Networks (CNN) and Support Vector Machines (SVM) in conjunction with Principal Component Analysis (PCA) to extract features from brain MRI images. Additionally, image segmentation methods are incorporated like image thresholding, contour detection, edge detection and histogram equalization for better identification of tumor.