Brain Tumor Prediction from Magnetic Resonance Images

Abstract

Globally, brain tumours are a leading cause of death. Depending on whether they are malignant or benign, tumours can interfere with neurological and body functions as well as cause severe health problems. Carrying out the diagnosis of brain tumours through the conventional methods of test, such as magnetic resonance imaging (MRI) scan and clinical examination, would be cumbersome and time consuming in most situations. With the latest machine learning hype, deep learning emerged as the medical image analysis game-changer. Early, precise identification of brain tumours has been made possible due to the remarkable performance of convolutional neural networks (CNNs) in extracting fine details from medical images. Support vector machines (SVMs) and CNNs can be combined to improve classification accuracy. The goal of this study is to achieve an improved model for classifying brain tumours using MRI images. This model extracts complex features using a CNN whilst improving classification with an SVM. The resulting model can enhance and speed up medical diagnosis. With a 99.2% training set success rate and a 96.1% test set success rate, the hybrid method is extremely accurate.