Brain Tumor Detection Using Magnetic Resonance Images Through Convolutional Neural Networks

Abstract

Nowadays, brain tumor classification is a crucial task for neurologists and radiologists. However, manually detecting brain tumors from magnetic resonance imaging (MRI) can be challenging and prone to errors. This study proposes a method using neural networks to detect brain tumors. This study uses a subset of the BRATS 2018 dataset that contains 1,992 brain MRI scans. The proposed model achieves an accuracy of 97% in the test set making it a tool for medical experts.