ST1372: Brain Tumor Detection Using CNN

Syabil Basyiruddin Bin Syamsul Bahrain MILA University

According to Global Cancer Registry (2020) 18.1 million case were suffering from cancer, with around 308,000 were brain related cancers, and a five-year mortality rate of 65% has been the highest recorded percentage. It is considered one of the most challenging problems in modern medicine where people are usually characterized to have low life expectancy resulting in a very low recovery rate. Even today, this kind of detection is based on manual classification of Magnetic Resonance Imaging (MRI) resulting in inaccurate conclusions driven especially by inexperienced doctors. We propose an automation tool using Convolutional Neural Network (CNN) based brain tumor detection through Deep Learning (DL). The proposed model processes the dataset containing real MRI with high testing accuracy and no losses, resulting in fast computation.