According to Global Cancer Registry, 18 million cases were suffering from brain tumors and a mortality rate of 73% 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 Magnetic Resonance Images (MRI) with nearly perfect testing accuracy and no losses, resulting in fast computation. This CNN model automatically trains itself based on the dataset we have provided for training and depicts the existence or absence of brain tumors by testing the given MRI images.
Software And Hardware