Brain Tumor Classification using Convolutional neural Network

Electronics and Communication Engineering

Brain cancer is one of the largest medical problems faced today. Therefore, it is an area where building new tools to more effectively diagnose the disease and cure the disease can have a major impact on a large proportion of the population. Today we have a lot of data at our disposal about the patient about the tissue sample and our goal is to use this data effectively to provide the most accurate diagnosis for the patient. The conventional method involves the classification of brain tumors by inspecting the MRI images of the patients. However, a large amount of data available for different specific types of brain tumors makes this method very time consuming and prone to human errors. In the paper, we tried to design a model using a convolution neural network to identify the tumor and non-tumor MRI images. The architecture proposed in the paper consisted of one of each convolution layer, max-pooling layer, fully connected layer, and hidden layer. The model was trained on Brain tumor dataset (BRATS 2017) consisting of 320 images in the training set and 80 images in the test set. We were able to achieve the training accuracy of 89.25and validation accuracy of 97.5 using this simple architecture.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.