Abstract

In this paper, two different Convolutional Neural Network (CNN) models are presented for identifying the different classes of brain tumors in MR image modality. The first model consists of sixteen layers that give validation accuracy of 85% and testing accuracy of 79.7%. The second model consists of twenty-five layers having validation accuracy of 87.7% and testing accuracy of 84.3% is obtained. Two different optimizers Sgdm (Stochastic Gradient Descent with Momentum) and Adam are used for training the network.