Paper Title
A Sharrow CNN Deep Learning Model to Detect Brain Tumors from Magnetic Resonance Image

Abstract
Brain tumor is among the deadliest disease worldwide. Once the patient is diagnosed with the disease, The tumor damage brain’ health tissues otherwise its intensified intra cranial pressure(ICP).Most death occurs from above causes or related adverse events. The earlier the tumor is diagnosed the luckier curing and survival probability increases. In this article, we presented a coherent techniques for detecting brain tumors in magnetic resonance images. Input gray-scale image are loaded inside our proposed method. A zero pading layer with a pool size of (2,2) is being considered and a double-convolutional layer with 32 neurons and a filter size of (9,9) with one unit stridal is applied, with two batch normalization layer. The normalizing pixel values speed up the computational process with a RelU activation function. Two max pooling layer with a filter size and stridal value four (4)has being used. Finally we applied a flatten layer to flatten our 3-dimensional matrix into a one-dimensional vector, this is then propelled into a Dense block (output Unit), which is a fully connected layer with one neuron using a sigmoid activation function . The classification task take place, where brain with tumor cells are being detected and classified as tumorous MR brain images and those without tumor cells are Non-tumorous brain MR images. Our method exhibits a higher Testing accuracy compared with the existing model Referring Tab.4 Keywords - Brain Tumor MRI(Magnetic Resonance Images), Convolution Neural Network, Tumor Classification Detection