Recognition of Brain Tumor Using Fully Convolutional Neural Network-Based Classifier

Abstract

The brain is a most essential body part in the human body. The brain neuron captures all moment of the memory which is happening in our lives. The brain tumor is a complex disease, and the tumor occurred in brain is called as a primary tumor. The two categories of brain tumor are benign and malignant tumor. Although many researchers and scientist have been implemented an algorithm to diagnose the tumor, still to diagnose the tumor in early stages is a challenging task. The limitation of existing system is time taken to recognize the tumor and the error rate is high. This research article develops a new methodology and system to identify the tumor in early stage. The classification technique of the proposed system is to diagnose the brain tumor with low error rates. To examine the recognition of brain tumor with less computational time, convolution neural network approach was used to detect the brain tumor and categorize the tumor using softmax activation function. Our results led to conclude that the proposed system is able to perform well to identify the brain tumor in less computational time with fine accuracy.