Classification of Brain Tumor Images using Deep Learning Methods

Abstract

Big data refer to all of the information and documents in the form of videos, photographs, text, created by gathering from different sources about a subject. Deep learning architectures are often used to reveal hidden information in the big data environment. Brain tumor is a fatal disease that negatively affects human life. Early diagnosis of the disease greatly increases the patient's chance of survival. For this reason, this study was conducted so that doctors could diagnose patients early. In this paper, deep learning architectures Alexnet, Googlenet, and Resnet50 architectures were used to detect brain tumor images. The highest accuracy rate was achieved in the Resnet50 architecture. The accuracy value of 85.71 percent obtained as a result of the experiments will be improved in our future studies. We will try to develop a new method based on convolutional neural networks in the near future. With this model, we will try to achieve higher accuracy than any known deep learning method.