Deep Learning Framework using CNN for Brain Tumor Classification

The brain tumor is one of the major concerns accounting to number of deaths caused by it worldwide. The available techniques to diagnose brain tumors requires human intervention, are time consuming and are subjective in nature. In the current scenario, automated methods for correct diagnosis of brain tumors is of utmost importance for timely detection and proper diagnosis. The existing frameworks are time consuming and intrusive, these limitations highlight the importance of using a fully automated deep learning system for brain tumour classification. The paper presents a deep learning framework for classification with minimal human intervention. Convolutional Neural Network is most preferred deep learning method for classification tasks owing to its flexible architecture and high accuracy. This paper aims to develop a deep learning based CNN framework for brain tumor classification. The proposed CNN framework has an edge over the other models by achieving classification accuracy of 99.4% and loss of 0.0030%. The Comparative analysis of the proposed framework with other state of art models shows a relative improvement in accuracy and test error.