Classification of Malignant Brain Tumor using Deep Transfer Learning-based Framework on Brain MRI Datasets

One of the leading causes of death in the world is Brain Tumors. In recent years, there has been an increase in cases of brain tumors, and it has become a serious problem in medical diagnosis. Deep Transfer Learning techniques are used in various medical diagnosis applications as they provide accurate and efficient results and require no human interventions. Brain MRI images are used for the effective diagnosis of Brain tumors. With these MRI images, effective datasets are prepared and Deep Transfer Learning models are applied to these datasets. This paper presents a transfer learning framework for predicting the brain tumor. In general, previous approaches were used for the detection of Benign tumors. In the proposed framework, a strategy was proposed to efficiently detect Malignant and Benign brain tumors. As the image datasets are involved in the proposed framework, a pre-trained Convolutional Neural Network will be used here for the feature extraction and pattern classification from the brain MRI image datasets.