|Title:||Deep Transfer Learning for Brain Magnetic Resonance Image Multi-class Classification|
|Keywords:||Brain Tumor; CAD; Convolutional Neural Network (CNN); Computer Vision; Deep Transfer Learning; Magnetic Resonance Imaging (MRI)|
Magnetic Resonance Imaging (MRI) is a principal diagnostic approach used in radiology to create images of a patient’s anatomical and physiological structures. MRI is the prevalent medical imaging practice to find abnormalities in soft tissues. Traditionally they are analyzed by a radiologist to detect abnormalities in soft tissues, especially the brain. However, the process of interpreting a massive volume of a patient's MRI is laborious. Hence, Machine Learning methodologies can aid in detecting abnormalities in soft tissues with considerable accuracy. This research has curated a novel dataset and developed a framework that uses Deep Transfer Learning to perform a multi-classification of tumors in the brain MRI images. This paper adapted the Deep Residual Convolutional Neural Network (ResNet-50) architecture for the experiments and discriminative learning techniques to train the model. Using the novel dataset and two publicly available MRI brain datasets, this proposed approach attained a classification accuracy of 86.40% on the curated dataset, 93.80% on the Harvard Whole Brain Atlas 97.05% accuracy on the School of Biomedical Engineering dataset. Our experimental results demonstrate the proposed framework for transfer learning is a potential and effective method for brain tumor multi-classification tasks.