Predicting Aphasia Type and Severity Using Machine Learning

Aphasia is a language disorder that can result from brain damage, often caused by a stroke or traumatic brain injury. The type and severity of aphasia can vary widely among individuals making accurate diagnosis and treatment challenging in real life conditions. In this paper, we propose an approach to determine the type and severity of aphasia using machine and deep learning on derivatives of brain MRI (Magnetic Resonance Imaging) scans. Our study involves training multiple machine and deep learning models to accurately classify and quantify the type and severity of aphasia and generate synthetic data in order to augment the existing dataset to combat data imbalance. We achieved a ROC-AUC score of 66% for aphasia type classification and 63% for severity prediction. With the use of data augmentation, we managed to achieve a ROC-AUC equal of 71% for type classification. The results of this research have the potential to improve the diagnosis and management of aphasia, leading to better outcomes for individuals affected by this debilitating condition.