Automated Screening of Parkinson's Disease Using Deep Learning Based Electroencephalography
Mohamed Shaban
Abstract
Parkinson's Disease (PD) is the second most common neurodegenerative disease after Alzheimer disease in the United States. It is characterized by motor symptoms such as slowness of movement, and tremor, and non-motor symptoms such as cognitive changes, anxiety, depression, and sleep problems. Diagnosis, and evaluation of the disease are based on clinical examination, making the diagnosis challenging, and subjective. In this paper a deep learning based framework was introduced that utilizes Artificial Neural Networks applied on three spatial channels of a resting state Electroencephalography (EEG) dataset. Using the proposed framework, it was feasible to successfully screen, and classify subjects into controls, and PD with an accuracy of 98%, a sensitivity of 97%, and specificity of 100%. The proposed framework is considered as a precise, and reliable computer aided diagnostic tool that supports clinicians' diagnosis, and therapeutic treatment recommendations.