Machine Learning Based Approaches in the Detection of Parkinson’s Disease – A Comparative Study

Parkinson’s disease (PD) is one of the enfeeble diseases that is incurable. It occurs when dopamine is not properly produced in substantia nigra of our central nervous system. Dopamine is responsible for our movement and activities like speaking, walking and writing. In the recent papers, researchers have found that speech, Freezing of Gait (FOG), writing problems are the most common symptoms for PD. However, most of the people who suffer from Parkinson have speech and FoG disorder. With the increase of severity, some other symptoms also arise gradually. Some works have been done by the researchers in early prediction and detection of PD using Machine Learning (ML) and Deep Learning (DL) classifiers. In this paper a detailed survey of the different works in this area has been carried out. We identified certain drawbacks of the existing works. Based on this, we tried to explore the future scope of works which can enhance the performance of the existing works in early detection of PD.