Research Analysis on Current Advances in Parkinson’s Disease Detection Using Signal Processing and Machine Learning-Based Techniques

Parkinson's Disease (PD) is a neurodegenerative condition characterized by the degeneration of dopaminergic neurons in the brain, resulting in both motor and non-motor symptoms. The timely and precise diagnosis of PD is of paramount importance for effective management and treatment strategies. Recent years have witnessed the emergence of diverse diagnostic approaches for PD, encompassing clinical assessments, advanced imaging techniques, and biochemical analyses. This comprehensive technical review aims to present a comprehensive overview of the various methodologies employed for diagnosing Parkinson's disease. Numerous scholarly articles have demonstrated the utilization of Deep Learning Neural Networks for classification purposes, while a select few have developed novel methods and Machine Learning models by incorporating existing Deep Learning techniques, often complemented by various optimization approaches like the Moth Flame Optimizer (MFO) or feature selection techniques to enhance the accuracy and precision of PD prediction. Notably, each article included in this study has harnessed Electrophysiological signals as the primary features of Machine Learning algorithms, showcasing their prominence in the field of PD classification and prediction, particularly within the biomedical domain.