Ensemble Technique based Parkison’s Disease Detection from FOG and EEG Signals

The Human Brain consists of neurons (nerve cells) and the neurotransmitter dopamine that solely orchestrates critical body functions. However, individuals suffering from Parkinson’s Disease (PD) are often plagued with unrestrained motor dysfunctions like bradykinesia, tremors, and rigidity leaving them with postural and cognitive impairments. In this research proposal, we have aimed to predict PD by utilizing the Electroencephalogram (EEG) signals, with a primary focus on PD symptoms like Freezing of Gait (FOG) episodes as a potential PD indicator. The research dataset comprises EEG recordings with a total of 53 PD patients and 41 Healthy Controls (HC) from Beijing University, the University of New Mexico (UNM), and the University of Iowa (UI). Different Ensemble techniques were employed for EEG signals analysis, and the outcomes were appraised based on their performance matrices and time complexity. Our findings demonstrated that ensemble method approaches yielded impressive results with some best accuracy, precision, and recall scores of 88.47%, 89.21%, and 87.42% respectively.