Machine Learning based Diagnostic System for Early Detection of Parkinson’s Disease

Early detection is an approach which enables initiation of diagnosis interventions. The objective of the study is to give an insight to diagnostic parameters of Biosignals which is used for finding brain and muscle defects of a Parkinson’s Disease(PD) patient. As PD occurs due to very less productionof dopamine in substantia nigra of brain leading to disturbed limb movements by affecting the muscles, Electroencephalogram and Electromyogram based GUI model would be an effective tool and true justification for early detection of PD. EEG and EMG was collected from early stages PD and healthy patients using Biophysical recording device. Features of EEG and EMG were extracted and classified using Artificial Neural Network. Designed model is a novel approach to classify PD from non-PD and monitor disease progression. There are various models available, but the work presented here gives the biosignals interpretations together with other parameters like those of radiological tools for disease diagnosis.