A New Approach to Detection of Parkinson’s Disease Using Variational Mode Decomposition Method and Deep Neural Networks

In this study, a new approach is proposed to detection of Parkinson’s disease by using Electroencephalography (EEG) signals and three different subband signals generated using Variational Mode Decomposition (VMD) method. In the proposed method, EEG signals and subband signals are applied separately to the generated 1D CNN model and the classification results are compared. The classification results showed that the VMD-sub band signals obtained from EEG signals were successful in diagnosing Parkinson’s. The highest classifier accuracy was obtained from second VMD subband data by 98.10%.