Frequency domain analysis for the classification of parkinson's disease patients

Abstract

In this paper, Emotional recognition in Parkinson's disease (PD) has been analyzed in frequency domain using Entropy, Energy-Entropy and Teager Energy-Entropy features. Classification results were observed using three classifiers namely Probabilistic Neural Network, K-Nearest Neighbors Algorithm and Support Vector Machine. Emotional EEG stimuli such as happiness, sadness, fear, anger, surprise, and disgust were used to categorize the PD patients and healthy controls (HC). For each EEG signal, the alpha, beta and gamma band frequency features are obtained for three different feature extraction methods (Spectral Entropy, Spectral Energy-Entropy, and Spectral Teager Energy-Entropy). The proposed Spectral Energy-Entropy feature performs well for all six emotions for different classifiers when compared to other features, whereas different features with classifiers give variant results for few emotions with highest accuracy of 96.8%.