An application of phonocardiography signals for psychological stress detection using non-linear entropy based features in empirical mode decomposition domain

The well-established association of psychological stress and pathogenesis emphasizes the need for early detection of psychological stress to prevent the progression of diseases and hence saving human lives. The purpose of this research paper is to present a new framework for using phonocardiography (PCG) signal to detect psychological stress based on non-linear entropy based features extracted using empirical mode decomposition (EMD). These PCG signals are used to extract time duration of cardiac cycles consisting of consecutive S1 peaks to form Inter-beat Interval (IBI) signal. The IBI signal is decomposed to sub-band signals using EMD to form Intrinsic Mode Functions (IMFs). Then non-linear features namely Permutation Entropy (PEn), Fuzzy Entropy (FzEn) and K-Nearest Neighbour (K-NN) entropy estimator are evaluated. Ranking methods namely - Entropy method, Bhattacharya space algorithm, Receiver Operating Characteristic (ROC) method and Wilcoxon method are then used in order to optimize the system. The extracted entropy features are fed to Least-Square Support Vector Machine (LS-SVM) for classification and highest accuracy, sensitivity and specificity obtained using the proposed system is 96.67%, 100% and 93.33% respectively. The proposed system opens a new research area of using PCG signal for psychological stress detection which can be easily used for home-care and is relatively inexpensive in comparison to other biophysical measures like Electroencephalography (EEG) and Electrocardiography (ECG). (C) 2019 Published by Elsevier B.V.