Wavelet Packet Analysis of ECG signals to Understand the Effect of a Motivating Song on Heart of Indian Male Volunteers
This chapter investigates the effect of a motivational song (acting as a stimulus) on the electrical activity of the heart using wavelet packet analysis of electrocardiogram (ECG) signals. ECG signals were acquired from 18 healthy male volunteers during the pre- and the post-stimulus conditions. Wavelet packet decomposition of the ECG signals was performed up to level 3 using db04 wavelet, which resulted in the formation of 8 wavelet packet coefficients. Linear (t-test) and nonlinear (classification and regression tree [CART], boosted tree [BT], and random forest [RF]) methods were used to identify the statistically significant parameters. The statistically significant parameters were used as categorical inputs for multilayer perceptron (MLP)-based artificial neural network (ANN) classification of the ECG signals. A classification efficiency of ≥ 80% was obtained, suggesting an alteration in the cardiac electrophysiology of the volunteers caused by the music stimulus.