Analyzing the effect of postural change on heart rate variability using multi-distance sample entropy (MDSE)

Abstract

One of the most studied physiological measures is heart-rate variability. In this research paper, a new nonlinear method called Multi Distance Sample Entropy (MDSE) is proposed to accurately detect the variation of Heart rate variability (HRV) for different body postures and examine its effect on the activity of the autonomic nervous system. The proposed method used a multi-distance function to calculate the distance between other matched vector pairs. The goal of employing considerable distances is to investigate the distance aspect for enhancing the accuracy of recognizing the proper trend of HRV for the postural shift. Thus, the ECG signals of 55 participants were recorded in the supine posture for 10 min and similarly, for 10 min in the standing posture using the BIOPAC®MP36 system. The noise signals have been simulated for varieddata lengths, and their error bars have been plottedusing the proposed and conventional sample entropy (SE) methods to demonstrate the superiority of the proposed MDSE method over the SE. Afterward, the proposed method (MDSE) and other existing methods were applied to the recorded dataset. From the obtained results, it is clear that the proposed method can accurately detect the variation of HRV in supine and standing postures than other applied methods. Using the MDSE method, HRV trends are 6.91% more accurate than Conventional SE and 10.85% more accurate than approximation entropy (AE) for the recorded dataset. The Spearman rank correlation between proposed and applied nonlinear methods has also been evaluated.