Treatment of Resampling Frequency and Epoch Length for RR interval in Autonomic Nervous System Analysis

This investigation aimed to clarify the effects of the frequency of resampling of the RR interval and length of the analysis interval to simplify the analysis of heart rate variability (HRV) for the evaluation of the autonomic nerve system (ANS). The subjects were 15 male university students, and the experiment consisted of 10 min of rest and 10 min of a pleasant-emotion-induction task, for a total of 20 min. The heart rate and pulse rate were detected from the measured electrocardiogram and pulse waves. RR intervals were used to analyze sympathetic and parasympathetic nerve activity. The resampling frequency was set to 2, 4, and 10 Hz. The analysis times were 30, 60, 120, 300, and 600 s for the 10-min resting period and 30, 60, and 300 s for the changes associated with the induction of pleasant emotions. The total power (TP), high-frequency (HF) power, and low-frequency (LF) power exhibited an interaction between the resampling frequency and analysis time. In the pleasant-emotion-induction task, significant differences were observed among HF/TP, LF/TP, and LF/HF at 2, 4, and 10 Hz. However, when the analysis time was set to 30 s, the number of measurement points where statistically significant differences were obtained was limited. The present investigation demonstrates that the sensitivity of the relative ANS index may decrease with decreasing analysis time. The results of this investigation suggest that while a shorter measurement time is consistent with adequate rest, the sensitivity of the ANS index to capture minute changes in the body needs to be considered.