Real-Time Analysis of Heart Rate Variability Patterns during Hemodialysis

This paper provides an analysis method for heart rate variability (HRV) that uses the information theory as a tool to predict sudden decreases in blood pressure during dialysis. Sympathetic and parasympathetic activities are closely related to blood pressure. Moreover, these indexes are obtained by calculating the integral of low- and high-frequency in the power spectrum of HRV components. Any study has not reported that these indexes contributed to the prediction of Intradialytic hypotension in real-time. In this paper, we focused on HRV patterns instead of the autonomic nervous system activity index. Confirmation of HRV-specific pattern changes before and after rapid decreases in blood pressure leads to predictors of dialysis hypotension. We used the variation of HRV patterns as the variation of the probability distribution in HRV. Moreover, we used Jensen-Shannon divergence (JSD) to investigate the change in probability distribution. The purpose of the HRV analysis method using JSD is to detect signs of sudden decreases in blood pressure in real-time during hemodialysis that cannot be obtained using the autonomic nervous system activity index.