A point process adaptive filter for time-variant analysis of heart rate variability

Estimating time-variant heart variability indices from R-R interval beat series has been widely investigated by current research involving cardiovascular control. Most of the currently accepted approaches in time-variant heart rate analysis ignore the underlying discrete structure of human heart beats, and usually require minutes of data. We derive an adaptive point process Bayes´ filter based on a statistical model which considers the stochastic structure of heart beat intervals as a point process. From the explicit inverse Gaussian probability density describing heart rate and heart rate variability we are able to extract and recursively update, at any time resolution, a set of indices related to the first and second moments of this probability density. We apply our algorithm in an analysis of human heart beat intervals from a tilt-table experiment. Our results describe real instantaneous estimates of heart rate variability and may have important implications for research studies of cardiovascular and autonomic regulation. Our algorithm is easy to implement for on-line analysis of heart rate variability in the intensive care unit, operating room or labor and delivery suits.