A point-process model of human heartbeat intervals: new definitions of heart rate and heart rate variability

Area of Science:

• Physiology
• Biostatistics
• Cardiovascular Regulation

Background:

• Heart rate variability (HRV) quantifies autonomic nervous system regulation of the cardiovascular system.
• Current HRV algorithms lack consideration for the point-process nature of heartbeats and do not provide instantaneous estimates.
• Existing methods fail to capture the inherent stochastic structure of heartbeat intervals.

Purpose of the Study:

• To develop a novel stochastic model for human heartbeat intervals.
• To derive new, instantaneous definitions of heart rate and heart rate variability.
• To assess the accuracy of the proposed model in describing heart rate dynamics.

Main Methods:

• Modeling heartbeat intervals using a history-dependent inverse Gaussian process.
• Deriving explicit probability densities for instantaneous R-R interval and heart rate standard deviations.
• Estimating time-varying model parameters via local maximum likelihood.
• Assessing model goodness-of-fit using Kolmogorov-Smirnov tests and the time-rescaling theorem.

Main Results:

• A new probability density for heartbeat intervals was derived, enabling instantaneous heart rate and heart rate variability definitions.
• The inverse Gaussian model accurately described human heartbeat intervals from 10 healthy subjects.
• The model provided a highly accurate description of heart rate dynamics at rest and under physiological stress during a tilt-table experiment.

Conclusions:

• The proposed elementary, physiologically based stochastic model offers a powerful new framework for analyzing heart rate and heart rate variability.
• This approach provides accurate, instantaneous measures of cardiovascular regulation, overcoming limitations of existing methods.
• The model effectively captures the stochastic nature of heartbeats, even in extreme physiological states.