An Efficient Framework for Solving a Convex, State-Space Heartbeat Dynamics Model.

Area of Science:

• Cardiovascular Physiology
• Biostatistics
• Computational Biology

Background:

• Traditional heart rate variability analysis methods using windowing and averaging do not fully capture the probabilistic nature of interbeat intervals.
• Existing point process models for interbeat intervals often involve nonconvex optimization, necessitating careful initialization or approximations for reliable parameter estimation.

Purpose of the Study:

• To develop a statistically rigorous and computationally efficient method for analyzing interbeat intervals.
• To address the limitations of existing methods by employing a convex optimization framework for model fitting.
• To accurately estimate heart rate dynamics and their relationship to physiological states.

Main Methods:

• A state-space point process model incorporating a latent Gauss-Markov process and a gamma generalized linear model for interbeat intervals was developed.
• The maximum a posteriori (MAP) estimation problem was proven to be convex.
• An efficient and exact solution for MAP estimation was achieved using the alternating direction method of multipliers (ADMM).

Main Results:

• The proposed method demonstrated a convex estimation problem, allowing for efficient and exact solutions.
• Application to electrocardiogram (ECG) recordings from a tilt table study showed good model fit using Kolmogorov-Smirnov plots.
• Estimated mean heart rate accurately reflected dynamic postural changes observed during the tilt table study.

Conclusions:

• The developed state-space point process model offers a statistically sound and computationally efficient approach to heart rate variability analysis.
• This method overcomes the limitations of nonconvexity in previous point process models.
• The accurate reflection of dynamic physiological changes highlights the clinical relevance of this novel approach for cardiovascular monitoring.