Nonlinear analysis of continuous ECG during sleep II. Dynamical measures

Area of Science:

• Cardiology
• Nonlinear Dynamics
• Sleep Science

Background:

• Cardiac rhythms are hypothesized to exhibit chaotic dynamics, indicating healthy physiological flexibility.
• Sleep significantly modulates the autonomic nervous system's control over cardiac dynamics.
• Understanding these modulations is crucial for interpreting physiological states via ECG.

Purpose of the Study:

• To investigate the utility of nonlinear system theory methods for analyzing continuous electrocardiogram (ECG) signals.
• To differentiate ECG signals recorded during various sleep stages using nonlinear measures.
• To assess the relationship between sleep stages and cardiac complexity.

Main Methods:

• Implementation of six nonlinear measures: correlation dimension (D2), Lyapunov exponent (L1), and Kolmogorov entropy (K2).
• Analysis of unstable periodic orbits from ECG signals.
• Comparison of nonlinear measures between different sleep stages (REM vs. slow wave sleep).

Main Results:

• Continuous ECG signals were successfully differentiated from linear stochastic surrogates using all implemented nonlinear measures.
• A significant increase in dominant chaoticity (L1) and a decrease in degrees of freedom (D2) were observed during REM sleep compared to slow wave sleep.
• Findings suggest nonlinear measures can distinguish sleep-related cardiac states.

Conclusions:

• Nonlinear analysis of continuous ECG signals provides valuable insights into cardiac dynamics during different sleep stages.
• Increased chaoticity during REM sleep correlates with increased heart rate variability.
• Reduced correlation dimension during REM sleep may indicate reduced respiratory influence on cardiac activity.