Characterization of dynamic interactions between cardiovascular signals by time-frequency coherence

Area of Science:

• Cardiovascular Physiology
• Biomedical Signal Processing
• Time-Frequency Analysis

Background:

• Understanding cardiovascular control relies on analyzing dynamic interactions between physiological signals.
• Time-frequency (TF) coherence is a valuable tool for characterizing these complex relationships.

Purpose of the Study:

• To describe and assess two novel methodologies for TF coherence analysis between cardiovascular signals.
• To evaluate the accuracy of these methods in localizing coupled TF regions.
• To investigate the impact of orthostatic stress on cardiovascular control using TF coherence.

Main Methods:

• Development and application of TF coherence analysis using smoothed pseudo-Wigner-Ville distribution (SPWVD) and multitaper spectrogram (MTSP).
• Automatic assessment of coherence estimate significance levels.
• Validation using computer-generated data and data from 14 healthy volunteers undergoing a tilt table test.

Main Results:

• The SPWVD method demonstrated higher accuracy (>96.9%) than MTSP (>84.4%) in localizing TF regions with significant coupling, especially at signal-to-noise ratios (SNR) ≥ 5 dB.
• Orthostatic stress significantly increased coupling between R-R variability (RRV) and systolic arterial pressure variability.
• No significant changes in RRV and respiration coupling were observed, while coupling between RRV and pulse interval variability in the high-frequency (HF) band decreased.

Conclusions:

• The SPWVD method provides a more accurate approach for TF coherence analysis of cardiovascular signals.
• TF coherence analysis effectively reveals dynamic changes in cardiovascular control during orthostatic stress.
• These findings enhance the understanding of autonomic nervous system regulation of cardiovascular function.