Multiple window correlation analysis of HRV power and respiratory frequency

Area of Science:

• Physiology
• Biomedical Engineering
• Signal Processing

Background:

• Heart rate variability (HRV) analysis is crucial for understanding autonomic nervous system function.
• Accurate estimation of respiratory influences on heart rate is vital, particularly in dynamic physiological states.
• Traditional HRV analysis often uses broad frequency bands, potentially limiting sensitivity to specific physiological drivers.

Purpose of the Study:

• To evaluate the correlation between respiratory frequency and high-frequency (HF) HRV power using a restricted frequency band.
• To determine if a narrow-banded HF HRV measure improves the estimation of respiratory-related heart rate variations compared to standard methods.
• To assess the robustness of correlation estimates under rapidly changing heart rate conditions.

Main Methods:

• Employed multiple window spectrum analysis to estimate respiratory center frequency.
• Defined a narrow HF HRV band as respiratory frequency +/-0.05 Hz.
• Compared mean square error of correlation estimates between a narrow HF band and the conventional 0.12-0.4 Hz band.
• Utilized peak-matched multiple windows and Welch methods for spectral estimation.

Main Results:

• A more restricted frequency band for HF HRV power estimation yielded a stronger correlation with respiratory frequency.
• The narrow-banded approach demonstrated a more robust estimation of HRV power related to respiration.
• This improved correlation suggests enhanced sensitivity to respiratory-induced heart rate fluctuations.

Conclusions:

• A narrow-banded approach to HF HRV power estimation significantly enhances the correlation with respiratory frequency.
• This method provides a more robust and sensitive measure of respiratory influences on heart rate, especially during dynamic changes.
• The findings support the use of tailored frequency bands in HRV analysis for improved physiological insights.