Instantaneous parameter estimation in cardiovascular time series by harmonic and time-frequency analysis

Area of Science:

• Cardiovascular physiology
• Signal processing
• Biomedical engineering

Background:

• Time-frequency distributions like SPWVD and CDM offer insights into time-varying spectral parameters.
• Individual methods have limitations, but their complementary features suggest joint utilization could be beneficial.
• Accurate analysis of cardiovascular variability is crucial for understanding physiological states.

Purpose of the Study:

• To validate the joint SPWVD-CDM method for analyzing cardiovascular time series.
• To assess the method's performance on synthetic and real data with varying cardiovascular variability.
• To introduce indexes for validating instantaneous parameter estimation.

Main Methods:

• Joint application of smoothed pseudo Wigner-Ville distribution (SPWVD) and complex demodulation (CDM).
• Analysis of synthetic and real cardiovascular time series, including those with normal and reduced variability (e.g., autonomic blockade).
• Development and application of two indexes to assess signal noise and power spectrum dispersion for parameter estimation validation.

Main Results:

• The joint SPWVD-CDM method demonstrates reliability in estimating instantaneous parameters in the high-frequency band, even with reduced cardiovascular variability.
• Interpretation of instantaneous frequency and phase in the low-frequency band is often unreliable in real-world cardiovascular data.
• Proposed indexes aid in validating the accuracy of instantaneous parameter estimations.

Conclusions:

• The joint SPWVD-CDM method is a valuable tool for analyzing cardiovascular time series, particularly in the high-frequency band.
• Caution is advised when interpreting low-frequency band parameters in cardiovascular signals.
• The method shows promise for applications in conditions affecting cardiovascular variability.