Beyond HRV: attractor reconstruction using the entire cardiovascular waveform data for novel feature extraction

Area of Science:

• Cardiovascular Physiology
• Biomedical Engineering
• Data Analysis

Background:

• Modern monitoring yields high-frequency blood pressure waveforms, but raw data is often underutilized.
• Traditional heart rate variability (HRV) analysis focuses on beat-to-beat intervals, discarding most waveform data.

Purpose of the Study:

• To develop a novel method for detecting subtle changes in blood pressure waveform shape within long data streams.
• To analyze the full raw waveform data, going beyond the limitations of existing HRV methods.

Main Methods:

• Extracting features from complex data by reconstructing a 3D phase space attractor using delay coordinates from waveform windows.
• Projecting the attractor onto a plane to remove baseline variation and derive new quantitative measures.
• Iterating the time window across data to capture various waveform shape aspects.

Main Results:

• Enabled visualization and quantification of blood pressure waveform shape changes.
• Successfully applied to both animal (mice) and human blood pressure data.
• Demonstrated detection of waveform changes undetectable by conventional HRV analysis.

Conclusions:

• A new method analyzes complete blood pressure waveform data, offering insights beyond traditional HRV.
• This approach enhances the understanding of cardiovascular dynamics by capturing waveform morphology.
• The method provides a more comprehensive analysis of blood pressure signals.