Enhancing coronary Wave Intensity Analysis robustness by high order central finite differences.

Area of Science:

• Cardiovascular Physiology
• Biomedical Engineering
• Medical Signal Processing

Background:

• Coronary Wave Intensity Analysis (cWIA) separates arterial hemodynamics from cardiac mechanics.
• cWIA indices predict disease and recovery but face standardization challenges.
• Signal processing, particularly noise removal via filters, impacts cWIA accuracy.

Purpose of the Study:

• To analyze the impact of filter parameter selection on cWIA output and clinical metrics.
• To introduce a novel, parameter-free approach for computing signal derivatives in cWIA.
• To enhance the robustness and reduce variability in cWIA measurements.

Main Methods:

• Sensitivity analysis of cWIA output to Savitzky-Golay filter parameters.
• Comparison of filter-based differentiation/smoothing versus a central finite difference scheme.
• Validation of the proposed method on human and animal in-vivo datasets.

Main Results:

• cWIA output and clinical metrics are significantly affected by filter parameters.
• The proposed central finite difference approach is parameter-free.
• Outcome variability in cWIA was reduced by 60% across diverse datasets.

Conclusions:

• Filter parameter selection critically influences cWIA results.
• The parameter-free central finite difference method enhances cWIA robustness.
• This novel approach improves the reliability of cWIA for clinical applications.