Acoustic characterization of mechanical valve condition and loading

Area of Science:

• Biomedical Engineering
• Mechanical Engineering
• Cardiovascular Research

Background:

• Björk-Shiley Convexo-Concave (BSCC) valve outlet strut fractures pose a significant risk.
• Fracture risk is influenced by pre-existing fractures and loading conditions.
• Non-invasive methods are needed for early detection.

Purpose of the Study:

• To evaluate the efficacy of passive acoustic measurement for assessing BSCC valve mechanical condition.
• To identify features in valve closing sounds indicative of outlet strut integrity.
• To develop a practical screening method for BSCC valve fractures.

Main Methods:

• Utilized non-invasive passive acoustic measurements of BSCC valve closing sounds.
• Employed cross-correlation to identify 13 characteristic waveform groups.
• Applied Mahalanobis distance and Volterra expansion for feature optimization and condition characterization.
• Validated against explanted valve data using blinded and leave-one-out analysis.

Main Results:

• Acoustic analysis identified features characterizing mechanical condition and predicting strut forces.
• The method demonstrated 100% sensitivity and specificity for strut fracture detection in 33 patients.
• Results were validated using blinded and leave-one-out cross-validation techniques.

Conclusions:

• Passive acoustic monitoring is a highly sensitive and specific non-invasive method for detecting BSCC valve outlet strut fractures.
• The technique shows potential as a practical and effective screening tool for valve integrity.
• Further analysis of blinded data is recommended to confirm robustness.