Personalized mitral valve closure computation and uncertainty analysis from 3D echocardiography

Area of Science:

• Biomedical Engineering
• Computational Mechanics
• Cardiovascular Research

Background:

• Accurate intervention planning for Mitral Valve (MV) repair is critical for successful outcomes.
• Current finite-element models (FEM) face challenges in clinical translation due to unknown parameters and simplified geometry, particularly chordae topology.
• Existing FEM approaches often use a simplified 'parachute model' for chordae, lacking anatomical detail.

Purpose of the Study:

• To develop a comprehensive MV FEM with a novel chordae representation for improved anatomical accuracy.
• To create a fully automated personalization approach for chordae rest length, reducing manual effort.
• To validate the computation of MV geometric configuration at peak systole (closure) using the developed model.

Main Methods:

• Developed a novel comprehensive MV model with a new chordae representation approximating regional connectivity.
• Implemented a fully automated personalization approach for chordae rest length.
• Computed peak systolic MV configuration from a mid-diastolic model using FEM.
• Evaluated the method on ten in vitro ovine cases using echocardiography and high-resolution μCT imaging for validation.

Main Results:

• The new MV model successfully approximates regional chordae connectivity.
• Automated personalization of chordae rest length was achieved, eliminating manual parameter selection.
• The FEM model accurately computed MV closure configurations.
• Validation using μCT imaging confirmed the accuracy of the closure computation.

Conclusions:

• The developed comprehensive MV FEM with novel chordae representation and automated personalization enhances intervention planning.
• This approach improves the accuracy of predicting MV geometric configuration during closure.
• The validated method shows promise for advancing clinical applications of MV modeling in repair procedures.