Parameterized shape optimization of a bi-leaflet heart valved conduit for pediatric applications.

Abstract

Congenital heart defects could affect the right ventricular outflow tract in pediatric patients. As a result, pediatric pulmonary valve replacements are often needed to effectively mediate unidirectional blood flow from the right ventricle to the pulmonary artery. The present work aims to optimize a parameterized shape of a bi-leaflet heart-valved conduit using multi-objective optimization algorithms. We developed an integrated framework that automatically facilitates the design, structural mechanics simulation, and design optimization of the prosthetic valve. Bezier curves are employed to represent the geometric profile of both the free edge and the attachment edge of the leaflets, while a genetic algorithm updates the parameterized design variables during optimization. A quasi-static finite element analysis (FEA) model simulates valve opening and closure mechanics under a prescribed hemodynamic pressure profile. The objective is to find the optimal leaflet shape, enhancing durability and functionality by minimizing the leaflets' maximum principal stress and the orifice area at valve closure. An optimized design is selected from the final Pareto fronts for prototyping. Numerical results demonstrate an improved performance of the optimized valve over the initial design, indicating the complex impact of the valve geometry on valve performance metrics and underscoring the imperative for design optimization. The performance of optimal valve design is further analyzed using fluid-structure interaction (FSI) modeling to evaluate its performance under dynamic loading conditions.