Model Validation with Global Sensitivity Analysis for Polymeric Scaffolds in Restorative Heart Valves
Please login to view abstract download link
In situ tissue engineering is an innovative technique for heart valve replacement. This method, also known as endogenous tissue restoration (ETR), involves implanting a bioabsorbable scaffold, typically obtained from polymer electrospinning, into the body. The polymeric scaffold has to maintain structural properties until it is gradually replaced by native tissue. In silico simulations are a powerful tool to understand the complex underlying mechanisms and optimize scaffold design. Recently, a constitutive model of the short-term nonlinear response of the polymeric scaffold was proposed [1]. The current work conducts a comprehensive global sensitivity analysis and uncertainty quantification to enhance the efficiency of the previously developed model. We identify influencing parameters that enable more targeted and cost-effective model optimization. The large correlated input domain is sampled to construct a surrogate model, namely a polynomial chaos expansion, which enables the computation of sensitivity indices for input variables. Using the surrogate model, ANCOVA (ANalysis Of COVAriance) indices determine influential input parameters and thus simplify the optimization process[2]. The identification of unbiased parameters is implemented in a thorough verification, validation, and uncertainty quantification plan as developed in the Horizon 2020 Project of the European Union SimInSitu. This project aims to simulate the entire ETR process through fluid-structure interaction simulations, including growth and remodeling and patient-specific geometry.
