Establishing In-silico Credibility of Patient-Specific Finite-Element Model in a Virtual Cohort

Abstract

The feasibility of transcatheter aortic valve implantation (TAVI) has been investigated by in-silico modeling to assist clinicians in the pre-operative planning of the device implantation. However, a rigorous standardization is needed to improve the reliability and effectiveness of the in-silico model. This study aims to verify and validate a structural finite-element model of TAVI using the ASME V&V40 protocol. After defining the context of use (COU) and model risk, verification activities were carried out for the computational model of the human aortic root as obtained from a cohort of twenty patients with TAVI. Validation involved comparing numerical predictions with post-procedural CT imaging measurements of both the aortic strain and valve orifice area at systole. Results demonstrated that the relative error between simulations and actual CT measurements ranged from 8.6% to 13.6% for the systolic strain whilst the calcific aortic valve presented differences in the range of 0.08% to 11.86%. This study presents a flexible and customizable framework for assessing the credibility of patient-specific modeling in TAVI patients, emphasizing the critical role of establishing credibility in the response of in-silico models.