MODELING OF TRANSCATHETER AORTIC VALVE INTERVENTION

Transcatheter aortic valve implantation (TAVI) has revolutionized the management of aortic valve diseases, providing a minimally invasive option for patients who are ineligible for open-heart surgery. Since its inception, TAVI has undergone significant advancements, extending its use to patients with intermediate and low surgical risk. However, the procedure is not devoid of risks, as complications such as paravalvular leakage, conduction disturbances, coronary obstruction, and structural valve deterioration can occur, potentially leading to fatal consequences.To address these challenges, this thesis focuses on the development of high-fidelity, patient- specific computational models capable of predicting the structural and hemodynamic performance of TAVI and its associated risks across various clinical scenarios. The modeling workflow was designed in compliance with the Verification, Validation, and Uncertainty Quantification (VVUQ) framework outlined by the ASME V&V 40-2018 standard, ensuring that the developed numerical model is sufficiently credible for clinical applications. The computational platform integrates finite element analysis (FEA) for structural simulation and fluid-structure interaction (FSI) for hemodynamic assessment, simulating the TAVI procedure and its biomechanical interaction with patient-specific anatomy. Advanced material calibration techniques were employed to accurately replicate the mechanical properties of native tissues of the investigated patient population, and a parametric modeling approach based on anatomical landmarks was used to reconstruct native valve leaflets. Validation of the patient-specific TAVI models against clinical data revealed a high correlation between model predictions and clinical outcomes, confirming the accuracy of the model in predicting key parameters such as valve orifice area, pressure gradients, and stent deformation.In addition to in-silico modeling, experimental methods such as 3D printing and Particle Image Velocimetry (PIV) were employed to investigate the role of transcatheter aortic valves (TAVs) in complex clinical scenarios, including transcatheter mitral valve replacement (TMVR) and valve-in-valve procedures. These investigations yielded critical insights into procedural success. For instance, TMVR models highlighted the importance of dynamic variations in the estimated neo-left ventricular outflow tract (neo-LVOT) area throughout the cardiac cycle, and how these variations are influenced by patient anatomy and procedural factors such as implantation depth, annular stiffness, and calcification degree. By accounting for these variables, these models can help clinicians in optimizing device selection and procedural planning, improving patient-specific outcomes.Overall, the computational platform developed in this thesis, in conjunction with experimental techniques, offers a robust tool for enhancing the safety and efficacy of TAVI procedures and supporting the development of next-generation TAVI devices. This work advances the field of personalized medicine by offering a reliable in-silico tool that can assist traditional clinical assessments and accelerates the development and optimization of new heart valve technologies.