A fast in silico model for preoperative risk assessment of paravalvular leakage.

In silico simulations can be used to evaluate and optimize the safety, quality, efficacy and applicability of medical devices. Furthermore, in silico modeling is a powerful tool in therapy planning to optimally tailor treatment for each patient. For this purpose, a workflow to perform fast preoperative risk assessment of paravalvular leakage (PVL) after transcatheter aortic valve replacement (TAVR) is presented in this paper. To this end, a novel, efficient method is introduced to calculate the regurgitant volume in a simplified, but sufficiently accurate manner. A proof of concept of the method is obtained by comparison of the calculated results with results obtained from in vitro experiments. Furthermore, computational fluid dynamics (CFD) simulations are used to validate more complex stenosis scenarios. Comparing the simplified leakage model to CFD simulations reveals its potential for procedure planning and qualitative preoperative risk assessment of PVL. Finally, a 3D device deployment model and the efficient leakage model are combined to showcase the application of the presented leakage model, by studying the effect of stent size and the degree of stenosis on the regurgitant volume. The presented leakage model is also used to visualize the leakage path. To generalize the leakage model to a wide range of clinical applications, further validation on a large cohort of patients is needed to validate the accuracy of the model's prediction under various patient-specific conditions.

Computational modeling of braided venous stents - Effect of design features and device-tissue interaction on stent performance.

Designing venous stents with desired properties is challenging due to the partly conflicting performance criteria, e.g., enhancing flexibility may be at odds with increasing patency. To evaluate the effect of design parameters on the mechanical performance of braided stents, computational simulations are performed using finite element analysis. Model validation is performed through comparison with measurements. Considered design features are stent length, wire diameter, pick rate, number of wires, and stent end-type, being either open-ended or closed looped. Based on the requirements of venous stents, tests are defined to study the effect of design variations with respect to the following key performance criteria: chronic outward force, crush resistance, conformability, and foreshortening. Computational modeling is demonstrated to be a valuable tool in the design process through its ability of assessing sensitivities of various performance metrics to the design parameters. Additionally, it is shown, using computational modeling, that the interaction between a braided stent and its surrounding anatomy has a significant impact on its performance. Therefore, taking into account device-tissue interaction is crucial for the proper assessment of stent performance.