Designing vascular stents through topology optimization: challenges and recent advances
Please login to view abstract download link
Stent implantation is a cornerstone treatment for vascular diseases, yet the performance and safety of stents heavily depend on their design. Conventional design approaches, which rely on incremental modi- fications to a limited set of stent geometries, often struggle to deliver innovative, patient-specific designs. However, optimization-based design methods must address diverse clinical and manufacturing require- ments to produce stents that are viable for practical applications. To tackle these challenges, we propose a topology optimization (TO) framework for designing vascular stents. In this approach, the stent is modeled as a cylindrical surface composed of repeating 2D unit cells. These cells are generated by solving a TO problem using the microSIMPATY algorithm [4], which employs inverse homogenization to create microscopic unit cells with desired macroscopic properties defined by a homogenized stiffness tensor. The algorithm also incorporates an anisotropic adaptive mesh to refine the designs. The resulting unit cells feature non-conventional designs that differ significantly from commercial stents [1]. The TO problem’s objective function and constraints can be tailored to achieve specific design properties, accounting for clinical and manufacturing requirements [2, 3]. In this talk, we focus on the formulation of the optimization problem. In this context, we also illustrate how the homogenized stiffness tensor’s components can be linked to clinically relevant quantities. We also highlight recent advances that bring TO-designed stents closer to practical application. For instance, we discuss the adaptation of an overhang constraint – originally developed in the context of TO for additive manufacturing – to improve the hemodynamic properties of the optimized designs. Acknowledgments: This study was carried out within the RESET project, funded by the European Union – NextGenerationEU, Italian Ministry of University and Research, Italy, within the PRIN 2022 PNRR program (D.D.1409 del 14/09/2022). REFERENCES [1] D. Carbonaro et al., Comput. Methods Appl. Mech. Engrg, 416:116288, 2023. [2] D. Carbonaro et al., Comput. Methods Programs Biomed., 257:108467, 2024. [3] N. Ferro et al., Finite Elem. Anal. Des., 244:104304, 2025. [4] N. Ferro, S. Micheletti and S. Perotto, Lecture Notes in Computational Science and Engineering, 132:211–221, 2020.