Full 3D blood flow simulation in curved deformable vessels using physics-informed neural networks

Investigating blood flow in the cardiovascular system is essential for assessing cardiovascular health. Numerical simulations are widely employed as a non-invasive alternative to traditional measurements, reducing risks for patients. In this study, we extend our previous research by introducing a flexible and ef-ficient approach for real-world simulations in a fully 3D framework. We employ physics-informed neural networks (PINNs) to solve the Navier-Stokes equations in a dynamic, deformable domain, focusing on simulating blood flow through elastic vessels with various bending degrees. The mechanics modeling of the interface of the fluid and structure also utilizes a full 3D model, providing more numerical charac-teristics for simulations of fluid mechanics within cardiovascular systems. Mesh-free approach circumvents the need for discretization and meshing, thus enhancing computational efficiency for complex geometries. Experiments on vessels with varying degrees of curvature are included. The analysis of blood flow mechanics indicates that highly curved vessels significantly reduce fluid velocity and exhibit diminished activity during the diastolic phase in a non-linear manner. © 2024, Comenius University in Bratislava. All rights reserved.