An experimental and numerical study on the predictive aortic emboli trajectory mapping and optimization in ventricular assist device settings*

Particle-laden flows in cardiovascular settings, especially in the aorta, are inherently challenging to study due to the presence of a high Reynolds number flow regime combined with a mix of large Stokes number embolic particles in a complex geometry. In continuation of our work presented last year, we perform a systematic mapping of inertial particles in four patient-specific aortic models grafted with the cannula of a left ventricular assist device (LVAD). This study experimentally and numerically investigates the Lagrangian particle trajectories in order to develop a predictive model for the fate of embolic particles at two clinically relevant flow rates. High-precision thin-wall phantoms of such models are 3D-printed and placed in a flow loop providing physiological flow conditions. Precision fluorescent beads ranging from 0.4 to 1.2 mm, replicating thrombi transport, are illuminated by a set of near-UV laser and LED light sources. The time-resolved particle tracking velocimetry (PTV) results are complemented by several hemodynamic parameters from a set of CFD simulations. The carefully validated numerical simulations based on our well-controlled experimental boundary conditions shed light on the appropriate choice of particle modeling approaches and numerical solver schemes.