Coupled lattice Boltzmann and Discrete Element Method simulations with applications to liver radioembolization
Tristan Vlogman is a PhD student in the Department of Engineering Fluid Dynamics. (Co)Promotors are prof.dr.ir. C.H. Venner, dr. K. Jain and dr.ir. R. Hagmeijer from the Faculty Engineering Technology and dr. F. Nijsen from the Radboud University Medical Center.
Trans-arterial radioembolization (TARE) is a treatment for liver cancer in which small radioactive particles called microspheres are injected into the hepatic artery. These microspheres travel with the bloodstream until they lodge in the liver tissue, where they deliver their radiation dose over time. The treatment’s success depends on the distribution of these microspheres: a large dose must be realized at the tumors while toxicity to the healthy tissue must be limited.
Unfortunately, optimal targeting is not always achieved in practice. As a result, the tumor dose needed for the treatment to be effective is not always reached and treatment outcome of TARE varies per patient. To find out why a favorable microsphere distribution is not always obtained, knowledge about the factors influencing the microsphere distribution is needed. An optimal set of treatment parameters may then be identified to maximize tumor dose while minimizing toxicity to healthy tissue.
The work in this thesis aims to contribute towards this goal by studying TARE through the development and implementation of a coupled lattice Boltzmann (LBM) and Discrete Element Method (DEM) code to solve the motion of solid particles suspended in a fluid. Both fully resolved and unresolved coupling approaches are implemented, validated and compared. The method used is then applied to study the effect of injection velocity, strategy and timing on the microsphere distributions in an idealized liver geometry. LBM-DEM simulations, validated with in vitro experiments, suggest that the sensitivity of the microsphere distributions to injection velocity depends on the ratio of jet breakdown length to the distance to the first bifurcation. Injection strategy (pulsed vs continuous injection) and the phase-shift between the injection and cardiac cycle were also shown to influence distributions.
