Spatiotemporal Analysis of Particle Spread to Assess the Hybrid Particle-Flow CFD Model of Radioembolization of HCC Tumors

Objective: Computational fluid dynamics(CFD) models can potentially aid in pre-operative planningof transarterial radioactive microparticle injections totreat hepatocellular carcinoma, but these models arecomputationally very costly. Previously, we introducedthe hybrid particle-flow model as a surrogate, less costlymodelling approach for the full particle distribution in trun-cated hepatic arterial trees. We hypothesized that highercross-sectional particle spread could increase the matchbetween flow and particle distribution. Here, we investigatewhether truncation is still reliable for selective injectionscenarios, and if spread is an important factor to considerfor reliable truncation.Methods:Moderate and severe up-and downstream truncation for selective injection servedas input for the hybrid model to compare downstreamparticle distributions with non-truncated models. In eachsimulation, particle cross-sectional spread was quantifiedfor 5-6 planes.Results:Severe truncation gave maximumdifferences in particle distribution of similar to 4-11% and similar to 8-9% fordown- and upstream truncation, respectively. For moderatetruncation, these differences were only similar to 1-1.5% and similar to 0.5-2%. Considering all particles, spread increased down-stream of the tip to 80-90%. However, spread was found tobe much lower at specific timepoints, indicating high time-dependency.Conclusion:Combining domain truncationwith hybrid particle-flow modelling is an effective method toreduce computational complexity, but moderate truncationis more reliable than severe truncation. Time-dependentspread measures show where differences might arisebetween flow and particle modelling.Significance:Thehybrid particle-flow model cuts down computational time significantly by reducing the physical domain, paving theway towards future clinical applications