Computational Modeling of Flow Diverting Devices in Intracranial Aneurysms

When evaluating unruptured cerebral aneurysms, it is necessary to assess their risk of rupture and to compare them to the risks involved in their treatment. Simulations using the patient-specific (P-S) geometry of the aneurysm may help in a better planning of the treatment and in a consequent reduction of the associated risks.;To have a better understanding of the rupture risks, we first review studies with the suggested hypotheses that connect the aneurysm risk factors and the mechanisms governing the aneurysm evolution. This literature review reveals a progressive wall degradation due to changing hemodynamic loading and biomechanic stress, affected by risk factors, that drives the geometrical evolution of the aneurysm until it stabilizes or ruptures. However, details of these interactions or their relative importance are still not clearly understood.;Second, to understand the influence of uncertainties involved in the P-S computational fluid dynamics (CFD) simulations, we compare the blood flow field in a growing cerebral aneurysm obtained with experimental particle image velocimetry (PIV) and CFD techniques. Despite small differences observed, mainly associated to the inherent limitations of each technique, the information derived is consistent and can be used to study the role of hemodynamics in the natural history of intracranial aneurysms.;The final objective of this work is to develop a methodology to carry on faster than current P-S simulations, allowing for their application to treatment planning and device design. We propose, validate, and implement a methodology for the simulation of flow diverting devices (FD) in aneurysms by using a porous medium method (PMM), which greatly reduces the computational cost of these simulations. The method relies on parameters from an empirical correlation derived from experimental observations in wire screens, consistent with CFD simulations. The validation of our PMM strategy was carried out by comparing the results of simulations in distinct P-S geometries and FDs, to those obtained under identical conditions by the immersed method (IMM) currently used. Overall, both quantitative and qualitative results are consistent between IMM and PMM in cases where the local porosity remains roughly uniform throughout the neck, with differences in the reduction of the observables lower than 10%. This PMM strategy is between 2 and 10 times faster than the IMM, which allows for a runtime of hours instead of days, bringing it closer for its application in the clinic.