Predictive modeling of blood flow in cerebral aneurysms following surgeries
Vitaliy L. Rayz, Ph.D.
Mechanical Engineering, University of Wisconsin – Milwaukee
Neurosurgery, Medical College of Wisconsin
Friday, May 6th
11:25- 11:45 am
Cerebral aneurysms are dilatations of arteries feeding the brain that can present a danger of rupture, brain compression, and thrombotic occlusion (clotting) of downstream vessels. In some cases, when an aneurysm cannot be completely removed from the circulation, it can be treated by altering the flow patterns with either a surgery or a flow diverting stent. Despite their advantages, such treatments introduce complications related to undesired occlusion of vital branch arteries with thrombus. Patient-specific models of cerebral aneurysms constructed from medical imaging data can help treatment planning by predicting flow fields that would result from alternative surgical procedures. In our studies, magnetic resonance imaging was used to obtain patient-specific vascular geometries as well as the flow inlet and outlet conditions. The models were then modified in order to simulate alternative surgeries considered in each case. Numerical solution of the unsteady Navier-Stokes equations was obtained with a finite-volume solver Fluent. In addition, the advection-diffusion equation was solved in order to assess the flow residence time and determine intra-aneurysmal regions that are likely to become occupied by thrombus following a procedure. The CFD-predicted flow fields were compared to phase-contrast MRI (PC-MRI) measurements, providing time-resolved, three-dimensional velocity fields. A close agreement was observed between the in vivo and in vitro PC-MRI measurements and CFD simulations. The results indicate that image-based flow models may help improve the outcome of surgeries in cerebral aneurysms.