SURROGATE MODELS OF BLOOD FLOW DYNAMICS IN BRAIN ANEURYSMS USING DYNAMIC MODE DECOMPOSITION

Abstract

The evolution of blood flow is vital in understanding the pathogenesis of brain aneurysms. Several past studies have shown evidence for a turbulent inflow jet at the aneurysm neck. Even though there is a great need for analyzing inflow jet dynamics in clinical practice, data summarized in non-invasive modalities such as Magnetic Resonance Imaging or Computed Tomography are usually limited in spatial and temporal resolutions, and thus cannot account for the hemodynamics. In this work, Dynamic Mode Decomposition (DMD) is used to pinpoint the dominant modes of the inflow jet in patient-specific models of internal sidewall aneurysms utilizing high-resolution data from Computational Fluid Dynamics. The purpose of this thesis is to prove that the dynamic modes are not only governed by the hemodynamics of the parent artery but by the inflow jet interaction with the distal wall. Our work indicates that DMD is an essential tool for analyzing blood flow patterns of brain aneurysms and is a promising tool to be used in in-vivo context.