Application of Complex Networks Theory and Computational Fluid Dynamics to the Study the Near-Wall Hemodynamics in the Aneurysm of the Abdominal Aorta

Onset and development of cardiovascular pathologies are strictly linked to flow features and hemodynamic forces' configuration. Specific Wall Shear Stress (WSS) patterns are highly responsible for most vascular wall alterations such as atherosclerosis and aneurysms' onset and progression. In this work, six abdominal aortic aneurysm (AAA) models are investigated, each one at a baseline and a follow-up time point, applying, for the first time in AAA flows, the Complex Networks (CNs) theory. A frequency domain analysis was also performed to isolate high-frequency WSS-based signals. The aim of the study is enriching the knowledge of AAA physiopathology unveiling the possible links between statistically related spatiotemporal flow patterns and the vascular disease progression. 4D-CT imaging allowed an accurate reconstruction of vessels' geometry, and unsteady-state large eddy simulations (LES) are carried out. WSS vector fields calculation is followed by the analysis of 3 major derived quantities, namely WSS magnitude, WSS axial component (WSSax) and WSS secondary component (WSSsc). Starting from the above-mentioned quantities’ time-series along the cardiac cycle, correlation-based networks are built for each AAA geometry and CNs metrics are computed to characterize the topological structure of the networks, and compared to classical WSS descriptors (Time Averaged Wall Shear Stress, Oscillatory Shear Index and Relative Residence Time). Among the main results, the CNs analysis of WSS magnitude and axial WSS shows well-defined and extended areas on the aneurysmal sac characterized by a large heterogeneity of the time-series. Such areas of low statistically related flow patterns are also partially co-localized with low WSS luminal surfaces in some of the investigated models. Finally, from the frequency domain analysis it emerges a moderate correspondence between heterogeneity of WSSax time-series, as measured by CNs, and high frequency spectral power density.