Combining Computational Hemodynamics and Complex Networks Theory for the Study of coherent fluid structures before and after endarterectomy in the carotid bifurcation

The study and understanding of arterial hemodynamics are essential to comprehend the onset and progression of atherosclerotic disease. Blood flow patterns in arteries are highly complex and spatiotemporally heterogeneous, and over the years many hemodynamic descriptors have been suggested to quantify and visualize them. However, the conventional hemodynamic tools allow a partial understanding of the spatiotemporal evolution and persistence length of the large-scale coherent flow structures into the vessels. Trying to overcome this lack of information, very recently the Complex Networks (CNs) theory has been employed as a tool of analysis of cardiovascular flows, integrating the classical Computational Fluid Dynamics (CFD) modeling. Inspired by a recent study on 10 ostensibly healthy carotid bifurcation (CB) models, in this thesis CNs are applied for the first time to a cohort of 13 patient-specific computational hemodynamics carotid models before and 1 month after carotid endarterectomy (CEA). By comparing the results of the CNs study performed on a dataset of healthy CBs with those obtained from the pre- and post-CEA models, this thesis aims at (1) investigating how the atherosclerotic disease modifies the spatiotemporal evolution of hemodynamic structures, and (2) establishing to what extent the physiological flow coherence of large-scale structures is restored by surgery. To do that, correlation-based networks were built from the time-histories of two fluid mechanics quantities of physiological significance, respectively (1) the blood velocity vector axial component locally aligned with the main flow direction and (2) the kinetic helicity density. It emerged that, in the pre-CEA models, spatiotemporal flow coherence of axial structures is only disrupted locally, near the stenosis. Similarly to the healthy CBs, in the post-CEA models the restored physiological bifurcation expansion disrupts the network topological connections between axial flow structures. On the other hand, the spatiotemporal coherence of helical flow patterns is overall less markedly influenced by the presence of the stenosis. The findings of this work confirm the potential of CNs theory applied to diseased blood vessels in capturing the effect of vascular pathology on the spatiotemporal evolution of large-scale blood flow patterns.