Atherosclerotic plaque in the carotid artery is an important risk factor for cerebrovascular events. The unique geometry of the carotid bifurcation, including a secondary helical curvature and non-planarity of the daughter vessels, influences local hemodynamics and contributes to plaque formation. Multiple imaging modalities, including inter-arterial angiography, Duplex ultrasound, computed tomographic angiography, and magnetic resonance imaging, are used to diagnose and stratify the severity of carotid stenosis. Experimental and computational models of carotid bifurcation have shown the presence of disturbed flow, decreased wall shear stress, and an increased oscillatory shear index, along the outer wall of the carotid bulb and the internal carotid artery that correlates with areas of plaque deposition. Carotid endarterectomy and carotid stent placement both change the local hemodynamics at the bifurcation by altering the geometry of the carotid sinus and internal carotid artery, impacting long-term outcomes. Machine learning approaches have been increasingly applied to reduce the computational costs of numerical simulations and prognosticate the effects of carotid artery stenosis.
Ho et al. (Fri,) studied this question.