Data and simulation-based exploration, analysis, and treatment of stenoses for prevention of ischemic strokes

Ischemic stroke is one of the primary causes of death and disability worldwide. Common origins of cerebral ischemia are the two carotid arteries that supply most of the intracranial circulation. Therefore, detecting and analyzing carotid stenoses, i.e., potentially lethal constrictions of these arteries, is a critical task in clinical stroke prevention and treatment. Carotid surgery can reduce the stroke risk associated with stenoses, however, the procedure entails risks itself. Therefore, a thorough evaluation of each case is necessary. Clinical evaluation requires assessing the vessel shape (morphology) and the properties of the internal blood flow (hemodynamics). The latter is typically assessed using Doppler ultrasonography, which requires specialized equipment and personnel to be carried out. With the emergence of computational fluid dynamics (CFD), the simulation of intricate biological transport processes became possible, including flow in actual physiological structures, such as blood vessels. These simulations of the internal blood flow field could facilitate the visual integration of hemodynamic and morphological information, provide a higher resolution on relevant parameters than current screening methods, and enable the analysis of new flow-derived parameters associated with stenosis progression. While hemodynamic simulations offer groundbreaking opportunities to enhance clinical decision-making, a successful translation of CFD to the clinical forefront is highly challenging. In this thesis, a self-contained pipeline is built, which integrates streamlined model extraction and pre-processing along with a collection of visual exploration tools into a single framework. Methods are proposed and evaluated to analyze the internal blood flow, anatomical context, and vessel wall composition, and to automatically and reliably classify stenosis candidates.