Analysis of Turbulence Models to Simulate Patient-Specific Vortex Flows in Aortic Coarctation

Coarctation of the aorta is a localized narrowing of the aortic lumen. This pathology leads to hypertension in upper extremity vessels, left ventricular hypertrophy and to impaired perfusion of the abdominal cavity and lower extremities. Along with traditional diagnostic methods, mathematical modeling is used for risk assessment and the prediction of disease outcomes. However, when applying numerical models to describe hemodynamic parameters, the choice of turbulence model to describe swirling flow occurring in the aorta in this pathology must be justified. Thus, three turbulence models, namely k-ε, k-ω, and SST were analyzed for the description of swirling flows in the study of coarctation’s effect on hemodynamic parameters and analysis of the mechanisms leading to various cardiovascular diseases caused by altered hemodynamics. The results revealed significant differences in swirling flow patterns between the k-ε and k-ω models, while the k-ω and SST models showed consistent results over the cardiac cycle. In the peak systolic phase, average velocity rises to 1.07–1.98 m·s−1 for the k-ε model, 0.82–2.12 m·s−1 for the k-ω model, 1.22–2.12 m·s−1 for the SST model and 0.8–2.12 m·s−1 for laminar flow. WSS values increase rapidly to 11–22 Pa in k-ε, 25–50 Pa in k-ω and SST models of turbulence, and 30–55 Pa for laminar flow. Significant differences were also evident in the prediction of wall shear stress, with the k-ε model giving values more than twice as high as the k-ω and SST models. The data obtained confirm the necessity of careful model selection for accurate hemodynamic parameter estimation, especially in coarctation. The findings of this study can be used for further physics-informed neural network analysis of evaluation of treatment evaluations for congenital heart disease patients.