Boundary and initial conditions

For the phantom, the imposed volumetric flow rate of 4.5 l/min is identical to the experimental conditions and corresponds to the inlet Reynolds number of Re=4539 (Re=V0·D0/ν). For the patient-specific aorta, the time-dependent inlet conditions are matched with MRI measurements during the entire cardiac cycle. We have extracted the measured volumetric flow rate at the inlet plane (Q0), and converted it to the characteristic mass flow rate (m˙=Q0·ρblood). The mass flow rates were fitted with a smooth spline with piecewise polynomial (with a smoothing parameter p=0.99999947 and R2=0.9995), Fig. 7b, which gives the following range of the inlet Reynolds number, 0≤Re≤6276, and corresponding Womersley number of Wo=D0ωf/ν1/2=29. In total, we have simulated five cardiac cycles, to obtain results without the influence of initial conditions. Only the last cycle was used for the analysis. For all turbulence parameters, the uniform inlet values were imposed with the following specifications: the intensity of turbulence of 5%, the ratio of turbulent and molecular viscosity (μt/μ) of 10, the isotropic assumption of normal turbulent stress components (uiui¯=2/3k), and zero values of the turbulent shear stress components (uiuj¯=0). At outlets, a zero diffusion flux was imposed for all transport variables. For the patient-specific aorta, a pre-defined fixed (MRI-based) percentage of the inlet flow rate was prescribed. The no-slip velocity boundary condition was imposed at the walls of blood vessels, and the model was assumed to be rigid.