MRI-Based Pressure Gradient Mapping in Patient-Specific Models of Coarctation of the Aorta

Fig. 1. (A) Model generation pipeline in SimVascular. (B) Use of the svMorph tool to increase the severity of the stenosis in the 3D model.

Fig. 2. (A) Patient-specific phantom generation process. (B)3D-printed patient-specific phantoms of CoA with compliant walls. White circles indicate location of CoA

Fig. 3. (A) MRI-compatible in vitro cardiovascular system emulator with an embedded 3D-printed compliant patient-specific CoA model instrumented for direct pressure and flow rate measurements. The system includes tunable fluid capacitors (C1 and C2) and resistors (R1 and R2) to match patient-specific hemodynamics. White arrows indicate the direction of flow. (B) Zoomed in view of the model-specific gel block with embedded aortic phantom and pressure catheter ports (blue) at the inlet and outlet. (C) The MRI-compatible cardiovascular system emulator prepared for a 3T MRI (Vida Fit, Siemens) study on the scanner bed.

Fig. 4. Velocity fields measured using conventional 4D-Flow (left), 4D-FlowP (middle) and FSI simulations (right) at the timepoint of peak velocity in two patients with CoA.

Fig. 5. Relative pressure fields derived from conventional 4D-Flow velocity estimates (left), velocity and acceleration measurements from 4D-FlowP (middle) and FSI simulations (right)

Fig. 6. Comparison of pressure drop estimated in CoA phantoms using conventional 4D-Flow (left), 4D-FlowP (middle), and FSI simulations (right) with catheter-derived pressure measurements. Dashed line represents the linear regression.

Fig. 7. Bland-Altman plots comparing pressure gradient (ΔP) estimates from conventional 4D-Flow (left), 4D-FlowP (middle) and FSI simulations (right) to catheter-derived ΔPCath. The solid lines represent the mean difference between the two methods, and the paired dotted lines correspond to the 95% limits of agreement(LoA).