Pulmonary insufficiency (PI) can render the right ventricle dysfunctional due to volume overloading and hypertrophy. The treatment requires a pulmonary valve replacement surgery. However, determining the right time for the valve replacement surgery has been difficult with currently employed clinical techniques such as, echocardiography and cardiac MRI. Therefore, there is a clinical need to improve the diagnosis of PI by using patient-specific (PS) hemodynamic endpoints. While there are many reported studies on the use of PS geometry with time varying boundary conditions (BC) for hemodynamic computation, few use spatially varying PS velocity measurement at each time point of the cardiac cycle. In other words, the gap is that, there are limited number of studies which implement both spatially- and time-varying physiologic BC directly with patient specific geometry. The uniqueness of this research is in the incorporation of spatially varying PS velocity data obtained from phase-contrast MRI (PC-MRI) at each time point of the cardiac cycle with PS geometry obtained from angiographic MRI. This methodology was applied to model the complex developing flow in human pulmonary artery (PA) distal to pulmonary valve, in a normal and a subject with PI.
To validate the methodology, the flow rates from the proposed method were compared with those obtained using QFlow software, which is a standard of care clinical technique. For the normal subject, the computed time average flow rates from this study differed from those obtained using the standard of care technique (QFlow) by 0.8 ml/s (0.9%) at the main PA, by 2 ml/s (3.4%) at the left PA and by 1.4 ml/s (3.8%) at the right PA. For the subject with PI, the difference was 7 ml/s (12.4%) at the main PA, 5.5 ml/s (22.6%) at the left PA and 4.9 ml/s (18.0%) at the right PA. The higher percentage differences for the subject with PI, was the result of overall lower values of the forward mean flow rate caused by excessive flow regurgitation. This methodology is expected to provide improved computational results when PS geometry from angiographic MRI is used in conjunction with PS PC-MRI data for solving the flow field.