Synthetic Database of Aortic Morphometry and Hemodynamics: Overcoming Medical Imaging Data Availability.

Modeling of hemodynamics and artificial intelligence have great potential to support clinical diagnosis and decision making. While hemodynamics modeling is extremely time- and resource-consuming, machine learning (ML) typically requires large training data that are often unavailable. The aim of this study was to develop and evaluate a novel methodology generating a large database of synthetic cases with characteristics similar to clinical cohorts of patients with coarctation of the aorta (CoA), a congenital heart disease associated with abnormal hemodynamics. Synthetic data allows use of ML approaches to investigate aortic morphometric pathology and its influence on hemodynamics. Magnetic resonance imaging data (154 patients as well as of healthy subjects) of aortic shape and flow were used to statistically characterize the clinical cohort. The methodology generating the synthetic cohort combined statistical shape modeling of aortic morphometry and aorta inlet flow fields and numerical flow simulations. Hierarchical clustering and non-linear regression analysis were successfully used to investigate the relationship between morphometry and hemodynamics and to demonstrate credibility of the synthetic cohort by comparison with a clinical cohort. A database of 2652 synthetic cases with realistic shape and hemodynamic properties was generated. Three shape clusters and respective differences in hemodynamics were identified. The novel model predicts the CoA pressure gradient with a root mean square error of 4.6 mmHg. In conclusion, synthetic data for anatomy and hemodynamics is a suitable means to address the lack of large datasets and provide a powerful basis for ML to gain new insights into cardiovascular diseases.

Publisher Correction: Model-Based Therapy Planning Allows Prediction of Haemodynamic Outcome after Aortic Valve Replacement.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

Model-Based Therapy Planning Allows Prediction of Haemodynamic Outcome after Aortic Valve Replacement.

Optimizing treatment planning is essential for advances in patient care and outcomes. Precisely tailored therapy for each patient remains a yearned-for goal. Cardiovascular modelling has the potential to simulate and predict the functional response before the actual intervention is performed. The objective of this study was to proof the validity of model-based prediction of haemodynamic outcome after aortic valve replacement. In a prospective study design virtual (model-based) treatment of the valve and the surrounding vasculature were performed alongside the actual surgical procedure (control group). The resulting predictions of anatomic and haemodynamic outcome based on information from magnetic resonance imaging before the procedure were compared to post-operative imaging assessment of the surgical control group in ten patients. Predicted vs. post-operative peak velocities across the valve were comparable (2.97 ± 1.12 vs. 2.68 ± 0.67 m/s; p = 0.362). In wall shear stress (17.3 ± 12.3 Pa vs. 16.7 ± 16.84 Pa; p = 0.803) and secondary flow degree (0.44 ± 0.32 vs. 0.49 ± 0.23; p = 0.277) significant linear correlations (p < 0.001) were found between predicted and post-operative outcomes. Between groups blood flow patterns showed good agreement (helicity p = 0.852, vorticity p = 0.185, eccentricity p = 0.333). Model-based therapy planning is able to accurately predict post-operative haemodynamics after aortic valve replacement. These validated virtual treatment procedures open up promising opportunities for individually targeted interventions.

Is MRI-based CFD able to improve clinical treatment of coarctations of aorta?

Pressure drop associated with coarctation of the aorta (CoA) can be successfully treated surgically or by stent placement. However, a decreased life expectancy associated with altered aortic hemodynamics was found in long-term studies. Image-based computational fluid dynamics (CFD) is intended to support particular diagnoses, to help in choosing between treatment options, and to improve performance of treatment procedures. This study aimed to prove the ability of CFD to improve aortic hemodynamics in CoA patients. In 13 patients (6 males, 7 females; mean age 25 ± 14 years), we compared pre- and post-treatment peak systole hemodynamics [pressure drops and wall shear stress (WSS)] vs. virtual treatment as proposed by biomedical engineers. Anatomy and flow data for CFD were based on MRI and angiography. Segmentation, geometry reconstruction and virtual treatment geometry were performed using the software ZIBAmira, whereas peak systole flow conditions were simulated with the software ANSYS(®) Fluent(®). Virtual treatment significantly reduced pressure drop compared to post-treatment values by a mean of 2.8 ± 3.15 mmHg, which significantly reduced mean WSS by 3.8 Pa. Thus, CFD has the potential to improve post-treatment hemodynamics associated with poor long-term prognosis of patients with coarctation of the aorta. MRI-based CFD has a huge potential to allow the slight reduction of post-treatment pressure drop, which causes significant improvement (reduction) of the WSS at the stenosis segment.

The impact of MRI-based inflow for the hemodynamic evaluation of aortic coarctation.

Aortic coarctation (CoA) accounting for 3-11% of congenital heart disease can be successfully treated. Long-term results, however, have revealed decreased life expectancy associated with abnormal hemodynamics. Accordingly, an assessment of hemodynamics is the key factor in treatment decisions and successful long-term results. In this study, 3D angiography whole heart (3DWH) and 4D phase-contrast magnetic resonance imaging (MRI) data were acquired. Geometries of the thoracic aorta with CoAs were reconstructed using ZIB-Amira software. X-ray angiograms were used to evaluate the post-treatment geometry. Computational fluid dynamics models in three patients were created to simulate pre- and post-treatment situations using the FLUENT program. The aim of the study was to investigate the impact of the inlet velocity profile (plug vs. MRI-based) with a focus on the peak systole pressure gradient and wall shear stress (WSS). Results show that helical flow at the aorta inlet can significantly affect the assessment of pressure drop and WSS. Simplified plug inlet velocity profiles significantly (p < 0.05) overestimate the pressure drop in pre- and post-treatment geometries and significantly (p < 0.05) underestimate surface-averaged WSS. We conclude that the use of the physiologically correct but time-expensive 4D MRI-based in vivo velocity profile in CFD studies may be an important step towards a patient-specific analysis of CoA hemodynamics.