SynthAorta: A 3D Mesh Dataset of Parametrized Physiological Healthy Aortas

TL;DR

SynthAorta bridges clinical and engineering perspectives by generating parametrized synthetic aortas from a single base model using literature-supported distributions, convolution-surface geometry, and a structured hex mesh suitable for simulations and data-driven analyses. The method yields a large, simulation-ready dataset with centerlines, surfaces, and multi-level meshes, enabling systematic studies of how geometry affects hemodynamics. Key contributions include a centerline–radius parametrization informed by weighted literature statistics, a KL-based radius sampling strategy, and a validated flow simulation workflow showing convergence and physiologic QoIs. This dataset supports uncertainty quantification, reduced-order modeling, and future exploration of pathological geometries, with open-source availability for broad adoption.

Abstract

The effects of the aortic geometry on its mechanics and blood flow, and subsequently on aortic pathologies, remain largely unexplored. The main obstacle lies in obtaining patient-specific aorta models, an extremely difficult procedure in terms of ethics and availability, segmentation, mesh generation, and all of the accompanying processes. Contrastingly, idealized models are easy to build but do not faithfully represent patient-specific variability. Additionally, a unified aortic parametrization in clinic and engineering has not yet been achieved. To bridge this gap, we introduce a new set of statistical parameters to generate synthetic models of the aorta. The parameters possess geometric significance and fall within physiological ranges, effectively bridging the disciplines of clinical medicine and engineering. Smoothly blended realistic representations are recovered with convolution surfaces. These enable high-quality visualization and biological appearance, whereas the structured mesh generation paves the way for numerical simulations. The only requirement of the approach is one patient-specific aorta model and the statistical data for parameter values obtained from the literature. The output of this work is SynthAorta, a dataset of ready-to-use synthetic, physiological aorta models, each containing a centerline, surface representation, and a structured hexahedral finite element mesh. The meshes are structured and fully consistent between different cases, making them imminently suitable for reduced order modeling and machine learning approaches.

Figures (5)

• Figure 1: The overview of the SynthAorta approach: starting from a single base model, physiologically motivated geometrical and radius parameters are modeled statistically, and varied on the level of the coarse skeleton. The surface is reconstructed purely from the skeleton using convolution surfaces, enabling the generation of a dataset of structured hexahedral meshes, which can be used in numerical simulations.
• Figure 2: (a) Schematic representation of the aorta, divided into three segments: the ascending aorta, the aortic arch, and the descending aorta. (b) The aortic centerline divided into the three aortic segments. The points $P_1$, $P_2$, and $P_3$ mark the locations where the carotid arteries intersect with the aortic path.
• Figure 3: Centerline curvature radius and descending aorta adjustment parameters on the plane-projected base centerline: (a) The locations of $\mathbf{A}$ and $\mathbf{B}$, as well as the visualization of the centerline curvature radius $R_\text{c}$ and the center point $\mathbf{x}_{\text{c}}$, (b) the variation in the centerline curvature radius modifies the points in the upper part of the aorta, in the direction of the center $\mathbf{x}_{\text{c}}$, and (c) the correction of the previously unmodified points.
• Figure 4: Examples from the SynthAorta dataset: structured hexahedral meshes of different aortic geometries, in different views. Note that the aortas are scaled for better visibility, but they are of varying heights/lengths as well.
• Figure 5: Temporal trace of the QoIs in the abdominal aorta: flow rate (left), mean pressure (middle) and spatial average of the wall shear stress (right) in the abdominal aorta. The deviation from the mean over the entire SynthAorta dataset (black dashed line) is largest in the systolic phase, while the pressure and flow show significantly less variation than the WSS.