Synthetic dataset of 1,500 anatomies representing the bifurcation of the left coronary artery
收藏资源简介:
该数据集由米兰理工大学等机构创建,主要用于研究冠状动脉疾病的血流动力学模拟。数据集包含1500个合成的左冠状动脉分叉模型,每个模型随机引入了1到3个狭窄,狭窄程度在20%到60%之间。数据集的生成过程包括从健康受试者重建分叉模型,并通过CFD模拟生成血流动力学数据。该数据集的应用领域主要集中在心血管疾病的诊断和治疗,旨在通过几何深度学习模型预测血流动力学标量场,从而替代传统的CFD模拟,减少计算成本和时间。
This dataset was developed by institutions including Politecnico di Milano, primarily for research on hemodynamic simulations of coronary artery disease. It contains 1500 synthetic left coronary artery bifurcation models, each of which randomly incorporates 1 to 3 stenoses with stenosis severity ranging from 20% to 60%. The dataset generation workflow involves reconstructing bifurcation models from healthy human subjects and generating hemodynamic data through CFD simulations. The core application scenarios of this dataset focus on the diagnosis and treatment of cardiovascular diseases, aiming to predict hemodynamic scalar fields using geometric deep learning models, thus replacing traditional CFD simulations to reduce computational costs and time.

- 1Learning Hemodynamic Scalar Fields on Coronary Artery Meshes: A Benchmark of Geometric Deep Learning Models米兰理工大学 · 2025年



