心脏激活时间映射预测数据集
收藏资源简介:
该数据集包含了从有限元(FE)模拟中生成的17,500个左心室(LV)几何形状的合成数据,这些数据被用于训练和测试深度学习模型。数据集涵盖了广泛的LV几何形状、起搏位点配置和心肌电导率。这些数据被用于训练几何深度学习模型,以预测心脏激活时间映射,从而优化心脏再同步治疗(CRT)的计划。数据集的创建过程涉及使用有限元方法对心脏电生理学进行模拟,并计算与随机选择的起搏位点相关的激活模式。数据集的应用领域是心脏再同步治疗,旨在解决患者个体化解剖结构的变异性和当前个性化计划策略的局限性所带来的挑战。
This dataset contains 17,500 synthetic datasets of left ventricle (LV) geometries generated from finite element (FE) simulations, which are utilized for training and testing deep learning models. It covers a broad spectrum of LV geometries, pacing site configurations, and myocardial conductivities. The dataset is employed to train geometric deep learning models for predicting cardiac activation time mapping, thereby optimizing the planning of cardiac resynchronization therapy (CRT). The development of this dataset entails simulating cardiac electrophysiology using the finite element method, and calculating activation patterns associated with randomly selected pacing sites. The target application field of this dataset is cardiac resynchronization therapy, which aims to address the challenges posed by the variability of individual patients' anatomical structures and the limitations of current personalized planning strategies.

- 1Rapid cardiac activation prediction for cardiac resynchronization therapy planning using geometric deep learning密歇根州立大学机械工程系 · 2025年



