Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data.
Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data.
复制标题
DOI：
10.1109/tbme.2021.3108164
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Wang L
中科院分区：
文献类型：
--
作者：
Gyawali PK;Murkute JV;Toloubidokhti M;Jiang X;Horacek BM;Sapp JL;Wang L
这项工作探讨了心电图（ECG）数据中受试者间解剖学变化的解纠缠表示学习的可能性。由于地面实况解剖因素通常是未知的，在临床心电图评估的模型的解缠能力，所提出的工作首先提出了SimECG数据集，12导联心电图数据集程序生成的一组受控的解剖生成因素。其次，为了执行这样的解纠缠，所提出的方法评估和比较深度生成模型与由非参数印度巴菲特过程建模的潜在密度，以解释ECG数据的复杂生成过程。在模拟数据中，实验首次证明了将ECG数据中的关键生成解剖因素与任务相关生成因素分离的可能性的具体证据。我们实现了92.1%的解开分数，同时解开五个解剖生成因子和任务相关的生成因子。在模拟和真实数据实验中，这项工作进一步提供了量化的证据，有利于解缠学习的下游临床任务定位心室激动的起源。总体而言，所提出的方法对于模拟数据集分别实现了约18.5%和11.3%的改进，对于真实的数据集分别实现了约7.2%和3.6%的改进，超过基线CNN和标准生成模型。这些结果证明了ECG数据中受试者间解剖学变化的解纠缠表示学习的重要性和可行性。这项工作提出了重要的研究方向，以处理由存在显着的受试者间的变化，在自动分析的ECG数据所构成的众所周知的挑战。
This work investigates the possibility of disentangled representation learning of inter-subject anatomical variations within electrocardiographic (ECG) data. Since ground truth anatomical factors are generally not known in clinical ECG for assessing the disentangling ability of the models, the presented work first proposes the SimECG data set, a 12-lead ECG data set procedurally generated with a controlled set of anatomical generative factors. Second, to perform such disentanglement, the presented method evaluates and compares deep generative models with latent density modeled by non-parametric Indian Buffet Process to account for the complex generative process of ECG data. In the simulated data, the experiments demonstrate, for the first time, concrete evidence of the possibility to disentangle key generative anatomical factors within ECG data in separation from task-relevant generative factors. We achieve a disentanglement score of 92.1% while disentangling five anatomical generative factors and the task-relevant generative factor. In both simulated and real-data experiments, this work further provides quantitative evidence for the benefit of disentanglement learning on the downstream clinical task of localizing the origin of ventricular activation. Overall, the presented method achieves an improvement of around 18.5%, and 11.3% for the simulated dataset, and around 7.2%, and 3.6% for the real dataset, over baseline CNN, and standard generative model, respectively. These results demonstrate the importance as well as the feasibility of the disentangled representation learning of inter-subject anatomical variations within ECG data. This work suggests the important research direction to deal with the well-known challenge posed by the presence of significant inter-subject variations during an automated analysis of ECG data.