Scholarlink

Abstract <p>Cardiovascular diseases remain the leading cause of mortality. Automated electrocardiogram (ECG) analysis can ease clinical workloads but is limited by scarce and imbalanced data. Synthetic ECGs can mitigate these issues, and while most methods use generative adversarial networks (GANs), recent work has shown that variational autoencoders (VAEs) perform comparably. This paper presents a cNVAE-ECG model, a conditional Nouveau VAE (NVAE) that generates high-resolution 10-s 12-lead ECGs with multiple pathologies. Leveraging a compact channel-generation scheme and class embeddings for multilabel conditioning, cNVAE-ECG improves downstream binary and multilabel classification, achieving an AUROC gain of up to 2% in transfer learning over GAN-based models. The model is publicly available at <a href="https://github.com/univanxx/cNVAE_ECG">https://github.com/univanxx/cNVAE_ECG</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Conditional Electrocardiogram Generation Using Hierarchical Variational Autoencoders

  • 期刊论文

  • 发表日期：

    2025年12月1日

查看全文
  • I. A. Sviridov,
  • K. S. Egorov

摘要

Abstract

Cardiovascular diseases remain the leading cause of mortality. Automated electrocardiogram (ECG) analysis can ease clinical workloads but is limited by scarce and imbalanced data. Synthetic ECGs can mitigate these issues, and while most methods use generative adversarial networks (GANs), recent work has shown that variational autoencoders (VAEs) perform comparably. This paper presents a cNVAE-ECG model, a conditional Nouveau VAE (NVAE) that generates high-resolution 10-s 12-lead ECGs with multiple pathologies. Leveraging a compact channel-generation scheme and class embeddings for multilabel conditioning, cNVAE-ECG improves downstream binary and multilabel classification, achieving an AUROC gain of up to 2% in transfer learning over GAN-based models. The model is publicly available at https://github.com/univanxx/cNVAE_ECG.