解码器设计对心电图波形界定至关重要
Decoder Design Matters for ECG Delineation
浏览论文内容
中文总结 AI 辅助
针对心电图波形界定中解码器设计被忽视的问题，提出R-U-Net（ResNet-18编码器+U-Net解码器），在SemiSegECG上显著优于基线，证明解码器设计比SSL方法更关键。
中文摘要 AI 辅助
心电图（ECG）波形界定识别P波、QRS波群和T波的边界，提供结构注释，可指导AI模型学习解读心电图。然而，训练准确的波形界定模型需要人工注释，这些注释稀缺且耗时。近期工作通过半监督学习（SSL）解决了这一局限，但架构设计，特别是解码器的设计，受到的关注较少。为此，我们提出R-U-Net，一种将ResNet-18编码器与U-Net解码器配对的心电图波形界定模型。在SemiSegECG上，R-U-Net在16个域内设置中的每一个都优于评估的最强ResNet-18+全卷积网络（FCN）头基线，mIoU提高3.3-13.0，并在跨域设置中达到82.6 mIoU，提高了8.1 mIoU。受控消融实验表明，解码器设计对性能提升的贡献大于所评估的SSL方法，这激励了对心电图波形界定架构的进一步探索。所有代码在此http URL开源。
英文摘要
Electrocardiogram (ECG) delineation identifies the boundaries of P waves, QRS complexes, and T waves, providing structural annotations that can guide AI models in learning to interpret ECGs. However, training accurate delineation models requires manual annotations that are scarce and time-consuming to obtain. Recent work addresses this limitation through semi-supervised learning (SSL), but the design of the architecture, particularly the decoder, has received less attention. To this end, we propose R-U-Net, an ECG delineation model that pairs a ResNet-18 encoder with a U-Net decoder. On SemiSegECG, R-U-Net outperforms the strongest evaluated ResNet-18 + fully convolutional network (FCN) head baseline in each of the 16 in-domain settings by 3.3-13.0 mIoU and achieves 82.6 mIoU in the cross-domain setting, an improvement of 8.1 mIoU. Controlled ablations show that decoder design contributes more to performance gains than the evaluated SSL methods, motivating further exploration of architectures for ECG delineation. All code is open-source at github.com/ELM-Research/ECG-Delineation.
发表机构
- Carnegie Mellon University（卡内基梅隆大学）
- Allegheny Health Network（阿勒格尼健康网络）
- University of Colorado（科罗拉多大学）
机构由 AI 辅助整理，请以论文原文为准。
