用标签引导的跨模态融合，让成人心电图模型更好识别儿童心电图。
Label-Conditioned Cross-Modal Fusion for Adult-to-Pediatric ECG Transfer via Curriculum-Gated Contrastive Alignment
- 通过标签条件对比学习对齐心电图与诊断语义，提升迁移效果。
- 零样本、50样本和全微调下在儿童数据集上分别达59.39%、81.74%、91.56%的AUC。
- 仅需心电图信号推理，但训练时用文本描述增强语义监督，适合标注稀缺场景。
自动化儿童心电图（ECG）解读仍具挑战，因心率、间期和波形发育差异限制了基于成人数据训练模型的可迁移性，而专家标注的儿童心电图数据集稀缺。本文提出PEACE（Pediatric-Adult ECG Alignment via Cross-modal Enhancement），一种基于MIMIC-IV心电图预训练并适配至儿童目标的成人到儿童心电图迁移框架。PEACE整合标签特定双向对比学习（LSBC）以对齐心电图表示与诊断语义，并采用课程自适应融合（CAF）在有限儿童标注下稳定优化过程。标签相关的短文本描述在训练中提供辅助语义监督，推理阶段仅需心电图信号。在ZZU-pECG数据集上，零样本、50样本和全微调设置下的宏平均AUC分别为59.39%、81.74%和91.56%，优于仅用心电图、多模态及通用域适应基线（如DANN、MMD）。在PTB-XL上，全微调后九个调和标签的宏平均AUC达96.90%。梯度注意力图显示，对于右室肥厚（RVH）关注QRS电压与形态区域，对于长QT综合征（LQTS）关注QRS-T/复极间期，与常规临床判读区域一致。结果表明，结合成人规模预训练与节律、形态及ST-T复极语义描述，可在标注稀缺条件下提升可迁移的儿童诊断性能，并保持临床可解释的波形关注点。
原文摘要 · Abstract (English)
Automated pediatric electrocardiogram (ECG) interpretation remains challenging because developmental differences in heart rate, intervals, and waveforms limit the transferability of models trained mainly on adult data, while expert-labeled pediatric ECG cohorts are scarce. We propose PEACE (Pediatric-Adult ECG Alignment via Cross-modal Enhancement), an adult-to-pediatric ECG transfer framework pretrained on MIMIC-IV ECGs and adapted to pediatric targets. PEACE integrates label-specific bidirectional contrastive learning (LSBC) to align ECG representations with diagnostic semantics and curriculum adaptive fusion (CAF) to stabilize optimization under limited pediatric supervision. Label-conditioned short text descriptors provide auxiliary semantic supervision during training, whereas inference requires ECG signals only. On ZZU-pECG, PEACE achieves macro-average AUCs of 59.39%, 81.74%, and 91.56% under zero-shot, 50-shot, and full fine-tuning settings, respectively, outperforming ECG-only, multimodal, and generic domain adaptation baselines including DANN and MMD. On PTB-XL, it reaches 96.90% macro-average AUC after full fine-tuning over nine harmonized labels with nonzero mapped incidence. Gradient-based attention maps show increased saliency around QRS voltage and morphology regions for chamber-related RVH and around QRS-to-T/repolarization intervals for LQTS, broadly consistent with ECG regions commonly inspected during routine interpretation. These results suggest that adult-scale ECG pretraining coupled with rhythm, morphology, and ST-T repolarization semantic descriptors improves transferable pediatric diagnosis under label scarcity while preserving clinically interpretable waveform focus.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。