arXiv：cs.LG· Robert A. Lewis, I-Min Chiu, Kyle Verrier, Karthik Jayaraman Raghuram, Francoise Marvel, Salar Abbaspourazad, Anshuman Mishra, Guillermo Sapiro, Andrew C. Miller, Joseph Futoma·· 4 小时前AI 评分37
MS-ECG-FM：利用多源对比学习打造更通用的心电图健康监测基础模型
MS-ECG-FM: Towards a More Universal Electrocardiogram Foundation Model for Health Monitoring using Multi-source Contrastive Learning
AI 导读
研究人员提出心电图基础模型 MS-ECG-FM，通过对比对齐 ECG、超声心动图、放射科和出院报告等多种临床记录类型进行训练。该模型在扩展的 ECG 检测基准上全面超越现有方法，覆盖 ECG 可检测的全部病症范围，在减少导联配置下同样表现优异。不同报告分别改善不同诊断领域的表征，多源对齐则捕捉其互补信息，在临床多样化任务上持续产出强表征。
正文
Abstract:Electrocardiography (ECG) records the electrical activity of the heart, aiding diagnosis by detecting abnormalities in cardiac function. ECG foundation models have demonstrated promising results, but are limited by a reliance on ECG interpretation reports as their sole supervision. Because interpretation reports only capture the subset of waveform information routinely recognized by clinicians, this constrains representation learning to overlook the broader diagnostic signals present in ECG. We introduce a new ECG foundation model --- MS-ECG-FM --- that is trained through contrastive alignment to multiple distinct clinical note types, including ECG, echocardiography, radiology, and discharge reports. We evaluate MS-ECG-FM on an extended set of ECG detection benchmarks, showing that it comprehensively outperforms existing methods on the full span of conditions that ECG can detect, including in reduced-lead configurations. Different reports improve representations for different diagnostic domains, while multi-source alignment captures their complementary information and produces consistently strong representations across clinically diverse tasks.
| Comments: | Andrew C. Miller, Joseph Futoma: equal contribution. 52 pages, 4 figures, 20 tables, including supplementary information |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07662 [cs.LG] |
| (or arXiv:2610.07662v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07662 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Robert A. Lewis [view email]
[v1]
Tue, 6 Oct 2026 02:58:14 UTC (4,873 KB)
来源：arXiv：cs.LG · arxiv.org