arXiv cs.AI· Cheol-Hui Lee, Wookyung Kim, Hyeonhoon Lee, Garam Lee, Da-in Eun, Hyung-Chul Lee·· 3 小时前AI 评分38
CARMEN：基于十种生理信号预训练的心肺基础模型，面向围术期与重症监护
A Cardiorespiratory Foundation Model Pretrained on Ten Physiological Signals for Perioperative and Critical Care
AI 导读
研究者发布 CARMEN，一个在 227 万模态小时、十种心肺生理波形上无标签预训练的单一编码器，支持 10 秒到 5 分钟任意组合的同步信号。其 LSCNorm 条件化机制可恢复归一化丢失的绝对信号水平，去掉后术中低血压预测 AUROC 从 0.894 降至 0.780。
正文 · 原文
Abstract:Arterial, venous, pulmonary artery and intracranial pressures and ventilator waveforms guide decisions in operating rooms and intensive care units, yet foundation models for physiological waveforms have been built almost entirely on the electrocardiogram and photoplethysmogram. We present CARMEN, a single encoder pretrained without labels on ten cardiorespiratory waveforms (2.27 million modality-hours). CARMEN accepts any combination of co-recorded signals from 10 s to 5 min, and a conditioning mechanism, LSCNorm, restores the absolute signal level that normalization removes; without it, the AUROC for intraoperative hypotension fell from 0.894 to 0.780. Across 19 tasks from Korean and US cohorts, the frozen encoder outperformed supervised models on 12 of 15 and three foundation models, one general-purpose and two for biosignals, on 14 to 17 of 19, with none exceeding it on any task. Given the electrocardiogram or photoplethysmogram alone, it matched or exceeded signal-specific foundation models in most comparisons, and it retained its performance when channels were missing.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.15004 [cs.LG] |
| (or arXiv:2611.15004v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15004 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Cheol-Hui Lee [view email]
[v1]
Wed, 11 Nov 2026 04:02:50 UTC (10,396 KB)
来源：arXiv cs.AI · arxiv.org