用自监督学习让血糖数据跨模态通用，提升糖尿病早期预警能力。
CGM-JEPA: Learning Consistent Continuous Glucose Monitor Representations via Predictive Self-Supervised Pretraining
- 通过预测隐藏表示而非原始值，实现多视角血糖数据的统一表征
- 在三类临床场景中均优于基线，最高提升6.5个百分点的识别准确率
- 特别擅长处理数据稀疏和人群差异问题，适合真实医疗场景部署
连续血糖监测（CGM）可发现早期代谢亚型（胰岛素抵抗、β细胞功能障碍），但大规模应用面临双重挑战：同一生理状态在不同视角（CGM时间序列、静脉葡萄糖耐量试验、葡萄糖密度图）下表现各异，导致单视角表征难以迁移；现有方法在视角或场景切换时性能不稳定。核心对策是构建脱离单一视角的高层时空与分布结构表征。本文提出CGM-JEPA，一种自监督预训练框架，通过预测被掩码的隐向量实现抽象表征，支持跨模态迁移。X-CGM-JEPA进一步引入掩码葡萄糖密度图的跨视图目标，补充分布信息。在约38.9万条未标注的CGM读数（来自228名受试者）上预训练，并在两个公开数据集（N=27和N=17）上，通过20轮×2折交叉验证，在三种范式（队列泛化、静脉到CGM迁移、家庭CGM）下评估。X-CGM-JEPA在所有范式中均排名第一或第二，显著超越最强基线：队列泛化中最高提升6.5个百分点，静脉到CGM迁移中提升3.6个百分点（配对Wilcoxon检验，p<0.001）。在模态切换下，其平均AUROC保持稳定，同时改善弱势群体表现（种族间AUROC差距缩小25%-54%）；在稀疏的域内静脉数据上，分布视图使标签感知聚类性能提升（ARI +39%，NMI +40%）。代码与权重见：https://github.com/cruiseresearchgroup/CGM-JEPA
原文摘要 · Abstract (English)
Continuous Glucose Monitoring (CGM) can detect early metabolic subphenotypes (insulin resistance, IR; $β$-cell dysfunction), but population-scale deployment faces two coupled problems. First, the same physiological state appears through multiple views (CGM time series, venous OGTT, Glucodensity summaries), so single-view representations fail to transfer when deployment shifts the modality or setting. Second, baselines perform inconsistently across these shifts. Both problems point to one remedy: representations that abstract away from any single view to capture higher-level temporal and distributional structure. We propose CGM-JEPA, a self-supervised pretraining framework which predicts masked latent representations rather than raw values, yielding abstraction that transfers across modalities. X-CGM-JEPA adds a masked Glucodensity cross-view objective for complementary distributional information. We pretrain on $\sim$389k unlabeled CGM readings from 228 subjects and evaluate on two clinical cohorts ($N=27$ and $N=17$ public-release subsets) across three regimes (cohort generalization, venous-to-CGM transfer, home CGM) under 20-iteration $\times$ 2-fold cross-validation. X-CGM-JEPA ranks first or second on AUROC for both endpoints across all three regimes while no baseline does, exceeding the strongest baseline by up to $+6.5$ pp in cohort generalization and $+3.6$ pp in venous-to-CGM transfer (paired Wilcoxon, $p<0.001$). Under modality shift, it matches mean AUROC while redistributing toward weaker subgroups (ethnicity AUROC gap shrinks 25-54%); on sparse in-domain venous data, the distributional view lifts label-aware clustering (ARI $+39\%$, NMI $+40\%$). Code and weights: https://github.com/cruiseresearchgroup/CGM-JEPA
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。