改进电子病历预训练，让每位患者贡献更均衡。
Pretraining EHR Foundation Models with Patient-Aware Sampling
- 按患者分组采样，控制训练信号分配
- 在MIMIC-IV上宏平均AUC提升显著
- 适合医疗数据建模与公平性研究者
针对电子病历（EHR）的自回归基础模型，传统预训练方法将患者轨迹拼接为单一词元流并随机采样窗口，导致不同患者贡献不均，长记录患者占据更多优化更新，可能引入偏差。本文提出患者感知采样（Patient Sampling），通过可控权重对患者进行分组采样，以调节训练信号分布。相比标准全局流（Global Stream）方法，在MIMIC-IV v2.2和v3.1的多个下游临床任务中，该方法显著提升宏平均AUROC与AUPRC。结果表明，序列构建方式是自回归EHR基础模型的重要且未被充分探索的设计选择。
原文摘要 · Abstract (English)
Autoregressive foundation models for electronic health records (EHRs) typically inherit pretraining methods from language modeling, where patient trajectories are concatenated into a single token stream and windows are sampled from that stream. In EHR data, this choice is consequential: windows may mix multiple patients, and patients with longer records contribute more optimization updates, potentially introducing bias. We propose Patient Sampling, a pretraining sequence-construction method that allows us to control how training signal is distributed across patients. We compare this method to the standard approach, which we refer to as Global Stream. We show that stochastic Patient Sampling with controllable weighting improves performance on real-world EHR data. Across downstream clinical tasks on MIMIC-IV v2.2 and v3.1, Patient Sampling improves Macro AUROC and AUPRC over the Global Stream baseline. These results identify training and validation sequence construction as important and underexplored design choices for autoregressive EHR foundation models.
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。