端到端紧急护理的统一多模式基础模型
A unified multi-modal foundation model for end-to-end emergency care.
摘要(中文译文)
在中国人口老龄化与慢性病负担不断上升的背景下，三级医院急诊科（ED）持续面临过度拥挤的问题。急诊科过度拥挤是一种多因素现象，源于急诊诊疗工作流程中输入、通过和输出因素的相互作用。现有用于急诊诊疗的人工智能方法大多针对特定任务。尽管这些方法在孤立的临床任务中有效，但缺乏以患者为中心的端到端整合，且在异质性和不完整输入条件下泛化能力较差，限制了其在资源异质的真实世界环境中的可扩展部署。在此，我们提出ED-Foundation，一种基于BEIT-3架构构建的统一急诊基础模型。通过联合利用对齐的图像-文本对和未对齐的纯文本数据，ED-Foundation在单一模型内学习连续的以患者为中心的表征，而两阶段自监督学习框架进一步提高了对普遍存在的缺失数据的鲁棒性。我们在九个下游验证数据集上评估ED-Foundation，这些数据集涵盖三个核心急诊场景，分别针对急诊诊疗工作流程中不同环节的瓶颈问题：急诊早期分诊、临床病程中的结局预测以及急诊治疗阶段的临床决策支持。ED-Foundation始终取得最先进的回顾性性能，并在信息有限和模态缺失条件下保持鲁棒性，在不同机构和跨系统评估设置中优于既往任务特定方法和现有基础模型。这些结果为ED-Foundation作为急诊诊疗任务统一表征学习框架的基准性能和表征可迁移性提供了回顾性证据。
Abstract
Against the backdrop of population aging and the rising burden of chronic diseases in China, emergency departments (EDs) in tertiary hospitals remain persistently overcrowded. ED overcrowding is a multifactorial phenomenon arising from the interplay of input, throughput, and output factors along the emergency care workflow. Existing artificial intelligence approaches for emergency care are predominantly task-specific. Although effective for isolated clinical tasks, they lack patient-centric, end-to-end integration and generalize poorly under heterogeneous and incomplete inputs, limiting scalable deployment in resource-heterogeneous real-world settings. Here we propose ED-Foundation, a unified emergency foundation model built on the BEIT-3 architecture. By jointly leveraging aligned image-text pairs and unaligned pure-text data, ED-Foundation learns continuous patient-centric representations within a single model, and a two-stage self-supervised learning framework further improves robustness to pervasive missingness. We evaluate ED-Foundation on nine downstream validation datasets spanning three core emergency scenarios that address bottlenecks at different points along the emergency care workflow: early emergency triage, outcome prediction during the clinical course, and clinical decision-making support in the emergency treatment phase. ED-Foundation consistently achieves state-of-the-art retrospective performance and maintains robustness under limited information and modality missingness, outperforming prior task-specific approaches and existing foundation models across diverse institutional and cross-system evaluation settings. These results provide retrospective evidence supporting the benchmark performance and representational transferability of ED-Foundation as a unified representation learning framework for emergency care tasks.
如何引用
AMA
Zhenwei Huang, Yunfeng Xu, Zixuan Nie, Huiyu Wang, Xiaodong Huang, Ling Huang, et al. A unified multi-modal foundation model for end-to-end emergency care.. NPJ digital medicine. 2026; doi:10.1038/s41746-026-02981-z.
APA
Zhenwei Huang, Yunfeng Xu, Zixuan Nie, Huiyu Wang, Xiaodong Huang, Ling Huang, et al (2026). A unified multi-modal foundation model for end-to-end emergency care.. NPJ digital medicine. https://doi.org/10.1038/s41746-026-02981-z
GB/T 7714
Zhenwei Huang, Yunfeng Xu, Zixuan Nie, Huiyu Wang, Xiaodong Huang, Ling Huang, et al. A unified multi-modal foundation model for end-to-end emergency care.[J]. NPJ digital medicine, 2026 doi:10.1038/s41746-026-02981-z.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出