可配置的临床信息提取与代理式RAG技术：哪些方法有效，哪些会失败，以及原因是什么
Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why
摘要
患者相关信息包含数百份不同的文档以及数千个结构化数据点，但人工智能系统所需的文档级元数据却缺失或不完整。传统的基于检索的生成式方法无法处理这些数据，因为它们无法正确处理时间关系、不同文档之间的依赖关系以及缺失的元数据问题。我们在埃森大学医学中心采用了ACIE技术——这是一种本地化的代理式RAG系统，它能够对完整的患者信息进行全面分析，并将每个结果都基于原始文本进行验证，以便临床医生能够确认其准确性。我们量化了元数据缺失的程度，分析了相关架构决策，并与另一项独立的淋巴瘤登记研究进行了对比分析，在研究中，核医学医生会验证所提取的数据是否来自正确的来源。在7,326次判断中，临床医生接受了96.5%的提取结果，各类型数据的接受率则介于80%到99%之间。
English Abstract
Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retrieval and triage is absent or incomplete. Standard retrieval-augmented generation fails on this data, mishandling temporal reasoning, cross-document dependencies, and missing metadata. We deploy ACIE (Agentic Clinical Information Extraction) at University Medicine Essen: an on-premise agentic RAG pipeline that reasons over complete patient contexts and grounds every answer in source passages for clinician verification. We quantify the metadata gap, trace the architectural decisions it shaped, and evaluate extraction alongside an independent retrospective lymphoma registry study, in which nuclear-medicine physicians verify every extracted value against its cited sources. Across 7,326 judgments, clinicians accepted 96.5\% of extractions, with per-type acceptance ranging from 80\% to 99\%.