arXiv 2605.264632026-05-26cs.CLcs.AI

迈向无差错的电子健康记录：临床笔记与结构化表格之间的推理密集型一致性验证

Towards Error-Free EHRs: Reasoning-Intensive Consistency Verification Between Clinical Notes and Structured Tables in Electronic Health Records

KAIST（韩国科学技术院）；Ghent University（根特大学）；Samsung Medical Center（三星医疗中心）；Samsung Changwon Hospital（三星昌原医院）；Asan Medical Center（亚山医疗中心）

Yeonsu Kwon, Jiho Kim, Junseong Choi, Paloma Rabaey, Minseo Kim, Sujeong Im, Jeewon Yang, Jun-Min Lee, Sangji Lee, Jiwon Kim, Hangyul Yoon, Hyunwook Kwon, Edward Choi

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对电子健康记录中临床笔记与结构化表格数据不一致的问题，提出推理密集型基准EHR-ReasonCon和基于大语言模型的框架EHR-Inspector，通过锚点实体提取、时间引用和表格探索工具实现高效一致性验证。

摘要：

电子健康记录中非结构化临床笔记与结构化表格之间的数据一致性对于患者安全和临床决策至关重要。然而，现有关于笔记-表格一致性验证的工作主要依赖于数值或简单事件的表面匹配。这些方法未能捕捉真实世界EHR文档背后的推理，包括临床解释、事件关系和时间变化。为弥补这一差距，我们引入了EHR-ReasonCon，一个用于笔记-表格一致性验证的推理密集型基准。它基于MIMIC-III构建，并经过专家指导的注释，包含来自临床笔记的8,048个实体，并提供高质量的真实标签。注释协议由专门的表格探索工具支持，以确保系统的证据检索和可靠的一致性评估。我们还提出了EHR-Inspector，一个基于LLM的框架，它分割笔记、提取锚点实体和时间引用，并使用表格探索工具与结构化表格进行一致性验证。在严格和宽松标准下，使用经过专家验证的LLM-as-a-judge指标进行评估，EHR-Inspector在多个模型骨干上实现了最先进的性能。进一步的分析证明了其组件的有效性，并突出了与人工验证的差异。

英文摘要：

Data consistency between unstructured clinical notes and structured tables in Electronic Health Records (EHRs) is essential for patient safety and clinical decision-making. However, existing work on note-table consistency verification mainly relies on surface-level matching of numeric values or simple events. Such approaches fail to capture the reasoning underlying real-world EHR documentation, including clinical interpretation, event relations, and temporal changes. To address this gap, we introduce EHR-ReasonCon, a reasoning-intensive benchmark for note-table consistency verification. Built on MIMIC-III with expert-guided annotations, it comprises 8,048 entities derived from clinical notes and provides high-quality ground-truth labels. The annotation protocol is supported by specialized table-exploration tools to ensure systematic evidence retrieval and reliable consistency assessment. We also propose EHR-Inspector, an LLM-based framework that segments notes, extracts anchor entities and temporal references, and uses table-exploration tools to verify consistency against structured tables. Evaluated using expert-validated LLM-as-a-judge metrics under harsh and lenient criteria, EHR-Inspector achieves state-of-the-art performance across multiple model backbones. Analyses further demonstrate the effectiveness of its components and highlight differences from human verification.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。