arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.01844cs.AI
收藏

基于多语言模型的临床知识图谱构建与评估:通过检索增强生成

Clinical Knowledge Graph Construction and Evaluation with Multi-LLMs via Retrieval-Augmented Generation

University of Missouri–Kansas City(密苏里大学-堪萨斯城分校);California State University, Fullerton(加州州立大学弗雷斯诺分校)

Udiptaman Das, Krishnasai B. Atmakuri, Duy Ho, Chi Lee, Yugyung Lee

AI总结:

本文提出基于多语言模型的临床知识图谱构建与评估框架,通过检索增强生成策略实现从自由文本到可解释、临床相关知识图谱的端到端构建与持续优化。

摘要:

大型语言模型(LLMs)为从非结构化临床叙述中构建知识图谱(KGs)提供了新的机会。然而,现有方法往往依赖于结构化输入,并缺乏对事实准确性和语义一致性的稳健验证,这些局限性在肿瘤学中尤为突出。我们介绍了一种端到端的框架,该框架通过多代理提示和一种约束模式的检索增强生成(KG-RAG)策略,直接从自由文本中构建和评估临床知识图谱。我们的流程整合了(1)提示驱动的实体、属性和关系提取;(2)基于熵的不确定性评分;(3)与本体对齐的RDF/OWL模式生成;以及(4)多LLM共识验证用于幻觉检测和语义优化。除了静态图谱构建外,该框架支持持续优化和自监督评估,使图谱质量能够不断改进。应用于两个肿瘤学队列(PDAC和BRCA)后,我们的方法生成了可解释、兼容SPARQL且具有临床依据的知识图谱,无需依赖黄金标准注释。实验结果表明,与基线方法相比,在精确度、相关性和本体兼容性方面均取得了持续提升。

英文摘要:

Large language models (LLMs) offer new opportunities for constructing knowledge graphs (KGs) from unstructured clinical narratives. However, existing approaches often rely on structured inputs and lack robust validation of factual accuracy and semantic consistency, limitations that are especially problematic in oncology. We introduce an end-to-end framework for clinical KG construction and evaluation directly from free text using multi-agent prompting and a schema-constrained Retrieval-Augmented Generation (KG-RAG) strategy. Our pipeline integrates (1) prompt-driven entity, attribute, and relation extraction; (2) entropy-based uncertainty scoring; (3) ontology-aligned RDF/OWL schema generation; and (4) multi-LLM consensus validation for hallucination detection and semantic refinement. Beyond static graph construction, the framework supports continuous refinement and self-supervised evaluation, enabling iterative improvement of graph quality. Applied to two oncology cohorts (PDAC and BRCA), our method produces interpretable, SPARQL-compatible, and clinically grounded knowledge graphs without relying on gold-standard annotations. Experimental results demonstrate consistent gains in precision, relevance, and ontology compliance over baseline methods.

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。

↑