基于PICO模型的中医药循证指南知识图谱构建与智能问答系统研究  

Construction of a knowledge graph and intelligent Q&A system for evidence-based TCM guidelines based on the PICO model

作　　者：崔唐明 孙美玲 孙华君 李海燕[1] CUI Tangming;SUN Meiling;SUN Huajun;LI Haiyan(Institute of Traditional Chinese Medicine Information,China Academy of Traditional Chinese Medicine,Beijing 100700,China)

机构地区：[1]中国中医科学院中医药信息研究所,北京100700

出　　处：《中国数字医学》2024年第5期20-27,共8页China Digital Medicine

基　　金：中国中医科学院科技创新工程创新团队项目-基于语义模型的中医循证临床指南知识图谱构建研究(JY202302-2017YFC1703505)。

摘　　要：目的:基于PICO模型构建中医药循证指南知识图谱,探索以此知识图谱为知识库搭建智能问答系统,增强临床决策支持。方法:采用自上而下与自下而上相结合的方法设计知识图谱的概念层和数据层,并以多囊卵巢综合征为例,利用“文心一言”大模型工具进行知识抽取,抽取结果通过Neo4j图数据库进行存储,实现知识图谱的可视化展示,同时对智能问答系统进行临床专家问卷调研和框架结构设计。结果:构建了包含13种类别、248个节点、12种关系类型、337条关系以及10种数据属性的中医药循证指南知识图谱,并展示了其可视化功能和应用示例。智能问答系统部分,已根据指南临床问题范围将其分为12大类,并预设了24种问题模板,为后续问题预处理奠定了基础,并对系统实现提供了有力支撑。结论:本研究以循证指南为数据源确保了临床辅助决策系统科学性和准确性,也为中医药领域的智能化信息服务提供了新的思路和方法。Objective To construct a knowledge graph of evidence-based guidelines of Chinese medicine based on PICO model,and to explore the use of this knowledge graph as a knowledge base to build an intelligent Q&A system to enhance clinical decision support.Methods The concept layer and data layer of the knowledge graph was designed by integrating top-down and bottom-up approaches.Taking polycystic ovary syndrome as an example,the“Wenxinyiyan”large model tool was used to extract knowledge,and the extraction results were stored in the Neo4j graph database,which realized the visual display of the knowledge graph.At the same time,clinical expert questionnaire survey and framework structure design of intelligent Q&A system were carried out.Results The knowledge graph of evidencebased TCM guidelines containing 13 categories,248 nodes,12 relationship types,337 relationships and 10 data attributes was constructed,and its visualization function and application examples have been demonstrated.In the part of the intelligent Q&A system,the guidelines have been divided into 12 major categories based on their clinical question ranges,and 24 kinds of question templates have been preset,which has laid the foundation for the subsequent question preprocessing and provided a strong support for the system implementation.Conclusion This study ensures the scientific and accurate clinical assistant decision-making system by using the evidence-based guidelines as the data source,and also provides new ideas and methods for intelligent information service in the field of traditional Chinese medicine.

关 键 词：PICO 循证指南 知识图谱 智能问答系统 大模型 

分 类 号：R197.32[医药卫生—卫生事业管理] R319[医药卫生—公共卫生与预防医学]