首页|基于知识元理论与知识图谱的中风病古籍医案研究路径探赜

基于知识元理论与知识图谱的中风病古籍医案研究路径探赜

扫码查看
目的 基于知识元理论与知识图谱揭示中风病古籍医案研究现状,并思考创新路径.方法 检索中国期刊全文数据库(CNKI)、维普数据库(VIP)、万方数据库(Wan Fang)、中国生物医学文献服务系统(SinoMed)等数据库,收集自建库至 2023 年10 月公开发表的中风病古籍医案研究文献,制定医案研究知识元体系,系统提取文献所涉及的知识元结构化数据,基于Neo4j图数据库构建知识图谱进行语义网络可视化,结合数据统计与图谱可视化综合分析研究脉络.结果 纳入相关文献 120 篇,形成医案知识元数据库,以此作为数据源构建包含301 个实体节点与1 100 条关系组成的知识图谱,剖析当前研究脉络为:医案以单一来源为主,多来源于单本古籍;医案研究多无特定分析角度;研究形式和研究方法受研究体量影响,小体量以个案研究结合研读领悟,大体量以群案研究结合知识发现;研究内容涉及13 种知识元类型、22 种语义类型和23 种语义关系,而当前主要内容仍局限在方剂知识元、中药语义与药对关系;知识发现技术采用以频数统计为主的数据挖掘.结论 根据研究现状,提出未来创新路径:增加医案来源纵横维度形成纵向时序或横向差异,应结合"知行合一"遴选医案分析焦点,开展基于个案解读式群案分析以提高研究复合性,构建领域知识图谱提高语义丰度与技术多样性.同时,本研究通过知识元理论与知识图谱的方法对接,实现语义数据分析与图谱可视化的"图-数互证",为中医重大疾病、优势病种的大数据源复杂成分分析提供方法参考.
Exploration of the Research Path of Stroke Ancient Medical Records Based on Knowledge Element Theory and Knowledge Graph
Objective Based on knowledge element theory and knowledge graph,the current status of research on ancient medical cases of stroke was revealed,and innovative paths were considered.Methods CNKI,VIP,Wan Fang,SinoMed,and other databases were retrieved,collected the publicly published research literature on ancient medical cases of stroke since the establishment of the library to October 2023,formulated the knowledge element system of medical case study,systematically extracted the knowledge elements of the text,and constructed the knowledge graph based on the Neo4j graph database to achieve the visual presentation of the complex network,and objectively revealed the research status and put forward the research optimization plan based on the data statistics and graph visualization analysis.Results 120 relevant documents were included to form a knowledge element database of medical cases,which was used as a data source to construct a knowledge graph consisting of 301 entity nodes and 1 100 relationships.The analysis of the current research context was as follows:medical cases are mainly from a single source,most of them come from single ancient books,and there is no specific point of view in medical case analysis.The research forms and methods are affected by the research volume,the small volume combines case study with reading comprehension,and the large volume combines group case study with knowledge discovery;the research content involves 13 types of knowledge elements,22 semantic types,and 23 semantic relations,but the current main content is still limited to prescription knowledge elements,Chinese medicine semantics and drug pair relations;knowledge discovery technology adopts data mining based on frequency statistics.Conclusion Based on the current status of the study,future innovation paths are proposed:increasing the vertical and horizontal dimensions of medical case sources to form a vertical time sequence or horizontal differences,selecting the focus of medical case analysis by combining"knowledge and action",conducting group case analysis based on individual case interpretation to improve the comprehensiveness of the study,and constructing a domain knowledge graph to improve the semantic abundance and technological diversity.At the same time,through the method of knowledge element theory and knowledge graph docking,this study realizes the semantic data analysis and graph visualization of"graph-data mutual evidence",which provides methodological references for the complex components analysis of big data sources for major diseases and dominant disease types in traditional Chinese medicine.

Knowledge element theoryStrokeMedical cases in ancient books of TCMKnowledge graphGraph visualization analysis

陈健、杨凤、任巧生、李颖、侯鉴宸、邢琛林、陶晓华、高颖、常静玲

展开 >

北京中医药大学东直门医院,北京 100007

中国中医科学院中医药信息研究所,北京 100700

北京中医药大学中医学院,北京 100029

北京邮电大学信息与通信工程学院,北京 100876

展开 >

知识元理论 中风病 中医古籍医案 知识图谱 可视化分析

国家重点研发计划国家重点研发计划国家自然科学基金北京中医药大学基本科研业务费项目(2022)

2019YFC17092002019YFC1709203819737902022-JYB-JBZR-032

2024

中国中医基础医学杂志
中国中医研究院基础理论研究所

中国中医基础医学杂志

CSTPCD
影响因子:0.779
ISSN:1006-3250
年,卷(期):2024.30(5)
  • 60