中医药领域人工智能应用研究文献可视化分析

Visualization Analysis of Research on Application of Artificial Intelligence in the Field of TCM

目的探寻中医药领域人工智能技术应用研究的热点及前沿,为相关研究提供思路。方法计算机检索中国知识资源总库(CNKI)、中文科技期刊数据库(VIP)、中国学术期刊数据库(万方数据)、中国生物医学文献数据库(CBM)建库至2022年5月17日收录的人工智能在中医药基础研究和临床实践应用相关文献,提取文献发表时间、作者、研究机构、来源期刊、关键词等信息,并对研究机构、来源期刊和关键词进行预处理,采用VOSviewer1.6.18软件构建关键词可视化网络并对纳入关键词进行聚类分析,以关键词出现的平均年度为依据构建关键词时间叠加网络。结果共纳入1852篇文献,来源于409种期刊,发文量前4位的期刊为《中华中医药杂志》《世界科学技术-中医药现代化》《中国中医药信息杂志》《中医杂志》;涉及2590名作者,来自北京中医药大学、上海中医药大学、中国中医科学院和成都中医药大学等272家单位,尚未发现区域性合作网络;纳入文献包含3361个关键词,共出现8074次,围绕数据挖掘、舌诊、专家系统共形成3个热点研究领域,知识图谱、卷积神经网络、注意力机制为近5年研究前沿。结论人工智能在中医药领域的应用研究随时间呈波动性增加,目前尚未形成稳定研究团队;将现代化技术与传统中医诊疗思维充分结合,更好地阐述传统中医思维并标准化表达,是该领域研究的重要方向和着力点。

Objective To explore the hotspots and frontiers of artificial intelligence(AI)technology in the field of TCM;To provide further ideas for related researches.Methods Literature about the application of AI in TCM basic research and clinical practice was retrieved from CNKI,Wanfang Data,CBM and VIP from the establishment to May 17,2022.The publication time,author,author unit,journal,keywords and other information were collected,and the author unit,journal and keywords were under preprocessed.VOSviewer 1.6.18 software was used to build a keyword visualization network and perform cluster analysis on the included keywords.Keyword time overlay network was constructed based on the average year of keyword appearance.Results A total of 1852 articles were finally included.The included articles were published in 409 journals,and the top four journals by the number of published papers were China Journal of Traditional Chinese Medicine and Pharmacy,Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology,Chinese Journal of Information on Traditional Chinese Medicine,and Traditional Chinese Medicine Journal;a total of 2590 authors of the 272 units from Beijing University of Chinese Medicine,Shanghai University of Traditional Chinese Medicine,China Academy of Chinese Medical Sciences and Chengdu University of Traditional Chinese Medicine,have not yet formed a mature regional cooperation network and team;the included articles covered 3361 keywords,with a total of 8074 occurrences,focusing on data mining,tongue diagnosis,and expert system,forming three hot research areas.Knowledge graph,convolutional neural network,attention mechanism have become the research frontier in the past five years.Conclusion The application research of AI in the field of TCM has fluctuated over time,and a relatively stable regional cooperation team has not yet formed;how to fully combine modern technology with traditional thinking of TCM diagnosis and treatment to better describe the thinking of TCM and give a standardized expression is an important direction and focus of AI application research in the field of TCM.