目的 探究近10年药物相互作用（DDI）领域预测模型的研究现状和热点。方法 检索2013年1月1日—2024年8月13日Web of Science核心合集数据库中发表的有关DDI预测模型研究文献，运用CiteSpace 6.3 R1软件绘制作者、研究机构、国家地区合作网络以及期刊、关键词可视化知识图谱，进行文献计量分析。结果 2013—2024年DDI预测模型研究相关文献754篇，发文量呈逐年上升趋势；美国发文量最多，中国其次；发文量最多的学者是Sugiyama，Yuichi；发文量第1的研究机构是RLUK-Research Libraries UK；热点关键词为抑制、代谢、药代动力学、清除率；发文量最多的期刊及被引频次最高的期刊是Bioinformatics；在高频被引文献中显示了对机器学习、深度学习的关注度较高。结论 分析2013—2024年DDI预测模型文献，可知此领域研究热点为药动学和体外实验研究，研究前沿为机器学习和深度学习，为未来研究提供新的视角。

Objective To explore the research status and hotspots of drug-drug interactions (DDI) prediction model in the last 10 years. Methods The relevant English literature on DDI prediction model published from 2013 to 2024 was retrieved from Web of Science (WOS) database. CiteSpace software was used to visual analysis. Results A total of 754 pieces of paper related to DDI prediction model were obtained, and the number of publications showed an increasing trend. The country with the most articles was the United States, followed by China. The person who published the most articles was Dr. Yuichi Sugiyama. The research institution with the largest number of publications was RLUK-Research Libraries UK. Hot keywords were inhibition, metabolism, pharmacokinetics, clearance. The journal with the highest number of publications and the most frequently cited journal was Bioinformatics. Machine learning and deep learning are gotten high attention by the high frequency cited literature. Conclusion CiteSpace software was used to analyze the literature on DDI prediction models, and it was found that in vitro experimental studies, pharmacokinetics, machine learning and deep learning are the research frontiers in this field, providing a new perspective for the study of DDI prediction models.

R969.2

中国药品监督管理研究会研究课题-基于多元证据体探索中成药安全性评价方法的研究(2024-Y-Y-006)；临床研究和成果转化能力提升试点项目-中药制剂研发——治疗胃轻瘫中药复方佛香散(DZMG-ZJXY-23002)