基于数据挖掘探究国家专利复方治疗骨结核的用药规律  

To Explore the Medication Rule of National Patent Compound Treatment for Bone Tuberculosis Based on Data Mining

作　　者：邓杨 李学清 乐思微 刘锦涛 王凯[1] 刘汝专[2] 

机构地区：[1]广西中医药大学研究生院,广西 南宁 [2]广西中医药大学附属瑞康医院脊柱外科,广西 南宁

出　　处：《中医学》2024年第9期2378-2386,共9页Traditional Chinese Medicine

摘　　要：目的:基于R语言平台分析国家专利复方治疗骨结核的用药规律,为骨结核的中医药治疗提供参考依据。方法:检索“国家知识产权局–中国专利公布公告”网站中治疗骨结核的复方专利数据,采用Excel软件提取纳入文献信息并进行四气五味归经统计,利用R语言进行频数统计、关联规则分析、聚类分析。结果:共纳入70首专利复方,共包含药物180味,用药频数大于10的药物分别是当归、甘草、全蝎、蜈蚣、黄芪、鳖甲、没药、乳香、穿山甲、白芨。药物药性以温、寒、平性为主,药味以甘、辛、苦味为主;归经以肝、脾、肾经为主。关联规则分析显示获得以当归、甘草、全蝎、蜈蚣、熟地、乳香、没药为核心的药物组合;层次聚类分析得到4类药物组合。结论:分析强直性脊柱炎临床用药的常用组合规律,为临床治疗强直性脊柱炎提供用药思路。Objective: To analyze the medication rule of national patent compound treatment based on R language platform and provide a reference for TCM treatment of bone tuberculosis. Methods: We searched the compound patent data for the treatment of bone tuberculosis in the website of “State Intellectual Property Office-China Patent Announcement”, extracted the included literature information and collected the four gas and five taste statistics, and used frequency statistics, association rule analysis and cluster analysis in R language. Results: A total of 70 patented compounds were included, including 180 drugs with drug frequency greater than 10 were angelica, licorice, scorpion, centipede, Astragalus, turtle nail, myrrh, frankincense, pangolin and Bletilla striata. Warm, cold, plain and sweet, spicy and bitter;mainly liver, spleen and kidney. The association rule analysis showed that the drug combinations with angelica, licorice, whole scorpion, centipede, ripening, frankincense and myrrh were obtained;the hierarchical clustering analysis obtained four types of drug c

关 键 词：数据挖掘 骨结核 用药规律 

分 类 号：R28[医药卫生—中药学]