基于数据挖掘探讨中医古籍中治疗云雾移睛的用药规律  

Research on Medication Patterns for Treating Vitreous Opacity(Moving Clouds and Mists)in Ancient Chinese Medical Books Based on Data Mining

作　　者：刘新宇 陈星宇 郭惠怡 莫亚欣 陈强[1] LIU Xinyu;CHEN Xingyu;GUO Huiyi;MO Yaxin;CHEN Qiang(Eye Hospital,China Academy of Chinese Medical Sciences,Beijing 100040,China)

机构地区：[1]中国中医科学院眼科医院,北京100040 [2]中国中医科学院中医基础理论研究所,北京100700

出　　处：《中国中医眼科杂志》2024年第6期507-512,共6页China Journal of Chinese Ophthalmology

基　　金：国家自然科学基金项目(81503621);中国中医科学院眼科医院横向课题(101120351);中国中医科学院眼科医院高水平中医医院课题(GSP5-32)。

摘　　要：目的利用数据挖掘技术分析中医古籍中治疗云雾移睛的用药规律。方法查阅《中医名词考证与规范》中有关云雾移睛的病名症状记载,搜集整理中医古籍中有关治疗云雾移睛的方剂,将方剂相关信息录入Excel软件,利用SPSS Modeler 18.0、Cytoscape 3.9.1、SPSS 25.0软件对16部中医古籍中治疗云雾移睛的方剂进行多维度分析,包括用药频次、性味归经、高频药物功效、关联规则、聚类分析及复杂网络分析。结果(1)一般情况:共纳入方剂68首,涉及153味中药,累计频次为709次,其中使用频次≥10次的中药共计27味。(2)性味归经:四气总频次153次,寒性及温性药物用药频次最高(皆为37次,24.18%);药味总频次247次,甘味药物用药频次最高(70次,28.34%);药物归经总频次374次,归肝经的药物用药频次最高(78次,20.86%)。(3)高频药物分析:使用频次排在前5位的药物分别为菊花(24次,3.39%)、防风(20次,2.82%)、羌活(19次,2.68%)、车前子(17次,2.40%)、茯苓(17次,2.40%)。(4)关联规则分析:置信度排名前5位的强关联对药为茯苓-泽泻、肉苁蓉-菟丝子、茯苓-山药、蒺藜-木贼、菊花-木贼;置信度排名前5位的强关联角药分别为防风-细辛-决明子、防风-车前子-决明子、细辛-玄参-决明子、玄参-细辛-决明子。(5)聚类分析:共得到7组核心药物,分别为补肝肾药、补气血药、引经-平肝熄风药、引经-清热药、引经-利水渗湿药、清热-平肝熄风-补气药、清热药。(6)复杂网络分析:核心的药物为防风、菊花、茯苓、甘草、川芎、蒺藜,其使用频率高且与其他药物关联性强。结论中医古籍中治疗云雾移睛的核心药物以肉苁蓉、菟丝子等补益药为主,以肝肾亏损为主要病机,治法以补益肝肾、补气养血兼辛散引经为主,同时结合辨证论治,辅以清热和利水渗湿之法,用药灵活。OBJECTIVE To analyze the medication patterns for treating vitreous opacity(moving clouds and mists)in ancient Chinese medical books using data mining techniques.METHODS Relevant records of vitreous opacity in ancient Chinese medical books were reviewed in Certification and Standardization of Chinese Medical Terms.Formulas related to treating vitreous opacity were collected and organized from ancient Chinese medical books.The relevant information of these formulas was entered into Excel software.SPSS Modeler 18.0,Cytoscape 3.9.1,and SPSS 25.0 software were utilized for multidimensional analysis of 68 formulas from 16 ancient Chinese medical books for treating vitreous opacity.The analysis included medication frequency,properties and flavors,high-frequency medicinal effects,association rules,cluster analysis,and complex network analysis.RESULTS(1)General situation:A total of 68 formulas were included,involving 153 Chinese medicines,with a total frequency of 709 occurrences,including 27 medicines with a frequency of≥10 times.(2)Properties and flavors:The total frequency of medicinal properties was 153 times,with cold and warm properties being the most frequently used(both 37 times,24.18%);The total frequency of flavors was 247 times,with sweet flavors being the most frequently used(70 times,28.34%);The total frequency of medicinal channels was 374 times,with medicines that entered liver channel being the most frequently used(78 times,20.86%).(3)High-frequency medicinal analysis:The top five most frequently used medicines were Chrysanthemum(24 times,3.39%),Saposhnikovia divaricata(20 times,2.82%),Rhizoma seu Radix Notopterygii(19 times,2.68%),Plantago asiatica(17 times,2.40%),and Poria cocos(17 times,2.40%).(4)Association rule analysis:The top five strong association pairs of medicines based on confidence level were Poria cocos-Alisma orientale,Cistanche deserticola-Cuscuta chinensis,Poria cocos-Dioscorea opposita,Tribulus terrestris-Desmodium styracifolium,Chrysanthemum-Desmodium styracifolium;The top five strong

关 键 词：数据挖掘 中医古籍 云雾移睛 用药规律 玻璃体混浊 

分 类 号：R276.7[医药卫生—中医五官科学]