基于数据挖掘探讨中药治疗糖尿病的组方规律

Exploring the prescription rules of traditional Chinese medicine in the treatment of diabetes mellitus based on data mining

目的基于数据挖掘探讨中药治疗糖尿病的用药规律,以期为临床提供相应参考。方法在药智网中检索收集治疗糖尿病的中药方剂,运用频数统计、关联规则、因子分析等方法探讨中药治疗糖尿病的组方规律。结果最终纳入344首方剂,涉及190味中药,共有3371次配伍联系以及13味高频中药。中药属性分析发现治疗糖尿病的中药多属补虚类、清热类、收涩类,且以性寒、味甘以及入脾经、肾经、胃经、肺经等为主。关联规则分析结果得到36项中药组合;因子分析提取到5个公因子。结论本研究基于数据挖掘发现了糖尿病治疗的高频使用中药、中药属性关联及常用配伍,可为临床治疗糖尿病提供参考。

Objective To explore the prescription rules of traditional Chinese medicine in the treatment of diabetes mellitus based on data mining,in order to provide relevant references for the clinical treatment.Methods The traditional Chinese medicine prescriptions in the treatment of diabetes mellitus were retrieved in Yaozhi network,and the prescription rules of traditional Chinese medicine were summarised by using frequency statistics,associated rules,and factor analysis.Results A total of 344 traditioanl Chinese medicine formula were included,involving 190 traditional Chinese medicine with 3371 compatibility and 13 high-frequency traditional Chinese medicine.The attribute analysis of traditional Chinese medicine found that these traditional Chinese medicine treating diabetes mellitus mostly belonged to the categories of tonifying,heat-clearing,and astringency,which were mainly cold,sweet,and entered spleen,kidney meridians,stomach meridians,and lung meridians.The association rules finally obtained 36 traditional Chinese medicine combinations,while factor analysis extracted 5 common factors.Conclusion Based on data mining,this study found the high-frequency traditional Chinese medicine,the attribute association and common compatibility in the treatment of diabetes mellitus,which can provide reference for clinical treatment of diabetes.