基于数据挖掘分析叶天士《临证指南医案》治疗脑系疾病的用药规律

Data mining analysis of the medication pattern in the treatment of cerebral diseases in Ye Tianshi’s Clinical Guide of Medical Records

目的利用数据挖掘方法分析叶天士《临证指南医案》中脑系疾病的用药规律,为中医临床治疗脑系疾病提供参考。方法筛选《临证指南医案》中治疗脑系疾病的医案,对医案处方中的药物进行规范化处理后录入中医传承辅助平台(V2.5),并对录入处方中药物四气、五味、归经、使用频次、常用药对进行分析,采用复杂系统熵聚类算法生成新处方。结果共录入处方116首,含中药151味,其中性寒、味甘之药所占比例最大,归经以肝经为主,使用频次较高的中药有茯苓、生地黄、人参、枸杞子、菊花等;得到支持度≥8、置信度≥0.5的关联规则24组,设置相关度10、惩罚度4、关联系数≥0.023,基于改进的互信息法挖掘得到药对26对,得到11首新处方。结论叶天士治疗脑系疾病重视从肾论治、填精益髓,擅用温补和化痰类中药,挖掘得到的新处方具有一定的临床应用价值。

Objective To analyze the medication pattern of brain diseases in Ye Tianshi’s Clinical Guide to Medical Cases,and to provide reference for the clinical treatment of brain diseases in Chinese medicine.Methods The medical cases for the treatment of brain disorders in the Clinical Guide of Medical Records was screened,the drugs in the prescriptions of the cases were standardized and entered them into the Traditional Chinese Medicine Heritage Assist Platform(V2.5),and the four qi,five tastes,the attribution,the frequency of use,and the common pairs of drugs in the entered prescriptions were analyzed,and the complex system entropy clustering algorithm to generate new prescriptions were used.Results A total of 116 prescriptions containing 151 herbal medicines were recorded,among which the proportion of cold and sweet medicines was the largest,and the liver meridian was the main attribute,while the more frequently used herbal medicines were Poria,Radix et Rhizoma,Ginseng,Lycium,and Chrysanthemum.Twenty-four sets of association rules with support≥8 and confidence≥0.5 were obtained,with correlation degree was 10,penalty degree was 4 and association coefficient≥0.023,and 26 pairs of medicine pairs were obtained based on the improved mutual information method of mining,while 11 new prescriptions were obtained.Conclusion Ye Tianshi’s treatment of brain disorders emphasizes treatment from the kidney and filling the marrow,and is good at using warming and phlegm-transforming herbs,and the new prescriptions obtained are of certain clinical application value.