维吾尔医治疗冠心病方剂的“药-效-量”配伍理论及新方数据挖掘研究

Data mining to Study the “drug-effect-dose” Compatibility Theory and New Prescriptions for the Treatment of Coronary Heart Disease from Uyghur Medical Prescriptions

目的:采用关联规则和聚类分析方法,对治疗冠心病维吾尔医方剂的组方配伍理论进行数据挖掘。方法:以古方为线索,借助于现代统计软件,建立冠心病方剂的Microsoft Excel数据库,应用关联规则的Apriori算法和样本聚类方法,分析处方中高频药组的关联规则及其剂量关系,挖掘核心药物组合和候选新方剂。结果:符合标准的127首方剂进行统计,药物之间具有强关联规则(支持度≥17%、置信度≥75%)的药物组合21组,挖掘出10个核心药物组合,得到6首候选新方剂。讨论:挖掘的核心组合和候选方剂所谓"木棑日亚提"的加减方,从结构可推断以强心为主药,配伍强筋、补脑、补肾以及相应成熟药的基本组方及其剂量配伍理论,继而为新药研发提供新方剂线索。

Objective： Conduct data mining the prescription compatibility theory of the treatment of coronary heart disease（ CHD ） by using association rules and cluster analysis methods. Methods： Ancient prescriptions as clue for establishing the Microsoft Excel database with the help of the modem computer technology, Analysis association rules and dose relationship of high frequency drugs group in prescriptions, Mining core drug combinations and candidate new prescriptions. Results ： Select 21 strong association rules （ support ≥ 17 % and confidence≥ 75% ） drug group, 10 core combinations and 6 new candidate prescriptions were obtained by statistics meet the standard of 127 prescriptions. Discussion ： Excaiated core combinations and candidate prescriptions are adds and subtracts of ＂mupariyat＂, From the structure of prescriptions deduce the compatibility theory that cardiotonic drug is the main group in prescriptions, compatibility of tonifying brain, kidney, strong tendon drugs and corresponding Munziq of Abnormal body fluids in the treated- prescriptions of CHD, after that provides new prescriptions and its dose compatibility clues for new drug research and development.