基于策略模式的中医药数据智能挖掘平台设计与应用
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

R273

基金项目:

国家自然科学基金项目(81660727): 基于策略模式的中药性效数据挖掘方法研究;江西省教育厅科学技术研究重点项目(GJJ190635): 基于网络药理学的抗癌中药“四气五味”物质基础研究;


Design and application of traditional Chinese medicine data intelligent mining platform based on strategy mode
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对目前大多数中医药数据挖掘研究中使用单一且基础的算法而出现的问题,设计并实现一种通过策略模式智能优选中医药数据挖掘方法的平台(TCM data strategy model analysis platform,TCMDP)。根据策略模式的思想,集成以下4个数据挖掘模块,统计学分析模块可对药物、药物种类、四气五味归经和药物功效进行统计学分析;关联规则挖掘模块可以分析处方中的药物关联性;聚类分析模块可获取药物组合得出聚类新方,通过分析结果来探讨肺癌处方的配伍规律;证型分类模块以电子病历的中医症状和四诊信息作为输入,将相关证型作为输出,构建证型分类模型。综上实现了基于策略模式的中医药数据智能挖掘平台,并运用该平台对中医临床治疗肺癌的中药处方进行用药规律和证型分类研究。结果表明:以痰瘀互结证肺癌病例为例,关联规则挖掘模块中WD-Get Rule算法的运行时间最少仅为0.038秒。聚类分析模块中CMC-DD算法分析时间略长但精确率高达87%。肺癌证型分类分析模块中PSO-ELM运行时间短为88.98秒,且模型平均精确率达88.44%,具有一定的临床参考价值。而本平台所采用的改进算法均优于传统算法,且能智能优中选优,效率更高、结果更准确;可见运用策略模式的思想,构建并实现中医药数据智能挖掘平台是研究中医药信息的有效途径。

    Abstract:

    Aiming at the problem that most of the current traditional Chinese medicine data mining research uses a single and basic algorithm, a platform (TCM data strategy model analysis platform, TCMDP) that intelligently selects traditional Chinese medicine data mining methods through strategy mode is designed and implemented. According to the idea of the strategy mode, the following four data mining modules are integrated. The statistical analysis module can perform statistical analysis on drugs, drug properties, four natures of drugs , the five flavors, channel tropism and drug efficacy; the association rule mining module can analyze the drug correlation in prescriptions ;The cluster analysis module can obtain the combination of drugs to obtain new clustering prescriptions, and explore the compatibility law of lung cancer prescriptions through the analysis results; the syndrome type classification module takes the TCM symptoms and four diagnosis information of the electronic medical record as the input, and the related syndromes as the output , to build a syndrome classification model. To sum up, the intelligent mining platform of traditional Chinese medicine data based on the strategy model is realized, and the platform is used to study the medication rules and syndrome classification of traditional Chinese medicine prescriptions for the clinical treatment of lung cancer. The results show that: taking the lung cancer case of phlegm and blood stasis as an example, the running time of the WD-Get Rule algorithm in the association rule mining module is at least 0.038 seconds. The CMC-DD algorithm in the cluster analysis module has a slightly longer analysis time but the accuracy rate is as high as 87%. The PSO-ELM in the lung cancer syndrome classification analysis module has a short running time of 88.98 seconds, and the average accuracy rate of the model is 88.44%, which has a certain clinical reference value. The improved algorithms used in this platform are all better than traditional algorithms, and can intelligently select the best, with higher efficiency and more accurate results; it can be seen that the idea of using the strategy model is effective way to build and realize the intelligent mining platform of traditional Chinese medicine data is the study of traditional Chinese medicine information.

    参考文献
    相似文献
    引证文献
引用本文

章新友,徐华康,唐琍萍,等. 基于策略模式的中医药数据智能挖掘平台设计与应用[J]. 科学技术与工程, 2023, 23(14): 5946-5954.
Zhang Xinyou, Xu Huakang, Kang Liping, et al. Design and application of traditional Chinese medicine data intelligent mining platform based on strategy mode[J]. Science Technology and Engineering,2023,23(14):5946-5954.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2022-08-31
  • 最后修改日期:2023-03-07
  • 录用日期:2022-12-02
  • 在线发布日期: 2023-06-01
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注