基于数据挖掘的针灸治疗卒中后认知障碍的选穴规律分析

Analysis of Acupoint Selection Rule of Acupuncture and Moxibustion for Post-Stroke Cognitive Impairment Based on Data Mining

摘要: 目的：运用数据挖掘技术对针灸治疗卒中后认知障碍(PSCI)的选穴规律进行探究和分析。方法：本文将查询中国期刊全文数据库(原名：CKNI)，访问万方数据知识服务平台(原称：Wanfang)，探索维普资讯中文期刊服务平台(旧称：VIP)，并查阅中国生物医学文献数据库(过去称为：SinoMed)、PubMed从建库至2024年12月15日所有关于针灸治疗卒中后认知障碍的相关文献，通过Excel软件建立针灸治疗卒中后认知障碍的处方数据库，我们利用SPSS Modeler 18.0版本中的Apriori算法，对研究中的穴位使用频率、所属经络、以及穴位处方内容进行了描述性统计与关联规则探索。接着，借助Cytoscape 3.9.1软件构建了复杂的网络图形，直观展示了高频穴位之间的关系。最后，通过SPSS 26.0软件对这些高频穴位进行了聚类分析，揭示了它们之间的相似性与差异性。结果：本次研究总共搜集并纳入了521篇相关文献，涵盖了143个不同的穴位，总频次3206次；腧穴使用频次排名前4位的是百会、神庭、四神聪、内关；所选经脉主要是督脉、胆经、胃经、脾经、膀胱经、心包经等；五输穴、交会穴、原穴、络穴依次是特定穴位中使用频次最高的；腧穴所处部位分析依次为头颈部占50.37%、下肢部占25.51%、上肢部占15.78%、背腰部占4.52%、胸腹部占3.81%；百会–风池–四神聪–神门四穴在腧穴配伍中，关联性最强；前20位高频腧穴可分为3个有效聚类群。结论：针对中风后认知功能受损，针灸疗法遵循补益根本、同时祛除病邪、畅通督脉以调理精神、安抚心神的原则进行治疗，以神庭–百会–四神聪为核心穴组。

Abstract: Objective: The rule of acupoint selection for the treatment of post-stroke cognitive impairment (PSCI) was investigated and analyzed using data mining technology. Method: This article will search the China National Knowledge Infrastructure (originally: CKNI), visit the Wanfang Data Knowledge Service Platform (originally: Wanfang), explore the VIP Chinese Journal Service Platform (old name: VIP) and consult the China Biomedical Literature Database (formerly: SinoMed), PubMed, and all relevant literature on the treatment of post-stroke cognitive impairment with acupuncture from the establishment of the database to December 15, 2024. Through Excel software, a database of acupuncture prescriptions for the of post-stroke cognitive impairment was established. We used the Apriori algorithm in SPSS Modeler 18.0 version to perform descriptive statistics and rule exploration on the frequency of acupoints used, the meridians to which they belong, and the content of the acupoint prescription in the study. Next, a complex network diagram was constructed using Cytoscape 3.9.1 software to intuitively show the relationship between high-frequency acupoint. Finally, these high-frequency acupoints were clustered using SPSS 26.0 software to reveal their similarities and differences. Results: A total of 521 related articles were collected and included in this study, covering 143 different acupoints with a total frequency of 3206 times; the top 4 acupoints in terms of frequency of use were Baihui, Shenting, Shishencong, and Neiguan; the selected meridians mainly included the Governor Vessel, Gallbladder Meridian, Stomach Meridian, Spleen Meridian, Badder Meridian, and Pericardium Meridian, etc.; the Five Shu points, Intersection points, Source points, and Collaterals points the highest in frequency of use in specific acupoints; the analysis of the parts where the acupoints were located was head and neck accounting for 50.37%, lower limbs accounting for 25.51%, upper limbs accounting for 15.78%, back and waist for 4.52%, and chest and abdomen accounting for 3.81%; the four acupoints Baihui-Fengchi-Sishencong-Shenmen had the strongest correlation in acupoint compatibility; the top 20 high-frequency acupoints could be divided 3 effective clustering groups. Conclusion: For post-stroke cognitive dysfunction, acupuncture therapy follows the principle of tonifying the root, dispelling pathogenic, dredging the Governor Vessel to regulate the spirit, and calming the mind to treat. The core acupoint group is Shenting-Baihu-Sishencong.