基于症状间关系及其对证候的贡献度初步建立新冠肺炎常见证候诊断依据  

Preliminarily Establishment of Common Syndromes Diagnosis Basis of Coronavirus Disease 2019 Based on Correlation between Symptoms and Their Contribution to Syndrome

作　　者：春柳 冯贞贞[1] 李建生[1,2] 谢洋[2] 赵虎雷 韩伟红 王晓春 Chun Liu;Feng Zhenzhen;Li Jiansheng;Xie Yang;Zhao Hulei;Han Weihong;Wang Xiaochun(Collaborative Innovation Center for Chinese Medicine and Respiratory Diseases Co-constructed by Henan Province&Education Ministry of P.R.China/Henan Key Laboratory of Chinese Medicine for Respiratory Disease,Henan University of Chinese Medicine,Zhengzhou 450046,China;Respiratory Department,The First Affiliated Hospital of Henan University of Chinese Medicine,Zhengzhou 450000,China)

机构地区：[1]河南中医药大学呼吸疾病中医药防治省部共建协同创新中心/河南省中医药防治呼吸病重点实验室,郑州450046 [2]河南中医药大学第一附属医院,郑州450000

出　　处：《世界科学技术-中医药现代化》2021年第3期874-882,共9页Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology

基　　金：河南省科学技术厅2020年新型冠状病毒防控应急攻关项目(201100310400):新型冠状病毒感染的肺炎的临床特征与证候规律研究,负责人:李建生;河南省科学技术厅2020年新型冠状病毒防控应急攻关项目(201100310500):新型冠状病毒感染的肺炎中西医结合治疗方案优化及评价,负责人:李素云;河南省科学技术厅中原学者科学家工作室项目(豫财行[2018]204号),负责人:李建生。

摘　　要：目的初步建立新型冠状病毒肺炎(简称"新冠肺炎")常见证候诊断依据。方法基于采集的654份河南省新冠肺炎临床资料,联合运用关联规则与贝叶斯网络分析常见证候中关联较强的症状群,并进一步分析常见症状群对该证候诊断的贡献度。结果通过关联规则(支持度>10%、置信度>80%及提升度>1的二项关联症状组合)和贝叶斯网络(存在直接因果关系且条件概率≥0.45的症状组合)分别得出新冠肺炎9个常见证候中关联密切的症状群;通过贝叶斯公式推导出常见症状群对证候的贡献度;初步建立了新冠肺炎常见证候诊断依据。结论根据常见证候症状间关系建立相关症状群及其对证候诊断的贡献,初步建立新冠肺炎常见证候诊断依据,可为中医证候诊断标准的建立提供新的思路与方法。Objective To preliminarily establish the diagnostic basis for common syndromes of coronavirus disease2019(COVID-19).Methods The correlation between symptoms and their contribution to common syndromes was analyzed using association rules combined with a Bayesian network based on the clinical data of 654 COVID-19 cases in Henan province.Results The correlated symptom clusters in 9 common syndromes of COVID-19 were summarized using association rules and Bayesian network;the binomial association combinations with support over 10%,confidence over 80%,and lift over 1 were extracted;the symptom combinations in the Bayesian network that had direct causality and conditional probabilities over 0.45 were extracted.Moreover,the contribution degree of common symptoms to syndromes was deducted by the Bayesian formula.The diagnostic basis of common syndromes was preliminarily established.Conclusion We preliminarily established the diagnostic basis for common syndromes of COVID-19 according to correlated symptom clusters and their contribution to syndromes,which could provide new thoughts and methods for differentiation standard.

关 键 词：新型冠状病毒 肺炎 常见证候 症状 贡献度 诊断依据 

分 类 号：R259[医药卫生—中西医结合]