慢性胃炎湿热证患者的口腔呼气气味图谱识别  

Recognition of oral breath odor map of chronic gastritis patients with syndrome of damp-heat

作　　者：林雪娟[1,2,3] 高雅 吴青海 田群辉[4] 罗志明 黄伟荣 李灿东 LIN Xue-juan;GAO Ya;WU Qing-hai;TIAN Qun-hui;LUO Zhi-ming;HUANG Wei-rong;LI Can-dong(Research Base of TCM Syndrome of Fujian University of Traditional Chinese Medicine,Fuzhou 350122,China;Key Laboratory of Chinese Medicine Health Status Differentiation of Fujian Province,Fuzhou 350122,China;Fujian Province 2011 TCM Health Management Collaborative Innovation Center,Fuzhou 350122,China;Xiamen University,Xiamen 361005,China;Jinjiang Hospital of Traditional Chinese Medicine Affiliated to Fujian University of Traditional Chinese Medicine,Jinjiang 362201,China)

机构地区：[1]福建中医药大学中医证研究基地,福州350122 [2]福建省中医健康状态辨识重点实验室,福州350122 [3]福建省2011中医健康管理协同创新中心,福州350122 [4]厦门大学,厦门361005 [5]福建中医药大学附属晋江中医院,晋江362201

出　　处：《中华中医药杂志》2023年第2期866-870,共5页China Journal of Traditional Chinese Medicine and Pharmacy

基　　金：国家自然科学基金项目(No.81973752,No.81373552);福建省自然科学基金项目(No.2018J01892);载人航天领域预先研究项目(No.020104)。

摘　　要：目的:运用电子鼻技术结合模式识别算法探讨慢性胃炎(CG)湿热证口腔呼气的气味图谱识别。方法:选择156例CG湿热证患者为研究对象,150例CG非湿热证患者和100名健康者为对照,运用基于阵列式气体传感器技术的电子鼻采集口腔呼气的气味图谱,采用模式识别的方法进行气味图谱识别。结果:(1)CG湿热证患者中不同湿热比重的频数分布从高到低依次为湿热并重、湿重于热、热重于湿;CG湿热证Hp阳性率为39.74%。(2)RF模型对CG和健康人气味图谱的分类最为准确,准确率可达98%;KNN聚类对CG湿热证和非湿热证气味图谱的分类最为准确,准确率为63%。(3)LR模型对湿热并重与非湿热并重气味图谱的分类最为准确,准确率为62%;LR模型对湿重于热与非湿重于热气味图谱的分类最为准确,准确率为71%;LR模型对热重于湿与非热重于湿气味图谱的分类最为准确,准确率达90%;SVM模型对CG湿热证Hp阳性与Hp阴性气味图谱的分类最为准确,准确率达60%。结论:运用阵列式气体传感器电子鼻检测人体口腔呼气,不仅对CG患者具有很高的识别准确率,而且也能初步判断CG湿热证及其不同湿热比重、Hp感染情况,为中医病证诊断和嗅诊客观化提供新方法。Objective: To explore the recognition of oral breath odor map of chronic gastritis(CG) patients with syndrome of damp-heat by electronic nose technology combined with pattern recognition algorithm. Methods: A total of 156CG patients with of damp-heat syndrome and 150 CG patients with non-damp-heat syndrome and 100 healthy persons were observed. Odor map of oral breath were collected by the electronic nose based on a film of gas sensor array. The method of pattern recognition was used to recognize the odor map. Results:(1)The frequency distribution of different proportion of damp and heat in CG patients with damp heat syndrome from high to low was equal damp and heat,more damp with less heat,more heat with less damp. The positive rate of Hp in CG damp-heat syndrome was 39.74%.(2)The RF model was the most accurate for the classification of odor maps of CG and healthy persons. The accuracy rate could reach 98%. KNN clustering was the most accurate for the classification of odor maps of damp-heat syndrome and non-damp-heat syndrome in CG. The accuracy rate was 63%.(3)LR model was the most accurate for the classification of odor maps of equal damp and heat syndrome and non-equal damp and heat syndrome. The accuracy rate was 62%. The LR model was the most accurate for the classification of odor maps of more damp with less heat syndrome and non-more damp with less heat syndrome. The accuracy rate was 71%. The LR model was the most accurate for the classification of odor maps of more heat with less damp syndrome and non-more heat with less damp syndrome.The accuracy rate was 90%. The SVM model was the most accurate for the classification of odor maps of Hp positive and Hp negative in CG damp-heat syndrome. The accuracy rate was 60%. Conclusion: The CG patient could be recognized accuracy by the electronic nose based on a film of gas sensor array. The damp-heat syndrome and its different proportion of damp and heat and Hp infection in CG also could be recognized preliminary. The electronic nose could provide a method for

关 键 词：慢性胃炎 湿热证 电子鼻 口腔呼气 气味图谱 嗅诊 模式识别算法 Hp感染 

分 类 号：R259[医药卫生—中西医结合]