基于电子鼻的慢性胃炎气滞证患者常见病位的口腔呼气气味图谱研究  

Study on oral exhalation odor pattern among common disease locations in patients of chronic gastritis with qi stagnation syndrome based on electronic nose

作　　者：林雪娟[1,2,3] 冯秀君 梁丽丽 吴青海[1,2,3] 郭锋 李灿东[1,2,3] 黄伟荣 LIN Xue-juan;FENG Xiu-jun;LIANG Li-li;WU Qing-hai;GUO Feng;LI Can-dong;HUANG Wei-rong(Research Base of Traditional Chinese Medicine Syndrome,Fujian University of Chinese Medicine,Fuzhou 350122,China;Key Laboratory on TCM Health Differentiation in Fujian Province,Fujian University of Chinese Medicine,Fuzhou 350122,China;Collaborative Innovation Center of Health Management of Traditional Chinese Medicine of Fujian Province in 2011,Fuzhou 350122,China;Intelligent Science and Technology Department,Xiamen University,Xiamen 361005,China;Jinjiang Hospital of Traditional Chinese Medicine Affiliated to Fujian University of TCM,Jinjiang 362200,China)

机构地区：[1]福建中医药大学中医证研究基地,福州350122 [2]福建中医药大学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]厦门大学智能科学与技术系,厦门361005 [5]福建中医药大学附属晋江中医院,福建晋江362200

出　　处：《中华中医药杂志》2018年第9期4057-4060,共4页China Journal of Traditional Chinese Medicine and Pharmacy

基　　金：国家自然科学基金项目(No.81373552); 福建省自然科学基金项目(No.2014J01362); 福建省中医药科研项目(No.wzpw201313); 福建省教育厅A类项目(No.JA14212); 载人航天领域预先研究项目(No.020104)

摘　　要：目的:运用电子鼻探讨慢性胃炎气滞证患者常见病位的口腔呼气气味图谱特征。方法:采用证素辨证的方法,筛选出397例慢性胃炎气滞证患者并判断病位证素,同时运用基于阵列式气体传感器技术的医用电子鼻(EN0l1103-A)采集其口腔呼气的气味图谱,选择气味图谱响应曲线的振幅、斜率作为图谱特征参数,借助分类器算法对慢性胃炎气滞证与非气滞证的口腔呼气气味图谱特征进行模式识别,比较慢性胃炎气滞证患者常见病位的口腔呼气气味图谱特征。结果:慢性胃炎气滞证的主要病位证素分布是胃(91.18%)、脾(38.29%)、肝(23.68%);主要病位为胃、脾胃、肝脾胃、肝胃;采用分类器算法对慢性胃炎气滞证的气味图谱进行模式识别时,运用随机森林算法,对气滞证的准确率可以达到65.85%;病位胃组、脾胃组的气味图谱响应曲线B、C、D、E、F、I、J的振幅均显著低于病位肝脾胃组(P<0.01);病位胃组、脾胃组的气味图谱响应曲线C、D、E、F、I的斜率均显著低于病位肝脾胃组(P<0.01),脾胃组曲线A的斜率低于病位肝脾胃组(P<0.05)。结论:运用电子鼻结合模式识别方法可初步判断慢性胃炎气滞证及其不同病位间的口腔呼气气味差异。Objective： to explore the oral exhalation odor pattern characteristics among the common disease location of Chronic gastritis（CG） patients with qi stagnation syndrome by electronic nose. Methods： A total of 397 cases of CG patients with qi stagnation syndrome were selected and disease locations were judged by the syndrome elements differentiation method. Oral exhalation odor pattern was collected by the electronic nose（EN0 l1103-A） based on array gas sensor technology. The amplitude and slope of response curves were selected as pattern characteristic parameters. Recognition of oral exhalation odor pattern characteristics between CG patients with qi stagnation syndrome and patients with non-qi stagnation syndrome was conducted by classifier algorithm. Oral exhalation odor pattern characteristics among the common disease locations of CG patients with qi stagnation syndrome were compared. Results： Distribution feature of disease location of qi stagnation syndrome elements in CG was 91.18% in the stomach, 38.29% in the spleen and 23.68% in the liver; the major disease location was stomach, spleen and stomach, liver and spleen stomach, liver and stomach. The pattern recognition of the oral exhalation odor pattern characteristics between patients with qi stagnation syndrome of CG and non-qi stagnation syndrome patients of CG was by classifier algorithm： using random forest algorithm, the accuracy of random forest algorithm can reach 65.85%. The amplitudes of oral exhalation odor pattern curves of B, C, D, E, I,F, J of the stomach group and the combination of stomach and spleen group of CG patients with qi stagnation syndrome were significantly lower than those in the combination stomach and spleen and liver group（P〈0.01）. The slopes of oral exhalation odor pattern curves of B, C, D, E, F, I of the stomach group and the combination of stomach and spleen group of CG patients with qi stagnation syndrome were significantly lower than those in the combination stomach and spleen and liver group（P〈0.

关 键 词：慢性胃炎 气滞证 病位 电子鼻 气味图谱 分类器算法 响应曲线振幅 响应曲线斜率 模式识别 

分 类 号：R259[医药卫生—中西医结合]