电子鼻检测常见伤口感染细菌实验研究  

Experimental study on detection of common bacteria in wound infection by electronic nose

作　　者：陆彦邑 曾琳 严博文 黎敏 何庆华 LU Yanyi;ZENG Lin;YAN Bowen;LI Min;HE Qinghua(State Key Laboratory of Trauma,Burns and Combined Injury/Daping Hospital,Army Medical University,Chongqing 400042,China)

机构地区：[1]陆军军医大学大坪医院/国家创伤烧伤复合伤重点实验室,重庆400042

出　　处：《中国医学物理学杂志》2021年第10期1268-1272,共5页Chinese Journal of Medical Physics

基　　金：国家国际科技合作专项(2014DFA31560);重庆市科技研发基地建设计划(国际科技合作)项目(cstc2013gjhz10003)。

摘　　要：为了进一步探讨电子鼻用于伤口细菌感染快速筛查的可能性,使用自制电子鼻检测鲍曼不动杆菌、大肠杆菌、肺炎克雷伯杆菌、金黄色葡萄球菌、铜绿假单胞菌的巯基乙酸酯(TH)培养液及纯培养液,经过预处理后,使用支持向量机、BP神经网络、逻辑回归3种算法进行分类。结果显示,使用BP算法,总体识别率可达93%以上。对于单个识别率,检测大肠杆菌、肺炎克雷伯杆菌、金黄色葡萄球菌可达98%以上,检测鲍曼不动杆菌、铜绿假单胞菌可达90%左右,表明电子鼻用于伤口感染细菌的快速检测和识别具有一定的可行性。To further explore the application of electronic nose in the rapid screening of bacterial infection of wounds,a selfmade electronic nose is used to detect Thioglycolate(TH)broth of Acinetobacter baumannii(A.Baumannii),Escherichia coli(E.Coli),Klebsiella pneumoniae(K.pneumoniae),Staphylococcus aureus(S.aureus),and Pseudomonas aeruginosa(P.aeruginosa)and sterile TH broth.After preprocessing,the data are classified by support vector machine,BP neural network and logistic regression,separately.The results show that BP neural network can achieve an overall recognition rate over 90%,and specific recognition rates over 98%for E.Coli,K.pneumoniae and S.aureus,and that its recognition rates for A.Baumannii and P.aeruginosa are around 90%,indicating that electronic nose is feasible for the rapid detection and recognition of bacteria in wound infection.

关 键 词：伤口感染细菌 电子鼻 分类识别 

分 类 号：R318[医药卫生—生物医学工程] Q93-331[医药卫生—基础医学]