面向肺癌筛查的呼出气检测电子鼻设计与应用  

Design and Application of a Breathalyzer Electronic Nose for Lung Cancer Screening

作　　者：周佳骏[1] 李华曜[1] 向润[2] 田博 王乐[1] 李龙[1] 吴蓉 李强[2] 刘欢[1] ZHOU Jiajun;LI Huayao;XIANG Run;TIAN Bo;WANG Le;LI Long;WU Rong;LI Qiang;LIU Huan(Optics Valley Laboratory,School of Integrated Circuits,Huazhong University of Science and Technology,Wuhan Hubei 430074,China;Department of Thoracic Surgery,Sichuan Cancer Hospital&Institute,Chengdu Sichuan 610044,China)

机构地区：[1]华中科技大学集成电路学院光谷实验室,湖北武汉430074 [2]四川省肿瘤医院胸外科,四川成都610044

出　　处：《传感技术学报》2024年第7期1277-1284,共8页Chinese Journal of Sensors and Actuators

基　　金：深圳市协同创新科技计划国际科技合作项目(GJHZ20200731095809028);湖北省自然科学基金项目(2022CFA035)。

摘　　要：人体呼出气包含的挥发性有机化合物(Volatile Organic Compounds,VOCs)与人体健康和疾病状态密切相关。电子鼻将气体传感器阵列与模式识别算法结合,可实现气味的识别与分类。针对肺癌患者诊治端口前移需求,开展呼出气检测电子鼻系统的设计与应用研究,基于半导体气体传感器阵列优化电路与气路设计,构建便携式电子鼻系统样机,对129例临床呼出气样本进行检测;通过数据处理与统计分析,研究健康对照(51例)与肺部疾病患者(78例)的分类效果。结果表明,设计的电子鼻对肺部疾病患者的识别准确率达89.1%(敏感性83.3%,特异性98.0%),展现出应用于肺癌早期筛查的潜力,有望成为一种无创、便捷的肺部疾病辅助诊断方法。Volatile Organic Compounds(VOCs)found in human exhalation are closely linked to human health and disease states.The electronic nose,which combines a gas sensor array and a pattern recognition algorithm,is used to recognize and classify odors.An electronic nose system is designed and Utilited to meet the demand of moving the diagnosis and treatment port of lung cancer patients forward,and a prototype of a portable electronic nose system is built using semiconductor gas sensor arrays to analyze 129 clinical exhaled breath samples.The results show that the designed electronic nose system has an accuracy of 89.1%(sensitivity 83.3%,specificity 98.0%)in identifying patients with lung diseases,showing potential application in the early screening of lung cancer.The designed electronic nose system is a promising non-invasive and convenient auxiliary diagnostic method for lung diseases.

关 键 词：电子鼻 呼出气检测 机器学习 统计分析 肺癌 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]