【摘要】 本研究介绍了一种新型的基于虚拟气体传感器阵列及图象识别方法的新型无创肺癌检测与诊断电子鼻。该电子鼻包含一个由固相微萃取和毛细管柱组成的前处理装置实现病人呼吸气体中有机气体成分的浓缩吸附、脱附和分离,通过一个表面涂覆聚异丁稀薄膜差动结构的声表面波传感器对分离后的有机气体成分进行定量检测。此外,采用由毛细管柱方法实现的虚拟气体传感器阵列及改进的人工神经网络算法实现对肺癌呼吸气味图像的有效识别。通过临床实验验证,表明该电子鼻仪器可以有效地识别出肺癌患者、肺癌疑似病人和健康人,因此将有望在包括肺癌等疾病的早期诊断仪器中发挥重要的作用。更多还原

【Abstract】 A novel non-invasive electronic nose for detection and diagnosis of lung cancer based on a kind of virtual gas sensors array and imaging recognition method is proposed in this paper.The electronic nose includes a gas path constructed by solid phase micro extraction(SPME) and capillary column to concentrate,desorb and separate volatile organic compounds(VOCs) in patients’ breath respectively.A pair of surface acoustic wave(SAW) sensors coated with a thin poly-isobutylene(PIB) film was used to detect chemical compounds.Besides,a virtual sensors array based on SAW sensor using gas chromatography(GC) technique and an artificial neural network(ANN) algorithm combined with imaging were proposed for the recognition of patients’ breath.Finally,the clinical experimental results show that the electronic nose could recognize lung cancer patents,suspected lung cancer patents and healthy persons.The electronic nose is expected to be used in the early diagnosis through further development.更多还原