The Research Of Diabetes Breath Detection Analyzer Based On The Principle Of Electronic Nose

The oxidation process speed of fatty acid in diabetic patients increases,which promotes the liver to produce ketone body and acetone is the final product of ketone body metabolism.Therefore,acetone could be used as a marker of diabetes breath detection.Compared with the traditional detection methods of diabetes,such as blood detection and urine detection,breath detection has the advantages of fast,convenient and noninvasive.In the research,a diabetes breath detection analyzer is designed based on the principle of electronic nose,and the performance of the system is verified through the experiments.The design of this system has positive significance for the noninvasive detection of diabetes.The main work is summarized as follows.（1）The Fabrication and Measurement of MOS Gas Sensor.The gas sensor array is the core of the electronic nose system.Two kinds of MOS nano gas sensing materials,Sn O₂and WO₃,were prepared.Based on the two kinds of gas sensing materials,the sensitivity and selectivity of MOS sensors were modified by noble metals and coated with zeolite films.The performance of the sensors was tested by a dynamic gas distribution test system.The results show that the sensitivity of noble metal loaded MOS gas sensors to acetone is significantly improved.The noble metal loaded and zeolite membrane technology have a significant effect on the selectivity of MOS gas sensors.（2）The Design and Implementation of Breath Analyzer Hardware.The hardware part of breath analyzer mainly includes gas path structure part and circuit part.The gas path structure involves the overall scheme of gas detection path,the selection of gas path components,the design of breath detection cavity and the design of breath injector.The circuit part mainly includes the working circuit of the sensors,the data acquisition circuit based on AD7616 and the design of the system power module.（3）Data Acquisition and Analysis of Breath Analyzer.A data acquisition software based on Lab VIEW for electronic nose system was developed,which realized the real-time display and storage of sensor data.The difference method is used to preprocess the sensor data,and the piecemeal features from the original data were extracted.PCA and BP neural network are used to recognize the breath data.（4）Validation Test of Breath Analyzer.In the study,the function of the breath analyzer was verified by simulation experiments and clinical experiments.The experimental results showed that PCA can effectively reduce the data dimension and the complexity of calculation.In the simulation experiment,the recognition accuracy of breath analyzer for two kinds of samples was 91.67%.The application of this system in clinical needs further optimization.