Research Into Exhaled Gas Detection And Disease Diagnostic Model Based On Conductive Polymer Gas Sensing Array

With the development of society and the improvement of people’s living standards,people’s health has become one of the key concerns.Among them,kidney disease,diabetes,halitosis and other diseases have become one of the diseases with high incidence.Real-time online detection of human diseases and protection of human health are important measure to improve the happiness index of residents.Human exhaled gas detection is an effective method for early screening and diagnosis of human diseases,and is of great significance for monitoring and diagnosing human health.Real-time monitoring of exhaled gas sensors is a key to early screening and diagnosis of human diseases.This paper is devoted to exploring nano-modified conductive polymer gas sensing array in collaboration with intelligent information processing technology to build human exhaled gas detection system.PANI/MWCNTs/MoS₂ sensor,PPy/Zn₂SnO₄ sensor,ZnO/S,N:GQDs/PANI sensor and SnO₂/rGO/PANI sensor were prepared by in-situ polymerization,sacrificial-template method and layer by layer self-assembly method.The main characteristic gases（NH₃,H₂S,acetone）in human exhaled gases were studied experimentally.The sensitivity,stability,selectivity and response/recovery time of the sensor were studied.The sensitive mechanism of nano-modified conductive polymer sensor to human exhaled gas was revealed by combining the characterization methods of SEM,TEM,FT-IR,BET,XPS,XRD,EDS with heterojunction band structure and charge transfer.At the same time,high-performance sensors were selected to form sensing array for human exhaled gas.The multi-dimensional response data of the gas sensing array under the simulated human illness state was obtained.Principal Component Analysis（PCA）was used to extract feature and reduce dimension of response data.On this basis,the Radial Basis Function（RBF）neural network model was constructed.Particle swarm optimization（PSO）was used to optimize RBF parameters.The algorithm was optimized to effectively reduce the impact of data dimension and cross-sensitivity.Thus,the high-precision prediction of the composition and concentration of exhaled gas could be realized.Finally,further research on disease diagnosis model based on human exhaled gas was carried out.Cluster analysis,Support Vector Machine（SVM）and Deep Belief Net（DBN）were used to construct human disease diagnosis model.It was found that the DBN network model had a high diagnostic accuracy and met the requirements of early diagnosis of human diseases.It had far-reaching significance to maintain human health.This paper focused on new material,new device and new application as the route.The preparation and gas sensing properties of nano-modified conductive polymer sensors were studied.The data prediction and disease diagnosis model based on gas sensing array collaborative intelligent algorithm were constructed which provided a new application for nano-sensor technology fusion intelligent algorithm in the field of early screening and diagnosis of human diseases.At the same time,it provided a new idea for human health monitoring.