Design of an Electronic Nose System for Exhaled Gas Markers of Acute Kidney Injury Patients

Abstract

In order to obtain the information of exhaled gas marker concentration of acute kidney injury patients quickly and accurately, a quantitative VOCs detection system based on metal oxide gas sensors array and BP neural network model is proposed in this paper. Firstly, four kinds of metal oxide gas sensors are used to form a sensor array to convert the measured gas components and concentrations into electrical signal waveforms, and the feature matrix is obtained by feature extraction of the sensor array's signals. Then, a BP neural network model is used to predict the target gas concentrations, and the topology design and parameters of the neural network are optimized.