Study of the effectiveness of diagnostic method for respiratory system diseases by analyzing the exhaled air using a gas analytical complex

Abstract

Aim: To study in patients the dependence of the exhaled air composition on pathological processes occurring in the respiratory system, including: lung cancer, community-acquired pneumonia and COVID-19.Material and Methods. The studies were carried out on the basis of a gas analytical complex using the method of neural network data analysis. The gas analytical complex includes semiconductor sensors that measure the concentrations of gas components in exhaled air with an average sensitivity of 1 ppm. Based on signals from sensors, the neural network classifies and identifies patients with certain pathological processes.Results. The statistical data set for training the neural network and testing the method included samples from 173 patients. Our study collected exhaled air samples from groups of patients with lung cancer, pneumonia, and COVID-19. In the case of lung cancer, the parameters of the diagnostic device have been determined at the level of sensitivity – 95.24%, specificity – 76.19%. For pneumonia and COVID-19, these parameters were 97.36% and 98.63, respectively.Conclusion. Taking into account the known value of diagnostic methods such as computed tomography (CT) and magnetic resonance imaging (MRI), the sensitivity and specificity indicators of the gas analytical complex achieved during the study reflect the promise of the proposed technique in the diagnosis of tumor processes in patients with lung cancer, COVID-19 and community-acquired pneumonia.