Acute respiratory viral infections in active-duty military personnel: Early diagnosis using biometric video analytics

Abstract

Acute respiratory viral infections (ARVIs) are an important cause of morbidity in the military setting for many decades. It has high rates of seeking medical care and job loss. Due to the stressful living conditions and activities and other abundance of risk factors, the most effective anti-epidemic measures in military teams should be considered the early detection of infected persons and their timely isolation. We used a modern data present a review that focuses on technical level of current software and hardware systems for biometric video analytics and artificial intelligence algorithms, which make it possible to detect early symptoms of infectious diseases. A list of most common symptoms in ARVIs that can be recognized using video surveillance and video analytics was determined. These symptoms can be used as initial empirical data for a comprehensive automated assessment of the person’s individual state parameters. A scheme for the operation of a software and hardware complex for video data analytics for the early detection of infected persons is proposed. This scheme is necessary because some infected persons appear healthy for some time but keep infecting others when they interact with them. The use of independent machine learning based on the principle of a retrospective statistical analysis of locomotor data and other signs identified in infected persons in the prodromal period of the disease can establish reliable diagnostic correlations based on big data. The data accumulation of the features of the preclinical stages of ARVIs using the proposed approach will lead to the formation of a minimum informative set of video analytical signs (markers) that allow them to be reliably recognized in the prodromal period for the purpose of timely isolation and additional examination of the infected persons and to protect non-infected persons. The implementation of the developed direction will improve the effectiveness of anti-epidemic measures through early localization and liquidation of the epidemic.