Infection Detection through Temperature Monitoring

This paper describes the theory behind a smartphone application designed to detect COVID-19 infections through temperature monitoring. COVID-19 infections can cause a temperature increase typically around 0.5 C over a slow, moderate, nonlinear multi-day course averaging 5 days. To enable detection of this increase, filtering techniques are applied to simultaneously establish a base temperature and to detect a COVID-typical deviation from that base temperature. Numerical simulations are developed to assess the effectiveness of the application in detecting the presence of an infection in individuals and in groups. The measure of effectiveness is the number of new-infections incurred before the initial infection is detected. It is observed that group testing and group analysis becomes increasingly effective as the infection rate is increased, e.g. R0 ≥ 3. If the infection rate is low, e.g. R0 ≤ 1.1, group analysis becomes ineffective because, at any given time, too few people are infected simultaneously to be detectable in the group average. The effectiveness of individual testing and the individual analysis is unaffected by the infection rate. This result is significant since it has to be expected that widespread vaccination as well as social distancing efforts significantly reduce the infection rate making outbreaks challenging to spot. © 2022, American Institute of Aeronautics and Astronautics Inc.. All rights reserved.