Real-time infection prediction with wearable physiological monitoring and AI to aid military workforce readiness during COVID-19

Abstract

Infectious threats, like the COVID-19 pandemic, hinder maintenance of a productive and healthy workforce. If subtle physiological changes precede overt illness, then proactive isolation and testing can reduce labor force impacts. This study hypothesized that an early infection warning service based on wearable physiological monitoring and predictive models created with machine learning could be developed and deployed. We developed a prototype tool, first deployed June 23, 2020, that delivered continuously updated scores of infection risk for SARS-CoV-2 through April 8, 2021. Data were acquired from 9381 United States Department of Defense (US DoD) personnel wearing Garmin and Oura devices, totaling 599,174 user-days of service and 201 million hours of data. There were 491 COVID-19 positive cases. A predictive algorithm identified infection before diagnostic testing with an AUC of 0.82. Barriers to implementation included adequate data capture (at least 48% data was needed) and delays in data transmission. We observe increased risk scores as early as 6 days prior to diagnostic testing (2.3 days average). This study showed feasibility of a real-time risk prediction score to minimize workforce impacts of infection.

A method for intelligent allocation of diagnostic testing by leveraging data from commercial wearable devices: a case study on COVID-19

“…To our knowledge, this is the first demonstration of this tradeoff in predictive model development for COVID-19 infection detection. The ITA model, in addition to using features of RHR and steps, can likely be further extended and improved with features from other digital biomarkers such as skin temperature, respiratory rate, blood oxygen saturation, and sleep duration 25 , 26 , 35 , 36 . It is anticipated that each of these distinct digital biomarkers would capture a physiological response to infection at different times during the detection period, thus improving the robustness and overall performance of the ITA approach.…”

A Method for Intelligent Allocation of Diagnostic Testing by Leveraging Data from Commercial Wearable Devices: A Case Study on COVID-19

“…To our knowledge, this is the rst demonstration of this tradeoff in predictive model development for COVID-19 infection detection. The ITA model, in addition to using features of RHR and steps, can likely be further extended and improved with features from other digital biomarkers such as skin temperature, respiratory rate, blood oxygen saturation, and sleep duration [21], [22], [33], [34]. It is anticipated that each of these distinct digital biomarkers would capture a physiological response to infection at different times during the detection period, thus improving the robustness and overall performance of the ITA approach.…”

“…For model development with the training dataset, we utilized nested CV over as traditional CV, which is a common approach in similar studies [26], [32], [34], [35], because it uses the same data for hyperparameter tuning and model performance evaluation [56]. In nested CV (also called double CV), the hyperparameter tuning procedure is nested (inner loop) under the model selection procedure (outer loop) and the inner loop is used for optimizing the hyperparameters of the model with inner CV, and the outer loop is used to compute the error of the optimized model with outer CV.…”

Beyond Pathogen Filtration: Possibility of Smart Masks as Wearable Devices for Personal and Group Health and Safety Management

“…These measures help monitor the soldier's physical and mental health status and decision-making. Moreover, breath analysis can predict and monitor the onset of pulmonary injury due to various environmental and infectious exposures [ 86 ].…”