Consumer-grade wearables identify changes in multiple physiological systems during COVID-19 disease progression.

Consumer-grade wearables are needed to track disease, especially in the ongoing pandemic, as they can monitor patients in real time. We show that decomposing heart rate from low-cost wearable technologies into signals from different systems can give a multidimensional description of physiological changes due to COVID-19 infection. We find that the separate physiological features of basal heart rate, heart rate response to physical activity, circadian variation in heart rate, and autocorrelation of heart rate are significantly altered and can classify symptomatic versus healthy periods. Increased heart rate and autocorrelation begin at symptom onset, while the heart rate response to activity increases soon after symptom onset and increases more in individuals exhibiting cough. Symptom onset is associated with a blunting of circadian variation in heart rate, as measured by the uncertainty in the phase estimate. This work establishes an innovative data analytic approach to monitor disease progression remotely using consumer-grade wearables.

Monitoring beliefs and physiological measures in students at risk for COVID-19 using wearable sensors and smartphone technology: Protocol for a mobile health study.

BACKGROUND: The COVID-19 pandemic has impacted lives significantly and greatly affected an already vulnerable population, college students, in relation to mental health and public safety. Social distancing and isolation have brought about challenges to student's mental health. Mobile health apps and wearable sensors may help to monitor students at risk for COVID-19 and support their mental well-being. OBJECTIVE: Through the use of a wearable sensor and smartphone-based survey completion, this study aimed to monitor students at risk for COVID-19. METHODS: We conducted a prospective study of students, undergraduate and graduate, at a public university in the Midwest. Students were instructed to download the Fitbit, Social Rhythms, and Roadmap 2.0 apps onto their personal mobile devices (Android or iOS). Subjects consented to provide up to 10 saliva samples during the study period. Surveys were administered through the Roadmap 2.0 app at five timepoints - at baseline, 1-month later, 2-months later, 3-months later, and at study completion. The surveys gathered information regarding demographics, COVID-19 diagnoses and symptoms, and mental health resilience, with the aim of documenting the impact of COVID-19 on the college student population. RESULTS: This study enrolled 2,158 college students between September 2020 and January 2021. Subjects are currently being followed on-study for one academic year. Data collection and analysis are ongoing. CONCLUSIONS: This study examined student health and well-being during the COVID-19 pandemic. It also assessed the feasibility of wearable sensor use and survey completion in a college student population, which may inform the role of our mobile health tools on student health and well-being. Finally, using wearable sensor data, biospecimen collection, and self-reported COVID-19 diagnosis, our results may provide key data towards the development of a model for the early prediction and detection of COVID-19. CLINICALTRIAL: ClinicalTrials.gov NCT04766788.

Monitoring Health Care Workers at Risk for COVID-19 Using Wearable Sensors and Smartphone Technology: Protocol for an Observational mHealth Study.

BACKGROUND: Health care workers (HCWs) have been working on the front lines of the COVID-19 pandemic with high risks of viral exposure, infection, and transmission. Standard COVID-19 testing is insufficient to protect HCWs from these risks and prevent the spread of disease. Continuous monitoring of physiological data with wearable sensors, self-monitoring of symptoms, and asymptomatic COVID-19 testing may aid in the early detection of COVID-19 in HCWs and may help reduce further transmission among HCWs, patients, and families. OBJECTIVE: By using wearable sensors, smartphone-based symptom logging, and biospecimens, this project aims to assist HCWs in self-monitoring COVID-19. METHODS: We conducted a prospective, longitudinal study of HCWs at a single institution. The study duration was 1 year, wherein participants were instructed on the continuous use of two wearable sensors (Fitbit Charge 3 smartwatch and TempTraq temperature patches) for up to 30 days. Participants consented to provide biospecimens (ie, nasal swabs, saliva swabs, and blood) for up to 1 year from study entry. Using a smartphone app called Roadmap 2.0, participants entered a daily mood score, submitted daily COVID-19 symptoms, and completed demographic and health-related quality of life surveys at study entry and 30 days later. Semistructured qualitative interviews were also conducted at the end of the 30-day period, following completion of daily mood and symptoms reporting as well as continuous wearable sensor use. RESULTS: A total of 226 HCWs were enrolled between April 28 and December 7, 2020. The last participant completed the 30-day study procedures on January 16, 2021. Data collection will continue through January 2023, and data analyses are ongoing. CONCLUSIONS: Using wearable sensors, smartphone-based symptom logging and survey completion, and biospecimen collections, this study will potentially provide data on the prevalence of COVID-19 infection among HCWs at a single institution. The study will also assess the feasibility of leveraging wearable sensors and self-monitoring of symptoms in an HCW population. TRIAL REGISTRATION: ClinicalTrials.gov NCT04756869; https://clinicaltrials.gov/ct2/show/NCT04756869. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/29562.

Machine learning-based cytokine microarray digital immunoassay analysis.

Serial measurement of a large panel of protein biomarkers near the bedside could provide a promising pathway to transform the critical care of acutely ill patients. However, attaining the combination of high sensitivity and multiplexity with a short assay turnaround poses a formidable technological challenge. Here, the authors develop a rapid, accurate, and highly multiplexed microfluidic digital immunoassay by incorporating machine learning-based autonomous image analysis. The assay has achieved 12-plexed biomarker detection in sample volume <15 µL at concentrations < 5 pg/mL while only requiring a 5-min assay incubation, allowing for all processes from sampling to result to be completed within 40 min. The assay procedure applies both a spatial-spectral microfluidic encoding scheme and an image data analysis algorithm based on machine learning with a convolutional neural network (CNN) for pre-equilibrated single-molecule protein digital counting. This unique approach remarkably reduces errors facing the high-capacity multiplexing of digital immunoassay at low protein concentrations. Longitudinal data obtained for a panel of 12 serum cytokines in human patients receiving chimeric antigen receptor-T (CAR-T) cell therapy reveals the powerful biomarker profiling capability. The assay could also be deployed for near-real-time immune status monitoring of critically ill COVID-19 patients developing cytokine storm syndrome.