Characterizing COVID-19 and Influenza Illnesses in the Real World via Person-Generated Health Data

Evaluation of Machine Learning to Detect Influenza Using Wearable Sensor Data and Patient-Reported Symptoms: Cohort Study

BACKGROUND

Machine learning offers quantitative pattern recognition analysis of wearable device data and has the potential to detect illness onset and monitor influenza-like illness (ILI) in patients who are infected.

OBJECTIVE

This study aims to evaluate the ability of machine-learning algorithms to distinguish between participants who are influenza positive and influenza negative in a cohort of symptomatic patients with ILI using wearable sensor (activity) data and self-reported symptom data during the latent and early symptomatic periods of ILI.

METHODS

This prospective observational cohort study used the extreme gradient boosting (XGBoost) classifier to determine whether a participant was influenza positive or negative based on 3 models using symptom-only data, activity-only data, and combined symptom and activity data. Data were collected from the Home Testing of Respiratory Illness (HTRI) study and FluStudy2020, both conducted between December 2019 and October 2020. The model was developed using the FluStudy2020 data and tested on the HTRI data. Analyses included participants in these studies with an at-home influenza diagnostic test result. Fitbit (Google LLC) devices were used to measure participants' steps, heart rate, and sleep parameters. Participants detailed their ILI symptoms, health care-seeking behaviors, and quality of life. Model performance was assessed by area under the curve (AUC), balanced accuracy, recall (sensitivity), specificity, precision (positive predictive value), negative predictive value, and weighted harmonic mean of precision and recall (F2) score.

RESULTS

An influenza diagnostic test result was available for 953 and 925 participants in HTRI and FluStudy2020, respectively, of whom 848 (89%) and 840 (90.8%) had activity data. For the training and validation sets, the highest performing model was trained on the combined symptom and activity data (training AUC=0.77; validation AUC=0.74) versus symptom-only (training AUC=0.73; validation AUC=0.72) and activity-only (training AUC=0.68; validation AUC=0.65) data. For the FluStudy2020 test set, the performance of the model trained on combined symptom and activity data was closely aligned with that of the symptom-only model (combined symptom and activity test AUC=0.74; symptom-only test AUC=0.74). These results were validated using independent HTRI data (combined symptom and activity evaluation AUC=0.75; symptom-only evaluation AUC=0.74). The top features guiding influenza detection were cough; mean resting heart rate during main sleep; fever; total minutes in bed for the combined model; and fever, cough, and sore throat for the symptom-only model.

CONCLUSIONS

Machine-learning algorithms had moderate accuracy in detecting influenza, suggesting that previous findings from research-grade sensors tested in highly controlled experimental settings may not easily translate to scalable commercial-grade sensors. In the future, more advanced wearable sensors may improve their performance in the early detection and discrimination of viral respiratory infections.

Optimizing COVID-19 testing resources use with wearable sensors

The timely identification of infectious pre-symptomatic and asymptomatic cases is key towards preventing the spread of a viral illness like COVID-19. Early identification has been done through routine testing programs, which are indeed costly and potentially burdensome for individuals who should be tested with high frequency. A supplemental tool is represented by wearable technology, that can passively monitor and identify individuals at high risk, alerting them to take a test. We designed a Markov chain model and simulated a routine testing and a wearable testing strategy to estimate the number of tests required and the average number of days in which an individual is infectious and undetected. According to our model, with 2 test per month available, we have that the number of infectious and undetected days is 4.1 in the case of routine testing, while it decreases by 46% and 27% with a wearable testing strategy in the presence or absence of self-reported symptoms. The proposed parametric model can be used for different viral illnesses by tuning its parameters. It shows that wearable technology informing a testing strategy can significantly reduce the number of infectious days in which an individuals can spread the virus. With the same number of infectious days, by using wearables we can potentially reduce the number of required tests and the cost of the testing strategy.

Study of Postacute Sequelae of COVID-19 Using Digital Wearables: Protocol for a Prospective Longitudinal Observational Study

BACKGROUND

Postacute sequelae of COVID-19 (PASC) remain understudied in nonhospitalized patients. Digital wearables allow for a continuous collection of physiological parameters such as respiratory rate and oxygen saturation that have been predictive of disease trajectories in hospitalized patients.

OBJECTIVE

This protocol outlines the design and procedures of a prospective, longitudinal, observational study of PASC that aims to identify wearables-collected physiological parameters that are associated with PASC in patients with a positive diagnosis.

METHODS

This is a single-arm, prospective, observational study of a cohort of 550 patients, aged 18 to 65 years, male or female, who own a smartphone or a tablet that meets predetermined Bluetooth version and operating system requirements, speak English, and provide documentation of a positive COVID-19 test issued by a health care professional within 5 days before enrollment. The primary end point is long COVID-19, defined as ≥1 symptom at 3 weeks beyond the first symptom onset or positive diagnosis, whichever comes first. The secondary end point is chronic COVID-19, defined as ≥1 symptom at 12 weeks beyond the first symptom onset or positive diagnosis. Participants must be willing and able to consent to participate in the study and adhere to study procedures for 6 months.

RESULTS

The first patient was enrolled in October 2021. The estimated year for publishing the study results is 2025.

CONCLUSIONS

This is a fully decentralized study investigating PASC using wearable devices to collect physiological parameters and patient-reported outcomes. The study will shed light on the duration and symptom manifestation of PASC in nonhospitalized patient subgroups and is an exemplar of the use of wearables as population-level monitoring health tools for communicable diseases.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04927442; https://clinicaltrials.gov/study/NCT04927442.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/57382.

Continuous Monitoring of Heart Rate Variability in Free-Living Conditions Using Wearable Sensors: Exploratory Observational Study

BACKGROUND

Wearable physiological monitoring devices are promising tools for remote monitoring and early detection of potential health changes of interest. The widespread adoption of such an approach across communities and over long periods of time will require an automated data platform for collecting, processing, and analyzing relevant health information.

OBJECTIVE

In this study, we explore prospective monitoring of individual health through an automated data collection, metrics extraction, and health anomaly analysis pipeline in free-living conditions over a continuous monitoring period of several months with a focus on viral respiratory infections, such as influenza or COVID-19.

METHODS

A total of 59 participants provided smartwatch data and health symptom and illness reports daily over an 8-month window. Physiological and activity data from photoplethysmography sensors, including high-resolution interbeat interval (IBI) and step counts, were uploaded directly from Garmin Fenix 6 smartwatches and processed automatically in the cloud using a stand-alone, open-source analytical engine. Health risk scores were computed based on a deviation in heart rate and heart rate variability metrics from each individual's activity-matched baseline values, and scores exceeding a predefined threshold were checked for corresponding symptoms or illness reports. Conversely, reports of viral respiratory illnesses in health survey responses were also checked for corresponding changes in health risk scores to qualitatively assess the risk score as an indicator of acute respiratory health anomalies.

RESULTS

The median average percentage of sensor data provided per day indicating smartwatch wear compliance was 70%, and survey responses indicating health reporting compliance was 46%. A total of 29 elevated health risk scores were detected, of which 12 (41%) had concurrent survey data and indicated a health symptom or illness. A total of 21 influenza or COVID-19 illnesses were reported by study participants; 9 (43%) of these reports had concurrent smartwatch data, of which 6 (67%) had an increase in health risk score.

CONCLUSIONS

We demonstrate a protocol for data collection, extraction of heart rate and heart rate variability metrics, and prospective analysis that is compatible with near real-time health assessment using wearable sensors for continuous monitoring. The modular platform for data collection and analysis allows for a choice of different wearable sensors and algorithms. Here, we demonstrate its implementation in the collection of high-fidelity IBI data from Garmin Fenix 6 smartwatches worn by individuals in free-living conditions, and the prospective, near real-time analysis of the data, culminating in the calculation of health risk scores. To our knowledge, this study demonstrates for the first time the feasibility of measuring high-resolution heart IBI and step count using smartwatches in near real time for respiratory illness detection over a long-term monitoring period in free-living conditions.

Assessing the current utilization status of wearable devices in clinical research

BACKGROUND/AIMS

Information regarding the use of wearable devices in clinical research, including disease areas, intervention techniques, trends in device types, and sample size targets, remains elusive. Therefore, we conducted a comprehensive review of clinical research trends related to wristband wearable devices in research planning and examined their applications in clinical investigations.

METHODS

As this study identified trends in the adoption of wearable devices during the planning phase of clinical research, including specific disease areas and targeted number of intervention cases, we searched ClinicalTrials.gov-a prominent platform for registering and disseminating clinical research. Since wrist-worn devices represent a large share of the market, we focused on wrist-worn devices and selected the most representative models among them. The main analysis focused on major wearable devices to facilitate data analysis and interpretation, but other wearables were also surveyed for reference. We searched ClinicalTrials.gov with the keywords "ActiGraph,""Apple Watch,""Empatica,""Fitbit,""Garmin," and "wearable devices" to obtain studies published up to 21 August 2022. This initial search yielded 3214 studies. After excluding duplicate National Clinical Trial studies (the overlap was permissible among different device types except for wearable devices), our analysis focused on 2930 studies, including simple, time-series, and type-specific assessments of various variables.

RESULTS

Overall, an increasing number of clinical studies have incorporated wearable devices since 2012. While ActiGraph and Fitbit initially dominated this landscape, the use of other devices has steadily increased, constituting approximately 10% of the total after 2015. Observational studies outnumbered intervention studies, with behavioral and device-based interventions being particularly prevalent. Regarding disease types, cancer and cardiovascular diseases accounted for approximately 20% of the total. Notably, 114 studies adopted multiple devices simultaneously within the context of their clinical investigations.

CONCLUSIONS

Our findings revealed that the utilization of wearable devices for data collection and behavioral interventions in various disease areas has been increasing over time since 2012. The increase in the number of studies over the past 3 years has been particularly significant, suggesting that this trend will continue to accelerate in the future. Devices and their evaluation methods that have undergone thorough validation, confirmed their accuracy, and adhered to established legal regulations will likely assume a pivotal role in evaluations, allowing for remote clinical trials. Moreover, behavioral intervention therapy utilizing apps is becoming more extensive, and we expect to see more examples that will lead to their approval as programmed medical devices in the future.

Detection of Common Respiratory Infections, Including COVID-19, Using Consumer Wearable Devices in Health Care Workers: Prospective Model Validation Study

BACKGROUND

The early detection of respiratory infections could improve responses against outbreaks. Wearable devices can provide insights into health and well-being using longitudinal physiological signals.

OBJECTIVE

The purpose of this study was to prospectively evaluate the performance of a consumer wearable physiology-based respiratory infection detection algorithm in health care workers.

METHODS

In this study, we evaluated the performance of a previously developed system to predict the presence of COVID-19 or other upper respiratory infections. The system generates real-time alerts using physiological signals recorded from a smartwatch. Resting heart rate, respiratory rate, and heart rate variability measured during the sleeping period were used for prediction. After baseline recordings, when participants received a notification from the system, they were required to undergo testing at a Northwell Health System site. Participants were asked to self-report any positive tests during the study. The accuracy of model prediction was evaluated using respiratory infection results (laboratory results or self-reports), and postnotification surveys were used to evaluate potential confounding factors.

RESULTS

A total of 577 participants from Northwell Health in New York were enrolled in the study between January 6, 2022, and July 20, 2022. Of these, 470 successfully completed the study, 89 did not provide sufficient physiological data to receive any prediction from the model, and 18 dropped out. Out of the 470 participants who completed the study and wore the smartwatch as required for the 16-week study duration, the algorithm generated 665 positive alerts, of which 153 (23.0%) were not acted upon to undergo testing for respiratory viruses. Across the 512 instances of positive alerts that involved a respiratory viral panel test, 63 had confirmed respiratory infection results (ie, COVID-19 or other respiratory infections detected using a polymerase chain reaction or home test) and the remaining 449 had negative upper respiratory infection test results. Across all cases, the estimated false-positive rate based on predictions per day was 2%, and the positive-predictive value ranged from 4% to 10% in this specific population, with an observed incidence rate of 198 cases per week per 100,000. Detailed examination of questionnaires filled out after receiving a positive alert revealed that physical or emotional stress events, such as intense exercise, poor sleep, stress, and excessive alcohol consumption, could cause a false-positive result.

CONCLUSIONS

The real-time alerting system provides advance warning on respiratory viral infections as well as other physical or emotional stress events that could lead to physiological signal changes. This study showed the potential of wearables with embedded alerting systems to provide information on wellness measures.

Information theory reveals physiological manifestations of COVID-19 that correlate with symptom density of illness

Algorithms for the detection of COVID-19 illness from wearable sensor devices tend to implicitly treat the disease as causing a stereotyped (and therefore recognizable) deviation from healthy physiology. In contrast, a substantial diversity of bodily responses to SARS-CoV-2 infection have been reported in the clinical milieu. This raises the question of how to characterize the diversity of illness manifestations, and whether such characterization could reveal meaningful relationships across different illness manifestations. Here, we present a framework motivated by information theory to generate quantified maps of illness presentation, which we term "manifestations," as resolved by continuous physiological data from a wearable device (Oura Ring). We test this framework on five physiological data streams (heart rate, heart rate variability, respiratory rate, metabolic activity, and sleep temperature) assessed at the time of reported illness onset in a previously reported COVID-19-positive cohort (N = 73). We find that the number of distinct manifestations are few in this cohort, compared to the space of all possible manifestations. In addition, manifestation frequency correlates with the rough number of symptoms reported by a given individual, over a several-day period prior to their imputed onset of illness. These findings suggest that information-theoretic approaches can be used to sort COVID-19 illness manifestations into types with real-world value. This proof of concept supports the use of information-theoretic approaches to map illness manifestations from continuous physiological data. Such approaches could likely inform algorithm design and real-time treatment decisions if developed on large, diverse samples.

COVID-19 Vaccine Effectiveness and Digital Pandemic Surveillance in Germany (eCOV Study): Web Application-Based Prospective Observational Cohort Study

BACKGROUND

The COVID-19 pandemic posed significant challenges to global health systems. Efficient public health responses required a rapid and secure collection of health data to improve the understanding of SARS-CoV-2 and examine the vaccine effectiveness (VE) and drug safety of the novel COVID-19 vaccines.

OBJECTIVE

This study (COVID-19 study on vaccinated and unvaccinated subjects over 16 years; eCOV study) aims to (1) evaluate the real-world effectiveness of COVID-19 vaccines through a digital participatory surveillance tool and (2) assess the potential of self-reported data for monitoring key parameters of the COVID-19 pandemic in Germany.

METHODS

Using a digital study web application, we collected self-reported data between May 1, 2021, and August 1, 2022, to assess VE, test positivity rates, COVID-19 incidence rates, and adverse events after COVID-19 vaccination. Our primary outcome measure was the VE of SARS-CoV-2 vaccines against laboratory-confirmed SARS-CoV-2 infection. The secondary outcome measures included VE against hospitalization and across different SARS-CoV-2 variants, adverse events after vaccination, and symptoms during infection. Logistic regression models adjusted for confounders were used to estimate VE 4 to 48 weeks after the primary vaccination series and after third-dose vaccination. Unvaccinated participants were compared with age- and gender-matched participants who had received 2 doses of BNT162b2 (Pfizer-BioNTech) and those who had received 3 doses of BNT162b2 and were not infected before the last vaccination. To assess the potential of self-reported digital data, the data were compared with official data from public health authorities.

RESULTS

We enrolled 10,077 participants (aged ≥16 y) who contributed 44,786 tests and 5530 symptoms. In this young, primarily female, and digital-literate cohort, VE against infections of any severity waned from 91.2% (95% CI 70.4%-97.4%) at week 4 to 37.2% (95% CI 23.5%-48.5%) at week 48 after the second dose of BNT162b2. A third dose of BNT162b2 increased VE to 67.6% (95% CI 50.3%-78.8%) after 4 weeks. The low number of reported hospitalizations limited our ability to calculate VE against hospitalization. Adverse events after vaccination were consistent with previously published research. Seven-day incidences and test positivity rates reflected the course of the pandemic in Germany when compared with official numbers from the national infectious disease surveillance system.

CONCLUSIONS

Our data indicate that COVID-19 vaccinations are safe and effective, and third-dose vaccinations partially restore protection against SARS-CoV-2 infection. The study showcased the successful use of a digital study web application for COVID-19 surveillance and continuous monitoring of VE in Germany, highlighting its potential to accelerate public health decision-making. Addressing biases in digital data collection is vital to ensure the accuracy and reliability of digital solutions as public health tools.

COVID-19 surveillance based on consumer wearable devices

Background

Consumer wearable devices such as wristbands and smartwatches have potential application value in communicable disease surveillance.

Objective

We investigated the ability of wearable devices to monitor COVID-19 patients of varying severity.

Methods

COVID-19 patients with mobile phones supporting wearable device applications were selected from Dalian Sixth People Hospital. Physiological parameters from the wearable devices and electronic questionnaires were collected from the device wearing until 14 days post-discharge. Clinical information during hospitalization was also recorded. Based on imaging data, the patients were categorized into the milder group without pneumonia and the more severe group with pneumonia. We plotted the curves of the physiological parameters of the two groups to compare the differences and changes.

Results

Ninety-eight patients were included in the analysis. The mean age was 39.6 ± 10.5 years, including 45 males (45.9%). There were 24 asymptomatic patients, 10 mild patients, 60 moderate patients, and 4 severe patients. Compared with the milder group, the more severe group had higher heart rate-related parameters, while the heart rate variability (HRV) was the opposite. In the more severe group, the heart rate-related parameters showed a downward trend from 0 to 7 days after the fever resolution. Among them, the resting heart rate and sleep heart rate decreased on the 25th day after the onset and were close to the milder group 1 week after discharge.

Conclusions

Consumer wearable devices have the potential to monitor respiratory infections. Heart rate-related parameters obtained from these devices can be sensitive indicators of COVID-19 severity and correlate with disease evolution.

Trial registration

ClinicalTrials.gov NCT04459637.

Unsupervised machine learning to investigate trajectory patterns of COVID-19 symptoms and physical activity measured via the MyHeart Counts App and smart devices

Previous studies have associated COVID-19 symptoms severity with levels of physical activity. We therefore investigated longitudinal trajectories of COVID-19 symptoms in a cohort of healthcare workers (HCWs) with non-hospitalised COVID-19 and their real-world physical activity. 121 HCWs with a history of COVID-19 infection who had symptoms monitored through at least two research clinic visits, and via smartphone were examined. HCWs with a compatible smartphone were provided with an Apple Watch Series 4 and were asked to install the MyHeart Counts Study App to collect COVID-19 symptom data and multiple physical activity parameters. Unsupervised classification analysis of symptoms identified two trajectory patterns of long and short symptom duration. The prevalence for longitudinal persistence of any COVID-19 symptom was 36% with fatigue and loss of smell being the two most prevalent individual symptom trajectories (24.8% and 21.5%, respectively). 8 physical activity features obtained via the MyHeart Counts App identified two groups of trajectories for high and low activity. Of these 8 parameters only 'distance moved walking or running' was associated with COVID-19 symptom trajectories. We report a high prevalence of long-term symptoms of COVID-19 in a non-hospitalised cohort of HCWs, a method to identify physical activity trends, and investigate their association. These data highlight the importance of tracking symptoms from onset to recovery even in non-hospitalised COVID-19 individuals. The increasing ease in collecting real-world physical activity data non-invasively from wearable devices provides opportunity to investigate the association of physical activity to symptoms of COVID-19 and other cardio-respiratory diseases.

A prospective natural history study of post acute sequalae of COVID-19 using digital wearables: Study protocol

Background

Post-acute sequelae of COVID-19 (PASC) is characterized by having 1 + persistent, recurrent, or emergent symptoms post the infection's acute phase. The duration and symptom manifestation of PASC remain understudied in nonhospitalized patients. Literature on PASC is primarily based on data from hospitalized patients where clinical indicators such as respiratory rate, heart rate, and oxygen saturation have been predictive of disease trajectories. Digital wearables allow for a continuous collection of such physiological parameters. This protocol outlines the design, aim, and procedures of a natural history study of PASC using digital wearables.

Methods

This is a single-arm, prospective, natural history study of a cohort of 550 patients, ages 18 to 65 years old, males or females who own a smartphone and/or a tablet that meets pre-determined Bluetooth version and operating system requirements, speak English, and provide documentation of a positive COVID-19 test issued by a healthcare professional or organization within 5 days before enrollment. The study aims to identify wearables collected physiological parameters that are associated with PASC in patients with a positive diagnosis. The primary endpoint is long COVID-19, defined as ≥ 1 symptom at 3 weeks beyond first symptom onset or positive diagnosis, whichever comes first. The secondary endpoint is chronic COVID-19, defined as ≥ 1 symptom at 12 weeks beyond first symptom onset or positive diagnosis. We hypothesize that physiological parameters collected via wearables are associated with self-reported PASC. Participants must be willing and able to consent to participate in the study and adhere to study procedures for six months.

Discussion

This is a fully decentralized study investigating PASC using wearable devices to collect physiological parameters and patient-reported outcomes. Given evidence on key demographics and risk profiles associated with PASC, the study will shed light on the duration and symptom manifestation of PASC in nonhospitalized patient subgroups and is an exemplar of use of wearables as population-level monitoring health tools for communicable diseases.

Trial registration

ClinicalTrials.gov NCT04927442.

Applying machine learning to consumer wearable data for the early detection of complications after pediatric appendectomy

When children are discharged from the hospital after surgery, their caregivers often rely on subjective assessments (e.g., appetite, fatigue) to monitor postoperative recovery as objective assessment tools are scarce at home. Such imprecise and one-dimensional evaluations can result in unwarranted emergency department visits or delayed care. To address this gap in postoperative monitoring, we evaluated the ability of a consumer-grade wearable device, Fitbit, which records multimodal data about daily physical activity, heart rate, and sleep, in detecting abnormal recovery early in children recovering after appendectomy. One hundred and sixty-two children, ages 3-17 years old, who underwent an appendectomy (86 complicated and 76 simple cases of appendicitis) wore a Fitbit device on their wrist for 21 days postoperatively. Abnormal recovery events (i.e., abnormal symptoms or confirmed postoperative complications) that arose during this period were gathered from medical records and patient reports. Fitbit-derived measures, as well as demographic and clinical characteristics, were used to train machine learning models to retrospectively detect abnormal recovery in the two days leading up to the event for patients with complicated and simple appendicitis. A balanced random forest classifier accurately detected 83% of these abnormal recovery days in complicated appendicitis and 70% of abnormal recovery days in simple appendicitis prior to the true report of a symptom/complication. These results support the development of machine learning algorithms to predict onset of abnormal symptoms and complications in children undergoing surgery, and the use of consumer wearables as monitoring tools for early detection of postoperative events.

Evidence for positive long- and short-term effects of vaccinations against COVID-19 in wearable sensor metrics

Vaccines are among the most powerful tools to combat the COVID-19 pandemic. They are highly effective against infection and substantially reduce the risk of severe disease, hospitalization, ICU admission, and death. However, their potential for attenuating long-term changes in personal health and health-related wellbeing after a SARS-CoV-2 infection remains a subject of debate. Such effects can be effectively monitored at the individual level by analyzing physiological data collected by consumer-grade wearable sensors. Here, we investigate changes in resting heart rate, daily physical activity, and sleep duration around a SARS-CoV-2 infection stratified by vaccination status. Data were collected over a period of 2 years in the context of the German Corona Data Donation Project with around 190,000 monthly active participants. Compared to their unvaccinated counterparts, we find that vaccinated individuals, on average, experience smaller changes in their vital data that also return to normal levels more quickly. Likewise, extreme changes in vitals during the acute phase of the disease occur less frequently in vaccinated individuals. Our results solidify evidence that vaccines can mitigate long-term detrimental effects of SARS-CoV-2 infections both in terms of duration and magnitude. Furthermore, they demonstrate the value of large-scale, high-resolution wearable sensor data in public health research.

Opportune warning of COVID-19 in a Mexican health care worker cohort: Discrete beta distribution entropy of smartwatch physiological records

We present a statistical study of heart rate, step cadence, and sleep stage registers of health care workers in the Hospital General de México "Dr. Eduardo Liceaga" (HGM), monitored continuously and non-invasively during the COVID-19 contingency from May to October 2020, using the Fitbit Charge 3® Smartwatch device. The HGM-COVID cohort consisted of 115 participants assigned to areas of COVID-19 exposure. We introduce a novel biomarker for an opportune signal for the likelihood of SARS-CoV-2 infection based on the Shannon Entropy of the Discrete Generalized Beta Distribution fit of rank ordered smartwatch registers. Our statistical test indicated infection for 94% of patients confirmed by positive polymer chain reaction (PCR+) test, 47% before the test, and 47% in coincidence. These results required innovative data preprocessing for the definition of a new biomarker index. The statistical method parameters are data-driven, confidence estimates were calibrated based on sensitivity tests using appropriately derived surrogate data as a benchmark. Our surrogate tests can also provide a benchmark for comparing results from other anomaly detection methods (ADMs). Biomarker comparison of the negative Immunoglobulin G Antibody (IgG-) subgroup with the PCR+ subgroup showed a statistically significant difference (p < 0.01, effect size = 1.44). The distribution of the uninfected population had a lower median and less dispersion than the PCR+ population. A retrospective study of our results confirmed that the biomarker index provides an early warning of the likelihood of COVID-19, even several days before the onset of symptoms or the PCR+ test request. The method can be calibrated for the analysis of different SARS-CoV-2 strains, the effect of vaccination, and previous infections. Furthermore, our biomarker screening could be implemented to provide general health profiles for other population sectors based on physiological signals from smartwatch wearable devices.

Feasibility of using intermittent active monitoring of vital signs by smartphone users to predict SARS-CoV-2 PCR positivity

Early detection of highly infectious respiratory diseases, such as COVID-19, can help curb their transmission. Consequently, there is demand for easy-to-use population-based screening tools, such as mobile health applications. Here, we describe a proof-of-concept development of a machine learning classifier for the prediction of a symptomatic respiratory disease, such as COVID-19, using smartphone-collected vital sign measurements. The Fenland App study followed 2199 UK participants that provided measurements of blood oxygen saturation, body temperature, and resting heart rate. Total of 77 positive and 6339 negative SARS-CoV-2 PCR tests were recorded. An optimal classifier to identify these positive cases was selected using an automated hyperparameter optimisation. The optimised model achieved an ROC AUC of 0.695 ± 0.045. The data collection window for determining each participant's vital sign baseline was increased from 4 to 8 or 12 weeks with no significant difference in model performance (F(2) = 0.80, p = 0.472). We demonstrate that 4 weeks of intermittently collected vital sign measurements could be used to predict SARS-CoV-2 PCR positivity, with applicability to other diseases causing similar vital sign changes. This is the first example of an accessible, smartphone-based remote monitoring tool deployable in a public health setting to screen for potential infections.

The WE SENSE study protocol: A controlled, longitudinal clinical trial on the use of wearable sensors for early detection and tracking of viral respiratory tract infections

Background

Viral respiratory tract infections (VRTI) are extremely common. Considering the profound social and economic impact of COVID-19, it is imperative to identify novel mechanisms for early detection and prevention of VRTIs, to prevent future pandemics. Wearable biosensor technology may facilitate this. Early asymptomatic detection of VRTIs could reduce stress on the healthcare system by reducing transmission and decreasing the overall number of cases. The aim of the current study is to define a sensitive set of physiological and immunological signature patterns of VRTI through machine learning (ML) to analyze physiological data collected continuously using wearable vital signs sensors.

Methods

A controlled, prospective longitudinal study with an induced low grade viral challenge, coupled with 12 days of continuous wearable biosensors monitoring surrounding viral induction. We aim to recruit and simulate a low grade VRTI in 60 healthy adults aged 18–59 years via administration of live attenuated influenza vaccine (LAIV). Continuous monitoring with wearable biosensors will include 7 days pre (baseline) and 5 days post LAIV administration, during which vital signs and activity-monitoring biosensors (embedded in a shirt, wristwatch and ring) will continuously monitor physiological and activity parameters. Novel infection detection techniques will be developed based on inflammatory biomarker mapping, PCR testing, and app-based VRTI symptom tracking. Subtle patterns of change will be assessed via ML algorithms developed to analyze large datasets and generate a predictive algorithm.

Conclusion

This study presents an infrastructure to test wearables for the detection of asymptomatic VRTI using multimodal biosensors, based on immune host response signature.

CliniclTrials.govregistration:NCT05290792

Systematic Review of Advanced AI Methods for Improving Healthcare Data Quality in Post COVID-19 Era

At the beginning of the COVID-19 pandemic, there was significant hype about the potential impact of artificial intelligence (AI) tools in combatting COVID-19 on diagnosis, prognosis, or surveillance. However, AI tools have not yet been widely successful. One of the key reason is the COVID-19 pandemic has demanded faster real-time development of AI-driven clinical and health support tools, including rapid data collection, algorithm development, validation, and deployment. However, there was not enough time for proper data quality control. Learning from the hard lessons in COVID-19, we summarize the important health data quality challenges during COVID-19 pandemic such as lack of data standardization, missing data, tabulation errors, and noise and artifact. Then we conduct a systematic investigation of computational methods that address these issues, including emerging novel advanced AI data quality control methods that achieve better data quality outcomes and, in some cases, simplify or automate the data cleaning process. We hope this article can assist healthcare community to improve health data quality going forward with novel AI development.

Historical development of accelerometry measures and methods for physical activity and sedentary behavior research worldwide: A scoping review of observational studies of adults

This scoping review identified observational studies of adults that utilized accelerometry to assess physical activity and sedentary behavior. Key elements on accelerometry data collection were abstracted to describe current practices and completeness of reporting. We searched three databases (PubMed, Web of Science, and SPORTDiscus) on June 1, 2021 for articles published up to that date. We included studies of non-institutionalized adults with an analytic sample size of at least 500. The search returned 5686 unique records. After reviewing 1027 full-text publications, we identified and abstracted accelerometry characteristics on 155 unique observational studies (154 cross-sectional/cohort studies and 1 case control study). The countries with the highest number of studies included the United States, the United Kingdom, and Japan. Fewer studies were identified from the continent of Africa. Five of these studies were distributed donor studies, where participants connected their devices to an application and voluntarily shared data with researchers. Data collection occurred between 1999 to 2019. Most studies used one accelerometer (94.2%), but 8 studies (5.2%) used 2 accelerometers and 1 study (0.6%) used 4 accelerometers. Accelerometers were more commonly worn on the hip (48.4%) as compared to the wrist (22.3%), thigh (5.4%), other locations (14.9%), or not reported (9.0%). Overall, 12.7% of the accelerometers collected raw accelerations and 44.6% were worn for 24 hours/day throughout the collection period. The review identified 155 observational studies of adults that collected accelerometry, utilizing a wide range of accelerometer data processing methods. Researchers inconsistently reported key aspects of the process from collection to analysis, which needs addressing to support accurate comparisons across studies.

Real-world data: a brief review of the methods, applications, challenges and opportunities

Background

The increased adoption of the internet, social media, wearable devices, e-health services, and other technology-driven services in medicine and healthcare has led to the rapid generation of various types of digital data, providing a valuable data source beyond the confines of traditional clinical trials, epidemiological studies, and lab-based experiments.

Methods

We provide a brief overview on the type and sources of real-world data and the common models and approaches to utilize and analyze real-world data. We discuss the challenges and opportunities of using real-world data for evidence-based decision making This review does not aim to be comprehensive or cover all aspects of the intriguing topic on RWD (from both the research and practical perspectives) but serves as a primer and provides useful sources for readers who interested in this topic.

Results and Conclusions

Real-world hold great potential for generating real-world evidence for designing and conducting confirmatory trials and answering questions that may not be addressed otherwise. The voluminosity and complexity of real-world data also call for development of more appropriate, sophisticated, and innovative data processing and analysis techniques while maintaining scientific rigor in research findings, and attentions to data ethics to harness the power of real-world data.

Current and Perspective Sensing Methods for Monkeypox Virus

The outbreak of the monkeypox virus (MPXV) in non-endemic countries is an emerging global health threat and may have an economic impact if proactive actions are not taken. As shown by the COVID-19 pandemic, rapid, accurate, and cost-effective virus detection techniques play a pivotal role in disease diagnosis and control. Considering the sudden multicountry MPXV outbreak, a critical evaluation of the MPXV detection approaches would be a timely addition to the endeavors in progress for MPXV control and prevention. Herein, we evaluate the current MPXV detection methods, discuss their pros and cons, and provide recommended solutions to the problems. We review the traditional and emerging nucleic acid detection approaches, immunodiagnostics, whole-particle detection, and imaging-based MPXV detection techniques. The insights provided in this article will help researchers to develop novel techniques for the diagnosis of MPXV.

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

Mass surveillance testing can help control outbreaks of infectious diseases such as COVID-19. However, diagnostic test shortages are prevalent globally and continue to occur in the US with the onset of new COVID-19 variants and emerging diseases like monkeypox, demonstrating an unprecedented need for improving our current methods for mass surveillance testing. By targeting surveillance testing toward individuals who are most likely to be infected and, thus, increasing the testing positivity rate (i.e., percent positive in the surveillance group), fewer tests are needed to capture the same number of positive cases. Here, we developed an Intelligent Testing Allocation (ITA) method by leveraging data from the CovIdentify study (6765 participants) and the MyPHD study (8580 participants), including smartwatch data from 1265 individuals of whom 126 tested positive for COVID-19. Our rigorous model and parameter search uncovered the optimal time periods and aggregate metrics for monitoring continuous digital biomarkers to increase the positivity rate of COVID-19 diagnostic testing. We found that resting heart rate (RHR) features distinguished between COVID-19-positive and -negative cases earlier in the course of the infection than steps features, as early as 10 and 5 days prior to the diagnostic test, respectively. We also found that including steps features increased the area under the receiver operating characteristic curve (AUC-ROC) by 7-11% when compared with RHR features alone, while including RHR features improved the AUC of the ITA model's precision-recall curve (AUC-PR) by 38-50% when compared with steps features alone. The best AUC-ROC (0.73 ± 0.14 and 0.77 on the cross-validated training set and independent test set, respectively) and AUC-PR (0.55 ± 0.21 and 0.24) were achieved by using data from a single device type (Fitbit) with high-resolution (minute-level) data. Finally, we show that ITA generates up to a 6.5-fold increase in the positivity rate in the cross-validated training set and up to a 4.5-fold increase in the positivity rate in the independent test set, including both symptomatic and asymptomatic (up to 27%) individuals. Our findings suggest that, if deployed on a large scale and without needing self-reported symptoms, the ITA method could improve the allocation of diagnostic testing resources and reduce the burden of test shortages.

Wearable sensor-based detection of influenza in presymptomatic and asymptomatic individuals

BACKGROUND

The COVID-19 pandemic highlighted the need for early detection of viral infections in symptomatic and asymptomatic individuals to allow for timely clinical management and public health interventions.

METHODS

Twenty healthy adults were challenged with an influenza A (H3N2) virus and prospectively monitored from 7 days before through 10 days after inoculation, using wearable electrocardiogram and physical activity sensors (Clinical Trial: NCT04204493; https://clinicaltrials.gov/ct2/show/NCT04204993). This framework allowed for responses to be accurately referenced to the infection event. For each participant, we trained a semi-supervised multivariable anomaly detection model on data acquired before inoculation and used it to classify the post-inoculation dataset.

RESULTS

Inoculation with this challenge virus was well-tolerated with an infection rate of 85%. With the model classification threshold set so that no alarms were recorded in the 170 healthy days recorded, the algorithm correctly identified 16 of 17 (94%) positive presymptomatic and asymptomatic individuals, on average 58 hours post inoculation and 23 hrs before the symptom onset.

CONCLUSION

The data processing and modeling methodology show promise for the early detection of respiratory illness. The detection algorithm is compatible with data collected from smartwatches using optical techniques but needs to be validated in large heterogeneous cohorts in normal living conditions.

Investigation of the use of a sensor bracelet for the presymptomatic detection of changes in physiological parameters related to COVID-19: an interim analysis of a prospective cohort study (COVI-GAPP)

OBJECTIVES

We investigated machinelearningbased identification of presymptomatic COVID-19 and detection of infection-related changes in physiology using a wearable device.

DESIGN

Interim analysis of a prospective cohort study.

SETTING, PARTICIPANTS AND INTERVENTIONS

Participants from a national cohort study in Liechtenstein were included. Nightly they wore the Ava-bracelet that measured respiratory rate (RR), heart rate (HR), HR variability (HRV), wrist-skin temperature (WST) and skin perfusion. SARS-CoV-2 infection was diagnosed by molecular and/or serological assays.

RESULTS

A total of 1.5 million hours of physiological data were recorded from 1163 participants (mean age 44±5.5 years). COVID-19 was confirmed in 127 participants of which, 66 (52%) had worn their device from baseline to symptom onset (SO) and were included in this analysis. Multi-level modelling revealed significant changes in five (RR, HR, HRV, HRV ratio and WST) device-measured physiological parameters during the incubation, presymptomatic, symptomatic and recovery periods of COVID-19 compared with baseline. The training set represented an 8-day long instance extracted from day 10 to day 2 before SO. The training set consisted of 40 days measurements from 66 participants. Based on a random split, the test set included 30% of participants and 70% were selected for the training set. The developed long short-term memory (LSTM) based recurrent neural network (RNN) algorithm had a recall (sensitivity) of 0.73 in the training set and 0.68 in the testing set when detecting COVID-19 up to 2 days prior to SO.

CONCLUSION

Wearable sensor technology can enable COVID-19 detection during the presymptomatic period. Our proposed RNN algorithm identified 68% of COVID-19 positive participants 2 days prior to SO and will be further trained and validated in a randomised, single-blinded, two-period, two-sequence crossover trial. Trial registration number ISRCTN51255782; Pre-results.

Smartphone apps in the COVID-19 pandemic

At the beginning of the COVID-19 pandemic, analog tools such as nasopharyngeal swabs for PCR tests were center stage and the major prevention tactics of masking and physical distancing were a throwback to the 1918 influenza pandemic. Overall, there has been scant regard for digital tools, particularly those based on smartphone apps, which is surprising given the ubiquity of smartphones across the globe. Smartphone apps, given accessibility in the time of physical distancing, were widely used for tracking, tracing and educating the public about COVID-19. Despite limitations, such as concerns around data privacy, data security, digital health illiteracy and structural inequities, there is ample evidence that apps are beneficial for understanding outbreak epidemiology, individual screening and contact tracing. While there were successes and failures in each category, outbreak epidemiology and individual screening were substantially enhanced by the reach of smartphone apps and accessory wearables. Continued use of apps within the digital infrastructure promises to provide an important tool for rigorous investigation of outcomes both in the ongoing outbreak and in future epidemics.

Detection and Monitoring of Viral Infections via Wearable Devices and Biometric Data

Mounting clinical evidence suggests that viral infections can lead to detectable changes in an individual's normal physiologic and behavioral metrics, including heart and respiration rates, heart rate variability, temperature, activity, and sleep prior to symptom onset, potentially even in asymptomatic individuals. While the ability of wearable devices to detect viral infections in a real-world setting has yet to be proven, multiple recent studies have established that individual, continuous data from a range of biometric monitoring technologies can be easily acquired and that through the use of machine learning techniques, physiological signals and warning signs can be identified. In this review, we highlight the existing knowledge base supporting the potential for widespread implementation of biometric data to address existing gaps in the diagnosis and treatment of viral illnesses, with a particular focus on the many important lessons learned from the coronavirus disease 2019 pandemic.

Measuring Health-Related Quality of Life With Multimodal Data: Viewpoint

The ability to objectively measure aspects of performance and behavior is a fundamental pillar of digital health, enabling digital wellness products, decentralized trial concepts, evidence generation, digital therapeutics, and more. Emerging multimodal technologies capable of measuring several modalities simultaneously and efforts to integrate inputs across several sources are further expanding the limits of what digital measures can assess. Experts from the field of digital health were convened as part of a multi-stakeholder workshop to examine the progress of multimodal digital measures in two key areas: detection of disease and the measurement of meaningful aspects of health relevant to the quality of life. Here we present a meeting report, summarizing key discussion points, relevant literature, and finally a vision for the immediate future, including how multimodal measures can provide value to stakeholders across drug development and care delivery, as well as three key areas where headway will need to be made if we are to continue to build on the encouraging progress so far: collaboration and data sharing, removal of barriers to data integration, and alignment around robust modular evaluation of new measurement capabilities.

Estimating the Burden of Influenza-like Illness on Daily Activity at the Population Scale Using Commercial Wearable Sensors

IMPORTANCE

The severity of viral infections can vary widely, from asymptomatic cases to complications leading to hospitalizations and death. Milder cases, despite being more prevalent, often go undocumented, and their public health burden is not accurately estimated.

OBJECTIVE

To estimate the true burden of influenza-like illness (ILI) in the US population using a surrogate measure of daily steps lost as measured by commercial wearable sensors.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study modeled data from 15 122 US adults who reported ILI symptoms during the 2018-2019 influenza season (before the COVID-19 pandemic) and who had a sufficient density of wearable sensor data at symptom onset. Participants' minute-level step data as measured by commercial wearable sensors were collected from October 1, 2018, through June 30, 2019. Minute-level activity time series were transformed into day-level time series per user, indicating the total number of steps daily.

MAIN OUTCOMES AND MEASURES

The primary end point was the number of steps lost during the period of 4 days before symptom onset (the latent phase) through 11 days after symptom onset (the symptomatic phase). The association between covariates and steps lost during this interval was also examined.

RESULTS

Of the 15 122 participants in this study, 13 108 (86.7%) were women, and the median age was 32 years (IQR, 27-38 years). For their ILI event, 2836 of 15 080 participants (18.8%) sought medical attention, and only 61 (0.4%) were hospitalized. Over the course of an ILI lasting 10 days, the mean cumulative loss was 4437 steps (95% CI, 4143-4731 steps). After weighting, there was an estimated overall nationwide reduction in mobility equivalent to 255.2 billion steps (95% CI, 232.9-277.6 billion steps) lost because of ILI symptoms during the study period. This finding reflects significant changes in routines, mobility, and employment and is equivalent to 15% of the active US population becoming completely immobilized for 1 day. Moreover, 60.6% of this reduction in steps (154.6 billion steps [95% CI, 138.1-171.2 billion steps]) occurred among persons who sought no medical care. Age and educational level were positively associated with steps lost.

CONCLUSIONS AND RELEVANCE

These findings suggest that most of the burden of ILI in this study would have been invisible to health care and public health reporting systems. This approach has applications for public health, health care, and clinical research, from estimating costs of lost productivity at population scale, to measuring effectiveness of anti-ILI treatments, to monitoring recovery after acute viral syndromes such as during long COVID-19.

A Micro-Level Analysis of Physiological Responses to COVID-19: Continuous Monitoring of Pregnant Women in California

Continuous monitoring of perinatal women in a descriptive case study allowed us the opportunity to examine the time during which the COVID-19 infection led to physiological changes in two low-income pregnant women. An important component of this study was the use of a wearable sensor device, the Oura ring, to monitor and record vital physiological parameters during sleep. Two women in their second and third trimesters, respectively, were selected based on a positive COVID-19 diagnosis. Both women were tested using the polymerase chain reaction method to confirm the presence of the virus during which time we were able to collect these physiological data. In both cases, we observed 3-6 days of peak physiological changes in resting heart rate (HR), heart rate variability (HRV), and respiratory rate (RR), as well as sleep surrounding the onset of COVID-19 symptoms. The pregnant woman in her third trimester showed a significant increase in resting HR (p = 0.006) and RR (p = 0.048), and a significant decrease in HRV (p = 0.027) and deep sleep duration (p = 0.029). She reported experiencing moderate COVID-19 symptoms and did not require hospitalization. At 38 weeks of gestation, she had a normal delivery and gave birth to a healthy infant. The participant in her second trimester showed similar physiological changes during the 3-day peak period. Importantly, these changes appeared to return to the pre-peak levels. Common symptoms reported by both cases included loss of smell and nasal congestion, with one losing her sense of taste. Results suggest the potential to use the changes in cardiorespiratory responses and sleep for real-time monitoring of health and well-being during pregnancy.

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.

•

We separate wearable heart rate into cardiopulmonary, circadian, and other signals

•

Parameters from different physiological systems enable disease tracking

•

Individual signals change in distinct ways around COVID-19 symptom onset

•

Together, the parameter changes can distinguish healthy from infection periods

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

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.

Fitbeat: COVID-19 estimation based on wristband heart rate using a contrastive convolutional auto-encoder

This study proposes a contrastive convolutional auto-encoder (contrastive CAE), a combined architecture of an auto-encoder and contrastive loss, to identify individuals with suspected COVID-19 infection using heart-rate data from participants with multiple sclerosis (MS) in the ongoing RADAR-CNS mHealth research project. Heart-rate data was remotely collected using a Fitbit wristband. COVID-19 infection was either confirmed through a positive swab test, or inferred through a self-reported set of recognised symptoms of the virus. The contrastive CAE outperforms a conventional convolutional neural network (CNN), a long short-term memory (LSTM) model, and a convolutional auto-encoder without contrastive loss (CAE). On a test set of 19 participants with MS with reported symptoms of COVID-19, each one paired with a participant with MS with no COVID-19 symptoms, the contrastive CAE achieves an unweighted average recall of 95.3 % , a sensitivity of 100 % and a specificity of 90.6 % , an area under the receiver operating characteristic curve (AUC-ROC) of 0.944, indicating a maximum successful detection of symptoms in the given heart rate measurement period, whilst at the same time keeping a low false alarm rate.

Flu@home: The comparative accuracy of an at-home influenza rapid diagnostic test, using a pre-positioned test kit, mobile app, mail-in reference sample, and symptom-based testing trigger

At-home testing with rapid diagnostic tests (RDTs) for respiratory viruses could facilitate early diagnosis, guide patient care, and prevent transmission. Such RDTs are best used near the onset of illness when viral load is highest and clinical action will be most impactful, which may be achieved by at-home testing. We evaluated the diagnostic accuracy of the QuickVue Influenza A + B RDT in an at-home setting. A convenience sample of 5,229 individuals who were engaged with an on-line health research platform were prospectively recruited throughout the United States. "flu@home" test kits containing a QuickVue RDT and reference sample collection and shipping materials were pre-positioned with participants at the beginning of the study. Participants responded to daily symptom surveys. If they reported experiencing cough along with aches, fever, chills, and/or sweats, they used their flu@home kit following instructions on a mobile app and indicated what lines they saw on the RDT. Of the 976 participants who met criteria to use their self-collection kit and completed study procedures, 202 (20.7%) were positive for influenza by qPCR. The RDT had a sensitivity of 28% (95% CI: 21-36) and specificity of 99% (98-99) for influenza A, and 32% (95% CI: 20-46) and 99% (95% CI: 98-99), for influenza B. Our results support the concept of app-supported, pre-positioned at-home RDT kits using symptom-based triggers, although it cannot be recommended with the RDT used in this study. Further research is needed to determine ways to improve the accuracy and utility of home-based testing for influenza.

Wearable Devices, Smartphones, and Interpretable Artificial Intelligence in Combating COVID-19

Physiological measures, such as heart rate variability (HRV) and beats per minute (BPM), can be powerful health indicators of respiratory infections. HRV and BPM can be acquired through widely available wrist-worn biometric wearables and smartphones. Successive abnormal changes in these indicators could potentially be an early sign of respiratory infections such as COVID-19. Thus, wearables and smartphones should play a significant role in combating COVID-19 through the early detection supported by other contextual data and artificial intelligence (AI) techniques. In this paper, we investigate the role of the heart measurements (i.e., HRV and BPM) collected from wearables and smartphones in demonstrating early onsets of the inflammatory response to the COVID-19. The AI framework consists of two blocks: an interpretable prediction model to classify the HRV measurements status (as normal or affected by inflammation) and a recurrent neural network (RNN) to analyze users' daily status (i.e., textual logs in a mobile application). Both classification decisions are integrated to generate the final decision as either "potentially COVID-19 infected" or "no evident signs of infection". We used a publicly available dataset, which comprises 186 patients with more than 3200 HRV readings and numerous user textual logs. The first evaluation of the approach showed an accuracy of 83.34 ± 1.68% with 0.91, 0.88, 0.89 precision, recall, and F1-Score, respectively, in predicting the infection two days before the onset of the symptoms supported by a model interpretation using the local interpretable model-agnostic explanations (LIME).

Passive detection of COVID-19 with wearable sensors and explainable machine learning algorithms

Individual smartwatch or fitness band sensor data in the setting of COVID-19 has shown promise to identify symptomatic and pre-symptomatic infection or the need for hospitalization, correlations between peripheral temperature and self-reported fever, and an association between changes in heart-rate-variability and infection. In our study, a total of 38,911 individuals (61% female, 15% over 65) have been enrolled between March 25, 2020 and April 3, 2021, with 1118 reported testing positive and 7032 negative for COVID-19 by nasopharyngeal PCR swab test. We propose an explainable gradient boosting prediction model based on decision trees for the detection of COVID-19 infection that can adapt to the absence of self-reported symptoms and to the available sensor data, and that can explain the importance of each feature and the post-test-behavior for the individuals. We tested it in a cohort of symptomatic individuals who exhibited an AUC of 0.83 [0.81-0.85], or AUC = 0.78 [0.75-0.80] when considering only data before the test date, outperforming state-of-the-art algorithm in these conditions. The analysis of all individuals (including asymptomatic and pre-symptomatic) when self-reported symptoms were excluded provided an AUC of 0.78 [0.76-0.79], or AUC of 0.70 [0.69-0.72] when considering only data before the test date. Extending the use of predictive algorithms for detection of COVID-19 infection based only on passively monitored data from any device, we showed that it is possible to scale up this platform and apply the algorithm in other settings where self-reported symptoms can not be collected.

Assessment of the Feasibility of Using Noninvasive Wearable Biometric Monitoring Sensors to Detect Influenza and the Common Cold Before Symptom Onset

IMPORTANCE

Currently, there are no presymptomatic screening methods to identify individuals infected with a respiratory virus to prevent disease spread and to predict their trajectory for resource allocation.

OBJECTIVE

To evaluate the feasibility of using noninvasive, wrist-worn wearable biometric monitoring sensors to detect presymptomatic viral infection after exposure and predict infection severity in patients exposed to H1N1 influenza or human rhinovirus.

DESIGN, SETTING, AND PARTICIPANTS

The cohort H1N1 viral challenge study was conducted during 2018; data were collected from September 11, 2017, to May 4, 2018. The cohort rhinovirus challenge study was conducted during 2015; data were collected from September 14 to 21, 2015. A total of 39 adult participants were recruited for the H1N1 challenge study, and 24 adult participants were recruited for the rhinovirus challenge study. Exclusion criteria for both challenges included chronic respiratory illness and high levels of serum antibodies. Participants in the H1N1 challenge study were isolated in a clinic for a minimum of 8 days after inoculation. The rhinovirus challenge took place on a college campus, and participants were not isolated.

EXPOSURES

Participants in the H1N1 challenge study were inoculated via intranasal drops of diluted influenza A/California/03/09 (H1N1) virus with a mean count of 106 using the median tissue culture infectious dose (TCID50) assay. Participants in the rhinovirus challenge study were inoculated via intranasal drops of diluted human rhinovirus strain type 16 with a count of 100 using the TCID50 assay.

MAIN OUTCOMES AND MEASURES

The primary outcome measures included cross-validated performance metrics of random forest models to screen for presymptomatic infection and predict infection severity, including accuracy, precision, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC).

RESULTS

A total of 31 participants with H1N1 (24 men [77.4%]; mean [SD] age, 34.7 [12.3] years) and 18 participants with rhinovirus (11 men [61.1%]; mean [SD] age, 21.7 [3.1] years) were included in the analysis after data preprocessing. Separate H1N1 and rhinovirus detection models, using only data on wearble devices as input, were able to distinguish between infection and noninfection with accuracies of up to 92% for H1N1 (90% precision, 90% sensitivity, 93% specificity, and 90% F1 score, 0.85 [95% CI, 0.70-1.00] AUC) and 88% for rhinovirus (100% precision, 78% sensitivity, 100% specificity, 88% F1 score, and 0.96 [95% CI, 0.85-1.00] AUC). The infection severity prediction model was able to distinguish between mild and moderate infection 24 hours prior to symptom onset with an accuracy of 90% for H1N1 (88% precision, 88% sensitivity, 92% specificity, 88% F1 score, and 0.88 [95% CI, 0.72-1.00] AUC) and 89% for rhinovirus (100% precision, 75% sensitivity, 100% specificity, 86% F1 score, and 0.95 [95% CI, 0.79-1.00] AUC).

CONCLUSIONS AND RELEVANCE

This cohort study suggests that the use of a noninvasive, wrist-worn wearable device to predict an individual's response to viral exposure prior to symptoms is feasible. Harnessing this technology would support early interventions to limit presymptomatic spread of viral respiratory infections, which is timely in the era of COVID-19.

Synthetic data in machine learning for medicine and healthcare

The proliferation of synthetic data in artificial intelligence for medicine and healthcare raises concerns about the vulnerabilities of the software and the challenges of current policy.