ABSTRACT
OBJECTIVE: Scalable strategies to reduce the time burden and increase contact tracing efficiency are crucial during early waves and peaks of infectious transmission. DESIGN: We enrolled a cohort of SARS-CoV-2-positive seed cases into a peer recruitment study testing social network methodology and a novel electronic platform to increase contact tracing efficiency. SETTING: Index cases were recruited from an academic medical center and requested to recruit their local social contacts for enrollment and SARS-CoV-2 testing. PARTICIPANTS: A total of 509 adult participants enrolled over 19 months (384 seed cases and 125 social peers). INTERVENTION: Participants completed a survey and were then eligible to recruit their social contacts with unique "coupons" for enrollment. Peer participants were eligible for SARS-CoV-2 and respiratory pathogen screening. MAIN OUTCOME MEASURES: The main outcome measures were the percentage of tests administered through the study that identified new SARS-CoV-2 cases, the feasibility of deploying the platform and the peer recruitment strategy, the perceived acceptability of the platform and the peer recruitment strategy, and the scalability of both during pandemic peaks. RESULTS: After development and deployment, few human resources were needed to maintain the platform and enroll participants, regardless of peaks. Platform acceptability was high. Percent positivity tracked with other testing programs in the area. CONCLUSIONS: An electronic platform may be a suitable tool to augment public health contact tracing activities by allowing participants to select an online platform for contact tracing rather than sitting for an interview.
Subject(s)
COVID-19 , Adult , Humans , COVID-19/epidemiology , COVID-19/prevention & control , Public Health , COVID-19 Testing , SARS-CoV-2 , Contact Tracing/methodsABSTRACT
A critical challenge to using longitudinal wearable sensor biosignal data for healthcare applications and digital biomarker development is the exacerbation of the healthcare "data deluge," leading to new data storage and organization challenges and costs. Data aggregation, sampling rate minimization, and effective data compression are all methods for consolidating wearable sensor data to reduce data volumes. There has been limited research on appropriate, effective, and efficient data compression methods for biosignal data. Here, we examine the application of different data compression pipelines built using combinations of algorithmic- and encoding-based methods to biosignal data from wearable sensors and explore how these implementations affect data recoverability and storage footprint. Algorithmic methods tested include singular value decomposition, the discrete cosine transform, and the biorthogonal discrete wavelet transform. Encoding methods tested include run-length encoding and Huffman encoding. We apply these methods to common wearable sensor data, including electrocardiogram (ECG), photoplethysmography (PPG), accelerometry, electrodermal activity (EDA), and skin temperature measurements. Of the methods examined in this study and in line with the characteristics of the different data types, we recommend direct data compression with Huffman encoding for ECG, and PPG, singular value decomposition with Huffman encoding for EDA and accelerometry, and the biorthogonal discrete wavelet transform with Huffman encoding for skin temperature to maximize data recoverability after compression. We also report the best methods for maximizing the compression ratio. Finally, we develop and document open-source code and data for each compression method tested here, which can be accessed through the Digital Biomarker Discovery Pipeline as the "Biosignal Data Compression Toolbox," an open-source, accessible software platform for compressing biosignal data.
Subject(s)
Data Compression , Algorithms , Electrocardiography , Photoplethysmography , Signal Processing, Computer-Assisted , Wavelet AnalysisABSTRACT
Smart portable devices- smartphones and smartwatches- are rapidly being adopted by the general population, which has brought forward an opportunity to use the large volumes of physiological, behavioral, and activity data continuously being collected by these devices in naturalistic settings to perform research, monitor health, and track disease. While these data can serve to revolutionize health monitoring in research and clinical care, minimal research has been conducted to understand what motivates people to use these devices and their interest and comfort in sharing the data. In this study, we aimed to characterize the ownership and usage of smart devices among patients from an expansive academic health system in the southeastern US and understand their willingness to share data collected by the smart devices. We conducted an electronic survey of participants from an online patient advisory group around smart device ownership, usage, and data sharing. Out of the 3021 members of the online patient advisory group, 1368 (45%) responded to the survey, with 871 female (64%), 826 and 390 White (60%) and Black (29%) participants, respectively, and a slight majority (52%) age 58 and older. Most of the respondents (98%) owned a smartphone and the majority (59%) owned a wearable. In this population, people who identify as female, Hispanic, and Generation Z (age 18-25), and those completing higher education and having full-time employment, were most likely to own a wearable device compared to their demographic counterparts. 50% of smart device owners were willing to share and 32% would consider sharing their smart device data for research purposes. The type of activity data they are willing to share varies by gender, age, education, and employment. Findings from this study can be used to design both equitable and cost-effective digital health studies, leveraging personally-owned smartphones and wearables in representative populations, ultimately enabling the development of equitable digital health technologies.
ABSTRACT
Recent advances in wearable technology through convenient and cuffless systems will enable continuous, noninvasive monitoring of blood pressure (BP), heart rate, and heart rhythm on both longitudinal 24-hour measurement scales and high-frequency beat-to-beat BP variability and synchronous heart rate variability and changes in underlying heart rhythm. Clinically, BP variability is classified into 4 main types on the basis of the duration of monitoring time: very-short-term (beat to beat), short-term (within 24 hours), medium-term (within days), and long-term (over months and years). BP variability is a strong risk factor for cardiovascular diseases, chronic kidney disease, cognitive decline, and mental illness. The diagnostic and therapeutic value of measuring and controlling BP variability may offer critical targets in addition to lowering mean BP in hypertensive populations.
Subject(s)
Cardiovascular Diseases , Hypertension , Humans , Blood Pressure/physiology , Blood Pressure Monitoring, Ambulatory , Cardiovascular Diseases/diagnosis , Risk Factors , Blood Pressure DeterminationABSTRACT
Advancements in AI enable personalizing healthcare, for example by investigating disease origins at the genetic or molecular level, understanding intraindividual drug effects, and fusing multi-modal personal physiological, behavioral, laboratory, and clinical data to uncover new aspects of pathophysiology. Future efforts should address equity, fairness, explainability, and generalizability of AI models.
Subject(s)
Medicine , Artificial IntelligenceABSTRACT
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.
ABSTRACT
BACKGROUND: The field of digital medicine has seen rapid growth over the past decade. With this unfettered growth, challenges surrounding interoperability have emerged as a critical barrier to translating digital medicine into practice. In order to understand how to mitigate challenges in digital medicine research and practice, this community must understand the landscape of digital medicine professionals, which digital medicine tools are being used and how, and user perspectives on current challenges in the field of digital medicine. OBJECTIVE: The primary objective of this study is to provide information to the digital medicine community that is working to establish frameworks and best practices for interoperability in digital medicine. We sought to learn about the background of digital medicine professionals and determine which sensors and file types are being used most commonly in digital medicine research. We also sought to understand perspectives on digital medicine interoperability. METHODS: We used a web-based survey to query a total of 56 digital medicine professionals from May 1, 2020, to July 10, 2020, on their educational and work experience, the sensors, file types, and toolkits they use professionally, and their perspectives on interoperability in digital medicine. RESULTS: We determined that the digital medicine community comes from diverse educational backgrounds and uses a variety of sensors and file types. Sensors measuring physical activity and the cardiovascular system are the most frequently used, and smartphones continue to be the dominant source of digital health information collection in the digital medicine community. We show that there is not a general consensus on file types in digital medicine, and data are currently handled in multiple ways. There is consensus that interoperability is a critical impediment in digital medicine, with 93% (52) of survey respondents in agreement. However, only 36% (20) of respondents currently use tools for interoperability in digital medicine. We identified three key interoperability needs to be met: integration with electronic health records, implementation of standard data schemas, and standard and verifiable methods for digital medicine research. We show that digital medicine professionals are eager to adopt new tools to solve interoperability problems, and we suggest tools to support digital medicine interoperability. CONCLUSIONS: Understanding the digital medicine community, the sensors and file types they use, and their perspectives on interoperability will enable the development and implementation of solutions that fill critical interoperability gaps in digital medicine. The challenges to interoperability outlined by this study will drive the next steps in creating an interoperable digital medicine community. Establishing best practices to address these challenges and employing platforms for digital medicine interoperability will be essential to furthering the field of digital medicine.
Subject(s)
Electronic Health Records , Smartphone , Humans , Surveys and QuestionnairesABSTRACT
INTRODUCTION: Diabetes prevalence continues to grow and there remains a significant diagnostic gap in one-third of the US population that has pre-diabetes. Innovative, practical strategies to improve monitoring of glycemic health are desperately needed. In this proof-of-concept study, we explore the relationship between non-invasive wearables and glycemic metrics and demonstrate the feasibility of using non-invasive wearables to estimate glycemic metrics, including hemoglobin A1c (HbA1c) and glucose variability metrics. RESEARCH DESIGN AND METHODS: We recorded over 25 000 measurements from a continuous glucose monitor (CGM) with simultaneous wrist-worn wearable (skin temperature, electrodermal activity, heart rate, and accelerometry sensors) data over 8-10 days in 16 participants with normal glycemic state and pre-diabetes (HbA1c 5.2-6.4). We used data from the wearable to develop machine learning models to predict HbA1c recorded on day 0 and glucose variability calculated from the CGM. We tested the accuracy of the HbA1c model on a retrospective, external validation cohort of 10 additional participants and compared results against CGM-based HbA1c estimation models. RESULTS: A total of 250 days of data from 26 participants were collected. Out of the 27 models of glucose variability metrics that we developed using non-invasive wearables, 11 of the models achieved high accuracy (<10% mean average per cent error, MAPE). Our HbA1c estimation model using non-invasive wearables data achieved MAPE of 5.1% on an external validation cohort. The ranking of wearable sensor's importance in estimating HbA1c was skin temperature (33%), electrodermal activity (28%), accelerometry (25%), and heart rate (14%). CONCLUSIONS: This study demonstrates the feasibility of using non-invasive wearables to estimate glucose variability metrics and HbA1c for glycemic monitoring and investigates the relationship between non-invasive wearables and the glycemic metrics of glucose variability and HbA1c. The methods used in this study can be used to inform future studies confirming the results of this proof-of-concept study.
Subject(s)
Prediabetic State , Wearable Electronic Devices , Blood Glucose , Glucose , Glycated Hemoglobin/analysis , Humans , Retrospective StudiesABSTRACT
Prediabetes affects one in three people and has a 10% annual conversion rate to type 2 diabetes without lifestyle or medical interventions. Management of glycemic health is essential to prevent progression to type 2 diabetes. However, there is currently no commercially-available and noninvasive method for monitoring glycemic health to aid in self-management of prediabetes. There is a critical need for innovative, practical strategies to improve monitoring and management of glycemic health. In this study, using a dataset of 25,000 simultaneous interstitial glucose and noninvasive wearable smartwatch measurements, we demonstrated the feasibility of using noninvasive and widely accessible methods, including smartwatches and food logs recorded over 10 days, to continuously detect personalized glucose deviations and to predict the exact interstitial glucose value in real time with up to 84% and 87% accuracy, respectively. We also establish methods for designing variables using data-driven and domain-driven methods from noninvasive wearables toward interstitial glucose prediction.
ABSTRACT
Recently, companies such as Apple Inc, Fitbit Inc, and Garmin Ltd have released new wearable blood oxygenation measurement technologies. Although the release of these technologies has great potential for generating health-related information, it is important to acknowledge the repercussions of consumer-targeted biometric monitoring technologies (BioMeTs), which in practice, are often used for medical decision making. BioMeTs are bodily connected digital medicine products that process data captured by mobile sensors that use algorithms to generate measures of behavioral and physiological function. These BioMeTs span both general wellness products and medical devices, and consumer-targeted BioMeTs intended for general wellness purposes are not required to undergo a standardized and transparent evaluation process for ensuring their quality and accuracy. The combination of product functionality, marketing, and the circumstances of the global SARS-CoV-2 pandemic have inevitably led to the use of consumer-targeted BioMeTs for reporting health-related measurements to drive medical decision making. In this viewpoint, we urge consumer-targeted BioMeT manufacturers to go beyond the bare minimum requirements described in US Food and Drug Administration guidance when releasing information on wellness BioMeTs. We also explore new methods and incentive systems that may result in a clearer public understanding of the performance and intended use of consumer-targeted BioMeTs.
Subject(s)
Fitness Trackers/trends , Pandemics , Wearable Electronic Devices/standards , COVID-19 , Humans , Wearable Electronic Devices/adverse effectsABSTRACT
INTRODUCTION: Personalized medicine has exposed wearable sensors as new sources of biomedical data which are expected to accrue annual data storage costs of approximately $7.2 trillion by 2020 (>2000 exabytes). To improve the usability of wearable devices in healthcare, it is necessary to determine the minimum amount of data needed for accurate health assessment. METHODS: Here, we present a generalizable optimization framework for determining the minimum necessary sampling rate for wearable sensors and apply our method to determine optimal optical blood volume pulse sampling rate. We implement t-tests, Bland-Altman analysis, and regression-based visualizations to identify optimal sampling rates of wrist-worn optical sensors. RESULTS: We determine the optimal sampling rate of wrist-worn optical sensors for heart rate and heart rate variability monitoring to be 21-64 Hz, depending on the metric. CONCLUSIONS: Determining the optimal sampling rate allows us to compress biomedical data and reduce storage needs and financial costs. We have used optical heart rate sensors as a case study for the connection between data volumes and resource requirements to develop methodology for determining the optimal sampling rate for clinical relevance that minimizes resource utilization. This methodology is extensible to other wearable sensors.
ABSTRACT
As wearable technologies are being increasingly used for clinical research and healthcare, it is critical to understand their accuracy and determine how measurement errors may affect research conclusions and impact healthcare decision-making. Accuracy of wearable technologies has been a hotly debated topic in both the research and popular science literature. Currently, wearable technology companies are responsible for assessing and reporting the accuracy of their products, but little information about the evaluation method is made publicly available. Heart rate measurements from wearables are derived from photoplethysmography (PPG), an optical method for measuring changes in blood volume under the skin. Potential inaccuracies in PPG stem from three major areas, includes (1) diverse skin types, (2) motion artifacts, and (3) signal crossover. To date, no study has systematically explored the accuracy of wearables across the full range of skin tones. Here, we explored heart rate and PPG data from consumer- and research-grade wearables under multiple circumstances to test whether and to what extent these inaccuracies exist. We saw no statistically significant difference in accuracy across skin tones, but we saw significant differences between devices, and between activity types, notably, that absolute error during activity was, on average, 30% higher than during rest. Our conclusions indicate that different wearables are all reasonably accurate at resting and prolonged elevated heart rate, but that differences exist between devices in responding to changes in activity. This has implications for researchers, clinicians, and consumers in drawing study conclusions, combining study results, and making health-related decisions using these devices.