Wrist-worn Wearables Research Articles

Monitoring Sleep and Nightly Recovery with Wrist-Worn Wearables: Links to Training Load and Performance Adaptations.

Previous studies on the effects of intensified training on sleep quality/quantity have been somewhat contradictory. Moreover, recreational athletes often track various sleep metrics, and those metrics' actual connections to training adaptations are unknown. This study explored the effects of intensified training on sleep and nightly recovery along with their associations with training adaptations. A total of 24 participants (10 females) performed a 3-week baseline training period (BL), a 2-week overload period (OL), and a 1-week recovery period (REC), which were followed by test days (T1-T3). The endurance performance was assessed with a 3000 m running test. Throughout all of the periods, the nightly recovery information was monitored with a wrist-worn wearable, including sleep quantity and quality, heart rate (HR) and HR variability (HRV), and proprietary parameters combining several parameters and scaling the results individually. In addition, the perceived strain and muscle soreness were evaluated daily. The 3000 m running performance improved from T1 to T2 (-1.2 ± 1.7%, p = 0.006) and from T1 to T3 (-1.7 ± 1.2%, p = 0.002). The perceived strain and muscle soreness increased (p < 0.001) from the final week of the BL to the final week of the OL, but the subjective sleep quality and nightly recovery metrics remained unchanged. The OL average of the proprietary parameter, autonomic nervous system charge ("ANS charge", combining the HR, HRV, and breathing rate), as well as the change in the sleep HR and HRV from the BL to the OL, were associated (p < 0.05) with a change in the 3000 m running time. In conclusion, the subjective recovery metrics were impaired by intensified training, while the sleep and nightly recovery metrics showed no consistent changes. However, there were substantial interindividual differences in nightly recovery, which were also associated with the training adaptations. Therefore, monitoring nightly recovery can help in recognizing individual responses to training and assist in optimizing training prescriptions.

Exploring the Applications of Explainability in Wearable Data Analytics: Systematic Literature Review.

Wearable technologies have become increasingly prominent in health care. However, intricate machine learning and deep learning algorithms often lead to the development of "black box" models, which lack transparency and comprehensibility for medical professionals and end users. In this context, the integration of explainable artificial intelligence (XAI) has emerged as a crucial solution. By providing insights into the inner workings of complex algorithms, XAI aims to foster trust and empower stakeholders to use wearable technologies responsibly. This paper aims to review the recent literature and explore the application of explainability in wearables. By examining how XAI can enhance the interpretability of generated data and models, this review sought to shed light on the possibilities that arise at the intersection of wearable technologies and XAI. We collected publications from ACM Digital Library, IEEE Xplore, PubMed, SpringerLink, JMIR, Nature, and Scopus. The eligible studies included technology-based research involving wearable devices, sensors, or mobile phones focused on explainability, machine learning, or deep learning and that used quantified self data in medical contexts. Only peer-reviewed articles, proceedings, or book chapters published in English between 2018 and 2022 were considered. We excluded duplicates, reviews, books, workshops, courses, tutorials, and talks. We analyzed 25 research papers to gain insights into the current state of explainability in wearables in the health care context. Our findings revealed that wrist-worn wearables such as Fitbit and Empatica E4 are prevalent in health care applications. However, more emphasis must be placed on making the data generated by these devices explainable. Among various explainability methods, post hoc approaches stand out, with Shapley Additive Explanations as a prominent choice due to its adaptability. The outputs of explainability methods are commonly presented visually, often in the form of graphs or user-friendly reports. Nevertheless, our review highlights a limitation in user evaluation and underscores the importance of involving users in the development process. The integration of XAI into wearable health care technologies is crucial to address the issue of black box models. While wrist-worn wearables are widespread, there is a notable gap in making the data they generate explainable. Post hoc methods such as Shapley Additive Explanations have gained traction for their adaptability in explaining complex algorithms visually. However, user evaluation remains an area in which improvement is needed, and involving users in the development process can contribute to more transparent and reliable artificial intelligence models in health care applications. Further research in this area is essential to enhance the transparency and trustworthiness of artificial intelligence models used in wearable health care technology.

Effects of epileptic seizures on the quality of biosignals recorded from wearables.

Wearable nonelectroencephalographic biosignal recordings captured from the wrist offer enormous potential for seizure monitoring. However, signal quality remains a challenging factor affecting data reliability. Models trained for seizure detection depend on the quality of recordings in peri-ictal periods in performing a feature-based separation of ictal periods from interictal periods. Thus, this study aims to investigate the effect of epileptic seizures on signal quality, ensuring accurate and reliable monitoring. This study assesses the signal quality of wearable data during peri-ictal phases of generalized tonic-clonic and focal to bilateral tonic-clonic seizures (TCS), focal motor seizures (FMS), and focal nonmotor seizures (FNMS). We evaluated accelerometer (ACC) activity and the signal quality of electrodermal activity (EDA) and blood volume pulse (BVP) data. Additionally, we analyzed the influence of peri-ictal movements as assessed by ACC (ACC activity) on signal quality and examined intraictal subphases of focal to bilateral TCS. We analyzed 386 seizures from 111 individuals in three international epilepsy monitoring units. BVP signal quality and ACC activity levels differed between all seizure types. We found the largest decrease in BVP signal quality and increase in ACC activity when comparing the ictal phase to the pre- and postictal phases for TCS. Additionally, ACC activity was strongly negatively correlated with BVP signal quality for TCS and FMS, and weakly for FNMS. Intraictal analysis revealed that tonic and clonic subphases have the lowest BVP signal quality and the highest ACC activity. Motor elements of seizures significantly impair BVP signal quality, but do not have significant effect on EDA signal quality, as assessed by wrist-worn wearables. The results underscore the importance of signal quality assessment methods and careful selection of robust modalities to ensure reliable seizure detection. Future research is needed to explain whether seizure detection models' decisions are based on signal responses induced by physiological processes as opposed to artifacts.

Laughter in everyday life: an event-based experience sampling method study using wrist-worn wearables.

Laughter is a universal, nonverbal vocal expression of broad significance for humans. Interestingly, rather little is known about how often we laugh and how laughter is associated with our personality. In a large, event-based, experience sampling method study (N = 52; k = 9,261 assessments) using wrist-worn wearables and a physical analogue scale, we analyzed belly laughs and fit of laughter events in participants' everyday life for 4 weeks. Additionally, we assessed associations with laughter frequency such as personality, happiness, life satisfaction, gelotophobia (i.e., fear of being laughed at), and cheerfulness. Validating our new measurement approach (i.e., wearables, physical analogue scale), laughter events elicited higher happiness ratings compared to reference assessments, as expected. On average, participants reported 2.5 belly laughs per day and on every fourth day a fit of laughter. As expected, participants who were happier and more satisfied with their life laughed more frequently than unhappier, unsatisfied participants. Women and younger participants laughed significantly more than men and older participants. Regarding personality, laughter frequency was positively associated with openness and conscientiousness. No significant association was found for gelotophobia, and results for cheerfulness and related concepts were mixed. By using state-of-the-art statistical methods (i.e., recurrent event regression) for the event-based, multi-level data on laughter, we could replicate past results on laughing.

A Motivational Technology Perspective on the Use of Smart Wrist-Worn Wearables for Postpartum Exercise and Weight Management

ABSTRACT Exercise and weight management is crucial in preventing postpartum depression and long-term obesity that carries the risk of chronic illness among postpartum women. Although communication devices, such as a smart wrist-worn wearable (SWW), can help users be more physically active, the extent to which postpartum women might benefit from this technology is unknown. We examined how SWWs promoted exercise and helped postpartum women return to pre-pregnancy weight. We tested a model based on the premise that a motivational device that prompts users to engage with it can establish healthy daily routines. An online survey of 309 postpartum women who were living in the United States and were current users of SWWs revealed that the device encouraged them to spend time completing workout goals. Technological affordances (i.e. customization, navigability, and interactivity) and subsequent user engagement with the device positively predicted total workout hours among postpartum women. We present practical implications for postpartum care programs and smart wearable developers.

Predicting cognitive scores from wearable-based digital physiological features using machine learning: data from a clinical trial in mild cognitive impairment

BackgroundContinuous assessment and remote monitoring of cognitive function in individuals with mild cognitive impairment (MCI) enables tracking therapeutic effects and modifying treatment to achieve better clinical outcomes. While standardized neuropsychological tests are inconvenient for this purpose, wearable sensor technology collecting physiological and behavioral data looks promising to provide proxy measures of cognitive function. The objective of this study was to evaluate the predictive ability of digital physiological features, based on sensor data from wrist-worn wearables, in determining neuropsychological test scores in individuals with MCI.MethodsWe used the dataset collected from a 10-week single-arm clinical trial in older adults (50–70 years old) diagnosed with amnestic MCI (N = 30) who received a digitally delivered multidomain therapeutic intervention. Cognitive performance was assessed before and after the intervention using the Neuropsychological Test Battery (NTB) from which composite scores were calculated (executive function, processing speed, immediate memory, delayed memory and global cognition). The Empatica E4, a wrist-wearable medical-grade device, was used to collect physiological data including blood volume pulse, electrodermal activity, and skin temperature. We processed sensors’ data and extracted a range of physiological features. We used interpolated NTB scores for 10-day intervals to test predictability of scores over short periods and to leverage the maximum of wearable data available. In addition, we used individually centered data which represents deviations from personal baselines. Supervised machine learning was used to train models predicting NTB scores from digital physiological features and demographics. Performance was evaluated using “leave-one-subject-out” and “leave-one-interval-out” cross-validation.ResultsThe final sample included 96 aggregated data intervals from 17 individuals. In total, 106 digital physiological features were extracted. We found that physiological features, especially measures of heart rate variability, correlated most strongly to the executive function compared to other cognitive composites. The model predicted the actual executive function scores with correlation r = 0.69 and intra-individual changes in executive function scores with r = 0.61.ConclusionsOur findings demonstrated that wearable-based physiological measures, primarily HRV, have potential to be used for the continuous assessments of cognitive function in individuals with MCI.

Real-Time Machine Learning for Human Activities Recognition Based on Wrist-Worn Wearable Devices

Wearable technologies have slowly invaded our lives and can easily help with our day-to-day tasks. One area where wearable devices can shine is in human activity recognition, as they can gather sensor data in a non-intrusive way. We describe a real-time activity recognition system based on a common wearable device: a smartwatch. This is one of the most inconspicuous devices suitable for activity recognition as it is very common and worn for extensive periods of time. We propose a human activity recognition system that is extensible, due to the wide range of sensing devices that can be integrated, and that provides a flexible deployment system. The machine learning component recognizes activity based on plot images generated from raw sensor data. This service is exposed as a Web API that can be deployed locally or directly in the cloud. The proposed system aims to simplify the human activity recognition process by exposing such capabilities via a web API. This web API can be consumed by small-network-enabled wearable devices, even with basic processing capabilities, by leveraging a simple data contract interface and using raw data. The system replaces extensive pre-processing by leveraging high performance image recognition based on plot images generated from raw sensor data. We have managed to obtain an activity recognition rate of 94.89% and to implement a fully functional real-time human activity recognition system.

A multimodal dataset of real world mobility activities in Parkinson’s disease

Parkinson’s disease (PD) is a neurodegenerative disorder characterised by motor symptoms such as gait dysfunction and postural instability. Technological tools to continuously monitor outcomes could capture the hour-by-hour symptom fluctuations of PD. Development of such tools is hampered by the lack of labelled datasets from home settings. To this end, we propose REMAP (REal-world Mobility Activities in Parkinson’s disease), a human rater-labelled dataset collected in a home-like setting. It includes people with and without PD doing sit-to-stand transitions and turns in gait. These discrete activities are captured from periods of free-living (unobserved, unstructured) and during clinical assessments. The PD participants withheld their dopaminergic medications for a time (causing increased symptoms), so their activities are labelled as being “on” or “off” medications. Accelerometry from wrist-worn wearables and skeleton pose video data is included. We present an open dataset, where the data is coarsened to reduce re-identifiability, and a controlled dataset available on application which contains more refined data. A use-case for the data to estimate sit-to-stand speed and duration is illustrated.

A scoping review of wrist-worn wearables in education

Within the past ten years, the public's use of wearables, such as smartwatches and fitness trackers, has greatly increased. Recently, researchers have published systematic reviews to examine how wearables are used in educational settings. However, not all wearables are alike. Head-mounted wearables and exoskeletons may be viewed as invasive, whereas smartwatches and wristbands are commonly worn by the public. In this scoping review, we systematically review the literature around wrist-worn wearables (WWs) in educational settings. Our systematic search revealed 1,932 unique research items, of which 46 met the inclusion criteria and were analyzed. Our findings reveal the trends in the field, we discuss the vast heterogeneity seen in the field, and we make a call for further in-depth systematic reviews in the areas of WWs as educational interventions, of studies where the data from WWs is used to predict or classify different types of outcomes, and about data privacy, data security, ethics, and legal concerns around WWs in educational settings.

Physical Activity Pattern of Adults With Metabolic Syndrome Risk Factors: Time-Series Cluster Analysis.

Physical activity plays a crucial role in maintaining a healthy lifestyle, and wrist-worn wearables, such as smartwatches and smart bands, have become popular tools for measuring activity levels in daily life. However, studies on physical activity using wearable devices have limitations; for example, these studies often rely on a single device model or use improper clustering methods to analyze the wearable data that are extracted from wearable devices. This study aimed to identify methods suitable for analyzing wearable data and determining daily physical activity patterns. This study also explored the association between these physical activity patterns and health risk factors. People aged >30 years who had metabolic syndrome risk factors and were using their own wrist-worn devices were included in this study. We collected personal health data through a web-based survey and measured physical activity levels using wrist-worn wearables over the course of 1 week. The Time-Series Anytime Density Peak (TADPole) clustering method, which is a novel time-series method proposed recently, was used to identify the physical activity patterns of study participants. Additionally, we defined physical activity pattern groups based on the similarity of physical activity patterns between weekdays and weekends. We used the χ2 or Fisher exact test for categorical variables and the 2-tailed t test for numerical variables to find significant differences between physical activity pattern groups. Logistic regression models were used to analyze the relationship between activity patterns and health risk factors. A total of 47 participants were included in the analysis, generating a total of 329 person-days of data. We identified 2 different types of physical activity patterns (early bird pattern and night owl pattern) for weekdays and weekends. The physical activity levels of early birds were less than that of night owls on both weekdays and weekends. Additionally, participants were categorized into stable and shifting groups based on the similarity of physical activity patterns between weekdays and weekends. The physical activity pattern groups showed significant differences depending on age (P=.004) and daily energy expenditure (P<.001 for weekdays; P=.003 for weekends). Logistic regression analysis revealed a significant association between older age (≥40 y) and shifting physical activity patterns (odds ratio 8.68, 95% CI 1.95-48.85; P=.007). This study overcomes the limitations of previous studies by using various models of wrist-worn wearables and a novel time-series clustering method. Our findings suggested that age significantly influenced physical activity patterns. It also suggests a potential role of the TADPole clustering method in the analysis of large and multidimensional data, such as wearable data.

Diagnostic accuracy of smart gadgets/wearable devices in detecting atrial fibrillation in primary prevention: a meta-analysis

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia and an important risk factor for stroke and heart. Recent technology advances have allowed for heart rhythm monitoring using smart gadgets/wearable devices which can be used for early AF diagnosis. Purpose We performed a systemic review and meta-analysis to assess the accuracy of AF diagnosis by smart gadgets/wearable devices in primary prevention. Methods We systematically searched Medline, Embase and Cochrane electronic databases up to September 22th, 2022 for observational studies of the diagnostic accuracy of smartphone application, wrist-worn wearables and external devices in detecting AF in primary prevention. Studies with 12-lead ECG or single-lead ECG interpreted by a cardiologist as gold standard were included. We calculated the area under the curve (AUC) of the summary receiver operating characteristic curves (SROC) and pooled sensitivities and specificities. Results A total of 28 studies were included enrolling 13463 patients, 57%(95% CI: 55-59%) male with average age of 65.3years (95% CI: 60.1 – 70.4). The pooled prevalence of AF was found to be 13% (95% CI 10 – 15%). In the overall analysis of all devices, the AUC was 0.99 (95% CI: 0.97-0.99)(Figure 1), the sensitivity 93%(95% CI: 89 – 95%) and the specificity 97%(95% CI: 95 – 98%) (Figure 2). Wrist-worn wearables had AUC of 0.97 (95% CI: 0.94-0.99), the sensitivity 83%(95% CI: 52 – 96%) and the specificity 98%(95% CI: 87 – 100%). Smartphone applications had AUC of 0.97 (95% CI: 0.96-0.98), the sensitivity 92%(95% CI: 77 – 98%) and the specificity 95%(95% CI: 91 – 97%). External devices had AUC of 0.99 (95% CI: 0.97-0.99), the sensitivity 94%(95% CI: 91 – 96%) and the specificity 97%(95% CI: 95 – 98%). Single-lead ECG had AUC of 0.99 (95% CI: 0.97- 0.99), the sensitivity 93%(95% CI: 88 – 96%) and the specificity 91%(95% CI: 78 – 97%). photoplethysmography pulse waveform technology (PPG) had AUC of 0.97 (95% CI: 0.96-0.98), the sensitivity 91%(95% CI: 78 – 97%) and the specificity 95%(95% CI: 91 – 97%). Pulse beat interval technology had AUC of 0.99 (95% CI: 0.98-0.99), the sensitivity 95%(95% CI: 90 – 98%) and the specificity 96%(95% CI: 93 – 98%) Conclusions Smartphone application, wrist-worn devices and external devices have excellent diagnostic accuracy in atrial fibrillation diagnosis in primary prevention.

AI-Driven sleep staging from actigraphy and heart rate.

Sleep is an important indicator of a person's health, and its accurate and cost-effective quantification is of great value in healthcare. The gold standard for sleep assessment and the clinical diagnosis of sleep disorders is polysomnography (PSG). However, PSG requires an overnight clinic visit and trained technicians to score the obtained multimodality data. Wrist-worn consumer devices, such as smartwatches, are a promising alternative to PSG because of their small form factor, continuous monitoring capability, and popularity. Unlike PSG, however, wearables-derived data are noisier and far less information-rich because of the fewer number of modalities and less accurate measurements due to their small form factor. Given these challenges, most consumer devices perform two-stage (i.e., sleep-wake) classification, which is inadequate for deep insights into a person's sleep health. The challenging multi-class (three, four, or five-class) staging of sleep using data from wrist-worn wearables remains unresolved. The difference in the data quality between consumer-grade wearables and lab-grade clinical equipment is the motivation behind this study. In this paper, we present an artificial intelligence (AI) technique termed sequence-to-sequence LSTM for automated mobile sleep staging (SLAMSS), which can perform three-class (wake, NREM, REM) and four-class (wake, light, deep, REM) sleep classification from activity (i.e., wrist-accelerometry-derived locomotion) and two coarse heart rate measures-both of which can be reliably obtained from a consumer-grade wrist-wearable device. Our method relies on raw time-series datasets and obviates the need for manual feature selection. We validated our model using actigraphy and coarse heart rate data from two independent study populations: the Multi-Ethnic Study of Atherosclerosis (MESA; N = 808) cohort and the Osteoporotic Fractures in Men (MrOS; N = 817) cohort. SLAMSS achieves an overall accuracy of 79%, weighted F1 score of 0.80, 77% sensitivity, and 89% specificity for three-class sleep staging and an overall accuracy of 70-72%, weighted F1 score of 0.72-0.73, 64-66% sensitivity, and 89-90% specificity for four-class sleep staging in the MESA cohort. It yielded an overall accuracy of 77%, weighted F1 score of 0.77, 74% sensitivity, and 88% specificity for three-class sleep staging and an overall accuracy of 68-69%, weighted F1 score of 0.68-0.69, 60-63% sensitivity, and 88-89% specificity for four-class sleep staging in the MrOS cohort. These results were achieved with feature-poor inputs with a low temporal resolution. In addition, we extended our three-class staging model to an unrelated Apple Watch dataset. Importantly, SLAMSS predicts the duration of each sleep stage with high accuracy. This is especially significant for four-class sleep staging, where deep sleep is severely underrepresented. We show that, by appropriately choosing the loss function to address the inherent class imbalance, our method can accurately estimate deep sleep time (SLAMSS/MESA: 0.61±0.69 hours, PSG/MESA ground truth: 0.60±0.60 hours; SLAMSS/MrOS: 0.53±0.66 hours, PSG/MrOS ground truth: 0.55±0.57 hours;). Deep sleep quality and quantity are vital metrics and early indicators for a number of diseases. Our method, which enables accurate deep sleep estimation from wearables-derived data, is therefore promising for a variety of clinical applications requiring long-term deep sleep monitoring.

Towards Nonintrusive and Secure Mobile Two-Factor Authentication on Wearables

Mobile devices are promising to apply two-factor authentication to improve system security. Existing solutions have certain limits of requiring extra user effort, which might seriously affect user experience and delay authentication time. In this paper, we propose PPGPass, a novel mobile two-factor authentication system, which leverages Photoplethysmography (PPG) sensors available in most wrist-worn wearables. PPGPass simultaneously performs a password/pattern/signature authentication and a physiological-based authentication. To realize both nonintrusive and secure, we design a two-stage algorithm to separate clean heartbeat signals from PPG signals contaminated by motion artifacts so that users do not have to deliberately keep their bodies still. In addition, to deal with noncancelable issues when biometrics are compromised, we design a repeatable and non-invertible method to generate cancelable feature templates as alternative credentials. We leverage the great power of Random Forest and Support Vector Data Description to detect adversaries and verify a user's identity. To the best of our knowledge, PPGPass is the first nonintrusive and secure mobile two-factor authentication based on PPG sensors. Extensive experiments demonstrate that PPGPass can achieve the false acceptance rate of 3.11% and the false recognition rate of 3.71%, which confirms its high effectiveness, security, and usability.

Study protocol for a pilot trial analysing the usability, validity and safety of an interventional health app programme for the structured prehabilitation of patients before major surgical interventions: the PROTEGO MAXIMA trial

IntroductionMajor surgery is associated with a high risk for postoperative complications, leading to an increase in mortality and morbidity, particularly in frail patients with a reduced cardiopulmonary reserve. Prehabilitation, including...

Noninvasive Hypoglycemia Detection in People With Diabetes Using Smartwatch Data.

To develop a noninvasive hypoglycemia detection approach using smartwatch data. We prospectively collected data from two wrist-worn wearables (Garminvivoactive 4S, Empatica E4) and continuous glucose monitoring values in adults with diabetes on insulin treatment. Using these data, we developed a machine learning (ML) approach to detect hypoglycemia (<3.9 mmol/L) noninvasively in unseen individuals and solely based on wearable data. Twenty-two individuals were included in the final analysis (age 54.5 ± 15.2 years, HbA1c 6.9 ± 0.6%, 16 males). Hypoglycemia was detected with an area under the receiver operating characteristic curve of 0.76 ± 0.07 solely based on wearable data. Feature analysis revealed that the ML model associated increased heart rate, decreased heart rate variability, and increased tonic electrodermal activity with hypoglycemia. Our approach may allow for noninvasive hypoglycemia detection using wearables in people with diabetes and thus complement existing methods for hypoglycemia detection and warning.

Understanding the relationship between acceptance of multifunctional health and fitness features of wrist-worn wearables and actual usage

PurposeThe purpose of this study was to identify existing users' acceptance of the multidimensional health and fitness features of wrist-worn wearable devices (WWDs) required for each stage of physical activity (i.e. before, during and after) and examine the relationship between its acceptance (i.e. knowledge acquisition, perceived usefulness and perceived ease of use) and the actual use of its health and fitness attributes.Design/methodology/approachBoth qualitative and quantitative approaches were taken to analyze the relationships. A focus group interview was conducted (N = 9) to design the research model, including the operationalized definition of the study constructs. A questionnaire survey was conducted with respondents in South Korea (N = 480). Partial least squares structural equation modeling via Smart PLS 3.0 was employed to test the hypotheses.FindingsWhen users learned to use fitness functions and perceived them as useful for physical activity without causing any difficulty, they tended to use those functions more, which provided enhanced health benefits in the digitalized interactive environment of WWDs.Originality/valueThis research is one of the first to examine the relationship between the perceived user value of WWDs and their actual usage within a digitalized and interactive environment. The results are expected to offer theoretical insights into how well users accept the health and fitness components of WWDs. Practically, it will build awareness of what makes users adopt and use WWDs, helping practitioners design better health promotions and campaigns associated with WWDs.

New Insights into Stroke from Continuous Passively Collected Temperature and Sleep Data Using Wrist-Worn Wearables.

Actigraphy may provide new insights into clinical outcomes and symptom management of patients through passive, continuous data collection. We used the GENEActiv smartwatch to passively collect actigraphy, wrist temperature, and ambient light data from 27 participants after stroke or probable brain transient ischemic attack (TIA) over 42 periods of device wear. We computed 323 features using established algorithms and proposed 25 novel features to characterize sleep and temperature. We investigated statistical associations between the extracted features and clinical outcomes evaluated using clinically validated questionnaires to gain insight into post-stroke recovery. We subsequently fitted logistic regression models to replicate clinical diagnosis (stroke or TIA) and disability due to stroke. The model generalization performance was assessed using a leave-one-subject-out cross validation method with the selected feature subsets, reporting the area under the curve (AUC). We found that several novel features were strongly correlated (|r|>0.3) with stroke symptoms and mental health measures. Using selected novel features, we obtained an AUC of 0.766 to estimate diagnosis and an AUC of 0.749 to estimate whether disability due to stroke was present. Collectively, these findings suggest that features extracted from the temperature smartwatch sensor may reveal additional clinically useful information over and above existing actigraphy-based features.

1120 DIAGNOSTIC ACCURACY OF SMART GADGETS/WEREABLE DEVICES IN ATRIAL FIBRILLATION DETECTION: A METANALYSIS

Abstract Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia and an important risk factor for stroke and heart. Recent technology advances have allowed for heart rhythm monitoring using smart gadgets/wearable devices which can be used for early AF diagnosis. Hypothesis We performed a systemic review and meta-analysis to assess the accuracy of AF diagnosis by smart gadgets/wearable devices. Methods We systematically searched Medline, Embase and Cochrane electronic databases up to April 15th, 2022 for observational studies of the diagnostic accuracy of smartphone application, wrist-worn wearables and external devices in detecting AF. We calculated the area under the curve (AUC) of the summary receiver operating characteristic curves (SROC) and pooled sensitivities and specificities. Results A total of 79 studies were included enrolling 36903 patients, 66.3% male with average age of 68.3±8 years. In the overall analysis of all devices, the AUC was 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(95% CI: 94–96%), the specificity 96%(95% CI: 96–97%). Wrist-worn wearables had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(9% CI: 92–97%), the specificity 97%(95% CI: 96–98%)(Figure 1A). Smartphone applications had AUC of 0.98 (95% CI: 0.96-0.99), the sensitivity 96%(9% CI: 94–97%), the specificity 96%(95% CI: 93–98%)(Figure 1B). External devices had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(9% CI: 93–97%), the specificity 96%(95% CI: 95–97%)(Figure 1C). Single-lead ECG had AUC of 0.99 (95% CI: 0.98- 1.00), the sensitivity 95%(9% CI: 92–96%), the specificity 96%(95% CI: 95–97%). PPG had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 96%(9% CI: 95–97%), the specificity 97%(95% CI: 95–98%). Conclusions Smartphone application, wrist-worn devices and external devices with PPG and single-lead ECG have excellent diagnostic accuracy in atrial fibrillation diagnosis. Figure 1. Summary Receiver Operating Characteristic curves and Areas Under The Curve of Wrist-worn wearables (A), Smartphone applications (B) and External devices (C).

1123 DIAGNOSTIC ACCURACY OF SMART GADGETS/WEREABLE DEVICES IN ATRIAL FIBRILLATION DETECTION: A METANALYSIS

Abstract Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia and an important risk factor for stroke and heart. Recent technology advances have allowed for heart rhythm monitoring using smart gadgets/wearable devices which can be used for early AF diagnosis. Hypothesis We performed a systemic review and meta-analysis to assess the accuracy of AF diagnosis by smart gadgets/wearable devices. Methods We systematically searched Medline, Embase and Cochrane electronic databases up to April 15th, 2022 for observational studies of the diagnostic accuracy of smartphone application, wrist-worn wearables and external devices in detecting AF. We calculated the area under the curve (AUC) of the summary receiver operating characteristic curves (SROC) and pooled sensitivities and specificities. Results A total of 79 studies were included enrolling 36903 patients, 66.3% male with average age of 68.3±8 years. In the overall analysis of all devices, the AUC was 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(95% CI: 94–96%), the specificity 96%(95% CI: 96–97%). Wrist-worn wearables had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(9% CI: 92–97%), the specificity 97%(95% CI: 96–98%)(Figure 1A). Smartphone applications had AUC of 0.98 (95% CI: 0.96-0.99), the sensitivity 96%(9% CI: 94–97%), the specificity 96%(95% CI: 93–98%)(Figure 1B). External devices had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 95%(9% CI: 93–97%), the specificity 96%(95% CI: 95–97%)(Figure 1C). Single-lead ECG had AUC of 0.99 (95% CI: 0.98- 1.00), the sensitivity 95%(9% CI: 92–96%), the specificity 96%(95% CI: 95–97%). PPG had AUC of 0.99 (95% CI: 0.98-1.00), the sensitivity 96%(9% CI: 95–97%), the specificity 97%(95% CI: 95–98%). Conclusions Smartphone application, wrist-worn devices and external devices with PPG and single-lead ECG have excellent diagnostic accuracy in atrial fibrillation diagnosis. Figure 1. Summary Receiver Operating Characteristic curves and Areas Under The Curve of Wrist-worn wearables (A), Smartphone applications (B) and External devices (C).

User characteristics associated with use of wrist-worn wearables and physical activity apps by adults with and without impaired speech-in-noise recognition: a cross-sectional analysis

Objective To study weekly use of smartwatches, fitness watches and physical activity apps among adults with and without impaired speech-in-noise (SIN) recognition, to identify subgroups of users. Design Cross-sectional study. Study sample Adults (aged 28–80 years) with impaired (n = 384) and normal SIN recognition (n = 341) as measured with a web-based digits-in-noise test, from the Netherlands Longitudinal Study on Hearing. Multiple logistic regression analyses were used to study differences and to build an association model. Results Employed adults in both groups are more likely to use each type of fitness technology (all ORs >3.4, all p-values < 0.004). Specific to fitness watch use, adults living with others use it more (OR 2.5, 95%CI 1.1;5.8, p = 0.033) whereas those abstaining from alcohol (OR 0.3, 95%CI 0.1;0.6) or consuming >2 glasses/week (OR 0.4, 95%CI 0.2;0.81, overall p = 0.006) and hearing aid users (OR 0.5, 95%CI 0.2;0.9, p = 0.024) make less use. Conclusions Subgroups of adults more and less likely to use fitness technology exist, but do not differ between adults with and without impaired SIN recognition. More research is needed to confirm these results and to develop interventions to increase physical activity levels among adults with hearing loss.