Early adverse physiological event detection using commercial wearables: challenges and opportunities DOI Creative Commons

Jesse F. Phipps,

Bryant Passage, Kaan Sel

и другие.

npj Digital Medicine, Год журнала: 2024, Номер 7(1)

Опубликована: Май 23, 2024

Abstract Data from commercial off-the-shelf (COTS) wearables leveraged with machine learning algorithms provide an unprecedented potential for the early detection of adverse physiological events. However, several challenges inhibit this potential, including (1) heterogeneity among and within participants that make scaling to a general population less precise, (2) confounders lead incorrect assumptions regarding participant’s healthy state, (3) noise in data at sensor level limits sensitivity algorithms, (4) imprecision self-reported labels misrepresent true values associated given event. The goal study was two-fold: characterize performance such presence these insights researchers on limitations opportunities, subsequently devise address each challenge offer future opportunities advancement. Our proposed include techniques build determining suitable baselines participant capture important changes label correction as it pertains participant-reported identifiers. work is validated potentially one largest datasets available, obtained 8000+ 1.3+ million hours wearable captured Oura smart rings. Leveraging extensive dataset, we achieve pre-symptomatic COVID-19 receiver operator characteristic (ROC) area under curve (AUC) 0.725 without techniques, 0.739 baseline correction, 0.740 training set, 0.777 both test set. Using same respective paradigms, ROC AUCs 0.919, 0.938, 0.943 0.994 fever, 0.574, 0.611, 0.601, 0.635 shortness breath. These improvements across almost all metrics events, PR AUC, 75% specificity, precision recall. ring allows continuous monitoring event onset, further demonstrate improvement average 3.5 days 4.1 before reported positive result.

Язык: Английский

Accuracy of Three Commercial Wearable Devices for Sleep Tracking in Healthy Adults DOI Creative Commons
Rebecca Robbins, Matthew D. Weaver, Jason P. Sullivan

и другие.

Sensors, Год журнала: 2024, Номер 24(20), С. 6532 - 6532

Опубликована: Окт. 10, 2024

Sleep tracking by consumers is becoming increasingly prevalent; yet, few studies have evaluated the accuracy of such devices. We sought to evaluate three devices (Oura Ring Gen3, Fitbit Sense 2, and Apple Watch Series 8) compared gold standard sleep assessment (polysomnography (PSG)). Thirty-five participants (aged 20-50 years) without a disorder were enrolled in single-night inpatient study, during which they wore Oura Ring, Fitbit, Watch, monitored with PSG. For detecting vs. wake, sensitivity was ≥95% for all discriminating between stages, ranged from 50 86%, as follows: ring 76.0-79.5% precision 77.0-79.5%; 61.7-78.0% 72.8-73.2%; 50.5-86.1% 72.7-87.8%. The not different PSG terms light sleep, deep or REM estimation. overestimated (18 min;

Язык: Английский

Процитировано

8

Performance of wearable sleep trackers during nocturnal sleep and periods of simulated real-world smartphone use DOI Creative Commons
Adrian R. Willoughby, Hosein Aghayan Golkashani, Shohreh Ghorbani

и другие.

Sleep Health, Год журнала: 2024, Номер unknown

Опубликована: Апрель 1, 2024

Goal and aimsTo test sleep/wake transition detection of consumer sleep trackers research-grade actigraphy during nocturnal simulated peri-sleep behavior involving minimal movement.Focus technologyOura Ring Gen 3, Fitbit Sense, AXTRO Fit Xiaomi Mi Band 7, ActiGraph GT9X.Reference technologyPolysomnography.SampleSixty-three participants (36 female) aged 20-68.DesignParticipants engaged in common (reading news articles, watching videos, exchanging texts) on a smartphone before after the period. They were woken up night to complete short questionnaire simulate responding an incoming message.Core analyticsDetection timing accuracy for onset times wake times.Additional analytics exploratory analysesDiscrepancy analysis both including excluding activity periods. Epoch-by-epoch rate extent misclassification periods.Core outcomesOura more accurate at detecting transitions than actigraph lower-priced tracker devices. Detection was less reliable with lower efficiency.Important additional outcomesWith inclusion periods, specificity Kappa improved significantly Oura Fitbit, but not ActiGraph. All devices misclassified motionless as some extent, this prevalent Fitbit.Core conclusionsPerformance is robust nights suboptimal bedtime routines or minor disturbances. Reduced performance low efficiency bolsters concerns that these are fragmented disturbed sleep.

Язык: Английский

Процитировано

7

Sleep Patterns Fluctuate Following Training and Games across the Season in a Semi-Professional, Female Basketball Team DOI Creative Commons
Cody J. Power, Jordan L. Fox, Masaru Teramoto

и другие.

Brain Sciences, Год журнала: 2023, Номер 13(2), С. 238 - 238

Опубликована: Янв. 31, 2023

Quantifying athlete sleep patterns may inform development of optimal training schedules and strategies, considering the competitive challenges faced across season. Therefore, this study comprehensively quantified a female basketball team examined variations in between nights. Seven semi-professional, players had their monitored using wrist-worn activity monitors perceptual ratings during 13-week in-season. Sleep variables were compared different nights (control nights, before games, non-congested game congested nights), generalized linear mixed models, as well Cohen’s d odds ratios effect sizes. Players experienced less on games to control (p < 0.05, = 0.43–0.69). also exhibited later onset times 0.01, 0.68), earlier offset following all other 0.74–0.79). Moreover, attaining better perceived quality was 88% lower than 0.001). While attained an adequate duration (7.3 ± 0.3 h) efficiency (85 2%) average in-season, they susceptible poor games. Although limited team-based case series design, these findings suggest coaches need reconsider scheduling team-based, on-court sessions prior consider implementing suitable psychological recovery strategies around optimize player sleep.

Язык: Английский

Процитировано

16

Performance evaluation of Fitbit Charge 3 and actigraphy vs. polysomnography: Sensitivity, specificity, and reliability across participants and nights DOI
Gal Eylon, Liat Tikotzky, Ilan Dinstein

и другие.

Sleep Health, Год журнала: 2023, Номер 9(4), С. 407 - 416

Опубликована: Июнь 1, 2023

Язык: Английский

Процитировано

15

Belun Ring (Belun Sleep System BLS-100): Deep learning-facilitated wearable enables obstructive sleep apnea detection, apnea severity categorization, and sleep stage classification in patients suspected of obstructive sleep apnea DOI Creative Commons
Z.B. Strumpf,

Wenbo Gu,

Chih-Wei Tsai

и другие.

Sleep Health, Год журнала: 2023, Номер 9(4), С. 430 - 440

Опубликована: Июнь 26, 2023

Our objective was to evaluate the performance of Belun Ring with second-generation deep learning algorithms in obstructive sleep apnea (OSA) detection, OSA severity categorization, and stage classification.Belun REFERENCE TECHNOLOGY: In-lab polysomnography (PSG) SAMPLE: Eighty-four subjects (M: F = 1:1) referred for an overnight study were eligible. Of these, 26% had PSG-AHI<5; 24% PSG-AHI 5-15; 23% 15-30; 27% ≥ 30.Rigorous evaluation by comparing concurrent in-lab PSG using 4% rule.Pearson's correlation coefficient, Student's paired t-test, diagnostic accuracy, sensitivity, specificity, positive predictive value, negative likelihood ratio, Cohen's kappa coefficient (kappa), Bland-Altman plots bias limits agreement, receiver operating characteristics curves area under curve, confusion matrix.The categorizing AHI 5 0.85, 0.92, 0.64, 0.58, respectively. The Kappa 15 0.89, 0.91, 0.88, 0.79, 30 0.83, 0.93, 0.76, BSP2 also achieved accuracy 0.88 detecting wake, 0.82 NREM, 0.90 REM sleep.Belun detected good demonstrated a moderate-to-substantial agreement classifying stages.

Язык: Английский

Процитировано

15

The impact of bedroom environment on sleep quality in winter and summer in the Qinghai-Tibetan plateau DOI
Chao Guo, Li Lan, Haodong Zhang

и другие.

Building and Environment, Год журнала: 2023, Номер 244, С. 110785 - 110785

Опубликована: Сен. 2, 2023

Язык: Английский

Процитировано

14

A proposed multi‐domain, digital model for capturing functional status and health‐related quality of life in oncology DOI Creative Commons
Elena S. Izmailova, John A. Wagner, Jessie P. Bakker

и другие.

Clinical and Translational Science, Год журнала: 2024, Номер 17(1)

Опубликована: Янв. 1, 2024

Whereas traditional oncology clinical trial endpoints remain key for assessing novel treatments, capturing patients' functional status is increasingly recognized as an important aspect supporting decisions and outcomes in trials. Existing assessments suffer from various limitations, some of which may be addressed by adopting digital health technologies (DHTs) a means collecting both objective self-reported outcomes. In this mini-review, we propose device-agnostic multi-domain model status, includes physical activity data, vital signs, sleep variables, measures related to health-related quality life enabled connected tools. By using DHTs all aspects data collection, our proposed allows high-resolution measurement patients navigate their daily lives outside the hospital setting. This complemented electronic questionnaires administered at intervals appropriate each instrument. Preliminary testing practical considerations address before adoption are also discussed. Finally, highlight multi-institutional pre-competitive collaborations successfully transitioning digitally collection feasibility studies interventional trials care management.

Язык: Английский

Процитировано

6

Advancing a U.S. navy shipboard infrastructure for sleep monitoring with wearable technology DOI Creative Commons
Andrew Kubala, Peter G. Roma, Jason Jameson

и другие.

Applied Ergonomics, Год журнала: 2024, Номер 117, С. 104225 - 104225

Опубликована: Янв. 14, 2024

Development of fatigue management solutions is critical to U.S. Navy populations. This study explored the operational feasibility and acceptability commercial wearable devices (Oura Ring ReadiBand) in a warship environment with 845 Sailors across five ship cohorts during at-sea operations ranging from 10 31 days. Participants were required wear both check-in daily research staff. Both functioned as designed reliably collected sleep-wake data. Over 10,000 person-days at-sea, overall prevalence Oura ReadiBand use was 69% 71%, respectively. Individual rates 71 ± 38% days underway for 59 34% ReadiBand. Analysis individual factors showed increasing device less interference age, more men than women found comfortable. provides initial support that wearables can contribute infrastructures naval environments.

Язык: Английский

Процитировано

6

A Protocol for Evaluating Digital Technology for Monitoring Sleep and Circadian Rhythms in Older People and People Living with Dementia in the Community DOI Creative Commons
Ciro della Monica, Kiran K G Ravindran, Giuseppe Atzori

и другие.

Clocks & Sleep, Год журнала: 2024, Номер 6(1), С. 129 - 155

Опубликована: Фев. 29, 2024

Sleep and circadian rhythm disturbance are predictors of poor physical mental health, including dementia. Long-term digital technology-enabled monitoring sleep rhythms in the community has great potential for early diagnosis, disease progression, assessing effectiveness interventions. Before novel technology-based can be implemented at scale, its performance acceptability need to evaluated compared gold-standard methodology relevant populations. Here, we describe our protocol evaluation technology which have applied cognitively intact older adults currently using people living with dementia (PLWD). In this protocol, test a range technologies simultaneously home (7–14 days) subsequently clinical research facility gold standard physiology is implemented. We emphasize importance both nocturnal diurnal (naps), valid markers physiology, that best achieved protocols mildly disturbed populations intended use-case. provide details on design, implementation, challenges, advantages along examples datasets.

Язык: Английский

Процитировано

5

Objective multi-night sleep monitoring at home: variability of sleep parameters between nights and implications for the reliability of sleep assessment in clinical trials DOI

Alexandre Chouraki,

Julia Tournant,

Pierrick J. Arnal

и другие.

SLEEP, Год журнала: 2022, Номер 46(5)

Опубликована: Дек. 30, 2022

Abstract Study Objectives In-laboratory polysomnography is the current gold standard for objective sleep measurements in clinical trials, but this does not capture night-to-night variability parameters. This study analyzed parameters recorded over multiple nights of an ecological setting using a portable monitor and then estimated minimum sample sizes required to reliably account inter- intra-individual Methods Participants were males who self-reported absence disorders, used monitoring device (Dreem Headband, Dreem, France) sleep. Night-to-night was determined five consecutive weeknights coefficients variation (CV), minimal number individuals needed determine each parameter assessed. Results whole group (n = 94; 470 nights) high (CV 0.44–0.58) N2, N3, onset persistent latencies, wake after (WASO), medium 0.22–0.28) N1 N3 percentage, awakenings REM latency, low 0.04–0.19) efficiency, N2 percentages, total time (TST) micro-arousal index. Minimum reliable assessment TST WASO 2 with 10 participants 4 50 participants, respectively. Conclusions underestimated under-recognized. These data on commonly will facilitate better estimation trials based outcomes interest.

Язык: Английский

Процитировано

20