0068 Estimation of Sleep Stages Using Cardiac and Accelerometer Data From A wrist-Worn Device

Beattie, Zachary; Pantelopoulos, Alexandros; Ghoreyshi, Atiyeh; Oyang, Y; Statan, A; Heneghan, Conor

doi:10.1093/sleepj/zsx050.067

Cited by 28 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the first generation Oura ring had a 96% sensitivity to detect sleep, and agreement of 65% for light sleep, 51% for deep sleep, and 61% for REM sleep in adolescents and young adults [ 11 ]. In another study on Fitbit devices, the estimated Cohen’s kappa was

and an epoch by epoch accuracy of

was reported [ 22 ]. NREM–REM classification of the Emfit bed sensor (based on ballistocardiography) achieved a total accuracy of

, but a kappa of only

, highlighting how accuracy is not an ideal metric for a classification problem with highly imbalanced data [ 23 ].…”

Section: Introductionmentioning

confidence: 99%

The Promise of Sleep: A Multi-Sensor Approach for Accurate Sleep Stage Detection Using the Oura Ring

Altini

Kinnunen²

2021

Sensors

100

View full text Add to dashboard Cite

Consumer-grade sleep trackers represent a promising tool for large scale studies and health management. However, the potential and limitations of these devices remain less well quantified. Addressing this issue, we aim at providing a comprehensive analysis of the impact of accelerometer, autonomic nervous system (ANS)-mediated peripheral signals, and circadian features for sleep stage detection on a large dataset. Four hundred and forty nights from 106 individuals, for a total of 3444 h of combined polysomnography (PSG) and physiological data from a wearable ring, were acquired. Features were extracted to investigate the relative impact of different data streams on 2-stage (sleep and wake) and 4-stage classification accuracy (light NREM sleep, deep NREM sleep, REM sleep, and wake). Machine learning models were evaluated using a 5-fold cross-validation and a standardized framework for sleep stage classification assessment. Accuracy for 2-stage detection (sleep, wake) was 94% for a simple accelerometer-based model and 96% for a full model that included ANS-derived and circadian features. Accuracy for 4-stage detection was 57% for the accelerometer-based model and 79% when including ANS-derived and circadian features. Combining the compact form factor of a finger ring, multidimensional biometric sensory streams, and machine learning, high accuracy wake-sleep detection and sleep staging can be accomplished.

show abstract

and an epoch by epoch accuracy of

was reported [ 22 ]. NREM–REM classification of the Emfit bed sensor (based on ballistocardiography) achieved a total accuracy of

, but a kappa of only

, highlighting how accuracy is not an ideal metric for a classification problem with highly imbalanced data [ 23 ].…”

Section: Introductionmentioning

confidence: 99%

The Promise of Sleep: A Multi-Sensor Approach for Accurate Sleep Stage Detection Using the Oura Ring

Altini

Kinnunen²

2021

Sensors

100

View full text Add to dashboard Cite

show abstract

“…Although the specific algorithms that calculate these values are proprietary to Fitbit, they have been found to accurately estimate sleep duration and quality in normal adult sleepers without the use of research-grade sleep staging equipment. 34 By collecting quantitative sleep data over the course of the semester on nearly 100 students, we aimed to relate objective measures of sleep duration, quality, and consistency to academic performance from test to test and overall in the context of a real, large university college course.…”

Section: Introductionmentioning

confidence: 99%

Sleep quality, duration, and consistency are associated with better academic performance in college students

et al. 2019

View full text Add to dashboard Cite

Although numerous survey studies have reported connections between sleep and cognitive function, there remains a lack of quantitative data using objective measures to directly assess the association between sleep and academic performance. In this study, wearable activity trackers were distributed to 100 students in an introductory college chemistry class (88 of whom completed the study), allowing for multiple sleep measures to be correlated with in-class performance on quizzes and midterm examinations. Overall, better quality, longer duration, and greater consistency of sleep correlated with better grades. However, there was no relation between sleep measures on the single night before a test and test performance; instead, sleep duration and quality for the month and the week before a test correlated with better grades. Sleep measures accounted for nearly 25% of the variance in academic performance. These findings provide quantitative, objective evidence that better quality, longer duration, and greater consistency of sleep are strongly associated with better academic performance in college. Gender differences are discussed.

show abstract

“…7,8 Several alternative techniques have been investigated to measure REM sleep, 9,10 with limited success. In 2017, Fitbit 11 introduced the capability to differentiate light, deep and REM sleep in wrist-worn actigraphy devices, but with a per-epoch accuracy of 69% compared with PSG, 12 the search for an accurate, costeffective ambulatory system to measure REM sleep continues.…”

Section: Introductionmentioning

confidence: 99%

<p>Finger Twitches are More Frequent in REM Sleep Than in Non-REM Sleep</p>

Reiter¹,

Roach²,

Sargent³

et al. 2020

NSS

View full text Add to dashboard Cite

Introduction: Abnormal rapid eye movement (REM) sleep is often symptomatic of chronic disorders, however polysomnography, the gold standard method to measure REM sleep, is expensive and often impractical. Attempts to develop cost-effective ambulatory systems to measure REM sleep have had limited success. As elevated twitching is often observed during REM sleep in some distal muscles, the aim of this study was to assess the potential for a finger-mounted device to measure finger twitches, and thereby differentiate periods of REM and non-REM (NREM) sleep. Methods: One night of sleep data was collected by polysomnography from each of 18 (3f, 15m) healthy adults aged 23.2 ± 3.3 (mean ± SD) years. Finger movement was detected using a piezoelectric limb sensor taped to the index finger of each participant. Finger twitch densities were calculated for each stage of sleep. Results: Finger twitch density was greater in REM than in NREM sleep (p < 0.001). Each sleep stage had a unique finger twitch density, except for REM and stage N1 sleep which were similar. Finger twitch density was greater in late REM than in early REM sleep (p = 0.005), and there was a time-state interaction: the difference between finger twitch densities in REM and NREM sleep was greater in late sleep than in early sleep (p = 0.007). Conclusion: Finger twitching is more frequent in REM sleep than in NREM sleep and becomes more distinguishable as sleep progresses. Finger twitches appear to be too infrequent to make definitive 30-second epoch determinations of sleep stage. However, an algorithm informed by measures of finger twitch density has the potential to detect periods of REM sleep and provide estimates of total REM sleep time and percentage.

show abstract

0068 Estimation of Sleep Stages Using Cardiac and Accelerometer Data From A wrist-Worn Device

Cited by 28 publications

References 0 publications

The Promise of Sleep: A Multi-Sensor Approach for Accurate Sleep Stage Detection Using the Oura Ring

The Promise of Sleep: A Multi-Sensor Approach for Accurate Sleep Stage Detection Using the Oura Ring

Sleep quality, duration, and consistency are associated with better academic performance in college students

<p>Finger Twitches are More Frequent in REM Sleep Than in Non-REM Sleep</p>

Contact Info

Product

Resources

About