Performance of a multisensor smart ring to evaluate sleep: in-lab and home-based evaluation of generalized and personalized algorithms

Abstract

Study Objectives Wearable sleep technology has rapidly expanded across the consumer market due to advances in technology and increased interest in personalized sleep assessment to improve health and mental performance. We tested the performance of a novel device, the Happy Ring, alongside other commercial wearables (Actiwatch 2, Fitbit Charge 4, Whoop 3.0, Oura Ring V2), against in-lab polysomnography (PSG) and at-home electroencephalography (EEG)-derived sleep monitoring device, the Dreem 2 Headband. Methods Thirty-six healthy adults with no diagnosed sleep disorders and no recent use of medications or substances known to affect sleep patterns were assessed across 77 nights. Subjects participated in a single night of in-lab PSG and two nights of at-home data collection. The Happy Ring includes sensors for skin conductance, movement, heart rate, and skin temperature. The Happy Ring utilized two machine-learning derived scoring algorithms: a “generalized” algorithm that applied broadly to all users, and a “personalized” algorithm that adapted to individual subjects’ data. Epoch-by-epoch analyses compared the wearable devices to in-lab PSG and to at-home EEG Headband. Results Compared to in-lab PSG, the “generalized” and “personalized” algorithms demonstrated good sensitivity (94% and 93%, respectively) and specificity (70% and 83%, respectively). The Happy Personalized model demonstrated a lower bias and more narrow limits of agreement across Bland-Altman measures. Conclusion The Happy Ring performed well at home and in the lab, especially regarding sleep/wake detection. The personalized algorithm demonstrated improved detection accuracy over the generalized approach and other devices, suggesting that adaptable, dynamic algorithms can enhance sleep detection accuracy.

Good sleep is a mood buffer for young women during menses

“…Previous research has repeatedly shown discrepancies between measures of subjective and objective sleep [ 31–34 ], suggesting that objective and subjective sleep outcomes may reflect different aspects of sleep. On the other hand, we cannot exclude that performance limitations of the ŌURA ring [ 35 ] may have affected the results.…”

Accuracy, Utility and Applicability of the WHOOP Wearable Monitoring Device in Health, Wellness and Performance - a systematic review

“…Where manual and automatic detection were assessed separately, result ranges were displayed, and the absolute error was converted to percentage by dividing by the PSG measurements (Miller et al, 2021). Where only figures and no numbers were provided, calculations of mean absolute error were not possible (Grandner et al, 2023). The classification of bias for heart rate (HR) and HRV measures was done according to the following criteria proposed by Miller et al 2022): 0.0–0.1, trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large; 0.7–0.9, very large; 0.9–1.0, nearly perfect (C. Bellenger et al, 2021; Hopkins et al, 2009).…”

“…Furthermore, both WHOOP manual and automatic two and four sleep categorization could provide a practical alternative to PSG and perform well against other commercially available devices (Miller et al, 2021). Grandner et al showed moderate and comparable sleep specificity with commercially available activity trackers and that personalized algorithms that adapt to user data over time can improve device performance (Grandner et al, 2023).…”

Remote Wearable Neuroimaging Devices for Health Monitoring and Neurophenotyping: A Scoping Review

“…In [43], the participants were asked to participate in one night in-lab PSG and two at-home EEG-monitoring sessions using the Dreem 2 Headband. Simulatnesously, the participants were asked to put on other wearable devices (rings or wristbands).…”

“…A Dreem headband (EEG-headband) was used in [39,57]. The Dreem 2 headband (EEG-headband) was used in [43]. The URGOnight device (an EEG headband) was used in [48].…”