The wearables industry is producing novel devices claiming to measure sleep/wake state and, most recently, sleep stage composition, by using information from several bio-signals in addition to motion. We evaluated the validity of a multi-sensor sleep-tracker (the ŌURA ring) against polysomnography (PSG) in measuring sleep/wake states, “light sleep” (PSG-N1+N2), “deep sleep” (PSG-N3) ... [Show full abstract] and rapid-eye-movement (REM) sleep.
We compared standard PSG and ŌURA ring sleep data obtained from a single laboratory overnight in forty-one healthy adolescents and young adults (13 females; Age: 17.2 ± 2.4 years).
ŌURA ring significantly underestimated PSG-N3 by about 20min, and overestimated PSG-REM sleep by about 17min (p<.05). It showed no significant bias for sleep onset latency (SOL), total sleep time (TST), and all-night wake after sleep onset (WASO). The PSG-ŌURA differences for TST and WASO lay within the ≤30min a priori set clinically satisfactory ranges for 87.8% and 85.4% of the sample, respectively. PSG-ŌURA discrepancies for WASO were greater in participants with more PSG-defined WASO (p<.001). The ring position affected the magnitude of the PSG-ŌURA discrepancies for “light sleep” and REM sleep (p<.05), with greatest discrepancy when placed on the ring finger. Epoch by epoch analysis indicated that the ŌURA ring had a sensitivity (ability to detect sleep) of 96%, specificity of 48% (ability to detect wake), agreement of 65% in detecting “light sleep”, agreement of 51% in detecting “deep sleep” and agreement of 61% in detecting REM sleep, relative to PSG. Importantly, similarly to PSG-N3 (p<.001), “deep sleep” detected with the ŌURA ring was negatively correlated with advancing age (p=.001), showing the ability of the device to capture a well-established effect in the literature. Finally, the percentage of participants the ŌURA ring correctly categorized into PSG-defined TST ranges of <6h, 6-7h, >7h were 90.9%, 81.3%, and 92.9%, respectively.
The ŌURA ring showed the potential for detecting sleep outcomes beyond “sleep” and “wake” by using multiple sources of information in addition to motion, including heart rate variability and pulse wave amplitude. The potential and reliability of a multisensory approach in assessing sleep stages needs to be further explored.
Support (If Any)