Scientists in China examined Huawei’s popular smartwatch against Lab Sleep Lab tests and found it amazingly accurate in sleep monitoring, though incomplete to diagnose sleep disorders.
Study: Validity and clinical utility of a device worn by the fruit against polymnography. Credit Picture: Andrey_Popov/Shutterstock.com
A recent Body A The study used data from participants from sleep clinics to evaluate the performance of a consumer sleep monitoring device in relation to polynomnography (PSG).
Sleep Tracking Using Portable Technology
The importance of sleep cannot be emphasized enough, with bad sleep associated with dementia, cardiovascular disease and other morbid conditions. PSG is used to measure sleep in clinical practice and is considered the golden model for sleeping disorders. However, it has known limitations, including the volatility of the night, the first night, the low cost efficiency and the need for trained professionals. Alternatives, such as portable monitoring and radiographs, are limited to specific populations and remain unclear.
Since the 1960s, sleeping technology has grown quickly, to a large extent by the developments of artificial intelligence (AI). These devices use built -in accelerometers to monitor movements and determine a person’s sleep status. Some also use mechanical learning to improve performance and provide information beyond sleep/awakening detection. However, product updates and changing algorithms often make it difficult to validate these products, making it difficult to integrate these devices into clinical practice.
This study represents the first validation of a Huawei Smartwatch in a Chinese clinical population, facing a significant gap, as most previous validation studies have focused on western brands such as Fitbit and Apple.
For the study
This study evaluated the performance of the Huawei Watch GT2 against the PSG. The Huawei Watch GT2 collects heart rate and sleeping variant signals for sleep detection. Smartwatch was tested in different sleep disorders. Adult participants who had completed the demographics and sleep questionnaires were hired from March 1, 2021 and April 30, 2023. People who worked night shifts in the last 6 months, slept less than 4 hours, had cognitive conditions.
All participants completed a one night PSG follow -up and the sleeping stages (N1, N2, N3 and Rapid Eye Sleep) were rated. The 30s seasons were used to analyze and abstain and sleep. Obstructive sleep apnea (OSA) was diagnosed using an apnea sub -sector (AHI) greater than or equal to 15/h. Other hypoxemia indicators were collected, such as the lower saturation of the oxygen pulse (LSPO2) and the Oxygen cutter (ODI).
Smartwatch provided four stages of sleep recording: awake, slight sleep (equivalent to N1 and N2), deep sleep (equals N3) and REM sleep. Measurements of interest included total sleep time (TST), awakening after sleeping (WASO) and sleep performance (SE), which is defined as TST divided by minutes between lights and lights. Because raw data could not be extracted from the device, the researchers manually exported information to sleep from the application graphs, ensuring synchronized PSG data timing.
Study findings
A total of 98 participants met the integration criteria, with about 84% being men. The average age and body mass index (BMI) was 45 and 26.0kg/m2, respectively. More than half of the participants complained of drowsiness during the day and poor sleep quality. The median SE and TST were 85% and 405.8 minutes, respectively. The results of the PSG showed that 47 patients had moderate to severe OSA, 33 patients were normal, 12 suffered from conspiratorial insomnia and sleep apnea and 30 had clinical insomnia.
PSG and Smart watches more agreed on Wake and Light sleep sorts. The smart watch seemed to sort the PSG REM as a slight sleep and the error rates in this case were high. In addition, deep sleep and REM sleep are often classified as a slight sleep. Among other possible stages, incorrect sorting errors were relatively lower.
Smartwatch agreed with the PSG for sleep EBE against WAKE states with a specialty of 44.5%and 95.3%sensitivity and the positive predictive value (PPV) was 72.20%. The total accuracy reached 87.3%and Cohen’s K Cohen 0.43 (K = 0.75) showed a moderate to downgrade between the two devices.
In addition to light sleep, the smart watch showed high accuracy for all sleep stages, ie more than 70%. The smart watch greatly overestimated SE, TST, deep sleep, REM sleep and delay in sleeping (SOL) while underestimating the WASO. Specifically, it overestimated the total sleep time by about +28.5 minutes and sleeping by +5.9 percentage points, while devaling up awakening after about -37 minutes.
After adapting to unstable sleep, the delay in persistent sleeping levels (LPS) between PSG and Smartwatch was not significantly different. In patients with sleep disorders, T tests have revealed lower precision in patients with insomnia and lower sensitivity to OSA patients than healthy tests. However, no significant differences were found between subgroups of disorder in the sleeping stage agreement.
Compared to the published criteria for accepted prejudice in portable validation studies (≤30 minutes for TST and ≤5% for SE), the performance of the performance “almost reached” the standards of rays of research quality, supporting its potential as a low -cost tool for sleeping/wake -up detection.
Conclusions
In summary, the Huawei Watch GT2 has a high deal in sleep/waking detection with PSG. While using smartwatches. Consumers and healthcare professionals need to be aware of the overestimation of the sleeping stage and underestimation.
A key restriction of the study revolves around the fact that it was carried out in a single center with a limited number of participants, thereby limiting generalization. Other sleep and mental disorders, such as narcolepsy, depression and periodic disorders of limb movement, were not evaluated.
The device’s algorithm was privately owned and can be changed by updates, requiring a review with any software repeat. In addition, some participants showed loss of data due to the removal or movement of the device during sleep. The rapidly updating of algorithms require new validation and this issue limits the clinical application of portable consumer devices.
Overall, the study suggests that while consumers wearing, such as the Huawei Watch GT2, it is not yet appropriate replacement of PSG in the diagnosis of sleep disorders, they can provide reliable monitoring of two stages (sleep/awakening) for general sleeping health monitoring.
