While there are already fitness trackers and smartwatches that are specifically designed to track the wearer's sleep patterns, an international team of scientists believe that with the right software, regular smartwatches are more than up to the job. To that end, they've created an app known as SleepGuard.
Developed by researchers from the UK's Lancaster University and China's Northwest University, the app utilizes the sensors on regular smartwatches to monitor sleep-affecting factors such as body movements, ambient light, and noise.
More specifically, it uses the watch's accelerometer and gyroscope to determine how often the wearer rolls over in the night (lots of rolls are a sign of restlessness), which of four sleeping postures they're adopting (front, back, side, or on the arm), and which of three positions their watch-wearing hand is in.
As far as the latter goes, apparently having the hand on the abdomen can indicate discomfort, placing it on the head may put pressure on the shoulder nerves and thus cause arm pain due to restricted blood flow, while putting it on the chest can result in nightmares caused by pressure on the heart.
Working with the watch's light sensor, the app is able to tell if the sleeping environment is too bright, with the mic being used to sense excessive noise from external sources – the mic is also utilized to monitor the person's breathing patterns, and to note if they're snoring or talking in their sleep.
Once all the data is analyzed, SleepGuard provides users with a sleep report, and suggests causes of (along with solutions for) any problems they may be having. It has been tested on 15 individuals so far, and was found to have an accuracy similar to that of consumer-grade sleep monitors.
"Our project aims to unlock the full potential of off-the-shelf consumer smartwatches, taking advantage of their sophisticated suite of sensors to gain a fuller understanding of a wearer's sleep patterns," says Lancaster's Dr. Petteri Nurmi, co-author of a paper on the study.
Source: Lancaster University via EurekAlert