Sleep problems are getting worse around the world, with one recent scientific paper describing sleep problems as a ‘global epidemic’. Around a third of U.S. adults are not getting enough sleep every week, according to research by the CDC. Part of the blame lies with our device addiction, the report found, along with issues such as hectic modern lifestyles meaning that people do not go to bed at the same time every day. For physicians, understanding and dealing with such issues requires accurate ways to measure how and when people are sleeping, and that could require a fresh technological approach.
Today, the standard way to analyze sleep problems is the polysomnography (PSG) test, which is used to identify and measure issues such as sleep apnea and insomnia. These comprehensive overnight tests measure everything from eye movements to oxygen levels, along with muscle activity and heart rate, offering doctors an accurate way to measure sleep stages and diagnose disorders such as sleep apnea. But PSG tests require large numbers of sensors attached to the body, and are carried out in a sleep clinic, rather than the patient’s home.
Understandably, some patients find it harder to sleep in a strange environment, and with wires attached to their bodies. The experience of a PSG test may actually disrupt sleep, making it harder to get accurate data and get to the root of patients’ problems. New technology, powered by AI and radar is helping to deal with this.
Why radar makes a difference
Sleep centers are turning to radar for a less cumbersome way to understand patients’ sleep problems. Radar is high-resolution enough to detect the tiny movements patients make when they sleep, and assess heart rate, breathing and movement throughout the night. Radar sensors do not disturb patient sleep in the way that traditional wired polysomnography tests do, with a small sensor sitting over the patient’s bed recording movements including breathing and heart rate.
Researchers have worked with traditional polysomnography data, comparing the output to radar data, and working with thousands of medical datasets to gain understanding of radar data outputs. This means that the accuracy levels of the radar output now approach the levels of polysomnography but delivered in a far less intrusive way. With a simple sensor at the end of a bed, data can be delivered to an app or interactive dashboard, offering doctors insights into how patients are sleeping.
How AI helps
Monitoring patients in a no-contact way is just the start. AI algorithms help to turn the tiny movements that humans make while sleeping into data that can be used to assess sleep quality. AI turns the raw data into sleep scores that allow clinicians to assess patients’ sleep accurately, working out how long patients spend in different sleep stages such as (REM) sleep and non-rapid eye movement (NREM) sleep. The output offers in-depth information on everything from when sleep starts and ends to sleep latency (the amount of time it takes to fall asleep) and the presence of sleep apnea.
Radar can offer physicians detailed insights into everything from respiratory rates to snoring, all without wires attached to the body. Going forward, radar systems could detect not just length and quality of sleep but assess other related conditions such as asthma and chronic obstructive pulmonary disease (COPD). Physicians believe that these non-contact sensors could also have important applications in dealing with major health conditions such as depression and anxiety.
Discreet monitoring
In contexts such as elder care, no-contact sensors such as radar bring important advantages, as well as offering sleep monitoring which can help track the effectiveness of treatments and offer insights into overall wellbeing. Unlike cameras, there is no surveillance element to radar ‘watching’ a patient, and patients can be reassured their privacy is secure. Radar sensors can be used in the home with supervision from medical professionals and can even be integrated with IoT equipment so that (for example) lights can go out once a radar system detects that an elderly patient has gone to bed. The sensors can also be used to monitor patient condition, and can alert medical professionals about falls, wandering, and other emergencies.
Sleeping more soundly
With sleep problems a growing issue in our connected, fast-moving world, radar offers a perfect way to measure people’s sleep in an accurate but non-contact way. Unlike today’s wired solutions, radar is discreet, privacy-protecting and can work with everything from smartphone applications to dashboards in hospitals, delivering highly accurate assessments of everything from sleep apnea to insomnia. In a world struggling to sleep, no-contact sensors offer a way for doctors to understand problems and help us all sleep better.
About Dr. Jae-Eun Lee
Dr. Jae-Eun Lee is the CEO at bitsensing, an innovative company specializing in advanced radar solutions. Prior to founding bitsensing, Jae-Eun was a Senior Research Engineer at Mando Corporation. He holds a PhD in Electrical Engineering from Seoul National University and a Master’s degree in Electronic Engineering from POSTECH, underlining his deep expertise in the field.
