AI Algorithm Remotely Monitors Sleep

Sleep is one of life’s most precious gifts—along with scented candles, tacos, and Barry Manilow.

Yet millions of Americans lay awake each night, counting sheep and wishing for their insomniac hell to end.

Sure, you could attach a bunch of sleep-monitoring sensors to yourself, but those will probably do more harm than good. The best alternative, it seems, comes down to science.

Researchers at Massachusetts Institute of Technology (MIT) and Massachusetts General Hospital (MGH) developed a way to remotely observe sleep stages, without applying any bits and bobs.

The device, mounted on a nearby wall, uses an artificial intelligence algorithm to analyze and translate radio signals around the user into sleep stages: light, deep, rapid eye movement (REM).

(Sounds like some kind of Voldemort-dark magic to me.)

“Imagine if your Wi-Fi router knows when you are dreaming, and can monitor whether you are having enough deep sleep, which is necessary for memory consolidation,” study lead Dina Katabi, the Andrew and Erna Viterbi Professor of Electrical Engineering and Computer Science, said in a statement.

“Our vision is developing health sensors that will disappear into the background and capture physiological signals and important health metrics, without asking the user to change her behavior in any way,” she added.

Katabi adopted previously developed radio-based sensors, which emit low-power radio frequency (RF) signals that reflect off the body, as a novel way to monitor sleep.

“The opportunity is very big because we don’t understand sleep well, and a high fraction of the population has sleep problems, MIT graduate student and study co-author Mingmin Zhao said. “We have the technology that, if we can make it work, can move us from a world where we do sleep studies once every few months in the sleep lab to continuous sleep studies in the home.”

To achieve that, the team incorporated a proprietary deep neural network-based AI algorithm, which automatically eliminates irrelevant information.

Just mount and sleep (via Shichao Yue @ MIT)

“The novelty lies in preserving the sleep signal while removing the rest” of the unwanted data,” according to Tommi Jaakkola, the Thomas Siebel Professor of Electrical Engineering and Computer Science.

The resulting 80 percent accuracy, MIT boasted, is comparable to that of sleep specialists based on EEG measurements.

“Our device allows you not only to remove all of these sensors that you put on the person, and make it a much better experience that can be done at home,” Katabi said.

It not only makes the job of the doctor and sleep technologist easier, it also opens new doors for studying how certain diseases, like Parkinson’s, affect sleep. “They don’t have to go through the data and manually label it.”

Katabi and Jaakkola partnered with Matt Bianchi, chief of the MGH Division of Sleep Medicine, to present their findings in a paper co-written by Zhao and grad student Shichao Yue.

Let us know what you like about Geek by taking our survey.

Rise of the Empire expands Star Wars: Rebellion in...
Merkel rounds on German auto executives in opening...

Related Posts