Skip to Content

Device Tracks How You’re Sleeping

A new system monitors your sleep patterns at home. But will it improve your sleep?
July 28, 2009

If you had asked me this morning how many times I woke up last night, I would have guessed four or five. But according to the Zeo, a new gadget that monitors a person’s sleep, it was a disturbing 15 times. I’m also getting considerably less sleep than I thought, averaging about six to seven hours rather than the seven to eight hours I had always estimated.

Sleeping soundly? Novel sensor technology embedded in a soft headband (shown here) allows a new sleep-tracking device, called the Zeo, to determine a user’s sleep state. The data is sent wirelessly to a bedside display unit, also shown here, which shows the user’s sleep patterns.

The Zeo Personal Sleep Coach, developed by a startup headquartered in Newton, MA, is the first at-home device that allows people to track their sleep cycles over time. With a simple headband recording system, the device represents a neat feat of engineering. And it certainly seems to tap into an interest–everyone I told about the Zeo wanted to test it out, a testament to our obsession with sleep, or lack thereof. It’s not yet clear that it will truly help users improve their sleep, but it may present a new opportunity in sleep research, allowing scientists to track normal variability in sleep much more cheaply and on a broader scale than before.

In essence, the Zeo is a highly simplified and automated version of the technology used to assess patients in sleep labs. (The company is careful to point out, however, that the Zeo is not a medical device and cannot diagnose sleep disorders.) While sleeping, users wear a sensor-laden headband that measures electrical activity in the brain. That data is wirelessly transmitted to a display unit like an alarm clock that sits next to the bed. In the morning, the display unit gives a summary of the previous night’s sleep, including how long a user slept, how many times she woke up, and the amount of time she spent in the various stages of sleep. A small memory card within the display unit stores the data, which can then be transferred to a computer and uploaded to a website that tracks the user’s sleep trends and offers advice for improving sleep.

“I see it more like an assessment, such as a blood-pressure monitor or weight scale, to monitor your physiology,” says Phyllis Zee, a sleep scientist at Northwestern University in Chicago, who is also a scientific advisor to the company. “Sleep is really the next vital sign.”

A growing pile of research emphasizes the crucial importance of sleep, linking it to everything from memory to obesity. A typical night’s sleep involves a repetitive cycle of light sleep followed by deep- or slow-wave sleep followed by REM (rapid eye movement) sleep–the time when we dream. These phases keep repeating, with the duration of each cycle shortening as the night progresses. “The duration of these cycles and the number of cycles are indications of sleep quality,” says Michael Twery, director of the National Center on Sleep Disorders Research, part of the National Institutes of Health in Bethesda, MD.

While the company won’t divulge specific details on its technology, the biggest innovation lies in the Zeo’s sensor and the algorithm used to process the information it records. In sleep labs, brain-wave activity is recorded via a number of electrodes attached to the scalp, a technique known as electroencephalography (EEG). Because this approach is highly susceptible to noise, the electrodes are placed in precise spots for optimal recording, and a conducting gel is usually smeared between the electrodes and the skin to improve the signal. Researchers at Zeo have developed a novel technology using dry, silver-coated fabric electrodes that sit on the forehead (typically not considered a great spot for recording EEG activity).

An algorithm that was developed using neural-network analysis processes the messy electrical information recorded by the sensor and automatically determines the wearer’s stage of sleep. (In a sleep lab, a technician watches the activity in real time, regularly scoring the patient’s stage of sleep.) “The algorithm is designed to mimic what an expert sleep scorer would report,” says John Shambroom, Zeo’s vice president of research. A comparison of the Zeo to traditional, technician-scored polysomnography–EEG and other measures used in a sleep lab–found that both performed similarly in healthy people, he says.

Deep sleep: Users can upload sleep patterns to a website that tracks and analyzes the data and sends e-mails with suggestions to improve sleep. Shown here is a graph of one user’s deep sleep over a week.

My own pattern of sleep seems to be slightly off-kilter. I get plenty of REM sleep–about 25 percent–but not enough deep sleep–about 10 percent, rather than the average 20 percent. Deep- or slow-wave sleep–the stage of sleep that is most difficult to wake from–has been linked to some types of memory, and a recent study suggests that lack of deep sleep can affect insulin sensitivity, a risk factor for type 2 diabetes.

An evening beer didn’t seem to have any effect on my sleep patterns, but a benzodiazepine sleeping pill did. As predicted by previous research, it dropped my slow-wave sleep percentage further, to about 5 percent. “Sleep is much more active and dynamic than people perceive,” says Shambroom. “This allows you to understand your personal sensitivity to caffeine and alcohol.” (As a highly caffeine-sensitive person, I didn’t risk a late afternoon latte, even for the sake of science.)

What exactly this means for me is unclear. I’ll probably stay away from that kind of sleeping pill. But the experts I spoke with were hesitant to comment on what my broader sleep patterns meant for my overall health, probably because the purpose of different types of sleep–indeed, the purpose of sleep in general–is still hotly contested. (There are few ways to increase slow-wave sleep specifically; increasing body temperature prior to the sleep cycle is one.) Most of the advice from Zeo centers on well-known recommendations for “sleep hygiene,” such as avoiding caffeine, alcohol, and stimulating activities right before going to sleep. (Disclaimer: I used the device only for a few nights, not long enough to test the online coaching program that comes with it. This program sends users e-mails with advice based on their individual sleep data.) “Really, the litmus test is how you feel during the day,” says Northwestern’s Zee.

As noted earlier, Zeo emphasizes that the device is not intended to be used as a medical device. It cannot, for example, diagnose sleep apnea, the most common sleep disorder, or disturbances such as periodic limb movements and seizures during sleep. But some worry that this warning might not be heeded. “If you have conventional insomnia, the Zeo would be fine,” says Jerome Siegel, director of the sleep research lab at the University of California, Los Angeles. “It would confirm you have insomnia and provide a way of quantifying and validating your sleep patterns as you do the conventional things to treat it.” But the more general issue, he says, “is that people who have the most severe sleep disorders won’t detect [them] with this device and then they may not go to their physician to be diagnosed.”

Siegel and others, however, are interested in the Zeo’s potential as a research tool. While sleep scientists have investigated numerous aspects of sleep–including the effects of drugs, age, and diseases such as depression–these studies are enormously expensive (a night in a sleep lab costs about $1,000). A relatively cheap home-use device would allow researchers to look at the variability of normal sleep patterns in a much larger number of people. “It’s simple to use, practical, and it can be monitored on a daily basis,” says Zee. “That has been a limitation in the field–recording sleep day after day.”

While a sleep lab generates much more detailed information, the Zeo might provide a way to study normative sleep patterns, such as variation with the season, adds Siegel. “It’s extremely interesting,” he says. “All sorts of things that would have been prohibitively expensive now become quite practical.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.