Skip to Content
Humans and technology

Wrist Sensor Tells You How Stressed Out You Are

Devices from two startups could be used to treat people with anxiety disorders—and one of the devices may eventually diagnose pain.
December 20, 2012

Amid rising concerns over post-traumatic stress disorder (PTSD) and other mental illnesses, two MIT startups are developing wrist-worn sensors that can detect physiological changes—including perspiration and elevated temperature—that may signal the onset of events like anxiety attacks.

Stress gauge: Neumitra’s wrist-worn sensor detects elevated perspiration and temperature associated with stress or excitement and can relay messages to the wearer.

The data collected by these devices can be fed into an algorithm that aims to learn what triggers anxiety, or when people may be about to engage in a risky behavior. One goal is to head off destructive behavior, from drug abuse to suicide and violent outbursts, and to help with treatment.

Although the technology is still experimental and the devices are used mainly in medical research, they herald the appearance of consumer versions and associated apps that let people monitor their mood and stress levels. Other emerging technologies aim to detect emotion using subtle cues from how people use smartphones (see “A Smartphone That Knows You’re Angry”).

The newest of these startups, Boston-based Neumitra, emerged from stealth mode earlier this year with a device that can measure proxies for excitement or stress, including increased motion, increased skin conductance from perspiration, and elevated skin temperature. The device, called bandu, sends readings to the wearer’s smartphone, which records them for later analysis. The device also includes a display that can be customized to suggest, say, that you take your medication, call a loved one, or listen to a song.

The challenge is teasing out exactly which physiological changes are linked to stressful events. Researchers also must filter out noise, such as perspiration from exercise or excitement from watching a favorite sports team. Some improvements will come from self-learning algorithms.

The Neumitra device is already being used in new research efforts at Massachusetts General Hospital, including ones to help patients suffering from PTSD and other anxiety disorders. The aim of the research is to create detailed records of what is triggering anxiety, says Darin Dougherty, director of MGH’s division of neurotherapeutics. While it is early days for the technology, he says, the device has the potential to fill a diagnostic gap.

“Previously, treaters had to rely on the patients’ subjective memory for the week, or weeks, between appointments as their sole measure of anxiety symptoms,” Dougherty says. The device “provides moment-to-moment objective data regarding a patient’s anxiety symptoms,” and both the doctor and patient can see it.

Robert Goldberg, a neuroscientist and Neumitra founder, points to a grim national backdrop: one in three U.S. adults has a form of mental disorder—ranging from attention deficit hyperactivity disorder to PTSD—making it the third-largest category of health-care spending. More soldiers now die of suicide than in battle. “My goal is to bring mobile portable technologies to brain health,” Goldberg says. 

Neumitra’s underlying technology is similar to that used by a commercial device made by Affectiva, a startup from MIT’s Media Lab that has made recent inroads in medical research.

The company’s Q sensor has been used for more than a year in trials with an undisclosed company that is attempting to develop a physiological measure of pain. This would be an enormous advance, as pain diagnoses currently rely entirely on patients’ self-reporting.

The device is also in clinical trials to help measure drug efficacy in sufferers of Rett Syndrome, a genetic neurological disorder that can cause loss of speech, problems with motor coördination, and impaired cognitive function. The goal there is to see whether the drug is producing any measurable changes in patients’ stress levels, movement, and sleep patterns. 

Affectiva has made recent advances in other medical applications. In a recent study, the device showed it can serve as a proxy detector for a type of epileptic attack associated with sudden death. More than 50,000 people a year die of seizures, and most are unexplained. But in a few cases where the victims died wearing headgear that can measure the brain’s electrical activity (an electroencephalogram or EEG), clinicians found that brain waves were unusually suppressed. 

Rosalind Picard, a cofounder of Affectiva and director of affective computing research at the Media Lab, says that physiological responses measured by the Q sensor are associated with this same type of suppressed brain activity (which is not always fatal). “We now have a correlate that is much easier to use than wearing an EEG all the time,” Picard says. Clinical trials using the device in epilepsy patients are being planned, she adds.

The Q sensor can also help predict outbursts in autistic children, who often have difficulty verbalizing their emotions (see “Sensor Detects Emotions through the Skin”). Picard says her company is also working on ways to detect changes that may signal that a person is about to engage in risky or destructive behavior, such as drug abuse or even an outburst of violence.

Affectiva is currently focusing on supplying researchers—a single device costs $2,000, but the price is in the mid-hundreds when they are sold in quantity—including the software and support. The company is not currently marketing to consumers, but a consumer version might be available in a year or two, Picard says.

Goldberg says the Neumitra device will sell for $249 to $1,499, depending on features and analytics, and will be made available to consumers, as well as researchers and other organizations, sometime in 2013.

Goldberg says the company was inspired in part by a war veteran who was having anxiety attacks. Three years into therapy he realized that one source of his stress triggers were his visits to a Whole Foods supermarket. For him, the experience—strangers, suddenly appearing from a maze of lanes, holding objects—felt threatening. “Whole Foods was a trigger point. We could have given him the same information in a week,” he says.

Goldberg made a map of his own stress triggers, including several hot spots on and around the MIT campus.  He says he realized those were from rushing to events, and from being asked to speak publicly, as he did recently at a health startup event

“We see in healthy individuals—stress associated with real-world events. I can point to my own—public speaking is one of the major categories of phobia,” he says. Ultimately, with enough people wearing stress monitors, he says, a company could find what was making employees stressed out, city officials could learn where women felt unsafe, and a nation could even take its collective stress pulse.

After the event, Goldberg invited me to put on the bracelet. He then asked me to recite the alphabet backwards in front of several strangers, who watched with a gleam of schadenfreude in their eyes. I lost my way around “W.” Goldberg, watching readings on his smartphone, said my wrist sweat and temperature indicated my stress was up 50 percent.

Keep Reading

Most Popular

How to opt out of Meta’s AI training

Your posts are a gold mine, especially as companies start to run out of AI training data.

The return of pneumatic tubes

Pneumatic tubes were supposed to revolutionize the world but have fallen by the wayside. Except in hospitals.

Why does AI hallucinate?

The tendency to make things up is holding chatbots back. But that’s just what they do.

How a simple circuit could offer an alternative to energy-intensive GPUs

The creative new approach could lead to more energy-efficient machine-learning hardware.

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 with a list of newsletters you’d like to receive.