Receive the Mach newsletter.
By Jessica Wapner
Imagine an application or portable device that could say one day in advance that a person at risk would have suicidal thoughts – and alert the person and his or her trusted contacts. This could soon become a reality, thanks to the nascent field of mood forecasting.
We have become accustomed to fitness trackers and other electronic devices that monitor our physical activity, and scientists are now saying that similar technologies can be used to track our psychological health in ways never before possible.
By identifying the first signs of emotional distress, new applications and wearable devices may soon help preserve our mental well-being.
"We rely on patients to tell us how they feel, and we make our decisions count," said Ipset Vahia, a psychiatrist at McLean Hospital in Belmont, Mass. But as it is not subject to the vagaries of our mood and our ability to assess our psychological state, technology could provide physicians with more reliable information.
Exploit mind-body links
The mood forecast exploits the connection between the mind and the body. Research has shown that changes in our mental state, including bouts of sadness or anxiety, affect our body in a perceptible way.
Heart rate is the best known of these emotional biomarkers – our pulse tends to accelerate when we are stressed. But our body also responds to emotional distress in different ways. "We know that reduced movement and sleep are indicators of depression," says Vahia. Sweating increases with stress. And the temperature of the skin often increases with what Matthew Nock, psychologist at Harvard, calls "emotional arousal."
At least theoretically, any portable device with sensors for pulse, skin temperature and movement can help track our mood. But even something as simple as how we use our smartphones can reveal useful information about our mental state.
The pace at which we text, call, and post on social media are all markers of what experts call our "digital phenotype." And these markers change with our mood: publish more photos when we are happy, for example, or less we feel blue.
As Sharath Guntuku, a researcher at the Center for Digital Health at the University of Pennsylvania in Philadelphia, explains, "we can predict our mental state based on the use of technology."
Detect and interpret biomarkers
To harness all this valuable data, academic researchers and private companies are working to develop devices and programs that not only detect and interpret our biomarkers, but also provide useful advice.
For example, a mood forecasting device or app may urge a person to call a friend after exchanging text messages or walking around when the device has not recorded movement for several hours. Alternatively, changing biomarkers or numerical behavior could be communicated directly to the patient's physician, who could then intervene if necessary.
"By making these day-to-day micro-adjustments, you might not be able to slide into depression and bounce back," said Rosalind Picard, a computer scientist at the MIT Media Lab in Cambridge, Massachusetts.
What shows research
Early studies provide insights into the potential of mood prediction.
Last June, a team led by Picard had announced that an experimental wrist sensor could detect stress. The team followed 201 students for five semesters. Each carried an experimental sensor, developed by Picard, that monitored body temperature, phone activity and skin conductance (how far the skin was conducting electricity). The research, published last year in the Journal of Medical Internet Research, showed that the sensor was 80% accurate in determining when students felt stressed.
In an unpublished research conducted from 2015 to 2018, Nock discovered that a wristband that tracked the movements of psychiatric patients, their skin conductance, and their body temperature could predict suicidal thoughts about a day before their onset. The prediction was accurate at about 75%.
Researchers at the University of Chicago and the University of Michigan also conducted an eight-week study of nine people with bipolar disorder. They discovered that changes in phone use predicted the symptoms of depression and mania. "Mood states in bipolar disorder seem to correlate to specific changes in the use of the mobile phone," the authors wrote in their 2018 publication in the Journal of Internet Medical Research.
This research is preliminary and the results are not perfect. But mental health professionals take note of it.
Last year, when the family of one of his patients contacted Vahia, the psychiatrist at McLean Hospital, to warn her of growing concern, he told her that convinced to start wearing a fitness monitor. Three weeks later, the data collected indicated something that had eluded the patient's family: the length of her nocturnal sleep had dropped by nearly an hour.
When the drug change did not resolve the problem, Vahia had the patient admitted for psychotherapy and adjusted thoroughly, saving her "at least two weeks" of anxious suffering. .
For his next study, Nock is planning to test the biosensitive bracelet on psychiatric patients after they leave the hospital as a result of a suicidal episode. In this case, the information of smartphones and bracelets go into a database. If the research team discovers a cause for concern, she will contact patients to encourage them to seek help, either by calling 911, a crisis service center or their doctor. .
Go ahead despite confidentiality issues
If the mood forecast is very promising, some see a danger in collecting and transmitting such intimate personal data – and this includes some of the scientists working on technology development.
Picard fears that employers can access data and discriminate against employees or job seekers. And Jukka-Pekka Onnela, director of the Harvard Scientific Health Data Program at T.H. The Chan School of Public Health warns that more research is needed to differentiate between a normal digital phenotype and a phenotype of concern.
Despite these and other concerns, mood prediction products are coming in or, in some cases, are already there.
Mindstrong, a company in Palo Alto, Calif., Is developing an app designed to track mental health by monitoring the activity of smartphones. Another company, SpireHealth, based in San Francisco, has developed a portable biomarker reading device that can be stored in a sock or bra. On the futuristic side of fashion, the Ger Mood sweater is characterized by a collar that changes color depending on the changes in the skin.
Many other products are in development. Picard hopes to adapt to the mood forecast of a bracelet that she originally developed to predict epileptic fits. A UK-based software developer hopes to bring to market its "MindYourself" mood forecast.
While mood forecasts are keeping their promise, experts are not convinced that apps and mood-tracking devices will replace mental health professionals.
"The essence of psychiatry remains between a person and a provider," Vahia said, adding that the technology only provided an "additional layer of meaningful information."
But Picard says that mood forecasts could make the difference between feeling good and suffering. "We would like to meet you before you become depressed," she says, "and help you put things back into your life before getting in trouble."
Want more stories on technology?
FOLLOW NBC NEWS MACH ON TWITTER, FACEBOOK AND INSTAGRAM.