Wearable sensor may help detect anxiety and depression in kids-摩登家庭人人影视网

当前位置：首页 > 关于我们 > Wearable sensor may help detect anxiety and depression in kids 正文

Wearable sensor may help detect anxiety and depression in kids

时间：2024-09-22 07:26:35 来源：摩登家庭人人影视网作者：行业动态阅读：717次

Welcome toSmall Humans, an ongoing series at Mashable that looks at how to take care of – and deal with – the kids in your life. Because Dr. Spock is nice and all, but it’s 2019 and we have the entire internet to contend with.

When Ellen McGinnis started her career as a psychologist several years ago, she realized just how hard it can be to spot anxiety and depression in young children. Though they have complex inner lives and sometimes develop mental health disorders, preschoolers don't always show the traditional symptoms you'd expect to see in older children or adults, nor do they always have the words to express their feelings.

McGinnis, now a clinical psychologist at the University of Vermont Medical Center's Vermont Center for Children, Youth and Families, also worked with parents experiencing their own mental illness, and they found it particularly difficult to gauge their children's emotions and moods. That made it even harder to screen kids. McGinnis wanted a fast yet objective way to identify which children might need help or treatment.

SEE ALSO:The skills that every teen should learn before they ever get a cellphone

On Wednesday, she and five other researchers published a study in the journal PLOS ONE detailing their efforts to do exactly that with a wearable sensor that analyzes a child's response to a task designed to elicit anxiety and fear. The researchers took a common sensor used in smartphones to measure movement and attached it to a belt that could be worn by the 63 children in their study.

Then they had the children, all between the ages of 3 and 7, complete a 90-second standardized research task. Each child walked into a dimly lit room where a trained adult followed a script meant to build anticipation, saying things like, "I have something in here to show you" and "Let’s be quiet so it doesn’t wake up." The child was led to the back of the room and stopped a foot in front of a covered terrarium, which the adult quickly unveiled to reveal a fake snake.

The sensor measured how the children moved as they perceived a potential threat, startled upon seeing the snake, and responded as they realized it was fake. Research shows that different types of physiological responses to an uncertain threat can indicate psychopathology, or simply put, anxiety or depression.

"It's important with these data-driven results that it matches with psychological research."

A child's physiological response to the potential threat, measured in a 20-second stretch of time, was analyzed by a machine learning algorithm, which successfully identified children as having or not having anxiety or depression 81 percent of the time. By comparison, a parent questionnaire used to identify those conditions in children and teenagers is accurate between 68 and 75 percent of the time.

"It's important with these data-driven results that it matches with psychological research," says McGinnis, who is married to the study's lead author, Ryan McGinnis, assistant professor in the department of electrical and biomedical engineering at the University of Vermont.

"Kids showing [anxiety and depression] turned away as fast as possible from an anticipatory threat, before they even saw this snake ... which matches well with how we understand anxiety disorders."

Mashable Top StoriesStay connected with the hottest stories of the day and the latest entertainment news.Sign up for Mashable's Top Stories newsletterBy signing up you agree to our Terms of Use and Privacy Policy.Thanks for signing up!

If the results can be replicated with larger samples, the researchers envision the possibility of creating a new screening tool that can help pediatricians quickly identify children who should be referred to a psychologist or psychiatrist for further evaluation. McGinnis and her colleagues see the technology supplementing current diagnostics as opposed to replacing them. They don't anticipate it becoming available to the public as a downloadable app.

John Torous, director of the digital psychiatry division at Beth Israel Deaconess Medical Center in Boston, said the study's results are promising but that much more research is needed. (Torous wasn't involved in the study.)

"It’s kind of exciting -- this digital monitoring that’s very quick, and it seems simple to do," says Torous. "The real question is how well will this work in the real world."

In the real world, of course, pediatricians can't use a fake snake to scare every child they're examining. That's why the researchers are considering isolating the task to the 20 seconds of introducing children to a perceived threat, like a dimly lit office with a covered object they'd walk toward. They're also studying other tasks, including having children wear the sensor while playing with bubbles or while speaking and being interrupted. They're looking for tasks that can be translated easily to different contexts, like school or a doctor's office.

"The real question is how well will this work in the real world."

"While the ‘snake’ task may not reflect an everyday occurrence, apprehension and fear about approaching a novel or unfamiliar situation is very real-word," Katherine L. Rosenblum, a professor in the departments of psychiatry and obstetrics & gynecology at Michigan Medicine and co-author of the study, wrote in an email.

Yet, Torous says that even if more clinicians were able to more effectively diagnose anxiety or depression in children, it's not clear what kind of services they'd receive or how they'd even access help. That's partly because of a shortage of child and adolescent psychiatrists. Previous research suggests that 2.6 million children in the U.S. have been diagnosed with anxiety and depression, but nearly 20 percent of children who experience those conditions don't get treatment.

McGinnis is hopeful that as research, including her own, draws attention to the children with undetected mental illness who are "unintentionally overlooked," it'll prompt bigger conversations about how to provide various types of treatment in diverse settings.

But McGinnis and her fellow researchers have much more research ahead of them before a wearable sensor becomes a reality. Since their machine learning algorithm is currently based on the small sample size in their study, they'll need to see if it's as effective with a more diverse population. Torous points out that a machine learning algorithm that's trained on data from white girls, for example, might not have the same accuracy when applied to black boys. (The study's sample size was 57 percent female, 65 percent white, and 82.5 percent or participants lived in two-parent households while 32 percent had an annual household income of greater than $100,000.)

"What kind of biases are present in [their] study," asks Torous. "Machine learning is only as good as the data it’s trained on."

Then there are the ethical implications of collecting data that can pinpoint exactly how a child responds to a potential threat. Ryan McGinnis prefers to looks at the positive aspects of getting that level of insight but admits "there is always the flip side." Now that we're increasingly aware of how apps and social media platforms sell and misuse our data, one could imagine a dystopian future in which someone found a way to access and monetize the kind of information the researchers hope to collect, or even use it against unsuspecting patients.

For now, Ellen McGinnis is optimistic about the difference a wearable sensor could make for millions of children experiencing undiagnosed anxiety and depression.

"People used to think kids didn’t have emotions that were stable ... so there was no depression in young people," she says. "[W]ith more objective data, we can better understand this early childhood period."

Featured Video For You

Bees with sensor backpacks may help farmers monitor crops

Related content from our partners: