Eye-tracking tool to spot autism sets sights on commercialization
Subscriber Benefit
As a subscriber you can listen to articles at work, in the car, or while you work out. Subscribe Now
When parents are worried about their child’s development, they often face an additional stress: they may have to wait months, or even years, for their child to be evaluated for autism. The delay is caused by the number of children who need an autism evaluation far outweighing the supply of specialists qualified to do the evaluations.
The answer, says a team of Hoosier researchers, is arming primary care doctors with a simple—but scientific—method to diagnose autism as accurately as specialists, and they’ve created a tool that uses the eyes as a window to the brain.
“Families experience very substantial stress when they’re concerned about their child’s development, but can’t get answers,” says Indiana University School of Medicine (IUSM) Assistant Professor of Pediatrics Dr. Rebecca McNally Keehn.
Several factors contribute to the backlog of families waiting for autism evaluations; specialists, already in short supply, often practice in metropolitan areas, limiting access for children in rural areas. Evaluation methods have also typically been expensive and require special training. While a child can be reliably diagnosed before age two, most aren’t diagnosed until four years old or later; Keehn notes that’s years of missed intervention.
“There’s emerging research suggesting that even the difference between starting the intervention at 18 months versus later, like [two or two-and-a-half years old], has a different effect on developmental outcomes,” says McNally Keehn.
McNally Keehn explains how IUSM’s Early Autism Evaluation Hub system has helped give more Hoosier children access to autism evaluation.
McNally Keehn believes the solution is arming primary care doctors with a tool that can diagnose autism with equal accuracy as specialists. Working with researchers at Purdue University, the team has developed an eye-tracking technology that analyzes a child’s “looking behavior.”
In its current form, the tool uses a series of high-tech cameras mounted on a laptop. The team plays videos on the screen and the cameras analyze where the child is looking.
“Some of the tasks show a social video—kids playing—and some show a non-social video, like a geometric pattern,” says Purdue Autism Research Center Associate Director Dr. Brandon Keehn. “We can, essentially, identify those [children] that might be at greater likelihood for an autism diagnosis just based on the pattern of their looking.”
Keehn says the research team is working to make the videos even more engaging for young children by “cartoon-ifying” them.
The eye-tracking tool also analyzes pupil diameter, another biomarker that’s been shown to indicate social and non-social attention and brain function.
“[Using conventional methods for autism diagnosis], pediatricians are pretty accurate; about 80% of the time, they agree with the gold standard expert evaluation,” says Keehn. “We’re trying to close that gap. Rather than 80%, with that extra tool—the eye-tracking biomarkers—they’re essentially equivalent to that expert evaluation.”
While eye-tracking techniques have been validated in studies, Keehn says there’s significant gap in translating eye-tracking methods to tools that could be used in the clinic to help families waiting months or years for an autism evaluation.
“As it is now, [our tool] is a research grade eye-tracking device. We want to turn that into a much more scalable device that can be used by pediatricians in their offices,” says Keehn. “One of our next steps is to develop [our device], so that it’s cheap, easy to use, and can be used at scale in pediatricians’ offices, measuring the same things that we measure with our research-grade eye-tracker, but doing it in a way that’s relatively automatic.”
And Keehn notes that’s “the real challenge”—bridging the gap where many other technologies have failed to become a usable tool. The team is working with engineers to develop the device and give it commercial legs “to scale this broadly in primary care practices across the U.S.,” says McNally Keehn.
“This approach, where we can increase capacity of primary care to conduct autism evaluations in a high-quality way at scale, has the potential to impact kids, families and healthcare providers,” says McNally Keehn. “And that’s what drives us every day in doing this work.”
