Health: Eye tracking app can diagnose children with autism from 16 months

0

Autism could soon be diagnosed via an APP: scientists develop an eye-tracking tool that can identify the condition in children from 16 months

  • A sign of autism in young people is the tendency to focus on objects, not people
  • The app uses the camera on an iPad / iPhone to track a toddler’s eye movements
  • This is done while the child is watching special videos with adults and toys
  • Machine learning then analyzes viewing patterns to make a diagnosis
  • The app-based test takes just ten minutes, the researchers said

An app that can diagnose autism in children from 16 months of age by monitoring eye movements has been developed by researchers from the United States.

A telltale sign of autism in young people is the tendency to draw attention to objects rather than people – a habit that can be revealed by analyzing gaze patterns.

The team’s app uses the camera on iPads / iPhones, combined with machine learning, to track and analyze such eye movements while children watch special videos.

For example, in one, a man blowing bubbles occupies the left side of the screen, while on the other, his bottle of bubble mixture and toys are on a shelf.

While eye follow has previously been used to diagnose autism, this is the first time it has been done without special hardware and an expert at interpreting gaze patterns.

Because it uses ubiquitous technology and only takes 10 minutes to test a child, the new approach could be easily applied in general practice or for home testing.

Scroll down for video

An app that can diagnose autism in children from 16 months old by tracking eye movements (photo) has been developed by researchers from the United States

An app that can diagnose autism in children from 16 months old by tracking eye movements (photo) has been developed by researchers from the United States

“We know that babies with autism pay attention to the environment differently and not so much about people,” says author and pediatric psychologist Geraldine Dawson of Duke University in North Carolina.

‘We can monitor eye patterns in toddlers to estimate the risk of autism.

“This is the first time that we have been able to provide these types of ratings using just a smartphone or tablet.

“This study served as a proof-of-concept, and we are very much encouraged.”

To test the tool, Professor Dawson and colleagues recruited 993 toddlers aged 16-38 months with an average age of 21 months, the period when autism is commonly diagnosed in children.

Forty of the toddlers were subsequently and independently identified with an autism spectrum disorder using the current gold standard diagnostic methods.

Next to the bubble-blowing video was another short clip of a cheerful woman playing with a top. As with the bubble-blowing man, the woman sat on one side of the screen and the toys on the other.

Using the eye-tracking technology, the team found that toddlers without autism scanned the entire screen while watching the videos, while in the case of the spinning top, people with autism, for example, turned their gaze to the side with the toy.

Similar differences in eye tracking patterns were observed between children with and without autism for different movies played in the app.

“This was the engineering achievement many years in the making,” Chang said.

“It is amazing how far we have come to achieve this ability to assess the gaze of the eyes without specialized equipment, using a common device many have in their pocket.”

In one video, a man blowing bubbles occupies the left side of the screen, while on the other side his bottle of bubble mixture and toys are on a shelf (as shown)

In one video, a man blowing bubbles occupies the left side of the screen, while on the other side his bottle of bubble mixture and toys are on a shelf (as shown)

With their initial work completed, the researchers are now conducting further validation studies – including those involving infants as young as six months old, to see if the app can also identify differences in children during the first year of life.

“We hope that this technology will eventually provide greater access to autism screening, which is an essential first step towards intervention,” said Professor Dawson.

‘Our long-term goal is to have a well-validated, easy-to-use app that healthcare professionals can download and use, both in a regular clinic and at home.

“We still have more steps to take, but this study suggests it will one day be possible.”

The full findings of the study are published in the journal JAMA Pediatrics.

THE SIGNS AND SYMPTOMS OF AUTISM

According to the Centers for Disease Control and Prevention, people with autism have problems with social, emotional, and communication skills that usually develop before the age of three and persist throughout a person’s life.

Specific signs of autism include:

  • Reactions to smell, taste, appearance, feel or sound are unusual
  • Difficulty adapting to changes in routine
  • Cannot repeat or echo what is being said to them
  • Difficulty expressing desires with words or movements
  • Unable to discuss their own feelings or those of others
  • Difficulty with affection, such as cuddling
  • Prefer to be alone and avoid eye contact
  • Difficulty interacting with other people
  • Cannot point or look at objects when others are pointing at them

Advertisement

.