Researchers from the United States have developed an application that can diagnose autism in children under the age of 16 months by tracking eye movement. It is reported by analyzing patterns of outlook.
The application uses the camera on iPad or iPhone devices with machine learning, to track and analyze these eye movements while children watch private videos.
Eye-tracking has been used to diagnose autism before, and it is the first time that this has been done without special devices and an expert to interpret gaze patterns.
The technique is used everywhere and takes only 10 minutes to test a child, and the new approach can easily be deployed in-home tests.
Lead research and child clinical psychologist Geraldine Dawson of Duke University in North Carolina said, we know that children with autism care for the environment differently and don't care about people as much, that we can track eye patterns in young children to assess the risk of autism.
This is the first time that we have been able to provide this type of assessment using only a smartphone or tablet.
Professor Dawson and her colleagues conducted the study on 993 children between the ages of 16 and 38 months with a mean age of 21 months, which is the period during which autism is most often recognized in children.
Forty young children were subsequently independently identified as having autism spectrum disorder using current gold-standard diagnostic methods.
The team found that young, non-autistic children scanned the entire screen while watching the videos, while in the case of the spinning top, for example, people with autism focused their gaze on the side of the game and not the person using it.
Similar differences in eye gaze patterns among autistic and non-autistic children have been seen in several films shown in the app.