Studies from Japan have demonstrated the use of gaze tracking techniques in early diagnosis of autism spectrum disorders.
Mikimasa Omori, an associate professor at Waseda University, sought to determine whether children with potential ASD exhibit preferences for predictable movements (behaviors that indicate neurodevelopmental disorders).
Survey results
Professor A/Omori has developed six sets of videos of 10-second videos showing predictable and unpredictable movements that create geometric shapes. Each video pair was displayed side by side with a pre-preferential paradigm, and the methods by which study participants observed them were compared.
These observations were captured and analyzed using an eye tracker system developed by Sweden-based company Tobii.
Survey resultsPublished in the Nature Journal Scientific Reports, children with potential autism “expended quite a lot of time observing predictable movements,” suggesting that this behavior can develop over time.
“Unlike typically developing children who did not show any change in the observational pattern, children with potential ASD showed that as the symptoms of the stimulation progressed, they gradually increased with a focus on predictable movement,” O/Prof Omori said.
Researchers say this repetitive behavior characteristic of autism “may be linked to the difficult causal relationship between movement trajectories and prediction of perfect shape.”
Meanwhile, this study demonstrated how predictable movement stimuli can be used as behavioral markers for early ASD screening.
Why is it important?
Up until this study, the reason behind children with autism was spent more time observing repetitive movements, and it was unclear how this behavior evolved over time. The current study focuses solely on social communication deficits such as eye contact and language delays.
A/Omori’s study suggests that identifying such behaviors serves as an early indicator of autism in children “younger as three years old.”
We also proposed introducing a simple video observation task as part of the routine developmental tests of young children between 18 and 36 months of age to identify people at risk of ASD. Professor A/OMORI A study procedure can also be adopted for children under 18 months.
Bigger trends
Over the past few years, several research and innovations have been published to advance the diagnosis of ASD around the world.
One of them was a device that also utilized eye tracking technology, receiving a US Food and Drug Administration 510(k) clearance. Georgia-based Earlitec Diagnostics solutions support ASD diagnosis by measuring the focus and responsiveness of a child while watching a short video.
California-based Cognoa also received the US FDA’s De Novo Clearance for AI-powered software that analyzes videos of children’s behavior to help diagnose autism.
In Australia, research The University of Southern Queensland is developing a cloud-based system that can automatically detect autism from a single brain scan.
Meanwhile, in Korea Seoul National University Hospital The National Center for Mental Health has also established a living lab to observe and collect data from children to discover biomarkers and develop AI models for early diagnosis of autism.
