Dubai researchers create AI tool for earlier autism screening in children

Dubai: A research team led by the University of Dubai has developed an artificial intelligence-powered screening tool designed to identify early signs of autism spectrum disorder (ASD) in children, providing a faster and more accessible way to support early intervention.

The technology combines eye-tracking analysis with developmental milestone data to assess whether a child may require a specialist evaluation. Researchers said the system could help reduce diagnostic waiting times by identifying children at higher risk of autism at an earlier stage.

Autism can often be detected during a child's first two years of life, yet many children are not formally diagnosed until around the age of four, delaying access to intervention programmes that are known to improve long-term developmental outcomes.

Unlike many existing screening methods, which rely primarily on parent-completed questionnaires such as the Modified Checklist for Autism in Toddlers (M-CHAT), the new system objectively analyses a child's eye movements while they watch short videos featuring both social and geometric scenes. 

Parents simultaneously complete a developmental milestone questionnaire, with the AI model combining both data sets to generate an early risk indicator.

The system evaluates raw eye-tracking data, including gaze patterns, fixation duration and visual scanning behaviour, without first converting the information into images, a process commonly used by conventional AI models. 

Researchers said the streamlined approach allows the software to operate more efficiently on tablets and smartphones without requiring high-performance computing equipment.

The screening process takes only two to three minutes, is non-invasive and has been designed for use in clinics, with the potential for future home-based screening to broaden access to early detection services.

Initial benchmark testing showed an accuracy of around 96 per cent, which researchers described as promising. 

They stressed, however, that the tool is intended to assist with early screening and prioritise referrals for specialist assessment, rather than replace a formal medical diagnosis. Clinical validation studies are continuing across different healthcare settings.

The project is funded through the Dubai Research, Development and Innovation (RDI) Grant Initiative, established within the Dubai Future Foundation, and is led by the College of Engineering and Information Technology at the University of Dubai in collaboration with Emirates Health Services (EHS), Al Amal Psychiatric Hospital, and researchers from the University of New South Wales, the University of Wollongong in Dubai and Macquarie University. The partners aim to translate the research into a practical screening tool that can be used in clinics and, eventually, in homes.