Dubai researchers create AI tool for earlier autism screening in children

Dubai researchers from the University of Dubai have created an AI-powered tool to identify early signs of autism spectrum disorder (ASD) in children. The innovative system uses eye-tracking analysis combined with developmental milestone data to assess whether a child may need a specialist evaluation. By combining eye movement data from short videos featuring social and geometric scenes with a parent-completed developmental milestone questionnaire, the AI model generates an early risk indicator.

This approach offers a faster, more accessible method for early screening, potentially reducing diagnostic waiting times. Unlike other methods that rely on questionnaires, this tool directly analyzes eye movements without converting the data into images, making it compatible with tablets and smartphones. The screening process takes only two to three minutes and is non-invasive.

Initial tests showed a 96 percent accuracy, though the tool is meant to assist with early screening and prioritize referrals for expert evaluation, not replace a formal diagnosis. The project, funded by the Dubai Research, Development and Innovation Grant Initiative, is led by the University of Dubai in collaboration with various healthcare institutions.

The goal is to translate this research into a practical screening tool for clinics and, eventually, for home use.