New deep learning model screens autism from facial features during attention tasks
Bioengineer.org
A new dual-branch deep learning model screens autism in children by analyzing facial features during attention tasks in a virtual reality classroom, offering a faster and more objective alternative to traditional assessments. The model, called GDFN, achieved strong performance using a diverse dataset and incorporates explainable AI to clarify its decisions. This approach addresses the limitations of current subjective and resource-intensive diagnostic methods, potentially enabling earlier and more consistent autism screening in educational settings.
All articles from 10 September 2026. Summaries are AI-generated and may contain errors; the linked article is the authority.