University of Louisville researchers say they have developed an AI-powered tool that could help diagnose autism at an earlier age.
According to their results published in the journal Biomedicines, the new tool has been shown to be 98.5% accurate in kids as young as two. The tests involved 226 children between the ages of 2 and 4 years old and were able to identify the around 120 children with autism.
The technology was co-invented by professor and chair in the J.B. Speed School of Engineering Ayman El-Baz, professor of neurology and executive director of the UofL Autism Center Gregory Barnes, and former UofL neuroscientist Manuel Casanova.
According to a release, less than half of kids are tested before age three and even few are diagnosed by age eight with research saying a big problem is too many patients and too few specialists to do interviews and examinations needed for diagnosis. A diagnosis could mean therapy, which research has shown to have the most impact if done in early childhood.
The release says that AI could reduce the specialist workload in the initial diagnosis by as much as 30%. The tool looks at the physical structure of the brain using MRIs instead of using interviews. The researchers believe this will lead to a more objective diagnosis and identify specific parts of the brain that could benefit most from therapy.
Source: WAVE
You must be logged in to post a comment.