Scientists have developed an artificial intelligence (AI) system with a near-perfect record of diagnosing early autism in toddlers. According to the University of Louisville, with specialized MRI brain scans, the tool can diagnose toddlers with 98.5 per cent accuracy.

The findings will be presented in Chicago next week at a meeting of the Radiological Society of North America (RSNA). "Our algorithm is trained to identify areas of deviation to diagnose whether someone is autistic or neurotypical," Mohamed Khudri, a visiting research scholar who was part of the team that developed the system, said in a news release.

Scientists say the system relies on DT-MRI - a special technique that detects how water travels through the brain along the white matter tracks and finds the level of connectivity. It then isolates images from the scans and compares patterns in the brains of children with autism to those who have normally developed brains.

"DT-MRI captures these abnormal connections that lead to the symptoms that children with autism often have, such as impaired social communication and repetitive behaviors," said Dr. Gregory Barnes, professor of neurology and director of the Norton Children's Autism Center in Louisville, in the news release.

According to Dr. Barnes, the results also showed positive results in identifying kids who will face more extensive issues with social communication, cognitive ability, and developmental delay in later life.

According to the researchers, they tested their method on 226 children aged between 24-48 months of age from the Autism Brain Imaging Data Exchange-II. Of those, 100 were developing normally, while the rest 126 were affected by autism.

The team thereafter concluded that the AI approach was 97 per cent precise in spotting real cases of autism and was 98 per cent right in identifying those children who did not have autism. "Our approach is a novel advancement that enables early detection of autism in infants under two years of age," Khudri said. "We believe that therapeutic intervention before the age of three can lead to better outcomes, including the potential for individuals with autism to achieve greater independence and higher IQs."

According to the US Centers for Disease Control and Prevention, less than half of the kids with autism receive a developmental evaluation by the age of three years, and at least 30 per cent of those who meet the criteria for autism are not diagnosed by age 8.

Researchers say there are several reasons behind late diagnosis, which include a lack of resources at testing centers, and that is where they believe the new AI system can help. Since it details which brain pathways are affected, the report can impact function and a severity score that can be used to guide intervention. "The idea behind early intervention is to take advantage of brain plasticity, or the ability of the brain to normalize function with therapy," Barnes said.

Researchers say they are now looking for approval for the AI software from the US Food and Drug Administration.

Get Latest News Live on Times Now along with Breaking News and Top Headlines from Health and around the world.