A new artificial intelligence system could help diagnose autism at earlier ages, researchers say. The AI system uses specialized scans called DTI-MRI and a machine learning algorithm to identify atypical areas in the brain. One study author explains that DTI-MRI captures abnormal connections that lead to symptoms commonly seen in children with autism, such as impaired social communication and repetitive behaviors.
Researchers tested the system using scans previously performed on 226 children between 24 and 48 months of age. The system was 98.5% accurate in identifying those with autism, according to the results. The researchers believe that their AI approach could lead to the early detection of autism in infants under the age of two. They are currently working on obtaining FDA clearance for their AI software.