West Virginia University researchers are using artificial intelligence and other advanced technologies to help diagnose people with autism.
The program is aimed at more easily identifying phenotypes related to Autism Spectrum Disorder. These phenotypes are noticeable traits or characteristics a person with ASD might have.
“Autism phenotyping is something we are still in the dark ages with. We have no clue how many different types of autism we are dealing with,” said WVU professor Xin Li, one of the project’s head researchers.
Technology like neural imaging and behavior imaging, along with eye-tracking data will help identify these specific traits. Li says he hopes this data will find different types of ASD and help reduce the gap between a child’s birth and their diagnosis. The average age of a child newly diagnosed with ASD is 4 years old — Li says part of the goal of this research is to reduce that age in half, aiming for diagnoses at 2 years old. The earlier the diagnosis, Li says, the more effective the treatment.
Li says this research is important because of how little is known about ASD compared to other disorders. The better the technology available to diagnose those with ASD, the better phenotypes can be successfully grouped into ASD subtypes.
“If we think about something we’re familiar with — for example, a butterfly… a butterfly can have different wings, have different patterns, colors… Those are the easy traits for laymen to tell a different species from one butterfly to another one,” Li said.
Recent data from the Centers for Disease Control and Prevention says 1 in 54 children in the U.S. are diagnosed with ASD.