Study: MRI can predict autism with 80 percent accuracy

Using magnetic resonance imaging (MRI) in infants who have older siblings with autism spectrum disorder (ASD), researchers were able to predict with 80 percent accuracy which infants would meet the criteria for an ASD diagnosis at age 2. Typically, the earliest an autism diagnosis can be made is between ages 2 and 3.

Intervening early could lead to improved outcomes, as the brain is more malleable in the first years of life compared with later in childhood.

“This research highlights the best of contemporary science,” says study coauthor Jed Elison, Ph.D., an assistant professor in the University of Minnesota’s Institute of Child Development and Department of Pediatrics, who worked with researchers from around the country and Canada on the study. “It’s collaborative and informed by technology and multiple areas of expertise, with the common goal of helping families.”

Researchers conducted MRI scans of infants who have an older sibling with an ASD diagnosis at 6, 12, and 24 months of age. They found that the babies who developed autism experienced a hyperexpansion of brain surface area from age 6 to 12 months, as compared with babies who had an older sibling with autism but did not show evidence of the condition at age 2. 

The increased growth rate of the brain’s surface area in a child’s first year of life was linked to an increased growth rate of overall brain volume in the second year of life. Brain overgrowth was tied to the emergence of autistic social deficits in the second year.

The researchers found that brain differences at 6 and 12 months of age in infants who have older siblings with ASD correctly predicted 8 out of 10 infants who would later meet criteria for autism at age 2. In addition, the researchers predicted with 89 percent accuracy which babies would not go on to receive an ASD diagnosis by age 2.

The study, published February 15 in the journal Nature, was led by researchers at the University of North Carolina with contributors from the University of Minnesota, University of Washington, Washington University in St. Louis, Children’s Hospital of Philadelphia, McGill University, University of Alberta, College of Charleston, and New York University. 

Published on March 28, 2017