Brain scan autism
© Rick Wilking/ Reuters
Scientists believe that there may be a correlation between brain growth visible on brain scans and the eventual development of autism spectrum disorder. These MRIs could potentially identify high-risk autism before the age of two.

Autism spectrum disorder (ASD) affects approximately 3.5 million Americans and is most likely to be identified by the age of four. However, a study published Wednesday in Nature, claims to have possibly discovered a formula to identify children with a high-risk for autism before they start to seriously lag in their social development.

ASD is a blanket term for a wide range of social impairments, from slightly to seriously disabling. As a result, children with autism often fall behind in developmental milestones related to social, communication and language skills.

The study claims that the brain scans revealed "hyperexpansions" of the brain surface in the first 12 months were a common feature in babies who would go on to develop ASD. The expansion was often followed by an increase in brain volume overgrowth that, according to the study, "was linked to the emergence and severity of autistic social deficits."

Researchers developed this conclusion by comparing developments of 106 infants who were at a high risk of autism due to family histories with 42 low-risk infants. Of the high-risk infants, 15 presented with brain overgrowth went on to develop autism by their 24th month.

Scientists then took the information from the scans and provided them to a machine learning algorithm and used that to predict autism in 24-month-old infants. Of those, 81 percent predicted to develop autism did so, and in about 3 percent, autism was predicted but did not develop.

The benefits of identifying autism before serious developments occur are numerous. Stephen Dager, a co-author at the University of Washington's Center on Human Development and Disability, told the Guardian. "Early intervention, before age two, can change the clinical course of those children whose brain development has gone awry and help them acquire skills that they would otherwise struggle to achieve."

This study could be the first step toward developing earlier intervention in ASD technology for pediatricians to apply on a larger scale.

Annette Estes, a co-author on the study and director of the University of Washington in Seattle, told the Guardian that while no tool yet exists to identify autism earlier, "researchers could start developing interventions to prevent these children from falling behind in social and communication skills