MENU

Computer Algorithm Can Predict Babies Likely To Develop Autism

16 February 2017, 8:40 am EST By Allan Adamson Tech Times
Close
Hamburg zoo welcomes four rare Siberian tiger cubs
An algorithm that takes into account measurements from brain scans of babies as young as 6 months old can predict who among the infants will develop autism. The research could pave way for early diagnosis and early intervention.  ( China Photos | Getty Images )

Using a computer algorithm and brain scans of babies as young as 6 months old, researchers were able to predict infants who are likely to develop autism when they get older.

1 In 68 American Children Has A Form Of Autism

About one in 68 children in the United States has some form of autism, a developmental disorder characterized by difficulties in social interaction and communication as well as repetitive behaviors or interests.

Behavioral symptoms often become evident when children reach 2 to 4 years old but research suggests that those who receive the earliest treatment are likely to reap the most benefits. Researchers have now developed a new computer algorithm that can improve early diagnosis of the condition among high-risk children.

Early Intervention For High-Risk Infants

Research is helping shed light on what causes autism. A 2016 study, for instance, found a link between mitochondrial DNA, which is passed from the mother to child, to a number of forms of autism spectrum disorder. Scientists have also found that infants who have older siblings with autism have a higher likelihood of developing the condition.

The new algorithm described in the journal Nature on Wednesday, Feb. 15 offers hope for earlier intervention and better treatment outcomes.

For their study, researchers used magnetic resonance imaging to capture brain images of babies considered to have an increased risk for autism by virtue of having an older sibling with the condition.

The researchers took measurements of the brains of children between 6 and 12 months of age. They gathered information on the surface area, overall volume, and thickness of the cerebral cortex in certain regions of the brain.

The researchers then used the algorithm to combine the measurements.The algorithm allowed the researchers to predict which of the babies would develop autism by the time they reach 2 years of age with more than 90 percent accuracy.

"A deep-learning algorithm that primarily uses surface area information from magnetic resonance imaging of the brain of 6-12-month-old individuals predicted the diagnosis of autism in individual high-risk children at 24 months," the researchers wrote in their study. "These findings demonstrate that early brain changes occur during the period in which autistic behaviors are first emerging."

Implications Of The Study On Treatment Of Autism

If researchers are able to replicate the results of the study, it could have a number of implications for the treatment of autism such as the development of new behavioral therapies that may help alter the brain development of infants with increased risk for the condition.

Using data gathered from the study, the researchers also hope to determine how certain environmental factors such exposure to toxic metals and air pollution can influence the likelihood of a baby to develop autism.

"While we have known for some time that autism emerges in subtle, gradual ways over the first few years of life, this study offers the first firm evidence before a child's first birthday predicting whether certain high-risk children are likely to be diagnosed with autism," said Robert Schultz from the Center for Autism Research.

© 2017 Tech Times, All rights reserved. Do not reproduce without permission.

From Our Sponsor

Eco-friendly Packaging On Demand Technology Changes The Way Retailers Think About Shipping Packages

On-demand packaging can be eco-friendly as well as save costs. Now, who would have thought of that? Packsize did, and it's changing the business of shipping packages.
Real Time Analytics