Researchers at the University of North Carolina, Chapel Hill have detected brain growth changes linked to autism in children as young as six months old. They then devised a deep-learning algorithm that used the data to predict whether a child at a high risk of autism would be diagnosed with the disorder at 24 months.
The algorithm correctly predicted the future diagnosis in high-risk children with 81-percent accuracy and 88-percent sensitivity.
The new technology outperformed behavioral questionnaires, which produce early autism diagnoses with only 50-percent accuracy.
The project, which is part of the U.S. National Institutes of Health Infant Brain Imaging Study, included 106 infants with an older sibling who had been given an autism diagnosis, and 42 infants with no family history of autism. The algorithm identified eight out of 10 kids with autism using just three variables--brain surface area, brain volume, and gender.
From IEEE Spectrum
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA