Back to Blog List

Predicting Autism in Infants: Early Identification and Early Treatment


According to a report from the Centers for Disease Control and Prevention in April this year, one in 59 children in the U.S. is diagnosed with autism spectrum disorder.

Autism is a very complex and variable condition. It is diagnosed based on observations of behavior in children as young as 2 years old, but it is often diagnosed much later. Interventions to treat autism symptoms are more effective the earlier they start, so identifying children earlier allows for starting treatment earlier, potentially lead to better outcomes.

Search for objective measure, biomarkers, for early detection of autism is difficult because of the complexity of the conditions, the wide range in characteristics and severity and the need for a simple test that could be used widely.

Researchers are exploring a variety of methods of identifying early markers of autism including eye gaze patterns, unusual vocalization and MRI-based imaging. Researcher William J. Bosl, Ph.D., associate professor of health Informatics and Clinical Psychology at the University of San Francisco, along with researchers at Harvard Medical School and Boston Children’s Hospital conducted a study aimed at determining if information from relatively short segments of resting state EEGs could be used to predict the development of autism from an early age. EEGs measure brain electrical activity reflecting differences in how the brain processes information.


The study, published in Scientific Reports in May 2018, involved comparing EEG measurements of 99 infants at high-risk for autism (having an older sibling with autism spectrum disorder) and 89 low-risk infants. They began taking measurements with infants at age 3 months and continued through 3 years old. The EEG was done placing a net with 128 sensors on the infants’ heads while the infants sat in their mothers’ laps.

The data was analyzed with computer algorithms. They found that prediction of a diagnosis of autism versus no diagnosis was highly accurate with a predictive accuracy more than 95 percent by 9 months. The researchers were also able to predict the severity of autism symptoms—predicted severity scores using the EEG measurements were significantly correlated with actual symptom severity scores at 3 months and older.

EEGs are relatively low-cost and non-invasive and could be potentially be incorporated into well-baby checkups.

Authors suggest that further research with a larger and more diverse group could help determine if this method would be work effectively in the general population. The research was supported by the National Institute of Mental Health, the National Institute on Deafness and Other Communication Disorders and the Simons Foundation.


Bosl, WJ, Tager-Flusberg H, Nelson CA. EEG Analytics for Early Detection of Autism Spectrum Disorder: A data-driven approach. Scientific Reports, 2018; 8(1).


AutismPatients and Families


Comments (0)