Brain Patterns Distinguish Autism With Striking Accuracy

Deborah Brauser

April 23, 2013

Analyzing brain activity patterns in children may allow confirmation of an autism spectrum disorder (ASD) diagnosis with striking accuracy, new research suggests.

The study showed that use of magnetoencephalography (MEG) to measure functional connectivity (the communication from 1 region of the brain to the other) and background noise in 19 children provided an accuracy rate of up to 94% in differentiating those with and those without an ASD.

The group with an ASD had significantly stronger connections between the rear and frontal areas of the brain and lower spatial complexity patterns.

"The result that was surprising was that with our method, we can estimate inputs to the brain. We can separate between neural activity vs the inputs that aren't driving that activity, which we refer to as 'background noise,' " principal investigator Roberto F. Galán, PhD, assistant professor of neuroscience at Case Western Reserve University School of Medicine in Cleveland, Ohio, and a trained electrophysiologist, told Medscape Medical News.

"The age range was pretty broad, and there were differences in their IQs," added Dr. Galán. "But despite the variability in the patient population, they all shared some brain pattern traits that made them very different from the control group."

Dr. Roberto Galán

The investigators noted in a release that their method greatly outperformed other algorithms in accurately discriminating between a typical and an atypical brain and "offers an efficient, quantitative way of confirming a clinical diagnosis of autism."

"Altogether our results demonstrate a promising potential of our approach as an efficient biomarker for altered brain dynamics associated with a cognitive phenotype," they write.

The study was published online April 17 in PLoS One.

Examining Neural Activity

Dr. Galán said that the researchers wanted to see if they could determine cognitive features of the brain just by examining neural activity.

He added that several techniques are currently being used to measure brain activity, including MEG, electroencephalography (EEG), and functional magnetic resonance imaging (fMRI). MEG was chosen for this study because of its ability to track "fast changes" in neural activity.

MEG measures magnetic fields generated by electric currents in neurons of the brain. Dr. Galán noted in a press release that currently, most measures of functional connectivity do not examine the directionality of a brain's interactions.

"It is not just who is connected to whom, but rather who is driving whom," he said.

For the current study, the researchers sought to examine both functional connectivity and the brain's background noise, or "the spontaneous input driving the brain's activity while at rest," as it related to autism in 19 children.

Of these, 9 had Asperger's syndrome ("ASD group"; 100% boys; mean age, 10.8 years) and 10 had no ASD ("control group"; 60% boys; mean age, 11.2 years).

Each participant underwent MEG, in which 141 sensors tracked brain cortex activity and recorded interactions between various regions while at rest, including the occipital, frontal, central, parietal, and temporal areas.

These interactions were then compared between the 2 groups of children.

Brain Differences

Results showed significant overall functional connectivity differences between the groups that did and those that did not have an ASD, with a discrimination accuracy rate of 84%.

The children with an ASD had significantly more "enhanced functional excitation from occipital to frontal areas along a parasagittal axis" than their healthy peers, report the investigators, adding that this may be caused by changes in white matter density.

In addition, analysis of background noise patterns showed that the group with ASD had significantly less variety and intricacy than did the group without ASD, who showed more spatial complexity and structure.

"Background noise in ASD patients is spatially correlated over wide areas, as opposed to control, where areas driven by correlated noise form smaller patches," write the researchers.

"In other words, there is more variability in the spatial pattern of the background activity (or noise) in children without autism," they add.

This measure was a stronger predictor of an ASD than functional connectivity alone, with a 94% discrimination accuracy rate.

"We asked the question, 'Can you distinguish an autistic brain from a nonautistic brain simply by looking at the patterns of neural activity?,' and indeed, you can," said Dr. Galán.

"This discovery opens the door to quantitative tools that complement the existing diagnostic tools for autism based on behavioral tests," he added.

Dr. Galán noted that he is excited about the possibility of other studies delving deeper into this topic.

"We are actually reaching out to other labs and other teams to see if they can replicate our results. Because if they can, I think this will show us something very fundamental about the autistic brain. I would like to collaborate with other teams who have access to this type of patient population and who can do imaging recordings. Then let's see if this is consistent across a wide population of autistic children or whether it's more specific to Asperger's children," he said.

He added that he would also like to see whether this measure would work for other psychiatric conditions, such as schizophrenia.

The study was funded in part by a grant from the Natural Sciences and Engineering Research Council of Canada and by a scholarship from the Mt Sinai Health Care Foundation. The study authors have reported no relevant financial relationships.

PLoS One. Published online April 17, 2013. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.