X

Brain scans could one day help diagnose autism earlier

Researchers say MRI scans show very specific brain activity that could help diagnose autism and aid people in determining early treatment options.

Elizabeth Armstrong Moore
Elizabeth Armstrong Moore is based in Portland, Oregon, and has written for Wired, The Christian Science Monitor, and public radio. Her semi-obscure hobbies include climbing, billiards, board games that take up a lot of space, and piano.
Elizabeth Armstrong Moore
3 min read
This brain scan shows weaker neural connectivity in participants with autism compared with the control participants. University of Alabama at Birmingham

Brain scans may reveal signs of autism, which could eventually aid in early intervention therapies, according to new research.

Researchers using a functional magnetic resonance imaging (fMRI) scanner measured the brain activity of volunteers with autism spectrum disorders against controls and say the comparisons reveal disrupted brain connectivity that could serve as a neural signature of autism.

While the study is both preliminary and small -- including only 30 volunteers -- the findings, which appear online Friday in the journal Frontiers in Human Neuroscience, joins a wider range of autism research that could ultimately help supplement current behavior-based diagnoses and possibly help in deciding early intervention therapies.

"This research suggests brain connectivity as a neural signature of autism and may eventually support clinical testing for autism," said Rajesh Kana, an associate professor of psychology and the project's senior researcher. "We found the information transfer between brain areas, causal influence of one brain area on another, to be weaker in autism. There's a very clear difference."

The joint work from the University of Alabama at Birmingham Department of Psychology and Auburn University looked at 15 high-functioning adolescents and adults with autism and 15 control participants ages 16 to 34 years. The team found that brain connectivity data from 19 paths was able to predict which volunteers had autism with 95.9 percent accuracy.

"These are sophisticated, exploratory techniques a long ways from clinical application, but [it's] encouraging that we are starting to get more and more accurate classifications," says Jeffrey Anderson, an assistant professor of neuroradiology at the University of Utah who has conducted similar scans. "It's a smaller step from there to understanding the disease."

For the study, the 30 participants watched a series of comic-strip stories while in an fMRI scanner that measured their brain activity. They were asked to choose the most logical ending out of three options -- with scenes ranging from a glass about to fall off a table to a man enjoying the music of a street violinist. The participants with autism had a harder time finding a logical conclusion to the violin scenario, which involved subtle social cues, and their scans reflected this difficulty.

Basically, Kana said, the disrupted connectivity results in "consistently weaker brain regions."

Kana said that over the next five to 10 years, the goal is to begin to supplement current behavior-based diagnoses (which starts at the very earliest at 18 months) with more objective medical testing, as well as analyze interventions that aim to improve these connectivity issues.

"Parents usually have a longer road before getting a firm diagnosis for their child now," Kana said. "You lose a lot of intervention time, which is so critical. Brain imaging may not be able to replace the current diagnostic measures, but if it can supplement them at an earlier age, that's going to be really helpful."

Update, 4:40 p.m. PT: Added a comment from Jeffrey Anderson, an assistant professor of neuroradiology at the University of Utah.