Can Brain Scans Diagnose Autism?

By Valerie Ross | June 16, 2011 3:32 pm

What’s the News: A number of recent studies have suggested that brain scans could be used to diagnose autism. Virginia Hughes investigated these claims in a report for the Simons Foundation Autism Research Initiative. While some researchers feel these tests could soon be ready for the clinic, she found, others feel that relying on the scans for diagnosis is at least premature, and perhaps entirely misguided. Some important points in her report:

How the Would-Be Autism Tests Work:

  • The studies have focused on a variety of possible indicators of autism and used several types of scans: measuring activation of brain regions as people do or experience particular things; examining neural anatomy; tracing connectivity between parts of the brain; and analyzing the electrical activity produced by neurons firing.
  • All the scans, however, have one central goal: picking out reliable, predictive differences between the brains of children with autism and unaffected children.

The Shortcomings:

  • For the scans to be useful in diagnosis, they have to be able to distinguish children with autism not only from healthy children but from children with other conditions who have similar symptoms, such as trouble with language or social interaction. Clinicians won’t be scanning children who seem perfectly healthy, after all, but children who seem to have linguistic, social, or developmental problems.
  • The current studies don’t address whether the scans are that specific, though, and many scientists doubt they can be. “Would this differentiate autism from severe speech [or] language disorder? Unlikely,” cognitive neuroscientist Susan Bookheimer said.
  • Researchers have also criticized the methodology of some of the studies. One study, for instance, compared scans from autistic children under sedation to controls who were wide awake.

The Potential Benefits:

  • Current methods of diagnosing autism rely on a clinician’s subjective evaluation of a child. A diagnostic scan would make the process more objective.
  • These behavioral exams are also expensive, running at least $600, and insurance companies often reimburse parents for only half the cost. Brain scans cost about the same, but they’re getting cheaper—and, since MRIs and EEGs count as medical tests, insurance companies would likely pay for a larger portion of the scan.
  • Even if the scans aren’t, or aren’t yet, ready as diagnostic tools, they can help scientists better understand the neural underpinnings of autism.

There’s plenty more in Hughes’ report on brain scans for detecting autism.

Image: Wikimedia Commons / Eubulides

CATEGORIZED UNDER: Health & Medicine, Mind & Brain
MORE ABOUT: autism, diagnosis, eeg, fMRI, MRI
  • Amos Zeeberg (Discover Web Editor)

    This post was edited (June 17) to give clearer credit to Virginia Hughes for her exclusive reporting.

  • Pippa

    What matters is the clinical picture, in order to ensure that children get treatment for challenging behaviours and communication deficits. There is always the risk that such a test will be used to deny treatment, which is expensive, to some children. As we don’t really know what autism is, and it may be a common behavioural end point with many different causes – I’d rather see such scans linked to treatment outcomes rather than diagnosis. Used to guide treatment they would be an asset.

  • Jay Fox

    It sounds like it is much too early for this. Keep studying it, though, and build up a much larger database. Include in that all those “healthy” kids, too. You need a better understanding of “normal” before you can point to specific areas of concern and say there’s a problem.

    Pie-in-the-sky healthcare: Scan every kid entering preschool. It would be hideously expensive, but imagine what could be learned?

  • Day Brown

    There’s a curious absence of autism in my neck of Ozark woods, which is still family farms cause the land is too steep for agribusiness. Similarly, the Amish rate is 1:15,000; which begs the question of whether there is a genetic endowment in family farm lineages, or whether the lower rates of contagious exposure, lower rates of exposure to chemical contamination in the food, air, water, and indoor environment, and/or whether kids raised on organic homegrown garden veggies do better.

  • Michelle

    Day Brown:
    Perhaps there isn’t a curious absence as much as a lack of diagnosis? People who don’t go to the doctor or can’t afford to are not necessarily healthier because they aren’t diagnosed.

  • Another Michelle

    Day Brown: Too broad a theory. Asperger’s runs in my own Ozark family farm lineage, and appears in folks raised on organic homegrown garden veggies too. It can’t be only the combinations you mention. Of course, this in turn begs the question of whether there’s only one kind of autism on one narrow spectrum, which I don’t think is the case, IMO.

  • custom wheels

    Nice read, I just passed this onto a colleague who was doing a little research on that. And he just bought me lunch since I found it for him smile Therefore let me rephrase that: Thanks for lunch!

  • JWS

    Another good reference site on this area of Autism

  • David

    As we don’t really know what autism is, and it may be a common behavioural end point with many different causes – I’d rather see such scans linked to treatment outcomes rather than diagnosis. Used to guide treatment they would be an asset.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


80beats is DISCOVER's news aggregator, weaving together the choicest tidbits from the best articles covering the day's most compelling topics.

See More

Collapse bottom bar