Anticipating autism through functional neuroimaging

by Nancy Fliesler on June 11, 2014

Mila-cap-ball-2014-05-08 12.24.14Is 9-month-old Mila Goshgarian at risk for developing autism spectrum disorder (ASD)? Her 4-year-old twin brothers are both on the spectrum, so statistically her chances are at least 20 percent.

Her mother, Tonia, brought her into Boston Children’s Hospital for the Infant Sibling Project, which works with babies who are at increased risk of developing ASD in hopes of discovering early brain biomarkers for the disorder. This is Mila’s fifth visit; she’s been coming to the Labs of Cognitive Neuroscience for testing since the age of 3 months.

“I wanted to keep a close eye on her development,” says Tonia, who learned of the study when she was pregnant with Mila. “I was very eager to get her in.”

The Infant Sibling Project team is exploring several potential biomarkers for ASD, including differences on brain electroencephalograms (EEGs), eye-tracking and now—the test Mila is having today—brain oxygen consumption, measured by functional near-infrared spectroscopy (fNIRS).

NIRS cap-flip side-2014-05-08 13.29.45Much like a pulse oximeter that you see in a doctor’s office, fNIRS technology uses light to monitor changes in blood oxygen. The cap Mila is wearing contains two dozen light emitters, as well as optical sensors that read how much oxygen different parts of her brain are using based on how much light reflects back. (The more oxygen being used, the more active an area is.) By performing fNIRS while children receive social stimuli—pictures of their mothers’ and strangers’ faces, both neutral and happy, and recordings of different speech sounds—the researchers hope to identify differences in how children like Mila, at relatively high risk for ASD, process such stimuli.

NIRs readouts

The greater the oxygenated hemoglobin response, the more active that brain area is likely to be.

Back in the lab’s control room, research assistant Sarah Mumanachit queues up the stimuli and oversees the recordings. While there’s a slight delay in the readout, fNIRS is able to localize brain activity at very high resolution and is less affected by eye-blinks and other motions than EEGs.

To date, about 275 children have been enrolled in the Infant Sibling Project, including control infants without an affected sibling, and about 80 have had the fNIRS testing. Ultimately, all will be followed to the age of 3 to see if they develop autism and whether this correlates with any of their early brain measures. So far, only about 150 children have known outcomes; complete data analysis is still to come.

But in preliminary results from 27 high-risk and 37 control infants, the control group showed increasing connectivity of NIRS responses from the back of the brain to the front between 3 and 12 months of age, while the high-risk group showed a reduction in connectivity. How individual findings may correlate with an actual ASD diagnosis remains to be seen, but the results are consistent with differences in brain networks and brain structure in ASD that the Nelson lab and others have observed with EEG and MRI.

“Currently, an ASD diagnosis is made based on certain behaviors, many of which don’t begin to appear until 2 or 3 years of age, even later in some cases,” says Charles Nelson, PhD, one of the principal investigators on the project. “The benefit of identifying a reliable brain biomarker for ASD is that we would no longer have to wait for those behaviors to come on line to make an accurate diagnosis. This would translate to much earlier interventions, leading to better outcomes and possibly even the prevention of certain behaviors typically associated with ASD.”

Leave a comment

Previous post:

Next post: