Physicians are encouraged to look at brain scan data in a new way. According to a study at Washington University School of Medicine in St. Louis, doctors should be able to analyze the development of a child’s brain and also keep track of any possible psychological or developmental disorders after a typical five-minute scan.
“Pediatricians regularly plot where their patients are in terms of height, weight and other measures, and then match these up to standardized curves that track typical developmental pathways,” says senior author Bradley Schlaggar, MD, PhD, a Washington University pediatric neurologist and the A. Ernest and Jane G. Stein Associate Professor of Neurology.
“When the patient deviates too strongly from the standardized ranges or veers suddenly from one developmental path to another, the physician knows there’s a need to start asking why.”
Schlaggar and his colleagues propose a new way of looking at brain scanning data that moves beyond observing the brain from only a structural point of view. This would be especially helpful in the monitoring and treating of patients with psychiatric and developmental disorders.
According to Schlaggar, he has sent children with obvious, profound psychiatric conditions for MRI scans and received results marked “no abnormalities noted.”
“That’s typically looking at the data from a structural point of view—what’s different about the shapes of various brain regions,” he adds. “But MRI also offers ways to analyze how different parts of the brain work together functionally.”
By comparing functional data with standardized models of how brain function or disease typically develops, Schlaggar says, a variety of new clinical insights becomes available.
Schlaggar and his colleagues use an approach to brain scanning called ‘resting state functional connectivity.’ As a patient rests in the scanner, scientists analyze increases and decreases in blood flow to the various brain regions and then determine if and how these regions work together in brain networks.
In a study published last year, Washington University scientists demonstrated how these brain networks change as the brain matures. In summary, they found that the overall brain system converts from closely-knit networks in the child’s brain to networks that are able to connect distant regions—the typical organization in an adult brain.
For the new study, lead author Nico Dosenbach, MD, PhD, a pediatric neurology resident at St. Louis Children’s Hospital, took this and other distinctions that mark the transition from child to adult brain and adapted them for use in a technique for mathematical analysis called a support vector machine.
“It’s a way that mathematicians have developed for predicting something with high specificity and sensitivity when you have huge amounts of data instead of one really good measurement,” Dosenbach explains.
“Any one of these measurements doesn’t tell you much, but if you put them together and use the right math to sift through and restructure them, you can get good predictive results.”
Dosenbach used data from five-minute MRI scans of 238 normal subjects between the ages of 7 and 30. The support vector machine analyzed approximately 13,000 functional brain connections and selected the best 200 to create a single index of the maturity of each subject. This information allowed researchers to predict whether subjects were children or adults and then form a curving line that tracks the path of normal functional brain development.
The plan is that patients with brain abnormalities will appear out of sync with this normal developmental curve.
“The beauty of this approach is that it lets you ask what’s different in the way that children with autism, for example, are off the normal development curve versus the way children with attention-deficit disorder are off that curve,” said Schlaggar.
He suggests that functional brain scans be conducted on children at risk but not yet suffering from a developmental disorder.
“When a fraction of them later develop that disorder, you can go back and construct an analysis like this one that will help predict the characteristics of the next child at highest risk of developing the disorder,” he says.
“That’s very powerful both clinically and from the perspective of understanding the causes of these disorders.”
This approach could enable treatment before any onset of symptoms, Schlaggar says, and should help physicians more quickly and closely track the results of clinical trials of new therapies.
“MRI scans are expensive, so this may not be what we use for everyone right now,” Dosenbach says. “But many children with these types of disorders already receive regular structural MRI scans, and five more minutes in the scanner won’t add that much to the cost.”
The study is featured this week in Science.