Studies relying on brain scans are often unreliable, analysis finds
THURSDAY, March 17, 2022 (HealthDay News) — Most brain studies that rely on MRI scans don’t include enough people to provide reliable results, researchers say.
These brain-wide association studies use MRI (magnetic resonance imaging) to see how brain structure and function are related to personality, behavior, thinking, neurological disorders and mental illness.
Such studies require thousands of participants to obtain accurate results, but typically recruit only a few dozen people, according to the authors of a report published March 16 in the journal Nature.
These “underpowered” studies may coincidentally show strong but incorrect associations while missing real but weaker associations, the researchers explained.
“For decades we have highlighted the potential of MRI to aid in clinical care – including diagnosis, risk, response to treatment, etc. – for mental health disorders and neurological conditions. However , the full potential has not been realized,” said lead author Damien Fair. He is director of the Masonic Institute for the Developing Brain at the University of Minnesota in Minneapolis.
“We now know our missteps and are redefining the required parameters, the so-called ‘special sauce,’ to move forward effectively,” Fair added in a university press release.
For the study, researchers analyzed data from nearly 50,000 participants in brain-wide association studies. In these studies, the median sample size was 25, which means that half had fewer participants, the other half had more.
Fair’s team found that the associations identified in such a small sample usually did not replicate in a separate sample.
Only as the sample size grew into the thousands were the associations more likely to be replicated.
The study’s lead author, Dr. Nico Dosenbach, said The New York Times that enrolling study participants can be time-consuming and expensive, ranging from $600 to $2,000 per hour. Studies that use MRI often include a note about short stature.
“It’s not a problem with any individual researcher or study. It’s not even unique to neuroimaging,” said Dosenbach, an associate professor of neurology at Washington University in St. Louis. “The field of genomics discovered a similar problem about a decade ago with genomic data and they took steps to address it.”
Dosenbach noted that the U.S. National Institutes of Health have begun funding larger data collection efforts and requiring data to be shared publicly. It reduces bias, he said, and genome science has improved a lot as a result.
“Sometimes you just need to change the research paradigm,” Dosenbach said. “Genomics showed us the way.”
If all the data from several small studies were pooled and analyzed together, the result would likely be reliable, he noted.
Fair said the future lies in sharing data and resources between institutions to make large datasets available to any scientist who wants to use them.
“This article is an amazing example of that,” Fair said.
There is more on MRI at the American Academy of Family Physicians.
SOURCE: University of Minnesota, press release, March 16, 2022