A significant new study shows that the human brain expands in childhood, and shrinks as people age, reports Max Kozlov in Nature. Although considered a work in progress, the collection of more than 120,000 MRI scans reveals remarkable clarity in important aspects of human brain structure. Issues of expense and access have previously prevented the compilation and assessment of such a large group of brain images, which provide illuminating data on brain development, and may eventually contribute to studies of brain abnormalities and disease.
- The first comprehensive brain development charts show how human brains expand and shrink over time.
- The study’s authors know their data could be more inclusive if they added scans from a more diverse population.
- The resulting charts plotted a number of brain metrics according to age.
- Giving credit to brain scan owners proved to be a sticky challenge.
The first comprehensive brain development charts show how human brains expand and shrink over time.
In their study published in Nature, University of Pennsylvania researcher Jakob Seidlitz and co-workers show that human brains quickly expand in youth, and shrink as people age. Rather than performing MRI scans for their study, the group collected scans from other scientists, who were willing to share data sets from their own studies. This allowed them to amass the world’s largest collection of magnetic resonance imaging (MRI) brain scans – more than 120,000.
“It is shocking how little biological information doctors have about this critical organ.” (Jakob Seidlitz, University of Pennsylvania)
This is in stark contrast to many imaging studies that feature small sample sizes and may have reproducibility issues. MRI scans are so expensive that researchers are typically limited in the number of study participants they enroll.
The study’s authors know their data could be more inclusive if they added scans from a more diverse population.
Seidlitz’s team knows that their work lacks diversity, as it relied on scans collected mostly from Europe and North America. The subjects were largely university-aged, white, affluent, and from urban areas. Three data sets came from South America, and one came from Africa – together representing about 1% of the brain images. The lack of diversity in the subjects makes it more difficult to generalize the findings. The authors maintain a website that updates their charts as they receive more scans. They caution that more information from more diverse sources must be compiled before the charts are deployed clinically – and that doctors must not misinterpret them. The charts still need to be adjusted before they are used in a clinical setting.
“If the charts are eventually rolled out to pediatricians, great care will be needed to ensure that they are not misinterpreted. A big brain is not necessarily a well-functioning brain.” (Hannah Tully, pediatric neurologist at University of Washington, Seattle)
Because of inherent brain structure variations, the scientists aggregated numerous scans from existing brain images to achieve statistically significant growth charts.
Richard Bethlehem, the study’s co-author and neuroscientist at the University of Cambridge, UK, worked with Seidlitz to ask researchers around the world to share neuroimaging information. They attributed the surprising number of positive replies to a slowing of clinical research due to the COVID-19 pandemic. In all, the team collected 123,894 MRI brain scans from 101,457 people ranging in age from fetuses to centurions. They compiled scans from normal brains and those with conditions such as autism and Alzheimer’s disease. They used statistical models that helped extract data from the images and ensure image compatibility from various types of MRI machines.
The resulting charts plotted a number of brain metrics according to age.
The researchers found that mean cortical thickness and gray matter volume peak in early human development, and that white matter found deeper in the brain peaks by approximately age 30. One surprising finding was that the amount of cerebrospinal fluid in the brain, characteristically associated with brain atrophy, increases more rapidly in older people than was previously known.
The study may allow neuroscientists to adopt a similar process to detect links between behavior and brain function, a field which a previous Nature article showed has suffered from unreliable conclusions due to a dearth of brain scans. Nico Dosenbach, a neuroscientist at Washington University in St. Louis, Missouri, called the study a “diplomatic masterpiece” that could encourage brain image aggregation on an optimum scale.
Giving credit to brain scan owners proved to be a sticky challenge.
Some scans were collected from open-access databases, but some scan owners shared closed-data images, and were named as study authors. Data sets are sometimes closed to protect health data privacy, or because researchers don’t have enough resources to release them publicly. In all, the paper cites approximately 200 authors and recognizes hundreds of others who provided scans.
Concerns about study credits illuminate issues about how researchers interact with “scientific enterprise.” In other words, such accolades have a direct effect upon funding, publication and recognition in the field. A Nature spokesperson contended that this issue was “considered carefully by the editors and authors according to our authorship policies” and that “all data sets were appropriately credited per our data citation policy.” According to Kaja LeWinn, a social epidemiologist at the University of California, it’s the responsibility of funders, publications and research institutions to reassess how contributions to brain science are properly acknowledged and rewarded, especially in large-scale studies.
About the Author
Max Kozlov is a science journalist at Nature. His work has also appeared in The Atlantic, Quanta Magazine, Science, The Scientist, St. Louis Post-Dispatch, Behavioral Scientist and The Public’s Radio.