Mental health disorders such as depression, anxiety, and ADHD affect millions worldwide and contribute significantly to the global burden of disease—impacting individuals' health and daily lives and placing a substantial strain on social systems and national and global economies.
These conditions often involve differences in brain development, particularly during childhood and adolescence. However, tracking these brain changes in relation to psychopathology—the study of mental health disorders—has been limited by the need for large, diverse neuroimaging datasets.
Now, a collaborative team led by Theodore D. Satterthwaite and Golia Shafiei of the Perelman School of Medicine and Michael P. Milham of the Child Mind Institute introduces Reproducible Brain Charts (RBC), a large-scale, open data resource for mapping brain development and its associations with mental health. Their results are published in Neuron .
"To create this big data resource, we did all the painful, unsexy stuff—data organization, image processing, and quality assurance," says Satterthwaite, the McLure II Professor of Psychiatry & Behavioral Research in the Department of Psychiatry and the director of the Penn Lifespan Informatics and Neuroimaging Center (PennLINC) .
RBC integrates data from five large studies of brain development in children and young adults across three continents and "harmonizes" the different mental health assessment tools they used. Satterthwaite notes that this type of task is labor intensive and requires computing power that is beyond the scope of the resources of most scientists. "We basically teed it up for everyone, so they just do science and run faster."
"The resource includes over 6,000 participants' brain MRIs," says Shafiei, a postdoctoral fellow in the Satterthwaite lab. Previously, she says, it had been difficult for researchers to fully harness the available data to map brain development from childhood through young adulthood—the challenge being that the brain scans from the studies are released and processed differently, making them hard to integrate.
Taking a similar approach with the mental health assessments, they harmonized the data for psychiatric symptoms, says Satterthwaite. The RBC includes data for "major symptom domains for each of the kids for whom we have measures of brain structure and brain function," he says. This will allow researchers to better understand how the brain develops normally and how variation in mental health symptoms is linked to variation in brain development, he adds.
"We just all put it together in one convenient place, accompanied by a website with simple instructions on how to get the data," says Satterthwaite. "This resource makes it really easy to look at brain development using massive samples of easy-to-use, de-identified data."
Beyond providing a large new open data resource for the community, RBC also offers a transparent and reproducible workflow for large-scale data integration and sharing and may serve as a model for future multistudy efforts that others can adopt or adapt.
"RBC serves as an important resource that future studies can build on," says Shafiei. "We already see it accelerating research—the data has been downloaded nearly 4,000 times even though it just came out."
Theodore D. Satterthwaite is the McLure II Professor of Psychiatry & Behavioral Research in the Department of Psychiatry and the director of the Penn Lifespan Informatics and Neuroimaging Center (PennLINC) at the Perelman School of Medicine at the University of Pennsylvania .
Golia Shafiei is a postdoctoral fellow in the Satterthwaite lab at Penn.
Other authors are Monica E. Calkins, Matthew Cieslak, Christos Davatzikos, Raquel E. Gur, Ruben C. Gur, Tyler M. Moore, Russell T. Shinohara, and Taylor Salo of the University of Pennsylvania; Xi-Nian Zuo of Beijing Normal University; Lei Ai, Jon Cluce, Nathalia B. Esper, Steven Giavasis, Gregory Kiar, and Connor Lane of the Child Mind Institute; Kahini Mehta and Nim Tottenham of Columbia University; Michael P. Milham of the Child Mind Center and Nathan Kline Institute; Alexandre R. Franco of the Child Mind Institute, Nathan Kline Institute, and NYU Grossman School of Medicine; Giovanni A. Salum of the Child Mind Institute, Universidade Federal do Rio Grande do Sul (UFRGS), National Institute of Developmental Psychiatry & National Center for Innovation and Research in Mental Health, and the Medical Council UNIFAJ & UNIMAX; Andrea P. Jackowski and Pedro M. Pan of Federal University of São Paulo (UNIFESP); Andrew A. Chen of the Medical University of South Carolina; Stanley Colcombe of Nathan Kline Institute and NYU Grossman School of Medicine; Tinashe M. Tapera of Northeastern University; Sydney Covitz of Standford University; Mauricio S. Hoffman of Universidade Federal de Santa Maria (UFSM), Universidade Federal do Rio Grande do Sul (UFRGS), National Institute of Developmental Psychiatry & National Center for Innovation and Research in Mental Health, and the London School of Economics and Political Science; Luis A. Rohde of Universidade Federal do Rio Grande do Sul (UFRGS); and Ariel Rokem of the University of Washington.
RBC was supported by the National Institute of Mental Health (R01MH120482) with additional support provided by R37MH125829, R01EB022573, R01MH112847, R01MH113550, RF1MH121867, R01MH123550, U24NS130411, P50MH109429, R01MH123440, and K08MH079364), the Canadian Institutes of Health Research (G.S. postdoctoral fellowship), and the AWS Open Data Sponsorship Program (storage support). The Developmental Chinese Color Nest Project (CCNP) study receives funding support from the STI 2030-the major projects of the Brain Science and Brain-Inspired Intelligence Technology (2021ZD0200500), the National Basic Science Data Center "Interdisciplinary Brain Database for In-vivo Population Imaging" (ID-BRAIN), the Key-Area Research and Development Program of Guangdong Province (2019B030335001), the Start-up Funds for Leading Talents at Beijing Normal University, the Beijing Municipal Science and Technology Commission (Z161100002616023 and Z181100001518003), the Major Project of National Social Science Foundation of China (20&ZD296), the CAS-NOW Programme (153111KYSB20160020), the Guangxi BaGui Scholarship (201621), the National Basic Research (973) Program (2015CB351702), the Major Fund for International Collaboration of National Natural Science Foundation of China (81220108014), the Chinese Academy of Sciences Key Research Program (KSZD-EW-TZ-002), and the National Basic Research Program (2015CB351702). The Philadelphia Neurodevelopmental Cohort was supported by NIMH RC2 MH089983 and RC2 MH089924. The Brazilian High Risk Cohort (BHRC) was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq grant numbers 573974/2008-0 and 465550/2014-2), Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP grant numbers: 2008/57896-8, 2013/08531-5, 2014/50917-0, 2020/06172-1, 2021/05332-8, and 2021/12901-9) to the National Institute of Development Psychiatric for Children and Adolescent (INPD), the European Research Council under the European Union's Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no 337673.
