1. Normative Modeling of Brain Morphometry Across the Lifespan using CentileBrain: Algorithm Benchmarking and Model Optimization
- Author
-
Ruiyang Ge, Yuetong Yu, Yi Xuan Qi, Yunan Vera Fan, Shiyu Chen, Chuntong Gao, Shalaila S Haas, Amirhossein Modabbernia, Faye New, Ingrid Agartz, Philip Asherson, Rosa Ayesa-Arriola, Nerisa Banaj, Tobias Banaschewski, Sarah Baumeister, Alessandro Bertolino, Dorret I Boomsma, Stefan Borgwardt, Josiane Bourque, Daniel Brandeis, Alan Breier, Henry Brodaty, Rachel M Brouwer, Randy Buckner, Jan K Buitelaar, Dara M Cannon, Xavier Caseras, Simon Cervenka, Patricia J Conrod, Benedicto Crespo-Facorro, Fabrice Crivello, Eveline A Crone, Liewe de Haan, Greig I de Zubicaray, Annabella Di Giorgio, Susanne Erk, Simon E Fisher, Barbara Franke, Thomas Frodl, David C Glahn, Dominik Grotegerd, Oliver Gruber, Patricia Gruner, Raquel E Gur, Ruben C Gur, Ben J Harrison, Sean N Hatton, Ian Hickie, Fleur M Howells, Hilleke E Hulshoff Pol, Chaim Huyser, Terry L Jernigan, Jiyang Jiang, John A Joska, Rene S Kahn, Andrew J Kalnin, Nicole A Kochan, Sanne Koops, Jonna Kuntsi, Jim Lagopoulos, Luisa Lazaro, Irina S Lebedeva, Christine Lochner, Nicholas G Martin, Bernard Mazoyer, Brenna C McDonald, Colm McDonald, Katie L McMahon, Tomohiro Nakao, Lars Nyberg, Fabrizio Piras, Maria J Portella, Jiang Qiu, Joshua L Roffman, Perminder S Sachdev, Nicole Sanford, Andrew J Saykin, Theodore D Satterthwaite, Sophia I Thomopolous, Carl M Sellgren, Kang Sim, Jordan W Smoller, Jair Soares, Iris E Sommer, Gianfranco Spalletta, Dan J Stein, Christian K Tamnes, Alexander S Tomyshev, Theo GM van Erp, Diana Tordesillas-Gutierrez, Julian N Trollor, Dennis van 't Ent, Odile A van den Heuvel, Neeltje EM van Haren, Daniela Vecchio, Dick J Veltman, Dongtao Wei, Henrik Walter, Yang Wang, Bernd Weber, Margaret J Wright, Wei Wen, Lars T Westlye, Lara M Wierenga, Paul M Thompson, Steven CR Williams, Sarah Medland, Mon-Ju Wu, Kevin Yu, Neda Jahanshad, and Sophia Frangou
- Abstract
Background: Normative modeling is a statistical approach to quantify the degree to which a particular individual-level measure deviates from the pattern observed in a normative reference population. When applied to human brain morphometric measures it has the potential to inform about the significance of normative deviations for health and disease. Normative models can be implemented using a variety of algorithms that have not been systematically appraised. Methods: To address this gap, eight algorithms were compared in terms of performance and computational efficiency using brain regional morphometric data from 37,407 healthy individuals (53% female; aged 3-90 years) collated from 87 international MRI datasets. Performance was assessed with the mean absolute error (MAE) and computational efficiency was inferred from central processing unit (CPU) time. The algorithms evaluated were Ordinary Least Squares Regression (OLSR), Bayesian Linear Regression (BLR), Generalized Additive Models for Location, Scale, and Shape (GAMLSS), Parametric Lambda, Mu, Sigma (LMS), Gaussian Process Regression (GPR), Warped Bayesian Linear Regression (WBLG), Hierarchical Bayesian Regression (HBR), and Multivariable Fractional Polynomial Regression (MFPR). Model optimization involved testing nine covariate combinations pertaining to acquisition features, parcellation software versions, and global neuroimaging measures (i.e., total intracranial volume, mean cortical thickness, and mean cortical surface area). Findings: Statistical comparisons across models at PFDR
- Published
- 2023
- Full Text
- View/download PDF