Brain age prediction in generalized anxiety disorder using a convolutional neural network

Higher predicted brain age difference has been associated with several psychiatric disorders. Generalized anxiety disorder (GAD) is associated with markers of accelerated aging. In this study, we determined brain predicted age difference (PAD) in individuals with GAD and healthy controls (HC) as well as group differences in PAD variability using voxel-wise structural MRI. The training dataset included 3511 controls, and the testing dataset included 1595 individuals with GAD and 4552 HC from the ENIGMA-Anxiety GAD Working Group. A convolutional neural network model using four input modalities per subject and a model ensemble approach was used to predict brain age. The PAD was then calculated by subtracting chronological age from the predicted age. Model performance was consistent with other image-based brain age prediction models with similar accuracy across the training set (mean absolute error (MAE) = 2.95 years) and HC in the testing set (MAE = 2.94). We found no evidence of accelerated brain aging in individuals with GAD compared to individuals without GAD, though we did find evidence for greater variation in PAD for individuals with GAD (Levene's test: W = 442.98, p < 0.001) and evidence for greater variability in PAD of those with GAD over 25 years of age. In several exploratory analyses, we found that symptom severity related significantly to PAD, even after controlling for medication and comorbid diagnoses, echoing previous brain age research. These findings underscore the need for consideration of heterogeneity and dimensionality of psychopathology when examining brain age predicted differences.

Authors

Institutions

Publication Details

Journal
Translational Psychiatry
Published
2026-05-23
DOI
https://doi.org/10.1038/s41398-026-04078-3
Primary Topic
Mental Health via Writing
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Brain age prediction in generalized anxiety disorder using a convolutional neural network

Katja Beesdo‐Baum, Jeffrey Robert Strawn, Marta Cano, Ulrike Lueken et al.
Translational Psychiatry
Mental Health via Writing
article

Brain age prediction in generalized anxiety disorder using a convolutional neural network

Katja Beesdo‐Baum, Jeffrey Robert Strawn, Marta Cano, Ulrike Lueken, Ellen Leibenluft, Gregory A. Fonzo, Narcis Cardoner, Elise M. Cardinale, Rebecca B. Price, C. Ottaviani, Gretchen J. Diefenbach, Lilianne R. Mujica‐Parodi, Helena van Nieuwenhuizen, Michal Assaf, Qiongru Yu, Grace Ringlein, Heidi K. Schroeder, Parmis Khosravi, Camilla Cividini, Anita Harrewijn, Nynke A. Groenewold, Nicholas L. Balderston, Antonia N. Kaczkurkin, Katharina Wittfeld, André Zugman, J. R. Blair, Chelsea Sawyers, Michael T. Perino, Janna Marie Bas‐Hoogendam, Frances Meeten, Carmen Andreescu, Monique Ernst, Ana Munjiza Jovanović, P. M. Pan, Benson Mwangi, Milutin Kostić, Brenda Benson, Daniel Porta-Casteràs, David Hofmann, Corey Richier, Sandra van der Auwera, Jair C. Soares, Giovana B. Zunta-Soares, Sophia I. Thomopoulos, Thomas Straube, Anderson M. Winkler, Mohammed R. Milad, Andrea P. Jackowski, Savannah Gosnell, Robin Bülow, Jennifer Harper, Karina S. Blair, Paolo Brambilla, Neda Jahanshad, Martin P. Paulus, Patrick McClure, Daniel S. Pine, Hans J. Grabe, Kathryn Werwath, Mon-Ju Wu, Jared A. Nielsen, Eleonora Maggioni, Ruben C. Gur, Elena Makovac, Giovanni A. Salum, Henry Völzke, Francisco Pereira, Gisele G. Manfro, Raquel E. Gur, Willem B. Bruin, Paul M. Thompson, Murray B. Stein, Dan J. Stein, Elisa Canu, Gabrielle F. Freitag, Dick J. Veltman, Hugo D. Critchley, Jordan W. Smoller, Hannah Zwiebel, Bart Larsen, Katie L. Burkhouse, Massimo Filippi, Ramiro Salas, Federica Agosta, Nic J. A. Van der Wee, Matteo Mancini, Moji Aghajani, Kevin Hilbert, Theodore D. Satterthwaite, K. Luan Phan, Chad M. Sylvester
article en

Abstract

Higher predicted brain age difference has been associated with several psychiatric disorders. Generalized anxiety disorder (GAD) is associated with markers of accelerated aging. In this study, we determined brain predicted age difference (PAD) in individuals with GAD and healthy controls (HC) as well as group differences in PAD variability using voxel-wise structural MRI. The training dataset included 3511 controls, and the testing dataset included 1595 individuals with GAD and 4552 HC from the ENIGMA-Anxiety GAD Working Group. A convolutional neural network model using four input modalities per subject and a model ensemble approach was used to predict brain age. The PAD was then calculated by subtracting chronological age from the predicted age. Model performance was consistent with other image-based brain age prediction models with similar accuracy across the training set (mean absolute error (MAE) = 2.95 years) and HC in the testing set (MAE = 2.94). We found no evidence of accelerated brain aging in individuals with GAD compared to individuals without GAD, though we did find evidence for greater variation in PAD for individuals with GAD (Levene's test: W = 442.98, p < 0.001) and evidence for greater variability in PAD of those with GAD over 25 years of age. In several exploratory analyses, we found that symptom severity related significantly to PAD, even after controlling for medication and comorbid diagnoses, echoing previous brain age research. These findings underscore the need for consideration of heterogeneity and dimensionality of psychopathology when examining brain age predicted differences.

Translational Psychiatry
Brigham Young University (US), Brighton and Sussex Medical School (GB), South African Medical Research Council (ZA), University of Southern California (US), Leiden University (NL), Universitat Autònoma de Barcelona (ES), University of Copenhagen (DK), Hartford Hospital (US), University of Minnesota (US), Universidade Federal do Rio Grande do Sul (BR), Boys Town (US), Pennsylvania State University (US), Harvard University (US), Johns Hopkins University (US), Vita-Salute San Raffaele University (IT), University of Cape Town (ZA), University of Illinois Urbana-Champaign (US), University of Sussex (GB), University of America (US), University of Pittsburgh (US), Baylor College of Medicine (US), King's College London (GB), Virginia Commonwealth University (US), University of Milan (IT), Vanderbilt University (US), University of Erfurt (DE), Washington University in St. Louis (US), University of Applied Sciences Leiden (NL), University of Münster (DE), Universitätsmedizin Greifswald (DE), Leiden University Medical Center (NL), Mental Health Services (DK), Instituto de Salud Carlos III (ES), German Center for Neurodegenerative Diseases (DE), Hospital de Sant Pau (ES), The University of Texas Rio Grande Valley (US), Copenhagen University Hospital (DK), Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico (IT), Yale University (US), University of California San Diego (US), Humboldt-Universität zu Berlin (DE), University of Belgrade (RS), Massachusetts General Hospital (US), Centro de Investigación Biomédica en Red de Salud Mental (ES), Hospital de Clínicas de Porto Alegre (BR), Boys Town National Research Hospital (US), Centro Ricerche Enrico Fermi (IT), Fondazione Santa Lucia (IT), Amsterdam University Medical Centers (NL), Istituti di Ricovero e Cura a Carattere Scientifico (IT), National Institute of Mental Health (US), Institute of Mental Health (RS), Laureate Institute for Brain Research (US), National Institute of Mental Health (CZ), Institut de Recerca Sant Pau (ES), Istituto di Ricovero e Cura a Carattere Scientifico San Raffaele (IT), HMU Health and Medical University Erfurt (DE), Hartford Financial Services (United States), The Ohio State University (US), New York University (US), Brunel University of London (GB), Stony Brook University (US), University of Cincinnati (US), Technische Universität Dresden (DE), University of Pennsylvania (US), Catholic University of America (US), Sapienza University of Rome (IT), The University of Texas at Austin (US), Universidade Federal de São Paulo (BR), Erasmus University Rotterdam (NL), The University of Texas Health Science Center at Houston (US), Stanford University (US)
National Institute of Mental Health
Openalex Percentile: Top 3%
Mental Health via Writing
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.