Exploring the predictive value of structural covariance networks for the diagnosis of schizophrenia

Clara S Vetter; Annika Bender; Dominic B Dwyer; Max Montembeault; Anne Ruef; Katharine Chisholm; Lana Kambeitz-Ilankovic; Linda A Antonucci; Stephan Ruhrmann; Joseph Kambeitz; Marlene Rosen; Theresa Lichtenstein; Anita Riecher-Rössler; Rachel Upthegrove; Raimo K R Salokangas; Jarmo Hietala; Christos Pantelis; Rebekka Lencer; Eva Meisenzahl; Stephen J Wood; Paolo Brambilla; Stefan Borgwardt; Peter Falkai; Alessandro Bertolino; Nikolaos Koutsouleris; PRONIA Consortium

doi:10.3389/fpsyt.2025.1570797

Exploring the predictive value of structural covariance networks for the diagnosis of schizophrenia

Front Psychiatry. 2025 Jun 9:16:1570797. doi: 10.3389/fpsyt.2025.1570797. eCollection 2025.

Authors

Clara S Vetter^{1

2

3}, Annika Bender¹, Dominic B Dwyer^{1

4

5}, Max Montembeault⁶, Anne Ruef¹, Katharine Chisholm⁷, Lana Kambeitz-Ilankovic⁸, Linda A Antonucci⁹, Stephan Ruhrmann⁸, Joseph Kambeitz⁸, Marlene Rosen⁸, Theresa Lichtenstein⁸, Anita Riecher-Rössler¹⁰, Rachel Upthegrove¹¹, Raimo K R Salokangas¹², Jarmo Hietala¹², Christos Pantelis^{13

14}, Rebekka Lencer^{15

16}, Eva Meisenzahl¹⁷, Stephen J Wood^{4

18}, Paolo Brambilla^{19

20}, Stefan Borgwardt¹⁶, Peter Falkai^{1

21}, Alessandro Bertolino⁹, Nikolaos Koutsouleris^{1

21

22}; PRONIA Consortium

Affiliations

¹ Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
² Munich Center of Machine Learning (MCML), Munich, Germany.
³ Helmholtz Association - Munich School for Data Science (MUDS), Munich, Germany.
⁴ Centre for Youth Mental Health, University of Melbourne, Melbourne, VIC, Australia.
⁵ The National Centre of Excellence for Youth Mental HealthOrygen, Melbourne, VIC, Australia.
⁶ Douglas Research Centre and Department of Psychiatry, McGill University, Montréal, QC, Canada.
⁷ School of Psychology, University of Sussex, Brighton, United Kingdom.
⁸ Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital of Cologne, University of Cologne, Germany, Cologne, Germany.
⁹ Department of Translational Biomedicine and Neuroscience (DiBraiN), University of Bari "Aldo Moro", Bari, Italy.
¹⁰ Faculty of Medicine, University of Basel, Basel, Switzerland.
¹¹ Department of Psychiatry, University of Oxford, Oxford, United Kingdom.
¹² Department of Psychiatry, University of Turku, Turku, Finland.
¹³ Melbourne Neuropsychiatry Centre, Department of Psychiatry, University of Melbourne, Melbourne, VIC, Australia.
¹⁴ NorthWestern Mental Health, Royal Melbourne Hospital, Melbourne, VIC, Australia.
¹⁵ Institute for Translational Psychiatry, University of Muenster, Muenster, Germany.
¹⁶ Department of Psychiatry and Psychotherapy, University of Lübeck, Lübeck, Germany.
¹⁷ Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany.
¹⁸ School of Psychology, University of Birmingham, Birmingham, United Kingdom.
¹⁹ Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Milan, Italy.
²⁰ Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy.
²¹ Department of Pathophysiology and Transplantation, Max Planck Institute for Psychiatry, Munich, Germany.
²² Department of Psychosis Studies, Institute of Psychiatry, Psychology Neuroscience, King's College London, London, United Kingdom.

Abstract

Introduction: Schizophrenia is a psychiatric disorder hypothesized to result from disturbed brain connectivity. Structural covariance networks (SCN) describe the shared variation in morphological properties emerging from coordinated neurodevelopmental processes, This study evaluates the potential of SCNs as diagnostic biomarker for schizophrenia.

Methods: We compared the diagnostic value of two SCN computation methods derived from regional gray matter volume (GMV) in 154 patients with a diagnosis of first episode psychosis or recurrent schizophrenia (PAT) and 366 healthy control individuals (HC). The first method (REF-SCN) quantifies the contribution of an individual to a normative reference group's SCN, and the second approach (KLS-SCN) uses a symmetric version of Kulback-Leibler divergence. Their diagnostic value compared to regional GMV was assessed in a stepwise analysis using a series of linear support vector machines within a nested cross-validation framework and stacked generalization, all models were externally validated in an independent sample (N_PAT=71, N_HC=74), SCN feature importance was assessed, and the derived risk scores were analyzed for differential relationships with clinical variables.

Results: We found that models trained on SCNs were able to classify patients with schizophrenia and combining SCNs and regional GMV in a stacked model improved training (balanced accuracy (BAC)=69.96%) and external validation performance (BAC=67.10%). Among all unimodal models, the highest discovery sample performance was achieved by a model trained on REF-SCN (balanced accuracy (BAC=67.03%). All model decisions were driven by widespread structural covariance alterations involving the somato-motor, default mode, control, visual, and the ventral attention networks. Risk estimates derived from KLS-SCNs and regional GMV, but not REF-SCNs, could be predicted from clinical variables, especially driven by body mass index (BMI) and affect-related negative symptoms.

Discussion: These patterns of results show that different SCN computation approaches capture different aspects of the disease. While REF-SCNs contain valuable information for discriminating schizophrenia from healthy control individuals, KLS-SCNs may capture more nuanced symptom-level characteristics similar to those captured by PCA of regional GMV.

Keywords: brain connectivity; machine learning; neuroimaging; precision psychiatry; schizophrenia; structural covariance.

Copyright © 2025 Vetter, Bender, Dwyer, Montembeault, Ruef, Chisholm, Kambeitz-Ilankovic, Antonucci, Ruhrmann, Kambeitz, Rosen, Lichtenstein, Riecher-Rössler, Upthegrove, Salokangas, Hietala, Pantelis, Lencer, Meisenzahl, Wood, Brambilla, Borgwardt, Falkai, Bertolino, Koutsouleris and PRONIA Consortium.