Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jan 17;9(1):12.
doi: 10.1038/s41398-018-0225-4.

Reproducible Grey Matter Patterns Index a Multivariate, Global Alteration of Brain Structure in Schizophrenia and Bipolar Disorder

Free PMC article

Reproducible Grey Matter Patterns Index a Multivariate, Global Alteration of Brain Structure in Schizophrenia and Bipolar Disorder

Emanuel Schwarz et al. Transl Psychiatry. .
Free PMC article


Schizophrenia is a severe mental disorder characterized by numerous subtle changes in brain structure and function. Machine learning allows exploring the utility of combining structural and functional brain magnetic resonance imaging (MRI) measures for diagnostic application, but this approach has been hampered by sample size limitations and lack of differential diagnostic data. Here, we performed a multi-site machine learning analysis to explore brain structural patterns of T1 MRI data in 2668 individuals with schizophrenia, bipolar disorder or attention-deficit/ hyperactivity disorder, and healthy controls. We found reproducible changes of structural parameters in schizophrenia that yielded a classification accuracy of up to 76% and provided discrimination from ADHD, through it lacked specificity against bipolar disorder. The observed changes largely indexed distributed grey matter alterations that could be represented through a combination of several global brain-structural parameters. This multi-site machine learning study identified a brain-structural signature that could reproducibly differentiate schizophrenia patients from controls, but lacked specificity against bipolar disorder. While this currently limits the clinical utility of the identified signature, the present study highlights that the underlying alterations index substantial global grey matter changes in psychotic disorders, reflecting the biological similarity of these conditions, and provide a roadmap for future exploration of brain structural alterations in psychiatric patients.

Conflict of interest statement

The IMAGEMEND Consortium

Francesco Bettella2, Christine L Brandt2, Toni-Kim Clarke26, David Coynel31,34, Franziska Degenhardt28,29, Srdjan Djurovic2,37, Sarah Eisenacher1, Matthias Fastenrath31,34, Helena Fatouros-Bergman11, Andreas J Forstner28,29,38,39,40, Josef Frank35, Francesco Gambi41, Barbara Gelao3, Leo Geschwind30,31, Massimo di Giannantonio41,42, Annabella Di Giorgio3,43, Catharina A Hartman44, Stefanie Heilmann-Heimbach28,29, Stefan Herms28,29,45, Pieter J Hoekstra46, Per Hoffmann28,29,45, Martine Hoogman5,18, Erik G Jönsson4,11, Eva Loos31,34, Eleonora Maggioni3,17, Jaap Oosterlaan47, Marco Papalino3, Antonio Rampino3, Liana Romaniuk26, Pierluigi Selvaggi3,48, Gianna Sepede3,41, Ida E Sønderby2, Klara Spalek31,34, Jessika E Sussmann26, Paul M Thompson49, Alejandro Arias Vasquez21, Christian Vogler30,31, Heather Whalley26 37Department of Medical Genetics, Oslo University Hospital, Oslo, Norway. 38Human Genomics Research Group, Department of Biomedicine, University of Basel, Basel, Switzerland. 39Department of Psychiatry (UPK), University of Basel, Basel, Switzerland. 40Institute of Medical Genetics and Pathology, University Hospital Basel, Basel, Switzerland. 41Department of Neuroscience, Imaging and Clinical Sciences “G. D’Annunzio” University Chieti-Pescara, Pescara, Italy. 42Department of Mental Health, National Health Trust, Chieti, Italy. 43Fondazione Casa Sollievo della Sofferenza IRCCS San Giovanni Rotondo (FG), San Giovanni Rotondo, Italy. 44Department of Psychiatry, Interdisciplinary Center Psychopathology and Emotion regulation, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 45Department of Biomedicine & Institute of Medical Genetics and Pathology, Human Genomics Research Group and Division of Medical Genetics, Department of Biomedicine, University and University Hospital Basel, Basel, Switzerland. 46Department of Child and Adolescent Psychiatry, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 47Emma Children’s Hospital, Academic Medical Center, Amsterdam, The Netherlands. 48Department of Neuroimaging, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK. 49Imaging Genetics Center, Stevens Institute for Neuroimaging & Informatics, University of Southern California, Los Angeles, CA, USA.

Karolinska Schizophrenia Project (KaSP) Consortium

Farde L11, Flyckt L11, Engberg G50, Erhardt S50, Fatouros-Bergman H11, Cervenka S11, Schwieler L50, Agartz I2,11,12, Collste K11, Victorsson P11, Malmqvist A50, Hedberg M50, Orhan F50 50Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden

Conflicts of interest

A.M.-L. has received consultant fees from Blueprint Partnership, Boehringer Ingelheim, Daimler und Benz Stiftung, Elsevier, F. Hoffmann-La Roche, ICARE Schizophrenia, K. G. Jebsen Foundation, L.E.K Consulting, Lundbeck International Foundation (LINF), R. Adamczak, Roche Pharma, Science Foundation, Synapsis Foundation – Alzheimer Research Switzerland, System Analytics, and has received lectures including travel fees from Boehringer Ingelheim, Fama Public Relations, Institut d’investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Janssen-Cilag, Klinikum Christophsbad, Göppingen, Lilly Deutschland, Luzerner Psychiatrie, LVR Klinikum Düsseldorf, LWL PsychiatrieVerbund Westfalen-Lippe, Otsuka Pharmaceuticals, Reunions i Ciencia S. L., Spanish Society of Psychiatry, Südwestrundfunk Fernsehen, Stern TV, and Vitos Klinikum Kurhessen. J.K.B. has been in the past 3 years a consultant to / member of advisory board of / and/or speaker for Roche, Medice and Servier. He is not an employee of any of these companies, and not a stock shareholder of any of these companies. He has no other financial or material support, including expert testimony, patents, royalties. A.B. is a stockholder of Roche and has received lecture fees from Otsuka. M.Z. has received unrestricted scientific grants from German Research Foundation (DFG), and Servier; further speaker and travel grants were provided by Otsuka, Servier, Lundbeck, Roche, Ferrer and Trommsdorff. S.C. has received grant support from AstraZeneca as a co-investigator, and has served as a one-off speaker for Otsuka-Lundbeck and Roche Pharmaceuticals. SCs spouse is an employee of SOBI pharmaceuticals. G.P. was an academic supervisor of a Hoffmann-La Roche collaboration grant (years 2015-16). B.F. has received educational speaking fees from Shire and Medice. All other authors declare no potential conflicts of interest.


Fig. 1
Fig. 1. Overview of analysis procedure. Subjects were first propensity score matched and VBM- / FreeSurfer-based features were then normalized against potential confounders.
Normalization models were built in training data only and these models were subsequently applied to adjust the test data. The same normalization strategy was applied for global structural parameters, which were subsequently used to remove the global structural signal from VBM- / FreeSurfer-based features. The resulting data was used for leave-site-out cross-validation analyses. For univariate analyses, as well as for machine learning analyses performed on the entire dataset, data were additionally corrected for a site factor, to account for the impact of site differences (see methods)
Fig. 2
Fig. 2. Accuracy of schizophrenia classifier using VBM- and FreeSurfer-based morphometry features.
a) Leave-site-out cross-validation performance measured as the ROC-AUC. b Specificity of schizophrenia-control classifier (trained on all SZ-HC cohorts) for prediction in independent cohorts. The red horizontal line demonstrates 50% ROC-AUC or specificity, respectively. The classification was based on random forest machine learning. SZ: schizophrenia; BD: bipolar disorder; ADHD: attention-deficit/ hyperactivity disorder; HC: healthy controls
Fig. 3
Fig. 3. VBM-based variable importance for classification.
a Random-forest variable importance for the schizophrenia vs. control (red, used to order the x-axis), the bipolar disorder vs control and the ADHD vs control comparisons. b Boxplot of random-forest variable importance measures, comparing the 14 most important schizophrenia predictors against the remaining predictors in bipolar disorder and ADHD. The asterisk indicates significance determined from permutation testing. Since variable importance was determined from the schizophrenia-control comparison, no significance estimate is shown for the corresponding boxplot
Fig. 4
Fig. 4. Effect of global structural covariates on classification.
a Comparison of associations between global structural features and the first principal components determined from the 14 selected VBM-based (orange; used to order the x-axis) and the 11 selected FreeSurfer-based (blue) features (see also Supplementary Table 1,0). b Effect of residualization against global structural features on classification performance and classification performance obtained from global features only. Notably, AUC values obtained from analyses with permuted diagnoses showed mean values > 0.5, which was due to chance associations in the comparatively small datasets. Furthermore, surface based features showed an increase in performance after residualization against permuted global features. This suggests features with poor cross-site reproducibility were coincidentally prioritized for classification in the original data and this was remedied in the residualized data. The two sets of global features were identical except for the addition of either a median VBM- or FreeSurfer-based feature

Similar articles

See all similar articles

Cited by 3 articles


    1. McGrath J, Saha S, Chant D, Welham J. Schizophrenia: a concise overview of incidence, prevalence, and mortality. Epidemiol. Rev. 2008;30:67–76. doi: 10.1093/epirev/mxn001. - DOI - PubMed
    1. Ross CA, Margolis RL, Reading SA, Pletnikov M, Coyle JT. Neurobiology of schizophrenia. Neuron. 2006;52:139–153. doi: 10.1016/j.neuron.2006.09.015. - DOI - PubMed
    1. Lewis DA, Lieberman JA. Catching up on schizophrenia: natural history and neurobiology. Neuron. 2000;28:325–334. doi: 10.1016/S0896-6273(00)00111-2. - DOI - PubMed
    1. Shepherd AM, Laurens KR, Matheson SL, Carr VJ, Green MJ. Systematic meta-review and quality assessment of the structural brain alterations in schizophrenia. Neurosci. Biobehav. Rev. 2012;36:1342–1356. doi: 10.1016/j.neubiorev.2011.12.015. - DOI - PubMed
    1. Okada N, et al. Abnormal asymmetries in subcortical brain volume in schizophrenia. Mol. Psychiatry. 2016;21:1460–1466. doi: 10.1038/mp.2015.209. - DOI - PMC - PubMed

Publication types