Ji Chen1, Kaustubh R Patil2, Susanne Weis1, Kang Sim3, Thomas Nickl-Jockschat4, Juan Zhou5, André Aleman6, Iris E Sommer7, Edith J Liemburg8, Felix Hoffstaedter1, Ute Habel9, Birgit Derntl10, Xiaojin Liu1, Jona M Fischer1, Lydia Kogler10, Christina Regenbogen9, Vaibhav A Diwadkar11, Jeffrey A Stanley11, Valentin Riedl12, Renaud Jardri13, Oliver Gruber14, Aristeidis Sotiras15, Christos Davatzikos16, Simon B Eickhoff1. 1. Institute of Neuroscience and Medicine, Brain and Behaviour (INM-7), Research Center Jülich, Jülich, Germany; Institute of Systems Neuroscience, Medical Faculty, Heinrich Heine University Düsseldorf, Düsseldorf, Germany. 2. Institute of Neuroscience and Medicine, Brain and Behaviour (INM-7), Research Center Jülich, Jülich, Germany; Institute of Systems Neuroscience, Medical Faculty, Heinrich Heine University Düsseldorf, Düsseldorf, Germany. Electronic address: k.patil@fz-juelich.de. 3. Department of General Psychiatry, Institute of Mental Health, Singapore; Research Division, Institute of Mental Health, Singapore. 4. Iowa Neuroscience Institute, Carver College of Medicine, University of Iowa, Iowa City, Iowa; Department of Psychiatry, Carver College of Medicine, University of Iowa, Iowa City, Iowa. 5. Center for Cognitive Neuroscience, Neuroscience and Behavioral Disorders Program, Duke-National University of Singapore Medical School, Singapore. 6. Department of Neuroscience, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 7. Department of Neuroscience, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands; BCN Neuroimaging Center, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 8. Rob Giel Research Center, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands. 9. Department of Psychiatry, Psychotherapy and Psychosomatics, Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen University, Aachen, Germany; Jülich Aachen Research Alliance-Institute Brain Structure Function Relationship, Research Center Jülich, and RWTH Aachen University, Aachen, Germany. 10. Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen, Germany. 11. Department of Psychiatry and Behavioral Neuroscience, Wayne State University, Detroit, Michigan. 12. Department of Neuroradiology, Rechts der Isar Hospital, Technical University of Munich, Munich, Germany. 13. University of Lille, National Centre for Scientific Research, UMR 9193, SCALab and CHU Lille, Fontan Hospital, CURE platform, Lille, France. 14. Section for Experimental Psychopathology and Neuroimaging, Department of General Psychiatry, Heidelberg University, Heidelberg, Germany. 15. Department of Radiology and Institute for Informatics, School of Medicine, Washington University in St. Louis, St. Louis, Missouri. 16. Center for Biomedical Image Computing and Analytics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; Department of Radiology, Section of Biomedical Image Analysis, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania.
Abstract
BACKGROUND: Disentangling psychopathological heterogeneity in schizophrenia is challenging, and previous results remain inconclusive. We employed advanced machine learning to identify a stable and generalizable factorization of the Positive and Negative Syndrome Scale and used it to identify psychopathological subtypes as well as their neurobiological differentiations. METHODS: Positive and Negative Syndrome Scale data from the Pharmacotherapy Monitoring and Outcome Survey cohort (1545 patients; 586 followed up after 1.35 ± 0.70 years) were used for learning the factor structure by an orthonormal projective non-negative factorization. An international sample, pooled from 9 medical centers across Europe, the United States, and Asia (490 patients), was used for validation. Patients were clustered into psychopathological subtypes based on the identified factor structure, and the neurobiological divergence between the subtypes was assessed by classification analysis on functional magnetic resonance imaging connectivity patterns. RESULTS: A 4-factor structure representing negative, positive, affective, and cognitive symptoms was identified as the most stable and generalizable representation of psychopathology. It showed higher internal consistency than the original Positive and Negative Syndrome Scale subscales and previously proposed factor models. Based on this representation, the positive-negative dichotomy was confirmed as the (only) robust psychopathological subtypes, and these subtypes were longitudinally stable in about 80% of the repeatedly assessed patients. Finally, the individual subtype could be predicted with good accuracy from functional connectivity profiles of the ventromedial frontal cortex, temporoparietal junction, and precuneus. CONCLUSIONS: Machine learning applied to multisite data with cross-validation yielded a factorization generalizable across populations and medical systems. Together with subtyping and the demonstrated ability to predict subtype membership from neuroimaging data, this work further disentangles the heterogeneity in schizophrenia.
BACKGROUND: Disentangling psychopathological heterogeneity in schizophrenia is challenging, and previous results remain inconclusive. We employed advanced machine learning to identify a stable and generalizable factorization of the Positive and Negative Syndrome Scale and used it to identify psychopathological subtypes as well as their neurobiological differentiations. METHODS: Positive and Negative Syndrome Scale data from the Pharmacotherapy Monitoring and Outcome Survey cohort (1545 patients; 586 followed up after 1.35 ± 0.70 years) were used for learning the factor structure by an orthonormal projective non-negative factorization. An international sample, pooled from 9 medical centers across Europe, the United States, and Asia (490 patients), was used for validation. Patients were clustered into psychopathological subtypes based on the identified factor structure, and the neurobiological divergence between the subtypes was assessed by classification analysis on functional magnetic resonance imaging connectivity patterns. RESULTS: A 4-factor structure representing negative, positive, affective, and cognitive symptoms was identified as the most stable and generalizable representation of psychopathology. It showed higher internal consistency than the original Positive and Negative Syndrome Scale subscales and previously proposed factor models. Based on this representation, the positive-negative dichotomy was confirmed as the (only) robust psychopathological subtypes, and these subtypes were longitudinally stable in about 80% of the repeatedly assessed patients. Finally, the individual subtype could be predicted with good accuracy from functional connectivity profiles of the ventromedial frontal cortex, temporoparietal junction, and precuneus. CONCLUSIONS: Machine learning applied to multisite data with cross-validation yielded a factorization generalizable across populations and medical systems. Together with subtyping and the demonstrated ability to predict subtype membership from neuroimaging data, this work further disentangles the heterogeneity in schizophrenia.
Authors: Mark van der Gaag; Anke Cuijpers; Tonko Hoffman; Mila Remijsen; Ron Hijman; Lieuwe de Haan; Berno van Meijel; Peter N van Harten; Lucia Valmaggia; Marc de Hert; Durk Wiersma Journal: Schizophr Res Date: 2006-05-26 Impact factor: 4.939
Authors: Martin Rozycki; Theodore D Satterthwaite; Nikolaos Koutsouleris; Guray Erus; Jimit Doshi; Daniel H Wolf; Yong Fan; Raquel E Gur; Ruben C Gur; Eva M Meisenzahl; Chuanjun Zhuo; Hong Yin; Hao Yan; Weihua Yue; Dai Zhang; Christos Davatzikos Journal: Schizophr Bull Date: 2018-08-20 Impact factor: 9.306
Authors: Aristeidis Sotiras; Jon B Toledo; Raquel E Gur; Ruben C Gur; Theodore D Satterthwaite; Christos Davatzikos Journal: Proc Natl Acad Sci U S A Date: 2017-03-13 Impact factor: 11.205
Authors: David M Gardner; Andrea L Murphy; Heather O'Donnell; Franca Centorrino; Ross J Baldessarini Journal: Am J Psychiatry Date: 2010-04-01 Impact factor: 18.112
Authors: Dwight Dickinson; Danielle N Pratt; Evan J Giangrande; MeiLin Grunnagle; Jennifer Orel; Daniel R Weinberger; Joseph H Callicott; Karen F Berman Journal: Schizophr Bull Date: 2018-01-13 Impact factor: 9.306
Authors: Ji Chen; Veronika I Müller; Juergen Dukart; Felix Hoffstaedter; Justin T Baker; Avram J Holmes; Deniz Vatansever; Thomas Nickl-Jockschat; Xiaojin Liu; Birgit Derntl; Lydia Kogler; Renaud Jardri; Oliver Gruber; André Aleman; Iris E Sommer; Simon B Eickhoff; Kaustubh R Patil Journal: Biol Psychiatry Date: 2020-10-03 Impact factor: 13.382
Authors: Xiaojin Liu; Simon B Eickhoff; Felix Hoffstaedter; Sarah Genon; Svenja Caspers; Kathrin Reetz; Imis Dogan; Claudia R Eickhoff; Ji Chen; Julian Caspers; Niels Reuter; Christian Mathys; André Aleman; Renaud Jardri; Valentin Riedl; Iris E Sommer; Kaustubh R Patil Journal: Neurosci Bull Date: 2020-07-23 Impact factor: 5.203
Authors: Jianxiao Wu; Simon B Eickhoff; Felix Hoffstaedter; Kaustubh R Patil; Holger Schwender; B T Thomas Yeo; Sarah Genon Journal: Cereb Cortex Date: 2021-07-05 Impact factor: 5.357
Authors: Hua Ye; Andrew Zalesky; Jinglei Lv; Samantha M Loi; Suheyla Cetin-Karayumak; Yogesh Rathi; Ye Tian; Christos Pantelis; Maria A Di Biase Journal: Schizophr Bull Date: 2021-07-08 Impact factor: 9.306
Authors: Ji Chen; Tobias Wensing; Felix Hoffstaedter; Edna C Cieslik; Veronika I Müller; Kaustubh R Patil; André Aleman; Birgit Derntl; Oliver Gruber; Renaud Jardri; Lydia Kogler; Iris E Sommer; Simon B Eickhoff; Thomas Nickl-Jockschat Journal: Neuroimage Clin Date: 2021-04-30 Impact factor: 4.881
Authors: Jie Lisa Ji; Markus Helmer; Clara Fonteneau; Joshua B Burt; Zailyn Tamayo; Jure Demšar; Brendan D Adkinson; Aleksandar Savić; Katrin H Preller; Flora Moujaes; Franz X Vollenweider; William J Martin; Grega Repovš; Youngsun T Cho; Christopher Pittenger; John D Murray; Alan Anticevic Journal: Elife Date: 2021-07-20 Impact factor: 8.713