Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach

Paris Alexandros Lalousis; Stephen J Wood; Lianne Schmaal; Katharine Chisholm; Sian Lowri Griffiths; Renate L E P Reniers; Alessandro Bertolino; Stefan Borgwardt; Paolo Brambilla; Joseph Kambeitz; Rebekka Lencer; Christos Pantelis; Stephan Ruhrmann; Raimo K R Salokangas; Frauke Schultze-Lutter; Carolina Bonivento; Dominic Dwyer; Adele Ferro; Theresa Haidl; Marlene Rosen; Andre Schmidt; Eva Meisenzahl; Nikolaos Koutsouleris; Rachel Upthegrove

doi:10.1093/schbul/sbaa185

Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach

Paris Alexandros Lalousis
, Stephen J Wood
, Lianne Schmaal
, Katharine Chisholm
, Sian Lowri Griffiths
, Renate L E P Reniers
, Alessandro Bertolino
, Stefan Borgwardt
, Paolo Brambilla
, Joseph Kambeitz
, Rebekka Lencer
, Christos Pantelis
, Stephan Ruhrmann
, Raimo K R Salokangas
, Frauke Schultze-Lutter
, Carolina Bonivento
, Dominic Dwyer
, Adele Ferro
, Theresa Haidl
, Marlene Rosen

Andre Schmidt, Eva Meisenzahl, Nikolaos Koutsouleris, Rachel Upthegrove

Institute for Mental Health and Centre for Human Brain Health, University of Birmingham, Birmingham, UK
Institute for Mental Health, University of Birmingham , Birmingham , United Kingdom; Orygen, The National Centre of Excellence for Youth Mental Health , Melbourne, VIC , Australia; Centre for Youth Mental Health, University of Melbourne , Parkville, VIC , Australia
Institute for Mental Health, University of Birmingham, Birmingham, UK,Centre for Human Brain Health, University of Birmingham, Birmingham, UK,Institute of Clinical Sciences, University of Birmingham, Birmingham, UK
Department of Basic Medical Sciences, Neuroscience and Sense Organs, University of Bari Aldo Moro, Bari, Italy
Department of Psychiatry, University of Basel, Basel, Switzerland,Department of Psychiatry and Psychotherapy, Center of Brain, Behavior and Metabolism, University of Lübeck, Germany
Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy,Department of Neurosciences and Mental Health, IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy
Department of Psychiatry and Psychotherapy, Ludwig Maxmilians University, Munich, Germany
grid.4562.5 0000 0001 0057 2672 Department of Psychiatry and Psychotherapy University of Lübeck Lübeck Germany; grid.5949.1 0000 0001 2172 9288 Department of Psychiatry University of Münster Münster Germany; grid.5949.1 0000 0001 2172 9288 Otto Creutzfeldt Center for Behavioral and Cognitive Neuroscience University of Münster Münster Germany
Melbourne Neuropsychiatry Centre, University of Melbourne and Melbourne Health, Melbourne, Australia
Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany
Department of Psychiatry, University of Turku, Turku, Finland
Department of Psychiatry and Psychotherapy, Heinrich-Heine University Düsseldorf, Düsseldorf, Germany,Department of Psychology and Mental Health, Faculty of Psychology, Airlangga University, Surabaya, Indonesia,University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland
IRCCS “E. Medea” Scientific Institute, San Vito al Tagliamento (Pn), Italy
Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, University of Milan, Milan, Italy
Department of Psychiatry, University Psychiatric Clinic, Psychiatric University Hospital, University of Basel, Basel, Switzerland
Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany
Institute for Mental Health, University of Birmingham, Birmingham, UK,Centre for Human Brain Health, University of Birmingham, Birmingham, UK,Early Intervention Service, Birmingham Women’s and Children’s NHS Foundation Trust, Birmingham, UK

Research output: Contribution to journal › Article › peer-review

43 Citations (SciVal)

Abstract

Abstract Diagnostic heterogeneity within and across psychotic and affective disorders challenges accurate treatment selection, particularly in the early stages. Delineation of shared and distinct illness features at the phenotypic and brain levels may inform the development of more precise differential diagnostic tools. We aimed to identify prototypes of depression and psychosis to investigate their heterogeneity, with common, comorbid transdiagnostic symptoms. Analyzing clinical/neurocognitive and grey matter volume (GMV) data from the PRONIA database, we generated prototypic models of recent-onset depression (ROD) vs. recent-onset psychosis (ROP) by training support-vector machines to separate patients with ROD from patients with ROP, who were selected for absent comorbid features (pure groups). Then, models were applied to patients with comorbidity, ie, ROP with depressive symptoms (ROP+D) and ROD participants with sub-threshold psychosis-like features (ROD+P), to measure their positions within the affective-psychotic continuum. All models were independently validated in a replication sample. Comorbid patients were positioned between pure groups, with ROP+D patients being more frequently classified as ROD compared to pure ROP patients (clinical/neurocognitive model: χ2 = 14.874; P < .001; GMV model: χ2 = 4.933; P = .026). ROD+P patient classification did not differ from ROD (clinical/neurocognitive model: χ2 = 1.956; P = 0.162; GMV model: χ2 = 0.005; P = .943). Clinical/neurocognitive and neuroanatomical models demonstrated separability of prototypic depression from psychosis. The shift of comorbid patients toward the depression prototype, observed at the clinical and biological levels, suggests that psychosis with affective comorbidity aligns more strongly to depressive rather than psychotic disease processes. Future studies should assess how these quantitative measures of comorbidity predict outcomes and individual responses to stratified therapeutic interventions.

Original language	English
Pages (from-to)	1130-1140
Number of pages	11
Journal	Schizophrenia Bulletin
Volume	47
Issue number	4
Early online date	5 Feb 2021
DOIs	https://doi.org/10.1093/schbul/sbaa185
Publication status	Published - 1 Jul 2021

Keywords

Psychosis
depression
transdiagnostic
machine learning
MRI
gray matter volume
comorbidity

Access to Document

10.1093/schbul/sbaa185

Cite this

Lalousis, P. A., Wood, S. J., Schmaal, L., Chisholm, K., Griffiths, S. L., Reniers, R. L. E. P., Bertolino, A., Borgwardt, S., Brambilla, P., Kambeitz, J., Lencer, R., Pantelis, C., Ruhrmann, S., Salokangas, R. K. R., Schultze-Lutter, F., Bonivento, C., Dwyer, D., Ferro, A., Haidl, T., ... Upthegrove, R. (2021). Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach. Schizophrenia Bulletin, 47(4), 1130-1140. https://doi.org/10.1093/schbul/sbaa185

@article{c2fbf7169b3b423189aa276ac9257e9e,

title = "Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach",

abstract = "Abstract Diagnostic heterogeneity within and across psychotic and affective disorders challenges accurate treatment selection, particularly in the early stages. Delineation of shared and distinct illness features at the phenotypic and brain levels may inform the development of more precise differential diagnostic tools. We aimed to identify prototypes of depression and psychosis to investigate their heterogeneity, with common, comorbid transdiagnostic symptoms. Analyzing clinical/neurocognitive and grey matter volume (GMV) data from the PRONIA database, we generated prototypic models of recent-onset depression (ROD) vs. recent-onset psychosis (ROP) by training support-vector machines to separate patients with ROD from patients with ROP, who were selected for absent comorbid features (pure groups). Then, models were applied to patients with comorbidity, ie, ROP with depressive symptoms (ROP+D) and ROD participants with sub-threshold psychosis-like features (ROD+P), to measure their positions within the affective-psychotic continuum. All models were independently validated in a replication sample. Comorbid patients were positioned between pure groups, with ROP+D patients being more frequently classified as ROD compared to pure ROP patients (clinical/neurocognitive model: χ2 = 14.874; P < .001; GMV model: χ2 = 4.933; P = .026). ROD+P patient classification did not differ from ROD (clinical/neurocognitive model: χ2 = 1.956; P = 0.162; GMV model: χ2 = 0.005; P = .943). Clinical/neurocognitive and neuroanatomical models demonstrated separability of prototypic depression from psychosis. The shift of comorbid patients toward the depression prototype, observed at the clinical and biological levels, suggests that psychosis with affective comorbidity aligns more strongly to depressive rather than psychotic disease processes. Future studies should assess how these quantitative measures of comorbidity predict outcomes and individual responses to stratified therapeutic interventions.",

keywords = "Psychosis, depression, transdiagnostic, machine learning, MRI, gray matter volume, comorbidity",

author = "Lalousis, \{Paris Alexandros\} and Wood, \{Stephen J\} and Lianne Schmaal and Katharine Chisholm and Griffiths, \{Sian Lowri\} and Reniers, \{Renate L E P\} and Alessandro Bertolino and Stefan Borgwardt and Paolo Brambilla and Joseph Kambeitz and Rebekka Lencer and Christos Pantelis and Stephan Ruhrmann and Salokangas, \{Raimo K R\} and Frauke Schultze-Lutter and Carolina Bonivento and Dominic Dwyer and Adele Ferro and Theresa Haidl and Marlene Rosen and Andre Schmidt and Eva Meisenzahl and Nikolaos Koutsouleris and Rachel Upthegrove",

year = "2021",

month = jul,

day = "1",

doi = "10.1093/schbul/sbaa185",

language = "English",

volume = "47",

pages = "1130--1140",

journal = "Schizophrenia Bulletin",

issn = "0586-7614",

publisher = "Oxford University Press",

number = "4",

}

Lalousis, PA, Wood, SJ, Schmaal, L, Chisholm, K, Griffiths, SL, Reniers, RLEP, Bertolino, A, Borgwardt, S, Brambilla, P, Kambeitz, J, Lencer, R, Pantelis, C, Ruhrmann, S, Salokangas, RKR, Schultze-Lutter, F, Bonivento, C, Dwyer, D, Ferro, A, Haidl, T, Rosen, M, Schmidt, A, Meisenzahl, E, Koutsouleris, N & Upthegrove, R 2021, 'Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach', Schizophrenia Bulletin, vol. 47, no. 4, pp. 1130-1140. https://doi.org/10.1093/schbul/sbaa185

TY - JOUR

T1 - Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach

AU - Lalousis, Paris Alexandros

AU - Wood, Stephen J

AU - Schmaal, Lianne

AU - Chisholm, Katharine

AU - Griffiths, Sian Lowri

AU - Reniers, Renate L E P

AU - Bertolino, Alessandro

AU - Borgwardt, Stefan

AU - Brambilla, Paolo

AU - Kambeitz, Joseph

AU - Lencer, Rebekka

AU - Pantelis, Christos

AU - Ruhrmann, Stephan

AU - Salokangas, Raimo K R

AU - Schultze-Lutter, Frauke

AU - Bonivento, Carolina

AU - Dwyer, Dominic

AU - Ferro, Adele

AU - Haidl, Theresa

AU - Rosen, Marlene

AU - Schmidt, Andre

AU - Meisenzahl, Eva

AU - Koutsouleris, Nikolaos

AU - Upthegrove, Rachel

PY - 2021/7/1

Y1 - 2021/7/1

N2 - Abstract Diagnostic heterogeneity within and across psychotic and affective disorders challenges accurate treatment selection, particularly in the early stages. Delineation of shared and distinct illness features at the phenotypic and brain levels may inform the development of more precise differential diagnostic tools. We aimed to identify prototypes of depression and psychosis to investigate their heterogeneity, with common, comorbid transdiagnostic symptoms. Analyzing clinical/neurocognitive and grey matter volume (GMV) data from the PRONIA database, we generated prototypic models of recent-onset depression (ROD) vs. recent-onset psychosis (ROP) by training support-vector machines to separate patients with ROD from patients with ROP, who were selected for absent comorbid features (pure groups). Then, models were applied to patients with comorbidity, ie, ROP with depressive symptoms (ROP+D) and ROD participants with sub-threshold psychosis-like features (ROD+P), to measure their positions within the affective-psychotic continuum. All models were independently validated in a replication sample. Comorbid patients were positioned between pure groups, with ROP+D patients being more frequently classified as ROD compared to pure ROP patients (clinical/neurocognitive model: χ2 = 14.874; P < .001; GMV model: χ2 = 4.933; P = .026). ROD+P patient classification did not differ from ROD (clinical/neurocognitive model: χ2 = 1.956; P = 0.162; GMV model: χ2 = 0.005; P = .943). Clinical/neurocognitive and neuroanatomical models demonstrated separability of prototypic depression from psychosis. The shift of comorbid patients toward the depression prototype, observed at the clinical and biological levels, suggests that psychosis with affective comorbidity aligns more strongly to depressive rather than psychotic disease processes. Future studies should assess how these quantitative measures of comorbidity predict outcomes and individual responses to stratified therapeutic interventions.

AB - Abstract Diagnostic heterogeneity within and across psychotic and affective disorders challenges accurate treatment selection, particularly in the early stages. Delineation of shared and distinct illness features at the phenotypic and brain levels may inform the development of more precise differential diagnostic tools. We aimed to identify prototypes of depression and psychosis to investigate their heterogeneity, with common, comorbid transdiagnostic symptoms. Analyzing clinical/neurocognitive and grey matter volume (GMV) data from the PRONIA database, we generated prototypic models of recent-onset depression (ROD) vs. recent-onset psychosis (ROP) by training support-vector machines to separate patients with ROD from patients with ROP, who were selected for absent comorbid features (pure groups). Then, models were applied to patients with comorbidity, ie, ROP with depressive symptoms (ROP+D) and ROD participants with sub-threshold psychosis-like features (ROD+P), to measure their positions within the affective-psychotic continuum. All models were independently validated in a replication sample. Comorbid patients were positioned between pure groups, with ROP+D patients being more frequently classified as ROD compared to pure ROP patients (clinical/neurocognitive model: χ2 = 14.874; P < .001; GMV model: χ2 = 4.933; P = .026). ROD+P patient classification did not differ from ROD (clinical/neurocognitive model: χ2 = 1.956; P = 0.162; GMV model: χ2 = 0.005; P = .943). Clinical/neurocognitive and neuroanatomical models demonstrated separability of prototypic depression from psychosis. The shift of comorbid patients toward the depression prototype, observed at the clinical and biological levels, suggests that psychosis with affective comorbidity aligns more strongly to depressive rather than psychotic disease processes. Future studies should assess how these quantitative measures of comorbidity predict outcomes and individual responses to stratified therapeutic interventions.

KW - Psychosis

KW - depression

KW - transdiagnostic

KW - machine learning

KW - MRI

KW - gray matter volume

KW - comorbidity

UR - https://academic.oup.com/schizophreniabulletin/advance-article/doi/10.1093/schbul/sbaa185/6129055

UR - https://www.scopus.com/pages/publications/85111078924

U2 - 10.1093/schbul/sbaa185

DO - 10.1093/schbul/sbaa185

M3 - Article

C2 - 33543752

SN - 0586-7614

VL - 47

SP - 1130

EP - 1140

JO - Schizophrenia Bulletin

JF - Schizophrenia Bulletin

IS - 4

ER -

Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this