Multimodal Machine Learning Workflows for Prediction of Psychosis in Patients With Clinical High-Risk Syndromes and Recent-Onset Depression.
Journal article

Multimodal Machine Learning Workflows for Prediction of Psychosis in Patients With Clinical High-Risk Syndromes and Recent-Onset Depression.

  • Koutsouleris N Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Dwyer DB Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Degenhardt F Institute of Human Genetics, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • Maj C Institute of Genomic Statistics and Bioinformatics, University of Bonn, Bonn, Germany.
  • Urquijo-Castro MF Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Sanfelici R Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Popovic D Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Oeztuerk O Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Haas SS Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York.
  • Weiske J Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Ruef A Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Kambeitz-Ilankovic L Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Antonucci LA Department of Education, Psychology, and Communication, University of Bari Aldo Moro, Bari, Italy.
  • Neufang S Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany.
  • Schmidt-Kraepelin C Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany.
  • Ruhrmann S Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Penzel N Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Kambeitz J Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Haidl TK Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Rosen M Department of Psychiatry and Psychotherapy, Faculty of Medicine and University Hospital, University of Cologne, Cologne, Germany.
  • Chisholm K Institute for Mental Health, University of Birmingham, Birmingham, United Kingdom.
  • Riecher-Rössler A Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland.
  • Egloff L Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland.
  • Schmidt A Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland.
  • Andreou C Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland.
  • Hietala J Department of Psychiatry, University of Turku, Turku, Finland.
  • Schirmer T GE Healthcare GmbH (previously GE Global Research GmbH), Munich, Germany.
  • Romer G Department of Child and Adolescent Psychiatry, University of Münster, Münster, Germany.
  • Walger P Department of Child and Adolescent Psychiatry, Psychotherapy and Psychosomatics, LVR Clinic Düsseldorf, Düsseldorf, Germany.
  • Franscini M Department of Child and Adolescent Psychiatry and Psychotherapy, University of Zürich, Zürich, Switzerland.
  • Traber-Walker N Department of Child and Adolescent Psychiatry and Psychotherapy, University of Zürich, Zürich, Switzerland.
  • Schimmelmann BG University Hospital of Child and Adolescent Psychiatry, University Hospital Hamburg-Eppendorf, Hamburg, Germany.
  • Flückiger R University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland.
  • Michel C University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland.
  • Rössler W Department of Psychiatry, Psychotherapy and Psychosomatics, University Hospital of Psychiatry Zurich, Zurich, Switzerland.
  • Borisov O Institute of Genomic Statistics and Bioinformatics, University of Bonn, Bonn, Germany.
  • Krawitz PM Institute of Genomic Statistics and Bioinformatics, University of Bonn, Bonn, Germany.
  • Heekeren K Department of Psychiatry, Psychotherapy and Psychosomatics, University Hospital of Psychiatry Zurich, Zurich, Switzerland.
  • Buechler R Department of Psychiatry, Psychotherapy and Psychosomatics, University Hospital of Psychiatry Zurich, Zurich, Switzerland.
  • Pantelis C Melbourne Neuropsychiatry Centre, University of Melbourne and Melbourne Health, Melbourne, Australia.
  • Falkai P Department of Psychiatry and Psychotherapy, Ludwig-Maximilian-University, Munich, Germany.
  • Salokangas RKR Department of Psychiatry, University of Turku, Turku, Finland.
  • Lencer R Department of Psychiatry and Psychotherapy, University of Münster, Münster, Germany.
  • Bertolino A Department of Basic Medical Science, Neuroscience and Sense Organs, University of Bari Aldo Moro, Bari, Italy.
  • Borgwardt S Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland.
  • Noethen M Institute of Human Genetics, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany.
  • Brambilla P Department of Neurosciences and Mental Health, Fondazione Istituto di Ricovero e Cura a Carattere Scientifico Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy.
  • Wood SJ Centre for Youth Mental Health, University of Melbourne, Melbourne, Australia.
  • Upthegrove R Institute for Mental Health, University of Birmingham, Birmingham, United Kingdom.
  • Schultze-Lutter F Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany.
  • Theodoridou A Department of Psychiatry, Psychotherapy and Psychosomatics, University Hospital of Psychiatry Zurich, Zurich, Switzerland.
  • Meisenzahl E Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany.
Show more…
  • 2020-12-02
Published in:
  • JAMA psychiatry. - 2020
English Importance
Diverse models have been developed to predict psychosis in patients with clinical high-risk (CHR) states. Whether prediction can be improved by efficiently combining clinical and biological models and by broadening the risk spectrum to young patients with depressive syndromes remains unclear.


Objectives
To evaluate whether psychosis transition can be predicted in patients with CHR or recent-onset depression (ROD) using multimodal machine learning that optimally integrates clinical and neurocognitive data, structural magnetic resonance imaging (sMRI), and polygenic risk scores (PRS) for schizophrenia; to assess models' geographic generalizability; to test and integrate clinicians' predictions; and to maximize clinical utility by building a sequential prognostic system.


Design, Setting, and Participants
This multisite, longitudinal prognostic study performed in 7 academic early recognition services in 5 European countries followed up patients with CHR syndromes or ROD and healthy volunteers. The referred sample of 167 patients with CHR syndromes and 167 with ROD was recruited from February 1, 2014, to May 31, 2017, of whom 26 (23 with CHR syndromes and 3 with ROD) developed psychosis. Patients with 18-month follow-up (n = 246) were used for model training and leave-one-site-out cross-validation. The remaining 88 patients with nontransition served as the validation of model specificity. Three hundred thirty-four healthy volunteers provided a normative sample for prognostic signature evaluation. Three independent Swiss projects contributed a further 45 cases with psychosis transition and 600 with nontransition for the external validation of clinical-neurocognitive, sMRI-based, and combined models. Data were analyzed from January 1, 2019, to March 31, 2020.


Main Outcomes and Measures
Accuracy and generalizability of prognostic systems.


Results
A total of 668 individuals (334 patients and 334 controls) were included in the analysis (mean [SD] age, 25.1 [5.8] years; 354 [53.0%] female and 314 [47.0%] male). Clinicians attained a balanced accuracy of 73.2% by effectively ruling out (specificity, 84.9%) but ineffectively ruling in (sensitivity, 61.5%) psychosis transition. In contrast, algorithms showed high sensitivity (76.0%-88.0%) but low specificity (53.5%-66.8%). A cybernetic risk calculator combining all algorithmic and human components predicted psychosis with a balanced accuracy of 85.5% (sensitivity, 84.6%; specificity, 86.4%). In comparison, an optimal prognostic workflow produced a balanced accuracy of 85.9% (sensitivity, 84.6%; specificity, 87.3%) at a much lower diagnostic burden by sequentially integrating clinical-neurocognitive, expert-based, PRS-based, and sMRI-based risk estimates as needed for the given patient. Findings were supported by good external validation results.


Conclusions and Relevance
These findings suggest that psychosis transition can be predicted in a broader risk spectrum by sequentially integrating algorithms' and clinicians' risk estimates. For clinical translation, the proposed workflow should undergo large-scale international validation.
Language
  • English
Open access status
hybrid
Identifiers
Persistent URL
https://sonar.ch/global/documents/132692
Statistics

Document views: 8 File downloads: