Using structural MRI to identify bipolar disorders - 13 site machine learning study in 3020 individuals from the ENIGMA Bipolar Disorders Working Group.

Journal: Molecular psychiatry

Volume: 25

Issue: 9

Year of Publication: 2021

Affiliated Institutions:  Department of Psychiatry, Dalhousie University, Halifax, Nova Scotia, Canada. Department of Psychiatry, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands. Interdepartmental Neuroscience Program, University of California, Los Angeles, CA, USA. NORMENT KG Jebsen Centre, University of Oslo, Oslo, Norway. Centre for Neuroimaging and Cognitive Genomics (NICOG), Clinical Neuroimaging Laboratory, NCBES Galway Neuroscience Centre, College of Medicine Nursing and Health Sciences, National University of Ireland Galway, Galway, Ireland. FIDMAG Germanes Hospitalàries Research Foundation, Barcelona, Spain. Institute of Clinical Radiology, Medical Faculty - University of Muenster - and University Hospital Muenster, Muenster, Germany. Department of Psychiatry, Melbourne Medical School, The University of Melbourne, Parkville, VIC, Australia. Department of Psychiatric Research, Diakonhjemmet Hospital, Oslo, Norway. Hospital Clinic, University of Barcelona, IDIBAPS, CIBERSAM, Barcelona, Catalonia, Spain. Laboratory of Psychiatric Neuroimaging (LIM-), Department and Institute of Psychiatry, Faculty of Medicine, University of São Paulo, São Paulo, Brazil. MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff, UK. Department of Psychiatry, University of Münster, Münster, Germany. Research Group in Psychiatry, Department of Psychiatry, Faculty of Medicine, Universidad de Antioquia, Medellín, Antioquia, Colombia. Department of Psychiatry and Psychotherapy, Philipps-University Marburg, Marburg, Germany. NeuroSpin, CEA, Paris-Saclay, Gif sur Yvette, France. Department of Psychiatry, University of California, San Diego, La Jolla, CA, USA. Cardiff University Brain Research Imaging Centre, Cardiff University, Cardiff, UK. Neuroscience Research Australia, Sydney, NSW, Australia. Department of Psychiatry, Yale University, New Haven, CT, USA. Institut Pasteur, Unité Perception et Mémoire, Paris, France. Imaging Genetics Center, Mark and Mary Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California, Marina del Rey, CA, USA. Neuroscience Institute, University of Cape Town, Cape Town, South Africa. School of Psychiatry, University of New South Wales, Sydney, NSW, Australia. Psychosomatic Unit, Division of Mental Health and Dependence, Oslo University Hospital and University of Oslo, Oslo, Norway. Department of Psychiatry, University of Texas Health Science Center at Houston, Houston, TX, USA. Research Group, Instituto de Alta Tecnología Médica (IATM), Medellín, Antioquia, Colombia. Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA. Department of Psychiatry, SA MRC Unit on Risk & Resilience in Mental Disorders, University of Cape Town, Cape Town, South Africa. Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia, Canada. Department of Psychiatry, Dalhousie University, Halifax, Nova Scotia, Canada. tomas.hajek@dal.ca.

Abstract summary 

Bipolar disorders (BDs) are among the leading causes of morbidity and disability. Objective biological markers, such as those based on brain imaging, could aid in clinical management of BD. Machine learning (ML) brings neuroimaging analyses to individual subject level and may potentially allow for their diagnostic use. However, fair and optimal application of ML requires large, multi-site datasets. We applied ML (support vector machines) to MRI data (regional cortical thickness, surface area, subcortical volumes) from 853 BD and 2167 control participants from 13 cohorts in the ENIGMA consortium. We attempted to differentiate BD from control participants, investigated different data handling strategies and studied the neuroimaging/clinical features most important for classification. Individual site accuracies ranged from 45.23% to 81.07%. Aggregate subject-level analyses yielded the highest accuracy (65.23%, 95% CI = 63.47-67.00, ROC-AUC = 71.49%, 95% CI = 69.39-73.59), followed by leave-one-site-out cross-validation (accuracy = 58.67%, 95% CI = 56.70-60.63). Meta-analysis of individual site accuracies did not provide above chance results. There was substantial agreement between the regions that contributed to identification of BD participants in the best performing site and in the aggregate dataset (Cohen's Kappa = 0.83, 95% CI = 0.829-0.831). Treatment with anticonvulsants and age were associated with greater odds of correct classification. Although short of the 80% clinically relevant accuracy threshold, the results are promising and provide a fair and realistic estimate of classification performance, which can be achieved in a large, ecologically valid, multi-site sample of BD participants based on regional neurostructural measures. Furthermore, the significant classification in different samples was based on plausible and similar neuroanatomical features. Future multi-site studies should move towards sharing of raw/voxelwise neuroimaging data.

Authors & Co-authors:  Nunes Abraham A Schnack Hugo G HG Ching Christopher R K CRK Agartz Ingrid I Akudjedu Theophilus N TN Alda Martin M Alnæs Dag D Alonso-Lana Silvia S Bauer Jochen J Baune Bernhard T BT Bøen Erlend E Bonnin Caterina Del Mar CDM Busatto Geraldo F GF Canales-Rodríguez Erick J EJ Cannon Dara M DM Caseras Xavier X Chaim-Avancini Tiffany M TM Dannlowski Udo U Díaz-Zuluaga Ana M AM Dietsche Bruno B Doan Nhat Trung NT Duchesnay Edouard E Elvsåshagen Torbjørn T Emden Daniel D Eyler Lisa T LT Fatjó-Vilas Mar M Favre Pauline P Foley Sonya F SF Fullerton Janice M JM Glahn David C DC Goikolea Jose M JM Grotegerd Dominik D Hahn Tim T Henry Chantal C Hibar Derrek P DP Houenou Josselin J Howells Fleur M FM Jahanshad Neda N Kaufmann Tobias T Kenney Joanne J Kircher Tilo T J TTJ Krug Axel A Lagerberg Trine V TV Lenroot Rhoshel K RK López-Jaramillo Carlos C Machado-Vieira Rodrigo R Malt Ulrik F UF McDonald Colm C Mitchell Philip B PB Mwangi Benson B Nabulsi Leila L Opel Nils N Overs Bronwyn J BJ Pineda-Zapata Julian A JA Pomarol-Clotet Edith E Redlich Ronny R Roberts Gloria G Rosa Pedro G PG Salvador Raymond R Satterthwaite Theodore D TD Soares Jair C JC Stein Dan J DJ Temmingh Henk S HS Trappenberg Thomas T Uhlmann Anne A van Haren Neeltje E M NEM Vieta Eduard E Westlye Lars T LT Wolf Daniel H DH Yüksel Dilara D Zanetti Marcus V MV Andreassen Ole A OA Thompson Paul M PM Hajek Tomas T

Study Outcome 

Source Link: Visit source

Statistics
Citations :  Gustavsson A, Svensson M, Jacobi F, Allgulander C, Alonso J, Beghi E, et al. Cost of disorders of the brain in Europe 2010. Eur Neuropsychopharmacol. 2011;21:718–79.
Authors :  75
Identifiers
Doi : 10.1038/s41380-018-0228-9
SSN : 1476-5578
Study Population
Male,Female
Mesh Terms
Bipolar Disorder
Other Terms
Study Design
Case Control Trial,Cross Sectional Study
Study Approach
Country of Study
Publication Country
England