Predicting alcohol dependence from multi-site brain structural measures.
Journal: Human brain mapping
Volume: 43
Issue: 1
Year of Publication: 2022
Affiliated Institutions:
Department of Psychiatry, University of Vermont College of Medicine, Burlington, Vermont, USA.
Department of Psychology, University of Amsterdam, Amsterdam, the Netherlands.
Department of Neuroscience & The Ernest J. Del Monte Institute for Neuroscience, University of Rochester School of Medicine and Dentistry, Rochester, New York, USA.
Department of Psychiatry and Psychotherapy, Charité-Universitätsmedizin Berlin, Berlin, Germany.
Melbourne School of Psychological Sciences, University of Melbourne, Melbourne, Australia.
Department of Psychology and Neuroscience, University of Colorado, Boulder, Colorado, USA.
Department of Addictive Behaviour and Addiction Medicine, Central Institute of Mental Health, Heidelberg University, Mannheim, Germany.
Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri, USA.
Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut, USA.
David Geffen School of Medicine, University of California at Los Angeles, Los Angeles, California, USA.
Monash Institute of Cognitive and Clinical Neurosciences & School of Psychological Sciences, Monash University, Melbourne, Australia.
Behavioural Science Institute, Radboud University, Nijmegen, the Netherlands.
Clinical NeuroImaging Research Core, Division of Intramural Clinical and Biological Research, National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland, USA.
VA San Diego Healthcare System and Department of Psychiatry, University of California San Diego, La Jolla, California, USA.
Orygen, The National Centre of Excellence in Youth Mental Health, Parkville, Australia.
Department of Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany.
SA MRC Unit on Risk & Resilience in Mental Disorders, Department of Psychiatry & Neuroscience Institute, University of Cape Town, Cape Town, South Africa.
Neuroimaging Research Branch, Intramural Research Program, National Institute on Drug Abuse, Baltimore, Maryland, USA.
Department of Psychiatry, Amsterdam UMC, Location AMC, University of Amsterdam, Amsterdam, the Netherlands.
Department of Psychiatry, VU University Medical Center, Amsterdam, the Netherlands.
Imaging Genetics Center, Stevens Institute for Neuroimaging & Informatics, Keck School of Medicine, University of Southern California, California, USA.
Department of Psychiatry, Université de Montreal, CHU Ste Justine Hospital, Montreal, Quebec, Canada.
Abstract summary
To identify neuroimaging biomarkers of alcohol dependence (AD) from structural magnetic resonance imaging, it may be useful to develop classification models that are explicitly generalizable to unseen sites and populations. This problem was explored in a mega-analysis of previously published datasets from 2,034 AD and comparison participants spanning 27 sites curated by the ENIGMA Addiction Working Group. Data were grouped into a training set used for internal validation including 1,652 participants (692 AD, 24 sites), and a test set used for external validation with 382 participants (146 AD, 3 sites). An exploratory data analysis was first conducted, followed by an evolutionary search based feature selection to site generalizable and high performing subsets of brain measurements. Exploratory data analysis revealed that inclusion of case- and control-only sites led to the inadvertent learning of site-effects. Cross validation methods that do not properly account for site can drastically overestimate results. Evolutionary-based feature selection leveraging leave-one-site-out cross-validation, to combat unintentional learning, identified cortical thickness in the left superior frontal gyrus and right lateral orbitofrontal cortex, cortical surface area in the right transverse temporal gyrus, and left putamen volume as final features. Ridge regression restricted to these features yielded a test-set area under the receiver operating characteristic curve of 0.768. These findings evaluate strategies for handling multi-site data with varied underlying class distributions and identify potential biomarkers for individuals with current AD.
Authors & Co-authors:
Hahn Sage S
Mackey Scott S
Cousijn Janna J
Foxe John J JJ
Heinz Andreas A
Hester Robert R
Hutchinson Kent K
Kiefer Falk F
Korucuoglu Ozlem O
Lett Tristram T
Li Chiang-Shan R CR
London Edythe E
Lorenzetti Valentina V
Maartje Luijten L
Momenan Reza R
Orr Catherine C
Paulus Martin M
Schmaal Lianne L
Sinha Rajita R
Sjoerds Zsuzsika Z
Stein Dan J DJ
Stein Elliot E
van Holst Ruth J RJ
Veltman Dick D
Walter Henrik H
Wiers Reinout W RW
Yucel Murat M
Thompson Paul M PM
Conrod Patricia P
Allgaier Nicholas N
Garavan Hugh H
Study Outcome
Source Link: Visit source