Population modeling with machine learning can enhance measures of mental health

Background Biological aging is revealed by physical measures, e.g., DNA probes or brain scans. Instead, individual differences in mental function are explained by psychological constructs, e.g., intelligence or neuroticism. These constructs are typically assessed by tailored neuropsychological tests that build on expert judgement and require careful interpretation. Could machine learning on large samples from the general population be used to build proxy measures of these constructs that do not require human intervention? Results Here, we built proxy measures by applying machine learning on multimodal MR images and rich sociodemographic information from the largest biomedical cohort to date: the UK Biobank. Objective model comparisons revealed that all proxies captured the target constructs and were as useful, and sometimes more useful than the original measures for characterizing real-world health behavior (sleep, exercise, tobacco, alcohol consumption). We observed this complementarity of proxy measures and original measures when modeling from brain signals or sociodemographic data, capturing multiple health-related constructs. Conclusions Population modeling with machine learning can derive measures of mental health from brain signals and questionnaire data, which may complement or even substitute for psychometric assessments in clinical populations. Key Points We applied machine learning on more than 10.000 individuals from the general population to define empirical approximations of health-related psychological measures that do not require human judgment. We found that machine-learning enriched the given psychological measures via approximation from brain and sociodemographic data: Resulting proxy measures related as well or better to real-world health behavior than the original measures. Model comparisons showed that sociodemographic information contributed most to characterizing psychological traits beyond aging.

[1]  Dustin Scheinost,et al.  Resample aggregating improves the generalizability of connectome predictive modeling , 2021, NeuroImage.

[2]  V. Arolt,et al.  Systematic misestimation of machine learning performance in neuroimaging studies of depression , 2019, Neuropsychopharmacology.

[3]  R. Adolphs,et al.  Personality beyond taxonomy , 2020, Nature Human Behaviour.

[4]  Danilo Bzdok,et al.  Inference and Prediction Diverge in Biomedicine , 2020, Patterns.

[5]  Danilo Bzdok,et al.  Different scaling of linear models and deep learning in UKBiobank brain images versus machine-learning datasets , 2020, Nature Communications.

[6]  Simon B. Eickhoff,et al.  Meta-matching: a simple framework to translate phenotypic predictive models from big to small data , 2020, bioRxiv.

[7]  Klaus P. Ebmeier,et al.  Prediction of brain age and cognitive age: Quantifying brain and cognitive maintenance in aging , 2020, Human brain mapping.

[8]  Gaël Varoquaux,et al.  Combining magnetoencephalography with magnetic resonance imaging enhances learning of surrogate-biomarkers , 2020, eLife.

[9]  Brian A. Gordon,et al.  Functional brain age prediction suggests accelerated aging in preclinical familial Alzheimer’s disease, irrespective of fibrillar amyloid-beta pathology , 2020, bioRxiv.

[10]  Tonio Ball,et al.  Machine-learning-based diagnostics of EEG pathology , 2020, NeuroImage.

[11]  O. Andreassen,et al.  Brain Connectome Mapping of Complex Human Traits and Their Polygenic Architecture Using Machine Learning , 2019, Biological Psychiatry.

[12]  Mark W. Woolrich,et al.  Optimising network modelling methods for fMRI , 2019, NeuroImage.

[13]  Gael Varoquaux,et al.  Establishment of Best Practices for Evidence for Prediction: A Review. , 2019, JAMA psychiatry.

[14]  Lloyd T. Elliott,et al.  Brain aging comprises many modes of structural and functional change with distinct genetic and biophysical associations , 2019, bioRxiv.

[15]  Wiro J Niessen,et al.  Gray Matter Age Prediction as a Biomarker for Risk of Dementia , 2019, Proceedings of the National Academy of Sciences.

[16]  Gael Varoquaux,et al.  Manifold-regression to predict from MEG/EEG brain signals without source modeling , 2019, NeurIPS.

[17]  Patrick G. Bissett,et al.  Uncovering the structure of self-regulation through data-driven ontology discovery , 2019, Nature Communications.

[18]  Gaël Varoquaux,et al.  Benchmarking functional connectome-based predictive models for resting-state fMRI , 2019, NeuroImage.

[19]  Stuart J. Ritchie,et al.  Structural brain imaging correlates of general intelligence in UK Biobank , 2019, Intelligence.

[20]  Patrick G. Bissett,et al.  Large-scale analysis of test–retest reliabilities of self-regulation measures , 2019, Proceedings of the National Academy of Sciences.

[21]  Stephen M. Smith,et al.  Estimation of brain age delta from brain imaging , 2019, NeuroImage.

[22]  Gaël Varoquaux,et al.  On the consistency of supervised learning with missing values , 2019, ArXiv.

[23]  J. Kable,et al.  Are Bigger Brains Smarter? Evidence From a Large-Scale Preregistered Study , 2018, Psychological science.

[24]  Wesley K. Thompson,et al.  A Nonlinear Simulation Framework Supports Adjusting for Age When Analyzing BrainAGE , 2018, bioRxiv.

[25]  Mert R. Sabuncu,et al.  Is deep learning better than kernel regression for functional connectivity prediction of fluid intelligence? , 2018, 2018 International Workshop on Pattern Recognition in Neuroimaging (PRNI).

[26]  Paola Galdi,et al.  A distributed brain network predicts general intelligence from resting-state human neuroimaging data , 2018, bioRxiv.

[27]  Delia Fuhrmann,et al.  The neural determinants of age-related changes in fluid intelligence: a pre-registered, longitudinal analysis in UK Biobank , 2018, Wellcome open research.

[28]  Danilo Bzdok,et al.  Points of Significance: Statistics versus machine learning , 2018, Nature Methods.

[29]  Paola Galdi,et al.  Resting-State Functional Brain Connectivity Best Predicts the Personality Dimension of Openness to Experience , 2018, Personality Neuroscience.

[30]  Peter B. Jones,et al.  Association of Childhood Infection With IQ and Adult Nonaffective Psychosis in Swedish Men , 2018, JAMA psychiatry.

[31]  A. Meyer-Lindenberg,et al.  Machine Learning for Precision Psychiatry: Opportunities and Challenges. , 2017, Biological psychiatry. Cognitive neuroscience and neuroimaging.

[32]  Gaël Varoquaux,et al.  Cross-validation failure: Small sample sizes lead to large error bars , 2017, NeuroImage.

[33]  Stuart J. Ritchie,et al.  Brain age predicts mortality , 2017, Molecular Psychiatry.

[34]  Ludovica Griffanti,et al.  Image processing and Quality Control for the first 10,000 brain imaging datasets from UK Biobank , 2017, NeuroImage.

[35]  C. Sudlow,et al.  Comparison of Sociodemographic and Health-Related Characteristics of UK Biobank Participants With Those of the General Population , 2017, American journal of epidemiology.

[36]  N. Wray,et al.  A direct test of the diathesis–stress model for depression , 2017, Molecular Psychiatry.

[37]  Nicola Filippini,et al.  Moderate alcohol consumption as risk factor for adverse brain outcomes and cognitive decline: longitudinal cohort study , 2017, British Medical Journal.

[38]  Luke J. Chang,et al.  Building better biomarkers: brain models in translational neuroimaging , 2017, Nature Neuroscience.

[39]  J. Ioannidis,et al.  Empirical assessment of published effect sizes and power in the recent cognitive neuroscience and psychology literature , 2017, PLoS biology.

[40]  T. Paus,et al.  Studying neuroanatomy using MRI , 2017, Nature Neuroscience.

[41]  A. Kaufman,et al.  Association of Fluid Intelligence and Psychiatric Disorders in a Population-Representative Sample of US Adolescents , 2017, JAMA psychiatry.

[42]  Giovanni Montana,et al.  Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker , 2016, NeuroImage.

[43]  Daniel S. Margulies,et al.  Predicting brain-age from multimodal imaging data captures cognitive impairment , 2016, NeuroImage.

[44]  Andres Hoyos Idrobo,et al.  Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines , 2016, NeuroImage.

[45]  Do P. M. Tromp,et al.  Dispositional negativity: An integrative psychological and neurobiological perspective. , 2016, Psychological bulletin.

[46]  P. Matthews,et al.  Multimodal population brain imaging in the UK Biobank prospective epidemiological study , 2016, Nature Neuroscience.

[47]  David J. Spiegelhalter,et al.  How old are you, really? Communicating chronic risk through ‘effective age’ of your body and organs , 2016, BMC Medical Informatics and Decision Making.

[48]  Nancy Kanwisher,et al.  Connectivity precedes function in the development of the visual word form area , 2016, Nature Neuroscience.

[49]  Josselin Houenou,et al.  Can neuroimaging disentangle bipolar disorder? , 2016, Journal of affective disorders.

[50]  Joanna M. Wardlaw,et al.  Brain volumetric changes and cognitive ageing during the eighth decade of life , 2015, Human brain mapping.

[51]  R. Power,et al.  Heritability estimates of the Big Five personality traits based on common genetic variants , 2015, Translational Psychiatry.

[52]  Robert Leech,et al.  Prediction of brain age suggests accelerated atrophy after traumatic brain injury , 2015, Annals of neurology.

[53]  P. Elliott,et al.  UK Biobank: An Open Access Resource for Identifying the Causes of a Wide Range of Complex Diseases of Middle and Old Age , 2015, PLoS medicine.

[54]  M. Crawford,et al.  Classification, assessment, prevalence, and effect of personality disorder , 2015, The Lancet.

[55]  M. Kosinski,et al.  Computer-based personality judgments are more accurate than those made by humans , 2015, Proceedings of the National Academy of Sciences.

[56]  T. Yarkoni.  Neurobiological substrates of personality: A critical overview. , 2015 .

[57]  Denis Bratko,et al.  Heritability of personality: A meta-analysis of behavior genetic studies. , 2015, Psychological bulletin.

[58]  Kathryn Asbury,et al.  The high heritability of educational achievement reflects many genetically influenced traits, not just intelligence , 2014, Proceedings of the National Academy of Sciences.

[59]  Christos Davatzikos,et al.  Accelerated brain aging in schizophrenia and beyond: a neuroanatomical marker of psychiatric disorders. , 2014, Schizophrenia bulletin.

[60]  Michael Eickenberg,et al.  Machine learning for neuroimaging with scikit-learn , 2014, Front. Neuroinform..

[61]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[62]  G. Andrew,et al.  arm: Data Analysis Using Regression and Multilevel/Hierarchical Models , 2014 .

[63]  Stefan Klein,et al.  Improving alignment in Tract-based spatial statistics: Evaluation and optimization of image registration , 2013, NeuroImage.

[64]  T. Insel,et al.  Why has it taken so long for biological psychiatry to develop clinical tests and what to do about it? , 2012, Molecular Psychiatry.

[65]  R. Collins.  What makes UK Biobank special? , 2012, The Lancet.

[66]  R. Perlis,et al.  Translating biomarkers to clinical practice , 2011, Molecular Psychiatry.

[67]  Jennifer H. Barnett,et al.  A quantitative meta-analysis of population-based studies of premorbid intelligence and schizophrenia , 2011, Schizophrenia Research.

[68]  Daniele Quercia,et al.  Our Twitter Profiles, Our Selves: Predicting Personality with Twitter , 2011, 2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing.

[69]  Thomas R. Carretta,et al.  Pilot Candidate Selection Method , 2011 .

[70]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[71]  Jonathan D. Power,et al.  Prediction of Individual Brain Maturity Using fMRI , 2010, Science.

[72]  Gaël Varoquaux,et al.  Detection of Brain Functional-Connectivity Difference in Post-stroke Patients Using Group-Level Covariance Modeling , 2010, MICCAI.

[73]  Pedro M. Valero-Mora,et al.  ggplot2: Elegant Graphics for Data Analysis , 2010 .

[74]  T. Insel,et al.  Wesleyan University From the SelectedWorks of Charles A . Sanislow , Ph . D . 2010 Research Domain Criteria ( RDoC ) : Toward a New Classification Framework for Research on Mental Disorders , 2018 .

[75]  Russell A. Matthews,et al.  The relationships of working memory, secondary memory, and general fluid intelligence: working memory is special. , 2010, Journal of experimental psychology. Learning, memory, and cognition.

[76]  Jörn Diedrichsen,et al.  A probabilistic MR atlas of the human cerebellum , 2009, NeuroImage.

[77]  B. Lahey,et al.  Public health significance of neuroticism. , 2009, The American psychologist.

[78]  Joseph Hilbe,et al.  Data Analysis Using Regression and Multilevel/Hierarchical Models , 2009 .

[79]  R. O’Brien,et al.  A Caution Regarding Rules of Thumb for Variance Inflation Factors , 2007 .

[80]  A. Toga,et al.  Tracking Alzheimer's Disease , 2007, Annals of the New York Academy of Sciences.

[81]  Anders M. Dale,et al.  An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest , 2006, NeuroImage.

[82]  K. Kendler,et al.  A population-based twin study of the relationship between neuroticism and internalizing disorders. , 2006, The American journal of psychiatry.

[83]  P. Sullivan,et al.  Heritability and nineteen-year stability of long and short EPQ-R neuroticism scales , 2006 .

[84]  D. Borsboom.  Measuring the mind: Conceptual issues in contemporary psychometrics , 2005 .

[85]  Antonio Terracciano,et al.  Smoking and the Five-Factor Model of personality. , 2004, Addiction.

[86]  M. Greicius,et al.  Default-mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI , 2004, Proc. Natl. Acad. Sci. USA.

[87]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[88]  D. Borsboom,et al.  The concept of validity. , 2004, Psychological review.

[89]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[90]  Timothy Edward John Behrens,et al.  Characterization and propagation of uncertainty in diffusion‐weighted MR imaging , 2003, Magnetic resonance in medicine.

[91]  Olivier Ledoit,et al.  Honey, I Shrunk the Sample Covariance Matrix , 2003 .

[92]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[93]  M. Raichle,et al.  Tracking neuronal fiber pathways in the living human brain. , 1999, Proceedings of the National Academy of Sciences of the United States of America.

[94]  R. Lynn,et al.  Gender differences in extraversion, neuroticism, and psychoticism in 37 nations. , 1997, The Journal of social psychology.

[95]  B. Biswal,et al.  Functional connectivity in the motor cortex of resting human brain using echo‐planar mri , 1995, Magnetic resonance in medicine.

[96]  Thomas R. Carretta,et al.  Pilot-Candidate Selection Method: Sources of Validity , 1994 .

[97]  P. Costa,et al.  Normal Personality Assessment in Clinical Practice: The NEO Personality Inventory. , 1992 .

[98]  G. Mcclearn,et al.  Neuroticism, extraversion, and related traits in adult twins reared apart and reared together. , 1988, Journal of personality and social psychology.

[99]  H. Eysenck,et al.  A revised version of the Psychoticism scale. , 1985 .

[100]  John L. Horn,et al.  Apprehension, Memory, and Fluid Intelligence Decline in Adulthood , 1981 .

[101]  R. Cattell.  Theory of fluid and crystallized intelligence: A critical experiment. , 1963 .

[102]  B. Kotkov,et al.  The Meaning and Measurement of Neuroticism and Anxiety , 1962 .

[103]  Eysenck Hj.  The continuity of abnormal and normal behavior. , 1958 .

[104]  L. Cronbach,et al.  Construct validity in psychological tests. , 1955, Psychological bulletin.

[105]  S S Stevens,et al.  On the Theory of Scales of Measurement. , 1946, Science.