{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T18:11:14Z","timestamp":1790273474522,"version":"4.1.0"},"reference-count":39,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:00:00Z","timestamp":1790208000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>Depression represents a complex, multifactorial disorder influenced by the interplay of biological, behavioral, and social determinants. While extensive health surveys, including the National Health and Nutrition Examination Survey (NHANES), gather comprehensive multidomain data, existing studies tend to analyze these domains independently, overlooking the potential to identify integrated, data-driven patterns of depression risk. We have developed SYNERGY-VAE, an interpretable and multimodal deep learning framework that is designed to identify latent health subgroups and facilitate the transparent prediction of depression risk, utilizing NHANES-derived multimodal data from a large analytic cohort that spans five distinct domains from 2005 to 2018. SYNERGY-VAE utilizes a variational autoencoder to learn a shared latent representation derived from five modalities of NHANES, specifically demographic, dietary, examination, laboratory, and questionnaire data. Clustering within this latent space revealed subpopulations exhibiting distinct health signatures. To enhance interpretability, we triangulated insights utilizing encoder weights, standardized mean differences, and permutation feature importance (PFI). Machine learning classifiers were trained within each cluster employing the top 30 PFI-ranked features, with performance evaluated through the receiver operating characteristic area under the curve (AUC) in a 70\/30 train-test split. Three latent clusters emerged, each demonstrating markedly different observed prevalence rates of depression within the analytic sample, ranging from 6.8% to 10.9%. Cluster-specific models consistently surpassed pooled models in performance, with the highest predictive accuracy identified in Cluster 2 utilizing XGBoost (AUC\u2009=\u20090.839, 95% CI:0.804\u20130.874). The importance of features varied across clusters, indicating unique depression risk profiles specific to each subgroup. SYNERGY-VAE demonstrates the power of generative, explainable deep learning for the identification of latent health phenotypes, thereby furthering the objectives of precision mental health. By integrating high-dimensional, multimodal data and facilitating transparent subgroup discovery, our model may inform future stratified and context-aware screening research and contribute to precision psychiatry-oriented research in large health survey datasets.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001719","type":"journal-article","created":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T17:44:58Z","timestamp":1790271898000},"page":"e0001719","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["SYNERGY-VAE: An explainable generative deep learning framework for discovering depression subgroups from multimodal population health data"],"prefix":"10.1371","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-5022-697X","authenticated-orcid":true,"given":"Divya","family":"Sharma","sequence":"first","affiliation":[{"name":"York University","place":["Toronto, Ontario, Canada"],"department":["Department of Mathematics and Statistics"]},{"name":"Princess Margaret Cancer Centre, University Health Network","place":["Toronto, Ontario, Canada"],"department":["Biostatistics Department"]},{"name":"University of Toronto","place":["Toronto, Ontario, Canada"],"department":["Biostatistics Department, Dalla Lana School of Public Health"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alice","family":"Rueda","sequence":"additional","affiliation":[{"name":"AI for Mental Health Program, St. Michael\u2019s Hospital, Unity Health Toronto","place":["Toronto, Ontario, Canada"]},{"name":"Toronto Metropolitan University","place":["Toronto, Ontario, Canada"],"department":["Department of Electrical and Computer Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qiaowei","family":"Lin","sequence":"additional","affiliation":[{"name":"AI for Mental Health Program, St. Michael\u2019s Hospital, Unity Health Toronto","place":["Toronto, Ontario, Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shakila","family":"Meshkat","sequence":"additional","affiliation":[{"name":"AI for Mental Health Program, St. Michael\u2019s Hospital, Unity Health Toronto","place":["Toronto, Ontario, Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Argyrios","family":"Perivolaris","sequence":"additional","affiliation":[{"name":"AI for Mental Health Program, St. Michael\u2019s Hospital, Unity Health Toronto","place":["Toronto, Ontario, Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Venkat","family":"Bhat","sequence":"additional","affiliation":[{"name":"AI for Mental Health Program, St. Michael\u2019s Hospital, Unity Health Toronto","place":["Toronto, Ontario, Canada"]},{"name":"Toronto Metropolitan University","place":["Toronto, Ontario, Canada"],"department":["Department of Electrical and Computer Engineering"]},{"name":"University of Toronto","place":["Toronto, Ontario, Canada"],"department":["Department of Psychiatry"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,24]]},"reference":[{"key":"pdig.0001719.ref001","volume-title":"Depression Disorder (Depression","author":"World Health Organization","year":"2025"},{"issue":"2","key":"pdig.0001719.ref002","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1016\/S2215-0366(21)00395-3","article-title":"1990\u20132019: a systematic analysis for the global burden of disease study 2019","volume":"9","author":"GDB 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories","year":"2022","journal-title":"The Lancet Psychiatry"},{"issue":"9","key":"pdig.0001719.ref003","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1046\/j.1525-1497.2001.016009606.x","article-title":"The PHQ-9: validity of a brief depression severity measure","volume":"16","author":"K Kroenke","year":"2001","journal-title":"J Gen Intern Med"},{"issue":"10161","key":"pdig.0001719.ref004","doi-asserted-by":"crossref","first-page":"2299","DOI":"10.1016\/S0140-6736(18)31948-2","article-title":"Depression. The","volume":"392","author":"GS Malhi","year":"2018","journal-title":"Lancet"},{"issue":"7","key":"pdig.0001719.ref005","doi-asserted-by":"crossref","first-page":"1133","DOI":"10.1176\/appi.ajp.159.7.1133","article-title":"Toward a comprehensive developmental model for major depression in women","volume":"159","author":"KS Kendler","year":"2002","journal-title":"Am J Psychiatry"},{"issue":"12","key":"pdig.0001719.ref006","doi-asserted-by":"crossref","first-page":"1958","DOI":"10.1017\/S0033291719001454","article-title":"Prevalence of low-grade inflammation in depression: a systematic review and meta-analysis of CRP levels","volume":"49","author":"EF Osimo","year":"2019","journal-title":"Psychol Med"},{"issue":"5","key":"pdig.0001719.ref007","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1016\/j.biopsych.2020.01.014","article-title":"Depression Heterogeneity and Its Biological Underpinnings: Toward Immunometabolic Depression","volume":"88","author":"Y Milaneschi","year":"2020","journal-title":"Biol Psychiatry"},{"key":"pdig.0001719.ref008","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1016\/j.sleep.2018.12.001","article-title":"Insomnia symptoms and short sleep predict anxiety and worry in response to stress exposure: a prospective cohort study of medical interns","volume":"55","author":"DA Kalmbach","year":"2019","journal-title":"Sleep Med"},{"issue":"2","key":"pdig.0001719.ref009","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1093\/aje\/kwf182","article-title":"Socioeconomic inequalities in depression: a meta-analysis","volume":"157","author":"V Lorant","year":"2003","journal-title":"Am J Epidemiol"},{"key":"pdig.0001719.ref010","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1186\/s12916-015-0325-4","article-title":"Depression sum-scores don\u2019t add up: why analyzing specific depression symptoms is essential","volume":"13","author":"EI Fried","year":"2015","journal-title":"BMC Med"},{"issue":"3","key":"pdig.0001719.ref011","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/S2215-0366(15)00471-X","article-title":"Cross-trial prediction of treatment outcome in depression: a machine learning approach","volume":"3","author":"AM Chekroud","year":"2016","journal-title":"Lancet Psychiatry"},{"key":"pdig.0001719.ref012","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1146\/annurev-clinpsy-032816-045037","article-title":"Machine Learning Approaches for Clinical Psychology and Psychiatry","volume":"14","author":"DB Dwyer","year":"2018","journal-title":"Annu Rev Clin Psychol"},{"key":"pdig.0001719.ref013","doi-asserted-by":"crossref","unstructured":"Suresh H, Guttag J. A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle. In: Equity and Access in Algorithms, Mechanisms, and Optimization, 2021. 1\u20139. https:\/\/doi.org\/10.1145\/3465416.3483305","DOI":"10.1145\/3465416.3483305"},{"key":"pdig.0001719.ref014","unstructured":"Centers for Disease Control and Prevention (CDC). NHANES Questionnaires, Datasets, and Related Documentation. 2011. Available from: https:\/\/wwwn.cdc.gov\/nchs\/nhanes\/"},{"key":"pdig.0001719.ref015","first-page":"1178925","article-title":"Prediction of depressive disorder using machine learning approaches.","volume":"16","author":"HN Vu","year":"2025","journal-title":"Frontiers in Psychiatry"},{"issue":"10","key":"pdig.0001719.ref016","doi-asserted-by":"crossref","DOI":"10.1002\/hsr2.1635","article-title":"Exploring Depression and Nutritional Covariates Amongst US Adults using Shapely Additive Explanations","volume":"6","author":"AA Huang","year":"2023","journal-title":"Health Sci Rep"},{"key":"pdig.0001719.ref017","unstructured":"Wang Q, et al. Machine learning model for depression based on heavy metals among aging people in NHANES. Environmental Science and Pollution Research, 2023;30:54862\u201354873."},{"issue":"1","key":"pdig.0001719.ref018","doi-asserted-by":"crossref","first-page":"10811","DOI":"10.1038\/s41598-025-95938-3","article-title":"The interpretable machine learning model for depression associated with heavy metals via EMR mining method","volume":"15","author":"S Xu","year":"2025","journal-title":"Sci Rep"},{"key":"pdig.0001719.ref019","volume-title":"Auto-Encoding Variational Bayes","author":"DP Kingma","year":"2014"},{"key":"pdig.0001719.ref020","doi-asserted-by":"crossref","first-page":"551299","DOI":"10.3389\/fpsyt.2020.551299","article-title":"Illuminating the Black Box: Interpreting Deep Neural Network Models for Psychiatric Research","volume":"11","author":"Y-H Sheu","year":"2020","journal-title":"Front Psychiatry"},{"issue":"1","key":"pdig.0001719.ref021","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1038\/s41380-020-00931-z","article-title":"Machine learning for psychiatry: getting doctors at the black box","volume":"26","author":"DM Hedderich","year":"2021","journal-title":"Molecular Psychiatry"},{"issue":"9","key":"pdig.0001719.ref022","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1046\/j.1525-1497.2001.016009606.x","article-title":"The PHQ-9: validity of a brief depression severity measure","volume":"16","author":"K Kroenke","year":"2001","journal-title":"J Gen Intern Med"},{"key":"pdig.0001719.ref023","doi-asserted-by":"crossref","first-page":"2068","DOI":"10.1001\/jama.2023.7787","article-title":"Depression and suicide risk screening: updated evidence report and systematic review for the US Preventive Services Task Force","author":"EA O\u2019Connor","year":"2023","journal-title":"Jama"},{"key":"pdig.0001719.ref024","unstructured":"Hoffman MD, Johnson MJ. Elbo surgery: yet another way to carve up the variational evidence lower bound. In Workshop in Advances in Approximate Bayesian Inference, NIPS 2016 Dec 9 (Vol. 1, No. 2)."},{"issue":"4","key":"pdig.0001719.ref025","doi-asserted-by":"crossref","first-page":"1683","DOI":"10.1109\/TKDE.2023.3309853","article-title":"Kullback-Leibler Divergence-Based Out-of-Distribution Detection With Flow-Based Generative Models","volume":"36","author":"Y Zhang","year":"2024","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"pdig.0001719.ref026","unstructured":"Dilokthanakul N, Mediano PAM, Garnelo M, Lee MCH, Salimbeni H, Arulkumaran K, Shanahan M. Deep unsupervised clustering with Gaussian mixture variational autoencoders. Proceedings of the 5th International Conference on Learning Representations (ICLR 2017). 2017:arXiv:1611.02648."},{"key":"pdig.0001719.ref027","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1186\/1471-2105-14-106","article-title":"SMOTE for high-dimensional class-imbalanced data","volume":"14","author":"R Blagus","year":"2013","journal-title":"BMC Bioinformatics"},{"key":"pdig.0001719.ref028","doi-asserted-by":"crossref","unstructured":"Suresh H, Guttag J. A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle. In: Equity and Access in Algorithms, Mechanisms, and Optimization, 2021. 1\u20139. https:\/\/doi.org\/10.1145\/3465416.3483305","DOI":"10.1145\/3465416.3483305"},{"issue":"25","key":"pdig.0001719.ref029","doi-asserted-by":"crossref","first-page":"3083","DOI":"10.1002\/sim.3697","article-title":"Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples","volume":"28","author":"PC Austin","year":"2009","journal-title":"Stat Med"},{"key":"pdig.0001719.ref030","first-page":"177","article-title":"All Models are Wrong, but Many are Useful: Learning a Variable\u2019s Importance by Studying an Entire Class of Prediction Models Simultaneously","volume":"20","author":"A Fisher","year":"2019","journal-title":"J Mach Learn Res"},{"issue":"9","key":"pdig.0001719.ref031","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0256902","article-title":"Classifying patients with depressive and anxiety disorders according to symptom network structures: A Gaussian graphical mixture model-based clustering","volume":"16","author":"J Kashihara","year":"2021","journal-title":"PLoS One"},{"issue":"5","key":"pdig.0001719.ref032","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1097\/JGP.0b013e31819431ff","article-title":"Profiles of depressive symptoms in older adults diagnosed with major depression: latent cluster analysis","volume":"17","author":"CF Hybels","year":"2009","journal-title":"Am J Geriatr Psychiatry"},{"issue":"1","key":"pdig.0001719.ref033","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1007\/s15010-022-01822-x","article-title":"Clustering analysis reveals different profiles associating long-term post-COVID symptoms, COVID-19 symptoms at hospital admission and previous medical co-morbidities in previously hospitalized COVID-19 survivors","volume":"51","author":"C Fern\u00e1ndez-de-Las-Pe\u00f1as","year":"2023","journal-title":"Infection"},{"issue":"5","key":"pdig.0001719.ref034","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1192\/bjp.2021.103","article-title":"Data-driven clustering differentiates subtypes of major depressive disorder with distinct brain connectivity and symptom features","volume":"219","author":"Y Wang","year":"2021","journal-title":"Br J Psychiatry"},{"key":"pdig.0001719.ref035","doi-asserted-by":"crossref","first-page":"397","DOI":"10.1016\/j.jad.2023.10.066","article-title":"Associations of adolescent substance use and depressive symptoms with adult major depressive disorder in the United States: NSDUH 2016-2019","volume":"344","author":"C Xu","year":"2024","journal-title":"J Affect Disord"},{"key":"pdig.0001719.ref036","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1016\/j.jad.2011.01.011","article-title":"Insomnia as a predictor of depression: a meta-analytic evaluation of longitudinal epidemiological studies","volume":"135","author":"C Baglioni","year":"2011","journal-title":"J Affect Disord"},{"key":"pdig.0001719.ref037","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1186\/s12875-014-0185-6","article-title":"Association of self-rated health with multimorbidity, chronic disease and psychosocial factors in a large middle-aged and older cohort from general practice: a cross-sectional study","volume":"15","author":"N Mavaddat","year":"2014","journal-title":"BMC Fam Pract"},{"issue":"6","key":"pdig.0001719.ref038","doi-asserted-by":"crossref","first-page":"1089","DOI":"10.1017\/S0033291720002846","article-title":"Investigating data-driven biological subtypes of psychiatric disorders using specification-curve analysis","volume":"52","author":"L Beijers","year":"2022","journal-title":"Psychol Med"},{"issue":"9","key":"pdig.0001719.ref039","doi-asserted-by":"crossref","first-page":"1631","DOI":"10.3390\/diagnostics11091631","article-title":"Machine Learning-Based Definition of Symptom Clusters and Selection of Antidepressants for Depressive Syndrome","volume":"11","author":"IB Kim","year":"2021","journal-title":"Diagnostics (Basel)"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001719","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T17:45:12Z","timestamp":1790271912000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001719"}},"subtitle":[],"editor":[{"given":"Jing","family":"Tian","sequence":"first","affiliation":[{"name":"National University of Singapore","place":["SINGAPORE"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,24]]},"references-count":39,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,24]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001719","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,24]]}}}