{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T06:11:19Z","timestamp":1783923079944,"version":"3.55.0"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T00:00:00Z","timestamp":1783900800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Public Health"],"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Occupational health examinations have long relied on conventional health classes to assess worker health status. However, individuals within the same examination class may differ substantially in risk sources, pathophysiological background, and potential progression patterns. This study aimed to determine whether stable hidden health subtypes exist beyond conventional occupational health classifications in a large real-world cohort and to evaluate the utility of graph-enhanced deep representation learning for subtype discovery and biomedical interpretation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We included 63,988 occupational health examination records collected from five organizational units between 2022 and 2025. A standardized master dataset was constructed by harmonizing anthropometric measures, blood pressure, glucose metabolism, lipid metabolism, liver function, kidney function, inflammatory markers, and vascular-related indicators. We first established a benchmark panel including conventional statistical models, tree-based models, and neural network models. We then developed a domain-aware prototype tabular network (DAPTN) and its graph-enhanced variant (Graph-DAPTN) to learn individual-level latent health representations, prototype distances, and hidden subtype structure. The identified subtypes were further interpreted in terms of clinical abnormality burden, risk gradients, temporal drift, and cross-domain stability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The cohort exhibited substantial heterogeneity across units and years, as well as strong coupling among biomedical indicators. Linear models showed limited performance in capturing complex occupational health phenotypes, whereas nonlinear models performed better overall. In subtype-discovery validation, Graph-DAPTN showed stronger cluster separation and class-subtype concordance than several conventional baselines. Based on this representation, six clinically interpretable hidden subtypes were identified beyond conventional health classes: resilient-healthy, hepato-inflammatory, uric-metabolic, gluco-lipotoxic, hypertensive-homocysteine, and ageing-vascular. These subtypes showed distinct abnormality spectra in glucose and lipid metabolism, blood pressure, liver enzymes, uric acid, homocysteine, and inflammatory burden, and their distributions varied across years and organizational units.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Conventional health classes in occupational health examinations do not fully capture the underlying heterogeneity of worker health status. Graph-enhanced deep representation learning can improve health-state modeling while identifying clinically meaningful hidden subtypes beneath routine classification. These findings may support more refined risk stratification, key-population screening, and unit-level health intervention planning in occupational health practice.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpubh.2026.1847173","type":"journal-article","created":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T05:35:31Z","timestamp":1783920931000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Identification of hidden subtypes in occupational health examinations and their biomedical characteristics using graph-enhanced deep representation learning"],"prefix":"10.3389","volume":"14","author":[{"given":"Zhaoli","family":"Wang","sequence":"first","affiliation":[{"name":"Department of Labor and Health, China Railway Harbin Group Co., Ltd.","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuhui","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Labor and Health, China Railway Harbin Group Co., Ltd.","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Xiang","sequence":"additional","affiliation":[{"name":"Harbin Railway Disease Prevention and Control Center, China Railway Harbin Group Co., Ltd.","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ying","family":"Wang","sequence":"additional","affiliation":[{"name":"Harbin Railway Disease Prevention and Control Center, China Railway Harbin Group Co., Ltd.","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Sun","sequence":"additional","affiliation":[{"name":"Heilongjiang Provincial Geriatric Hospital","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Pan","sequence":"additional","affiliation":[{"name":"School of Medicine, University of Electronic Science and Technology of China","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunhuo","family":"Zhang","sequence":"additional","affiliation":[{"name":"Heilongjiang Provincial Health Commission","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Ye","sequence":"additional","affiliation":[{"name":"Harbin Railway Disease Prevention and Control Center, China Railway Harbin Group Co., Ltd.","place":["Harbin, Heilongjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,13]]},"reference":[{"key":"ref1","first-page":"301","article-title":"Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi","author":"Gao","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"194","DOI":"10.1080\/11038128.2018.1441322","article-title":"Occupational health care return-to-work practices for workers with job burnout","volume":"26","author":"K\u00e4rkk\u00e4inen","year":"2019","journal-title":"Scand J Occup Ther"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"199","DOI":"10.1377\/hlthaff.2017.0059","article-title":"The work\/health relationship","volume":"36","author":"Weil","year":"2017","journal-title":"Health Aff (Millwood)"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"165","DOI":"10.5271\/sjweh.4088","article-title":"Post-COVID condition or \"long COVID\", return-to work, and occupational health research","volume":"49","author":"Descatha","year":"2023","journal-title":"Scand J Work Environ Health"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"2923","DOI":"10.1002\/jcop.22806","article-title":"Occupational health intervention for work-life balance and burnout management among teachers in rural communities","volume":"50","author":"Ogakwu","year":"2022","journal-title":"J Community Psychol"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"e26541","DOI":"10.1097\/MD.0000000000026541","article-title":"Rational emotive occupational health coaching for quality of work-life among primary school administrators","volume":"100","author":"Agu","year":"2021","journal-title":"Medicine (Baltimore)"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1016\/j.socscimed.2018.02.028","article-title":"Healthy working days: the (positive) effect of work effort on occupational health from a human capital approach","volume":"202","author":"Urtasun","year":"2018","journal-title":"Soc Sci Med"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1037\/ocp0000140","article-title":"Interventions in occupational health psychology","volume":"24","author":"Beehr","year":"2019","journal-title":"J Occup Health Psychol"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"11290","DOI":"10.3390\/ijerph182111290","article-title":"The impact of shift work on occupational health indicators among professionally active adults: a comparative study","volume":"18","author":"Pereira","year":"2021","journal-title":"Int J Environ Res Public Health"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"428","DOI":"10.2105\/AJPH.2006.086900","article-title":"Work, obesity, and occupational safety and health","volume":"97","author":"Schulte","year":"2007","journal-title":"Am J Public Health"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"420","DOI":"10.1001\/jamaoncol.2021.6987","article-title":"Cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life years for 29 cancer groups from 2010 to 2019: a systematic analysis for the global burden of disease study 2019","volume":"8","author":"Kocarnik","year":"","journal-title":"JAMA Oncol"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"917","DOI":"10.1002\/ajim.23037","article-title":"Artificial intelligence: implications for the future of work","volume":"62","author":"Howard","year":"2019","journal-title":"Am J Ind Med"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"101936","DOI":"10.1016\/j.eclinm.2023.101936","article-title":"Global burden of chronic respiratory diseases and risk factors, 1990-2019: an update from the global burden of disease study 2019","volume":"59","year":"2019","journal-title":"EClinicalMedicine"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"6726","DOI":"10.1038\/s41467-025-61650-z","article-title":"Machine learning in Alzheimer's disease genetics","volume":"16","author":"Bracher-Smith","year":"2025","journal-title":"Nat Commun"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1007\/s10926-020-09910-1","article-title":"Machine learning for work disability prevention: introduction to the special series","volume":"30","author":"Gross","year":"2020","journal-title":"J Occup Rehabil"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1067","DOI":"10.1136\/thorax-2022-219158","article-title":"Pulmonary emphysema subtypes defined by unsupervised machine learning on CT scans","volume":"78","author":"Angelini","year":"2023","journal-title":"Thorax"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"135814","DOI":"10.1016\/j.jhazmat.2024.135814","article-title":"Investigating the potential risk of cadmium exposure on seizure severity and anxiety-like behaviors through the ferroptosis pathway in epileptic mice: an integrated multi-omics approach","volume":"480","author":"Chang","year":"2024","journal-title":"J Hazard Mater"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1093\/joccuh\/uiag014","article-title":"Machine learning in the analysis of mental health at work: a scoping review","volume":"68","author":"Varje","year":"2026","journal-title":"J Occup Health"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"e690","DOI":"10.1097\/JOM.0000000000003468","article-title":"Occupational mental health: an investigation of risk indicators using interpretable machine learning techniques","volume":"67","author":"Schneider","year":"2025","journal-title":"J Occup Environ Med"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1007\/s10926-020-09888-w","article-title":"Intelligent robotics incorporating machine learning algorithms for improving functional capacity evaluation and occupational rehabilitation","volume":"30","author":"Fong","year":"2020","journal-title":"J Occup Rehabil"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"706","DOI":"10.1136\/ard-2024-225531","article-title":"EULAR recommendations for the management of psoriatic arthritis with pharmacological therapies: 2023 update","volume":"83","author":"Gossec","year":"2024","journal-title":"Ann Rheum Dis"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"2100","DOI":"10.1016\/S0140-6736(24)00367-2","article-title":"Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the global burden of disease study 2021","volume":"403","year":"","journal-title":"Lancet"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"104754","DOI":"10.1016\/j.compbiomed.2021.104754","article-title":"A systematic literature review on obesity: understanding the causes and consequences of obesity and reviewing various machine learning approaches used to predict obesity","volume":"136","author":"Safaei","year":"2021","journal-title":"Comput Biol Med"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"523","DOI":"10.1007\/s11906-022-01212-6","article-title":"Machine learning for hypertension prediction: a systematic review","volume":"24","author":"Silva","year":"2022","journal-title":"Curr Hypertens Rep"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"1502599","DOI":"10.3389\/fpubh.2025.1502599","article-title":"Machine learning and public health policy evaluation: research dynamics and prospects for challenges","volume":"13","author":"Li","year":"2025","journal-title":"Front Public Health"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1847173\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T05:35:31Z","timestamp":1783920931000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1847173\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,13]]},"references-count":25,"alternative-id":["10.3389\/fpubh.2026.1847173"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1847173","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,13]]},"article-number":"1847173"}}