{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T06:28:18Z","timestamp":1786429698370,"version":"build-2736575974"},"reference-count":38,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T00:00:00Z","timestamp":1786406400000},"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>Background<\/jats:title>\n                    <jats:p>Machine learning-based risk classification of type 2 diabetes mellitus (T2DM) has become a prevailing research direction in intelligent healthcare. Nevertheless, current studies suffer from two key limitations. First, most existing models are built on cross-sectional data and cannot capture the dynamic progression characteristics of T2DM. Second, traditional longitudinal time-series models are commonly adopted without sufficiently exploring the deep interactive relationships among multi-temporal physical examination features, limiting model predictive performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To tackle the above drawbacks, this study proposes a novel multi-dimensional feature extraction and fusion network (MFFNet) for T2DM early risk prediction based on sequential annual physical examination data. The proposed architecture comprehensively mines latent feature information from three complementary dimensions: the horizontal interaction of multiple examination indicators within a single period, the longitudinal dynamic evolution pattern of individual indicators across consecutive years, and the cross-temporal synergistic correlations among different indicators across multiple periods. Experiments were conducted on a real-world longitudinal dataset consisting of three consecutive years of physical examination records from the Health Management Center of Beijing Hospital, which presents an extremely imbalanced sample distribution with a positive\u2013negative ratio of 1:8.34.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Comparative experiments with state-of-the-art time-series models (RNN, LSTM, GRU, and Transformer) demonstrate the superiority of MFFNet. The proposed model achieved a sensitivity of 0.8766, a specificity of 0.7082, a negative predictive value (NPV) of 0.9799, and a PR-AUC of 0.4322. Furthermore, SHAP interpretability analysis identified core predictive features for T2DM risk assessment, including age, creatinine (Cr), waist circumference (WC), alanine aminotransferase (ALT), triglycerides (TG), and systolic blood pressure (SBP).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The proposed MFFNet can effectively implement accurate T2DM classification and early risk prediction using longitudinal physical examination sequences. This multi-dimensional feature fusion strategy substantially improves the mining capability of temporal healthcare data. The lightweight and cost-effective MFFNet provides a reliable artificial intelligence-assisted solution for large-scale early pre-screening and risk intervention of T2DM in primary clinical healthcare.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpubh.2026.1841719","type":"journal-article","created":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T05:40:07Z","timestamp":1786426807000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Multidimensional feature fusion in longitudinal physical examination data: a machine learning framework for classification of type 2 diabetes"],"prefix":"10.3389","volume":"14","author":[{"given":"Yan","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of General Practice, Beijing Hospital, National Center of Gerontology, National Clinical Research Center for Gerontology, The Key Laboratory of Geriatrics of NHC, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bijun","family":"Sun","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, School of Artificial Intelligence, Beijing Forestry University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiang","family":"Cai","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, School of Artificial Intelligence, Beijing Forestry University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiandong","family":"Qi","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, School of Artificial Intelligence, Beijing Forestry University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,11]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"28","DOI":"10.1038\/s41574-022-00737-9","article-title":"Endosomal trafficking in metabolic homeostasis and diseases","volume":"19","author":"Gilleron","year":"2023","journal-title":"Nat Rev Endocrinol"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"109119","DOI":"10.1016\/j.diabres.2021.109119","article-title":"IDF diabetes atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045","volume":"183","author":"Sun","year":"2022","journal-title":"Diabetes Res Clin Pract"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"12063","DOI":"10.1038\/s41598-026-45276-9","article-title":"Advancing cardiovascular screening: deep learning-based heart-sound classification using SMOTE and temporal modeling","volume":"16","author":"Ameen","year":"2026","journal-title":"Sci Rep"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1385167","DOI":"10.3389\/fendo.2024.1385167","article-title":"Utilizing machine learning for early screening of thyroid nodules: a dual-center cross-sectional study in China","volume":"15","author":"Weng","year":"2024","journal-title":"Front Endocrinol"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1186\/s40537-024-01011-7","article-title":"Advances in ECG and PCG-based cardiovascular disease classification: a review of deep learning and machine learning methods","volume":"11","author":"Ameen","year":"2024","journal-title":"J Big Data"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"37170","DOI":"10.1038\/s41598-025-24850-7","article-title":"Optimizing YOLOv11 for automated classification of breast cancer in medical images","volume":"15","author":"Abd El-Hafeez","year":"2025","journal-title":"Sci Rep"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1007\/s11882-023-01097-8","article-title":"Computer-assisted clinical diagnosis and treatment","volume":"23","author":"Hamid","year":"2023","journal-title":"Curr Allergy Asthma Rep"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1186\/s12874-022-01577-x","article-title":"Methodological conduct of prognostic prediction models developed using machine learning in oncology: a systematic review","volume":"22","author":"Dhiman","year":"2022","journal-title":"BMC Med Res Methodol"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"4223","DOI":"10.2147\/JPR.S494238","article-title":"Beyond the pain management clinic: the role of AI-integrated remote patient monitoring in chronic disease management\u2013a narrative review","volume":"17","author":"Patel","year":"2024","journal-title":"J Pain Res"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"03000605231159335","DOI":"10.1177\/03000605231159335","article-title":"Artificial intelligence in public health: the potential of epidemic early warning systems","volume":"51","author":"MacIntyre","year":"2023","journal-title":"J Int Med Res"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1186\/s13098-021-00767-9","article-title":"Machine learning and deep learning predictive models for type 2 diabetes: a systematic review","volume":"13","author":"Fregoso-Aparicio","year":"2021","journal-title":"Diabetol Metab Syndr"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"100458","DOI":"10.1016\/j.hjdsi.2020.100458","article-title":"The longitudinal epidemiologic assessment of diabetes risk (LEADR): unique 1.4 M patient electronic health record cohort","volume":"8","author":"Fishbein","year":"2020","journal-title":"Healthcare"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1038\/s41746-023-00933-5","article-title":"A scoping review of artificial intelligence-based methods for diabetes risk prediction","volume":"6","author":"Mohsen","year":"2023","journal-title":"NPJ Digit Med"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"2864","DOI":"10.3390\/healthcare11212864","article-title":"Revolutionizing diabetes diagnosis: machine learning techniques unleashed","volume":"11","author":"Shaukat","year":"2023","journal-title":"Healthcare"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"706","DOI":"10.1016\/j.procs.2020.03.336","article-title":"Prediction of type 2 diabetes using machine learning classification methods","volume":"167","author":"Tigga","year":"2020","journal-title":"Procedia Comput Sci"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"663","DOI":"10.1038\/s41598-022-27340-2","article-title":"Prediction of type 2 diabetes mellitus using hematological factors based on machine learning approaches: a cohort study analysis","volume":"13","author":"Mansoori","year":"2023","journal-title":"Sci Rep"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1808","DOI":"10.3390\/healthcare11121808","article-title":"Ensemble learning for disease prediction: a review","volume":"11","author":"Mahajan","year":"2023","journal-title":"Healthcare"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"13101","DOI":"10.1038\/s41598-023-40170-0","article-title":"Comparisons of the prediction models for undiagnosed diabetes between machine learning versus traditional statistical methods","volume":"13","author":"Choi","year":"2023","journal-title":"Sci Rep"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1186\/s13098-023-01112-y","article-title":"Machine learning for predicting diabetes risk in western China adults","volume":"15","author":"Li","year":"2023","journal-title":"Diabetol Metab Syndr"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"144777","DOI":"10.1109\/access.2019.2945129","article-title":"Development of disease prediction model based on ensemble learning approach for diabetes and hypertension","volume":"7","author":"Fitriyani","year":"2019","journal-title":"IEEE Access"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"103737","DOI":"10.1109\/access.2021.3098691","article-title":"Machine learning tools for long-term type 2 diabetes risk prediction","volume":"9","author":"Fazakis","year":"2021","journal-title":"IEEE Access"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"2744","DOI":"10.1109\/jbhi.2020.3040225","article-title":"Deep learning for diabetes: a systematic review","volume":"25","author":"Zhu","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"3989","DOI":"10.3390\/app12083989","article-title":"An optimization-based diabetes prediction model using CNN and bi-directional LSTM in real-time environment","volume":"12","author":"Madan","year":"2022","journal-title":"Appl Sci"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"3989","DOI":"10.3390\/app14103989","article-title":"Attention-oriented cnn method for type 2 diabetes prediction","volume":"14","author":"Zhao","year":"2024","journal-title":"Appl Sci (Basel)"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"6683","DOI":"10.35444\/ijana.2025.16609","article-title":"Leveraging diabetes prediction using the deep learning-based hybrid ANN-CNN architecture","volume":"16","author":"Abed","year":"2025","journal-title":"Int J Adv Netw Appl"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"2737","DOI":"10.3390\/electronics11172737","article-title":"Deep LSTM model for diabetes prediction with class balancing by SMOTE","volume":"11","author":"Alex","year":"2022","journal-title":"Electronics"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"ref29","author":"Cho","year":"2014"},{"key":"ref30","author":"Fawaz","year":"2020"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"105737","DOI":"10.1016\/j.ijmedinf.2024.105737","article-title":"Interpretable time-series neural turing machine for prognostic prediction of patients with type 2 diabetes in physician-pharmacist collaborative clinics","volume":"195","author":"Xiao","year":"2025","journal-title":"Int J Med Inform"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"3317","DOI":"10.3390\/ijerph18063317","article-title":"Prediction of type 2 diabetes based on machine learning algorithm","volume":"18","author":"Deberneh","year":"2021","journal-title":"Int J Environ Res Public Health"},{"key":"ref33","author":"Ravaut","year":"2019"},{"key":"ref34","volume-title":"Annual Report on Health Management and Health Industry Development in China No.5 (2022)","author":"Wu","year":"2022"},{"key":"ref35","volume-title":"Classification of Diabetes mellitus","year":"2019"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"111","DOI":"10.1111\/j.2517-6161.1974.tb00994.x","article-title":"Cross-validatory choice and assessment of statistical predictions","volume":"36","author":"Stone","year":"1974","journal-title":"J R Stat Soc B Stat Methodol"},{"key":"ref37","author":"Vaswani","year":"2017"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"1937","DOI":"10.1007\/s11063-018-09977-1","article-title":"Learning from imbalanced data sets with weighted cross-entropy function","volume":"50","author":"Aurelio","year":"2019","journal-title":"Neural Process Lett"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1841719\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T05:40:08Z","timestamp":1786426808000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1841719\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,11]]},"references-count":38,"alternative-id":["10.3389\/fpubh.2026.1841719"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1841719","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,11]]},"article-number":"1841719"}}