{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:44:04Z","timestamp":1785545044716,"version":"3.56.0"},"reference-count":42,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Early prediction of gestational diabetes mellitus (GDM) is critical for improving maternal health outcomes. However, predictive models are often challenged by limited early-pregnancy samples, severe class imbalance in datasets, and complex interrelationships among clinical features.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and evaluate a unified dual-dimensional enhancement framework integrating data augmentation and feature engineering. By addressing data imbalance and leveraging medical prior knowledge, this framework significantly improves early GDM prediction performance.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We proposed a framework combining Generative Adversarial Network (GAN)\u2013based data augmentation with large language model\u2013inspired feature engineering. GAN sampling was used to generate clinically plausible synthetic minority class samples to mitigate data imbalance. The large language model was guided to organize features into domains (eg, basic demographics, metabolic syndrome, and core liver biomarkers) and generate higher-order composite features, integrating medical prior knowledge. Machine learning models were subsequently developed, and interpretability analyses were performed using Shapley additive explanations to identify key predictors.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n This study used a final analytical cohort of 8214 pregnant women, divided into dataset A comprising 966 out of 5251 (18.4%) participants with GDM, and dataset B comprising 598 out of 2963 (20.2%) participants with GDM. The random forest model enhanced by Tabular Variational Autoencoder\u2013based feature augmentation demonstrated the best performance. On the test dataset, it achieved a recall of 0.7559, an accuracy of 0.8444, and an area under the receiver operating characteristic curve (AUROC) of 0.8873. Statistical evaluation confirmed that the Tabular Variational Autoencoder method significantly outperformed the baseline (Cohen\n d<\/jats:italic>\n =2.894;\n P<\/jats:italic>\n <.001) and the Conditional Tabular Generative Adversarial Network method (Cohen\n d<\/jats:italic>\n =1.637;\n P<\/jats:italic>\n =.02) in recall enhancement. Shapley additive explanations analysis identified the following 5 features as the most influential predictors: fasting blood glucose, the composite feature (fasting blood glucose+triglycerides)\u00d7prepregnancy BMI, activated partial thromboplastin time, leukocyte count, and neutrophil count.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n The proposed dual-dimensional enhancement framework effectively alleviates data limitations and captures complex feature interactions in early GDM prediction. This strategy not only improves model performance, particularly in recall, but also provides interpretable biological evidence to support rapid clinical screening, stratified management, and early intervention in pregnancy.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/85335","type":"journal-article","created":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T18:55:08Z","timestamp":1776365708000},"page":"e85335-e85335","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Early Prediction of Gestational Diabetes Mellitus Through Data Augmentation and Feature Guidance: Model Development and Validation Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-1006-0862","authenticated-orcid":false,"given":"Xiekun","family":"Chen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1817-6413","authenticated-orcid":false,"given":"Zhifa","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-5330-2816","authenticated-orcid":false,"given":"Dong","family":"Su","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-1795-8665","authenticated-orcid":false,"given":"Xiaoping","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9549-9698","authenticated-orcid":false,"given":"Aiping","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4207-9728","authenticated-orcid":false,"given":"Zhen","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-0350-2355","authenticated-orcid":false,"given":"Huabin","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,5,25]]},"reference":[{"issue":"5","key":"R1","doi-asserted-by":"publisher","first-page":"763","DOI":"10.1210\/endrev\/bnac003","article-title":"A clinical update on gestational diabetes mellitus","volume":"43","author":"Sweeting","journal-title":"Endocr Rev"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","first-page":"3","DOI":"10.4093\/dmj.2021.0335","article-title":"Gestational diabetes mellitus: diagnostic approaches and maternal-offspring complications","volume":"46","author":"Moon","journal-title":"Diabetes Metab J"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","first-page":"3","DOI":"10.2337\/dc17-2197","article-title":"The effect of a smartphone-based, patient-centered diabetes care system in patients with type 2 diabetes: a randomized, controlled trial for 24 weeks","volume":"42","author":"Kim","journal-title":"Diabetes Care"},{"issue":"8","key":"R4","doi-asserted-by":"publisher","first-page":"895","DOI":"10.1016\/j.amjmed.2020.03.033","article-title":"Artificial intelligence: the future for diabetes care","volume":"133","author":"Ellahham","journal-title":"Am J Med"},{"issue":"7","key":"R5","doi-asserted-by":"publisher","first-page":"1151","DOI":"10.2337\/diacare.24.7.1151","article-title":"Gestational diabetes mellitus diagnosed with a 2-h 75-g oral glucose tolerance test and adverse pregnancy outcomes","volume":"24","author":"Schmidt","journal-title":"Diabetes Care"},{"issue":"3","key":"R6","doi-asserted-by":"publisher","first-page":"419","DOI":"10.1016\/j.diabres.2013.12.009","article-title":"Validation of early risk-prediction models for gestational diabetes based on clinical characteristics","volume":"103","author":"Th\u00e9riault","journal-title":"Diabetes Res Clin Pract"},{"issue":"3","key":"R7","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1016\/j.ajog.2010.01.024","article-title":"Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study: preeclampsia","volume":"202","author":"Yogev","journal-title":"Am J Obstet Gynecol"},{"issue":"6","key":"R8","doi-asserted-by":"publisher","first-page":"982","DOI":"10.2337\/dc16-0160","article-title":"Accelerated fetal growth prior to diagnosis of gestational diabetes mellitus: a prospective cohort study of nulliparous women","volume":"39","author":"Sovio","journal-title":"Diabetes Care"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","first-page":"8028","DOI":"10.1038\/s41598-025-92299-9","article-title":"Early prediction of postpartum dyslipidemia in gestational diabetes using machine learning models","volume":"15","author":"Jiang","journal-title":"Sci Rep"},{"issue":"1","key":"R10","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1038\/s41591-019-0724-8","article-title":"Prediction of gestational diabetes based on nationwide electronic health records","volume":"26","author":"Artzi","journal-title":"Nat Med"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejogrb.2016.03.032","article-title":"Prediction and prevention of gestational diabetes: an update of recent literature","volume":"202","author":"Kennelly","journal-title":"Eur J Obstet Gynecol Reprod Biol"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.diabres.2020.108139","article-title":"Early pregnancy prediction of gestational diabetes mellitus risk using prenatal screening biomarkers in nulliparous women","volume":"163","author":"Snyder","journal-title":"Diabetes Res Clin Pract"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.1016\/j.diabres.2021.109001","article-title":"A risk prediction model of gestational diabetes mellitus before 16 gestational weeks in Chinese pregnant women","volume":"179","author":"Wu","journal-title":"Diabetes Res Clin Pract"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2022.101637","article-title":"Development, validation and clinical utility of a risk prediction model for adverse pregnancy outcomes in women with gestational diabetes: the PeRSonal GDM model","volume":"52","author":"Cooray","journal-title":"EClinicalMedicine"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2023.105228","article-title":"Comparison of machine learning and conventional logistic regression-based prediction models for gestational diabetes in an ethnically diverse population; the Monash GDM machine learning model","volume":"179","author":"Belsti","journal-title":"Int J Med Inform"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1016\/j.diabres.2024.111865","article-title":"A prediction model for gestational diabetes mellitus based on steroid hormonal changes in early and mid-down syndrome screening: a multicenter longitudinal study","volume":"217","author":"Zhu","journal-title":"Diabetes Res Clin Pract"},{"issue":"10","key":"R17","doi-asserted-by":"publisher","first-page":"104065","DOI":"10.1016\/j.numecd.2025.104065","article-title":"Development and validation of a prediction model for gestational diabetes mellitus based on clinical characteristics and laboratory biomarkers among Chinese women","volume":"35","author":"Lyu","journal-title":"Nutr Metab Cardiovasc Dis"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1016\/j.cgpj.2024.03.008","article-title":"Dilemma in primary care facility development:a qualitative study based on grounded theory in Guilin city","volume":"1","author":"Wang","journal-title":"Chin Gen Pract"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102272","article-title":"RA-GCN: graph convolutional network for disease prediction problems with imbalanced data","volume":"75","author":"Ghorbani","journal-title":"Med Image Anal"},{"issue":"9","key":"R20","doi-asserted-by":"publisher","first-page":"1074","DOI":"10.1111\/j.1464-5491.2011.03351.x","article-title":"New International Association of the Diabetes and Pregnancy Study Groups (IADPSG) recommendations for diagnosing gestational diabetes compared with former criteria: a retrospective study on pregnancy outcome","volume":"28","author":"Lapolla","journal-title":"Diabet Med"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","DOI":"10.2196\/59423","article-title":"Development and testing of a mobile app for management of gestational diabetes in Nepal: protocol for a user-centered design study and exploratory randomized controlled trial","volume":"13","author":"Berube","journal-title":"JMIR Res Protoc"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02537-9","article-title":"Interpretable machine learning model for predicting acute kidney injury in critically ill patients","volume":"24","author":"Li","journal-title":"BMC Med Inform Decis Mak"},{"key":"R23","unstructured":"Xu L Skoularidou M Cuesta-Infante A Veeramachaneni K . Modeling tabular data using conditional GAN. Presented at: Advances in Neural Information Processing Systems (NeurIPS 2019); Dec 8-14, 2019:7335-7345; Vancouver, Canada. 2019.[doi: 10.48550\/arXiv.1907.00503] [Medline: 32103879]"},{"key":"R24","unstructured":"Lopez R Regier J Jordan MI Yosef N . Information constraints on auto-encoding variational bayes. arXiv. Preprint posted online on Nov 29, 2018. [doi: 10.48550\/arXiv.1805.08672]"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02799-3","article-title":"Predicting gestational diabetes mellitus in the first trimester using machine learning algorithms: a cross-sectional study at a hospital fertility health center in Iran","volume":"25","author":"Bigdeli","journal-title":"BMC Med Inform Decis Mak"},{"issue":"1","key":"R26","doi-asserted-by":"publisher","DOI":"10.1186\/s40001-025-02823-w","article-title":"Cross-talk between mitophagy pathways in pre-eclampsia and gestational diabetes mellitus: a systematic analysis of shared molecular mechanisms","volume":"30","author":"He","journal-title":"Eur J Med Res"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s44196-022-00110-8","article-title":"Analysis and prediction of gestational diabetes mellitus by the ensemble learning method","volume":"15","author":"Wang","journal-title":"Int J Comput Intell Syst"},{"key":"R28","unstructured":"Lundberg SM Lee SI . A unified approach to interpreting model predictions. arXiv. Preprint posted online on Nov 25, 2017. [doi: 10.48550\/arXiv.1705.07874]"},{"issue":"12","key":"R29","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2024.e32709","article-title":"A stepwise prediction and interpretation of gestational diabetes mellitus: foster the practical application of machine learning in clinical decision","volume":"10","author":"Zhou","journal-title":"Heliyon"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1186\/s12913-021-06197-7","article-title":"The costs of delivering emergency care at regional referral hospitals in Uganda: a micro-costing study","volume":"21","author":"Werner","journal-title":"BMC Health Serv Res"},{"issue":"5","key":"R31","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1016\/j.annepidem.2017.12.005","article-title":"Excess mortality patterns during 1918-1921 influenza pandemic in the state of Arizona, USA","volume":"28","author":"Dahal","journal-title":"Ann Epidemiol"},{"issue":"3","key":"R32","first-page":"249","volume":"108","author":"Will","journal-title":"Am Fam Physician"},{"key":"R33","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2021.100830","article-title":"Association of maternal lipid profile and gestational diabetes mellitus: a systematic review and meta-analysis of 292 studies and 97,880 women","volume":"34","author":"Hu","journal-title":"EClinicalMedicine"},{"key":"R34","doi-asserted-by":"publisher","DOI":"10.3389\/fpubh.2022.850191","article-title":"Application value of predictive model based on maternal coagulation function and glycolipid metabolism indicators in early diagnosis of gestational diabetes mellitus","volume":"10","author":"Zheng","journal-title":"Front Public Health"},{"issue":"4","key":"R35","doi-asserted-by":"publisher","DOI":"10.1002\/jcla.23129","article-title":"The variation degree of coagulation function is not responsible for extra risk of hemorrhage in gestational diabetes mellitus","volume":"34","author":"Dong","journal-title":"J Clin Lab Anal"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0016470","article-title":"Diabetes mellitus Is associated with shortened activated partial thromboplastin time and increased fibrinogen values","volume":"6","author":"Zhao","journal-title":"PLoS One"},{"issue":"11","key":"R37","doi-asserted-by":"publisher","DOI":"10.3390\/jcm10112419","article-title":"Effects of hyperglycemia and diabetes mellitus on coagulation and hemostasis","volume":"10","author":"Li","journal-title":"J Clin Med"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","DOI":"10.1186\/s12884-020-03455-0","article-title":"Comparison of factor XII levels in gestational diabetes, fetal macrosomia, and healthy pregnancies","volume":"20","author":"Ozbasli","journal-title":"BMC Pregnancy Childbirth"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","DOI":"10.1186\/s12884-023-05766-4","article-title":"Development of machine learning models to predict gestational diabetes risk in the first half of pregnancy","volume":"23","author":"Cubillos","journal-title":"BMC Pregnancy Childbirth"},{"issue":"1","key":"R40","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-46735-3","article-title":"Machine learning vs. traditional regression analysis for fluid overload prediction in the ICU","volume":"13","author":"Sikora","journal-title":"Sci Rep"},{"key":"R41","doi-asserted-by":"publisher","DOI":"10.1155\/2020\/4168340","article-title":"Comparison of machine learning methods and conventional logistic regressions for predicting gestational diabetes using routine clinical data: a retrospective cohort study","volume":"2020","author":"Ye","journal-title":"J Diabetes Res"},{"key":"R42","doi-asserted-by":"publisher","DOI":"10.1109\/RBME.2023.3242261","article-title":"Digital health and machine learning technologies for blood glucose monitoring and management of gestational diabetes","volume":"17","author":"Lu","journal-title":"IEEE Rev Biomed Eng"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e85335","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:25:50Z","timestamp":1785536750000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e85335"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,25]]},"references-count":42,"URL":"https:\/\/doi.org\/10.2196\/85335","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,25]]},"article-number":"v14i5e85335"}}