{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T18:13:40Z","timestamp":1788459220061,"version":"build-2803163510"},"reference-count":33,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"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>Blood glucose prediction is a critical component of next-generation diabetes technologies, such as artificial pancreas systems, where reliable performance is essential for safety and effectiveness. Although deep learning methods have achieved promising advances in this area, a critical gap remains in understanding the reproducibility and generalizability of these methods. To contextualize the gap, this study reviewed 67 recent papers that proposed a deep learning method for glucose prediction to identify key reproducibility challenges. Next, we adopted a standardized framework, encompassing technical, statistical, and conceptual reproducibility evaluations, to experimentally assess the reproducibility of eight representative deep learning methods. To achieve this, we reimplemented and evaluated these eight deep learning methods using over 1.36 million continuous glucose monitoring samples (5,061 days) from 128 individuals with type 1 diabetes across three public datasets: OhioT1DM, DiaTrend, and T1DEXI. We found that even though these models demonstrated good technical and statistical reproducibility, their conceptual reproducibility\u2014the ability to generalize to datasets with different diabetes management patterns\u2014was limited. Further analyses revealed that each model\u2019s overall prediction performance was strongly influenced by individual glycemic control, with higher prediction errors observed among participants with lower time with blood glucose in the target range (70\u2013180 mg\/dL). This study identified key reproducibility challenges associated with current blood glucose prediction methods within type 1 diabetes populations, highlighting the need for increased transparency, dataset diversity, standardized evaluation practices, and code accessibility to ensure reproducible and reliable models for blood glucose prediction.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001633","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T17:22:02Z","timestamp":1788456122000},"page":"e0001633","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep learning for blood glucose prediction: Reproducibility challenges and factors affecting differential performance"],"prefix":"10.1371","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6345-235X","authenticated-orcid":true,"given":"Baiying","family":"Lu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Biratal","family":"Wagle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhaohui","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanjun","family":"Cui","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6750-202X","authenticated-orcid":true,"given":"Temiloluwa","family":"Prioleau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"issue":"10397","key":"pdig.0001633.ref001","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1016\/S0140-6736(23)01301-6","article-title":"Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021","volume":"402","author":"GBD 2021 Diabetes Collaborators","year":"2023","journal-title":"Lancet"},{"issue":"1","key":"pdig.0001633.ref002","article-title":"Number of People with Type 2 Diabetes Mellitus in 2035 and 2050: A Modelling Study in 188 Countries","volume":"21","author":"WC Guzman-Vilca","year":"2024","journal-title":"Curr Diabetes Rev"},{"issue":"8","key":"pdig.0001633.ref003","doi-asserted-by":"crossref","first-page":"704","DOI":"10.1001\/jama.2021.9883","article-title":"Trends in prevalence of diabetes and control of risk factors in diabetes among US adults, 1999-2018","volume":"326","author":"L Wang","year":"2021","journal-title":"JAMA"},{"key":"pdig.0001633.ref004","doi-asserted-by":"crossref","first-page":"3086167","DOI":"10.1155\/2018\/3086167","article-title":"Complications of Diabetes 2017","volume":"2018","author":"K Papatheodorou","year":"2018","journal-title":"J Diabetes Res"},{"issue":"2","key":"pdig.0001633.ref005","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1089\/dia.2018.0384","article-title":"State of Type 1 Diabetes Management and Outcomes from the T1D Exchange in 2016-2018","volume":"21","author":"NC Foster","year":"2019","journal-title":"Diabetes Technol Ther"},{"key":"pdig.0001633.ref006","article-title":"Diabetes technology: standards of care in diabetes\u20142024","volume":"47","year":"2024","journal-title":"Diabetes Care"},{"issue":"1","key":"pdig.0001633.ref007","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.ecl.2019.10.009","article-title":"Evolution of Diabetes Technology","volume":"49","author":"K Dovc","year":"2020","journal-title":"Endocrinol Metab Clin North Am"},{"issue":"9","key":"pdig.0001633.ref008","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1371\/journal.pone.0310801","article-title":"Deep learning for blood glucose level prediction: How well do models generalize across different data sets?","volume":"19","author":"S Ghimire","year":"2024","journal-title":"PLOS ONE"},{"issue":"11","key":"pdig.0001633.ref009","doi-asserted-by":"crossref","first-page":"1101","DOI":"10.1001\/jama.2018.11100","article-title":"Deep Learning-A Technology With the Potential to Transform Health Care","volume":"320","author":"G Hinton","year":"2018","journal-title":"JAMA"},{"issue":"7","key":"pdig.0001633.ref010","doi-asserted-by":"crossref","first-page":"2744","DOI":"10.1109\/JBHI.2020.3040225","article-title":"Deep Learning for Diabetes: A Systematic Review","volume":"25","author":"T Zhu","year":"2021","journal-title":"IEEE J Biomed Health Inform"},{"issue":"5","key":"pdig.0001633.ref011","doi-asserted-by":"crossref","first-page":"3819","DOI":"10.1109\/JBHI.2025.3630214","article-title":"Systematic Review on Deep Learning Algorithms for Blood Glucose Forecasting in Type 1 Diabetes","volume":"30","author":"A Calzavara","year":"2026","journal-title":"IEEE J Biomed Health Inform"},{"key":"pdig.0001633.ref012","first-page":"71","article-title":"The OhioT1DM Dataset for Blood Glucose Level Prediction: Update 2020","volume":"2675","author":"C Marling","year":"2020","journal-title":"CEUR Workshop Proc"},{"issue":"586","key":"pdig.0001633.ref013","article-title":"Reproducibility in machine learning for health research: Still a ways to go","volume":"13","author":"MBA McDermott","year":"2021","journal-title":"Sci Transl Med"},{"issue":"1","key":"pdig.0001633.ref014","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1038\/s41597-023-02469-5","article-title":"DiaTrend: A dataset from advanced diabetes technology to enable development of novel analytic solutions","volume":"10","author":"T Prioleau","year":"2023","journal-title":"Sci Data"},{"key":"pdig.0001633.ref015","unstructured":"Jaeb Center for Health Research. Type 1 Diabetes EXercise Initiative: The Effect of Exercise on Glycemic Control in Type 1 Diabetes Study. https:\/\/doi.org\/10.25934\/PR00008428"},{"issue":"3","key":"pdig.0001633.ref016","doi-asserted-by":"crossref","first-page":"575","DOI":"10.1177\/1932296818812062","article-title":"Measures of Accuracy for Continuous Glucose Monitoring and Blood Glucose Monitoring Devices","volume":"13","author":"G Freckmann","year":"2019","journal-title":"J Diabetes Sci Technol"},{"issue":"8","key":"pdig.0001633.ref017","doi-asserted-by":"crossref","first-page":"1593","DOI":"10.2337\/dci19-0028","article-title":"Clinical Targets for Continuous Glucose Monitoring Data Interpretation: Recommendations From the International Consensus on Time in Range","volume":"42","author":"T Battelino","year":"2019","journal-title":"Diabetes Care"},{"issue":"5","key":"pdig.0001633.ref018","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0252158","article-title":"Time in blood glucose range 70 to 180 mg\/dL and survival rate in critically ill patients: A retrospective cohort study","volume":"16","author":"H Naraba","year":"2021","journal-title":"PLoS One"},{"key":"pdig.0001633.ref019","doi-asserted-by":"crossref","first-page":"273","DOI":"10.4093\/dmj.2015.39.4.273","article-title":"Glycemic variability: how do we measure it and why is it important?","volume":"39","author":"S Suh","year":"2015","journal-title":"Diabetes & Metabolism Journal"},{"issue":"1","key":"pdig.0001633.ref020","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s41666-019-00059-y","article-title":"Blood Glucose Prediction with Variance Estimation Using Recurrent Neural Networks","volume":"4","author":"J Martinsson","year":"2019","journal-title":"J Healthc Inform Res"},{"issue":"6","key":"pdig.0001633.ref021","article-title":"Machine learning-based glucose prediction with use of continuous glucose and physical activity monitoring data: The Maastricht Study","volume":"16","author":"WPTM van Doorn","year":"2021","journal-title":"PLoS One"},{"issue":"2","key":"pdig.0001633.ref022","doi-asserted-by":"crossref","first-page":"414","DOI":"10.1109\/JBHI.2019.2931842","article-title":"GluNet: A Deep Learning Framework for Accurate Glucose Forecasting","volume":"24","author":"K Li","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"issue":"1","key":"pdig.0001633.ref023","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1038\/s41746-021-00480-x","article-title":"Deep transfer learning and data augmentation improve glucose levels prediction in type 2 diabetes patients","volume":"4","author":"Y Deng","year":"2021","journal-title":"NPJ Digit Med"},{"issue":"1","key":"pdig.0001633.ref024","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1186\/s12911-021-01462-5","article-title":"Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction","volume":"21","author":"MF Rabby","year":"2021","journal-title":"BMC Med Inform Decis Mak"},{"issue":"3","key":"pdig.0001633.ref025","doi-asserted-by":"crossref","first-page":"1600","DOI":"10.1109\/JBHI.2023.3236822","article-title":"Glucose Transformer: Forecasting Glucose Level and Events of Hyperglycemia and Hypoglycemia","volume":"27","author":"S-M Lee","year":"2023","journal-title":"IEEE J Biomed Health Inform"},{"issue":"8","key":"pdig.0001633.ref026","doi-asserted-by":"crossref","first-page":"5424","DOI":"10.1109\/JBHI.2024.3428921","article-title":"Multi-Horizon Glucose Prediction Across Populations With Deep Domain Generalization","volume":"29","author":"T Zhu","year":"2025","journal-title":"IEEE J Biomed Health Inform"},{"key":"pdig.0001633.ref027","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1016\/j.ifacol.2025.06.027","article-title":"A comparative study of transformer-based models for multi-horizon blood glucose prediction","volume":"59","author":"MA Karagoz","year":"2025","journal-title":"IFAC-PapersOnLine"},{"issue":"4","key":"pdig.0001633.ref028","doi-asserted-by":"crossref","first-page":"538","DOI":"10.2337\/dc16-2482","article-title":"REPLACE-BG: A Randomized Trial Comparing Continuous Glucose Monitoring With and Without Routine Blood Glucose Monitoring in Adults With Well-Controlled Type 1 Diabetes","volume":"40","author":"G Aleppo","year":"2017","journal-title":"Diabetes Care"},{"key":"pdig.0001633.ref029","doi-asserted-by":"crossref","first-page":"1707","DOI":"10.1056\/NEJMoa1907863","article-title":"Six-Month Randomized, Multicenter Trial of Closed-Loop Control in Type 1 Diabetes","volume":"381","author":"SA Brown","year":"2019","journal-title":"New England Journal of Medicine"},{"key":"pdig.0001633.ref030","unstructured":"Nie Y, Nguyen NH, Sinthong P, Kalagnanam J. A time series is worth 64 words: Long-term forecasting with transformers. In: 2023. https:\/\/arxiv.org\/abs\/2211.14730"},{"key":"pdig.0001633.ref031","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1109\/RBME.2023.3331297","article-title":"Artificial Intelligence and Machine Learning for Improving Glycemic Control in Diabetes: Best Practices, Pitfalls, and Opportunities","volume":"17","author":"PG Jacobs","year":"2024","journal-title":"IEEE Reviews in Biomedical Engineering"},{"issue":"19","key":"pdig.0001633.ref032","doi-asserted-by":"crossref","first-page":"6235","DOI":"10.3390\/s24196235","article-title":"Experimental Study of Long Short-Term Memory and Transformer Models for Fall Detection on Smartwatches","volume":"24","author":"ST Haque","year":"2024","journal-title":"Sensors (Basel)"},{"key":"pdig.0001633.ref033","first-page":"11121","article-title":"Are Transformers Effective for Time Series Forecasting?","volume":"37","author":"A Zeng","year":"2022","journal-title":"Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001633","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T17:22:11Z","timestamp":1788456131000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001633"}},"subtitle":[],"editor":[{"given":"Amara","family":"Tariq","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":33,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,3]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001633","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]}}}