ARTIFICIAL INTELLIGENCE FOR PREDICTION OF EXERCISE-ASSOCIATED HYPOGLYCEMIA IN PHYSICALLY ACTIVE PEOPLE WITH TYPE 1 DIABETES

Aleksander Krupski; Michał  Marusza; Victoria Stielow; Antonina  Zatyka

doi:10.31435/ijitss.3(51).2026.5802

Authors

Aleksander Krupski National Medical Institute of the Ministry of the Interior and Administration in Warsaw, ul. Wołoska 137, 02-507 Warsaw, Poland https://orcid.org/0000-0002-6432-4982
Michał Marusza National Medical Institute of the Ministry of the Interior and Administration in Warsaw, ul. Wołoska 137, 02-507 Warsaw, Poland https://orcid.org/0009-0008-0065-9928
Victoria Stielow Faculty of Medicine, Lazarski University, ul. Świeradowska 43, 02-662 Warsaw, Poland https://orcid.org/0009-0007-0954-0918
Antonina Zatyka Faculty of Medicine, Lazarski University, ul. Świeradowska 43, 02-662 Warsaw, Poland https://orcid.org/0009-0009-7849-6978

DOI:

https://doi.org/10.31435/ijitss.3(51).2026.5802

Keywords:

Artificial Intelligence, Machine Learning, Continuous Glucose Monitoring, Type 1 Diabetes, Physical Activity

Abstract

Research objectives. Physical activity continues to be a major barrier to achieving ideal blood glucose control in individuals with type 1 diabetes due to the ongoing threat of hypoglycemia. Traditional alerts from Continuous Glucose Monitoring (CGM) systems are mostly reactive as they fail to consider the specific context of physical activity. The purpose of this project is to provide a critical overview of evidence regarding AI and ML techniques for identifying when exercise will cause low blood glucose in individuals who are physically active and have type 1 diabetes.

Methods. The primary databases searched for this narrative review with a structured literature search were PubMed/MEDLINE and PubMed Central.

Key findings. Exercise-specific models based on repeated-measures random forest and logistic regression achieved AUROC values around 0.83, while broader explainable XGBoost models using CGM data showed excellent short-term predictive performance. The use of simple translation of clinical knowledge into tool-based formats such as GlucoseGO has been shown to retain high levels of performance when only using limited number of easily obtained variables.

Conclusions. AI will allow for a meaningful integration with traditional monitoring in order to shift clinical practice from detecting problems after they occur toward predicting the level of risk each patient has when participating in an activity. The direction in which this will be clinically beneficial will be through developing cost-effective, interpretable, validated tools for use within the current CGM and automated insulin delivery systems.

References

Adolfsson, P., Taplin, C. E., Zaharieva, D. P., Pemberton, J., Davis, E. A., Riddell, M. C., et al. (2022). ISPAD Clinical Practice Consensus Guidelines 2022: Exercise in children and adolescents with diabetes. Pediatric Diabetes, 23(8), 1341–1372. https://doi.org/10.1111/pedi.13452

American Diabetes Association Professional Practice Committee. (2025). 7. Diabetes technology: Standards of Care in Diabetes—2025. Diabetes Care, 48(Suppl. 1), S146–S166. https://doi.org/10.2337/dc25-S007

Bergford, S., Riddell, M. C., Jacobs, P. G., Gal, R. L., Grimes, C., DuBose, S. N., et al. (2023). The Type 1 Diabetes and EXercise Initiative: Predicting hypoglycemia risk during exercise for participants with type 1 diabetes using repeated measures random forest. Diabetes Technology & Therapeutics, 25(9), 602–611. https://doi.org/10.1089/dia.2023.0140

Bergford, S., Riddell, M. C., Gal, R. L., Willi, S. M., Sherr, J. L., Pinsker, J. E., et al. (2024). Predicting hypoglycemia and hyperglycemia risk during and after activity for adolescents with type 1 diabetes. Diabetes Technology & Therapeutics, 26(10), 728–738. https://doi.org/10.1089/dia.2024.0061

Cavallo, M., De Fano, M., Barana, L., et al. (2024). Nutritional management of athletes with type 1 diabetes: A narrative review. Nutrients, 16(6), 907. https://doi.org/10.3390/nu16060907

Denes-Fazakas, L., Siket, M., Szilagyi, L., Kovacs, L., & Eigner, G. (2022). Detection of physical activity using machine learning methods based on continuous blood glucose monitoring and heart rate signals. Sensors, 22(21), 8568. https://doi.org/10.3390/s22218568

Duckworth, C., Guy, M. J., Kumaran, A., Boniface, M., et al. (2024). Explainable machine learning for real-time hypoglycemia and hyperglycemia prediction and personalized control recommendations. Journal of Diabetes Science and Technology, 18(2), 506–517. https://doi.org/10.1177/19322968221103561

Dziekonski, K., Popiel, M., Wieczorek, I., Cybulski, P., Smala, K., Ostojewska, J., et al. (2025). The influence of physical activity and carbohydrate management on glycemic control in type 1 diabetes: Exploring the potential of artificial intelligence in diabetes care. Quality in Sport, 41, 60303. https://doi.org/10.12775/QS.2025.41.60303

Gaszynska, M., Drzewiecka, A., Drzewiecki, A., Korczak, W., Madejska, M. A., Rudzka, A., et al. (2025). The role of physical activity in type 1 diabetes: A review of the literature. Journal of Education, Health and Sport, 80, 59766. https://doi.org/10.12775/JEHS.2025.80.59766

Gozhenko, A., Zukow, W., Olenovych, O., & Gozhenko, O. (2026). Renal threshold for glucose: Physiological basis and relationship with water metabolism. A narrative review. Pedagogy and Psychology of Sport, 31, 69828. https://doi.org/10.12775/PPS.2026.31.69828

Helleputte, S., Yardley, J. E., Senior, P. A., et al. (2023). Effects of postprandial exercise on blood glucose levels in adults with type 1 diabetes: A review. Diabetologia, 66, 1179–1191. https://doi.org/10.1007/s00125-023-05910-x

Katsarou, D. N., Georga, E. I., Christou, M. A., Christou, P. A., Tigas, S., Papaloukas, C., & Fotiadis, D. I. (2025). Optimizing hypoglycaemia prediction in type 1 diabetes with ensemble machine learning modeling. BMC Medical Informatics and Decision Making, 25(1), 46. https://doi.org/10.1186/s12911-025-02867-2

Leutheuser, H., Voegeli, J. E., Burckhardt, M. A., Vogt, J. E., et al. (2024). Predicting risk for nocturnal hypoglycemia after physical activity in children with type 1 diabetes. Frontiers in Medicine, 11, 1439218. https://doi.org/10.3389/fmed.2024.1439218

Ma, S., Coopergard, R., Clements, M., & Chow, L. (2025). Managing exercise-related glycemic events in type 1 diabetes: Development and validation of predictive models for a practical decision support tool. JMIR Diabetes, 10, e68948. https://doi.org/10.2196/68948

Maytham, K., Hagelqvist, P. G., Engberg, S., Andersen, A., et al. (2024). Accuracy of continuous glucose monitoring during exercise-related hypoglycemia in individuals with type 1 diabetes. Frontiers in Endocrinology, 15, 1352829. https://doi.org/10.3389/fendo.2024.1352829

McClure, R. D., Alcantara-Cordero, F. J., Weseen, E., & Yardley, J. E. (2023). Systematic review and meta-analysis of blood glucose response to high-intensity interval exercise in adults with type 1 diabetes. Canadian Journal of Diabetes, 47(2), 171–179. https://doi.org/10.1016/j.jcjd.2022.11.006

Moser, O., Zaharieva, D. P., Adolfsson, P., Clements, M. A., Francescato, M. P., Gallen, I. W., et al. (2025). The use of automated insulin delivery around physical activity and exercise in type 1 diabetes: A position statement of the European Association for the Study of Diabetes and the International Society for Pediatric and Adolescent Diabetes. Diabetologia, 68(2), 255–280. https://doi.org/10.1007/s00125-024-06308-z

Mosquera-Lopez, C., Ramsey, K. L., Roquemen-Echeverri, V., & Jacobs, P. G. (2023). Modeling risk of hypoglycemia during and following physical activity in people with type 1 diabetes using explainable mixed-effects machine learning. Computers in Biology and Medicine, 155, 106670. https://doi.org/10.1016/j.compbiomed.2023.106670

Murillo, S., Brugnara, L., Servitja, J. M., et al. (2022). HIIT and hypoglycemia in type 1 diabetes. Diabetes & Metabolism, 48(6), 101361. https://doi.org/10.1016/j.diabet.2022.101361

Parent, C., Lespagnol, E., Berthoin, S., et al. (2023). Barriers to physical activity in children and adults living with type 1 diabetes: A complex link with real-life glycemic excursions. Canadian Journal of Diabetes, 47(2), 124–132. https://doi.org/10.1016/j.jcjd.2022.10.006

Perkins, B. A., Turner, L. V., & Riddell, M. C. (2024). Applying technologies to simplify strategies for exercise in type 1 diabetes. Diabetologia, 67(10), 2045–2058. https://doi.org/10.1007/s00125-024-06229-x

Riddell, M. C., Gal, R. L., Bergford, S., Willi, S. M., DuBose, S. N., Sherr, J. L., et al. (2024). The acute effects of real-world physical activity on glycemia in adolescents with type 1 diabetes: The Type 1 Diabetes Exercise Initiative Pediatric (T1DEXIP) study. Diabetes Care, 47(1), 132–139. https://doi.org/10.2337/dc23-1548

Riddell, M. C., Li, Z., Gal, R. L., Grimes, C., Dimeglio, L. A., Pinsker, J. E., et al. (2023). Examining the acute glycemic effects of different types of structured exercise sessions in type 1 diabetes in a real-world setting: The Type 1 Diabetes and Exercise Initiative (T1DEXI). Diabetes Care, 46(4), 704–713. https://doi.org/10.2337/dc22-1721

Riddell, M. C., & Peters, A. L. (2023). Exercise in adults with type 1 diabetes mellitus. Nature Reviews Endocrinology, 19(2), 98–111. https://doi.org/10.1038/s41574-022-00756-6

Rilstone, S., Oliver, N., Godsland, I. F., et al. (2024). A randomized controlled trial assessing the impact of continuous glucose monitoring with a predictive hypoglycemia alert function on hypoglycemia in physical activity for people with type 1 diabetes (PACE). Diabetes Technology & Therapeutics, 26(2). https://doi.org/10.1089/dia.2023.0376

Russon, C. L., Allen, M. J., Cockcroft, E., et al. (2026). GlucoseGo: A simple tool to predict hypoglycaemia during exercise in type 1 diabetes. Diabetologia. Advance online publication. https://doi.org/10.1007/s00125-026-06692-8

Schubert-Olesen, O., Kroger, J., Siegmund, T., Thurm, U., & Halle, M. (2022). Continuous glucose monitoring and physical activity. International Journal of Environmental Research and Public Health, 19(19), 12296. https://doi.org/10.3390/ijerph191912296

Tsichlaki, S., Koumakis, L., & Tsiknakis, M. (2022). Type 1 diabetes hypoglycemia prediction algorithms: Systematic review. JMIR Diabetes, 7(3), e34699. https://doi.org/10.2196/34699

Tyler, N. S., Mosquera-Lopez, C., Young, G. M., El Youssef, J., Castle, J. R., & Jacobs, P. G. (2022). Quantifying the impact of physical activity on future glucose trends using machine learning. iScience, 25(3), 103888. https://doi.org/10.1016/j.isci.2022.103888

Zhang, L., Yang, L., & Zhou, Z. (2023). Data-based modeling for hypoglycemia prediction: Importance, trends, and implications for clinical practice. Frontiers in Public Health, 11, 1044059. https://doi.org/10.3389/fpubh.2023.1044059