Keywords
Diabetes Mellitus
Sorensen
How to Cite
Métricas de PLUMX
Abstract
The objective of this research is to develop a platform that controls the insulin dose required to maintain glucose levels in patients with type 1 diabetes mellitus. The study is based on the implementation of the Sorensen mathematical model and a Reinforcement Learning-Based Controller (RLBC) named Deep Q-Learning on a Raspberry Pi 2B development board, with real-time execution. The obtained results are quantitatively compared with a µ-synthesis-based robust controller, concluding that the RLBC exhibits superior performance. The main implication of this research is the potential to improve glucose control in type 1 diabetes mellitus patients through an automated and low-cost system. The original contribution of this study lies in the integration of a complex mathematical model with accessible hardware for diabetes management. In conclusion, this work provides a foundation for the development of more efficient and accessible insulin control devices, although further research is required to validate its effectiveness in clinical settings.
References
Dénes-Fazakas, L., Siket, M., Kertész, G., Szilágyi, L., Kovács, L., & Eigner, G. (2022). Control of type 1 diabetes mellitus using direct reinforcement learning based controller. In 2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 1512-1517. IEEE. http://dx.doi.org/10.1109/SMC53654.2022.9945084
Fox, I., & Wiens, J. (2019). Reinforcement learning for blood glucose control: Challenges and opportunities. Reinforcement Learning for Real Life (RL4RealLife) Workshop in the 36th International Conference on Machine Learning. https://openreview.net/forum?id=ByexVzSAs4
Katsarou, A., Gudbjörnsdottir, S., Rawshani, A., Dabelea, D., Bonifacio, E., Anderson, B. J., Jacobsen, L. M., Schatz, D. A., & Lernmark, Å. (2017). Type 1 diabetes mellitus. Nature Reviews Disease Primers, 3, 1-17. http://dx.doi.org/10.1038/nrdp.2017.16
Kovács, L., Benyó, B., Bokor, J., & Benyó, Z. (2011). Induced L2-norm minimization of glucose–insulin system for Type I diabetic patients. Computer Methods and Programs in Biomedicine, 102(2), 105-118. http://dx.doi.org/10.3182/20090812-3-DK-2006.0068
Magliano, D. J., & Boyko, E. J. (2022). IDF diabetes atlas. Federación Internacional de Diabetes. https://www.diabetesatlas.org
Marchetti, A., Sasso, D., D’Antoni, F., Morandin, F., Parton, M., Matarrese, M. A. G., & Merone, M. (2025). Deep reinforcement learning for Type 1 Diabetes: Dual PPO controller for personalized insulin management. Computers in Biology and Medicine, 191, 110-147. https://doi.org/10.1016/j.compbiomed.2025.110147
Navarrete, G. A., & Vidrios-Serrano, C., & Maldonado-Fregoso, B., & García, R. J. A., & Chocoteco, J. A. (2024). Implementación del modelo y control en tiempo discreto para diabetes mellitus tipo 1 en una plataforma embebida de bajo costo. In Ingeniantes, Instituto Tecnológico Superior de Misantla. 3(1), 97-102. https://citt.itsm.edu.mx/ingeniantes/articulos/ingeniantes11no1vol3/13.pdf
Ngo, P. D., Wei, S., Holubová, A., Muzik, J., & Godtliebsen, F. (2018). Control of Blood Glucose for Type‐1 Diabetes by Using Reinforcement Learning with Feedforward Algorithm. Computational and mathematical methods in medicine, 2018, 4091497. https://doi.org/10.1155/2018/4091497
Parker, R. S., Doyle III, F. J., Ward, J. H., & Peppas, N. A. (2000). Robust H∞ glucose control in diabetes using a physiological model. AIChE Journal, 46(12), 2537-2549. https://aiche.onlinelibrary.wiley.com/doi/abs/10.1002/aic.690461220
Ramprasad, Y., Rangaiah, G. P., & Lakshminarayanan, S. (2004). Robust PID controller for blood glucose regulation in type I diabetics. Industrial & engineering chemistry research, 43(26), 8257-8268. http://dx.doi.org/10.1021/ie049546a
Ruiz-Velázquez, E., García-Rodríguez, J., Quiroz, G., & Femat, R. (2020). Robust μ-synthesis: Towards a unified glucose control in adults, adolescents and children with T1DM. Journal of the Franklin Institute, 357(14), 9633-9653. http://dx.doi.org/10.1016/j.jfranklin.2020.07.030
Sánchez-Peña, R., Colmegna, P., Grosembacher, L., Breton, M., De Battista, H., Garelli, F., & Cherñavvsky, D. (2017). Artificial pancreas: first clinical trials in Argentina. IFAC-PapersOnLine, 50(1), 7731-7736. https://doi.org/10.1016/j.ifacol.2017.08.1151
Sorensen, J. T. (1985). A physiologic model of glucose metabolism in man and its use to design and assess improved insulin therapies for diabetes (Doctoral dissertation, Massachusetts Institute of Technology). MIT Libraries. https://dspace.mit.edu/bitstream/handle/1721.1/15234/13180776-MIT.pdf?sequence=2
Syed, F. Z. (2022). Type 1 diabetes mellitus. Annals of internal medicine, 175(3), ITC33-ITC48. http://dx.doi.org/10.7326/AITC202203150
Tanzanakis, A., & Lygeros, J. (2023). Data-Driven Robust Reinforcement Learning Control of Uncertain Nonlinear Systems: Towards a Fully-Automated, Insulin-Based Artificial Pancreas. arXiv preprint arXiv:2312.04503. https://doi.org/10.48550/arXiv.2312.04503
Van Hasselt, H., Guez, A., & Silver, D. (2016). Deep reinforcement learning with double q-learning. In Proceedings of the AAAI conference on artificial intelligence , 30(1). https://doi.org/10.1609/aaai.v30i1.10295
Viroonluecha, P., Egea-Lopez, E., & Santa, J. (2022). Evaluation of blood glucose level control in type 1 diabetic patients using deep reinforcement learning. Plos one, 17(9), e0274608. http://dx.doi.org/10.1371/journal.pone.0274608
Wang, Z., Xie, Z., Tu, E., Zhong, A., Liu, Y., Ding, J., & Yang, J. (2021). Reinforcement learning-based insulin injection time and dosages optimization. In 2021 International Joint Conference on Neural Networks (IJCNN), 1-8. IEEE. http://dx.doi.org/10.1109/IJCNN52387.2021.9533957
Watkins, C. J., & Dayan, P. (1992). Q-learning. Machine learning, 8, 279-292. https://doi.org/10.1007/BF00992698
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2026 Revista Bio Ciencias