AI-Driven Blood Glucose Forecasting in Real-World Diabetes Care: Evaluating Wearable-Based Predictive Models
DOI:
https://doi.org/10.63282/3050-922X.IJERET-V7I2P122Keywords:
Blood Glucose Prediction, Wearable Sensors, Continuous Glucose Monitoring (CGM), Recurrent Neural Networks, Diabetes Management, Machine Learning, Personalized HealthcareAbstract
Wearable technology has rapidly advanced the landscape of diabetes self-management by enabling continuous, real-time monitoring of glucose levels in daily life. This paper leverages data collected exclusively through wearable continuous glucose monitors to develop and compare advanced forecasting models for short-term blood glucose prediction in individuals with diabetes. By implementing and benchmarking artificial intelligence-based recurrent neural networks against traditional statistical approaches, the research investigates both generalized and patient-tailored prediction strategies. Outcomes highlight the effectiveness of population-level machine learning models for predicting glucose fluctuations, even with limited user history, and discuss the implications for proactive and personalized intervention. The findings underscore the growing potential of AI-enhanced wearables to deliver actionable insights, optimize insulin dosing, and mitigate acute events, signaling a new era of digitally-empowered diabetes management through wearable technologies.
References
[1] E. I. Georga, V. C. Protopappas, D. Ardigo, M. Marina, I. Zavaroni, D. Polyzos, and D. I. Fotiadis, “Multivariate prediction of subcutaneous glucose concentration in type 1 diabetes patients based on support vector regression,” IEEE Journal of Biomedical and Health Informatics, vol. 17, no. 1, pp. 71–81, 2012. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Multivariate prediction of subcutaneous glucose concentration in type 1 diabetes patients based on support vector regression&btnG= | https://doi.org/10.1109/TITB.2012.2219876
[2] T. Zhu, K. Li, P. Herrero, and P. Georgiou, “Basal glucose control in type 1 diabetes using deep reinforcement learning: An in silico validation,” IEEE Journal of Biomedical and Health Informatics, vol. 25, no. 4, pp. 1223–1232, 2020. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Basal glucose control in type 1 diabetes using deep reinforcement learning: An in silico validation&btnG= | https://doi.org/10.1109/JBHI.2020.3014556
[3] T. W. Bickmore, H. Trinh, S. Olafsson, T. K. O’Leary, R. Asadi, N. M. Rickles, and R. Cruz, “Patient and consumer safety risks when using conversational assistants for medical information: An observational study of Siri, Alexa, and Google Assistant,” Journal of Medical Internet Research, vol. 20, no. 9, p. 11510, 2018. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Patient and consumer safety risks when using conversational assistants for medical information: An observational study of Siri, Alexa, and Google Assistant&btnG= | https://doi.org/10.2196/11510
[4] R. N. Bergman, L. S. Phillips, C. Cobelli, et al., “Physiologic evaluation of factors controlling glucose tolerance in man: Measurement of insulin sensitivity and betacell glucose sensitivity from the response to intravenous glucose,” The Journal of Clinical Investigation, vol. 68, no. 6, pp. 1456–1467, 1981. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Physiologic evaluation of factors controlling glucose tolerance in man: Measurement of insulin sensitivity and betacell glucose sensitivity from the response to intravenous glucose&btnG=| https://dm5migu4zj3pb.cloudfront.net/manuscripts/110000/110398/JCI81110398.pdf
[5] M. A. Makroum, M. Adda, H. Bouzouane, and A. Ibrahim, “Machine learning and smart devices for diabetes management: Systematic review,” Sensors, vol. 22, no. 5, p. 1843, 2022. |https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Machine learning and smart devices for diabetes management: Systematic review&btnG= | https://doi.org/10.3390/s22051843
[6] K. Saiti, M. Macaš, L. Lhotská, K. Štechová, and P. Pithová, “Ensemble methods in combination with compartment models for blood glucose level prediction in type 1 diabetes mellitus,” Computer Methods and Programs in Biomedicine, vol. 196, p. 105628, 2020. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Ensemble methods in combination with compartment models for blood glucose level prediction in type 1 diabetes mellitus&btnG= | https://doi.org/10.1016/j.cmpb.2020.105628
[7] M. F. Rabby, Y. Tu, M. I. Hossen, I. Lee, A. S. Maida, and X. Hei, “Stacked LSTM based deep recurrent neural network with Kalman smoothing for blood glucose prediction,” BMC Medical Informatics and Decision Making, vol. 21, no. 1, p. 101, 2021. | https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Stacked LSTM based deep recurrent neural network with Kalman smoothing for blood glucose prediction&btnG= | https://doi.org/10.1186/s12911-021-01462-5
[8] C. Mosquera-Lopez, R. Dodier, N. Tyler, N. Resalat, and P. Jacobs, “Leveraging a big dataset to develop a recurrent neural network to predict adverse glycemic events in type 1 diabetes,” IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 9, pp. 2527–2536, 2020. https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Leveraging a big dataset to develop a recurrent neural network to predict adverse glycemic events in type 1 diabetes&btnG= | https://doi.org/10.1109/JBHI.2019.2911701
[9] E. Acuna, R. Aparicio, and V. Palomino, “Analyzing the performance of transformers for the prediction of the blood glucose level considering imputation and smoothing,” Big Data and Cognitive Computing, vol. 7, no. 1, p. 41, 2023. https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Analyzing the performance of transformers for the prediction of the blood glucose level considering imputation and smoothing&btnG= | https://doi.org/10.3390/bdcc7010041
[10] A. Ahmed, S. Aziz, A. Abd-alrazaq, F. Farooq, and J. Sheikh, “Overview of artificial intelligence-driven wearable devices for diabetes: Scoping review,” Journal of Medical Internet Research, vol. 24, no. 8, p. e36010, 2022. https://scholar.google.com/scholar?hl=en&as_sdt=0,44&q=Overview of artificial intelligence-driven wearable devices for diabetes: Scoping review&btnG= | https://doi.org/10.2196/36010
