Risk of having hypoglycemia is one of the biggest barriers preventing Type 1 Diabetes (T1D) patients from performing exercise. In addition, management of diet and exercise levels needs to be personalized for each patient to avoid hypoglycemia and to achieve a good glycemic control. In this paper, we developed a model-based diet and exercise recommender system that could be used to provide an (optimal) personalized intervention on diet and exercise for T1D patients. The recommender system makes prediction of blood glucose at each intervention time based on a patient-specific model of glucose dynamics, and then provides the optimal meal sizes, target heart rates during exercise, pre-exercise carbohydrate and bedtime snack by minimizing a clinical risk function under constraints. Patient-specific models of glucose dynamics were identified for 30 virtual subjects generated from a modified UVa/Padova simulator with an added exercise-glucose subsystem. The performance of the recommender system was then compared to two self-management schemes (the Starter and the Skilled). The latter represents an off-line optimal solution providing a lower bound on the risk index. The average clinical risk under the recommender system was reduced by 49% compared to that under the Starter, and it was comparable to the risk of the Skilled. In addition, the recommender system had less number of post-exercise/nocturnal hypoglycemia events occurred than that under the Starter or the Skilled. Such recommender system can be implemented as an “App” patient advisor to improve T1D patients’ self-management of glucose control.
- Dynamic Systems and Control Division
A Personalized Diet and Exercise Recommender System in Minimizing Clinical Risk for Type 1 Diabetes: An In Silico Study
Xie, J, & Wang, Q. "A Personalized Diet and Exercise Recommender System in Minimizing Clinical Risk for Type 1 Diabetes: An In Silico Study." Proceedings of the ASME 2017 Dynamic Systems and Control Conference. Volume 1: Aerospace Applications; Advances in Control Design Methods; Bio Engineering Applications; Advances in Non-Linear Control; Adaptive and Intelligent Systems Control; Advances in Wind Energy Systems; Advances in Robotics; Assistive and Rehabilitation Robotics; Biomedical and Neural Systems Modeling, Diagnostics, and Control; Bio-Mechatronics and Physical Human Robot; Advanced Driver Assistance Systems and Autonomous Vehicles; Automotive Systems. Tysons, Virginia, USA. October 11–13, 2017. V001T08A003. ASME. https://doi.org/10.1115/DSCC2017-5136
Download citation file: