Neural Predictive Controller for Closed-Loop Control of Glucose Using the Subcutaneous Route: A Simulation Study

Abstract The objective of this study was to develop and evaluate a strategy for closed-loop control of glucose using subcutaneous (s.c.) glucose measurement and s.c. infusion of monomeric insulin analogues. The method was based on off-line identification of the glucoregulatory system using neural networks and a nonlinear model predictive controller. Numerical studies on system identification and closed-loop control of glucose were carried out using a comprehensive model of glucose regulation. The proposed control strategy was robust against noise and time delays, and enabled stable control also for slow time variations of the controlled process. In conclusion, closed-loop control of glucose is feasible using the s.c. route and a neural predictive controller.