A novel adaptive run-to-run control strategy for insulin pump therapy optimization

Good glycemic control is very important for individuals with type 1 diabetes mellitus. They need adjustment of insulin dose time after time to maintain blood glucose in a safe range, so the regulation procedure is very troublesome and it can put extra pressure on patients. Therefore, an automatic methodology is needed to complete this procedure. In this study, a novel run-to-run control strategy with adaptive learning gain was proposed to deal with this issue. In this algorithm, it updates the basal insulin and bolus insulin simultaneously, and also can get a suitable gain for different subjects, Furthermore, the proposed algorithm shows great robustness with respect to variations on meal time and meal amount.