Control-Oriented Model Including Hyperinsulinemia Induced Insulin Resistance in Type 1 Diabetes

Abstract

Insulin Resistance (IR) is an important risk factor for health complications in subjects with Type 1 Diabetes Mellitus (TIDM) that is induced by Hyperglycemia (HG) and Hyperinsulinemia (HI) conditions, among other triggers. Considering the relevance of impaired insulin action in closed-loop glycemic control in TIDM, in this work, a previous controloriented Linear Parameter-Varying (LPV) model is augmented by adding the estimate of the patient’s Insulin On Board (IOB) as a model input. The model is obtained based on the distribution version of the UVA/Padova metabolic simulator, and allows to include the effect of active insulin on the patient’s insulin sensitivity $(\mathrm {S}_{\mathrm {I}})$ and cope with inter-patient variability by using a priori patient information. The efficacy of this model is evaluated in comparison with the previous control-oriented LPV model which in turn is an improvement of previous representations, in terms of their open- and closed-loop differences with respect to the UVA/Padova model. In silico results evidence the importance of including the effects of insulin stacking at the controller-design stage.