Chance-Constrained Model Predictive Control for Blood Glucose Management in Diabetes

Patients with type 1 diabetes mellitus (T1DM) need to supply their body with insulin from external sources in order to manage their blood glucose (BG) concentration and mitigate the long–term effects of a chronically increased BG level without falling into a potentially life-threatening hypoglycemia. Doing so is challenging and a heavy burden for those patients, which led to efforts of automating (parts of)this task, most notably in Artificial Pancreas (AP) systems.In standard AP approaches a (typically constant) reference BG value is tried to be tracked as closely as possible and often leads to satisfactory results in terms of BG management.However, requiring a constant BG can be an excessive requirement. Differently from that, this paper proposes a different framework, in which the unavoidable uncertainty is modeled in probabilistic terms and the control goal is defined not in terms of proximity to a specific BG target but as keeping the risk of leaving the euglycemic range under a given threshold. The degree of freedom gained by this problem relaxation can be used for other purposes, e.g. the minimization of total insulin intake. In the current paper an AP controller based on chanceconstrained Model Predictive Control (MPC) is proposed for this purpose.