Development of adaptive artificial pancreas systems for personalized glucose management Public Deposited

Type 1 diabetes (T1D) is a chronic disease where the pancreas does not produce sufficient insulin. Exogenous insulin is required to be delivered. Over delivery of insulin can cause hypoglycemia, which can cause coma or death if not treated. Under delivery can cause hyperglycemia; long-term exposure can cause chronic health problems including neuropathy, retinopathy, and limb loss. The artificial pancreas (AP) is an automated technology for helping people with T1D control their glucose. A single-hormone (SH) AP consists of three main components: a glucose sensor that is inserted in the body, a control algorithm for calculating insulin infusion based on the glucose levels, and an insulin pump. A dual-hormone (DH) AP system includes a glucagon pump; glucagon stimulates endogenous glucose production. Exercise is challenging for people with T1D as it can cause hypoglycemia. Glucagon can help avoid exercise-induced hypoglycemia. In AP systems, the control algorithm is a critical component that affects how well glucose is managed including avoidance of hypoglycemia and hyperglycemia. Model predictive control (MPC) is a state-of-the-art method for controlling glucose levels in people with T1D. There is currently not an adaptive MPC for single and dual-hormone AP control systems that can effectively handle exercise.

  • resalat.navid.2019.pdf
Publication Date
  • 2019
Document type