A nonlinear model predictive controller has been developed to maintain normoglycemia in subjects with type 1 diabetes during fasting conditions such as during overnight fast. The controller employs a compartment model, which represents the glucoregulatory system and includes submodels representing absorption of subcutaneously administered short-acting insulin Lispro and gut absorption. The controller uses Bayesian parameter estimation to determine time-varying model parameters. Moving target trajectory facilitates slow, controlled normalization of elevated glucose levels and faster normalization of low glucose values. The predictive capabilities of the model have been evaluated using data from 15 clinical experiments in subjects with type 1 diabetes. The experiments employed intravenous glucose sampling (every 15 min) and subcutaneous infusion of insulin Lispro by insulin pump (modified also every 15 min). The model gave glucose predictions with a mean square error proportionally related to the prediction horizon with the value of 0.2 mmol L(-1) per 15 min. The assessment of clinical utility of model-based glucose predictions using Clarke error grid analysis gave 95% of values in zone A and the remaining 5% of values in zone B for glucose predictions up to 60 min (n = 1674). In conclusion, adaptive nonlinear model predictive control is promising for the control of glucose concentration during fasting conditions in subjects with type 1 diabetes.

No abstract

The objective of the project Advanced Insulin Infusion using a Control Loop (ADICOL) was to develop a treatment system that continuously measures and controls the glucose concentration in subjects with type 1 diabetes. The modular concept of the ADICOL's extracorporeal artificial pancreas consisted of a minimally invasive subcutaneous glucose system, a handheld PocketPC computer, and an insulin pump (D-Tron, Disetronic, Burgdorf, Switzerland) delivering subcutaneously insulin lispro. The present paper describes a subset of ADICOL activities focusing on the development of a glucose controller for semi-closed-loop control, an in silico testing environment, clinical testing, and system integration. An incremental approach was adopted to evaluate experimentally a model predictive glucose controller. A feasibility study was followed by efficacy studies of increasing complexity. The ADICOL project demonstrated feasibility of a semi-closed-loop glucose control during fasting and fed conditions with a wearable, modular extracorporeal artificial pancreas.

The aim of the study was to realize a mathematical model of insulin-glucose relationship in type I diabetes and test its effectiveness for the design of control algorithms in external artificial pancreas. A new mathematical model, divided into glucose and insulin sub-models, was developed from the so-called "minimal model". The key feature is the representation of insulin sensitivity so as to permit the personalisation of the parameters. Real-time applications are based on an insulin standardised model. Clinical data were used to estimate model parameters. Root mean square error between simulated and real blood glucose profiles (G(rms)) was used to evaluate system efficacy. Results from parameter estimation and insulin standardisation showed a good capability of the model to identify individual characteristics. Simulation results with a G(rms) 1.30 mmol/l in the worst case testified the capacity of the model to accurately represent glucose-insulin relationship in type 1 diabetes allowing self tuning in real time.

No abstract