Enhancing parameter precision and the minimal modeling approach in type I diabetes

An evaluation of a simple model including external perturbations was evaluated for its usefulness in predicting diabetic patients´ behavior. The model proposed has been derived from Cobelli and Marl´s comprehensive model and is structurally identifiable. The optimization was carried out on data gathered using CGMS (Medtronic MiniMed) on 3 subjects. The model was also optimized after performing a model-based signal enhancement. The results obtained before and after signal enhancement showed a promising reduction in the variation coefficient of the estimated parameters. This reduction is expected to be useful in the design of a closed loop controller for subcutaneous insulin delivery.