Article ID Journal Published Year Pages File Type
718241 IFAC Proceedings Volumes 2012 6 Pages PDF
Abstract

Identification of individualized models in type 1 diabetes is of vital importance for the development of a successful artificial pancreas and other model-based strategies. The availability of reliable models has been proven very difficult in the past, most probably due to the large intra-patient variability.In this work, a new approach to cope with intra-patient variability in the identification of postprandial models is presented. Uncertainties due to physiological variability and input errors in insulin infusion rate and meal estimation are characterized by means of interval models, which predict a glucose envelope containing all possible patient responses. Multi-objective optimization is performed over a cohort of virtual patients, minimizing both the fitting error and the output glucose envelope width. A Pareto Front is then built ranging from classical identification representing average behaviors to interval identification guaranteeing full enclosure of the measurements. A method for the selection of the best individual in the Pareto Front for identification from home monitoring data with a continuous glucose monitor is presented, reducing the overestimation of patient's variability due to monitor inaccuracies and noise. The proposed method shows good prediction capabilities in a five-hour time horizon as compared to gold standard measurements demonstrating its feasibility.

Related Topics
Physical Sciences and Engineering Engineering Computational Mechanics