System identification of a nonlinear glucose insulin dynamic model

In this paper a modified glucose insulin dynamic model is proposed based on Bergman's minimal model. A glucose and insulin dependent net hepatic glucose balance term as used in AIDA is incorporated into Bergman's minimal model. The adopted representation is nonlinear and can be given in a closed-form approximation. An extended nonlinear least-squares algorithm is used for the parameter estimation of the proposed nonlinear model using experimental data sets. The identifiability of the proposed model and effectiveness of the proposed method are demonstrated by computer simulations. The results are validated by the experimental data.