Real-Time Estimation of Plasma Insulin Concentration in Patients with Type 1 Diabetes
Iman Hajizadeh, MSc1; Kamuran Turksoy, PhD2; Sediqeh Samadi, MSc1; Jianyuhan Feng, MSc1; Ali Cinar, PhD1,2
1Department of Chemical and Biological Engineering, Illinois Institute of Technology
2Department of Biomedical Engineering, Illinois Institute of Technology
Closed-loop control of blood glucose concentration, known as artificial pancreas (AP) system, is an alternative treatment for patients with type 1 diabetes, by automatically controlling their blood glucose levels –. These control systems use continuous glucose measurements to calculate optimum amount of insulin to be infused with an insulin pump in order to keep plasma glucose concentration within a safe target range to avoid health problems. Real-time estimations for plasma insulin concentration is beneficial for increasing the efficiency of AP control algorithms. This enables calculation of more realistic insulin infusion rates and prevention of hypoglycemia that would be caused by over dosing of insulin.
Our objective is to fulfill a real-time estimation of plasma insulin concentration from continuous glucose monitoring data by using a mathematical model. Hovorka’s model which has been developed to describe glucose-insulin dynamic in different parts of the human body, has been incorporated into a continuous-discrete extended Kalman filter (CDEKF) to provide an estimate of the plasma insulin concentration. The CDEKF is the generalization of the Kalman filter to nonlinear systems with continuous-time state equations and discrete-time measurements. Furthermore, because of variability in system dynamics, uncertain parameters have been considered as new states in Hovorka’s model to be estimated by CDEKF. This methodology has been evaluated by using clinical data from patients with Type 1 diabetes who underwent a closed-loop AP experiment. Real-time insulin estimations have been compared to plasma insulin measurements to evaluate performance and validity of the proposed methodology.
Finally, based on simulation results, it has been shown that the proposed method is able to estimate the plasma insulin concentration in real time. Furthermore, the results for estimated continuous glucose measurements and blood glucose concentration have been compared to their real values to acknowledge performance of the proposed method. This method will be beneficial for an AP system in terms of real time estimation of non-measurable variables such as plasma glucose and insulin concentrations.
 K. Turksoy, L. T. Quinn, E. Littlejohn, and A. Cinar, “An Integrated Multivariable Artificial Pancreas Control System,” J. Diabetes Sci. Technol., vol. 8, no. 3, pp. 498–507, 2014.
 K. Turksoy, L. Quinn, E. Littlejohn, and A. Cinar, “Multivariable Adaptive Identification and Control for Artificial Pancreas Systems,” IEEE Trans. Biomed. Eng., vol. 61, no. 3, pp. 883–891, 2014.
See more of this Group/Topical: Computing and Systems Technology Division