Page 80 - Control Theory in Biomedical Engineering

Adaptive control of artificial pancreas systems for treatment of type 1 diabetes 67

Fig. 2 A summary of the developed techniques for the PMM-AP system.

time. The designed PIC estimator is individualized by appropriately initial-
izing some of the parameters in the insulin compartmental model using the
partial least squares regression. The demographic information of individuals,
such as weight, body mass index, and duration with diabetes, is used to pre-
dict the initial values for these parameters (Hajizadeh et al., 2017a, 2018c).
After initialization, the time-varying parameters are estimated online. The
proposed PIC estimator is able to capture the intersubject and intrasubject
variabilities to provide accurate PIC estimates. Furthermore, the gut