Observer-based control for plasma glucose regulation in type 1 diabetes mellitus patients with unknown input delay

Area of Science:

• Biomedical Engineering
• Control Systems Theory
• Endocrinology

Background:

• Type 1 diabetes mellitus (T1DM) management requires precise glucose regulation.
• Artificial pancreas systems face challenges with unpredictable factors like time delays and meal disturbances.
• Existing control strategies may not fully address realistic physiological variations.

Purpose of the Study:

• To develop an observer-based control strategy for plasma glucose regulation in T1DM patients.
• To enhance artificial pancreas performance by accounting for unknown time-varying delays and meal disturbances.
• To design a control system that mimics the natural pancreas's dynamic response.

Main Methods:

• Utilized an extended Bergman minimal model, a nonlinear glucose-insulin model.
• Designed an observer-based state feedback controller to handle unknown long delays.
• Employed linear matrix inequalities (LMIs) based on Lyapunov conditions for gain computation and stability assurance.

Main Results:

• The proposed control strategy effectively regulates plasma glucose levels.
• The system demonstrated robustness against unknown meal disturbances and time-varying delays.
• Numerical simulations confirmed the avoidance of hypoglycemic episodes and achievement of stable glycemic control.

Conclusions:

• The observer-based control strategy offers a promising approach for advanced artificial pancreas systems.
• This method enhances glucose regulation in T1DM by addressing realistic system uncertainties.
• The designed controller provides a soft, natural dynamic response, improving patient outcomes and safety.