Fuzzy-based controller for glucose regulation in type-1 diabetic patients by subcutaneous route

Area of Science:

• Biomedical Engineering
• Control Systems Engineering
• Endocrinology

Background:

• Type-1 diabetes mellitus (TIDM) requires intensive insulin therapy for blood glucose level (BGL) regulation.
• Managing BGL variability in TIDM patients on multiple daily injections (MDIR) presents significant challenges.
• Existing control strategies often struggle to adapt to inter-patient physiological differences.

Purpose of the Study:

• To develop and evaluate an advisory/control algorithm for TIDM patients on MDIR.
• To enhance BGL regulation using expert knowledge embedded in fuzzy logic controllers.
• To address glucose-insulin dynamics variability through a robust two-loop feedback strategy.

Main Methods:

• Implementation of Mamdani-type fuzzy logic controllers for BGL regulation.
• Design of a two-loop feedback control architecture: an inner-loop for insulin dosing and an outer-loop for supervisory adjustment.
• Utilizing a detailed compartmental model of insulin-glucose dynamics for extensive closed-loop simulations.

Main Results:

• The proposed algorithm effectively regulates BGL in simulated TIDM patients.
• The two-loop strategy successfully compensates for patient-specific glucose-insulin dynamics variability.
• Fuzzy logic controllers demonstrated proficiency in managing rapid/short acting insulin (RSAI) and intermediate/long acting insulin (ILAI) delivery.

Conclusions:

• The developed fuzzy logic-based control algorithm offers a promising approach for TIDM management under MDIR.
• The supervisory outer-loop enhances the robustness and adaptability of the insulin delivery system.
• Simulation results support the clinical potential of this advanced BGL control strategy.