Dynamic Insulin Basal Needs Estimation and Parameters Adjustment in Type 1 Diabetes

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Medicine
• Endocrinology

Background:

• Current closed-loop systems for diabetes management often rely on fixed basal insulin profiles, requiring precise manual adjustments.
• Adapting to daily physiological changes or system malfunctions (e.g., infusion site issues) remains a challenge for existing algorithms.
• Accurate basal insulin delivery is critical for effective glycemic control in type 1 diabetes management.

Purpose of the Study:

• To develop and evaluate a novel closed-loop control algorithm that dynamically determines and adjusts basal insulin needs.
• To enhance the adaptability and safety of artificial pancreas systems by incorporating real-time physiological data.
• To reduce the burden on patients by minimizing the need for manual basal profile adjustments.

Main Methods:

• A novel algorithm was developed to dynamically determine basal insulin needs using linear regression on historical insulin dosing and glycemic data.
• The algorithm integrates dynamic adjustments of insulin sensitivity factor (ISF) and glycemic targets.
• Simulations were conducted on 30 virtual patients (adults, adolescents, children) using the FDA-approved UVa/Padova Simulator in Python.

Main Results:

• The proposed system successfully estimated patients' dynamic basal insulin needs.
• The algorithm demonstrated adaptability to simulated partial insulin infusion site blockages, maintaining comparable glycemic control (time in range) to unobstructed scenarios.
• Dynamic adjustments in basal needs, ISF, and glycemic targets improved overall algorithm safety.

Conclusions:

• Dynamic basal insulin needs determination is feasible and beneficial for closed-loop diabetes management systems.
• The proposed algorithm can enhance existing artificial pancreas technologies by providing real-time basal need estimation and adaptation.
• This approach offers a potential monitoring channel to detect deviations in basal insulin requirements due to illness, exercise, or technical issues.