AID Control Parameter Adaptation for Persistent Glucose Excursions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Automatic Insulin Delivery (AID) systems respond conservatively to persistent low-level blood glucose excursions due to fixed weight coefficients, failing to adequately adjust insulin delivery to manage these excursions effectively, which can have negative health consequences.
Innovation Solution
Adaptive adjustment of the cost function parameters in AID systems based on current and projected blood glucose level performance, modifying weight coefficients to enhance responsiveness to persistent low-level excursions, and incorporating a parallel integral control approach to request additional insulin as needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed weight coefficients Q and R are used in the cost function, then the system is simple and stable, but the system responds conservatively to persistent low-level blood glucose excursions and cannot adapt to changing user needs
Solution Approach 1:
The patent applies dynamics by transitioning from fixed weight coefficients to adaptive weight coefficients that change over time based on user performance. The controller dynamically adjusts the weight coefficients Q and R in the cost function based on whether the user has been achieving target blood glucose levels, allowing the system to become more aggressive or conservative as needed. This dynamic adjustment resolves the contradiction by enabling adaptability while maintaining a relatively simple base structure.
Solution Approach 2:
The patent implements feedback by continuously monitoring user blood glucose control performance and using this information to adjust the weight coefficients in the cost function. The controller receives feedback about whether the user is achieving target glucose levels and modifies the weighting between glucose cost and insulin cost accordingly. This feedback mechanism enables the system to adapt to changing needs while maintaining stability through structured adjustment protocols.
2Reliability
If the system responds aggressively to blood glucose excursions, then blood glucose control improves, but the risk of hypoglycemia increases
Solution Approach 1:
The patent applies parameter changes by modifying the weight coefficients Q and R in the cost function based on user performance. When users are achieving target glucose levels reliably, the system increases the weight on glucose cost (making Q larger relative to R), which makes the system more aggressive in correcting excursions. Conversely, when hypoglycemia risk is detected, the system decreases the aggression parameter. This dynamic parameter adjustment allows the system to optimize between control effectiveness and hypoglycemia risk reduction.
Solution Approach 2:
The system applies self-service by automatically adjusting its own aggressiveness parameter based on observed user performance without requiring manual reconfiguration. The controller monitors whether users are achieving target glucose levels and self-modifies the weight coefficients accordingly. This self-adjusting mechanism enables the system to respond appropriately to changing conditions while maintaining safety, resolving the contradiction between aggressive control and hypoglycemia prevention.
3Object-affected harmful factors
If the system uses conservative response to blood glucose excursions, then the risk of hypoglycemia is reduced, but persistent low-level excursions are not adequately addressed
Solution Approach 1:
The patent applies dynamics by enabling the system's responsiveness to change based on user performance. When users consistently achieve target glucose levels, the system dynamically increases its aggressiveness parameter, making it more effective at managing excursions. When performance is inconsistent, the system maintains conservatism to prevent hypoglycemia. This dynamic behavior allows the system to optimize productivity while maintaining safety through conditional responsiveness.
Solution Approach 2:
The system implements feedback by continuously monitoring user blood glucose control and using this information to adjust the weight coefficients in the cost function. The controller receives feedback about glucose control effectiveness and modifies the balance between glucose cost and insulin cost weighting accordingly. This feedback mechanism enables the system to increase its management effectiveness for excursions when safe to do so, while maintaining conservatism when hypoglycemia risk is present.
Data Source
AI summary
Exemplary embodiments may modify the cost function parameters based on current and projected mean outcomes in blood glucose level control performance. The exemplary embodiments may modify the weight coefficient R for the insulin cost so that the value of R is not fixed and is not based solely on clinical determined values. Exemplary embodiments may also adjust the cost function to address persistent low-level blood glucose level excursions for users. The exemplary embodiments may reduce the penalty of the insulin cost by the sum of the converted insulin cost of the glucose excursions above target for a period divided by a number of cycles of average insulin action time. The AID system reduces the insulin cost by the lack of insulin in previous cycles.