Adaptive Model Predictive Control for Physiological Glucose
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing artificial pancreas systems lack control methods that can effectively handle both abrupt and slow variations in glucose dynamics, often relying on limited or fixed models that fail to provide adequate glycemic control in dynamic conditions.
Innovation Solution
A closed-loop system employing a model predictive controller with adaptable model parameters that adjust based on changes in glycemic patterns and medication type, incorporating multiple state vectors and models to enhance responsiveness to varying insulin pharmacokinetics and patient conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed control algorithm is used in artificial pancreas systems, then the system structure is simple, but the system cannot adequately handle abrupt and slow variations in glucose dynamics
Solution Approach 1:
The control algorithm transitions from a fixed structure to a dynamic adaptive structure that can adjust its parameters in real-time. The system dynamically adapts to both abrupt and slow variations in glucose dynamics by continuously updating its control strategy based on current physiological conditions, thereby resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The system changes its control parameters dynamically based on detected glycemic patterns. By adjusting parameters such as insulin sensitivity factors, carbohydrate ratios, and control horizons according to the current glucose dynamics characteristics, the system achieves high adaptability without requiring an overly complex structural framework.
2Reliability
If a single fixed model is used for insulin pharmacokinetics, then the model is simple, but it fails to provide adequate control when pharmacokinetic profiles change
Solution Approach 1:
The system employs a universal control framework that can handle multiple insulin types and pharmacokinetic profiles through a single adaptive model structure. This multi-functional approach allows the system to reliably control glycemia across different conditions without requiring separate dedicated models for each scenario, thus balancing reliability with manageable complexity.
Solution Approach 2:
The system performs preliminary identification and characterization of the patient's insulin pharmacokinetic profile during initial treatment phases. By establishing baseline parameters in advance and preparing adaptive mechanisms beforehand, the system ensures reliable control when pharmacokinetic profiles change, avoiding the need for complex real-time model switching.
3Ease of operation
If conventional insulin therapy with repeated monitoring is used, then the system is simple, but it is burdensome to the patient requiring vigilant control
Solution Approach 1:
The artificial pancreas system performs self-service by automatically monitoring glucose levels through continuous sensors and autonomously adjusting insulin delivery based on detected glycemic patterns. This eliminates the need for patients to perform repeated finger sticks and manually calculate insulin doses, significantly reducing patient burden while implementing high-level automation for safety-critical functions.
Solution Approach 2:
The system implements continuous feedback loops where glucose sensor data is constantly monitored and fed back to the control algorithm, which automatically adjusts insulin delivery in real-time. This automated feedback mechanism replaces manual monitoring and control actions, reducing patient burden while maintaining appropriate automation levels for clinical safety.
Data Source
AI summary
The present disclosure relates to systems and methods for controlling physiological glucose concentrations in a patient.