Automated Blood Glucose Control with Physiological Model Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current artificial pancreas systems using MPC type regulation struggle to accurately predict blood sugar levels, leading to risks of hyperglycemia or hypoglycemia, and are vulnerable to model failures, which can result in poor insulin regulation.
Innovation Solution
An automated system that includes a blood glucose sensor, an insulin injection device, and a processing and control unit that uses a physiological model described by differential equations to predict blood sugar evolution, with an automatic calibration process to estimate initial state variables and parameters, and a quality check mechanism to ensure model reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a physiological model is used to predict future blood sugar levels, then insulin regulation can be automated, but the prediction accuracy is insufficient leading to hyperglycemia or hypoglycemia risks
Solution Approach 1:
The system performs preliminary calibration of the physiological model using past blood sugar measurements before making future predictions. This pre-adjustment of model parameters based on historical data improves prediction accuracy while maintaining automated regulation
Solution Approach 2:
The system continuously compares predicted blood sugar levels with actual sensor measurements and uses this feedback to adjust and recalibrate the physiological model parameters, thereby improving prediction accuracy over time while maintaining automated control
2Reliability
If a physiological model is used for prediction, then insulin regulation is improved, but the system becomes vulnerable to model failures
Solution Approach 1:
The system performs self-validation by automatically assessing the quality and reliability of its own physiological model predictions. When the model quality falls below thresholds, the system automatically switches to alternative control strategies, thereby maintaining reliable insulin regulation without manual intervention
Solution Approach 2:
The system prepares backup control strategies in advance that can be activated when the physiological model fails or produces unreliable predictions. This pre-prepared alternative ensures continuous safe operation and protects against model failures
3Measurement precision
If manual calibration of the physiological model is performed, then prediction accuracy can be improved, but the system complexity and operation difficulty increase
Solution Approach 1:
The system automatically calibrates its own physiological model parameters using historical blood sugar measurements without requiring manual intervention. This self-calibration process improves prediction accuracy while keeping the system easy to operate by eliminating complex manual calibration steps
Data Source
Figure 1~3
Figure 4~5
Figure 6
AI summary
The invention relates to an automated system for regulating a patient's blood glucose, comprising: a blood glucose sensor (101); an insulin injection device (103); and a processing and control unit (105) adapted to predict the future evolution of the patient's blood glucose from a physiological model and to control the insulin injection device (103) taking into account this prediction, in which: the physiological model comprises a system of differential equations describing the evolution of a plurality of state variables as a function of time;and the processing and control unit (105) is adapted to implement an automatic calibration step of the physiological model comprising a step of estimating initial values of the state variables by minimizing a quantity representative of the error, during a past observation period, between the blood glucose estimated from the physiological model and the blood glucose measured by the sensor (101).