Insulin Control Algorithm Validation Using Real CGM Action Differences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control algorithms for insulin delivery in diabetes management are not adequately validated, posing risks of hypoglycemia and hyperglycemia, and clinical trials for validation are time-consuming and potentially dangerous.
Innovation Solution
A method using real data from continuous glucose monitoring sensors to validate control algorithms by comparing action differences with reference algorithms, ensuring accuracy and safety without the need for simulators or clinical trials.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If clinical trials are used to validate control algorithms, then validation reliability is improved, but validation time and safety risk increase
Solution Approach 1:
The patent creates virtual copies of patients (virtual patients) with simulated physiological models that replicate real patient behavior. These virtual patients are used to validate control algorithms through extensive simulations, eliminating the need for actual clinical trials while maintaining validation reliability. The virtual patients are generated from real patient data and can be extensively simulated without time or safety constraints.
Solution Approach 2:
The patent replaces the mechanical process of clinical trials with a computational simulation system. Instead of physically testing algorithms on real patients through clinical trials, the system uses computer-based virtual patients with physiological models to simulate and evaluate algorithm performance, thereby reducing validation time and eliminating safety risks associated with real-world testing.
2Reliability
If clinical trials are used to validate control algorithms, then validation reliability is improved, but safety risk increases
Solution Approach 1:
The patent creates virtual copies of patients (virtual patients) with simulated physiological models that replicate real patient behavior. These virtual patients are used to validate control algorithms through extensive simulations, eliminating the need for actual clinical trials while maintaining validation reliability. The virtual patients are generated from real patient data and can be extensively simulated without time or safety constraints.
Solution Approach 2:
The patent introduces virtual patients as an intermediary between the control algorithm and real patients. The virtual patients serve as a safe intermediary medium that allows algorithm validation without directly exposing real patients to potential harm. The physiological models of virtual patients mediate the testing process, enabling safety evaluation while maintaining validation integrity.
3Loss of time
If simulators are used to validate control algorithms, then validation time is reduced, but simulation accuracy decreases due to modeling biases
Solution Approach 1:
The patent dynamically adjusts and refines the physiological model parameters of virtual patients based on real patient data. Instead of using fixed, biased simulator parameters, the system continuously updates model parameters to accurately reflect individual patient characteristics, thereby maintaining high simulation accuracy while benefiting from the time efficiency of virtual simulation.
Solution Approach 2:
The patent implements feedback mechanisms where simulation results are continuously compared with actual patient outcomes, and the virtual patient models are updated accordingly. This feedback loop ensures that the simulation accuracy is maintained at high levels by continuously refining the physiological models based on real-world performance data, eliminating the modeling biases inherent in static simulators.
Data Source
AI summary
A method for validating a control algorithm, the method being implemented by a computer and comprising the steps of receiving a plurality of states (50); receiving a plurality of reference actions (52); receiving a plurality of reference outputs (54); processing the plurality of states (56); computing at least an action difference (58) consisting of computing at least a difference between at least one control action and at least one reference action; evaluating the at least an action difference (60) by computing at least an evaluation score; and validating the control algorithm (62), the control algorithm being validated if at least the evaluation score satisfies an evaluation score criteria.
