Adaptive On-Board Estimation of Insulin and Carbs From Glucose Trends
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diabetes management systems struggle to accurately determine the real-time concentration or mass of exogenous pharmacons like insulin and carbohydrates due to the complexity of human metabolism, which varies significantly between individuals and over time, making it impractical to use predefined action curves for estimation.
Innovation Solution
An adaptive on-board estimation system that dynamically estimates the equivalent amount of exogenous pharmacons based on glucose time-series data, incorporating glucose concentration readings, insulin injection data, meal data, exercise, and stress data to provide accurate insulin and carbohydrate on-board estimates for diabetes management systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed time-action curves are used to estimate exogenous pharmacon on board, then the estimation process is simple, but the accuracy is poor due to varying human metabolism
Solution Approach 1:
The patent transitions from static fixed time-action curves to dynamic adaptive curves that continuously adjust based on real-time glucose monitoring data. The system updates the action curve parameters dynamically to reflect individual metabolic variations, thereby improving estimation accuracy while maintaining computational feasibility through iterative optimization processes.
Solution Approach 2:
The system implements feedback mechanisms by continuously comparing predicted glucose responses with actual CGM readings and adjusting the pharmacon on board estimation accordingly. This closed-loop approach allows the estimation algorithm to learn from measurement discrepancies and refine its predictions, resolving the accuracy issue without requiring overly complex manual adjustment processes.
2Measurement precision
If real-time measurement of pharmacon absorption and distribution is performed, then accuracy is improved, but the system complexity and impracticality increase
Solution Approach 1:
The patent uses glucose concentration as an intermediary marker to indirectly estimate pharmacon on board status. Instead of directly measuring insulin or carbohydrate concentrations (which would require complex invasive sensors), the system leverages glucose monitoring data as a proxy that reflects pharmacon effects, thereby achieving accurate estimation with simpler, non-invasive CGM technology.
Solution Approach 2:
The system replaces complex direct measurement mechanisms with computational modeling approaches. Rather than using sophisticated sensors to directly detect pharmacon concentrations, the patent substitutes these with algorithmic processing of glucose time-series data, transforming a physical measurement problem into an information processing solution that is both accurate and practically implementable.
3Measurement precision
If personalized metabolic models are used for each subject, then estimation accuracy is improved, but the computational requirements and system complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-calculating and storing subject-specific baseline characteristics and metabolic parameters during an initial calibration phase. This preliminary data collection and processing reduces the computational burden during real-time estimation, as the system only needs to update parameters dynamically rather than performing complex calculations from scratch for each prediction, thereby balancing personalization with computational efficiency.
Data Source
AI summary
Adaptive on board estimation of exogenous pharmacon responsive to transient (i.e., impermanent) physiological effects is provided. Dynamically estimating an equivalent amount of an exogenous pharmacon on board (XOB), such as insulin and/or carbohydrates, left in the subject, is based on predictions of glucose time-series data. These estimated values, such as insulin on board (IOB), are useful for diabetes management software, including decision support and/or artificial pancreas (AP) algorithms, for example.