Insulin Pump Basal Delivery Estimation Using Recent IOB
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional insulin pump systems inaccurately determine basal insulin needs due to deviations in manual deliveries, leading to glucose level management issues and prolonged compensation for glucose excursions.
Innovation Solution
The system adjusts the proportionality parameter based on recent manual insulin deliveries and insulin on board (IOB) within a specific duration, calculates a reliability metric for TDI variance, and distinguishes between post-prandial and non-post-prandial basal insulin deliveries to improve basal insulin need estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional insulin pump systems calculate basal insulin needs as a proportion of TDI using fixed proportionality parameters, then the system maintains simple calculation methods, but the accuracy of basal insulin delivery determination deteriorates due to deviations in manual deliveries
Solution Approach 1:
The system dynamically adjusts the proportionality parameter P based on recent manual bolus deliveries and insulin on board (IOB) within a sliding time window (e.g., past 5 hours). Instead of using a fixed historical period, the parameter adapts in real-time to reflect current insulin activity, improving accuracy while maintaining computational efficiency through a focused time window approach.
Solution Approach 2:
The system changes the proportionality parameter P from a static value to a dynamic value that varies based on recent manual deliveries and IOB. The parameter is recalculated continuously within a moving time window, allowing the system to adapt to changing insulin needs without requiring complex long-term historical analysis.
2Measurement precision
If the system accounts for insulin on board (IOB) from manual boluses in proportionality parameter calculation, then the accuracy of basal needs estimation improves, but the computational complexity and data processing requirements increase
Solution Approach 1:
The system extracts only the relevant portion of historical data (manual boluses and their IOB) within a specific recent time window, rather than analyzing complete historical records. This selective extraction focuses computational resources on the most impactful recent deliveries while ignoring older, less relevant data, thereby improving accuracy without proportionally increasing complexity.
Solution Approach 2:
The system uses a partial time window (e.g., past 5 hours) rather than analyzing complete historical data. This partial action approach captures the most relevant recent insulin activity sufficient for accurate basal calculation without the burden of processing excessive historical information, achieving good accuracy with manageable computational effort.
3Reliability
If conventional systems use fixed historical periods for calculating proportionality parameters, then the system maintains operational simplicity, but the reliability of basal insulin delivery deteriorates during glucose excursions
Solution Approach 1:
The system incorporates feedback from recent manual bolus deliveries and IOB status to continuously adjust the proportionality parameter. This feedback mechanism allows the system to respond to changing conditions (such as glucose excursions) by adapting the parameter based on actual recent insulin activity patterns, improving reliability while maintaining a relatively simple computational framework.
Data Source
AI summary
Exemplary embodiments may attempt to account for more substantial deviations in a user's basal insulin needs and reduce or eliminate glucose level excursions that may result from such deviations more quickly than conventional insulin pump systems. The exemplary embodiments may rely upon a user's recent basal insulin delivery history to establish a current estimate of the user's basal insulin needs. Exemplary embodiments may calculate a reliability metric for a TDI value that reflects that degree of variance in TDI value for a user over a recent period, such as over multiple days or weeks. The reliability indicator may be used to adjust the TDI value that is used to establish a user's daily basal insulin needs and hourly basal insulin needs. In determining a user's basal insulin needs, the exemplary embodiments may not only look to recent basal insulin deliveries but also may account for correction boluses and/or meal boluses.