Fluid Infusion Control Device Using Time-Based Data Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current closed-loop systems for managing blood glucose levels in diabetes patients fail to accurately account for individual glycemic behavior patterns, leading to inaccurate insulin dosing predictions and potential hypoglycemia or hyperglycemia due to unconsidered habits, external factors, and transient physiological changes.
Innovation Solution
A control device that retrieves user data, segments it into time-based categories, calculates segment correction values, and updates coefficients to provide accurate recommendations for future insulin dosing, incorporating recent physiological values and meal data to adapt to individual changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If time-based data is used to predict future glycemia concentration, then insulin dosing can be calculated automatically, but prediction accuracy deteriorates due to inability to account for individual glycemic behavior patterns and physiological changes
Solution Approach 1:
The patent segments historical user data into multiple time periods (e.g., recent period vs. historical period) and calculates separate correction coefficients for each segment. This allows the system to adapt to changing physiological patterns while maintaining automatic operation, resolving the contradiction between automation and prediction accuracy.
Solution Approach 2:
The system dynamically updates correction coefficients based on recent physiological data, allowing the insulin dosing algorithm to adapt to transient physiological changes and evolving glycemic patterns. This dynamic adjustment maintains prediction accuracy while preserving automatic operation.
2Measurement precision
If correction coefficients are updated frequently to adapt to physiological changes, then prediction accuracy improves, but system complexity increases
Solution Approach 1:
The patent updates only specific parameters (correction coefficients) that are most critical for adapting to physiological changes, rather than recalculating the entire insulin dosing model. This partial update approach improves prediction accuracy while limiting the increase in system complexity.
3Measurement precision
If individual glycemic behavior patterns are incorporated into the model, then prediction accuracy improves, but the number of parameters to manage increases
Solution Approach 1:
The patent applies different correction coefficients to different time periods and physiological conditions, allowing the system to capture individual glycemic behavior patterns locally rather than requiring a comprehensive global model. This reduces the overall number of parameters while improving prediction accuracy for specific conditions.
Data Source
AI summary
A control device for determining a recommendation value of a control parameter of a fluid infusion device. The control device comprises a retrieving unit configured to retrieve user data, a segmentation unit, the segmentation unit configured to create a plurality of time segments in order to group user data, a segment correction unit configured to create a segment correction value for each time segment and a period correction unit configured to create an updated correction coefficient for a future determined period of time. The control device also comprises a recommendation unit configured to determine the recommendation value.