Insulin Delivery Settings Adaptation by Glucose Control Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated insulin delivery (AID) systems employ a one-size-fits-all approach to initial settings, which are not optimal for individual users, leading to suboptimal glucose control and a prolonged adjustment period.
Innovation Solution
The system allows for personalized adjustment of insulin delivery settings based on the user's quality of blood glucose level control, enabling more aggressive settings for good control and less aggressive settings for poor control, with options for automatic or user-initiated changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a one-size-fits-all approach to initial settings is used, then device complexity is reduced and ease of manufacture is improved, but adaptability to individual user needs deteriorates and productivity in achieving optimal glucose control decreases
Solution Approach 1:
The system pre-calculates and stores multiple optimized setting profiles (aggressive, moderate, conservative) before the user needs them. These profiles are prepared in advance based on simulated or population data, so when a user demonstrates competent glucose control, the appropriate profile is already ready for immediate activation without requiring complex real-time calculations or user configuration.
Solution Approach 2:
The system changes key control parameters (such as insulin delivery aggressiveness, constraint tightness, cost function weights) based on the user's demonstrated performance. By transitioning between pre-defined parameter sets corresponding to different competence levels, the system achieves adaptability without requiring complex individualized tuning for each user.
2Productivity
If conventional initial settings are used, then ease of operation is improved, but the time required to achieve optimal glucose control increases
Solution Approach 1:
The system continuously monitors the user's glucose control quality and uses this feedback to determine when to transition between setting profiles. By measuring actual performance (time in range, glucose variability) and comparing it against thresholds, the system automatically accelerates the adaptation process, reducing the time to reach optimal settings without requiring manual intervention or extended trial periods.
Solution Approach 2:
The system dynamically adjusts the level of aggressiveness in insulin delivery based on real-time assessment of user competence. Rather than using fixed conservative settings indefinitely, the system transitions from conservative initial settings to more aggressive optimized settings as the user demonstrates ability, thereby reducing the adjustment period while maintaining safety.
3Reliability
If aggressive settings are used from the beginning, then productivity in achieving glucose control is improved, but the risk of harmful outcomes such as hypoglycemia increases
Solution Approach 1:
The system applies conservative settings with built-in safety margins during the initial period when user competence is unknown. These conservative settings act as a cushion protecting against hypoglycemia while the system monitors user performance. Once the user demonstrates adequate glucose control, the system transitions to more aggressive settings that maintain effectiveness while the user's demonstrated competence provides a form of validation cushion.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
The exemplary embodiments allow a user to have more aggressive settings or less aggressive settings for an AID system after demonstrating good blood glucose level control. This allows the settings to be more quickly customized to users that demonstrate good quality blood glucose level control than conventional systems. As these users have demonstrated good quality glucose level control there is less of a need to constrain the settings and provide a high margin of safety. Conversely, users demonstrating poor quality blood glucose level control may have less aggressive settings imposed, representing a higher margin of safety.