Optimal Insulin Infusion Profile via Objective Function Minimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current insulin delivery systems, such as Continuous Subcutaneous Insulin Infusion (CSII) and Multiple Daily Injections (MDI), fail to accurately replicate physiological insulin delivery, leading to unsatisfactory postprandial blood glucose control due to the complexity of meal absorption characteristics, resulting in both short-term and long-term complications.
Innovation Solution
A system and method that computes an optimal insulin infusion profile by minimizing an objective function, taking into account pre-meal blood glucose concentration, meal size and composition, and individual insulin infusion history, using a dynamical minimal model to determine the optimal insulin dosage and timing sequence for improved postprandial blood glucose control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single-shot bolus is used to control postprandial blood glucose, then the insulin delivery is simple to implement, but the control accuracy is insufficient due to inability to account for complex meal absorption characteristics
Solution Approach 1:
The patent divides the single-shot bolus into multiple distributed boluses administered over time. The system segments the total insulin dose into multiple smaller doses delivered at different time points (e.g., immediate bolus followed by extended bolus), allowing the insulin delivery to match the complex absorption kinetics of different meal components more accurately.
Solution Approach 2:
The patent implements dynamic insulin delivery by adjusting the insulin infusion rate over time based on real-time glucose monitoring data. The system continuously adapts the bolus distribution pattern according to the actual glucose response, making the insulin delivery dynamic rather than static, thereby improving postprandial control accuracy.
2Measurement precision
If distributed boluses are used to account for meal absorption characteristics, then the postprandial blood glucose control is improved, but the system complexity increases making it difficult to operate
Solution Approach 1:
The patent implements a closed-loop control system where the insulin pump automatically calculates and administers the optimal distributed bolus pattern based on glucose sensor data. The system self-adjusts the insulin delivery without requiring manual intervention or complex user decisions, making the sophisticated distributed bolus functionality as easy to use as a simple bolus command.
Solution Approach 2:
The patent incorporates continuous glucose monitoring feedback to automatically adjust the distributed bolus delivery. The system uses real-time glucose measurements to determine the optimal timing and amount of subsequent insulin doses, eliminating the need for users to manually calculate complex dosing schedules while maintaining high control accuracy.
3Device complexity
If conventional insulin therapy is used, then the device complexity is low, but the reliability of glycemic control is insufficient leading to complications
Solution Approach 1:
The patent replaces manual insulin dosing calculations and decisions with an automated control algorithm that processes glucose sensor data and determines optimal insulin delivery. This substitution of manual mechanical dosing with automated electronic control significantly improves the reliability and consistency of glycemic control while maintaining ease of use through simple user interfaces.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system (200) which is designed to determine an optimal insulin infusion profile useful as a postprandial blood glucose control for a patient/meal combination is disclosed. The system comprises: a) an input device configured to receive information about a pre-meal measurement of blood glucose concentration, and size and composition of a meal to be consumed; b) a memory which contains additional information about a pre-meal insulin infusion history, a chosen insulin pattern for the postprandial blood glucose control, and a dynamical minimal model having model parameters which are predetermined for the patient/meal combination; c) a processor programmed to use the received information and the additional information to compute automatically the optimal insulin infusion profile by minimizing an objective function, wherein said minimizing of the objective function results in the optimal insulin infusion profile; and d) a display configured to display the optimal insulin infusion profile. Furthermore, a method for the determination of an optimal insulin infusion profile useful as a postprandial blood glucose control for a patient/meal combination is disclosed.