Meal Response Modeling for Predictive Glucose Bolus Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blood glucose management systems struggle to accurately determine insulin doses for meals containing macronutrients other than carbohydrates, leading to unpredictable glucose level fluctuations due to inaccurate calculation of residual carbohydrates and reactive correction boluses.
Innovation Solution
A physiological model is fitted to a person's glycemic response to a specific meal, allowing for the prediction of residual carbohydrates and future glucose levels, enabling precise calculation of correction boluses to maintain desired glucose ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive correction boluses are used to manage glucose levels, then glucose control can be achieved, but glucose level fluctuations become unpredictable and clinical outcomes deteriorate
Solution Approach 1:
The system performs preliminary action by predicting future glucose levels and residual carbohydrates before the glucose excursion occurs. The physiological model forecasts glucose levels at future time points (e.g., 1, 2, 3 hours ahead) and calculates the required correction bolus in advance, rather than reacting after glucose levels have already deviated from the target range.
Solution Approach 2:
The system implements feedback by continuously monitoring actual glucose levels and comparing them with predicted values from the physiological model. This feedback loop allows the system to assess prediction accuracy and adjust future predictions, ensuring reliable glucose control through iterative refinement of the control strategy.
2Ease of operation
If traditional insulin calculation methods are used, then insulin dosing can be performed, but the calculation of residual carbohydrates becomes inaccurate leading to unpredictable glucose fluctuations
Solution Approach 1:
The physiological model serves as an intermediary between traditional insulin calculation methods and accurate residual carbohydrate prediction. The model incorporates macronutrient composition data and individual physiological parameters to mediate the calculation process, transforming simple input data into accurate predictions of future glucose levels and residual carbohydrates.
Solution Approach 2:
The system applies parameter changes by using individualized physiological parameters (insulin sensitivity factor, carbohydrate absorption rate, macronutrient composition) specific to each patient and meal type. These parameter variations enable accurate prediction of residual carbohydrates for different meal compositions without complicating the overall dosing process.
3Reliability
If correction boluses are delivered reactively, then glucose excursions can be addressed, but time-in-range metrics deteriorate due to delayed response
Solution Approach 1:
The system delivers correction boluses in advance based on predicted future glucose levels. By calculating the required insulin dose before the glucose excursion occurs (proactive rather than reactive), the system prevents hyperglycemia rather than correcting it after the fact, thereby maintaining better time-in-range metrics.
Solution Approach 2:
The system implements dynamic insulin dosing by continuously updating predictions based on real-time glucose measurements and changing physiological conditions. The correction bolus timing and dosage are dynamically adjusted according to the predicted glucose trajectory, allowing optimal response timing that maintains glucose stability.
Data Source
AI summary
A processor-implemented method comprises obtaining measured glucose values of a person, fitting a physiological model to a portion of the measured glucose values within a time window after a start of a meal to determine meal-specific values of parameters of the physiological model that characterizes the person's glycemic response to the meal, and predicting a future blood glucose level of the person at a first time after the time window using the physiological model and the meal-specific values of the parameters of the physiological model. In one example, an alert or a notification can be sent to a user or an electronic device based on the predicted future blood glucose level of the person.