Infusion Site Failure Detection Using Glucose and Insulin Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infusion sites in insulin delivery systems can become less effective over time, leading to issues such as leakage, hyperglycemia, and ketones in the blood, which are not easily detected and can prolong hyperglycemic events if not addressed promptly.
Innovation Solution
A system utilizing a model-based approach with a trained machine learning model and a rule-based approach to analyze physiological glucose and insulin delivery data to predict infusion site failure, providing real-time alerts and recommendations for site replacement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If infusion sets are used for extended periods to reduce replacement frequency, then patient convenience and system cost are improved, but the reliability of insulin delivery deteriorates as sites become less effective over time
Solution Approach 1:
The system performs preliminary detection of infusion site failure by continuously monitoring glucose data and insulin delivery patterns, identifying signs of site deterioration before complete failure occurs. This allows proactive site replacement while the infusion set is still functional, resolving the contradiction by enabling extended usage duration while maintaining reliability through early failure detection.
2Reliability
If infusion site failure is detected earlier using advanced monitoring, then reliability is improved, but device complexity increases due to additional sensors and processing requirements
Solution Approach 1:
The system uses the existing glucose sensor and insulin pump data to detect infusion site failure, rather than adding separate detection sensors. The glucose data and insulin delivery records already collected for therapy management are repurposed for failure detection, eliminating the need for additional hardware and keeping device complexity low while maintaining high reliability.
3Measurement precision
If continuous monitoring of glucose and insulin data is performed to detect site failure, then detection accuracy is improved, but energy consumption increases
Solution Approach 1:
The system performs partial monitoring by selectively analyzing glucose and insulin data only when infusion site failure is suspected or at scheduled intervals, rather than continuously processing all data. This reduces energy consumption while maintaining sufficient detection accuracy by focusing computational resources on critical detection moments rather than constant analysis.
Data Source
AI summary
Systems, methods, and devices are provided for predicting a status of an infusion site. Approaches include applying a regression model to physiological glucose data and insulin delivery data to generate predictive data, operating a trained machine learning model to process the predictive data to generate an output, and determining that the infusion site has failed or is likely to have failed based on the output.