Predictive Infusion Control With Glucose Data Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infusion pump devices face challenges in maintaining accurate blood glucose regulation due to corrupted or invalid glucose data samples, leading to unpredictable operation and potential delays in insulin delivery, especially when recursive prediction algorithms are reset to eliminate 'bad' data.
Innovation Solution
A processor-implemented method that monitors physiological conditions, transitions between operational modes based on glucose level thresholds and elapsed time, and uses modified measurement sequences to calculate predicted values, allowing continuous operation even with unusable data samples by interpolating or replacing them with more recent values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If blood glucose data samples are continuously monitored to maintain accurate regulation, then control precision is improved, but battery life deteriorates
Solution Approach 1:
The system performs preliminary actions by continuously maintaining a modified measurement data sequence that includes interpolated or replaced values, so that when actual glucose samples are needed, the predictive algorithm can immediately use pre-prepared data without requiring continuous real-time sampling, thus reducing energy consumption while maintaining regulation accuracy
Solution Approach 2:
The patent introduces an intermediary mechanism that modifies the measurement data sequence by interpolating or replacing corrupted samples with estimated values from the modified sequence itself. This intermediary data processing layer allows the system to maintain continuous operation and predictive capability without requiring uninterrupted actual sensor readings, thereby extending battery life while preserving measurement precision
2Reliability
If predictive algorithms are reset to eliminate corrupted data, then data reliability is improved, but response time deteriorates
Solution Approach 1:
The system maintains continuous useful action by keeping a modified measurement data sequence that continuously incorporates interpolated or replaced values. Instead of resetting the predictive algorithm when corrupted data is detected, the system continuously updates the measurement sequence with valid or estimated values, allowing the algorithm to operate without interruption and eliminating the lag time that would otherwise occur during algorithm reset and re-initialization
Solution Approach 2:
The modified measurement data sequence serves itself by using interpolated or replaced values to maintain its own continuity. When corrupted data is detected, the sequence automatically generates replacement values from existing valid data points, allowing the system to self-correct without external intervention or algorithm reset, thus maintaining both reliability and continuous operation
3Duration of action of stationary object
If corrupted data samples are used in predictive algorithms, then operational continuity is maintained, but prediction accuracy deteriorates
Solution Approach 1:
The patent introduces an intermediary data processing layer that modifies the measurement data sequence by identifying and replacing corrupted samples with interpolated values derived from surrounding valid data points. This intermediary mechanism allows the predictive algorithm to receive clean, accurate data continuously, maintaining both operational continuity and prediction accuracy simultaneously
Solution Approach 2:
The system creates a copy of the measurement data sequence that is modified to include interpolated or replaced values. This modified copy serves as the input to the predictive algorithm, allowing the system to maintain operational continuity while ensuring prediction accuracy by using the cleaned, interpolated data sequence rather than the original corrupted data
Data Source
AI summary
Techniques for fluid delivery are provided. In some embodiments, the techniques may involve monitoring a value representative of a physiological condition in a body of a user. The techniques may further involve operating in a first operational mode of a fluid delivery device based on the value representative of the physiological condition. The techniques may further involve transitioning to a second operational mode of the fluid delivery device based at least in part on the value representative of the physiological condition relative to a threshold and a time that has been spent in the first operational mode of the fluid delivery device.