Glucose Sensor Error Mitigation via Translation Model Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Continuous glucose monitoring (CGM) sensors face challenges in accuracy and reliability due to measurement errors, noise, and physiological lag, which increase patient burden and complexity, and are influenced by factors like manufacturing variations and transient changes.
Innovation Solution
A processor-implemented method involving a translation model that identifies error metrics, generates modulated values, and updates the model with reduced weighting to mitigate errors, thereby improving the accuracy and reliability of glucose measurements without increasing patient burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration measurements are performed using fingerstick measurements, then measurement accuracy is improved, but patient burden and perceived complexity increase
Solution Approach 1:
The patent extracts and eliminates the need for fingerstick calibration measurements by using alternative calibration approaches that rely on interstitial fluid glucose measurements alone, removing the harmful element (patient burden) while preserving measurement accuracy through different calibration methodologies
Solution Approach 2:
The patent employs disposable test strips that do not require fingerstick calibration, using the test strip itself as a calibration reference through its internal chemistry, thereby eliminating the need for separate calibration measurements and reducing patient burden
2Measurement precision
If signal processing techniques are applied to compensate for physiological lag, then measurement accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary filtering and correction algorithms during the manufacturing and initial setup phases, pre-compensating for known physiological lag characteristics so that minimal real-time signal processing is needed during actual glucose monitoring
Solution Approach 2:
The patent introduces intermediary reference measurements and calibration standards that mediate between the raw sensor signals and the final glucose values, simplifying the correction of physiological lag through standardized intermediate steps rather than complex direct processing
3Reliability
If multiple measurement parameters are used to improve accuracy, then measurement reliability is improved, but manufacturing precision requirements increase
Solution Approach 1:
The patent changes the measurement parameters from requiring multiple precisely-manufactured sensor components to using a single primary measurement parameter (interstitial fluid glucose) supplemented by easily-obtainable contextual parameters (patient data, environmental factors), reducing manufacturing precision requirements while maintaining reliability
Solution Approach 2:
The patent introduces intermediary computational models and algorithms that process readily-available data to compensate for variations in manufacturing precision, using software-based corrections to achieve reliable measurements without requiring extremely precise manufacturing
Data Source
AI summary
Techniques disclosed herein relate generally to sensor error mitigation. In some embodiments, the techniques involve identifying an error metric associated with an input variable to a translation model, determining a reference output of the translation model by providing a reference input value for the input variable to the translation model, generating a modulated value for the input variable based on the reference input value using the error metric, determining a simulated output of the translation model by providing the modulated value for the input variable to the translation model, and updating the translation with a reduced weighting applied to the input variable when a difference between the simulated output and the reference output is greater than a threshold.