Adaptive Glucose Sensor Algorithm Balancing Lag and Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analyte monitoring systems face errors due to lag and noise between interstitial fluid and blood glucose measurements, affecting accuracy in glucose level and rate of change data.
Innovation Solution
A combination of algorithms is used to balance signal responsiveness and noise reduction, including lag correction and smoothing algorithms that minimize error and noise amplification, utilizing time derivative estimates and historical data to generate accurate glucose values and rates of change.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lag correction algorithms are applied to eliminate blood-to-interstitial glucose dynamics error, then point-wise accuracy of glucose values is improved, but noise amplification increases
Solution Approach 1:
The system dynamically adjusts the degree of lag correction by changing parameters such as the time derivative estimates and historical data weights based on current glucose rate of change conditions. When glucose is stable, stronger lag correction is applied; when glucose is changing rapidly, weaker correction is applied to avoid noise amplification.
Solution Approach 2:
The lag correction algorithm transitions from static to dynamic operation by continuously adapting correction strength based on real-time glucose rate of change. The system monitors whether glucose is within a predefined rate threshold and adjusts correction parameters accordingly, making the correction process adaptive rather than fixed.
2Measurement precision
If smoothing algorithms are applied to eliminate noise and sample time-to-sample time variation, then rate of change accuracy is improved, but responsiveness to true glucose change decreases
Solution Approach 1:
The smoothing algorithm dynamically adjusts its parameters based on the detected glucose rate of change. When glucose is stable, stronger smoothing is applied to eliminate noise; when glucose changes rapidly, smoothing is reduced to maintain responsiveness. This adaptive parameter adjustment resolves the contradiction between noise reduction and responsiveness.
Solution Approach 2:
The smoothing process transitions from static to dynamic by continuously monitoring glucose rate of change and adjusting smoothing intensity in real-time. The system adapts the balance between noise reduction and responsiveness based on current metabolic conditions, making the smoothing algorithm intelligent rather than fixed.
3Productivity
If higher sample rate is desired for more frequent glucose measurements, then monitoring frequency is improved, but computational complexity and processing load increase
Solution Approach 1:
The system dynamically adjusts the effective sampling frequency based on glucose rate of change. When glucose is stable, the system can use lower sampling rates with stronger lag correction and smoothing. When glucose changes rapidly, the system increases sampling frequency to capture the changes accurately. This adaptive approach optimizes the balance between monitoring frequency and computational load.
Data Source
AI summary
The present disclosure provides methods of processing data provided by a transcutaneous or subcutaneous glucose sensor utilizing different algorithms to strike a balance between signal responsiveness accompanied by signal noise and the introduction of error associated with that noise. The methods utilize the strengths of a lag correction algorithm and a smoothing algorithm to optimize the quality and value of the resulting data (glucose concentrations and the rates of change in glucose concentrations) to a continuous glucose monitoring system. Also provided are systems and kits.