Analyte Sensor Fault Detection Using Rate and Threshold Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing analyte sensors are prone to unexpected functional failures due to moisture ingress, leading to inaccurate readings that can mislead patients into administering incorrect medication dosages.
Innovation Solution
Implementing a method for sensor fault detection using calculated metrics and thresholds, such as rate of analyte level change and sum of analyte levels, to generate an indication of suspected sensor faults, thereby reducing false positives and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing data quality checks are used to detect sensor faults, then some artifacts can be removed, but not all cases of sensor faults are detected
Solution Approach 1:
The patent divides sensor fault detection into multiple independent check mechanisms: (1) rate of change analysis comparing current reading to previous readings, (2) plausibility checks against physiological ranges, and (3) consistency checks across multiple sensors. Each mechanism operates independently to detect different types of faults, collectively achieving comprehensive detection that single methods cannot achieve alone.
Solution Approach 2:
The patent applies multiple layers of detection beyond what single existing methods provide. It performs both rate-of-change detection and plausibility range checks simultaneously, and applies checks at multiple time scales (individual readings and rolling averages). This excessive application of detection mechanisms ensures that even if one check fails to detect a fault, other checks will catch it.
2Reliability
If multiple detection mechanisms are implemented to improve fault detection, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent merges multiple detection functions into a unified sensor control device that processes data from multiple sensors simultaneously. The device combines rate-of-change analysis, plausibility checks, and consistency checks into a single integrated system that evaluates all metrics and generates unified fault indications, avoiding the complexity of separate independent systems.
Solution Approach 2:
The sensor control device performs multiple functions: it monitors individual sensor readings, calculates rates of change, validates plausibility against physiological ranges, compares readings across multiple sensors, and generates fault indications. This multi-functional approach consolidates what would otherwise require separate specialized systems into a single universal detection platform.
Data Source
AI summary
Various embodiments of systems, devices and methods for detecting a sensor fault in an analyte sensor are disclosed. These embodiments utilize analyte metrics and thresholds.