Indwelling Sensor Factory Calibration via Predictive Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Continuous glucose monitoring sensors require frequent recalibration due to changes in sensor sensitivity and baseline signals over time, leading to inaccuracies in blood glucose monitoring, especially in diabetic patients, who need reliable and convenient glucose level tracking.
Innovation Solution
The implementation of predictive prospective modeling of sensor behavior and physiology to achieve factory calibration, using mathematical models defined by a small number of parameters, which include sensitivity and baseline signals, to estimate blood glucose concentration in real time, and a calibration workflow that adjusts for individual sensor variability through the 'cal check' test.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If factory calibration is implemented using predictive prospective modeling, then calibration accuracy is improved, but device complexity increases due to mathematical models and parameter tracking
Solution Approach 1:
The patent performs calibration computations and determines sensitivity profiles during the manufacturing process before the sensor is deployed. By pre-calculating calibration parameters and storing them in memory, the complex mathematical modeling is done in advance rather than requiring real-time computational resources during operation, thus improving measurement precision while managing device complexity.
Solution Approach 2:
The patent creates a digital copy or model of the sensor's sensitivity characteristics through mathematical modeling. Instead of physically adjusting each sensor during calibration, the system uses computational models to represent and predict sensor behavior, allowing accurate calibration without complex physical adjustment mechanisms.
2Measurement precision
If frequent recalibration is performed to maintain accuracy, then measurement precision is improved, but loss of time increases due to repeated calibration procedures
Solution Approach 1:
By performing comprehensive calibration during manufacturing and pre-determining sensitivity profiles, the patent eliminates the need for frequent recalibration during operation. The sensor is calibrated once in advance with high precision, and this calibration is maintained throughout the sensor's operational life, thus improving measurement precision while avoiding repeated calibration time losses.
Solution Approach 2:
The sensor system performs self-calibration through internal reference measurements and automatic compensation algorithms. The sensor monitors its own performance characteristics and adjusts its output based on pre-determined sensitivity profiles, eliminating the need for external recalibration procedures and user intervention.
3Manufacturing precision
If individual sensor variability is addressed through cal check testing, then manufacturing precision is improved, but productivity decreases due to additional testing steps
Solution Approach 1:
The patent uses computational modeling to create digital representations of each sensor's characteristics during manufacturing. Instead of extensive physical testing, the system uses mathematical models to predict and verify sensor performance, allowing individual sensor variability to be addressed through efficient computational methods rather than time-consuming physical measurements.
Solution Approach 2:
The patent determines sensitivity profiles and calibration parameters during manufacturing by measuring a limited set of key parameters. These parameters are then used to characterize individual sensor variability and adjust calibration settings, achieving high manufacturing precision through efficient parameter-based characterization rather than exhaustive testing.
Data Source
AI summary
Systems and methods are disclosed which provide for a “factory-calibrated” sensor. In doing so, the systems and methods include predictive prospective modeling of sensor behavior, and also include predictive modeling of physiology. With these two correction factors, a consistent determination of sensitivity can be achieved, thus achieving factory calibration.