Glucose Sensor Accuracy Optimization

Compensating for varying accuracy of medical sensors over their lifetime to improve performance of devices like insulin pumps. The technique involves adjusting sensor readings based on estimated accuracy levels at different points in the sensor's lifespan. This accounts for the fact that sensors can have lower accuracy early on and higher accuracy later. By taking into account the sensor age, the adjustment aims to provide more accurate readings to devices like insulin pumps to improve dosing decisions.

Source

Detecting compression artifacts in continuous glucose monitoring (CGM) sensors in real time to prevent false hypoglycemia alarms, insulin shutoff, and other negative impacts on diabetes treatment. The method involves comparing clearance values between consecutive CGM readings to normal distributions. If the clearance values fall outside the normal distributions, it indicates compression artifacts from sensor compression. This allows accurate and real-time detection of CGM sensor compression and compression artifacts that can then be used to improve diabetes treatment by preventing false alarms, insulin stoppage, etc. during compression.

Source

Improving the performance of medical devices like glucose sensors by determining personalized calibration information specific to each sensor. The calibration is based on parameters measured during manufacturing. This allows individualized calibration values for each sensor, rather than a single value for a group. It involves modifying sensor surfaces, measuring sensor properties, and using machine learning to predict calibration. This improves accuracy and reduces variation compared to a single calibration value.

Source

Non-invasive glucose monitoring system that allows accurate and reliable glucose readings without piercing the skin. The system uses near-infrared spectroscopy to detect glucose levels through the skin. It involves placing a sensor against the skin that emits near-infrared light and measures how it is absorbed and reflected. This allows glucose concentration to be estimated without breaking the skin. The system also has a calibration step and an alert mechanism if levels stray outside a set range.

Source

Reducing blanking of sensor glucose signals in continuous glucose monitoring (CGM) systems by using multiple machine learning models trained for specific conditions. The CGM device inputs sensor data into multiple models, each trained for different data characteristics or abnormal conditions. The models generate predicted glucose values. The device combines the predictions to generate the displayed glucose value. This improves accuracy and reduces blanking compared to using a single model.

Source

Monitoring of multiple analytes using a continuous multi-analyte monitoring and sensor system that facilitates better management of disease, e.g., diabetes. The monitoring includes monitoring multiple analytes with a working electrode, detecting and compensating for artifacts and related events, and determining whether a compression event has occurred based on the measured first and second analyte data and the physiological state of the patient.

Source

A dynamic blood glucose monitoring system that automatically starts testing without user intervention to improve accuracy and prevent reuse of implanted sensors. The system checks the initial sensor readings to determine if the sensor is still inside the body. If the initial readings are zero, indicating the sensor has not touched glucose, the system starts monitoring. If the initial readings are non-zero, indicating the sensor is already implanted, the system waits for normal operation conditions with sequential increasing readings before starting monitoring. This prevents false triggers from resetting the system without removing the sensor.

Source

Enhancing the accuracy of glucose sensors and detecting sensor faults using secondary physiological measurements involves comparing glucose readings from the sensor with secondary metrics such as ketone levels and heart rate. This method helps confirm true hypo/hyperglycemia versus false conditions. Additionally, secondary measurements are used to correct glucose readings during sensor attenuation. By leveraging secondary sensors, this approach distinguishes between physiological and false glucose trends and optimizes lag correction.

Source

Personalized calibration method for glucose sensing devices that improves accuracy by accounting for subject-specific differences in glucose diffusion between sensors. It involves using two glucose sensors at different depths in a subject to estimate personalized time constants for glucose diffusion from blood to each sensor site. This allows more accurate glucose level estimation using the sensors by accounting for the subject's unique interstitial glucose dynamics. The method involves obtaining glucose measurements from both sensors over a time interval, estimating the personalized time constants based on the measurements, and using the estimated time constants to calculate glucose levels from the sensors.

Source

Improved analyte monitoring systems for diabetes management that provide personalized glucose metrics, user-friendly interfaces, and more accurate data handling. The systems calculate personalized metrics like adjusted HbA1c based on individual factors like glucose uptake and red cell turnover. The interfaces display actionable alerts and trend graphs. The systems also improve data integrity by backfilling, merging, and handling disconnections. This allows more accurate, flexible, and robust monitoring.

Source

A method for accurately determining continuous glucose levels from a glucose sensor without a separate blood glucose reference involves subtracting the time-dependent zero-signal level of the sensor from the continuous sensor signal. This compensates for drift and interference, resulting in a more accurate representation of the actual glucose level in the body fluid.

Source

A wearable medical device that can adjust the frequency of glucose measurements from a continuous glucose monitor (CGM) based on the user's glucose trend. The device processes the CGM data to determine the rate of change in glucose levels over time. If the rate indicates a rapid change, the device instructs the CGM to provide more frequent glucose readings. This allows faster detection of extreme glucose events to enable timely intervention. The device can also suspend insulin delivery based on the CGM data.

Source

Data-stream bridging to provide uninterrupted glucose monitoring during sensor transitions. The technique involves generating estimated glucose values during the period between when an old sensor expires and a new sensor is properly warmed up. It does this by using historical glucose data from the old sensor and current but potentially inaccurate data from the new sensor to predict glucose levels. This allows continuous glucose monitoring without gaps when swapping sensors. It also shortens warmup times by ending them when the new sensor's accuracy meets a threshold.

Source

Improving the accuracy of continuous glucose monitoring (CGM) devices by using sensor progression parameters to correct for errors and delays in glucose measurements. The method involves calculating ratios between current and previous glucose signals from a CGM sensor. These ratios are stored in a gain function and used to adjust the CGM glucose value at the current point. This reduces errors from sources like sensor drift and interstitial fluid delays.

Source

More accurate and efficient method for determining glucose levels in real-time using a continuous glucose monitoring (CGM) system. The method involves estimating the current glucose level in the blood based on interstitial fluid glucose measurements using a Kalman filter. It improves the estimation accuracy compared to previous methods by leveraging the dynamics of glucose diffusion between blood and tissue. The filter accounts for the time delay between blood and interstitial glucose changes. This allows more precise estimation of blood glucose levels from interstitial measurements.

Source

Continuous glucose monitoring (CGM) system that enables accurate, stable, and uninterrupted glucose monitoring without the need for frequent fingerstick calibrations. The system involves overlapping sensor placements during sensor swaps. After calibrating the first sensor, a second sensor is placed while the first is still in the body. The second sensor's calibration is based on data from the first sensor, eliminating the need for fingerstick calibrations. This allows continuous calibration and monitoring without gaps as sensors are replaced.

Source

Continuous glucose monitoring system that reduces user pain and provides more accurate readings compared to invasive devices like lancets. The system uses both invasive and non-invasive glucose sensors to measure blood sugar. The invasive sensor takes an initial measurement, and the non-invasive sensor continuously monitors changes. This provides a more complete picture of glucose levels. The invasive sensor's initial reading is sent to a server along with the non-invasive sensor's continuous data. The server correlates the data to predict future glucose levels and trends. This allows users to proactively manage their diabetes without frequent invasive tests.

Source

Improving accuracy and reliability of glucose monitoring systems by providing data processing and control features. The method involves sampling a batch of in vivo glucose sensors, determining the sensitivity of each sensor, calculating a mean sensitivity, and using that mean value to validate sensor data. Sensor codes are assigned based on sensitivity ranges. During monitoring, the sensor code is entered and the system checks data against the corresponding mean sensitivity. If outside a tolerance, it disables glucose trend calculations and alerts the user. This accounts for sensor drift and ensures accurate trend estimation.

Source

Method for improving continuous glucose monitoring (CGM) systems to accurately measure blood glucose levels despite sensor data limitations and outlier conditions. The method involves training multiple CGM models for different scenarios like sensor availability, accuracy, outlier conditions. When CGM sensor data is input, the appropriate model is selected based on signatures in the sensor data. This allows the system to use the best model for the given conditions and produce more reliable glucose estimates.

Source

Improving the accuracy of continuous glucose monitoring systems by combining algorithms to compensate for lag and noise. The system monitors glucose levels over time and generates lag-corrected signals to improve point-wise accuracy. It also generates smoothed signals to reduce noise. These signals are then used separately to estimate glucose concentration and rate of change. This two-step process balances responsiveness and noise reduction.

Source