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

Measuring in vivo properties and identifying physiological changes in a host using implantable analyte sensors. The sensors can detect analytes like glucose and also measure properties like impedance, pH, temperature, oxygen concentration, and baseline current. By monitoring these additional parameters, the sensors can identify physiological responses like post-implantation sensitivity loss, biofouling, encapsulation, and vasospastic events. The sensors can respond to these identified conditions by adjusting processing of the analyte data or alerting the user.

Source

Calibrating blood glucose measurements from a continuous glucose monitoring system to improve accuracy and reliability. The calibration method selects between two modes based on the difference between blood glucose values measured by the continuous monitor and a reference device. In the first mode, the reference value must fall within a calculated range based on the continuous measurement to calibrate. In the second mode, multiple reference values are used to calibrate if the first mode is not applicable. This prevents inaccurate calibration when single reference values are outliers.

Source

A blood glucose monitoring system that adapts measurement intervals based on user events to provide more accurate readings. The system detects glucose levels at regular intervals. When an event occurs like exercise, food intake, sleep, hormone change, etc., it increases the glucose measurement frequency. This captures rapid glucose changes after events. By using dynamic intervals, the system can better track user glucose trends and alert them to potentially dangerous levels.

Source

Glucose sensing device with wide range and low noise for accurate glucose concentration measurement. The device uses a unique sensing technique to mitigate noise and expand the sensing range. It generates a reference current with variable amplitude based on the digital data. This reference current pulse width is compared with the sensing current pulse width to generate digital measurement data. The sensing current is generated by toggling between voltages, one matching the reference and one different. This avoids noise when measuring glucose concentration by subtracting the noise component from the sensing current.

Source

Reducing the need for field calibration of continuous glucose monitors by using periodic scrutiny potential modulation during sensor operation. The method involves applying a scrutiny potential sequence after each primary voltage step during continuous monitoring. This generates scrutiny currents that provide indicators for sensor condition. Predictive equations using scrutiny currents can be stored in the sensor to compensate for drift, sensitivity changes, and background interference. This allows accurate glucose measurement without calibration during continuous monitoring.

Source

Closed-loop insulin infusion systems using orthogonally redundant glucose sensors for improved accuracy and reliability. The system has two glucose sensors, one optical and one electrochemical, to provide orthogonal redundancy. An algorithm combines the sensor data to improve accuracy and reliability. If one sensor fails, the other can provide glucose values. The sensors have features like distributed electrodes and membrane barriers to reduce drift and fouling. The system uses on-demand calibration rather than frequent fingersticks.

Source

Cross-calibration method for biosensors that reduces background artifact and improves accuracy by using a second sensor inserted after the first. The method involves measuring currents from both sensors before and after insertion, converting them to analyte values, comparing the values, and using the comparison to calibrate the second sensor. This provides consistent readings without relying on blood tests for calibration.

Source

Enhancing glucose monitoring during exercise to improve the accuracy of continuous glucose monitoring (CGM) devices in people with diabetes. The method involves using signals from wearable devices that track physical activity during exercise to compensate for the error in CGM glucose estimation during exercise. The wearable signals are analyzed to calculate an enhancement factor to adjust the CGM glucose readings during exercise. This improves accuracy during exercise compared to relying solely on the CGM sensor.

Source

Dynamic glucose monitoring in patients adaptively determines the frequency of fingerstick glucose tests based on factors like sensor reliability, glucose variability, and current level. It aims to reduce unnecessary fingersticks while still providing appropriate monitoring for individual patients. The system uses continuous glucose monitoring to track patients' levels, and based on factors like sensor reliability, glucose variability, and current level, determines the time for the next fingerstick. This allows more frequent checks when needed, and less frequent checks when stable.

Source

Improving accuracy of glucose sensors and detecting sensor faults by using secondary physiological measurements. The method involves comparing glucose sensor readings with measurements from secondary sensors like lactate, ketone, or heart rate monitors. If the secondary sensor indicates a true physiological condition matching the glucose sensor's low glucose reading, it confirms the low glucose. But if the secondary sensor indicates a false condition, it suggests the glucose sensor may have faults. This allows more aggressive lag correction of glucose levels when secondary sensors confirm true hypoglycemia versus false low readings.

Source

Reducing noise in continuous glucose monitoring (CGM) systems by applying time-varying filtering to the signals during the monitoring period. The filtering smooths out the signals to counteract increasing noise levels as the CGM sensor degrades over time. The filter parameters are adjusted during the monitoring period to compensate for sensor degradation and other component changes that cause noise.

Source

Long-term implantable glucose sensor for continuous monitoring of tissue glucose levels in subjects with diabetes. The sensor is fully implantable, compact, hermetically sealed, and wirelessly transmits glucose data outside the body. It uses a detector array with enzyme-coated membranes, electronics, power source, and telemetry portal. The sensor is implanted subcutaneously for months to years and provides stable, accurate glucose readings for diabetes management. The sensor design allows long-term implantation by minimizing tissue irritation, variability in microvascular perfusion, and enzyme inactivation.

Source

Automatically initializing and calibrating a new glucose sensor when deployed by using sensor data from the old sensor. During sensor overlap, the old sensor continues providing readings while the new sensor initializes. The new sensor receives sensor data from the old sensor and uses it for calibration without needing a new blood sample. This reduces user burden, improves experience, and allows uninterrupted therapy when switching sensors.

Source

A body attachable unit for continuous glucose monitoring that can be inserted and attached to the skin using an applicator. The attachable unit contains the glucose sensor, electronics, and communication components. The user inserts the sensor into their skin through the applicator. After attachment, the user manually makes contact between the sensor and PCB to initiate operation. This allows precise sensor insertion timing, avoids contamination, and improves accuracy compared to pre-assembled sensors.

Source

Real-time detection of early sensor signal attenuation (ESA) in continuous glucose monitoring (CGM) systems to accurately detect episodes of low sensor sensitivity that cause clinically significant errors. The method involves estimating the probability of ESA based on analyte sensor current signals, and then confirming ESA presence using a single blood glucose measurement. This provides accurate ESA detection with minimal false alarms.

Source

Improving the accuracy of continuous glucose monitoring by calibrating the sensor data using the most frequently occurring signal points as a reference for the normal physiological glucose level. The method involves collecting glucose sensor data over time, identifying the signal data points that correspond to known physiological levels, and deriving glucose levels from the collected data using those identified points as a reference. This calibration improves accuracy by accounting for the sensor's drift and variability compared to the true physiological levels.

Source

Compensating for errors during continuous glucose monitoring (CGM) to improve accuracy and reduce lag. The method involves computing sensor progression parameters (SPPs) based on ratios between current and previous glucose signals. These SPPs are used in a gain function to adjust the glucose determination gain and compensate for errors like signal drift and ISF lag. By referencing past signals to present ones, the method leverages the continuous nature of CGM to extract information from the data continuum that can improve glucose determination accuracy.

Source

Intelligent real-time dynamic blood glucose monitoring system using cloud big data to improve accuracy and overcome limitations of implantable glucose sensors. The system leverages historical blood glucose data from a user to correct sensor signals in real-time. It accesses the user's stored data from a cloud server and uses algorithms to establish regression equations relating sensor signals, impedance, and historical glucose levels. This compensates and modifies the calculated blood glucose results from the sensor to improve accuracy by accounting for individual variations and sensor artifacts.

Source

Method to stabilize a continuous glucose monitoring system to improve accuracy by adjusting the stabilization time before displaying the readings. The stabilization involves the communication terminal receiving data from the sensor transmitter and controlling the stabilization time. The first step is a fixed time stabilization. If not stabilized, a second step allows a variable time. If still not stabilized, a third step further extends the time. If stabilization is not reached by the maximum time, the connection is terminated. This allows faster stabilization compared to a fixed time.

Source

Orthogonally redundant glucose sensing for continuous glucose monitoring (CGM) in diabetes management. The system uses two types of glucose sensors, electrochemical and optical, to improve accuracy and reliability compared to single sensor CGMs. It involves calibration, integrity checks, and fusion techniques to leverage the redundancy. The sensors are calibrated separately when possible, and if one fails integrity, the other's output is corrected and calibrated. Fused sensor glucose values are calculated based on weighted values from each sensor using reliability indices. This provides robust glucose monitoring with redundant sensors.

Source

Continuously determining glucose levels in blood using a sensor in tissue surrounding the blood and a mathematical model. The model correlates blood glucose to tissue glucose with process noise. Sensor measurements are correlated to tissue glucose with measurement noise. A moving horizon estimation method uses past sensor and tissue readings to estimate current blood glucose. This provides more accurate blood glucose estimates than just sensor readings due to accounting for the diffusion delay.

Source

Determining a patient's blood glucose level using a continuous interstitial sensor with improved accuracy by accounting for individual time lag differences between interstitial and capillary glucose levels. The method involves measuring interstitial glucose, correcting for lag, offset, and sensitivity based on prior measurements, then calculating the true capillary glucose level at the time of the interstitial measurement. This provides a more accurate estimate compared to using uncorrected interstitial glucose. The correction factors are personalized to each patient based on their unique glucose dynamics.

Source

A method for monitoring blood glucose levels using continuous glucose monitors that reduces the frequency of calibrations needed. The method involves selecting the best calibration model from a set of candidate models based on a Bayesian statistical framework. The models describe different cases of the device-physiology interface state and sensor characteristics. The selection is based on a priori probabilities of accurately predicting blood glucose levels using sensor signals and references. By dynamically adapting the calibration model to the sensor state, accuracy is improved compared to static global calibration.

Source

Transcutaneous glucose sensor with a membrane that resists glucose influx to allow long-term wear without calibration. The sensor has a membrane with layers applied by vapor deposition. The membrane controls glucose entry and has an enzyme domain, electrode domain, and interference domain. The vapor deposition allows consistent membrane properties across sensors. The sensor can be manufactured in batches with consistent performance. The membrane resistance prevents glucose diffusion and enables long-term wear without recalibration. The sensor can be used in a system with a receiver and key to limit usage time.

Source

Communication protocol for data communication between a continuous glucose monitoring transmitter and receiver. It involves generating a radio frequency (RF) data stream based on the monitored glucose data using a low-power, low-noise logic circuit consisting of finite-state machines and digital circuits. This avoids a high-power ASIC and enables accurate glucose monitoring for diabetes treatment. The transmitter sends glucose data in a synchronized window over RF to the receiver, which identifies the transmitter and displays the glucose levels.

Source

Improving the accuracy of continuous in-body analyte monitoring by calibrating the sensor readings using known physiological levels. The method involves collecting sensor data over a period, identifying the most frequently occurring signal points, and using those as reference points to derive calibrated analyte levels. This leverages the fact that the body naturally maintains stable analyte levels for most of the time. By identifying the most frequently occurring signals, it can determine the normal physiological level and use that as a reference to calibrate the readings.

Source

Method and device for accurately determining if blood glucose sensor data is correct or erroneous. The method involves analyzing sensor readings based on factors like sampling values, slopes, peaks, and trends to detect potential errors. If criteria like insufficient points, missed peaks, inconsistent slopes, or excessive fluctuations are met, the data is flagged as error prone. The device collects sensor data, applies the error detection criteria, and can flag faulty data for retesting or replacement.

Source

Enhancing the accuracy of glucose sensors used in continuous glucose monitoring (CGM) systems for diabetes management by leveraging information from insulin pumps to improve sensor accuracy, particularly during hypoglycemia where CGM accuracy is lowest. The method involves using insulin delivery data along with glucose sensor readings and a filtering algorithm to estimate glucose levels. This estimated glucose is then weighted more heavily than the sensor reading during hypoglycemia to account for the sensor's accuracy issues in that range. The weighting scheme balances sensor and estimated glucose based on factors like insulin delivery and sensor error indices.

Source

Accurately measuring physiological parameters like blood analyte levels using implantable sensors by reducing errors from spatial arrangement of sensing elements. The sensor design has analyte-modulated sensing electrodes surrounded by multiple background electrodes. This configuration mitigates errors compared to prior sensors with just one background electrode. It also uses machine learning to model and correct errors during normal operation based on training data.

Source

Automated blood glucose monitoring system that adaptively determines the patient's insulin sensitivity coefficient based on physical activity levels. The system has a blood glucose sensor, a physical activity sensor, and a processing unit. It calculates an influence signal representing how physical activity affects insulin sensitivity. Using a mathematical model, it estimates the patient's insulin sensitivity coefficient from the activity signal. This adaptive coefficient is then used to control insulin injections. The system periodically recalibrates the model using glucose, insulin, and activity history.

Source