US-12626819-B2 - Machine learning models for detecting outliers and erroneous sensor use conditions and correcting, blanking, or terminating glucose sensors

US12626819B2US 12626819 B2US12626819 B2US 12626819B2US-12626819-B2

Abstract

Techniques for improving continuous glucose monitoring (“CGM”) are described herein. In some embodiments, the techniques involve obtaining sensor data; applying, to the sensor data, a machine learning model trained to identify sensor data error patterns; and detecting an erroneous sensor use condition based on output of the machine learning model indicating an error pattern identified in the sensor data.

Inventors

Elaine Gee
Jeffrey Nishida
Peter Ajemba
Keith Nogueira
ANDREA VARSAVSKY

Assignees

MEDTRONIC MINIMED, INC.

Dates

Publication Date: 20260512
Application Date: 20210129

Claims (18)

1 . A system comprising: one or more processors; and one or more processor-readable media storing instructions which, when executed by the one or more processors, cause performance of: obtaining sensor data from a glucose sensor inserted into subcutaneous tissue of a user; identifying a sensor data error pattern in real time; identifying an erroneous sensor use condition from two or more different erroneous sensor use conditions that each could cause the sensor data error pattern using a machine learning model; in response to identifying the erroneous sensor use condition, predicting, using the machine learning model, a resolution from a plurality of possible resolutions that is most effective for resolving the erroneous sensor use condition, wherein the machine learning model was trained with training data comprising pairs of erroneous sensor use conditions and corresponding resolutions, wherein predicting the resolution from the plurality of possible resolutions comprises identifying a possible resolution that is most effective for resolving the erroneous sensor use condition among the plurality of possible resolutions as the predicted resolution; causing implementation of the resolution; causing delivery of insulin to the user by an insulin delivery device based on the sensor data manipulated by the implementation of the resolution; and updating a configuration of the machine learning model based on the predicted resolution, an assessment of the predicted resolution, and reference feedback information, wherein the updated configuration is used for future predictions.
2 . The system of claim 1 , wherein the one or more processor-readable media further store instructions which, when executed by the one or more processors, cause performance of: blanking display of the sensor data in response to identifying the sensor data error pattern by the machine learning model.
3 . The system of claim 1 , wherein the one or more processor-readable media further store instructions which, when executed by the one or more processors, cause performance of: generating an alert informing of the erroneous sensor use condition in response to identifying the erroneous sensor use condition using the machine learning model.
4 . The system of claim 1 , wherein the implementation of the resolution comprises: implementing the resolution by manipulating the sensor data via a user interface of a sensor device.
5 . The system of claim 1 , wherein: the sensor data includes impedance spectroscopy signals measured at a plurality of signal frequencies using the glucose sensor; and applying the machine learning model comprises generating, from the sensor data that includes the impedance spectroscopy signals, a multi-dimensional feature input to the machine learning model.
6 . The system of claim 1 , wherein the one or more processor-readable media further store instructions which, when executed by the one or more processors, cause performance of: obtaining input indicating context information relating to the sensor data; and applying the machine learning model to the context information relating to the sensor data, wherein outputs of the machine learning model are further based on the context information relating to the sensor data.
7 . The system of claim 6 , wherein the context information relating to the sensor data includes historic information relating to the sensor data over a time period.
8 . The system of claim 1 , wherein the machine learning model is trained based on clinical data on the erroneous sensor use condition.
9 . A processor-implemented method comprising: obtaining sensor data from a glucose sensor inserted into subcutaneous tissue of a user; identifying a sensor data error pattern in real time; identifying an erroneous sensor use condition from two or more different erroneous sensor use conditions that each could cause the sensor data error pattern using a machine learning model; in response to identifying the erroneous sensor use condition, predicting, using the machine learning model, a resolution from a plurality of possible resolutions that is most effective for resolving the erroneous sensor use condition, wherein the machine learning model was trained with training data comprising pairs of erroneous sensor use conditions and corresponding resolutions, wherein predicting the resolution from the plurality of possible resolutions comprises identifying a possible resolution that is most effective for resolving the erroneous sensor use condition among the plurality of possible resolutions as the predicted resolution; causing implementation of the resolution; causing delivery of insulin to the user by an insulin delivery device based on the sensor data manipulated by the implementation of the resolution; and updating a configuration of the machine learning model based on the predicted resolution, an assessment of the predicted resolution, and reference feedback information, wherein the updated configuration is used for future predictions.
10 . The processor-implemented method of claim 9 , wherein the resolution comprises at least one of: adjusting a signal of the sensor data; adding a filter to a signal of the sensor data, adjusting a filter of a signal of the sensor data, replacing a signal of the sensor data; removing a signal from the sensor data; or replacing the glucose sensor with a replacement glucose sensor.
11 . The processor-implemented method of claim 9 , further comprising generating an alert informing of the erroneous sensor use condition in response to identifying the erroneous sensor use condition using the machine learning model.
12 . The processor-implemented method of claim 9 , wherein the implementation of the resolution comprises implementing the resolution by manipulating the sensor data via a user interface of a sensor device.
13 . The processor-implemented method of claim 9 , wherein: the sensor data includes impedance spectroscopy signals measured at a plurality of signal frequencies using the glucose sensor; and applying the machine learning model comprises generating, from the sensor data that includes the impedance spectroscopy signals, a multi-dimensional feature input to the machine learning model.
14 . The processor-implemented method of claim 9 , further comprising: obtaining input indicating context information relating to the sensor data; and applying the machine learning model to the context information relating to the sensor data, wherein outputs of the machine learning model are further based on the context information relating to the sensor data.
15 . The processor-implemented method of claim 14 , wherein the context information relating to the sensor data includes historic information relating to the sensor data over a time period.
16 . The processor-implemented method of claim 9 , wherein the machine learning model is trained based on clinical data on the erroneous sensor use condition.
17 . A processor-implemented method comprising: obtaining sensor data from a glucose sensor inserted into subcutaneous tissue of a user, wherein the glucose sensor includes a single working electrode; identifying a sensor data error pattern in real time; identifying an erroneous sensor use condition from two or more different erroneous sensor use conditions that each could cause the sensor data error pattern using a machine learning model; in response to identifying the erroneous sensor use condition, predicting, using the machine learning model, a resolution from a plurality of possible resolutions that is most effective for resolving the erroneous sensor use condition, wherein the machine learning model was trained with training data comprising pairs of erroneous sensor use conditions and corresponding resolutions, wherein predicting the resolution from the plurality of possible resolutions comprises identifying a possible resolution that is most effective for resolving the erroneous sensor use condition among the plurality of possible resolutions as the predicted resolution; causing implementation of the resolution; causing delivery of insulin to the user by an insulin delivery device based on the sensor data manipulated by the implementation of the resolution; and updating a configuration of the machine learning model based on the predicted resolution, an assessment of the predicted resolution, and reference feedback information, wherein the updated configuration is used for future predictions.
18 . The processor-implemented method of claim 17 , further comprising: obtaining input indicating context information relating to the sensor data, wherein the sensor data error pattern comprises a low signal error pattern, and wherein the context information comprises historic sensor data over a time period that is used to disambiguate between the two or more different erroneous sensor use conditions comprising temporary signal loss and sensitivity loss.

Description

RELATED APPLICATION DATA This application is a continuation of U.S. patent application Ser. No. 17/121,624, filed Dec. 14, 2020, which is incorporated herein by reference in its entirety. FIELD The present technology is generally related to sensor technology, including sensors used for sensing a variety of physiological parameters, e.g., glucose concentration. BACKGROUND Over the years, a variety of sensors have been developed for detecting and/or quantifying specific agents or compositions in a patient's blood, which enable patients and medical personnel to monitor physiological conditions within the patient's body. Illustratively, subjects may wish to monitor blood glucose levels in a subject's body on a continuing basis. Thus, glucose sensors have been developed for use in obtaining an indication of blood glucose levels in a diabetic patient. Such readings are useful in monitoring and/or adjusting a treatment regimen which typically includes the regular administration of insulin to the patient. Presently, a patient can measure his/her blood glucose (“BG”) using a BG measurement device (i.e., glucose meter), such as a test strip meter, a continuous glucose measurement system (or a continuous glucose monitor), or a hospital BG test. BG measurement devices use various methods to measure the BG level of a patient, such as a sample of the patient's blood, a sensor in contact with a bodily fluid, an optical sensor, an enzymatic sensor, or a fluorescent sensor. When the BG measurement device has generated a BG measurement, the measurement is displayed on the BG measurement device. SUMMARY Current continuous glucose monitoring (“CGM”) systems use CGM calibration algorithms to determine when measurements are accurate. For example, accuracy may change based on wear, battery life, and other factors. Current CGM systems determine accuracy based on an independent input signal, whereas systems and methods described herein utilize a multi-dimensional input signal for determining accuracy. The multi-dimensional input signal may include an Interstitial Current Signal (“Isig”), an Electrochemical Impedance Spectroscopy Signal (“EIS”), a counter voltage (“Vcntr”), and/or other signals. The multi-dimensional input signal improves CGM performance, leading to more reliable determinations of accuracy, wear, battery life, and other factors and providing a user with more accurate data. Though the use of a multi-dimensional input signal improves CGM performance, present CGM requirements are difficult to scale with the multi-dimensional inputs. Government agencies (e.g., the Federal Drug Administration (“FDA”)) impose restrictions and requirements for the sensitivity and accuracy of CGMs. For example, CGM devices are required to meet numerous criteria (e.g., FDA's integrated continuous glucose monitoring (“iCGM”) criteria) in order for the sensor data to be considered accurate enough to qualify for preferential treatment during regulatory the regulatory approval process. The iCGM criteria is designed to characterize the relative distribution of error of a CGM system to balance the often-competing needs of an overall mean error value and short tails in the error distribution. In addition, current systems used as predicate devices in fashioning the iCGM criteria are based on single signal analysis, so the criteria may be difficult to generalize to systems based on multi-dimensional input signals as described herein. To solve these problems, methods, systems, and devices described herein may train a machine learning model to classify multi-dimensional input signals in accordance with the iCGM criteria. The outputs from the trained machine learning model may be used to blank (i.e., remove, ignore) measurements, during computation within a glucose estimation device, which do not meet the iCGM criteria. Thus, the methods, systems, and devices described herein allow for improved CGM techniques that are compatible with the FDA's iCGM criteria and related criteria designed to balance gross measures of accuracy with error distributions featuring shorter tail. More particularly, the methods, systems, and devices describe training a machine learning model to identify outlier measurements based on behavior signatures and informed by the iCGM and similar criteria. The machine learning model may take as inputs multi-dimensional CGM sensor data and may use training data to set model parameters. The training data may include clinical data on iCGM performance. In some embodiments, the system may classify the training data according to known classifications (e.g., large negative bias, large positive bias, nominal accuracy, poor accuracy, intermediate accuracy, good accuracy, or other classifications). The system may receive multi-dimensional CGM sensor data from a sensor electrode or another computing device and may input the sensor data into the machine learning model it contains or contained in another computing device. Outputs from the machine le