Computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring, computer system, computer program product, and continuous glucose monitoring system

Abstract

The present disclosure refers to a computer implemented method for detecting an operation status of an analyte sensor (10) for continuous monitoring an analyte, comprising: receiving continuous monitoring data for an analyte detected by an analyte sensor (10); detecting a loss of sensitivity for the analyte sensor (10), comprising determining a sensitivity indicator indicative of a sensor sensitivity of the analyte sensor (10) from the continuous monitoring data, providing a critical measure for the sensitivity indicator, and determining the loss of sensitivity for the analyte sensor (10), if the sensitivity indicator matches the critical measure for the sensitivity indicator; and detecting a defect status for the analyte sensor (10), comprising receiving an impedance value indicative of an impedance measured for the analyte sensor (10), providing a critical measure for the impedance of the analyte sensor (10), and determining the defect status for the analyte sensor (10), if the impedance value matches the critical measure for the impedance of the analyte sensor (10). A failure of the analyte sensor (10) is detected if for the analyte sensor (10) at least one of the loss of sensitivity and the defect status is determined, and, in response to detecting the failure of the analyte sensor (10), at least one of providing failure data indicative of the failure of the analyte sensor (10) and deactivating the analyte sensor (10) is conducted. Further, a computer system, a computer program product, and a continuous glucose monitoring system are provided.

Description

[0001] Computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring, computer system, computer program product, and continuous glucose monitoring system

[0002] The present disclosure refers to a computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring, a computer system, a computer program product, and a continuous glucose monitoring system.

[0003] Background

[0004] Analyte sensors are applied for detecting characteristic for an analyte, such as a level of analyte. For example, an analyte sensor may be applied for continuously monitoring a glucose level for a patient, such as diabetes patient. For reliably monitoring the analyte, correct function of the analyte sensor is important. Correct function of the analyte sensor may be checked by determining an operation status of or operation condition for the analyte sensor during operation of the analyte sensor for continuously monitoring the analyte.

[0005] Document EP 3 422 222 A1 refers to a method for detecting an operation status for a sensor, the method comprising: receiving continuous monitoring data related to an operation of a sensor; providing a trained learning algorithm for detecting an operation status for the sensor which signifies a sensor function, wherein the learning algorithm is trained according to a training data set comprising historical data; detecting an operation status for the sensor by analyzing the continuous monitoring data with the trained learning algorithm; and providing output data indicating the detected operation status for the sensor. Further, a state machine system is provided, the state machine having one or more processors configured for data processing and for performing the method for detecting an operation status for a sensor.

[0006] Document US 20141 0182350 A1 discloses a system and method for processing sensor data and end of life detection of a CGM (continuous glucose monitoring) sensor. The method for determining the end of life of a continuous analyte sensor includes evaluating a plurality of risk factors using an end of life function to determine an end of life status of the sensor and providing an output related to the end of life status of the sensor. The plurality of risk factors is selected from a list including the number of days the sensor has been in use, whether there has been a decrease in signal sensitivity, whether there is a predetermined noise pattern, whether there is a predetermined oxygen concentration pattern, and error between reference BG (blood glucose) values and EGV (Estimated Glucose Value) sensor values. Referring to Document US 2019 I 0274604 A1 , it Is proposed to calculate a single, optimal, fused sensor glucose value based on respective sensor glucose values of a plurality of redundant working electrodes (WEs) of a glucose sensor. Respective electrochemical impedance spectroscopy (EIS) procedures are performed for each of the WEs to obtain values of membrane resistance (Rmem) for each WE. A noise value and a calibration factor (CF) value are calculated for each WE, and respective fusion weights are calculated for Rmem, noise, and CF for each WE. An overall fusion weight is then calculated based on the WE's Rmem fusion weight, noise fusion weight, and CF fusion weight, such that a single, optimal, fused sensor glucose value are calculated based on the respective overall fusion weight and sensor glucose value of each of the plurality of redundant working electrodes.

[0007] Document US 2021 1 0386331 A1 discloses a method, a system, and a device for continuous glucose monitoring. The method, system, and devices describe a working electrode with a GOx sensor and a background electrode in which the background electrode has no GOx sensor. The system is then comparing the first signal and the second signal to detect ingestion of a medication by the user. The system generates a sensor glucose value based on the comparison.

[0008] Document US 202410031039 A1 refers to an apparatus, comprising: a sensor comprising an insertion portion and a contact portion, wherein the insertion portion is configured to be in contact with a biological fluid and to detect an analyte level in the biological fluid; a data storage unit; and a processing unit operatively coupled to the data storage unit and operatively coupled to the contact portion, the processing unit programmed to perform a sliding window analysis on a set of analyte sensor data, wherein the set of analyte sensor data is taken over a first time period after initialization of a sensor. In performance of the sliding window analysis, the processing unit is programmed to: extract a first sensor data characteristic for a first window of the set of analyte sensor data, wherein the first window starts at a first start time, ends at a first end time, and has a first duration less than the first time period; extract a second sensor data characteristic for a second window of the set of analyte sensor data, wherein the second window starts at a second start time after the first start time, ends at a second end time after the first end time, and has a second duration less than the first time period; and determine a probable existence of sensor current abnormalities based on the first and second sensor data characteristics. Summary

[0009] It is an object to provide a computer implemented method for detecting an operation status of an analyte sensor for continuous monitoring an analyte, a computer system, a computer program product, and a continuous glucose monitoring system which allow for improved checking or monitoring of the operation status of the analyte sensor, thereby, reliably identifying potential failure of the analyte sensor in operation.

[0010] For solving the object, a computer implemented method for detecting an operation status of an analyte sensor for continuous monitoring an analyte, a computer system, and a computer program according to claims 1 , 11 , and 12, respectively, are provided. Further, a continuous glucose monitoring system according to claim 13 is provided. Further aspects are disclosed in dependent claims.

[0011] According to one aspect, a computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring is provided, the method comprising: receiving continuous monitoring data for an analyte detected by an analyte sensor; detecting a loss of sensitivity for the analyte sensor; detecting a defect status for the analyte sensor; detecting a failure of the analyte sensor if for the analyte sensor at least one of the loss of sensitivity and the defect status is determined; and in response to detecting the failure of the analyte sensor, conducting at least one of providing failure data indicative of the failure of the analyte sensor and deactivating the analyte sensor. The detecting of the loss of sensitivity for the analyte sensor is comprising: determining a sensitivity indicator indicative of a sensor sensitivity of the analyte sensor from the continuous monitoring data, providing a critical measure for the sensitivity indicator, and determining the loss of sensitivity for the analyte sensor, if the sensitivity indicator matches the critical measure for the sensitivity indicator. The detecting of the defect status for the analyte sensor is comprising: receiving an impedance value indicative of an impedance measured for the analyte sensor; providing a critical measure for the impedance of the analyte sensor, and determining the defect status for the analyte sensor, if the impedance value matches the critical measure for the impedance of the analyte sensor.

[0012] According to another aspect, a computer system is provided, the computer system having one or more processors configured for data processing and for performing a method for detecting an operation status of an analyte sensor for continuous analyte monitoring. According to a further aspect, a computer program product is provided, the computer program product being configured to be operable to, when loaded on a computer, to perform a method for detecting an operation status of an analyte sensor for continuous analyte monitoring.

[0013] According to still another aspect, a continuous glucose monitoring system is provided, the system comprising an analyte sensor configured for continuous glucose monitoring, and the computer system having one or more processors and being configured for data processing and for performing a method for detecting an operation status of an analyte sensor for continuous analyte monitoring.

[0014] The technology proposed allows for reliably checking or monitoring an operation status of or operation condition for the analyte sensor during operation for continuously monitoring the analyte. Different parameters indicative of characteristics of the operation status of the analyte sensor, namely loss of sensitivity and defect status, are analyzed. Failure of the analyte sensor is detected if for the analyte sensor if at least one of the following is determined: (i) the loss of sensitivity and (ii) the defect status.

[0015] The critical measure for the impedance may define a threshold value for the impedance of the analyte sensor. The impedance of the analyte sensor may be matching the critical measure for the impedance, if the threshold value is matched or crossed. Measurement of the impedance may be indicative of small or large crack of the analyte sensor within one or more conductive layers of the analyte sensor.

[0016] Different types of analyte sensors are known as such. For example, an operation status for a glucose sensor operable for continuously monitoring a glucose level may be detected.

[0017] The step of detecting the loss of sensitivity for the analyte sensor may further comprise the following: determining of the sensitivity indicator comprising calculating a level of aggregated continuous monitoring data over a first predefined time period; providing of the critical measure for the sensitivity indicator comprising a first critical level of aggregated continuous monitoring data; and determining of the loss of sensitivity for the analyte sensor, if the level of aggregated continuous monitoring data matches the first critical level of aggregated continuous monitoring data. The level of aggregated continuous monitoring data, for example, may be one of a mean value of continuous monitoring data and a median value of the continuous monitoring data over the first predefined time period. For example, the first predefined time period may be a period of time or time window of 12 hours or 24 hours. The level of aggregated continuous monitoring data over the first predefined time period may be determined for a plurality of n (n > 2) first predefined time periods. The level of aggregated continuous monitoring data determined for a first predefined time period m (m < n) may define the first critical level of aggregated continuous monitoring data for one or more following first predefined time periods k (k > m; k < n). In such case, the level of aggregated continuous monitoring data determined for the first predefined time period m may also be referred to as reference level of aggregated continuous monitoring data. The first critical level of aggregated continuous monitoring data for one or more following first predefined time periods k may be defined by a level being lower than the reference level of aggregated continuous monitoring data.

[0018] Detection of a first matching between the level of aggregated continuous monitoring data and the first critical level of aggregated continuous monitoring data may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of detecting a second matching between the level of aggregated continuous monitoring data and the first critical level of aggregated continuous monitoring data, such event may provide for verification of loss of sensitivity. In response, loss of sensitivity for the analyte sensor is determined.

[0019] In the step of detecting the loss of sensitivity for the analyte sensor, the determining of the sensitivity indicator may further comprise the following: calculating a first probability of point sensitivities belonging to a distribution around an expected batch sensitivity level from the continuous monitoring data; calculating a second probability of point sensitivities belonging to a distribution around a sensitivity level being lower than the expected batch sensitivity level from the continuous monitoring data; and determining a ratio between the first probability of point sensitivities and the second probability of point sensitivities. In the step of providing the critical measure for the sensitivity indicator, a critical ratio between the first probability of point sensitivities and the second probability of point sensitivities may be provided. Further, the loss of sensitivity for the analyte sensor may be determined, if the ratio matches the critical ratio. The ratio between the first probability of point sensitivities and the second probability of point sensitivities may be matching the critical measure for the sensor sensitivity, if the ratio is equal to or lower than the critical ratio.

[0020] Detection of a first matching of the ratio with the critical ratio may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of detecting a second matching of the ratio with the critical ratio, such event may provide for verification of loss of sensitivity. In response, loss of sensitivity for the analyte sensor is determined.

[0021] The step of detecting the loss of sensitivity for the analyte sensor may further comprise: determining of the sensitivity indicator comprising calculating a level of aggregated continuous monitoring data over a second predefined time period as a moving average over time, i.e. moving along time scale during operation of the analyte sensor; providing the critical measure for the sensitivity indicator comprising a second critical level of aggregated continuous monitoring data; and determining the loss of sensitivity for the analyte sensor, if the level of aggregated continuous monitoring data matches the second critical level of aggregated continuous monitoring data.

[0022] The level of aggregated continuous monitoring data, for example, may be one of a mean value of continuous monitoring data and a median value of the continuous monitoring data over the second predefined time period. The second predefined time period which may also be referred to as (moving) time window may be different from or equal to the first predefined time period. For example, the second predefined time period may be a period of time or time window of 12 hours or 24 hours. The second critical level of aggregated continuous monitoring data may be equal to or different from the first critical level of aggregated continuous monitoring data.

[0023] With respect to the second critical level of aggregated continuous monitoring data, a moderate threshold and a critical threshold may be provided for the second critical level. A matching of the level of aggregated continuous monitoring data with the second critical level being defined by the moderate threshold may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of a matching between the level of aggregated continuous monitoring data with the second critical level being defined by the critical threshold is determined, such event may be identifying loss of sensitivity.

[0024] The detecting of loss of sensitivity for the analyte sensor may further comprise verifying the loss of sensitivity for the analyte sensor, if the loss of sensitivity is determined. With respect to such embodiment, the loss of sensitivity detected for the analyte sensor by means of applying the sensitivity indicator may provide for preliminary or provisional indication for loss of sensitivity for the analyte sensor. Such preliminary indication is to be verified in the step of verifying the loss of sensitivity. For example, one or more additional characteristics indicative of sensor sensitivity may be analyzed for verification. The step of verifying may further comprise verifying the loss of sensitivity for the analyte, if at least one of the following is determined: (i) the continuous monitoring data are indicating a monitoring value for the analyte being below a first critical threshold for a first predefined verification time period; and (ii) the level of aggregated continuous monitoring data over a predefined time period determined as moving average over time is below a second critical threshold for a second predefined verification time period.

[0025] In case of continuous glucose monitoring, for example, the loss of sensitivity may be verified, if the continuous glucose monitoring data indicate glucose values below a glucose threshold, such as about 40 mg 1 1, for a predefined monitoring time period, such as about 3 hours. Alternatively or in addition, the loss of sensitivity may be verified, if the aggregated continuous glucose monitoring data for a predefined moving time window, such as about 6 hours, exceed a predefined threshold for a predefined monitoring time period, such as about 3 hours.

[0026] The detecting of the defect status for the analyte sensor may further comprise the following: determining, from the continuous monitoring data, a signal noise characteristic; providing a critical measure for the signal noise characteristic, and determining the defect status for the analyte sensor, if the signal noise characteristic matches the critical measure for the signal noise characteristic, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor.

[0027] For this example, no defect status of the analyte sensor is detected, if at least one of the following is not determined: (i) the signal noise characteristic matches the critical measure for the signal noise characteristic, and (ii) the impedance value matches the critical measure for the impedance. The critical measure for the signal noise characteristic is identifying a signal noise characteristic categorized “defect sensor status”. The noise characteristic may be determined over a predefined noise observing time period during operation of the analyte sensor, for example, over a predefined noise observing time between about one and about ten minutes.

[0028] The critical measure for the signal noise characteristic may define a threshold for a level of noise for the signals provided by the analyte sensor. Such level of noise for the signal may be determined from the continuous monitoring data, such as continuous glucose monitoring data. The signal noise characteristic may be matching the critical measure for the signal noise characteristic, if the threshold for the level of noise is crossed. The step of detecting of the defect status for the analyte sensor may further comprise the following: determining, from the continuous monitoring data, a polarization voltage; providing a critical measure for the polarization voltage; and determining the defect status for the analyte sensor, if the polarization voltage matches the critical measure for the polarization voltage, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor.

[0029] For this example, no defect status of the analyte sensor is detected, if at least one of the following is not determined: (i) the polarization voltage matches the critical measure for the polarization voltage, and (ii) the impedance value matches the critical measure for the impedance. The critical measure for the polarization voltage is identifying a polarization voltage categorized “defect sensor status”. The polarization voltage may be determined or monitored over a predefined voltage observing time period during operation of the analyte sensor, for example, over a predefined voltage observing time between one and ten minutes.

[0030] With respect to the polarization voltage, a relation of a working electrode of the analyte sensor a counter or reference electrode of the analyte sensor may be determined. The analyte sensor may provide measurement signals identifying a current value proportional to a concentration of the analyte such as glucose concentration with a determined voltage between the working electrode and the counter electrode, such voltage is referred to as polarization voltage.

[0031] The step of receiving of the continuous monitoring data may comprise receiving continuous glucose monitoring data detected by the analyte sensor. The continuous glucose monitoring data may be detected for a patient having diabetes. Several types or embodiments of continuous glucose monitoring sensors are known as such.

[0032] In some embodiment the failure of the analyte sensor may be detected if for the analyte sensor both is determined the loss of sensitivity and the defect status. Otherwise, for such embodiment no failure of the analyte sensor is detected.

[0033] For the computer system, the computer program product, and the continuous glucose monitoring system, respectively, the aspects or embodiments disclosed for the method above may apply mutatis mutandis. In an alternative embodiment, a computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring may be provided, the method comprising: receiving continuous monitoring data for an analyte detected by an analyte sensor; and detecting a loss of sensitivity for the analyte sensor. The step of detecting the loss of sensitivity for the analyte sensor is comprising the following: determining a sensitivity indicator indicative of a sensor sensitivity of the analyte sensor from the continuous monitoring data; providing a critical measure for the sensor sensitivity; and determining the loss of sensitivity for the analyte sensor, if the sensitivity indicator matches the critical measure for the sensor sensitivity. If the loss of sensitivity is determined for the analyte sensor, a failure of the analyte sensor is detected. In response to detecting the failure of the analyte sensor, at least one of providing failure data indicative of the failure of the analyte sensor and deactivating the analyte sensor is conducted.

[0034] With respect to detecting the loss of sensitivity for the analyte sensor, if the step of determining the sensitivity indicator comprises determining the level of aggregated continuous monitoring data, a first matching between the level of aggregated continuous monitoring data and the first critical level of aggregated continuous monitoring data may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of detecting a second matching between the level of aggregated continuous monitoring data and the first critical level of aggregated continuous monitoring data, such event may provide for verification of loss of sensitivity. In response, loss of sensitivity for the analyte sensor is determined. In response, the analyte sensor may be deactivated (shut off).

[0035] With respect to detecting the loss of sensitivity for the analyte sensor, if the step of determining the sensitivity indicator comprises determining the ratio between the first probability of point sensitivities and the second probability of point sensitivities, a first matching of the ratio with the critical ratio may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of a second matching of the ratio with the critical ratio is determined, such event may provide for verification of loss of sensitivity. In response, the analyte sensor may be deactivated (shut off).

[0036] Regarding the detecting the loss of sensitivity for the analyte sensor, if the step of determining the sensitivity indicator comprises determining the ratio between the first probability of point sensitivities and the second probability of point sensitivities, a moderate threshold and a critical threshold may be provided for the critical ratio. A matching of the ratio with the critical ratio being defined by the moderate threshold may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of a matching of the ratio with the critical ratio being defined by the critical threshold is determined, such event may provide for indication of loss of sensitivity. In response, the analyte sensor may be deactivated (shut off).

[0037] In another alternative embodiment, a computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring may be provided, the method comprising: receiving continuous monitoring data for an analyte detected by an analyte sensor; and detecting a defect status for the analyte sensor. The step of detecting the defect status for the analyte sensor is comprising the following: receiving an impedance value indicative of an impedance measured for the analyte sensor; providing a critical measure for the impedance of the analyte sensor; and determining the defect status for the analyte sensor, if the impedance value matches the critical measure for the impedance of the analyte sensor. If the defect status is determined for the analyte sensor, a failure of the analyte sensor is detected. In response to detecting the failure of the analyte sensor, at least one of providing failure data indicative of the failure of the analyte sensor and deactivating the analyte sensor is conducted.

[0038] With respect to the second critical level of aggregated continuous monitoring data, a moderate threshold and a critical threshold may be provided for the second critical level. A matching of the level of aggregated continuous monitoring data with the second critical level being defined by the moderate threshold may identify an indication for possible loss of sensitivity. In response, operation of the analyte sensor may continue. In case of a matching between the level of aggregated continuous monitoring data with the second critical level being defined by the critical threshold is determined, such event may provide for indication of loss of sensitivity. In response, the analyte sensor may be deactivated (shut off).

[0039] The detecting of the defect status for the analyte sensor may further comprise the following: determining, from the continuous monitoring data, a signal noise characteristic; providing a critical measure for the signal noise characteristic, and determining the defect status for the analyte sensor, if the signal noise characteristic matches the critical measure for the signal noise characteristic, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor.

[0040] For this example, no defect status of the analyte sensor is detected, if at least one of the following is not determined: (i) the signal noise characteristic matches the critical measure for the signal noise characteristic, and (ii) the impedance value matches the critical measure for the impedance. The critical measure for the signal noise characteristic is identifying a signal noise characteristic categorized “defect sensor status”. The noise characteristic may be determined over a predefined noise observing time period during operation of the analyte sensor, for example, over a predefined noise observing time between about one and about ten minutes.

[0041] The critical measure for the signal noise characteristic may define a threshold for a level of noise for the signals provided by the analyte sensor. Such level of noise for the signal may be determined from the continuous monitoring data, such as continuous glucose monitoring data. The signal noise characteristic may be matching the critical measure for the signal noise characteristic, if the threshold for the level of noise is crossed.

[0042] The step of detecting of the defect status for the analyte sensor may further comprise the following: determining, from the continuous monitoring data, a polarization voltage; providing a critical measure for the polarization voltage; and determining the defect status for the analyte sensor, if the polarization voltage matches the critical measure for the polarization voltage, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor.

[0043] For this example, no defect status of the analyte sensor is detected, if at least one of the following is not determined: (i) the polarization voltage matches the critical measure for the polarization voltage, and (ii) the impedance value matches the critical measure for the impedance. The critical measure for the polarization voltage is identifying a polarization voltage categorized “defect sensor status”. The polarization voltage may be determined or monitored over a predefined voltage observing time period during operation of the analyte sensor, for example, over a predefined voltage observing time between one and ten minutes.

[0044] With respect to the polarization voltage, a relation of a working electrode of the analyte sensor a counter or reference electrode of the analyte sensor may be determined.

[0045] With respect to the alternative embodiments, regarding the step of detecting the loss of sensitivity for the analyte sensor and I or the step of detecting the defect status for the analyte sensor the aspects or embodiments disclosed above may apply mutatis mutandis. Similarly, a computer system, a computer program product, and I or A continuous glucose monitoring system may be provided with a configuration for performing at least one of the methods according to the alternative embodiments. of embodiments

[0046] Following further embodiments are described with reference to figures. In the figures show: Fig. 1 a schematic representation of a system for continuous analyte monitoring such as continuous glucose monitoring;

[0047] Fig. 2 a schematic block diagram for a computer implemented method for detecting an operation status of an analyte sensor for continuous analyte monitoring;

[0048] Fig. 3 a schematic representation of continuous glucose monitoring data detected over a time period;

[0049] Fig. 4 a schematic representation of continuous glucose monitoring data detected over time period;

[0050] Fig. 5 a graphical representation depicting probability density functions of two (Gaussian) distributions in sensor sensitivity; and

[0051] Fig. 6 a schematic representation of continuous glucose monitoring data detected over time period.

[0052] Fig. 1 shows a schematic representation of an arrangement comprising an analyte sensor 10 configured for continuous analyte monitoring of an analyte. For example, the analyte sensor 10 may be a glucose sensor configured for continuous glucose monitoring.

[0053] In operation, the analyte sensor 10 is continuously detecting measurement data for the analyte such as glucose level which are provided as continuous monitoring data, for example, continuous glucose monitoring data. The continuous monitoring data may be provided to a computer system 11 by means of wireless data communication 12. Alternatively, there may be a wired or cable-based connection between the analyte sensor 10 and the computer system 11 , at least for data communication. The computer system 11 may be one of a mobile computer such as mobile phone or laptop computer, server computer system, and desktop computer system. The computer system 11 is provided with one or more processors configured for data processing. Suitable software applications may be implemented on the computer system 11 for data processing such as processing the continuous monitoring data received from the analyte sensor 10. Wireless data communication 12 may apply data communication over the world wide web. The computer system 11 may be configured for data exchange with other computer systems (not shown). Fig. 2 shows a schematic block diagram for a method for detecting an operation status of the analyte sensor 10 in operation. Specifically, the operation status is indicative of whether the analyte sensor 10 is functioning correctly or not while continuously monitoring the analyte such as glucose for a diabetes patient. Monitoring the operation status may be aiming at detecting failure of the analyte sensor, thereby, ensuring correct monitoring by the analyte sensor 10 in operation.

[0054] In step 20 continuous monitoring data, such as continuous glucose monitoring data, are received in the computer system 11 from the analyte sensor 10. A loss of sensitivity for the analyte sensor 10 is detected in step 21. Such detecting of the loss of sensitivity for the analyte sensor 10 comprises, in step 22, determining a sensitivity indicator indicative of a sensor sensitivity for the analyte sensor 10 from the continuous monitoring data received in the computer system 11. The sensitivity indicator provides a measure for the sensitivity of the analyte sensor 10 provided during operation. In step 23, a critical measure for the sensitivity indicator is provided in the computer system 11. The loss of sensitivity for the analyte sensor 10 is determined in step 24, if the sensitivity indicator matches the critical measure for the sensitivity indicator. For example, the sensitivity indicator determined may be matching the critical measure, if the sensitivity indicator is reaching or crossing a threshold defined by the critical measure for the sensitivity indicator.

[0055] According to Fig. 2, the method for detecting the operation status of the analyte sensor 10 further comprises a step 25 for detecting a defect status for the analyte sensor 10. Thus, in addition to detecting loss of sensitivity, the operation status of the analyte sensor 10 is analyzed with respect to the presence of a defect status for the analyte sensor 10. The detecting of the defect status comprises receiving the impedance value indicative of an impedance measured for the analyte sensor 10 in step 26. In step 27, a critical measure for the impedance of the analyte sensor 10 is provided in the computer system 11. The defect status is determined for the analyte sensor 10, if the impedance value matches the critical measure for the impedance of the analyte sensor 10. For example, the impedance value is matching the critical measure, if the impedance value is reaching or crossing a critical impedance defined by the critical measure.

[0056] Following, in step 28 a failure of the analyte sensor 10 is detected, if for the analyte sensor 10 both is determined: (i) loss of sensitivity, and (ii) defect status. Thus, failure of the analyte sensor 10 is detected or determined, if the analyte sensor 10 does no longer provide some required sensitivity and if the defect status is detected for the analyte sensor 10. Failure of the analyte sensor 10 will be determined more reliably.

[0057] In response to detecting the failure of the analyte sensor 10, at least one of the following is conducted in the step 29: (i) providing failure data indicative of the failure of the analyte sensor 10, and (ii) deactivating the analyte sensor 10. The failure data are provided in the computer system 11. In response, information about the failure of the analyte sensor 10 may be outputted through a user interface 13 to a user, the user interface 13 comprising, for example, at least one of a display and a speaker of the computer system 11 .

[0058] As an alternative or in addition, in response to detecting the failure of the analyte sensor 10, the analyte sensor 10 may be deactivated (shut off).

[0059] The failure of the analyte sensor 10 may not be detected or determined, if only one of the following is determined: loss of sensitivity and defect status.

[0060] With respect to detecting the loss of sensitivity for analyte sensor (step 21), different embodiments may apply. For example, the step of detecting the loss of sensitivity for the analyte sensor 10 may comprise calculating a level of aggregated continuous monitoring data over a first predefined time period, the level of aggregated continuous monitoring data providing for the sensitivity indicator. The critical measure for the sensitivity indicator may be provided with a first critical level of aggregated continuous monitoring data. The loss of sensitivity for the analyte sensor may be determined, if the level of aggregated continuous monitoring data matches the first critical level of aggregated continuous monitoring data. The level of aggregated continuous monitoring data, for example, may be one of a mean level of the continuous monitoring data and a median value of the continuous monitoring data over the first predefined time period, which, for example, may be a period or time window of 12 hours or 24 hours.

[0061] Fig. 3 shows a graphical representation of continuous glucose monitoring data 30 over time. A plurality of first predefined time periods 31 is shown, each first predefined time period covering 24 hours. A mean or average value 32 is determined as the level of aggregated continuous monitoring data for each first predefined time period 31 . For the example shown, for following first predefined time periods the mean value 32 is below a threshold level defined by the first critical level of aggregate continuous monitoring data starting with a first predefined time period indicated by reference numeral 33 in Fig. 3. In response, loss of sensitivity is detected for the analyte sensor 10. As an alternative or in addition, the step of detecting of loss of sensitivity for the analyte sensor 10 may further comprise calculating a level of aggregated continuous monitoring data over a second predefined time period which is moving a long time scaled during operation of the analyte sensor 10. The second predefined time period may be equal to or different from the first predefined time period. For example, the second predefined time period may be 24 hours. The level of aggregated continuous monitoring data being, for example, an average I mean level of continuous monitoring data and / or a median value of the continuous monitoring data over the second predefined time period, is calculated for the continuous monitoring data within the second predefined time period moving along the time scale. Thereby, a moving level of aggregated continuous monitoring data is determined in operation of the analyte sensor 10.

[0062] Fig. 4 shows a graphical representation for continuous glucose monitoring data 40 over a time scale of 6 days. In addition, spot-monitoring (non-continuous) glucose data 41 detected for a glucose level during the 6 days are depicted. An additional curve 42 represents an average or mean level of the continuous glucose monitoring data 40 for a moving time window of 24 hours.

[0063] For determining curve 42, a portion of low continuous glucose monitoring data may progressively emphasized, wherein a glucose value may be categorized “low continuous glucose monitoring data” if the value is below a specific threshold, for example, below 100 mg / dl. The low continuous glucose monitoring data being below the specific threshold are weighted non- equally (differently). E.g., a glucose value of 70 mg I dl is processed to contribute more to an aggregated score value than a glucose value of 90 mg I dl. By such processing of the continuous glucose monitoring data, glucose values categorized “low continuous glucose monitoring data” are progressively emphasized.

[0064] Referring to Fig. 5, the step of detecting the loss of sensitivity for the analyte sensor 10, alternatively or in addition, may comprise calculating a first probability of point sensitivities belonging to a distribution around an expected batch sensitivity level, and calculating a second probability of point sensitivities belonging to a distribution around a sensitivity level being lower than the expected batch sensitivity level from the continuous monitoring data.

[0065] Fig. 5 shows a graphical representation depicting probability density functions of two (Gaussian) distributions. A first Gaussian distribution 50 refers to a distribution of sensor sensitivity around an expected batch or lot sensitivity 51 of analyte sensors. As second Gaussian distribution 52 refers to a distribution of sensitivity around a reduced sensor sensitivity 53. A probability value of a point sensitivity determined from continuous monitoring data can be estimated (illustrated by a dot-dashed line 54 and a dashed line 55) for belonging to each of the first and second distributions 50, 52. Finally, the ratio of these point probabilities (e.g. for a point sensitivity estimate of 0.195 (dashed line, 55) the point probability that this values belongs to the first distribution 50 is close to 1 , while the probability that it belongs to the second distribution 52 is <0.5) can be analyzed. The expected batch sensor sensitivity 51 is deduced from in-vitro measurements of samples from the lot or batch. In some embodiment, a failure of the analyte sensor in operation may be determined, for example, if point sensitivity is found to belong more likely to the second distribution 52 than to the first distribution 50.

[0066] Fig. 6 shows a graphical representation of continuous glucose monitoring data 60 detected over a time period of 24 hours for Day 1 and a different day, i.e. Day 2. Further, spot-monitoring glucose data 61 detected by spot-monitoring of glucose level over the time period of 24 hours are depicted.

[0067] In addition, point sensitivities 62 determined from the continuous glucose monitoring data 60 are depicted. A further curve refers to the daily mean sensor sensitivity 63, and a range band 64 shows the mean plus I minus the standard deviation for the daily mean sensor sensitivity 63 which is lower on Day 2 compared to Day 1.

[0068] Further, a batch sensitivity 65 is depicted. Also, a batch sensitivity range 66 is shown. On Day 1 the point sensitivities 62 are (mostly) close to the batch sensitivity 65 and (mostly) within the batch sensitivity range 66. On Day 2, the point sensitivities 62 are noticeably reduced and with higher number outside the batch sensitivity range 66 (even though a spread of the point sensitivities 62 is still rather small). Because of such finding, a failure of the analyte sensor 10 in operation may be determined.

Claims

Claims1. A computer implemented method for detecting an operation status of an analyte sensor (10) for continuous monitoring an analyte, comprising:- receiving continuous monitoring data for an analyte detected by an analyte sensor (10);- detecting a loss of sensitivity for the analyte sensor (10), comprising- determining a sensitivity indicator indicative of a sensor sensitivity of the analyte sensor (10) from the continuous monitoring data,- providing a critical measure for the sensitivity indicator, and- determining the loss of sensitivity for the analyte sensor (10), if the sensitivity indicator matches the critical measure for the sensitivity indicator;- detecting a defect status for the analyte sensor (10), comprising- receiving an impedance value indicative of an impedance measured for the analyte sensor (10),- providing a critical measure for the impedance of the analyte sensor (10), and- determining the defect status for the analyte sensor (10), if the impedance value matches the critical measure for the impedance of the analyte sensor (10);- detecting a failure of the analyte sensor (10) if for the analyte sensor (10) at least one of the loss of sensitivity and the defect status is determined; and- in response to detecting the failure of the analyte sensor (10), conducting at least one of providing failure data indicative of the failure of the analyte sensor (10) and deactivating the analyte sensor (10).

2. Method of claim 1 , wherein the detecting of the loss of sensitivity for the analyte sensor (10) is further comprising:- determining of the sensitivity indicator comprising calculating a level of aggregated continuous monitoring data over a first predefined time period;- providing of the critical measure for the sensitivity indicator comprising a first critical level of aggregated continuous monitoring data; and- determining of the loss of sensitivity for the analyte sensor (10), if the level of aggregated continuous monitoring data matches the first critical level of aggregated continuous monitoring data.

3. Method of claim 1 or 2, wherein the detecting of the loss of sensitivity for the analyte sensor (10) is further comprising:- determining the sensitivity indicator comprising- calculating a first probability of point sensitivities belonging to a distribution around an expected batch sensitivity level from the continuous monitoring data;- calculating a second probability of point sensitivities belonging to a distribution around a sensitivity level being lower than the expected batch sensitivity level from the continuous monitoring data; and- determining a ratio between the first probability of point sensitivities and the second probability of point sensitivities;- providing of the critical measure for the sensitivity indicator comprising a critical ratio between the first probability of point sensitivities and the second probability of point sensitivities; and- determining of the loss of sensitivity for the analyte sensor (10), if the ratio matches the critical ratio.

4. Method of at least one of the preceding claims, wherein the detecting of the loss of sensitivity for the analyte sensor (10) is further comprising:- determining of the sensitivity indicator comprising calculating a level of aggregated continuous monitoring data over a second predefined time period as a moving average over time;- providing of the critical measure for the sensitivity indicator comprising a second critical level of aggregated continuous monitoring data; and- determining of the loss of sensitivity for the analyte sensor (10), if the level of aggregated continuous monitoring data matches the second critical level of aggregated continuous monitoring data.

5. Method of at least one of the preceding claims, wherein the detecting of loss of sensitivity for the analyte sensor (10) is further comprising verifying the loss of sensitivity for the analyte sensor (10), if the loss of sensitivity is determined.

6. Method of claim 5, wherein the verifying is further comprising verifying the loss of sensitivity for the analyte, if at least one of the following is determined:- the continuous monitoring data are indicating a monitoring value for the analyte being below a first critical threshold for a first predefined verification time period; and- the level of aggregated continuous monitoring data over a predefined time period determined as moving average over time is below a second critical threshold for a second predefined verification time period.

7. Method of at least one of the preceding claims, wherein the detecting of the defect status for the analyte sensor (10) is further comprising- determining, from the continuous monitoring data, a signal noise characteristic;- providing a critical measure for the signal noise characteristic, and- determining of the defect status for the analyte sensor (10), if the signal noise characteristic matches the critical measure for the signal noise characteristic, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor (10).

8. Method of at least one of the preceding claims, wherein the detecting of the defect status for the analyte sensor (10) is further comprising- determining, from the continuous monitoring data, a polarization voltage;- providing a critical measure for the polarization voltage; and- determining of the defect status for the analyte sensor (10), if the polarization voltage matches the critical measure for the polarization voltage, in addition to the impedance value matching the critical measure for the impedance of the analyte sensor (10).

9. Method of at least one of the preceding claims, wherein the receiving of the continuous monitoring data is comprising receiving continuous glucose monitoring data detected by the analyte sensor (10).

10. Method of at least one of the preceding claims, wherein the failure of the analyte sensor (10) is detected if for the analyte sensor (10) both is determined the loss of sensitivity and the defect status.11 . A computer system (11), having one or more processors configured for data processing and for performing a method for detecting an operation status of an analyte sensor (10) for continuous monitoring an analyte of at least one of the preceding claims.

12. Computer program product, configured to be operable to, when loaded on a computer, to perform a method for detecting an operation status of an analyte sensor (10) for continuous monitoring an analyte of at least one of the claims 1 to 10.

13. A continuous glucose monitoring system, comprising an analyte sensor (10) configured for continuous glucose monitoring and a computer system (11) of claim 11.

Images