System for calibrating a glucose sensor and glucose detection system - Patents.com

Abstract

To provide systems and methods for processing sensor data and self-calibration.SOLUTION: Systems and methods are provided which are capable of calibrating a continuous analyte sensor based on an initial sensitivity, and then continuously performing self-calibration without using, or with reduced use of, reference measurements. In specific embodiments, a sensitivity (110) of the analyte sensor is determined by applying an estimative algorithm (120) that is a function of certain parameters. A sensor property can be used to compensate sensor data for sensitivity drift, or determine another property associated with the sensor, such as temperature, sensor membrane damage, moisture ingress into sensor electronics, and scaling factors.SELECTED DRAWING: Figure 1A

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 61 / 476,145, filed April 15, 2011. The above-referenced application is hereby incorporated by reference in its entirety and expressly made a part of this specification.

[0002] SUMMARY OF THE DISCLOSURE The embodiments described herein generally relate to systems and methods for processing and automatically calibrating sensor data from continuous analyte sensors. [Background technology]

[0003] Diabetes mellitus is a chronic disease that occurs when the pancreas does not produce enough insulin (Type I) or when the body cannot effectively use the insulin that the pancreas does produce (Type II). This condition typically leads to increased glucose concentrations in the blood (hyperglycemia), which can result in a number of physiological disorders related to the deterioration of small blood vessels (e.g., kidney failure, skin ulcers, or bleeding in the vitreous of the eye). Sometimes, a hypoglycemic reaction (low blood sugar) is precipitated by an inadvertent overdose of insulin, or after a regular dose of insulin or glucose-lowering medication accompanied by excessive exercise or inadequate food intake.

[0004] A variety of sensor devices have been developed to continuously measure blood glucose concentrations. Traditionally, diabetics carry self-monitoring blood glucose (SMBG) monitors, which typically involve uncomfortable finger cell prick procedures. Lack of comfort and convenience often results in a diabetic measuring his or her glucose levels two to four times per day. Unfortunately, these measurements can be spaced too far apart, sometimes causing the diabetic to learn of hypoglycemic or hyperglycemic events too late, thereby incurring potentially dangerous side effects. In fact, not only are diabetics unlikely to make timely SMBG measurements, but even if the diabetic is able to obtain timely SMBG values, based on SMBG alone the diabetic cannot tell if his or her blood glucose levels are increasing or decreasing. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] US Patent Application Publication No. 2007 / 0299617 Summary of the Invention [Problem to be solved by the invention]

[0006] In the past, various glucose sensors have been developed to continuously measure glucose levels. Many implantable glucose sensors suffer from internal complications and provide short-term and inaccurate blood glucose measurements. Similarly, transcutaneous sensors have suffered from problems in accurately detecting and continuously reporting glucose levels over long periods of time. Although some efforts have been made to obtain blood glucose data from implantable devices and to retrospectively determine blood glucose trends for analysis, these efforts do not assist diabetics in determining real-time blood glucose information. Some efforts have also been made to obtain blood glucose data from transcutaneous devices for prospective data analysis, but similar problems arise. [Means for solving the problem]

[0007] In a first aspect, a method is provided for calibrating sensor data generated by a continuous analyte sensor, the method including: generating the sensor data using the continuous analyte sensor; iteratively determining, by an electronic device, a sensitivity value of the continuous analyte sensor as a function of time by applying a priori information including sensor sensitivity information; and calibrating the sensor data based at least in part on the determined sensitivity value.

[0008] In one embodiment of the first aspect or any other embodiment thereof, calibrating the sensor data is performed repeatedly throughout substantially an entire sensor session.

[0009] In one embodiment of the first aspect, or any other embodiment thereof, the iteratively determining the sensitivity values ​​is performed periodically or at irregular intervals as determined by the a priori information.

[0010] In one embodiment of the first aspect, or any other embodiment thereof, the iteratively determining the sensitivity value is performed substantially throughout the entire sensor session.

[0011] In one embodiment of the first aspect, or any other embodiment thereof, determining the sensitivity value is performed substantially in real time.

[0012] In one embodiment of the first aspect or any other embodiment thereof, the a priori information relates to at least one predetermined sensitivity value associated with a predetermined time after a sensor session begins.

[0013] In one embodiment of the first aspect, or any other embodiment thereof, the at least one predetermined sensitivity value relates to a correlation between sensitivity determined from in vitro analyte concentration measurements and sensitivity determined from in vivo analyte concentration measurements at a predetermined time.

[0014] In one embodiment of the first aspect or any other embodiment thereof, the a priori information relates to a predetermined sensitivity function that uses time as an input.

[0015] In one embodiment of the first aspect, or any other embodiment thereof, the time corresponds to a time after the sensor session begins.

[0016] In one embodiment of the first aspect, or any other embodiment thereof, the time corresponds to at least one of a time of manufacture or a time since manufacture.

[0017] In one embodiment of the first aspect or any other embodiment thereof, the sensitivity value of the continuous analyte sensor is also a function of at least one other parameter.

[0018] In one embodiment of the first aspect or any other embodiment, the at least one other parameter is selected from the group consisting of temperature, pH, hydration level or duration, curing conditions, analyte concentration of a fluid surrounding the continuous analyte sensor while the sensor is started up, and combinations thereof.

[0019] In one embodiment of the first aspect, or other embodiments thereof, calibrating the sensor data is performed without the use of reference blood glucose data.

[0020] In one embodiment of the first aspect or any other embodiment, the electronic device is configured to provide a level of accuracy corresponding to a mean absolute relative difference of not more than about 10% over a sensor session of at least about 3 days, and a reference measurement related to calculation of the mean absolute relative difference is determined by analysis of blood.

[0021] In one embodiment of the first aspect or any other embodiment thereof, the sensor session is at least about 4 days.

[0022] In one embodiment of the first aspect or any other embodiment thereof, the sensor session is at least about 5 days.

[0023] In one embodiment of the first aspect or any other embodiment thereof, the sensor session is at least about 6 days.

[0024] In one embodiment of the first aspect or any other embodiment thereof, the sensor session is at least about 7 days.

[0025] In one embodiment of the first aspect or any other embodiment thereof, the sensor session is at least about 10 days.

[0026] In one embodiment of the first aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 7% over the sensor session.

[0027] In one embodiment of the first aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 5% over the sensor session.

[0028] In one embodiment of the first aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 3% over the sensor session.

[0029] In one embodiment of the first aspect or any other embodiment thereof, the a priori information relates to a calibration code.

[0030] In one embodiment of the first aspect, or any other embodiment thereof, the a priori sensitivity information is stored in sensor electronics prior to use of the sensor.

[0031] In a second aspect, there is provided a system for carrying out the method of the first aspect or any embodiment thereof.

[0032] In a third aspect, a method of calibrating sensor data generated by a continuous analyte sensor is provided, the method including generating sensor data using the continuous analyte sensor, determining by an electronic device a plurality of distinct sensitivity values ​​for the continuous analyte sensor as a function of time and as a function of sensitivity information related to a priori information, and calibrating the sensor data based at least in part on at least one of the plurality of distinct sensitivity values.

[0033] In one embodiment of the third aspect, or other embodiments thereof, calibrating the continuous analyte sensor is performed repeatedly throughout substantially the entire sensor session.

[0034] In one embodiment of the third aspect or any other embodiment thereof, the plurality of different sensitivity values ​​are stored in a look-up table in a computer memory.

[0035] In one embodiment of the third aspect or any other embodiment thereof, determining the different sensitivity values ​​is performed only once throughout the entire sensor session.

[0036] In one embodiment of the third aspect or any other embodiment thereof, the a priori information relates to at least one predetermined sensitivity value associated with a predetermined time after a sensor session begins.

[0037] In one embodiment of the third aspect or any other embodiment thereof, the at least one predetermined sensitivity value relates to a correlation between sensitivity determined from in vitro analyte concentration measurements and sensitivity determined from in vivo analyte concentration measurements at a given time.

[0038] In one embodiment of the third aspect or any other embodiment thereof, the a priori information relates to a predetermined sensitivity function that uses time as an input.

[0039] In one embodiment of the third aspect, or any other embodiment thereof, the time corresponds to a time after the sensor session begins.

[0040] In one embodiment of the third aspect, or any other embodiment, the time corresponds to the time at or since production.

[0041] In one embodiment of the third aspect or any other embodiment thereof, the plurality of sensitivity values ​​is also a function of at least one parameter other than time.

[0042] In one embodiment of the third aspect or any other embodiment thereof, the at least one other parameter is selected from the group consisting of temperature, pH, level or duration of hydration, curing conditions, analyte concentration of a fluid surrounding the continuous analyte sensor during activation of the sensor, and combinations thereof.

[0043] In one embodiment of the third aspect or any other embodiment thereof, calibrating the continuous analyte sensor is performed without the use of reference blood glucose data.

[0044] In one embodiment of the third aspect or any other embodiment, the electronic device is configured to provide a level of accuracy corresponding to a mean absolute relative difference of not more than about 10% over a sensor session of at least about 3 days, and a reference measurement associated with calculation of the mean absolute relative difference is determined by analysis of blood.

[0045] In one embodiment of the third aspect or any other embodiment thereof, the sensor session is at least about 4 days.

[0046] In one embodiment of the third aspect or any other embodiment thereof, the sensor session is at least about 5 days.

[0047] In one embodiment of the third aspect or any other embodiment thereof, the sensor session is at least about 6 days.

[0048] In one embodiment of the third aspect or any other embodiment thereof, the sensor session is at least about 7 days.

[0049] In one embodiment of the third aspect or any other embodiment thereof, the sensor session is at least about 10 days.

[0050] In one embodiment of the third aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 7% over the sensor session.

[0051] In one embodiment of the third aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 5% over the sensor session.

[0052] In one embodiment of the third aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 3% over the sensor session.

[0053] In one embodiment of the third aspect or any other embodiment thereof, the a priori information relates to a calibration code.

[0054] In a fourth aspect, there is provided a system for carrying out the method of the third aspect or any embodiment thereof.

[0055] In a fifth aspect, a method is provided for processing data from a continuous analyte sensor, the method comprising receiving, by an electronic device, sensor data from a continuous analyte sensor including at least one sensor data point, iteratively determining a sensitivity value of the continuous analyte sensor as a function of time and at least one predetermined sensitivity value associated with a predetermined time after initiating a sensor session, forming a conversion function based at least in part on the sensitivity values, and determining an analyte output value by applying the conversion function to the at least one sensor data point.

[0056] In one embodiment of the fifth aspect or any other embodiment thereof, the iteratively determining the sensitivity of the continuous analyte sensor is performed continuously.

[0057] In one embodiment of the fifth aspect, or any other embodiment thereof, the iteratively determining the sensitivity is performed substantially in real time.

[0058] In one embodiment of the fifth aspect or any other embodiment thereof, the method further includes determining a baseline for the continuous analyte sensor, and the conversion function is based at least in part on the baseline.

[0059] In one embodiment of the fifth aspect, or any other embodiment thereof, determining the baseline of the continuous analyte sensor is performed continuously.

[0060] In one embodiment of the fifth aspect, or any other embodiment thereof, determining the sensitivity of the continuous analyte sensor and determining the baseline of the analyte sensor are performed substantially simultaneously.

[0061] In an embodiment of the fifth aspect or any other embodiment thereof, the at least one predetermined sensitivity value is set at a manufacturing facility for the continuous analyte sensor.

[0062] In one embodiment of the fifth aspect or any other embodiment, the method further includes accepting at least one calibration code and applying the at least one calibration code to the electronic device at a predetermined time after the sensor session begins.

[0063] In one embodiment of the fifth aspect, or any other embodiment thereof, the repeatedly determining the sensitivity is performed periodically or at irregular intervals determined by the at least one calibration code.

[0064] In one embodiment of the fifth aspect or any other embodiment thereof, the at least one calibration code is associated with at least one predetermined sensitivity.

[0065] In one embodiment of the fifth aspect or any other embodiment thereof, the at least one calibration code is associated with a predetermined sensitivity function that uses the time of the time function as an input.

[0066] In one embodiment of the fifth aspect, or any other embodiment thereof, the time corresponds to a time after starting the sensor session.

[0067] In one embodiment of the fifth aspect, or any other embodiment, the time corresponds to the time at or since production.

[0068] In one embodiment of the fifth aspect or any other embodiment thereof, the sensitivity value of the continuous analyte sensor is also a function of at least one other parameter.

[0069] In one embodiment of the fifth aspect or any other embodiment thereof, the at least one other parameter is selected from the group consisting of temperature, pH, level or duration of hydration, curing conditions, analyte concentration of a fluid surrounding the continuous analyte sensor during activation of the sensor, and combinations thereof.

[0070] In a sixth aspect, there is provided a system for carrying out the method of the fifth aspect or any embodiment thereof.

[0071] In a seventh aspect, there is provided a method for calibrating a continuous analyte sensor, the method including accepting sensor data from the continuous analyte sensor, forming or accepting a predetermined sensitivity profile corresponding to a change in sensor sensitivity to an analyte substantially throughout the sensor session that is a function of at least one predetermined sensitivity value associated with a predetermined time after initiating the sensor session, and applying the sensitivity profile by an electronic device in a real-time calibration.

[0072] In one embodiment of the seventh aspect, or any other embodiment thereof, the at least one predetermined sensitivity value, the predetermined sensitivity profile, or both, is established at a manufacturing facility for the continuous analyte sensor.

[0073] In one embodiment of the seventh aspect or any other embodiment, the method includes accepting at least one calibration code and applying the at least one calibration code to the electronic device at a predetermined time after the sensor session begins.

[0074] In one embodiment of the seventh aspect or any other embodiment thereof, the at least one calibration code is associated with at least one predetermined sensitivity.

[0075] In an embodiment of the seventh aspect or any other embodiment thereof, the at least one calibration code is associated with a predetermined sensitivity function that uses time as an input.

[0076] In one embodiment of the seventh aspect or any other embodiment thereof, the sensitivity profile is a function of time.

[0077] In one embodiment of the seventh aspect, or any other embodiment thereof, the time corresponds to a time after starting the sensor session.

[0078] In one embodiment of the seventh aspect, or any other embodiment, the time corresponds to the time at or since production.

[0079] In one embodiment of the seventh aspect or any other embodiment, the sensitivity value is a function of time, a predetermined sensitivity value, and at least one parameter selected from the group consisting of temperature, pH, hydration level or duration, curing conditions, analyte concentration of a fluid surrounding the continuous analyte sensor during activation of the sensor, and combinations thereof.

[0080] In an eighth aspect, there is provided a system for carrying out the method of the seventh aspect or any embodiment thereof.

[0081] In a ninth aspect, there is provided a method of processing data from a continuous analyte sensor, the method comprising: accepting, by an electronic device, sensor data from the continuous analyte sensor including at least one sensor data point; accepting or forming a predetermined sensitivity profile corresponding to changes in sensor sensitivity over substantially an entire sensor session; forming a conversion function based at least in part on the sensitivity profile; and determining an analyte output value by applying the conversion function to the at least one sensor data point.

[0082] In one embodiment of the ninth aspect or any other embodiment thereof, the sensitivity profile is established at a manufacturing facility for the continuous analyte sensor.

[0083] In one embodiment of the ninth aspect or any other embodiment, the method includes accepting at least one calibration code and applying the at least one calibration code to the electronic device at a predetermined time after the sensor session begins.

[0084] In one embodiment of the ninth aspect or any other embodiment thereof, the at least one calibration code is associated with at least one predetermined sensitivity.

[0085] In one embodiment of the ninth aspect or any other embodiment thereof, the at least one calibration code is associated with a sensitivity profile.

[0086] In one embodiment of the ninth aspect or any other embodiment thereof, the sensitivity profile is a function of time.

[0087] In one embodiment of the ninth aspect, or any other embodiment thereof, the time corresponds to a time after starting the sensor session.

[0088] In one embodiment of the ninth aspect or any other embodiment, the time corresponds to the time at or since production.

[0089] In one embodiment of the ninth aspect or any other embodiment thereof, the sensitivity is a function of time, and the at least one parameter is selected from the group consisting of temperature, pH, hydration level or duration, curing conditions, analyte concentration of a fluid surrounding the continuous analyte sensor while the sensor is started up, and combinations thereof.

[0090] In a tenth aspect, there is provided a system for carrying out the method of the ninth aspect or any embodiment thereof.

[0091] In an eleventh aspect, a system is provided for monitoring an analyte concentration in a recipient, the system comprising a continuous analyte sensor configured to measure an analyte concentration in a recipient and provide factory-calibrated sensor data, the factory-calibrated sensor data being calibrated without the use of reference blood glucose data, the system configured to provide an accuracy level corresponding to a mean absolute relative difference of not more than about 10% over a sensor session of at least about 3 days, and a reference measurement relevant to calculation of the mean absolute relative difference is determined by analysis of blood.

[0092] In an embodiment of the eleventh aspect or any other embodiment thereof, the sensor session is at least about 4 days.

[0093] In an embodiment of the eleventh aspect or any other embodiment thereof, the sensor session is at least about 5 days.

[0094] In an embodiment of the eleventh aspect or any other embodiment thereof, the sensor session is at least about 6 days.

[0095] In an embodiment of the eleventh aspect or any other embodiment thereof, the sensor session is at least about 7 days.

[0096] In an embodiment of the eleventh aspect or any other embodiment thereof, the sensor session is at least about 10 days.

[0097] In an embodiment of the eleventh aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 7% over the sensor session.

[0098] In an embodiment of the eleventh aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 5% over the sensor session.

[0099] In an embodiment of the eleventh aspect or any other embodiment thereof, the average absolute relative difference does not exceed about 3% over the sensor session.

[0100] In a twelfth aspect, there is provided a method of determining a characteristic of a continuous analyte sensor comprising applying a bias voltage to the analyte sensor, applying a voltage step above the bias voltage to the analyte sensor, measuring a signal response of the voltage step using sensor electronics, determining a peak current of the signal response using sensor electronics, and determining a characteristic of the sensor by relating the peak current to the predetermined relationship using sensor electronics.

[0101] In one embodiment of the twelfth aspect or any other embodiment thereof, relating the peak current to the predetermined relationship includes calculating an impedance of the sensor based on the peak current and relating the sensor impedance to the predetermined relationship.

[0102] In an embodiment of the twelfth aspect or any other embodiment thereof, the characteristic of the sensor is a sensitivity of the sensor or a temperature of the sensor.

[0103] In one embodiment of the twelfth aspect or any other embodiment thereof, the peak current is the difference between the magnitude of the response prior to a voltage step and the largest measured magnitude of the response due to the voltage step.

[0104] In an embodiment of the twelfth aspect or any other embodiment thereof, the predetermined relationship is a relationship between impedance and sensor sensitivity, and the characteristic of the sensor is the sensitivity of the sensor.

[0105] In an embodiment of the twelfth aspect or any other embodiment thereof, the method further includes compensating sensor data using the determined sensor characteristics.

[0106] In one embodiment of the twelfth aspect or any other embodiment thereof, the compensating includes relating a predetermined relationship of the peak current to a sensor sensitivity or a change in sensor sensitivity, and modifying a value of the sensor data in response to the associated sensor sensitivity or change in sensor sensitivity.

[0107] In an embodiment of the twelfth aspect or any other embodiment thereof, the predetermined relationship is a linear relationship over a period of use of the analyte sensor.

[0108] In an embodiment of the twelfth aspect or any other embodiment thereof, the predetermined relationship is a non-linear relationship over a period of use of the analyte sensor.

[0109] In one embodiment of the twelfth aspect or any other embodiment thereof, the predetermined relationship is determined by pre-testing a sensor similar to the analyte sensor.

[0110] In one embodiment of the thirteenth aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the twelfth aspect or any embodiment thereof.

[0111] In a fourteenth aspect, there is provided a method for calibrating an analyte sensor comprising: applying a time-varying signal to the analyte sensor; measuring a signal response to the applied signal; determining a sensitivity of the analyte sensor using sensor electronics, said determining including relating at least one characteristic of the signal response to a predetermined sensor sensitivity profile; and generating an estimated analyte concentration value using the determined sensitivity and sensor data generated by the analyte sensor using sensor electronics.

[0112] In one embodiment of the fourteenth aspect or any other embodiment thereof, the sensitivity profile comprises sensitivity values ​​that change over a period of time since implantation of the sensor.

[0113] In an embodiment of the fourteenth aspect or any other embodiment thereof, the predetermined sensitivity profile comprises a plurality of sensitivity values.

[0114] In one embodiment of the fourteenth aspect or any other embodiment thereof, the predetermined sensitivity profile is based on sensor sensitivity data generated from examining sensitivity variations of analyte sensors similar to the analyte sensor.

[0115] In an embodiment of the fourteenth aspect or any other embodiment thereof, the method includes applying a bias voltage to a sensor, and the time-varying signal includes a step voltage above the bias voltage or a sinusoidal voltage superimposed on the bias voltage.

[0116] In one embodiment of the fourteenth aspect or any other embodiment thereof, the determining further includes calculating an impedance value based on the measured signal response and relating the impedance value to a sensitivity value of a predetermined sensitivity profile.

[0117] In one embodiment of the fourteenth aspect or any other embodiment thereof, the method further includes applying a DC bias voltage to the sensor to generate sensor data, and generating corrected sensor data using the determined sensitivity to estimate the analyte concentration value.

[0118] In an embodiment of the fourteenth aspect or any other embodiment thereof, the method further includes applying a conversion function to the modified sensor data to generate the estimated analyte concentration value.

[0119] In one embodiment of the fourteenth aspect or any other embodiment thereof, the method further includes forming a conversion function based at least in part on the determined sensitivity, and applying the conversion function to the sensor data to generate an estimated analyte concentration value.

[0120] In an embodiment of the fourteenth aspect or any other embodiment thereof, the characteristic is a peak current value of the signal response.

[0121] In one embodiment of the fourteenth aspect or any other embodiment thereof, the determining further includes using at least one of performing a fast Fourier transform on the signal response data, incorporating at least a portion of a curve of the signal response, and determining a peak current of the signal response.

[0122] In one embodiment of the fourteenth aspect or any other embodiment thereof, the determining further includes selecting a predetermined sensitivity profile from a plurality of different predetermined sensitivity profiles based on the determined sensor characteristic.

[0123] In one embodiment of the fourteenth aspect or any other embodiment thereof, the selecting includes performing a data association analysis to determine a correlation between the determined sensor characteristic and each of the plurality of different predetermined sensitivity profiles, and the selected predetermined sensitivity profile has the highest correlation.

[0124] In an embodiment of the fourteenth aspect or any other embodiment thereof, the method further includes generating an estimated analyte concentration value using the selected sensitivity profile.

[0125] In one embodiment of the fourteenth aspect or any other embodiment thereof, the method further includes determining a second sensitivity value using the selected sensitivity profile, wherein a first set of estimated analyte concentration values ​​are generated using the determined sensitivity value and sensor data associated with a first time period, and a second set of concentration values ​​are generated using the second sensitivity value and sensor data associated with a second time period.

[0126] In a fifteenth aspect, there is provided a sensor system for carrying out the method of the fourteenth aspect or any embodiment thereof.

[0127] In one embodiment of the fifteenth aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the fourteenth aspect or any embodiment thereof.

[0128] In a sixteenth aspect, there is provided a method for determining whether the analyte sensor system is functioning properly, the method comprising applying a stimulation signal to the analyte sensor, measuring a response to the stimulation signal, estimating a value of a sensor characteristic based on the signal response, relating the sensor characteristic value to a predetermined relationship of the sensor characteristic and a predetermined sensor sensitivity profile, and initiating an error routine if the correlation does not exceed a predetermined correlation threshold.

[0129] In one embodiment of the sixteenth aspect or any other embodiment thereof, the correlating includes performing a data association analysis.

[0130] In an embodiment of the sixteenth aspect or any other embodiment thereof, the error routine includes displaying a message to a user indicating that the analyte sensor is not functioning properly.

[0131] In an embodiment of the sixteenth aspect or any other embodiment thereof, the sensor characteristic is an impedance of the sensor membrane.

[0132] In a seventeenth aspect, there is provided a sensor system configured to perform the method of the sixteenth aspect or any embodiment thereof.

[0133] In one embodiment of the seventeenth aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the sixteenth aspect or any embodiment thereof.

[0134] In an eighteenth aspect, there is provided a method for determining a temperature associated with a continuous analyte sensor, comprising applying a stimulation signal to the analyte sensor, measuring a signal response of the stimulation signal, and determining a temperature associated with the analyte sensor, wherein the determining comprises relating at least one characteristic of the signal response to a predetermined relationship of a sensor characteristic to temperature.

[0135] In an embodiment of the eighteenth aspect or any other embodiment thereof, the method further includes generating an estimated analyte concentration value using the determined temperature and sensor data generated from the analyte sensor.

[0136] In one embodiment of the eighteenth aspect or any other embodiment thereof, the generating includes compensating the sensor data using the determined temperature and converting the compensated sensor data using a conversion function to the generated estimated analyte value.

[0137] In one embodiment of the eighteenth aspect or any other embodiment thereof, the generating includes generating or modifying a conversion function using the determined temperature, and converting the sensor data to the generated estimated analyte value using the generated or modified conversion function.

[0138] In one embodiment of the 18th aspect or any other embodiment thereof, the method includes measuring a temperature using a second sensor, and the determining further includes determining a temperature associated with the analyte sensor using the measured temperature.

[0139] In an embodiment of the eighteenth aspect or any other embodiment thereof, the second sensor is a thermistor.

[0140] In a nineteenth aspect, there is provided a system configured to perform the method of the eighteenth aspect or any embodiment thereof.

[0141] In one embodiment of the 19th aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 18th aspect or any embodiment thereof.

[0142] In a twentieth aspect, there is provided a method of determining moisture ingress into an electronic sensor system comprising applying a stimulus signal having a particular frequency, or a signal including a frequency spectrum, to an analyte sensor, measuring a response to the stimulus signal, using sensor electronics to calculate an impedance based on the measured signal response, using sensor electronics to determine whether the impedance falls within predefined levels corresponding to moisture ingress, and using sensor electronics to initiate an error routine if the impedance exceeds one or both of the predefined levels.

[0143] In one embodiment of the twentieth aspect or any other embodiment, the method further includes an error routine that includes activating one or more audible alarms and visual alarms on a display screen to alert a user that the sensor system may not be functioning properly.

[0144] In an embodiment of the twentieth aspect or any other embodiment thereof, the stimulation signal has a predetermined frequency.

[0145] In an embodiment of the twentieth aspect or any other embodiment thereof, the stimulation signal has a spectrum of frequencies.

[0146] In one embodiment of the twentieth aspect or any other embodiment thereof, the calculated impedance has a magnitude value and a phase value, and the determining includes comparing the impedance magnitude value to a predefined impedance magnitude threshold and the phase value to a predefined phase threshold.

[0147] In an embodiment of the twentieth aspect or any other embodiment thereof, the calculated impedance is a complex impedance value.

[0148] In a twenty-first aspect, there is provided a sensor system for carrying out the method of the twentieth aspect or any embodiment thereof.

[0149] In one embodiment of the 21st aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 20th aspect or any embodiment thereof.

[0150] In a twenty-second aspect, there is provided a method of determining membrane damage in an analyte sensor using a sensor system, the method comprising: applying a stimulation signal to the analyte sensor; measuring a response to the stimulation signal; using sensor electronics calculating an impedance based on the signal response; using sensor electronics determining whether the impedance falls within a predefined level corresponding to membrane damage; and using sensor electronics initiating an error routine if the impedance exceeds the predefined level.

[0151] In one embodiment of the twenty-second aspect or any other embodiment thereof, the error routine includes activating one or more audible alarms and visual alarms on a display screen.

[0152] In an embodiment of the twenty-second aspect or any other embodiment thereof, the stimulation signal has a predetermined frequency.

[0153] In an embodiment of the twenty-second aspect or any other embodiment thereof, the stimulation signal has a spectrum of frequencies.

[0154] In one embodiment of the twenty-second aspect or any other embodiment thereof, the calculated impedance includes a magnitude value and a phase value, and the determining includes comparing the impedance magnitude value to a predefined impedance magnitude threshold and the phase value to a predefined phase threshold.

[0155] In an embodiment of the twenty-second aspect or any other embodiment thereof, the calculated impedance is a complex impedance value.

[0156] In a twenty-third aspect, there is provided a sensor system for carrying out the method of the twenty-second aspect or any embodiment thereof.

[0157] In one embodiment of the 23rd aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 22nd aspect or any embodiment thereof.

[0158] In a twenty-fourth aspect, there is provided a method of determining reuse of an analyte sensor, the method comprising applying a stimulation signal to an analyte sensor, measuring a response to the stimulation signal, calculating an impedance response based on the response, comparing the calculated impedance to a predetermined threshold, and initiating a sensor reuse routine if the impedance is determined to exceed the threshold.

[0159] In an embodiment of the twenty-fourth aspect or any other embodiment thereof, the sensor reuse routine includes activating an audible and / or visual alarm to notify a user of improper sensor reuse.

[0160] In one embodiment of the twenty-fourth aspect or any other embodiment thereof, the routine for recycling the sensor includes completely or partially shutting down the sensor system and / or ceasing to display sensor data on a user interface of the sensor system.

[0161] In a twenty-fifth aspect, there is provided a sensor system configured to perform the method of the twenty-fourth aspect or any embodiment thereof.

[0162] In one embodiment of the twenty-fifth aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the twenty-fourth aspect or any embodiment thereof.

[0163] In a twenty-sixth aspect, there is provided a system for determining reuse of an analyte sensor, the system including applying a stimulation signal to an analyte sensor, measuring a response to the stimulation signal, calculating an impedance based on the response, determining a correlation between the calculated impedance and one or more recorded impedance values ​​using a data relation function, and initiating a sensor reuse routine if the correlation is determined to exceed a predetermined threshold.

[0164] In an embodiment of the twenty-sixth aspect or any other embodiment thereof, the sensor reuse routine includes activating an audible and / or visual alarm to notify a user of improper sensor reuse.

[0165] In one embodiment of the twenty-sixth aspect or any other embodiment thereof, the routine for recycling the sensor includes completely or partially shutting down the sensor system and / or ceasing to display sensor data on a user interface of the sensor system.

[0166] In one embodiment of the 26th aspect or other embodiments thereof, the sensor system comprises instructions stored in a computer memory that, when executed by one or more processors of the sensor system, causes the sensor system to perform the method of the 25th aspect.

[0167] In a twenty-seventh aspect, there is provided a method of applying an excess potential to an analyte sensor comprising applying a stimulation signal to the analyte sensor, measuring a response to the stimulation signal, determining a sensitivity of the sensor or a change in sensitivity of the sensor based on the response, and applying an excess potential to the sensor based on the determined sensitivity or change in sensitivity.

[0168] In one embodiment of the twenty-seventh aspect or any other embodiment thereof, the determining further includes calculating an impedance based on the response, and determining a sensitivity or a change in sensitivity based on the impedance.

[0169] In an embodiment of the twenty-seventh aspect or any other embodiment thereof, determining the sensitivity or change in sensitivity further includes relating an impedance-sensitivity relationship to a predetermined impedance.

[0170] In an embodiment of the twenty-seventh aspect or any other embodiment thereof, the applying includes determining or modifying an amount of time the excess potential is applied to the sensor.

[0171] In an embodiment of the twenty-seventh aspect or any other embodiment thereof, the applying includes determining or modifying a magnitude of an over-potential applied to the sensor.

[0172] In a twenty-eighth aspect, there is provided a sensor system configured to perform the method of the twenty-seventh aspect or any embodiment thereof.

[0173] In one embodiment of the 28th aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 27th aspect or any embodiment thereof.

[0174] In a twenty-ninth aspect, there is provided a method for determining a characteristic of a continuous analyte sensor comprising applying a stimulation signal to a first analyte sensor having a first working electrode and a first reference electrode, measuring a signal response of the stimulation signal using a second analyte sensor having a second working electrode and a second reference electrode, and determining a characteristic of the first sensor by relating the response to a predetermined relationship.

[0175] In an embodiment of the twenty-ninth aspect or any other embodiment thereof, the method further includes applying a bias voltage to the first working electrode to generate sensor data and measuring a response to the bias voltage.

[0176] In an embodiment of the twenty-ninth aspect or any other embodiment thereof, the method further includes calibrating sensor data using the determined characteristics.

[0177] In an embodiment of the twenty-ninth aspect or any other embodiment thereof, the determined characteristic is one of impedance and temperature.

[0178] In an embodiment of the twenty-ninth aspect or any other embodiment thereof, the method further includes determining damage to the sensor membrane using the determined characteristic.

[0179] In one embodiment of the twenty-ninth aspect or any other embodiment thereof, the method further includes determining moisture ingress into a sensor system surrounding the first and second analyte sensors using the determined characteristic.

[0180] In a thirtieth aspect, there is provided a sensor system configured to perform the method of the twenty-ninth aspect or any embodiment thereof.

[0181] In one embodiment of the 30th aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 29th aspect or any embodiment thereof.

[0182] In a thirty-first aspect, there is provided a method of determining a scaling for use in a continuous analyte sensor system, the method comprising: applying a first stimulation signal to a first working electrode of an analyte sensor, measuring a response to the first stimulation signal, applying a second stimulation signal to a second working electrode of the analyte sensor, measuring a response to the second stimulation signal, determining using sensor electronics a scaling factor based on the measured responses to the first and second stimulation signals, and using the scaling factor to generate an estimated analyte value based on sensor data generated by the analyte sensor.

[0183] In an embodiment of the thirty-first aspect or any other embodiment thereof, the method further includes the method being performed periodically.

[0184] In one embodiment of the thirty-first aspect or any other embodiment thereof, the determining includes calculating a first impedance using a response to the first stimulation signal and calculating a second impedance using a response to the second stimulation signal, and the scaling factor is a ratio of the first impedance and the second impedance.

[0185] In one embodiment of the thirty-first aspect or any other embodiment thereof, a first working electrode has a thin film including an enzyme configured to react with the analyte and a second working electrode has a thin film that does not include the enzyme.

[0186] In an embodiment of the thirty-first aspect or any other embodiment thereof, determining the scaling factor includes updating a previous scaling factor based on measured responses to the first and second stimulus signals.

[0187] In one embodiment of the thirty-first aspect or any other embodiment thereof, the scaling factor is an acetaminophen scaling factor, and the method includes updating a further scaling factor based on the acetaminophen scaling factor, and applying the further scaling factor to the sensor data to generate the estimated analyte value.

[0188] In a thirty-second aspect, there is provided a sensor system configured to perform the method of the thirty-first aspect or any embodiment thereof.

[0189] In one embodiment of the 32nd aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory which, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 31st aspect or any embodiment thereof.

[0190] In a thirty-third aspect, there is provided a method of calibrating an analyte sensor comprising applying a predetermined signal to the analyte sensor, measuring a response to the applied signal, determining using sensor electronics a change in impedance associated with a thin film of the analyte sensor based on the measured response, calculating a change in sensitivity of the analyte sensor based on the determined impedance, calculating a revised sensitivity based on the calculated change in sensitivity and a previously used sensitivity of the analyte sensor, and generating an estimated analyte value using the revised sensitivity.

[0191] In an embodiment of the thirty-third aspect or any other embodiment thereof, calculating the change in sensitivity includes applying a non-linear compensation function.

[0192] In an embodiment of the thirty-third aspect or any other embodiment thereof, the nonlinear compensation function is expressed as the following equation: △S=(a*log(t)+b)*△I where ΔS is the change in sensitivity, t is the time since the analyte sensor was calibrated, ΔI is the change in determined impedance, and a and b are predetermined coefficients.

[0193] In one embodiment of the thirty-third aspect or any other embodiment thereof, a and b are determined by pre-testing similar analyte sensors.

[0194] In a thirty-fourth aspect, there is provided a sensor system configured to perform the method of the thirty-third aspect or any embodiment thereof.

[0195] In one embodiment of the 34th aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory that, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 33rd aspect or any embodiment thereof.

[0196] In a thirty-fifth aspect, there is provided a method for calibrating an analyte sensor, the method comprising generating sensor data using a subcutaneous analyte sensor, forming or modifying a conversion function using pre-implant information, internal diagnostic information, and / or external reference information as inputs to form or modify the conversion function, and calibrating the sensor data using the conversion function.

[0197] In one embodiment of the thirty-fifth aspect or any other embodiment, the pre-implant information includes information selected from the group consisting of a predetermined sensitivity profile associated with the analyte sensor, a predetermined relationship between a measured sensor attribute and sensor sensitivity, and one or more predetermined relationships between a measured sensor attribute and sensor temperature, sensor data obtained from a previously used analyte sensor, a calibration code associated with the analyte sensor, a patient-specific relationship between the analyte sensor and one or more of sensitivity, baseline, drift, and impedance, information representative of the implantation site of the sensor, the time since manufacture of the analyte sensor, and information representative of the analyte exposed to temperature or humidity.

[0198] In one embodiment of the thirty-fifth aspect or any other embodiment thereof, the internal diagnostic information includes information selected from the group consisting of output of a stimulation signal, sensor data representative of an analyte concentration measured by the sensor, temperature measurements using the sensor or a separate sensor, sensor data generated by a redundant sensor designed substantially similar to the analyte sensor, sensor data generated by an auxiliary sensor having a different format than the analyte sensor, time since the sensor was implanted or connected to sensor electronics connected to the sensor, data generated by a pressure sensor representative of pressure on the sensor or sensor system, data generated by an accelerometer, moisture ingress criteria, and analyte concentration signal noise criteria.

[0199] In one embodiment of the thirty-fifth aspect or any other embodiment, the reference information includes information selected from the group consisting of real-time and / or previous analyte concentration information obtained from a reference monitor, information regarding the type / brand of the reference monitor used to provide the reference data, information regarding the amount of carbohydrates consumed by the user, information received from a drug delivery device, glucagon sensitivity information, and information collected from population-based data.

[0200] In a thirty-sixth aspect, there is provided a sensor system configured to perform the method of the thirty-fifth aspect or any embodiment thereof.

[0201] In one embodiment of the 36th aspect or any other embodiment thereof, the sensor system comprises instructions stored in a computer memory that, when executed by one or more processors of the sensor system, cause the sensor system to perform the method of the 35th aspect or any embodiment thereof.

[0202] These and other features and advantages will be better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings. [Brief description of the drawings]

[0203] [Figure 1A] 1 shows a schematic diagram of sensor sensitivity as a function of time during a sensor session, according to one embodiment. [Figure 1B] 1B shows a schematic diagram of a conversion function for different time periods of a sensor session according to the embodiment of FIG. 1A. [Figure 2A] 1 illustrates one embodiment of a process for generating a sensor sensitivity profile. [Figure 2B] 1 illustrates another embodiment of a process for generating a sensor sensitivity profile. [Figure 2C] 1 illustrates another embodiment of a process for generating a sensor sensitivity profile. [Figure 2D] 1 illustrates yet another embodiment of a process for generating a sensor sensitivity profile. [Figure 3A] 1 is a Bland-Altman plot showing the difference between a YSI reference measurement and a specific factory calibrated in vivo continuous analyte sensor, according to one embodiment. [Figure 3B]3B is a Clarke error grid associated with data from the continuous analyte sensor associated with FIG. 3A. [Figure 4] 1 illustrates data from a study testing the accuracy levels of a continuous analyte sensor that received one baseline measurement approximately one hour after insertion into a patient, according to one embodiment. [Diagram 5] 1 shows a diagram illustrating different types of information that can be input into a sensor system to define a sensor sensitivity profile over time. [Figure 6] 1 illustrates a schematic diagram of sensor sensitivity as a function of time between completion of sensor fabrication and start of a sensor session, according to one embodiment. [Figure 7A] FIG. 2 is a schematic diagram illustrating a distribution curve of sensor sensitivity corresponding to a Bayesian learning process according to one embodiment. [Figure 7B] FIG. 7B is a schematic diagram illustrating confidence levels associated with the sensor sensitivity profile corresponding to the distribution curve shown in FIG. 7A. [Figure 8] 1 shows a graph providing a comparison between estimated and detected glucose equivalent baselines from one study. [Figure 9] FIG. 2 is a schematic diagram of a model sensor circuit, according to one embodiment. [Figure 10] FIG. 2 is a Bode plot of an analyte sensor according to one embodiment. [Figure 11] 1 is a flow chart describing a process for determining impedance of a sensor, according to one embodiment. [Figure 12] 13 is a flowchart describing a process for determining the impedance of a sensor based on a derivative response according to one embodiment. [Figure 13] 1 is a flow chart describing a process for determining the impedance of a sensor based on a peak current response according to one embodiment. [Figure 14A] 4 illustrates a step voltage applied to a sensor according to one embodiment. [Figure 14B] 14B shows the response to the step voltage of FIG. 14A. [Figure 15] 1 is a flow chart describing a process for estimating an analyte concentration value using a corrected signal based on impedance measurements, according to one embodiment. [Figure 16] 1 is a flow chart describing a process for estimating an analyte concentration value using a predetermined sensitivity profile selected based on impedance measurements, according to one embodiment. [Figure 17] 11 is a flow chart describing a process for determining an error based on whether a sensitivity determined using impedance measurements sufficiently corresponds to a predetermined sensitivity, according to one embodiment. [Figure 18] 11 is a flowchart describing a process for determining a temperature associated with a sensor by correlating impedance measurements with a predetermined temperature-impedance relationship, according to one embodiment. [Figure 19] 1 is a flowchart describing a process for determining moisture egress in sensor electronics associated with an analyte sensor, according to one embodiment. [Figure 20] 1 is a flow chart describing a process for determining damage to a membrane associated with an analyte sensor, according to one embodiment. [Figure 21] 1 is a flowchart describing a first process associated with determining sensor reuse, according to one embodiment. [Figure 22] 11 is a flow chart describing a second process associated with determining sensor reuse, according to one embodiment. [Figure 23] FIG. 2 is a schematic diagram of a two-electrode configuration used to determine sensor characteristics, according to one embodiment. [Figure 24] FIG. 1 is a diagram of a calibration process using various inputs to form a conversion function, according to one embodiment. [Diagram 25]Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 26] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 27] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 28] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 29] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 30] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 31] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 32] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 33] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 34] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 35] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 36] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 37] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 38] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 39] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 40] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 41] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 42] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 43] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 44] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 45] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Diagram 46] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 47] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 48] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 49] Results of studies using stimulation signals to determine sensor characteristics are summarized. [Figure 50] 1 is a flow chart describing a process for generating an estimated analyte value using a compensation algorithm based on measured changes in impedance, according to one embodiment. [Figure 51] 1 is a graph of a study comparing the use of a compensation algorithm based on measured changes in impedance. [Figure 52] 1 is a graph of a study comparing the use of a compensation algorithm based on measured changes in impedance. [Figure 53] 13 is a graph of a study comparing the use of a compensation algorithm based on measured changes in impedance. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0204] definition To facilitate understanding of the embodiments described herein, a number of terms are defined below.

[0205] The term "analyte" as used herein is a broad term that is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, any chemical constituent in a substance or biological fluid that can be analyzed (e.g., blood, interstitial fluid, cerebrospinal fluid, lymphatic fluid, or urine). Analytes can include naturally occurring substances, man-made substances, metabolic products, and / or reaction products. In some embodiments, the analyte for measurement by the sensor heads, devices, and methods disclosed herein is glucose. However, other analytes were predicted as well, including carboxyprothrombin; acylcarnitines; adenine phosphoribosyltransferase; adenosine deaminase; albumin; alpha-fetoprotein; amino acid profile (arginine (Krebs cycle), histidine / urocanic acid, homocysteine, phenylalanine / tyrosine, tryptophan); androstenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; C-reactive protein; carnitine; carnosinase; CD4; ceruloplasmin; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1β-hydroxycholic acid; cortisol; creatine kinase; creatine kinase; MM isoenzyme;cyclosporine A;d-penicillamine;deethylchloroquine;dehydroepiandrosterone sulfate;DNA (acetylation polymorphism, alcohol dehydrogenase, alpha 1-antitrypsin, cystic fibrosis, Duchenne / Becker muscular dystrophy, analyte-6-phosphate dehydrogenase, hemoglobinopathies, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin E, D Punjab type, beta thalassemia, hepatitis B virus, HCMV, HIV-1, HTLV-1, Leber hereditary optic neuropathy, medium-chain acyl-CoA dehydrogenase (MCAD), ribonucleic acid (RNA), phenylketonuria (PKU), Plasmodium vivax, sex differentiation, 21-deoxycortisol);desbutylhalofantrine;dihydropteridine reductase;diphtheria / tetanus antitoxin;Erythrocyte arginase;Erythrocyte protoporphyrin;Esterase D;Fatty acids / acylglycines;Free beta-human chorionic gonadotropin;Free erythrocyte porphyrin;Free thyroxine (FT4);Free triiodothyronine (FT3);Fumarylacetoacetase;Galactose / gal-1-phosphate;Galactose-1-phosphate uridyltransferase;Gentamicin;Analyte-6-phosphate dehydrogenase;Glutathione;Glutathione hemoglobin;glycocholate;glycosylated hemoglobin;halofantrine;hemoglobin modification;hexosaminidase A;human erythrocyte decarboxylase I;17α-hydroxyprogesterone;hypoxanthine phosphoribosyltransferase;immunoreactive trypsin;lactic acid;lead;lipoprotein ((a),B / A-1,β);lysozyme;mefloquine;netilmicin;phenobarbitone;phenytoin;phytanic acid / pris Tannic acid;progesterone;luteinizing hormone;prolidase;purine nucleoside phosphorylase inhibitors;quinine;reverse triiodothyronine (rT3);selenium;serum pancreatic lipase;sisomicin;somatomedin C;specific antibodies (adenovirus, antinuclear antibody, antizeta antibody, arbovirus, Aujezsky's disease virus, dengue virus, guinea worm, tapeworm, entamoeba histolytica, enterovirus, Giardia lamblia, Helicobacter pylori, Hepatitis B virus, Herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, Leptospirosis, Measles / Mumps / Rubella, Mycobacterium leprae, Mycoplasma pneumoniae, Myoglobin, Onchocerca volvulus, Parainfluenza virus, Plasmodium falciparum, Poliovirus, Pseudomonas aeruginosa, Respiratory syncytial virus, Rickettsia (tsutsugamushi disease), Schistosoma mansoni, Toxoplasma gondii, Treponema pallidum , Trypanosoma cruzi / Trypanosoma Langerii, Vesicular stomatitis virus, Wuchereria bancrofti, Yellow fever virus);Specific antigens (Hepatitis B virus, HIV-1);Succinylacetone;Sulfadoxine;Theophylline;Thyroid-stimulating hormone (TSH);Thyroxine (T4);Thyroxine-binding globulin;Trace elements;Transferrin;UDP-galactose-4-epimerase;Urea;Uroporphyrinogen I synthase;Vitamin A;Leukocytes;Analytes include, but are not limited to, zinc protoporphyrin, and zinc protoporphyrin. Salts, carbohydrates, proteins, fats, vitamins and hormones that occur naturally in blood or interstitial fluid may also constitute analytes in some embodiments. Analytes may be naturally occurring in biological fluids, such as metabolites, hormones, antigens, antibodies, etc. Alternatively, the specimen may be introduced into the body, for example, a contrast agent for imaging, a radioisotope, a chemical agent, a fluorocarbon-based synthetic blood, or the following drugs or pharmaceutical compositions, including, but not limited to: insulin; ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (laughing gas, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, Ritalin, Cylert, Preludin, Didrex, PreState, Voranil, Sandrex, Plegine); depressants (barbiturates, methaqualone; tranquilizers (varium, librium, miltuna, serax, meprobamate, transgene); hallucinogens (phencyclidine, lysergidine, acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tussionex, fentanyl, Darvon, Talwin, Lomotil); synthetic antibiotics (fentanyl, meperidine, amphetamine, methamphetamine, and phencyclidine, such as analogs of ecstasy); muscle enhancers; and nicotine. Metabolites of drugs and pharmaceutical compositions are also anticipated analytes. Analytes such as neurochemicals and other chemicals produced in the body, such as ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), and 5-hydroxyindoleacetic acid (FHIAA), are also analyzed.

[0206] As used herein, the terms "continuous analyte sensor" and "continuous glucose sensor" are broad terms and are given their ordinary and accustomed meaning to those of ordinary skill in the art (and are not limited to any special or customized meanings), and refer to, but are not limited to, a device that continuously or continuously measures analyte / glucose concentration and / or calibrates (e.g., by continuously or continuously adjusting or determining the sensitivity and background of the sensor) over time intervals ranging from, for example, a fraction of a second up to, for example, 1, 2, or 5 minutes, or longer.

[0207] The term "biological sample" as used herein is intended to be a broad term and is to be given its ordinary and accustomed meaning to those of skill in the art (and is not limited to any special or customized meaning), and refers to a sample derived from the body or tissues of a recipient, such as, but not limited to, blood, interstitial fluid, spinal fluid, saliva, urine, tears, sweat, or other similar fluids.

[0208] The term "host" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to animals, including, but not limited to, humans.

[0209] The term "membrane system" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a permeable or semi-permeable membrane that is permeable to oxygen and selectively permeable to glucose, which may be composed of two or more regions and is typically composed of a material several microns thick or thicker. In one embodiment, the membrane system includes an immobilized glucose oxidase enzyme that allows for the occurrence of an electrochemical reaction to measure glucose concentration.

[0210] The term "domain" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, a region of a film or a portion of a film that may be a layer of uniform or non-uniform gradient (e.g., anisotropic) functional aspects of a material.

[0211] The term "sensing region" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the region of a monitoring device responsible for detecting a particular analyte. In one embodiment, the sensing region generally comprises a non-conductive body, at least one electrode, a reference electrode and an optional counter electrode extending through and secured within the body to form an electroactive surface at one location of the body and an electronic connection at another location of the body, and a membrane system secured to the body and covering the electroactive surface.

[0212] The term "electroactive surface" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, an electrode surface where an electrochemical reaction occurs. In one embodiment, the working electrode measures hydrogen peroxide (H2O2), which generates a measurable current.

[0213] The term "baseline" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, the component of the analyte sensor signal that is not related to analyte concentration. In one embodiment of a glucose sensor, this baseline consists essentially of signal contributions due to factors other than glucose (e.g., interfering species, unreacted associated hydrogen peroxide, or other electroactive species with an oxidation potential that overlaps with hydrogen peroxide). In some embodiments where the calibration is determined by solving the equation y=mx+b, the value of b represents the baseline of the signal. In certain embodiments, this value of b (i.e., baseline) can be zero or near zero. This can be the result of, for example, a baseline-subtracted electrode or a low bias potential setting. As a result, in these embodiments, the calibration can be determined by solving the equation y=mx.

[0214] The term "inactive enzyme" as used herein is a broad term and is given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, an enzyme (e.g., glucose oxidase, GOx) that has been rendered inactive (e.g., by denaturing the enzyme) and has substantially no enzymatic activity. Enzymes can be inactivated using a variety of techniques known in the art, including, but not limited to, heating, freeze-thawing, denaturation in organic solvents, acid or base, cross-linking, genetic modification of enzymatically important amino acids, etc. In some embodiments, a solution containing an active enzyme can be applied to the sensor, followed by heating or treatment with an inert solvent to substantially inactivate the applied enzyme.

[0215] The term "non-enzymatic" as used herein is a broad term and is given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the absence of enzyme activity. In some embodiments, the "non-enzymatic" membrane portion does not contain an enzyme, while in other embodiments, the "non-enzymatic" membrane portion contains an inactive enzyme. In some embodiments, an enzyme solution containing inactive enzyme or no enzyme is added.

[0216] The term "substantially" as used herein is intended to be a broad term and is to be given its ordinary and accustomed meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, most but not necessarily all of what is specified.

[0217] The term "about" as used herein is a broad term and is to be given its ordinary and accustomed meaning to those of ordinary skill in the art (and is not limited to any specific or customized meaning), and when in connection with any numerical value or range, refers to, but is not limited to, the understanding that the quantity or condition that the term modifies can vary somewhat more than the stated amount so long as the disclosed function is achieved.

[0218] The term "ROM" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, read-only memory, a type of data storage device manufactured with a fixed capacity. ROM includes, for example, Electrically Erasable Programmable ROM (EEPROM).

[0219] The term "RAM" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, a data storage device in which the ordering of accesses to different locations does not affect the speed of access. RAM is broad enough to include, for example, SRAM, which is a static random access memory that retains data bits in its memory as long as power is applied.

[0220] The term "A / D converter" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, hardware and / or software that converts an analog electrical signal into a corresponding digital signal.

[0221] As used herein, the terms "raw data stream" and "data stream" are broad terms and are given their ordinary and customary meanings to those of ordinary skill in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, analog or digital signals directly related to the analyte concentration measured by an analyte sensor. In one embodiment, a raw data stream is digital data in counts converted by an A / D converter from an analog signal (e.g., voltage or amperage) representative of the analyte concentration. The terms broadly encompass multiple time-spaced data points from a substantially continuous analyte sensor, including single measurements taken at time intervals ranging from a fraction of a second up to, for example, 1, 2, or 5 minutes, or longer.

[0222] The term "counts" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, a unit of measurement for a digital signal. In one embodiment, the raw data stream measured in counts is directly related to a voltage (e.g., converted by an A / D converter) that is directly related to a current from the working electrode.

[0223] The term "sensor electronics" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, elements of a device (e.g., hardware or software) configured to process data. In the case of an analyte sensor, the data includes biological information regarding the analyte concentration in a biological fluid acquired by the sensor. U.S. Patent Nos. 5,233,633, 5,233,625 and 5,393,663 describe suitable electronic circuitry that may be utilized in certain embodiment devices.

[0224] The term "potentiostat" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning) and refers, without limitation, to an electrical system that applies a set amount of potential between a working electrode and a reference electrode of a two or three electrode battery and measures the current flowing through the working electrode. The potentiostat forces the current required to flow between the working and counter electrodes to maintain the desired potential, as long as the required battery voltage and current do not exceed the range of the potentiostat.

[0225] The term "operably connected" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, one or more components that are coupled to other components to allow for the transmission of signals between those components. For example, one or more electrodes may be used to detect the amount of glucose in a sample and convert that information into a signal that can then be transmitted to an electronic circuit. In this case, the electrodes are "operably connected" to the electronic circuit. These terms are broad enough to include wired and wireless connections.

[0226] The term "filtering" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, modifications that make a set of data smoother and more continuous, and remove or reduce spread out points, for example, by performing a moving average of the raw data stream.

[0227] The term "algorithm" as used herein is intended to be a broad term and is to be given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, a computational process (e.g., a program) involving the transformation of information from one state to another, e.g., using a computer.

[0228] The term "calibration" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the process of determining the calibration of a sensor that results in a quantitative measurement (e.g., analyte concentration). By way of example, calibration can be updated or readjusted over time to compensate for changes associated with the sensor, such as changes in sensor sensitivity and sensor background. Additionally, calibration of a sensor can include automatic and self-calibration, e.g., without the use of a reference analyte value after the point of use.

[0229] The term "sensor data" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and refers to, but is not limited to, data received from a continuous analyte sensor that includes one or more time-spaced sensor data points.

[0230] As used herein, the terms "reference analyte values" and "reference data" are broad terms and are given their ordinary and customary meanings to those of ordinary skill in the art (and are not limited to any special or customized meanings), and refer to, but are not limited to, reference data from a reference analyte monitor, such as a blood glucose meter or the like, that includes one or more reference data points. In some embodiments, the reference glucose values ​​are obtained from a self-monitored blood glucose (SMBG) test (e.g., a finger or forearm blood test) or from, for example, a Yellow Springs Instrument (YSI) test.

[0231] The terms "interferents" and "interfering species" as used herein are broad terms and are given their ordinary and customary meaning to those of ordinary skill in the art (and are not limited to any special or customized meaning), and refer to, but are not limited to, effects or species that interfere with the measurement of an analyte of interest in a sensor to produce a signal that does not accurately represent the analyte measurement. In one embodiment of an electrochemical sensor, an interfering species is a compound that has an oxidation potential that overlaps with the analyte being measured, producing a false positive signal.

[0232] The term "sensor session" as used herein is a broad term and is given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the period during which a sensor is applied (e.g., implanted) to a recipient or used to obtain sensor values. For example, in some embodiments, a sensor session extends from the time the sensor is implanted (e.g., including insertion of the sensor into subcutaneous tissue and placing the sensor in fluid communication with the recipient's circulatory system) to the time the sensor is removed.

[0233] As used herein, the terms "sensitivity" or "sensor sensitivity" are broad terms and are given their ordinary and customary meanings to those of ordinary skill in the art (and are not limited to any special or customized meanings), and refer to, but are not limited to, a given concentration of a measured analyte or the amount of signal generated by a measured chemical species (e.g., H2O2) in relation to the analyte (e.g., glucose) being measured. For example, in one embodiment, the sensor has a sensitivity of about 1 to about 300 picoamps of current for every 1 mg / dL of glucose analyte.

[0234] The terms "sensitivity profile" or "sensitivity curve" as used herein are broad terms and are given their ordinary and customary meaning to those of ordinary skill in the art (and are not limited to any special or customized meaning), and refer to, but are not limited to, a representation of the change in sensitivity over a period of time.

[0235] overview Conventional in vivo continuous analyte sensing techniques typically rely on a reference measurement performed during a sensor session to calibrate the continuous analyte sensor. This reference measurement is combined with substantially time-corresponding sensor data to produce a combined data pair. A regression is then performed (e.g., using least squares regression) on the combined data pair to generate a transfer function that defines the relationship between the sensor signal and an estimated glucose concentration.

[0236] In critical care settings, calibration of continuous analyte sensors is often performed by using a calibration solution with a known concentration of analyte as a reference. This calibration technique can be cumbersome because a calibration bag separate from (and in addition to) the IV (intravenous) bag is typically used. In ambulatory settings, calibration of continuous analyte sensors is traditionally performed by capillary blood glucose measurements (e.g., finger stick glucose tests) whereby reference data is obtained and input into the continuous analyte sensor system. This calibration technique typically involves frequent finger stick measurements, which can be inconvenient and painful.

[0237] To date, systems and methods for in vitro calibration of continuous analyte sensors by manufacturers (e.g., factory calibration) have not relied on periodic recalibration and have often been inadequate with respect to high levels of sensor accuracy. This can be attributed to changes in sensor characteristics (e.g., sensor sensitivity) that occur during sensor use. As a result, calibration of continuous analyte sensors typically requires periodic input of reference data, whether related to calibration solutions or finger stick measurements. This can be a significant burden for patients in ambulatory settings or for hospital staff in critical care settings.

[0238] The continuous analyte sensors described herein can achieve a high level of accuracy while being capable of continuous, automatic self-calibration during a sensor session without (or with reduced) reliance on reference data from a reference analyte monitor (e.g., a blood glucose meter). In some embodiments, the continuous analyte sensor is an invasive, minimally invasive, or non-invasive device. The continuous analyte sensor can be a subcutaneous, transcutaneous, or intravascular device. In some embodiments, one or more of these devices can form a continuous analyte sensor system. For example, the continuous analyte sensor system can include a combination of subcutaneous and transcutaneous devices, a combination of subcutaneous and intravascular devices, a combination of transcutaneous and intravascular devices, or a combination of subcutaneous, transcutaneous, and intravascular devices. In some embodiments, the continuous analyte sensor can analyze multiple intermittent biological samples (e.g., blood samples). The continuous analyte sensor can use any glucose measurement method, including methods involving enzymatic, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, iontophoretic, and radiometric mechanisms, etc.

[0239] In some embodiments, a continuous analyte sensor includes one or more working electrodes and one or more reference electrodes that work together to measure a signal related to an analyte concentration in a recipient. The output signal from the working electrodes is typically a raw data stream that is calibrated, processed, and used to generate an estimated analyte (e.g., glucose) concentration. In some embodiments, the continuous analyte sensor can measure an additional signal related to the baseline and / or sensitivity of the sensor, thereby allowing for monitoring of the baseline and / or additional monitoring of sensitivity changes or drift that occur in the continuous analyte sensor over time.

[0240] In some embodiments, the sensor extends through the housing, which maintains the sensor on the skin and provides the sensor with an electrical connection to the sensor electronics. In one embodiment, the sensor is formed from a wire. For example, the sensor can include an elongated conductive body, such as a bare elongated conductive core (e.g., a metal wire) or an elongated conductive core coated with one, two, three, four, five or more layers of material, each of which may be conductive or non-conductive. The elongated sensor can be long and thin, yet flexible and strong. For example, in some embodiments, the minimum dimension of the elongated conductive body is less than about 0.1 inches, 0.075 inches, 0.05 inches, 0.025 inches, 0.01 inches, 0.004 inches, or 0.002 inches. Other embodiments of elongated conductive bodies are disclosed in U.S. Patent No. 6,399,433, which is incorporated herein by reference in its entirety. Preferably, a membrane system is deposited on at least a portion of the electroactive surfaces of the sensor 102 (including the working electrode and optionally the reference electrode) to provide protection of exposed electrode surfaces from the biological environment, resistance (limitation) of analyte diffusion if necessary, catalyst to enable enzymatic reactions, limit or block interferents, and / or hydrophilicity of the electrochemically reactive surfaces of the sensor interface. Disclosures regarding different membrane systems that can be used in the embodiments described herein are described in U.S. Patent No. 5,399,633, which is incorporated herein by reference in its entirety.

[0241] Calibrating sensor data from a continuous analyte sensor typically involves determining a relationship between a measurement generated by the sensor (e.g., in nA or digital counts after A / D conversion) and one or more reference measurements (e.g., in mg / dL or mmol / L). In certain embodiments, one or more reference measurements taken immediately after the analyte sensor is manufactured and before the sensor is used for calibration. The reference measurements can be taken in many forms. For example, in certain embodiments, the reference measurements can be determined from a ratio or correlation between the sensitivity of a sensor (e.g., from a sensor lot) for in vivo analyte concentration measurements over a period of time and the sensitivity of other sensors (e.g., from the same lot manufactured in substantially the same manner under substantially the same conditions) for in vitro analyte concentration measurements. By providing a continuous analyte sensor with a predetermined in vivo to in vitro ratio and a predetermined sensitivity profile (as described in more detail elsewhere herein), self-calibration of the sensor can be achieved in conjunction with a high level of sensor accuracy.

[0242] Self-calibration can eliminate or otherwise reduce the need for recalibration using reference data during a sensor session, and can require recalibration only in limited circumstances, such as when a sensor failure is detected. Additionally or alternatively, in some embodiments, a continuous analyte sensor can be configured to request and accept one or more reference measurements (e.g., from a finger stick glucose measurement or a calibration solution) at the start of a sensor session. In some embodiments, using a reference measurement associated with a predetermined sensor sensitivity profile at the start of a sensor session can eliminate or substantially reduce the need for further reference measurements.

[0243] For some implantable enzyme-based electrochemical glucose sensors, the sensing mechanism relies on certain phenomena that have a generally linear relationship with glucose concentration, such as (1) the diffusion of the analyte through a membrane system located between the implantation site (e.g., the subcutaneous space) and the electroactive surface, (2) the rate of the enzyme-catalyzed reaction of the analyte to produce the measured species inside the membrane system (e.g., the rate of the glucose oxidase-catalyzed reaction of glucose with O2 to produce gluconic acid and H2O2), and (3) the diffusion of the measured species (e.g., H2O2) to the electroactive surface. Due to this generally linear relationship, calibration of the sensor is obtained by solving the equation: y=mx+b where y represents the sensor signal (counts), x represents the estimated glucose concentration (mg / dL), m represents the sensor sensitivity to analyte concentration (counts / mg / dL), and b represents the baseline signal (counts). As described elsewhere herein, in some embodiments, the value of b (i.e., the baseline) can be zero or near zero. As a result, for these embodiments, the calibration can be determined by solving the equation y=mx.

[0244] In some embodiments, the continuous analyte sensor system is configured to estimate the change or drift in sensitivity of the sensor throughout the sensor session as a function of time (e.g., time elapsed since the start of the sensor session). As described elsewhere herein, this sensitivity function plotted against time can resemble a curve. Additionally or alternatively, the system can be configured to determine the change or drift in sensor sensitivity as a function of time and one or more other parameters that may affect the sensor sensitivity or provide additional information regarding the sensor sensitivity. These parameters may affect the sensor sensitivity or provide additional information regarding the sensor sensitivity prior to the sensor session, such as parameters related to the manufacture of the sensor (e.g., the material used to manufacture the sensor film, the thickness of the sensor film, the temperature at which the sensor film was cured, the amount of time the sensor was immersed in a particular coating solution, etc.). In some embodiments, some of the parameters include information obtained prior to the sensor session that may account for a calibration code associated with a particular sensor lot. Other parameters can relate to conditions surrounding the sensor after it is manufactured but prior to a sensor session, such as, for example, the level of exposure of the sensor to a certain level of humidity, or the temperature while the sensor was in its package during transport from the manufacturing facility to the patient. Still other parameters (e.g., the permeability of the sensor membrane, the temperature at the sample location, the pH at the sample location, the oxygen level at the sample location, etc.) can also affect sensor sensitivity or provide additional information regarding sensor sensitivity during a sensor session.

[0245] Determining the sensor sensitivity at different times of a sensor session based on a predetermined sensor sensitivity profile can be performed prior to the sensor session or when starting the sensor session. Additionally, in certain embodiments, determining the sensor sensitivity based on the sensor sensitivity profile can be continuously adjusted to compensate for parameters that affect the sensor sensitivity or provide additional information regarding the sensor sensitivity during the sensor session. These determinations of sensor sensitivity change or drift can be used to provide self-calibration, update the calibration, supplement the calibration based on measurements (e.g., from a reference analyte monitor), and / or confirm or reject the reference analyte measurement from the reference analyte monitor. In some embodiments, the confirmation or rejection of the reference analyte measurement can be based on whether the reference analyte measurement is within a range of values ​​associated with the predetermined sensor sensitivity profile.

[0246] Some of the continuous analyte sensors described herein can be configured to measure a signal related to a non-analyte constant of the recipient. Preferably, the non-analyte constant signal is measured below the membrane system on the sensor. In one example of a continuous glucose sensor, a non-glucose constant that may be measured is oxygen. In some embodiments, changes in oxygen delivery, which may indicate a change or drift in the sensitivity of the glucose signal, can be measured by switching the working electrode, an auxiliary electrode that measures oxygen, an oxygen sensor, or other bias potentials.

[0247] In addition, some of the continuous analyte sensors described herein can be configured to measure a change in the amount of background noise in the signal. Detection of a change above a certain threshold can provide the basis for initiating a calibration, updating the calibration, and / or confirming or rejecting an inaccurate reference analyte value from the reference analyte monitor. In one embodiment of a continuous glucose sensor, the background noise includes signal contributions from factors substantially other than glucose (e.g., interfering species, non-reactive related hydrogen peroxide, or other electroactive species with an oxidation potential that overlaps with hydrogen peroxide). That is, the continuous glucose sensor is configured to measure a signal related to a baseline (including substantially all of the non-glucose related currents generated) measured by the host sensor. In some embodiments, an auxiliary electrode located under a non-enzymatic portion of the membrane system is used to measure the baseline signal. To obtain a signal related entirely or substantially entirely to the glucose concentration, the baseline signal can be subtracted from the glucose+baseline signal. This subtraction can be accomplished electronically in the sensor using a differential amplifier, digitally in the receiver, and / or in the hardware or software of the sensor or receiver as described in more detail elsewhere herein.

[0248] Together, by determining sensor sensitivity based on the sensitivity profile and by measuring a baseline signal, a continuous analyte sensor can continuously self-calibrate during a sensor session, eliminating (or reducing) reliance on a reference measurement from a reference analyte monitor or calibration solution.

[0249] Determining Sensor Sensitivity As described elsewhere herein, in one embodiment, self-calibration of an analyte sensor system can be performed by determining the sensor sensitivity based on the sensitivity profile (and a measured or estimated baseline) such that the following equation can be solved: y=mx+b where y represents the sensor signal (counts), x represents the estimated glucose concentration (mg / dL), m represents the sensor sensitivity to the analyte (counts / mg / dL), and b represents the baseline signal (counts). From this equation, a transfer function can be formed, which converts the sensor signal to an estimated glucose concentration.

[0250] What has been found is that the sensitivity of a sensor to analyte concentration during a sensor session often changes or drifts as a function of time. FIG. 1A illustrates this phenomenon and provides a plot of sensor sensitivity 110 as a function of time during a sensor session of a group of continuous glucose sensors. FIG. 1B provides a plot of the transfer function for three different periods of the sensor session. As shown in FIG. 1B, the three transfer functions have different slopes, each corresponding to a different sensor sensitivity. Thus, the difference in slope over time indicates that a change or drift in sensor sensitivity occurs over the sensor session.

[0251] Referring again to the study associated with FIG. 1A, the sensors were manufactured in substantially the same manner under substantially the same conditions. The sensor sensitivities associated with the y-axis of the diagram are expressed as a percentage of the substantially steady-state sensitivity reached approximately 3 days after the start of the sensor session. Additionally, these sensor sensitivities correspond to measurements obtained from YSI testing. As shown in the diagram, the measured sensitivity of each sensor (expressed as a percentage of the steady-state sensitivity) is very close to the sensitivity of the other sensors in the group at any given time point of the sensor session. Without wishing to be limited by theory, the observed upward trend (over time) in sensitivity, especially evident in the early part of the sensor session, can be attributed to conditioning and hydration of the sensing area of ​​the working electrode. The glucose concentration of the fluid surrounding the continuous glucose sensor during the start-up of the sensor is also believed to affect the drift in sensitivity.

[0252] For the sensors tested in this study, the change in sensor sensitivity (expressed as a percentage of substantially steady-state sensitivity) over time as defined by a sensor session resembles a logarithmic growth curve. It should be understood that other continuous analyte sensors manufactured with different technologies, with different specifications (e.g., different film thicknesses or compositions), or under different manufacturing conditions may exhibit different sensor sensitivity profiles (e.g., one related to a linear function). Nevertheless, a high level of reproducibility was achieved through improved control of the operating conditions of the sensor manufacturing process such that the sensitivity profiles exhibited by individual sensors of a sensor population (e.g., a sensor lot) are substantially similar, and sometimes nearly identical.

[0253] What has been discovered is that not only is the change or drift in sensor sensitivity over a sensor session substantially consistent for sensors manufactured in a substantially identical manner under substantially the same conditions, but it is also possible to model this change or drift with a mathematical function that can accurately estimate it. As shown in FIG. 1A, an evaluation algorithm function 120 can be used to define the relationship between time during a sensor session and sensor sensitivity. The evaluation algorithm function can be generated by testing a sample set (including one or more sensors) from a sensor lot under in vivo and / or in vitro conditions. Alternatively, the evaluation algorithm function can be generated by testing each sensor under in vivo and / or in vitro conditions.

[0254] In some embodiments, the sensor may be subjected to an in vitro sensor sensitivity drift test, in which the sensor is exposed to a change in condition (e.g., a step change in glucose concentration in a solution) to generate an in vitro sensitivity profile of the sensor over a period of time. The duration of the test may match the entire sensor session of the corresponding in vivo sensor, or may include a portion of the sensor session (e.g., the first day, first two days, first three days, etc., of the sensor session). It is contemplated that the above test may be performed on an individual sensor, or on one or more sample sensors of a sensor lot. From this test, an in vitro sensitivity profile may be generated from which an in vivo sensitivity profile may be modeled and / or created.

[0255] From this in vivo or in vitro testing, one or more data sets can be generated and plotted, each containing data points related to sensitivity over time. A sensitivity profile or curve can then be fitted to the data points. If the curve fit is determined to be satisfactory (e.g., the standard deviation of the generated data points is less than a certain threshold), it can be determined that the sensor sensitivity profile or curve has passed quality control and is suitable for release. From there, the sensor sensitivity profile can be converted into an evaluation algorithm function or into a look-up table. The algorithm function or look-up table can be stored in a computer readable memory and accessed, for example, by a computer processor.

[0256] The evaluation algorithm function can be formed by applying curve fitting techniques that backward fit a curve to the data points by adjusting the function (e.g., by adjusting the constants of the function) until an optimal fit to the available data points is obtained. Briefly, a "curve" (i.e., a function sometimes called a "model") that relates one data value to one or more other data is fitted, and the parameters of the curve are selected such that the curve estimates the relationship of the data values. By way of example, the selection of the parameters of the curve can include the selection of coefficients of a polynomial function. In some embodiments, the curve fitting process can include an evaluation of how closely the curve determined in the curve fitting process estimates the relationship between the data values ​​to determine the optimal fit. The term "curve" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to a function or a graph of the function, which can include a rounded or straight curve (i.e., a line).

[0257] The curve may be formed by any of a variety of curve fitting techniques, such as linear least squares fitting, nonlinear least squares fitting, Nelder-Mead simplex, Levenberg-Marquardt, and variations thereof. Additionally, the curve may be fitted using any of a variety of functions, including, but not limited to, linear functions (including constant functions), logarithmic functions, quadratic functions, cubic functions, square root functions, power functions, polynomial functions, rational functions, exponential functions, sinusoidal functions, and variations and combinations thereof. For example, in some embodiments, the evaluation algorithm includes a linear function element that is assigned a first weight w1, a logarithmic function element that is assigned a second weight w2, and an exponential function element that is assigned a third weight w3. In further embodiments, the weights associated with each element may vary as a function of time and / or other parameters, while in alternative embodiments, one or more of these weights are constants as a function of time.

[0258] In some embodiments, the correlation of the evaluation algorithm function with the data obtained from the sample sensor (e.g., R value), which is a measure of the quality of the curve fit to the data points, can be one metric used to determine whether the function is optimal. In some embodiments, the evaluation algorithm function formed from the curve fit analysis can be adjusted to compensate for other parameters, such as other parameters that may affect or provide additional information regarding the sensor sensitivity. For example, the evaluation algorithm function can be adjusted to compensate for the sensitivity of the sensor to hydrogen peroxide or other chemical species.

[0259] The evaluation algorithms developed and used to accurately estimate the sensitivity of an individual sensor can be based on factory calibration and / or a single early reference measurement (e.g., using a single-point blood glucose monitor) at any time during the sensor session. In some embodiments, the entire sensor of a population of serial analyte sensors manufactured in substantially the same manner under substantially the same conditions exhibit a substantially fixed relationship of in vivo to in vitro sensitivity. For example, in one embodiment, the in vivo sensitivity of a sensor at a certain time after sensor use begins (e.g., t=5, 10, 15, 30, 60, 120, or 180 minutes after sensor use) is consistently equal to the measured in vitro sensitivity of that sensor or an equivalent sensor. From this relationship, an initial value of the in vivo sensitivity can be generated, from which an algorithmic function corresponding to the sensor sensitivity profile can be developed. In other words, from this initial value (representing one point of the sensor sensitivity profile), the entire remainder of the sensor sensitivity profile can be determined and plotted. The initial value of the in vivo sensitivity can be associated with any portion of the sensor sensitivity profile. In one embodiment, multiple initial values ​​corresponding to multiple in vitro sensitivities spaced apart in time from the in vivo sensitivity can be calculated and combined together to generate a sensor sensitivity profile.

[0260] In some embodiments, as shown in FIG. 2A, the initial value 210 of in vivo sensitivity is associated with a time corresponding to (close to) the beginning of the sensor session. Based on this initial value 210, the remaining sensor sensitivity profile 220 is plotted (i.e., plotted back and forth across the x-axis in response to time), as shown in FIG. 2B. However, as shown in FIG. 2C, in some embodiments, the initial value 210' can be associated with any other time of the sensor session. For example, as shown in FIG. 2C, in one embodiment, the initial value 210' of in vivo sensitivity is associated with a time when the sensitivity reaches a substantially steady state (e.g., about 3 days). From the initial value 210', the remaining sensor sensitivity profile 220' is plotted as shown in FIG. 2D.

[0261] According to other embodiments, the in vitro to in vivo sensitivity relationships were not equal, but the relationships nevertheless included a consistent, fixed ratio. By making the in vitro to in vivo sensitivity relationships substantially constant, some of the sensors described herein can be factory calibrated by evaluating the in vitro sensitivity characteristics of sensors from a particular sensor lot (e.g., one or more sensitivity values ​​measured over a period of time) at a manufacturing facility, determining the in vivo sensitivity characteristics of other sensors in the same sensor lot based on the relationship to the measured in vitro sensitivity characteristics, and storing the calculated in vivo sensitivity characteristics on electronics associated with the sensor (e.g., a computer memory of the sensor electronics, discussed elsewhere herein, configured to be operatively connected to the sensor during use of the sensor).

[0262] Thus, with information regarding the in vivo vs. in vitro sensor sensitivity relationship obtained prior to the sensor session and the predefined sensor sensitivity profile, factory calibration with a high level of sensor accuracy is achieved. For example, in some embodiments, the sensor was able to achieve an accuracy corresponding to a mean absolute relative difference of not more than about 10% over a sensor session of at least about 3 days, and sometimes at least about 4, 5, 6, 7, or 10 days. In some embodiments, the sensor was able to achieve an accuracy corresponding to a mean absolute relative difference of not more than about 7%, about 5%, or about 3% over a sensor session of at least about 3 days. With factory calibration, the need for recalibration can be eliminated or is only required in certain circumstances, such as in response to detection of sensor failure.

[0263] Referring back to the work associated with FIG. 1A, a sensor is manufactured with a working electrode configured to measure a glucose + baseline signal and a corresponding auxiliary electrode configured to measure only the baseline signal. The sensor electronics of the sensor system subtracts the baseline signal from the glucose + baseline signal to obtain a signal that is completely or substantially completely related to the glucose concentration. In addition, an algorithmic function is generated and stored in the sensor electronics associated with the sensors to estimate the sensitivity of the sensors over their lifetime. This algorithmic function is shown plotted in FIG. 1A and closely overlaying the measured sensor sensitivity of the sensor. By determining the baseline and sensitivity at any given time during the life of the sensor, a conversion function is formed, which converts the sensor signal to an estimated glucose concentration.

[0264] As shown in Figure 3A, which is a Bland-Altman plot illustrating the difference between the YSI reference measurements and one factory calibrated in vivo continuous analyte sensor, the measurements from these sensors demonstrate extremely high accuracy. The lines in Figure 3A represent accuracy standards that correspond to deviations from actual measurements (using the YSI test) of less than ±10 mg / dL at glucose concentrations between about 40-75 mg / dL and less than ±15% at glucose concentrations between about 75 mg / dL and 400 mg / dL. In fact, the difference between estimated glucose concentration values ​​calculated using a predefined sensor sensitivity profile and actual measurements (using the YSI test) over the life of the sensor does not exceed about 10 mg / dL at glucose concentrations between about 40-75 mg / dL and does not exceed 15% at glucose concentrations between about 75 mg / dL and 400 mg / dL. Furthermore, at glucose concentrations between about 40 mg / dL and 75 mg / dL, about 97% of the estimated glucose concentration values ​​were within ±5 mg / dL of the corresponding YSI measurement, and at glucose concentrations between about 70 mg / dL and 400 mg / dL, about 99% of the estimated glucose concentrations were within ±10% of the corresponding YSI measurement.

[0265] FIG. 3B shows the Clarke Error Grid associated with the factory calibration study associated with FIG. 3A. The Clarke Error Grid in FIG. 3B is based on a correlation plot of the performance of the factory calibrated method described above against a reference method in the form of a YSI measurement. If the correlation were perfect, all points would fall on a 45 degree line. The area surrounding this line is divided into zones that predict the clinical outcome for an action taken by the patient depending on where the factory calibrated method measurement falls on the line. Zone A corresponds to a clinically correct decision (e.g., take insulin, take glucose, or do nothing), Zone B corresponds to a clinically acceptable decision, and Zone D corresponds to a clinically incorrect decision. As shown in FIG. 3B, all of the data points from the factory calibration study fell into either Zone A or Zone B. In fact, nearly all of the data points fell into Zone A, thus establishing that the factory calibration study described above results in highly accurate glucose concentration measurements.

[0266] It has been found that individual sensors of a sensor group manufactured under substantially the same conditions generally exhibit substantially similar or nearly identical sensor sensitivity profiles and have substantially similar or nearly identical in vitro to in vivo sensor sensitivity relationships, but that sometimes the actual sensor sensitivity (i.e., sensitivity expressed as an actual sensitivity value, not a percentage of the substantially steady state sensitivity) may vary between sensors. For example, individual sensors may be manufactured under substantially the same conditions, but may have different sensitivity characteristics over the time between sensor manufacture and sensor use if, during use, they are exposed to different environmental conditions (e.g., exposure to radiation, extreme temperatures, abnormal dehydration conditions, or any environment that would damage the enzymes of the sensor membrane or other parts of the sensor, etc.).

[0267] Thus, to compensate for potential effects resulting from these conditions, in certain embodiments, the continuous analyte sensor is configured to request and accept one or more reference measurements (e.g., from a finger stick glucose measurement or from a calibration solution) at the beginning of a sensor session. For example, the request for one or more reference measurements can be made at about 15 minutes, 30 minutes, 45 minutes, 1 hour, 2 hours, 3 hours, etc. after activating the sensor. In some embodiments, the sensor electronics is configured to process and use the reference data to generate (or adjust) a sensor sensitivity profile in response to input of one or more reference measurements to the sensor. For example, if a reference measurement of glucose concentration is taken and input to the sensor at time=x, an algorithmic function of the sensor sensitivity can be generated by matching the sensor sensitivity profile at time=x to the reference measurement. Using one of the one or more reference measurements in conjunction with a predefined sensor sensitivity profile at the start of the sensor allows for self-calibration of the sensor without or with reduced need for further reference measurements.

[0268] FIG. 4 shows a Bland-Altman plot showing the difference between the YSI reference measurements and an in vivo continuous analyte sensor that received one reference measurement approximately one hour after insertion into a patient. The lines in FIG. 4 represent accuracy criteria of less than about ±20 mg / dL at glucose concentrations between about 40 mg / dL and 75 mg / dL, and less than about +20% at glucose concentrations between about 75 mg / dL and 400 mg / dL, corresponding to deviations from the actual measurements (using the YSI test). From this reference measurement, an initial value of the in vivo sensor sensitivity is generated, which in turn allows for the formation of an algorithmic function corresponding to the sensitivity profile for the remainder of the sensor session. The sensor is manufactured with a working electrode and an auxiliary electrode that are used to measure a baseline signal that is subtracted from the glucose plus baseline signal acquired by the working electrode. As shown, about 85% of the estimated glucose concentrations were within a range 410 defined as ±20 mg / dL from the corresponding YSI measurement for glucose concentrations between about 40 mg / dL and 75 mg / dL and ±20% from the corresponding YSI measurement for glucose concentrations between about 75 mg / dL and 400 mg / dL. In addition, about 95% of the estimated glucose concentration values ​​were within ±20 mg / dL of the corresponding YSI measurement for glucose concentration conditions between about 40 mg / dL and 75 mg / dL. The sensors in this study achieved a level accuracy corresponding to a mean absolute relative difference of about 12% over at least a 7-day sensor session, and a first-day accuracy level corresponding to a mean absolute relative difference of about 11%. For both the overall and first-day accuracy levels, the median absolute relative difference achieved was about 10%.

[0269] FIG. 5 illustrates different types of information that can be input to the sensor system to define a sensor sensitivity profile over a period of time in one embodiment. This input information can include information acquired prior to the sensor session 510 and information acquired during the sensor session 520. While in the embodiment illustrated in FIG. 5, both information acquired prior to the sensor session 510 and information acquired during the sensor session 520 are used to generate, adjust and update the function 530 associated with the sensor sensitivity profile, in other embodiments, the sensor system can be configured to use only information acquired prior to the sensor session. In some embodiments, the formation of the initial sensor sensitivity profile can occur prior to the sensor session, at the beginning of the sensor session, or immediately after starting the sensor session. Additionally, in some embodiments, the sensor sensitivity profile can be continuously adjusted, regenerated, or updated to compensate for parameters that may affect the sensor sensitivity or provide additional information regarding the sensor sensitivity during the sensor session. Information acquired prior to the sensor session can include, for example, a sensor sensitivity profile generated prior to or at the beginning of the sensor session, as previously described. It can also include sensitivity values ​​associated with a substantially fixed relationship of in vivo to in vitro sensor sensitivity, as previously described.

[0270] Alternatively, instead of a fixed sensitivity value, the relationship of in vitro to in vivo sensor sensitivity can be determined as a function of time between completion of sensor fabrication (or the time when a calibration test is performed on sensors from the same lot) and the start of a sensor session. As shown in FIG. 6, it has been found that the sensitivity of a sensor to an analyte concentration can change as a function of time between completion of sensor fabrication and the start of a sensor session. FIG. 6 illustrates this phenomenon with a plot line, which resembles a downward trend in sensitivity over a period of time between completion of sensor fabrication and the start of a sensor session. Similar to the discovered change or drift in sensitivity over a period of time between completion of sensor fabrication and the start of a sensor session, this change or drift in sensitivity over a period of time between completion of sensor fabrication and the start of a sensor session is generally consistent among sensors that are not only manufactured under substantially the same conditions and in substantially the same manner, but also avoid exposure to certain conditions (e.g., exposure to radiation, extreme temperatures, abnormal dehydration conditions, or any environment that may damage the enzymes of the sensor membrane or other parts of the sensor, etc.). Thus, the change or drift in sensitivity over a period of time between the completion of sensor manufacturing and the start of a sensor session may also be modeled by a mathematical function 620 that accurately estimates this change or drift. The evaluation algorithm function 620 may be any of a variety of functions, such as, for example, linear functions (including constant functions), logarithmic functions, quadratic functions, cubic functions, square root functions, power functions, polynomial functions, rational functions, exponential functions, sinusoidal functions, and combinations thereof.

[0271] Information acquired prior to a sensor session may also include information regarding characteristics or attributes of a sensor. By way of example and not limitation, information acquired prior to a sensor session may include the particular material used to manufacture the sensor (e.g., material used to form the sensor membrane), the thickness of the sensor membrane, the permeability of the membrane to glucose or other chemical species, in vivo or ex vivo sensor sensitivity profiles of other sensors manufactured in substantially the same manner under substantially the same conditions, etc. In certain embodiments, information acquired prior to a sensor session may include information regarding the process conditions under which the sensor was manufactured. This information may include, for example, the temperature at which the sensor membrane was cured, the amount of time the sensor was immersed in a particular coating solution, etc. In other embodiments, information acquired prior to a sensor session may relate to physiological information of the patient. For example, the patient's age, body mass index, sex, and / or historical patient sensitivity profile may be used as parameters to form the sensor sensitivity profile. Other information that may be acquired prior to a sensor session and used includes information regarding the insertion of the sensor, such as the location (e.g., abdominal vs. back) or the depth of sensor insertion.

[0272] In general, the sensor sensitivity function can be generated by theoretical or empirical methods or both and stored as a function look-up table, thereby enabling self-calibration of the sensor, eliminating (or substantially reducing) the need for reference measurements. The sensor sensitivity function can be generated at a manufacturing facility and shipped with the system, or generated by the system just prior to (or during) use. The term "self-calibration" as used herein is a broad term and is given its ordinary and customary meaning to those skilled in the art (and is not limited to a special or customized meaning), and it refers to, but is not limited to, calibration of a sensor or device performed by someone other than the manufacturer or installer of the control system, the manufacturer of the sensor, or the user of the sensor. The sensor calibration can be performed on the individual sensor on which the control system is mounted, or it can be performed on a reference sensor, for example one from the same sensor lot, and the calibration function can be transferred from one control system to another. In some embodiments, the system can be shipped with some self-calibration functions, and others can be added by the sensor user. Additionally, the sensor system can be shipped with a self-calibration function and only requires adjustment or touch-up calibration during use.

[0273] In certain embodiments, the sensor sensitivity profile can be adjusted during sensor use to compensate in real time for parameters that may affect or provide additional information regarding the sensor sensitivity. These parameters include, but are not limited to, parameters related to sensor attributes, such as the permeability of the sensor membrane or the level of hydration of the sensor, or parameters related to patient physiological information, such as the patient's temperature (e.g., the temperature of the sample site or the temperature of the skin), the pH of the sample site, the hematocrit level, or the oxygen level of the sample site. In some embodiments, the continuous analyte sensor can be fitted with a thermistor on the ex vivo portion of the sensor and thermally conductive traces extending from the thermistor to the sample site.

[0274] In some embodiments, the calibration method can be improved by algorithmically quantifying each unique sensor / user environment. Quantification can be achieved by generating or adjusting a sensitivity profile, which can include an inference engine including a causal probability network, such as a Bayesian network. A Bayesian network includes conditional probability-based networks that rely on the Bayes theorem to characterize the likelihood of different outcomes based on known prior probabilities (e.g., the prevalence of observed in vivo sensitivities under certain parameter conditions) and newly acquired information (e.g., the occurrence or non-occurrence of the above conditions).

[0275] In some embodiments, quantification is achieved by applying an analytical Bayesian framework to data already within the scope of the system algorithm. Quantification includes (1) wedge (e.g., maximum or minimum) values ​​for parameters related to sensitivity and baseline; (2) sensitivity values ​​calculated during the sensor session; and (3) baselines calculated during the sensor session. In some embodiments, a first set of wedge values ​​is based on parameter distributions of "prior" data (e.g., data collected from previous clinical studies). Using a Bayesian network, the system can learn from data collected during the sensor session and adapt its wedge values ​​(or other algorithm parameters and / or constraints) to the specific sensor / user environment. As a result, the calibration method of the system can be improved to have better accuracy, reliability, and overall performance.

[0276] In one embodiment, the Bayesian framework used involves establishing a framework using known prior information, followed by inference from the combination (i.e., combination of prior and new information) in light of collected new information to generate subsequent information. When prior and new information are combined to form subsequent information that is mathematically related and can be functionally represented, the relationship is defined as a conjunct. The Bayesian framework of one embodiment uses conjunct distribution relationships that result in subsequent distribution parameters (e.g., mean, variance) that are directly available. This may be advantageous in computationally limited space where time and resources are limited.

[0277] In some embodiments, under a Bayesian framework, the wedge parameters follow a priori substantially normal distributions for each parameter, and the algorithmically calculated sensitivity and baseline values ​​follow those substantially normal distributions representing new information, allowing for the generation of independent posterior distributions for sensitivity and / or baseline. These two posterior distributions are then used in parallel to create glucose thresholds that are used to accept, reject, or modify a user's manual calibration input (e.g., input from fingerstick measurements). In some embodiments, the posterior distributions of sensitivity and baseline can also be used for semi-self-calibration, thereby reducing the number of fingerstick measurements required for manual calibration.

[0278] In one exemplary embodiment, from prior information, the minimum / maximum wedge values ​​for the sensor sensitivity to glucose are known to be A and B with a 95% confidence interval in a substantially normal distribution. By using the algorithmically calculated sensitivity values ​​as new information, the previous three, four, five, six, seven or more sensitivity values ​​calculated by manual calibration can be used as new information to generate a subsequent distribution, which typically has reduced variability than the distribution based on prior information and is a direct quantification of the unique sensor / user environment. Due to the reduced variability, the subsequent distribution generated typically has wedge values ​​that are closer to A and B, i.e., the difference between the wedge values ​​of the distribution generated from the prior information. This tightening of the difference between the wedge values ​​between the prior and subsequent distributions allows the system to more reliably reject manual calibration inputs that are clearly erroneous, such as a calibration input of 60 mg / dL that was intended by the user to be an input of 160 mg / dL.

[0279] Bayesian networks use causal knowledge between different events and model probabilistic dependent or independent relationships. FIG. 7A shows a distribution curve of sensor sensitivity corresponding to a Bayesian learning process according to one embodiment. FIG. 7B shows confidence levels associated with the sensor sensitivity profile corresponding to the distribution curve shown in FIG. 7A. The distribution curve 720 and confidence levels 730 (e.g., 25%, 33%, 50%, 75%, 95%, or 99% confidence levels) are associated with a lack of initial knowledge about a parameter that affects or provides additional information about the sensor sensitivity. For example, the distribution curve 720 can be associated with factory information. As information about a parameter is obtained, the distribution curve 720' becomes steeper and the confidence interval 730' becomes narrower as the certainty of the sensor sensitivity profile 710 increases. Examples of information that can be used to change the distribution curve include reference analyte values, calibration tests of the sensor at the factory, patient medical history information, and any other information described elsewhere herein that can affect or provide information about the sensor sensitivity. As information about additional parameters is obtained, the distribution curve 720'' becomes steeper and the confidence interval 730'' becomes narrower as the certainty of the sensor sensitivity profile 710 increases further.

[0280] During use of the sensor, the confidence interval curves 730, 730' and / or 730" can be used to form a sensitivity profile 710 that provides an estimated sensitivity value at a given time. The estimated sensitivity value can then be used to calibrate the sensor, which allows the sensor data to be processed to generate a glucose concentration value to display to a user. In some embodiments, the first estimated sensitivity profile 710, formed from the confidence interval curves 730, 730' and / or 730", can be used to monitor and display glucose concentrations. Additionally, the confidence interval curves 730, 730', 730", or a combination thereof, can be used to form a second estimated sensitivity profile that is less accurate than the first estimated sensitivity profile, but is nevertheless more likely to result in the detection of a hypoglycemic or hyperglycemic range than the first estimated sensitivity profile 710.

[0281] Determining the Baseline When a sensor is implanted in a recipient, various types of noise may occur. Some implantable sensors measure a signal (e.g., counts) that is composed of two components: a baseline signal and an analyte signal. The baseline signal is composed substantially of signal contributions from factors other than the measured analyte (e.g., interfering species, non-reacted associated hydrogen peroxide, or other electroactive species that have an oxidation potential that overlaps with the analyte or co-analyte). The analyte signal (e.g., glucose signal) is composed substantially of signal contributions from the analyte. Thus, since the signal includes these two components, a calibration can be performed to determine the analyte (e.g., glucose) concentration by solving the equation y=mx+b, where the value of b represents the baseline of the signal. In some situations, the baseline is composed of constant and non-constant non-analyte factors. In general, it is desirable to reduce or remove the background signal to provide a more accurate analyte concentration to the patient or health care professional.

[0282] In one embodiment, the analyte sensor (e.g., a glucose sensor) is configured for insertion into a host for measuring an analyte within the host. The sensor includes a working electrode disposed under an active enzyme portion of a membrane on the sensor, an auxiliary electrode disposed under an inactive or non-enzyme portion of the membrane on the sensor, and sensor electronics operably connected to the working and auxiliary electrodes. The sensor electronics is configured to process signals from the electrodes to generate an estimate of the analyte (e.g., glucose) concentration that substantially excludes signal contributions from non-glucose related noise artifacts.

[0283] In some embodiments, the working electrode is configured to generate, via the sensor electronics, a first signal related to the analyte and a non-analyte-related electroactive compound having an oxidation potential less than or similar to the first oxidation potential. The auxiliary electrode is configured to generate a second signal related to the non-analyte-related electroactive compound. The non-analyte-related electroactive compound is any compound present in the local environment of the sensor and has an oxidation potential less than or similar to the oxidation potential of the species being measured (e.g., H2O2). Without wishing to be bound by theory, it is believed that both a signal directly related to the enzyme-catalyzed reaction of glucose (producing oxidized H2O2 at the first working electrode) and a signal from unknown compounds in the extracellular environment surrounding the sensor can be measured by the electrode measuring glucose. These unknown compounds may or may not be constant (e.g., intermittent or transient) in concentration and / or effect. In some circumstances, some of these unknown compounds are believed to be related to a disease state of the recipient. For example, it is known that blood chemistry changes dramatically during / after a heart attack (e.g., changes in pH, changes in concentrations of various blood components / proteins, etc.). As another example, the transdermal insertion of a needle-type sensor may initiate a sequence of events including the release of various reactive molecules by macrophages. Other components that may contribute to non-glucose related signals are compounds associated with the wound healing process, which may be initiated by the implantation / insertion of the sensor into the recipient, as described in more detail with reference to US Pat. No. 6,399,623.

[0284] As described above, the auxiliary electrode is configured to generate a second signal related to a non-analyte-associated electroactive compound having an oxidation potential less than or similar to the first oxidation potential. Non-analyte-associated electroactive species may include interfering species, non-reactive associated species corresponding to the measured species (e.g., H2O2), and other electroactive species. Interfering species include any compounds that are not directly related to the electrochemical signal generated by the enzyme-catalyzed reaction of the analyte, such as electroactive species in the local environment generated by other body processes (e.g., cellular metabolism, wound healing, disease processes, etc.). Non-reactive associated species include any compounds from sources other than the enzyme-catalyzed reaction, such as H2O2 released by nearby cells during the course of cellular metabolism, H2O2 generated by other enzyme reactions (e.g., extracellular enzymes that may be released during the death of cells surrounding or nearby the sensor, or that may be released by activated macrophages), etc. Other electroactive species include any compounds that have an oxidation potential less than or similar to that of H2O2.

[0285] Non-analyte signals generated by compounds other than the analyte (e.g., glucose) are considered background noise and can obscure the analyte-related signal, thereby contributing to sensor inaccuracies. As described in more detail elsewhere herein, background noise includes both constant and non-constant components and can be removed to accurately calculate the analyte concentration.

[0286] In certain embodiments, the analyte sensor system is configured (e.g., symmetrical, coaxial design, and / or integral construction) such that the working and auxiliary electrodes are affected by substantially the same external / environmental factors, thereby allowing substantially equivalent measurements of constant and non-constant species / noise. This allows for substantial elimination of noise on the sensor signal using the sensor electronics described elsewhere herein. As a result, substantial reduction or elimination of the effects of noise-related signals, including non-constant noise (e.g., transient, unpredictable biologically relevant noise), increases the accuracy of the continuous sensor signal.

[0287] In some embodiments, the sensor electronics is operatively connected to the working and auxiliary electrodes. The sensor electronics can be configured to measure a current (or voltage) to generate first and second signals. Collectively, the first and second signals can be used to generate glucose concentration data without substantial signal contribution from non-glucose related noise. This can be done, for example, by subtracting the second signal from the first signal to generate a signal related to the analyte concentration without substantial noise contribution, or by alternative data analysis techniques.

[0288] In other embodiments, auxiliary electrodes are not required, and the sensor electronics are operatively connected only to one or more working electrodes. For example, the sensor membrane of some embodiments can be constructed from a polymer and include a mediator and an enzyme chemically attached to the polymer. The mediator used can be oxidized at a lower potential than hydrogen peroxide, and thus few interferents that are oxidizable at these low potentials are oxidized. Thus, in some embodiments, a very low baseline (i.e., a baseline that approaches a zero baseline and substantially does not accept signal contributions from non-glucose-related noise) can be achieved, thereby potentially eliminating (or reducing) the need for auxiliary electrodes to measure signal contributions from non-glucose-related noise.

[0289] The sensor electronics may include a potentiostat, an A / D converter, RAM, ROM, a transceiver, a processor, and / or others. The potentiostat may be used to provide a bias to the electrodes and convert raw data (e.g., raw counts) collected from the sensor into an analyte concentration value (e.g., a glucose concentration value expressed in units of mg / dL). A transmitter may be used to communicate the first and second signals to a receiver, where additional data analysis and / or calibration of the analyte concentration may be processed. In certain embodiments, the sensor electronics may perform additional operations, such as data filtering and noise analysis.

[0290] In some embodiments, the sensor electronics can be configured to analyze an analyte-equivalent baseline or normalized baseline instead of the baseline. The normalized baseline is calculated as the y-intercept divided by b / m or slope (from the equation y=mx+b). The units of the analyte-equivalent baseline can be expressed as units of analyte concentration (mg / dL) related to the output of the continuous analyte sensor. Using an analyte-equivalent baseline (normalized baseline) can eliminate the effect of glucose sensitivity on the baseline, thereby allowing the baselines of different sensors (e.g., from the same sensor lot or from different sensor lots) to be evaluated with different glucose sensitivities.

[0291] While some embodiments are described herein that use an auxiliary electrode to allow subtraction of the baseline signal from the glucose+baseline signal, it should be understood that the use of this electrode is optional and may not be used in other embodiments. For example, in some embodiments, a membrane system covering the working electrode can substantially block interferents, substantially reducing the baseline to a negligible level, allowing the baseline to be estimated. The baseline estimation can be based on an assumption of the sensor's baseline in physiological conditions associated with a typical patient. For example, the baseline estimation can be modeled after in vivo or ex vivo measurement of the baseline according to certain physiological levels of interferents characteristic of the body. FIG. 8 is a graph providing a comparison between the estimated glucose equivalent baseline and the detected glucose equivalent baseline according to one study. The estimated glucose equivalent baseline was formed by conducting ex vivo measurements of the glucose sensor's baseline in a solution that mimics physiological levels of interferents in humans. The interferents included uric acid and ascorbic acid at concentrations of about 4 mg / dL, with the ascorbic acid concentration being about 1 mg / dL. In this study, the estimated baseline was found to closely resemble the detected baseline, as shown in Figure 8. Thus, with the possibility of accurate assessment of the baseline and / or with a negligible baseline, a single working electrode alone (i.e., without the use of an auxiliary electrode) for a predefined sensor sensitivity profile is sufficient to provide self-calibration for the sensor system.

[0292] In accordance with some embodiments, it has been found that not only does the sensitivity of a sensor tend to drift over time, but the baseline of the sensor also drifts over time. Thus, in certain embodiments, the concepts behind the methods and systems used to predict drift in sensitivity can be applied to generate models for predicting drift in baseline over time. Without wishing to be bound by theory, it is believed that the overall signal received by the sensor electrode is composed of a glucose signal component, an interference signal component, and an electrode-associated baseline signal component that is a function of the electrode and is substantially independent of the environment surrounding the electrode (e.g., extracellular matrix). As noted above, the term "baseline" as used herein refers to, but is not limited to, the component of the analyte sensor signal that is not related to the analyte concentration. Thus, the baseline as the term is defined herein is composed of the interference signal component and the electrode-associated baseline component signals. Without wishing to be bound by theory, it is believed that an increase in membrane permeability typically results in an increase in the glucose diffusion rate across the sensor membrane, as well as an increase in the interferent diffusion rate across the sensor membrane. Thus, changes in sensor membrane permeability over time that cause sensor sensitivity drift can also cause baseline interference signal components to drift. Simply put, baseline interference signal components are not static and typically change as a function of time, which in turn causes baseline drift over time. By analyzing how each of the above-mentioned components of the baseline respond to changes in conditions and time (e.g., as a function of time, temperature), predictive models can be developed to predict how much a sensor's baseline will drift during a sensor session. The ability to predict sensor sensitivity and baseline prospectively is believed to enable self-calibrating continuous analyte sensors, i.e., sensors that do not require the use of reference measurements (e.g., fingerstick measurements) for calibration.

[0293] Calibration Code The process of manufacturing continuous analyte sensors may sometimes be subject to some degree of sensor lot-to-lot variation. To compensate for this variation, one or more calibration codes may be assigned to each sensor or set of sensors to define parameters that affect sensor sensitivity or provide additional information regarding the sensitivity profile. The calibration codes may reduce the variation between different sensors, and by applying an algorithm to adjust for the differences, ensure that results obtained from using sensors from different sensor lots are generally equivalent and consistent. In one embodiment, the analyte sensor system may be configured such that one or more calibration codes are manually entered into the system by a user. In another embodiment, the calibration codes may be part of a calibration-encoded label that is affixed to (or inserted into) a package of multiple sensors. The calibration-encoded label itself may be readable or interrogated by any of a variety of technologies, including, but not limited to, optical technologies, RFID (radio frequency identification), and others, and combinations thereof. These techniques for transferring the codes to the sensor system may be more automatic, accurate, and convenient for the patient, and less prone to error compared to manual entry. Manual entry has an inherent risk of error, for example caused by the patient or hospital staff entering an incorrect code, which may lead to a false calibration and therefore an inaccurate glucose concentration reading, which may result in the patient or hospital staff taking inappropriate action (e.g., being in a hypoglycemic state and injecting insulin).

[0294] In some embodiments, the calibration codes assigned to the sensor may include a first calibration code associated with a predefined logarithmic function corresponding to the sensitivity profile, a second calibration code associated with an initial in vivo sensitivity value, and other calibration codes, each defining a parameter that affects or provides information about the sensor sensitivity. The other calibration codes may be associated with any a priori information or parameters described elsewhere herein and / or any parameters that help define the mathematical relationship between the measured signal and the analyte concentration.

[0295] In some embodiments, packaging used to store and transport a continuous analyte sensor (or sensor set) includes a detector configured to measure a parameter that can affect sensor sensitivity or provide additional information regarding sensor sensitivity or other sensor characteristics. For example, in one embodiment, the sensor package can include a temperature detector configured to provide calibration information regarding whether the sensor has been exposed to a temperature condition above (and / or below) one or more predetermined temperature values. In some embodiments, the one or more predetermined temperature values ​​can be greater than about 75 degrees Fahrenheit, greater than about 80 degrees Fahrenheit, greater than about 85 degrees Fahrenheit, greater than about 90 degrees Fahrenheit, greater than about 95 degrees Fahrenheit, greater than about 100 degrees Fahrenheit, greater than about 105 degrees Fahrenheit, and / or greater than about 110 degrees Fahrenheit. Additionally or alternatively, the one or more predetermined temperature values ​​can be values ​​less than about 75 degrees Fahrenheit, less than about 70 degrees Fahrenheit, less than about 60 degrees Fahrenheit, less than about 55 degrees Fahrenheit, less than about 40 degrees Fahrenheit, less than about 32 degrees Fahrenheit, less than about 10 degrees Fahrenheit, and / or less than about 0 degrees Fahrenheit. In certain embodiments, the sensor package can include a humidity exposure indicator configured to provide calibration information regarding whether the sensor has been exposed to a humidity level that is greater than or less than the one or more predetermined humidity values. In some embodiments, the one or more predetermined humidity values ​​can be values ​​greater than about 60% relative humidity, greater than about 70% relative humidity, greater than about 80% relative humidity, and / or greater than about 90% relative humidity. Alternatively or additionally, the one or more predetermined humidity values ​​can be values ​​less than about 30% relative humidity, less than about 20% relative humidity, and / or less than about 10% relative humidity.

[0296] When exposure of the sensor to a certain level of temperature and / or humidity is detected, the corresponding calibration code can be altered to compensate for the possible effect of this exposure on the sensor sensitivity or other sensor characteristics. This alteration of the calibration code can be performed automatically by a control system associated with the sensor package. Alternatively, in other embodiments, an indicator (e.g., a color change indicator) adapted to change (e.g., change color) upon exposure to an environment can be used. By way of example and not limitation, the sensor package can include an indicator that irreversibly changes color from blue to red upon exposure of the package to a temperature greater than about 85 degrees Fahrenheit, and includes instructions to the user to enter a calibration code when the indicator is red in color. Although exposure to temperature and humidity is described herein as an example of a condition that can be detected by the sensor package and used to trigger an alteration of the calibration code information, it should be understood that other conditions can also be detected and used to trigger a change of the calibration code information.

[0297] In some embodiments, the continuous analyte system can comprise a library of stored sensor sensitivity or calibration functions associated with one or more calibration codes. Each sensitivity or calibration function results in calibrating the system to a different set of conditions. The different conditions during sensor use can be related to temperature, body mass index, various conditions, or any of the parameters that can affect or provide additional information regarding the sensor sensitivity. The library can also include sensitivity profiles or calibrations for different types of sensors or different sensor lots. For example, a single sensitivity profile library can include sub-libraries of sensitivity profiles for different sensors manufactured from different sensor lots and / or with different design configurations (e.g., different design configurations customized for patients with different body mass indexes).

[0298] Determining sensor attributes and calibrating sensor data using one or more stimulus signals - Patents.com To determine attributes of the sensor and / or calibrate the sensor data, some embodiments apply one or more stimulus signals to the sensor. As used herein, the term "stimulus signal" is a broad term that is given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a signal (e.g., a time-varying or space-varying quantity such as voltage, current, or field strength) applied to a system (e.g., an analyte sensor) used to produce or elicit a response.

[0299] Non-limiting examples of stimulation signals that can be used in the embodiments described herein can be waveforms including one or more of: a step increase in voltage of a first magnitude, a step decrease in voltage of a second magnitude (where the first and second magnitudes can be the same or different), an increase in voltage at a first rate for a period of time, a gradual decrease in voltage at a second rate for a period of time (where the first and second rates can be different or the same), one or more sine waves of the same or different frequency and / or amplitude superimposed on the input signal. Responses to the stimulation signals can then be measured and analyzed (the responses are also referred to herein as "signal responses"). The analysis can include one or more of calculating impedance values, capacitance values, and relating the signal responses to one or more predetermined relationships. The term "impedance value" as used herein can refer to a value that represents an electrical impedance, including, but not limited to, a value that represents only the magnitude of the impedance or a value that represents the magnitude and phase of the impedance when the impedance is represented in a polar form, and a value that represents only the real impedance or both the real and complex impedance when the impedance is represented in a Cartesian form. Based on the calculated impedance values, capacitance values ​​and / or predetermined relationships, various sensor attributes may be determined and / or characterized, such as one or more of the sensor attributes discussed herein.

[0300] The sensor information can then be used to determine if the analyte sensor is functioning properly and / or to calibrate the sensor. For example, the techniques described herein can be used to generate calibration information (e.g., one or more of baseline, sensor sensitivity, and temperature information), which is then used to create or modify a conversion function or calibration factor used to convert sensor data (e.g., in units of current) to blood glucose data (e.g., glucose concentration values ​​in units of mg / dL or mmol / L), as described in more detail elsewhere herein. The sensor information can alternatively or additionally be used to first correct uncalibrated sensor data (e.g., raw sensor data), and then apply a conversion function to convert the corrected, uncalibrated data to calibrated sensor data (e.g., glucose concentration values ​​in units of glucose concentration).

[0301] For example, one technique that can be used to determine attributes of a system of use (e.g., an analyte sensor) is Electrochemical Impedance Spectroscopy (EIS). EIS is an electrochemical technique based on measuring the electrical impedance of a system of use over a range of different frequencies. Changes in the system of use may reflect changes in the frequency spectrum. As an example, if the sensitivity of a system of use changes over a period of time, a decrease in impedance at a particular frequency may be observed over that period of time. As discussed further below, other techniques can also be used to determine attributes of a system of use.

[0302] As one illustrative example of how a stimulation signal can be used to determine a sensor attribute, reference is now made to the schematic diagram of an equivalent sensor circuit model 900 shown in FIG. 9. The sensor circuit model 900 can represent the electrical properties of an analyte sensor, such as an embodiment of a continuous glucose sensor. The circuit 900 includes a working electrode 904 and a reference electrode 902. Operably connected in series with the reference electrode is Rsolution, which represents the resistance of the bulk between the working and reference electrodes. This bulk can be a liquid or other medium within which the sensor is placed, such as a buffer solution in bench test research examples, but in an example application of a sensor placed subcutaneously, this bulk can represent the resistance of the subcutaneous tissue between the working and reference electrodes. Operably connected to Rsolution is Cmembrane, which represents the capacitance of the sensor membrane, and Rmembrane, which represents the resistance of the sensor membrane. A parallel network of Cdouble layers and Rpolarization is operably connected to Rmembrane. The parallel network of Cdouble layer and Rpolarization represents the reactions occurring on the surface of the platinum interface of the working electrode. In particular, Cdouble layer represents the charge built up when the platinum electrode is in the bulk, and Rpolarization is the polarization resistance of the electrochemical reactions occurring at the platinum interface.

[0303] FIG. 10 is a Bode plot of an analyte sensor of one embodiment (i.e., |Z real | vs logω, where Z real(ω=2πf, where f is frequency). The analyte sensor may have the attributes of the sensor circuit model 900 of FIG. 9. Returning to the Bode plot of FIG. 10, the x-axis is the frequency of the stimulus signal applied to the analyte sensor, and the y-axis is the impedance resulting from the response signal of the analyte sensor. Without wishing to be bound by theory, it is believed that different frequencies may be used to measure or determine different material attributes of the sensor. For example, in the diagram of FIG. 10, an impedance value having a frequency of about 7 Hz resulting from the measured response to the input signal may represent the Cdouble layer, a frequency of about 1 kHz may represent the Rmembrane, and a frequency in the range of about 10-20 kHz may represent the Cmembrane.

[0304] Based on this information, the state of a particular attribute of the sensor can be determined by applying a stimulus signal to the sensor at a particular frequency or including multiple frequencies and determining the sensor impedance based on the signal response. For example, the capacitance of a sensor having the characteristics of the Bode plot of Figure 10 can be determined using a stimulus signal at a frequency in the range of about 1 Hz to 100 Hz, e.g., 10 Hz or greater than 10 kHz. Additionally, the resistance of the sensor can be determined using a stimulus signal at a frequency in the range of about 100 Hz to 10 kHz, e.g., 1 kHz.

[0305] FIG. 11 is a flow chart illustrating a process 1100 for determining the impedance of a sensor, according to one embodiment. In step 1102, a stimulation signal in the form of an effective current (AC) voltage of a given frequency is applied to the working electrode of the sensor being studied. The AC voltage can be superimposed on a bias potential and can be relatively small compared to the bias voltage, such as a voltage in the range of about 1% to 10% of the bias voltage. In one embodiment, the AC voltage is a sine wave with an amplitude in the range of 10 to 50 mV and a frequency in the range of 100 to 1 kHz. The sine wave can be superimposed on a bias potential of 600 mV. The response signal (e.g., in units of current) can then be measured in step 1104 and analyzed in step 1106 to determine the impedance at the given frequency. If the impedance of the sensor over a range of frequencies is of interest, the process 1100 can be repeated by applying an AC voltage at each frequency of interest and analyzing the corresponding output response.

[0306] 12, a process is described according to one embodiment for determining the impedance or impedances of a sensor under study by applying one or more stimulus signals and transforming the response signal into the frequency domain. The data can be transformed into the frequency domain using a Fourier transform technique such as fast Fourier transform (FFT), discrete time Fourier transform (DTFT) or others. In step 1202, a stimulus signal in the form of a voltage step can be applied to a bias voltage of the sensor. The voltage step can be in the range of 10-50 mV, for example 10 mV, and the bias voltage can be 600 mV. Then, in step 1204, a signal response can be measured and recorded (e.g., output current), and in step 1206, a derivative of the response can be obtained. In step 1208, a Fourier transform of the derivative of the response can be subsequently calculated to obtain the alternating current in the frequency domain. The impedance or impedances of the sensor over a wide spectrum of frequencies can be calculated based on the alternating current in step 1210.

[0307] 13 is a flow chart of a process 1300 for determining the impedance of a studied sensor, such as the impedance of a sensor's membrane, according to one embodiment. In step 1302, a stimulus signal in the form of a voltage step above a bias voltage is applied to the sensor. The signal response is measured in step 1304, and the peak current of the response is determined in step 1306. One or more impedance characteristics (e.g., resistance) of the sensor's membrane (e.g., Rmembrane) are then calculated based on the peak current in step 1308. The one or more impedance characteristics can then be associated with an attribute of the sensor.

[0308] In an alternative embodiment, instead of calculating the sensor impedance based on the peak current, the peak current may be related to one or more predefined sensor relationships to determine an attribute of the sensor, such as sensor sensitivity, i.e., in an alternative embodiment, the step of calculating one or more impedance characteristics is omitted.

[0309] The relationship between the signal response resulting from a stimulation signal in the form of a voltage step and the sensor membrane resistance of embodiments of the analyte sensor is discussed further herein with reference to FIGS. 14A, 14B and 9.

[0310] 14A is a graph of an input voltage 1400 applied to an analyte sensor over a period of time, according to one embodiment. The input voltage 1400 initially applied to the analyte sensor corresponds to a bias voltage, which in one embodiment is approximately 600 mV. Subsequently, a stimulus signal in the form of a voltage step is applied to the input voltage at time t1. The magnitude of the voltage step Δv can be in the range of 10-50 mV, for example 10 mV.

[0311] Figure 14B is a graph of the current response 1402 of the analyte sensor to the input voltage 1400 of Figure 14A. As shown in Figure 14B, the current response 1402 can include a sharp spike in current beginning at time t2, which corresponds to the time the voltage step begins to affect the response. The current response 1402 has a peak current at point 1404, and then the current response 1402 gradually decreases and stabilizes at a slightly higher level due to the increase in input voltage 1400 compared to before the voltage step.

[0312] In one embodiment, the change in current, Δi, measured as the difference between the magnitude of the current response 1402 prior to the voltage step and the magnitude of the peak current 1404 resulting from the voltage step, can be used to estimate the sensor membrane resistance, such as Rmembrane in Figure 9. In one embodiment, the estimated sensor membrane resistance can be calculated based on Ohm's law, where: Rmembrane = △v / △i It is.

[0313] As stated above, Δv is the voltage step increase, and Δi is the change in current response due to the step voltage increase.

[0314] Without wishing to be bound by theory, it is believed that certain embodiments of the sensor provide a direct relationship between the change in current in response to a voltage step to the impedance characteristics (eg, resistance) of the sensor's membrane.

[0315] As a non-limiting example of such a relationship, the following description returns to FIG. 9 and refers to the sensor circuit model 900. In some embodiments, the capacitance Cmembrane is much smaller than the capacitance Cdouble layer. For example, the value of Cmembrane is about 1 / 1000 smaller than the Cdouble layer. The bulk resistance Rsolution of the sensor circuit 900 is typically much smaller than the resistance Rmembrane, and the resistance Rpolarization may be very large, such as about 3 megaohms. With such sensor attributes, a voltage step Δv applied to the circuit 900 causes a current to flow substantially through the circuit 900 along a path from conductor 902 through Rsolution, Rmembrane, the Cdouble layer, and finally to conductor 904. Specifically, the capacitance resistance of Cmembrane is initially very large due to the voltage step, since the capacitance resistance is inversely proportional to capacitance and frequency, and theoretically the voltage step is at a very high frequency. Substantially all of the current flows through Rmembrane and not through Cmembrane due to the high capacitance resistance of Cmembrane. Furthermore, the current will flow substantially through the Cdouble layer instead of Rpolarization because the capacitive resistance of Cdouble is initially small due to the voltage step (the high capacitance value of the Cdouble layer results in a low capacitive resistance at high frequencies, e.g., voltage steps) and the relatively large resistance of Rpolarization. Thus, the initial total resistance through which substantially all of the current flows through the circuit 900 when a step voltage is applied to the circuit 900 can be summarized as Rsolution plus the series resistance of Rmembrane. However, since Rsolution is much smaller than Rmembrane in this example, the total resistance can be estimated as the resistance of the membrane, Rmembrane.

[0316] It has been found that since the resistance of the membrane Rmembrane at the time of the voltage step is essentially the resistance of the circuit 900, the value of Rmembrane can be estimated using Ohm's Law using the known values ​​of the step increase Δv and the measured change Δi in the current response due to the voltage step.

[0317] Sensitivity As discussed herein, the sensitivity of a sensor to analyte concentration during a sensor session will often change as a function of time. This change in sensitivity may be manifested as an increase in current for a particular level of sensitivity. In some embodiments, the sensitivity increases as a relative change of tens of percent over the first 24-48 hours. System calibration using a reference measurement (e.g., a strip-based blood glucose measurement) may be required to provide the user with an accurate analyte concentration measurement. Typically, the calibration rate can be once, twice, or more per day.

[0318] As discussed further below, a relationship between sensitivity and impedance has been observed in embodiments of the analyte sensor. Without wishing to be bound by theory, it is believed that embodiments of the analyte sensor have a relationship between the impedance of the sensor membrane and the diffusion coefficient of the membrane. For example, a change in the impedance of the analyte sensor may indicate a proportional change in the diffusion coefficient of the analyte sensor membrane. Furthermore, an increase in the diffusion coefficient causes an increase in the transport of the analyte being measured (e.g., glucose) through the membrane, resulting in an increase in the sensor output current. That is, a change in the diffusion coefficient may result in a proportional change in the sensor sensitivity. It should be noted that, depending on the characteristics of the sensor and the environment in which the sensor is used, other factors besides the change in the diffusion coefficient of the sensor membrane may actually contribute to the change in sensitivity.

[0319] The relationship between sensitivity and impedance can be used to estimate a sensor's sensitivity value and / or to correct for changes in the sensor's sensitivity over a period of time, resulting in improved accuracy, reduced calibration requirements, or both. In addition to detecting sensitivity, some embodiments can detect other characteristics of the analyte sensor system based on measurements of electrical impedance over one or more frequencies. These characteristics include, but are not limited to, temperature, moisture ingress into the sensor electronics, and damage to the sensor membrane.

[0320] In some exemplary embodiments, the relationship between the impedance of the sensor and the sensitivity of the sensor can be used to calculate and compensate for changes in sensitivity of the analyte sensor. For example, a change in impedance of the analyte sensor may correspond to a proportional change in the sensitivity of the sensor. In addition, the absolute value of the impedance of the analyte sensor can correspond to the absolute value of the sensitivity of the analyte sensor, and a corresponding sensitivity value can be determined based on a predetermined relationship determined from prior studies of similar sensors. The sensor data can be compensated for changes in sensitivity based on the relationship between impedance and sensitivity.

[0321] FIG. 15 is a flow chart of an exemplary process 1500 for compensating sensor data for changes in sensitivity, according to one embodiment. In step 1502, a stimulus signal may be applied to the sensor, such as a signal having a given frequency, which may be used to determine the impedance of the sensor's membrane, as discussed with reference to FIG. 11. A response to the applied signal is then measured in step 1504, and the impedance of the sensor's membrane is determined based on the response in step 1506. The determined impedance is then compared to an established impedance vs. sensor sensitivity relationship in step 1508. This established relationship may be determined from prior studies of analyte sensors that exhibit a similar impedance vs. sensor sensitivity relationship to the currently used analyte sensor, e.g., sensors manufactured in substantially the same manner under substantially the same conditions as the currently used sensor. In step 1510, the sensor signal (e.g., in units of current or counts) of the currently used sensor is modified using the impedance vs. sensitivity relationship. Then, in step 1512, an estimated analyte concentration value is calculated based on the modified sensor signal, e.g., using a conversion function. This estimated analyte concentration value can then be used for further processing and / or output, such as triggering an alarm, displaying information representing the estimated value on a user device, and / or outputting information to an external device.

[0322] It should be understood that process 1500 is one example of using sensor impedance to compensate for changes in sensor sensitivity, and various modifications may be made to process 1500 that fall within the scope of the embodiment. For example, an established impedance-sensitivity relationship may be used to determine a sensitivity value for a sensor in use. The sensitivity value may then be used to modify or form a conversion function used to convert the sensor signal of the sensor in use to one or more estimated glucose concentration values. Additionally, instead of calculating an impedance value based on the stimulation signal response, one or more attributes of the stimulation signal response (e.g., peak current value, counts, etc.) may be directly related to sensitivity based on a predetermined relationship between the stimulation signal attribute and sensitivity.

[0323] Some embodiments use one or more impedance values ​​of a sensor to create, modify, or select a sensitivity profile for an analyte sensor. As described above, a sensor may have a sensitivity profile that indicates changes in the sensitivity of the sensor over a period of time. Although sensors manufactured in substantially the same manner under substantially the same conditions may exhibit similar sensitivity profiles, the profile may still vary. For example, the environment in which a particular sensor is used may cause the sensitivity profile of the sensor to differ from other similar sensors. Thus, some embodiments may select a sensitivity profile from a number of predefined sensitivity profiles based, for example, on a correlation between the calculated impedance value or values ​​and the selected sensitivity profile. Additionally, some embodiments modify a sensor sensitivity profile already associated with an analyte sensor to be used based on the one or more impedance values ​​to more closely predict the sensitivity profile of the sensor.

[0324] FIG. 16 is a flow chart of an exemplary process 1600 for determining a predicted sensitivity profile using one or more sensor membrane impedance measurements, according to one embodiment. In step 1602, a stimulus signal is applied to an analyte sensor of interest, and a response is measured in step 1604. One or more sensor membrane impedance values ​​are then calculated based on the response in step 1606. Various techniques for calculating sensor membrane impedance values ​​based on the response that can be used in process 1600, such as one or more of the techniques discussed with reference to FIGS. 11-14, are described elsewhere herein. A sensitivity profile is then determined in step 1608 based on the one or more calculated impedance values. Process 1600 (which may include retrospective correction and / or prospective calculation) then calculates an estimated analyte concentration value using the determined sensitivity profile. The estimated analyte concentration value may then be used for further processing and output, such as displaying information representing the estimated value on a user device and / or outputting information to an external calculator.

[0325] In addition to step 1608, various techniques can be used to determine the sensitivity profile. One exemplary technique is to compare one or more calculated impedance values ​​to a number of different predicted sensitivity profiles and select the predicted sensitivity profile that best fits the one or more calculated impedance values. The number of different predicted sensitivity profiles can be predetermined and stored, for example, in a computer memory of the sensor electronics. Other techniques that can be used include using an evaluation algorithm to predict or determine a sensitivity profile based on one or more calculated impedance values. Further techniques include determining the sensitivity profile by modifying a sensitivity profile associated with the sensor being used (e.g., a sensor profile previously used to generate an estimated glucose value using the sensor). Modifying the sensitivity profile includes using an evaluation algorithm to modify the sensitivity profile to more closely match the sensitivity profile of the sensor being used based on one or more calculated impedance values.

[0326] In some embodiments, to determine whether a sensor is functioning properly, one or more impedance values ​​of the analyte sensor being used are compared to a predetermined or predicted sensitivity profile associated with the sensor. As described above, a sensor may be predicted to have a particular sensitivity profile based, for example, on a study of sensitivity changes over a period of time for sensors manufactured in substantially the same manner and used under substantially the same conditions. However, if the sensor is deemed not to adequately track its predicted sensitivity profile based on sensitivity derived from impedance measurements of the sensor, e.g., due to improper sensor insertion, damage to the sensor during shipping, manufacturing defects, etc., it may be determined that the sensor is functioning improperly. In other words, if one or more impedance values ​​of the sensor membrane do not adequately correspond to the predicted sensitivity profile of the sensor (e.g., since the impedance of the sensor membrane can be indicative of the sensitivity of the sensor), it may be determined that the sensor is not functioning properly.

[0327] FIG. 17 is a flow chart of an exemplary process 1700 for determining whether an analyte sensor in use is functioning properly based on a predicted sensitivity profile and one or more impedance measurements. At step 1702, a stimulus signal is applied to the analyte sensor in use, and a response is measured at step 1704. One or more sensor membrane impedance values ​​are then calculated at step 1706 based on the signal response. Various stimulus signals and techniques for calculating sensor membrane impedance values ​​based on the signal response that can be used in process 1700, such as any one of the techniques discussed with reference to FIGS. 11-14, are described elsewhere herein. Process 1700 then determines a correspondence between the one or more calculated impedance values ​​and the sensitivity profile at step 1708. At decision step 1710, process 1700 then queries whether the one or more calculated impedance values ​​sufficiently correspond to the predicted sensitivity profile. If it is determined that the one or more calculated impedance values ​​sufficiently correspond to the predicted sensitivity profile, process 1700 continues by verifying proper operation of the analyte sensor in use. If found appropriate at step 1710, process 1700 may be repeated after a predetermined delay ranging from about one minute to one day, e.g., about 10 minutes, one hour, 12 hours, or one day. However, if it is determined that one or more calculated impedance values ​​do not sufficiently correspond to the predicted sensitivity profile, process 1700 may initiate an error routine 1712. Error routine 1712 may include one or more of activating an audible alarm, displaying an error message on a user display, ceasing to display the sensor data on the user display, sending a message to a remote communication device over a communication network, such as a mobile phone over a cellular telephone network or a remote computer over the Internet, etc.The error routine may include, for example, modifying the predicted sensitivity profile based on one or more impedance measurements or selecting a new predicted sensitivity profile based on one or more impedance measurements. The modified or new predicted sensitivity profile may be a sensitivity profile that more closely corresponds to the sensitivity changes of the sensor being used when compared to an unmodified or previously used predicted sensitivity profile based on one or more impedance measurements.

[0328] In addition to step 1708 of process 1700, various statistical analysis techniques can be used to determine correspondence between one or more impedance values ​​and the predicted sensitivity profile. As one example, correspondence can be determined based on whether a sensitivity value derived from a calculated impedance value (e.g., derived from a predetermined relationship between impedance and sensitivity) differs from a predicted sensitivity value determined from the predicted sensitivity profile by more than a predetermined threshold amount. The predetermined threshold amount can be by absolute value or percentage. As another example, correspondence can be determined based on a data association function. As used herein, the term "data association function" is a broad term and is used in its ordinary sense, including, but not limited to, statistical analysis of data and in particular correlation to or deviation from a particular curve. Data association functions can be used to indicate association of data. For example, sensor sensitivity data derived from impedance measurements described herein can be mathematically analyzed to determine correlation to or deviation from a curve (e.g., a line or set of lines) that defines the sensitivity profile of the sensor, and this correlation or deviation is association of the data. Examples of data association functions that may be used include, but are not limited to, linear regression, non-linear mapping / regression, rank (e.g., non-parametric) correlation, least mean squares fit, mean absolute deviation (MAD), and mean absolute relative difference. In one such example, the correlation coefficient of a linear regression represents the amount of data association of the sensitivity data derived from impedance measurements from a sensitivity profile, and thus the quality of the data and / or sensitivity profile. Of course, other statistical analysis methods that determine the correlation of one or more points to a curve in addition to those described herein can be used in process 1700.

[0329] As described above, the processes 1600 and 1700 can use one or more impedance values. When more than one impedance value is used, each impedance value can be spaced in time from the other impedance values. In other words, one impedance value can be taken at an initial time t1 (representing the sensor impedance at time t1), a second impedance value can be taken at a second, later time t2 (representing the sensor impedance at time t2), and a third impedance value can be taken at a third, still later time t3 (representing the sensor impedance at time t3). Furthermore, the time between t1 and t2 can be a first time period, and the time between t2 and t3 can be a second time period that is the same or different from the first time period. The time-spaced impedance values ​​can then be used separately or combined using a statistical algorithm (e.g., calculating the average or median of the time-spaced values). The separate or combined values ​​can then be used, for example, to determine a sensitivity value and / or a sensitivity profile in step 1608 of process 1600 or to determine a correspondence with a sensitivity profile in step 1708 of process 1700. Additionally or alternatively, more than one impedance value can be obtained at substantially the same time, but each is derived using a different measurement technique, such as two of the measurement techniques described herein. For example, a first impedance can be calculated using a step voltage method as described in the process of FIG. 13, and a second impedance can be calculated using a sine wave overlay technique as described in the process of FIG. 11. The impedance values ​​derived with the different measurement techniques can then be applied to a statistical algorithm (e.g., to calculate an average or median) to determine a processed impedance value. The processed impedance value can then be used, for example, to determine a sensitivity value and / or a sensitivity profile in step 1608 of process 1600 or to determine a correspondence with a sensitivity profile in step 1708 of process 1700.

[0330] b. Temperature Some embodiments can use signal processing techniques to determine the temperature of the sensor. For example, a stimulus signal can be applied to the sensor, the signal response can be measured, and the temperature of the sensor can be derived based on the signal response.

[0331] For example, the impedance of the sensor membrane, determined using one of the techniques described with reference to Figures 11-14, can be used to estimate the temperature of the sensor, according to one embodiment. Without wishing to be bound by theory, it is believed that the sensitivity of the sensor is affected by temperature, with higher temperatures being associated with higher sensitivity and lower temperatures being associated with lower sensitivity. Similarly, it is believed that the impedance of the sensor membrane has a direct relationship to the sensitivity of the sensor, such that higher temperatures are associated with lower impedance and lower temperatures are associated with higher impedance. That is, the sensitivity and impedance can have a direct relationship to the temperature of the sensor. Thus, using the known relationship between impedance and temperature, for example based on previously conducted studies of substantially similar sensors, the temperature of the sensor can be estimated based on the impedance measurements of the sensor.

[0332] FIG. 18 is a flow chart of an exemplary process 1800 for determining sensor temperature, according to one embodiment. At step 1802, a stimulation signal is applied to an analyte sensor in use, and at step 1804, a response is measured and recorded. An impedance is calculated based on the signal response at step 1806. The impedance may be calculated using any of the techniques described herein, such as, for example, the techniques described with reference to FIGS. 11-14. Then, at step 1808, the temperature of the sensor is estimated based on a predetermined relationship between impedance and temperature. The temperature may then be used to estimate an analyte concentration value (e.g., glucose concentration) using the sensor data, or for other further processing and / or output. For example, the temperature may be used to compensate for the effect of temperature on the sensitivity of the sensor, a more accurate analyte concentration value may be estimated based on the sensitivity compensation, and the more accurate analyte concentration may be output to a display, or used to activate an alarm using a glucose monitoring system.

[0333] The relationship between sensor sensitivity and different temperatures can be mathematically modeled (e.g., by fitting a mathematical curve to the data using one of the modeling techniques described herein), and this mathematical model can then be used to compensate for the effect of temperature on the sensitivity of the sensor. That is, the sensitivity of the sensor (as affected by the temperature of the sensor) can be determined based on relating the measured impedance of the sensor to the mathematical curve. A predetermined relationship between impedance and temperature can be determined by studying the impedance of similar sensors at a range of temperatures. The sensor data can then be converted into an estimated analyte concentration value based on the determined sensor sensitivity.

[0334] As a non-limiting example, some embodiments of the analyte sensor may have an essentially linear relationship between temperature and impedance after a sensor break-in period (e.g., a period during which the sensor stabilizes after implantation, which in some embodiments may last from 1 to 5 hours). The slope of the linear correlation may be established by studying sensors manufactured in substantially the same manner as the sensor to be used over a range of temperatures. The sensor temperature may then be estimated by measuring the impedance value of the sensor membrane and applying the impedance value to the established linear correlation. Other embodiments may have a non-linear relationship between impedance and temperature, and for these other embodiments, the relationship is characterized by use of the established non-linear relationship.

[0335] Some embodiments may compare a first sensor temperature to a second sensor temperature, where the first temperature is derived from an impedance measurement of the analyte sensor and the second sensor temperature is derived independently from the impedance measurement. The second estimated temperature may be measured, for example, using a thermistor. In examples using a thermistor, the thermistor may be configured to measure temperature in vivo or ex vivo and may be located on or remote from the analyte sensor. As non-limiting examples, the thermistor may be integral with the analyte sensor and located on the surface of the recipient's skin adjacent to an insertion site within which the analyte sensor is implanted, on the recipient's skin at a location remote from the insertion site, or completely spaced from the recipient, such as on a portable device carried by the recipient. Factors causing changes in sensor sensitivity or other sensor attributes may then be determined or ascertained based at least in part on a comparison of the first and second temperatures.

[0336] c. Moisture ingress In some embodiments, moisture ingress into the sensor electronics can be determined based on measuring the impedance of the sensor at a particular frequency or range of frequencies. If the measured impedance does not sufficiently correspond to a predefined impedance value, the sensor system operatively connected to the sensor can initiate a moisture ingress error routine. Correspondence can be determined using one or more thresholds, data related functions, and the like. Additionally, it should be noted that impedance phase information can provide useful information in determining moisture release. Thus, in some embodiments, the impedance measurement can be separated into separate impedance magnitude and phase components, and one or both of the impedance components can be compared to predefined values ​​to determine correspondence.

[0337] FIG. 19 is a flow chart of an exemplary process 1900 for determining moisture ingress. In step 1902, a stimulus signal having a particular frequency or a signal having a spectrum of frequencies (e.g., voltage steps) is applied to an analyte sensor used, and a signal response is measured and recorded in step 1904. The impedance magnitude and phase are calculated based on the signal response in step 1906. The process 1900 then determines whether the impedance magnitude and phase values ​​are each within a predefined level in a decision step 1908. If the impedance magnitude and phase values ​​exceed one or both of the predefined levels, respectively, the process 1900 then begins an error routine in step 1910. The error routine may include one or more activations of an audible alarm and / or a visual alarm on a display screen to alert a user that the sensor system may not function properly. The alarm may, for example, inform a user that there is a defect in the current sensor system. On the other hand, if one or both of the impedance and phase values ​​fall within the respective predefined levels, the process 1900 ends.

[0338] While the above describes calculating separate impedance and phase values, it is understood that the above process 1900 can determine the complex impedance value by using a data related function or by comparing the determined complex impedance to a predetermined complex impedance value threshold or range, and can determine a correspondence between the determined complex impedance value and one or more predetermined complex impedance values. An error routine can be initiated in response to the correspondence.

[0339] d. Membrane damage In some embodiments, membrane damage can be detected based on measuring impedance at a particular frequency or range of frequencies. If the measured impedance does not sufficiently correspond to one or more predetermined impedance values, a sensor system operatively connected to the sensor can then initiate a membrane damage error routine. The correspondence can be determined using a data association function.

[0340] FIG. 20 is a flow chart of an exemplary process 2000 for determining membrane damage. In step 2002, a stimulus signal of a particular frequency, a plurality of signals having different frequencies, and / or a signal having a spectrum of frequencies is applied to the analyte sensor in use, and in step 2004, the signal response is measured and recorded. In step 2006, both the magnitude and phase of the impedance are calculated based on the signal response. The process 2000 then determines in decision step 2008 whether the impedance magnitude and phase values ​​are each within a range of predefined levels. If the impedance magnitude and phase values ​​each exceed one or both of the predefined levels, the process 2000 then begins an error routine in step 2010. The error routine may include one or more of the activation of an audible alarm and a visual alarm on a display screen to alert the user that the sensor system is not functioning properly. The alarm may, for example, inform the user that the currently used sensor is damaged and needs to be replaced. On the other hand, if one or both of the impedance magnitude and phase values ​​fall within the respective predefined levels, the process 2000 ends.

[0341] Although the above description describes the use of separate impedance magnitude and phase values, it is understood that the above process 2000 can use complex impedance values ​​and determine a correspondence between the complex impedance values ​​and predetermined values ​​or levels. An error routine can be initiated in response to the determined correspondence.

[0342] e. Reusing sensors In some embodiments, sensor reuse can be detected. Glucose sensor embodiments described herein may have a defined lifespan during which the sensor can provide reliable sensor data. After the defined lifespan, the sensor may no longer be reliable and provide inaccurate sensor data. To prevent use beyond a predefined lifespan, some embodiments notify the user to replace the sensor after it is determined that the sensor should no longer be used. Various methods can be used to determine whether a sensor should no longer be used. For example, a predefined period of time that has elapsed since the sensor was first used (e.g., when it was first implanted in the user or when it was first electrically connected to the sensor electronics module), or a determination that the sensor is defective (e.g., a ruptured membrane, unstable sensitivity, etc.). Once it is determined that the sensor should no longer be used, the sensor system can notify the user to use a new sensor, for example, by audibly and / or visually prompting the user to use a new sensor and / or by shutting down the display or ceasing to display the sensor data on the display. However, the user may attempt to reuse the same sensor instead of using a new sensor. This may be dangerous for the user. This is because the user relies on the inaccurate data that the sensors may provide.

[0343] Thus, some embodiments can be configured to determine sensor reuse based at least in part on one or more measurements of the impedance of the sensor. As described in more detail elsewhere herein, the impedance associated with the membrane resistance of the sensor is typically high initially, but gradually decreases as the sensor breaks in. Thus, in one embodiment, sensor reuse can be detected if the impedance measured shortly after implantation of the sensor exceeds what the sensor should typically have when it is first implanted, indicating that the sensor has already been used.

[0344] FIG. 21 is a flow chart of an exemplary process 2100 for determining sensor reuse, according to one embodiment. In step 2102, a sensor insertion event is initiated. The insertion event may be one of many possible events that indicate that a new sensor has been implanted, such as a user inputting to the sensor system that a new sensor has been implanted, the sensor system detecting an electrical connection to the sensor, a predefined period of time that has occurred since the system prompted the user to use the new sensor, etc. A stimulation signal is then applied to the analyte sensor being used in step 2104, and a response is measured and recorded in step 2106. In step 2108, an impedance is calculated based on the signal response. The stimulation signal and techniques for calculating the impedance in steps 2106 and 2108 may be any of the signals and techniques described herein, such as those described with reference to FIGS. 11-14. Then, in decision step 2110, the calculated impedance is compared to a predefined threshold. If it is determined that the impedance exceeds the threshold, then the sensor reuse routine is initiated in step 2112. If the impedance does not exceed the threshold, as determined at decision step 2110 , then process 2100 ends at step 2114 .

[0345] The sensor reuse routine of step 2112 includes activating an audible and / or visual alarm to notify the user of improper sensor reuse. The alarm may also inform the user why sensor reuse is undesirable, such as potentially outputting inaccurate and unreliable sensor readings. The sensor reuse routine 2112 may alternatively or additionally cause a complete or partial shutdown of the sensor system and / or cease displaying sensor data on the user interface of the sensor system.

[0346] In one embodiment, recent impedance measurement information (e.g., one or more recent impedance measurements) of a previously used sensor (e.g., a sensor used immediately prior to the newly implanted sensor) or a predefined sensor profile can be stored in a computer memory and compared to the impedance measurement information of the newly implanted sensor (e.g., the presumed newly used sensor). If the comparison, e.g., using a data association function, then indicates that the impedance of the previously used sensor is too similar to the impedance of the newly inserted sensor at a time close to when the use of the previous sensor was discontinued (e.g., removed), then it can be determined that the sensor is being reused, since the new sensor should have an impedance significantly different from the impedance of the previously used sensor substantially prior to the discontinuation of its use shortly after it is first implanted. If it is determined that the sensor is being reused, then the sensor system can initiate an error routine, which may include notifying the user of the improper sensor reuse via an audible and / or visual alarm using the user interface of the sensor system, as described above with reference to step 2112, and prompting the user to use a new sensor.

[0347] FIG. 22 is a flow chart of another exemplary process 2200 for determining sensor reuse, according to one embodiment. In step 2202, a sensor insertion event is initiated. The insertion event can be one of many possible events indicating that a new sensor has been implanted, such as a user inputting to the sensor system that a new sensor has been implanted, the sensor system detecting an electrical connection to the sensor, a predefined period of time occurring since the system prompted the user to use the new sensor, etc. Next, in step 2204, a stimulation signal is applied to the analyte sensor being used, and in step 2206, the response is measured and recorded. An impedance is calculated in step 2208 based on the signal response. The stimulation signal and techniques for calculating the impedance in steps 2206 and 2208 can be any of the signals and techniques described herein, such as those described with reference to FIGS. 11-14. Then, in decision step 2210, the calculated impedance is compared to one or more previously measured impedance values ​​measured using one or more previously implanted sensors (at least whatever sensors have already been implanted according to the sensor system). If it is determined that the calculated impedance correlates to one or more previously measured impedance measurements within a predetermined amount, then a sensor reuse routine is initiated at step 2112. Correlation may be determined using a data relation function, such as one of the data relation functions described herein. If, at decision step 2110, it is determined that the impedance does not correlate to a previously measured impedance value within a predetermined amount, then process 2100 ends at step 2114 and the system continues using the sensor to measure the recipient's glucose concentration.

[0348] The sensor reuse routine of step 2212 may include initiating an audible and / or visual alarm to notify the user of improper sensor reuse. The alarm may also inform the user why sensor reuse is undesirable, such as potentially outputting inaccurate and unreliable sensor readings. The sensor reuse routine 2212 may alternatively or additionally cause a complete or partial shutdown of the sensor system and / or cease displaying sensor data on the sensor system's display.

[0349] f. Sensor over-potential Some embodiments apply an over-potential routine based on the measured impedance or impedances of the sensor membrane. It has been found that applying an over-potential (e.g., a voltage higher than the bias voltage applied to the sensor when used to continuously sense the analyte) to some embodiments of the analyte sensor helps stabilize the analyte sensor, thereby shortening the break-in period of the sensor. The over-potential should be discontinued once the sensor is sufficiently stabilized, otherwise damage to the sensor may occur. Thus, one or more impedance measurements of the sensor can be used to determine the sensitivity or change in sensitivity of the sensor. Any of the techniques described herein, such as those described with reference to Figures 11-14, can be used to measure the impedance of the sensor. The determined sensitivity or change in sensitivity can then be used to indicate whether the sensor has stabilized or will stabilize within a determined period of time, for example, by determining a correspondence between the measured impedance and a pre-defined sensitivity-impedance relationship. Once it is determined that the sensor is stable or will stabilize within a determined period of time, application of the over-potential can be discontinued or reduced. Additionally, the magnitude of the over-potential and / or the amount of time for which the over-potential should be applied to the sensor may be determined or modified based on one or more impedance measurements taken prior to or during application of the over-potential, i.e., the over-potential routine, according to some embodiments, may be modified or interrupted according to one or more impedance measurements.

[0350] g. Multi-electrode or multi-sensor configurations Some embodiments of the sensor system include multiple sensor electrodes. For example, in addition to the analyte sensing electrodes as described above, some embodiments may include auxiliary electrodes to allow subtraction of a baseline signal from the analyte+baseline signal. Some embodiments may also include one or more redundant analyte sensors.

[0351] According to one embodiment, a first stimulation signal can be applied to a first sensor and a second stimulation signal can be applied to a second sensor. The first sensor can be configured to detect an analyte concentration (e.g., glucose) and in this regard can generate an analyte+baseline signal. The second sensor can be an auxiliary sensor configured to measure a baseline signal that can be subtracted from the signal of the first sensor. In addition, the first stimulation signal can have the same or a different waveform as the second stimulation signal. The response to the first stimulation signal and the response to the second stimulation signal can be measured and recorded, respectively. The first response and the second response can be processed, and sensor characteristics can be determined based on the processing, such as an impedance value associated with each sensor. This processing can include comparing the first and second response signals to each other (e.g., using a data-related function described herein) and / or comparing each of the first and second response signals to an established relationship. The sensor characteristics may include any of the sensor characteristics described herein, including sensitivity values ​​or changes in sensitivity of the first and / or second electrodes, temperatures of the first and / or second electrodes, magnitude of detected membrane damage of the first electrode, etc. Additionally, the sensor data may be compensated for changes in the sensor characteristics, e.g., changes in sensor sensitivity, based on processing of the first and second response signals.

[0352] 23 is a schematic diagram of a dual sensor configuration that can be used in a sensor system according to some embodiments. The dual sensor configuration includes a first sensor 2302 having a working electrode 2304, a reference electrode 2306, and a membrane 2308, and a second sensor 2310 having a working electrode 2312, a reference electrode 2314, and a membrane 2316. Each sensor can be essentially the same, including having the same type of membrane, or each sensor can be different, such as having a different membrane or even one of the sensors having no membrane. In one embodiment, each sensor 2302 and 2310 is configured to measure an analyte concentration in a recipient. In another embodiment, the second sensor 2310 does not have a separate reference electrode. In this alternative embodiment, the reference electrode of the first sensor can also function as the reference electrode of the second sensor.

[0353] 23, a stimulus signal can be applied to the first sensor 2302 using a signal generator / detector 2318 (the stimulus signal can be any stimulus signal described herein, such as a voltage step). The stimulus signal can induce electric field lines 2320 emanating from the first sensor 2302 and cause an electrical response in the second sensor 2310. The electrical response can be measured using a signal generator / detector 2322 electrically connected to the second sensor 2314. The signal generator / detectors 2318 and 2322 can include any known electrical circuitry capable of generating the desired stimulus signal described herein and measuring the response to the stimulus signal.

[0354] 23, the response measured at the second sensor 2310 can then be used to determine a sensor attribute, such as any sensor attribute described herein. For example, the response can be used to calculate the impedance of the first sensor 2302, which can then be used to determine the sensitivity of the first sensor 2302 and / or to correct the sensor data generated by the first sensor using one of the processes described herein.

[0355] Alternatively or additionally, a first stimulus signal can be applied to the first sensor 2302 and measured using the second sensor 2310, and a second stimulus signal can be applied to the second sensor 2310 and measured using the first sensor 2302. The first and second stimulus signals can be essentially the same or different. The response to each of the first and second stimulus signals can then be used to determine a sensor attribute, such as any of the sensor attributes described herein.

[0356] The sensor that applies the stimulation signal can also be used to measure a response to the stimulation signal in addition to other sensors that measure responses to the stimulation signal. For example, a first stimulation signal can be applied to a first sensor 2302, and first and second responses to the first stimulation signal can be measured by the first sensor 2302 and the second sensor 2310, respectively. The first and second responses can then be used to determine sensor attributes, as described herein, of either or both of the first sensor 2302 and the second sensor 2310.

[0357] h. Scaling factor In some embodiments, a scaling factor can be used to correct for differences in the response of two-electrode analyte sensors. In some embodiments, a two-electrode analyte sensor can be used as a reference sensor / system, thereby providing reference data (e.g., internal to the system) for calibration purposes without the use of an external (e.g., separate from the system) analyte measurement device. In some embodiments, a two-electrode analyte sensor has a first electrode that contains an enzyme that reacts with a particular analyte (which can be referred to as a positive enzyme electrode) and a second electrode that does not contain an enzyme (which can be referred to as a negative enzyme electrode).

[0358] In some embodiments, the sensor system (e.g., a sensor electronics module of the sensor system) is configured to determine a scaling factor (k). Simply stated, the scaling factor defines the relationship between the electrodes of a two-electrode analyte sensor. Thus, in some embodiments, the sensor system is configured to calibrate the analyte sensor data using the scaling factor such that the calibrated sensor data is free of inaccuracies caused by differences between the respective first and second working electrodes. That is, the scaling factor can be used to calculate an estimated analyte value based on data generated by the sensor system.

[0359] US Patent Publication No. 2007 / 0133990 describes in more detail a two-electrode analyte sensor system, scaling factors and methods of using the scaling factors that can be used in some embodiments, the contents of which are incorporated herein by reference in their entirety.

[0360] According to some embodiments, the membrane impedance of each electrode of a two-electrode system can be used to determine or update a scaling factor that can then be used to calculate an estimated analyte concentration value based on the data generated by the sensor system.

[0361] The following illustrates an exemplary process for determining a scaling factor using impedance, according to one embodiment. First, the membrane impedance is measured for both electrodes of a two-electrode sensor system. Techniques described herein, such as those described under the heading "Multi-electrode or multiple sensor configurations" with reference to Figures 11-14, can be used to measure the membrane impedance of each of the electrodes of a two-electrode analyte sensor. The impedance can be measured periodically during use of the sensor. A scaling factor can be generated using the ratio of the measured membrane impedances of the two electrodes (e.g., the ratio of the membrane impedance of the positive enzyme electrode to the membrane impedance of the negative enzyme electrode). A scaling factor for generating an estimated analyte value can then be determined or updated based on the generated scaling factor. The determined or updated scaling factor can then be used to generate an estimated analyte value.

[0362] It has been found that acetaminophen can interfere with glucose measurements using some sensor embodiments. It has been found that the response of acetaminophen can be proportional to its diffusion through the sensor membrane and that the sensor impedance can represent the membrane impedance. Thus, the process described above for determining a scaling factor can be used to periodically determine a scaling factor for acetaminophen by determining the ratio of the impedances of the positive and negative enzyme electrodes. The acetaminophen scaling factor can be used to update a scaling factor, such as one of those described above, and can be used to calculate an estimated glucose concentration.

[0363] i. Calibration An exemplary calibration process according to some embodiments will now be described with reference to FIG. 24. The calibration process 2400 can use one or more of pre-implant information 2402, internal diagnostic information 2404, and external reference information 2406 as inputs to form or modify a conversion function 2408. The conversion function 2408 can be used to convert the sensor data (e.g., in units of current or counts) to an estimated analyte value 2410 (e.g., in units of analyte concentration). The information representing the estimated analyte value can then be output 2412, such as displayed on a user display, transmitted to an external device (e.g., an insulin pump, a PC computer, a mobile computing device, etc.), and / or further processed. The analyte can be, for example, glucose.

[0364] In process 2400, pre-implant information 2402 can refer to information generated prior to implantation of the currently calibrated sensor. Pre-implant information 2402 can include any of the following types of information: - a predefined sensitivity profile associated with a currently used (e.g. implanted) sensor, such as a predicted profile of change in sensitivity of the sensor over time; - a previously determined relationship between a particular stimulation signal output (e.g., an output representing the impedance, capacitance, or other electrical or chemical property of the sensor) and the sensor sensitivity (e.g., determined from prior in vivo and / or in vitro studies); - a previously determined relationship between a particular stimulation signal output (e.g., an output representing impedance, capacitance, or other electrical or chemical property of the sensor) and sensor temperature (e.g., determined from prior in vivo and / or in vitro studies); - sensor data obtained from previously implanted analyte concentration sensors; - a calibration code associated with a sensor to be calibrated as described herein; - Patient-specific relationships between sensors and sensitivity, baseline, drift, impedance, impedance / temperature relationships (e.g., with characteristics common to the patient determined from the patient or prior studies of other patients); - specific relation to the sensor implantation site (abdomen, arm, etc.) (different sites may have different vascular densities); - time since sensor manufacture (e.g., how long the sensor was on the shelf, the date the sensor was manufactured and / or shipped, the time between when the sensor was manufactured and / or shipped and when the sensor was implanted); and - Exposure of shelved sensors to temperature, humidity and external factors.

[0365] In process 2400, internal diagnostic information 2402 can refer to information generated by the sensor system in which the implanted analyte sensor (whose data is calibrated) is used. Internal diagnostic information 2402 can include any of the following types of information: - output of a stimulation signal (e.g., the output of which represents the impedance of the sensor) of the sensor using any of the stimulation signal techniques described herein (wherein the output of the stimulation signal may be acquired and processed in real time); - sensor data representing analyte concentrations measured by the implanted sensors (real-time data using the currently implanted sensors and / or previously generated sensor data); - Temperature measurements using an implanted sensor or an auxiliary sensor (such as a thermistor) located in the same location as the implanted analyte sensor or separate from the implanted analyte sensor; - sensor data from a multi-electrode sensor, for example where one electrode of the sensor is designed to determine a baseline signal as described herein; - sensor data generated by redundant sensors, where one or more of the redundant sensors are designed to be substantially identical to at least some, if not all, of the other redundant sensors (e.g., have the same sensor membrane type); - sensor data generated by one or more auxiliary sensors having a different modality (such as optical, thermal, capacitive, etc.) and either co-located with the analyte sensor or remotely located from the analyte sensor; - the time since the sensor was implanted and / or connected (e.g. physically or electronically) to the sensor electronics of the sensor system; - data representing the pressure over a sensor / sensor system, for example generated by a pressure sensor (e.g. to detect compression artefacts); - data generated by an accelerometer (e.g. representing the recipient's movement / mobility / activity); - a certain amount of moisture ingress (e.g., indicative of the integrity of the moisture seal of the sensor system); and - A certain amount of noise in the analyte concentration signal (which in some embodiments can be referred to as the residual between the raw and filtered signals).

[0366] In process 2400, external reference information 2402 can refer to information generated from a source while the implanted analyte sensor (whose data is calibrated) is in use. External reference information 2402 can be any of the following types of information: - real-time and / or preceding analyte concentration information obtained from a reference monitor (e.g., analyte concentration values ​​obtained from a separate sensor, e.g., a finger stick glucose meter); - Reference meter type / brand (different meters may have different bias / accuracy); - information describing the carbohydrates consumed by the patient; - Information from drug pens / pumps such as insulin on board, insulin sensitivity, glucagon on board; - Glucagon sensitivity information; and - Information collected from population-based data (e.g., based on data collected from sensors with similar characteristics, such as sensors from the same lot).

[0367] Exemplary Sensor System Configurations The embodiments of the present invention are described above and below with reference to flowcharts of methods, devices, and computer program products. It is understood that each block of the flowchart illustrations, and combinations of blocks in the flowchart illustrations, can be implemented by execution of computer program instructions. These computer program instructions can be loaded onto a computer or other programmable data processing machine (e.g., controller, microcontroller, microprocessor, or other) in a sensor electronics system to produce a machine, such that the instructions executing on the computer or other programmable data processing machine produce instructions to perform the functions identified in the flowchart blocks. These computer program instructions can also be stored in a computer readable memory that instructs the computer or other programmable data processing machine to function in a particular manner, such that the instructions stored in the computer readable memory produce a product including instructions to perform the functions identified in the flowchart blocks. The computer program instructions can also be loaded onto a computer or other programmable data processing machine to produce a computer-implemented process, such that the steps to perform a sequence of functions are executed on the computer or other programmable device, such that the instructions executing on the computer or other programmable device provide steps to perform the functions identified in the flowchart blocks illustrated herein.

[0368] In some embodiments, a sensor system is provided for continuously measuring an analyte (e.g., glucose) within a recipient, the sensor system including a continuous analyte sensor configured to continuously measure a concentration of an analyte within a recipient and a sensor electronics module physically connected to the continuous analyte sensor during use of the sensor. In one embodiment, the sensor electronics module includes electronics configured to process a data stream associated with an analyte concentration measured by the continuous analyte sensor, and processes the sensor data to generate displayable sensor information including, for example, raw sensor data, transformed sensor data, and / or any other sensor data. The sensor electronics module can include electronics configured to process a data stream associated with an analyte concentration measured by the continuous analyte sensor, and processes the sensor data to generate displayable sensor information including, for example, raw sensor data, transformed sensor data, and / or any other sensor data. The sensor electronics module can include a processor and computer program instructions for implementing the processes described herein, including the functions identified in the blocks of the flowcharts illustrated herein.

[0369] In some embodiments, a receiver, which may also be referred to as a display device, communicates (e.g., via wired or wireless communication) with the sensor electronics module. The receiver may be a purpose-built portable device or a general-purpose device, such as a personal computer, a smart phone, a tablet computer, etc. In one embodiment, the receiver may be in data communication with the sensor electronics module to accept sensor data, such as raw and / or displayable data, and includes a processing module for processing and / or displaying the received sensor data. The receiver may also include an input module configured to accept input from a user via a keyboard or touch-sensitive display screen, such as calibration codes, reference analyte values, and any other information described herein, and may be configured to accept information from external devices, such as insulin pumps and reference meters, via wired or wireless data communication. This input may be processed alone or in combination with information received from the sensor electronics module. The receiver's processing module may include a processor and computer program instructions for performing any of the processes described herein, including the functions identified in the blocks of the flowcharts illustrated herein.

[0370] Working Example The following examples further describe the embodiments, which are provided as examples and are not intended to limit the invention in any manner. Studies were performed using a Gamry potentiostat system to analyze the complex impedance of a glucose sensor placed in a buffer solution as disclosed herein. The Gamry potentiostat system is commercially available from Gamry under the model name Ref600. The buffer solution is a modified Isolyte (isotonic electrolyte formulation) with a known glucose concentration. In many examples, the glucose concentration is about 100 mg / dL. Testing of the system with known impedances found the impedance measurement error in the examples to be about 1-5%.

[0371] Although some of the following examples are performed on a laboratory bench, the sensors under test are configured for in vivo use to continuously or substantially continuously measure a recipient's glucose concentration.

[0372] The analyte sensors used in the following examples were selected from different types of sensors. The sensors under test included sensors selected from different sensor lots, where the sensors from the first lot were manufactured in different ways and under different conditions, which may result in sensors from different lots exhibiting different sensitivity profiles. Furthermore, some of the sensors studied in these examples are configured for placement in a recipient's transdermal tissue to measure the recipient's glucose concentration, while other sensors are configured to measure the recipient's venous blood glucose concentration. In the following experiments, sensors intended for use in measuring venous blood glucose concentrations may be referred to as "IVBG sensor type" and sensors intended for measuring blood glucose concentrations in a recipient's transdermal tissue may be referred to as "transdermal sensor type."

[0373] [Example 1] Relationship between sensitivity and impedance Example 1 shows the relationship between sensor sensitivity and sensor impedance. In this example, an IVBG sensor was connected to a Gamry potentiostat system and placed in a buffer solution of modified isolyte with a glucose concentration of 100 mg / dL. The temperature during the experiment was 37° C. Impedance spectra were acquired at fixed intervals of time. The impedance spectra analyzed in this experiment ranged from 1 Hz to 100 kHz, and measurements of sensor impedance and sensor sensitivity were acquired at 15 minute intervals over a period of approximately 1200 minutes.

[0374] 25, a graph shows the absolute values ​​of the sensor's sensitivity and impedance based on an input signal with a frequency of 1 kHz. Data point 2502 shows the measurement of the sensor's sensitivity over a period of 1200 minutes (20 hours), where t=0 corresponds to the time when the sensor was first placed in the buffer solution. Data point 1104 represents the impedance value measured over the same period.

[0375] The sensitivity and impedance values ​​in Figure 25 appear to have an inverse correlation: the sensitivity increases rapidly at first, but then the rate of increase slows down and levels off, and the impedance decreases rapidly at first, but then the rate of decrease gradually slows down and levels off. Without wishing to be bound by theory, it is believed that the initial increase in sensitivity and decrease in impedance are due to break-in of the sensor.

[0376] Figure 26 is a plot of the data from Figure 25, but with sensitivity and impedance as rates of increase or decrease per hour instead of absolute values. As can be seen, the relative changes in sensitivity 2602 and impedance 2604 per hour also appear to be inversely related.

[0377] [Example 2] Retrospective compensation of sensitivity drift using impedance. 27 is a plot of sensitivity and impedance points measured over various time periods using seven different sensors, sensors A-G. Sensors A-G are transcutaneous type sensors, but were selected from several different sensor lots. Thus, although sensors A-G are all transcutaneous sensors, sensors from different lots may be manufactured in slightly different ways or under slightly different conditions, resulting in sensors from different lots exhibiting different sensitivity profiles. In this example, sensors A and D were selected from a first lot, sensor B was selected from a second lot, and sensors C, E, F, and G were selected from a third lot.

[0378] 27, the plotted data points are the sensitivity and impedance values ​​of each of the sensors A through G. Because the sensitivity of each of the sensors A through G increases gradually over time, the rightmost plotted data points for each sensor tend to correspond to values ​​measured around t=0, and the leftmost plotted data points tend to correspond to values ​​around t=24 hours.

[0379] As can be seen from Figure 27, the impedance and sensitivity values ​​of each sensor A-G have an essentially linear relationship, with sensitivity gradually increasing and impedance decreasing over time. Data obtained from all seven sensors A-G generally follows this linear relationship, although the data points for each sensor may be shifted relative to the data points for the other sensors. In other words, although the initial impedance and sensitivity values ​​for each sensor may be different, the change in sensitivity and impedance for each sensor changes at the same linear rate.

[0380] Figure 28 is a graph further illustrating the linear correlation between impedance and sensitivity for sensors A through G. The graph in Figure 28 is based on the same data used in Figure 27, but the data is graphed as percentage change rather than absolute values. As can be seen in Figure 28, sensors A through G appear to show a very similar correspondence between change in impedance and change in sensitivity.

[0381] A computer-implemented evaluation algorithm function can be used to model the relationship between changes in sensitivity of the sensor data generated by sensors A-G and changes in impedance values. The evaluation algorithm function can be formed by applying curve fitting techniques, such as those described above for forming evaluation curves of sensitivity profiles, to recursively fit a curve to the data points by adjusting the function (e.g., by adjusting constants in the function) until a best fit to the available data points is obtained. Additionally or alternatively, this relationship can be modeled in a look-up table stored in computer memory.

[0382] Additionally in Figure 28, an evaluation curve 2802 of the combined sensor data of sensors A-G is plotted. The curve in Figure 28 is a straight line, but other types of curves may occur depending on the relationship between the impedance and sensitivity of the sensors. As discussed herein, curve 2802 can be used to compensate for drift in sensor sensitivity.

[0383] 29 and 30 show compensation of sensor data taken with the same sensors used to derive evaluation curve 2802. The measurements in FIG. 29 and FIG. 30 were taken at 37° C. (Note that evaluation curve 2802 was derived based on sensor measurements taken at 37° C., as discussed above with reference to FIG. 27 and FIG. 28.) FIG. 29 is a plot of uncompensated measurements using sensors A-G. FIG. 30 is a plot of the percent estimated sensitivity error of measurements compensated using the sensitivity vs. impedance relationship based on evaluation curve 2802. The Mean Absolute Relative Difference (MARD) of the uncompensated sensor data is 21.8%. The MARD of the compensated data is 1.8%, a notable improvement over the uncompensated data.

[0384] 31 and 32 are plots of percent sensitivity error for uncompensated and compensated data, respectively. The data in FIG. 31 and FIG. 32 are based on measurements using sensors H-L. Sensors H, J, K, and L were selected from the same lot as sensors C, F, and G described with reference to FIG. 27, and sensor I was selected from the same lot as sensors A and D described with reference to FIG. 27. Additionally, the measurements plotted in FIG. 31 and FIG. 32 were taken at 25° C. instead of 37° C. However, an evaluation curve 2802 derived from sensors A-G was used to compensate the data measured by sensors H-L (note that evaluation curve 2802 is based on measurements taken at 37° C.). The MARD for the uncompensated data is 21.9%, which is nearly the same as the 21.8% MARD calculated for the sensor data in FIG. 29. The MARD for the compensated data in FIG. 18 is 4.4%, which is slightly higher than but close to the MARD for the compensated data in FIG. 30, and much less than the uncompensated MARD.

[0385] 33 and 34 are graphs of percent sensitivity error for uncompensated and compensated data, respectively. The data in FIG. 33 and FIG. 34 are based on data obtained using sensors M through Q. Sensors M, O, P, and Q were selected from the same lot as sensors C, F, and G described with reference to FIG. 27, and sensor N was selected from the same lot as sensors A and D described with reference to FIG. 27. Measurements for sensors M through Q were taken at 42° C. An evaluation curve 2802 derived from sensors A through G was also used to compensate the data in FIG. 34, where the MARD for the uncompensated data is 13.1% and the MARD for the compensated data is 4.6%.

[0386] Thus, Example 2 shows that the relationship between change in sensitivity and change in impedance determined at a first temperature can be used to compensate for drift in sensitivity at temperatures different from the first temperature.

[0387] [Example 3] Predictive Calibration of Sensor Data Using Impedance Measurements - Patent application Example 5 relates to predictive calibration. Moreover, in this experiment, the calibration of sensor data is based on a relationship between sensitivity change and impedance change previously derived from sensors from different sensor lots. That is, in Example 3, the evaluation curve 2802 was used to compensate data acquired using sensors R to U selected from a fourth sensor lot, which was not included in the group of sensors used to derive the evaluation curve 2802. Example 5 shows that the relationship between sensitivity change and impedance change derived from a sensor of a type different from the type of sensor being calibrated can be used to calibrate data. This shows that it is not necessary to use the factory calibration code of the sensor to compensate for sensitivity drift.

[0388] Figures 35 and 36 show predictive calibration of sensor data obtained from sensors R through U. Figure 35 is a plot of percent sensitivity change versus 30 minutes over nearly 1400 minutes for each sensor. Figure 36 shows the percent estimated sensitivity error for the compensated data. The MARD for the uncompensated sensor data is 24.8% (the MARD for the uncompensated data in the example of Figure 15 was 21.8%) and the MARD for the compensated sensor data is 6.6% (the MARD for the compensated data in the example of Figure 16 was 1.8%).

[0389] Predictive calibration of the sensors will now be discussed with reference to Figures 37-39, where sensitivity and impedance data was collected from sensors VX, Z, and AA with immersion-applied films and sensor Y with spray-applied films. With reference to Figure 37, an evaluation curve 3702 was calculated based on the sensitivity and impedance data from all six sensors VZ and AA (i.e., using sensor data obtained from both immersion-applied and spray-applied film sensors). Figure 39 is a graph of the percent sensitivity change versus 60 minutes for the uncompensated data for each of the six sensors. Figure 39 is a graph of the percent estimated sensitivity error for all six sensors' data after compensation using the evaluation curve 3702 discussed above with respect to Figure 37. The MARD for the uncompensated data is 25.3% and the MARD for the compensated data is 5.2%.

[0390] Thus, Example 5 shows that using the relationship between sensitivity change and impedance change derived from sensors selected from different lots, rather than calibrated sensors, can nevertheless compensate for drift in sensor sensitivity to a greater extent. Note that curve 3702 is a straight line. The contemplated finding is that a non-linear fit or correlation could be used instead, which may yield better results.

[0391] [Example 4] Effect of Temperature Figure 40 shows the relationship between impedance and sensitivity of the sensor and temperature. Point 4002 is the sensitivity value of the sensor measured over a period of three days, and point 4004 is the impedance value of the sensor measured over the same period. In Example 4, the sensor is a transcutaneous type sensor. As shown in Figure 40, the temperature is first set and maintained at 37°C, then increased to 45°C, and finally decreased to 25°C.

[0392] As shown in FIG. 40, both the sensitivity and impedance of the sensor appear to have an inverse relationship as the temperature changes.

[0393] Figure 41 is a plot of the sensitivity measurements versus impedance measurements of Figure 40. Figure 41 shows points measured during sensor break-in as diamonds and points measured after break-in as squares.

[0394] [Example 5] temperature compensation 42 shows the effect of temperature after sensor break-in on the compensation of analyte concentration data measured by the sensor of Example 4. Here, a relationship between impedance and temperature was used to compensate the sensor data. In this example, the relationship is based on a characterization curve derived from the data of FIG.

[0395] The relationship between the sensor sensitivity and different temperatures can then be mathematically modeled (e.g., by fitting a mathematical curve to the data using one of the modeling techniques used herein), and this mathematical model can then be used to compensate for the effect of temperature on the sensor sensitivity. That is, the sensitivity of the sensor (as affected by the temperature of the sensor) can be determined based on the measured impedance of the sensor to which the mathematical curve is applied. The sensor data can then be converted to an estimated glucose value based on the determined sensor sensitivity.

[0396] Further adding to FIG. 42, the MARD for the uncompensated data was calculated to be 9.3%, and the MARD for the compensated data was calculated to be 2.8%.

[0397] [Example 6] Moisture ingress detection Example 6 involves the detection of moisture ingress in the sensor electronics components used to drive the sensor. In this example, an input signal with a frequency ranging from 100 Hz to 1 kHz is applied to the sensor. The response of the signal is then measured over the range of frequencies and both the impedance and phase changes are derived therefrom. Next, contacts typically located inside the sealed portion of the sensor electronics module are wetted. The frequency of the input signal is varied between 100 Hz and 1000 Hz. The response is again measured and both the impedance and phase changes are derived.

[0398] In the first experiment, the contacts were wetted to a point that caused significant failure of the sensor; the sensor current increased from 2.3 nA when dry to 36 nA when wet. In the second experiment, the contacts were only slightly wetted, and the sensor current increased from 2.3 nA when dry to 6 nA when wet.

[0399] Figure 43 is a graph of the change in impedance and phase for a first experiment, where the contacts were wetted to a point that caused significant failure of the sensor system. Curves 4302 and 4304 are the impedance and phase values ​​of the sensor, respectively, before wetting the contacts. Curves 4306 and 4308 are the impedance and phase values ​​of the sensor, respectively, after wetting the contact surfaces. As shown in Figure 43, the impedance of the dry curve 4302 and the wet curve 4306 are significantly different around 1000 Hz, and the phase of the dry curve 4304 and the wet curve 4308 are significantly different at about 100 Hz.

[0400] Figure 44 is a graph of the change in impedance and phase for a second experiment, where the contacts are only slightly wetted. Curves 4402 and 4404 are the impedance and phase, respectively, of the sensor before wetting the contacts. Curves 4406 and 4408 are the impedance and phase, respectively, of the sensor after wetting the contacts. As shown in Figure 44, the impedance of the dry curve 4402 and the wet curve 4406 are significantly different around 100 Hz, and the phase of the dry curve 4404 and the wet curve 4408 are significantly different around 1 kHz.

[0401] [Example 7] Detection of membrane damage Example 7 involves detection of damage to a sensor membrane using impedance measurements. In this example, the impedance of an analyte sensor was measured over a range of frequencies spanning 100 to 1 kHz. A portion of the sensor membrane was then cut using a razor blade to cause membrane damage. The impedance of the sensor was measured again over the same frequency range. To determine whether impedance can be used to detect membrane damage, the impedance measurements of the sensor before cutting the membrane were then compared to the measurements of the sensor after cutting the membrane.

[0402] 45 is a graph of sensor measurements before and after membrane damage. Curves 4502 and 4504 are the impedance and phase of the sensor, respectively, before membrane damage, and curves 4506 and 4508 are the impedance and phase of the sensor, respectively, after membrane damage. Both the impedance and phase relationships appear to change with frequency per kHz. This relationship can then be used to detect membrane damage.

[0403] [Example 8] A study on using FFT to calculate impedance FIG. 46 is a graph plotting impedance calculated using a Fast Fourier Transform (FFT) and impedance measured using Gamry for comparison. FIG. 46 shows the FFT and corresponding Gamry measurements for 30 minutes, 1.5 hours, 3 hours, and 13 hours of input data. The FFT data substantially tracked the Gamry data up to a frequency of about 1 kHz. The discrepancy in FIG. 33 after the 1 kHz frequency is believed to be due to known limitations of the system used to calculate the FFT. Thus, it is believed that the use of the FFT can provide accurate impedance data even beyond the 1 kHz spectrum, thereby providing a good correspondence to the impedance measured using Gamry.

[0404] [Example 9] Compensation for in vivo sensitivity changes 47-49 relate to compensation for sensitivity variations within humans using an in vivo glucose sensor.

[0405] FIG. 47 is a graph showing in vivo data of sensor impedance and sensitivity over a 25 hour period for a subcutaneous continuous glucose sensor (referred to herein as a CGM sensor). FIG. 47 is similar to FIG. 25, but uses in vivo data instead of in vitro data. Sensitivity data was generated using data obtained from a blood glucose fingerstick detector and raw sensor current measurements using a CGM sensor (i.e., raw CGM sensor current in counts was divided by time-corresponding meter glucose value in mg / dL to generate sensitivity values). Impedance data was generated by applying a step voltage to the CGM sensor and calculating impedance based on the peak current of the response, as discussed herein with reference to FIG. 13. As shown in FIG. 47, the impedance and sensitivity data generally appear to follow a similar profile to the in vitro data of FIG. 25.

[0406] FIG. 48 is a graph of glucose values ​​estimated using data generated by a CGM sensor before and after sensitivity compensation using impedance measurements of FIG. 47. The compensated and uncompensated data were calibrated once at the one hour mark using blood glucose reference meter measurements. Compensation was performed in a manner similar to that discussed above with respect to the in vitro study of FIG. 30. FIG. 48 also shows the recipient's blood glucose reference measurements. Based on the data in FIG. 48, the compensated CGM sensor data appears to correspond more closely to the reference measurement data than the uncompensated CGM sensor data.

[0407] 49 is an accuracy plot showing the difference between uncompensated and compensated CGM sensor data and finger stick meter reference data, with the reference data corresponding to the zero line of the plot. The MARD for the uncompensated CGM sensor data is 28.5% and the MARD for the compensated CGM sensor data is 7.8%. Thus, this compensation appears to increase the accuracy of the CGM sensor data.

[0408] [Example 10] Comparison of linear sensitivity-impedance correlation with nonlinear sensitivity-impedance correlation Laboratory experiments have shown that there is an inverse relationship between the change in sensor sensitivity (drift) and the change in impedance. Based on the measured impedance, the drift in sensor sensitivity can be compensated for using a linear or nonlinear sensor sensitivity-impedance correlation.

[0409] Linear correlation can be expressed as an equation: ΔS=a*ΔI+b, where a and b can be pre-determined coefficients determined from prior testing of similar sensors. Non-linear correlation can be expressed as: △S = (a*log(t)+b)*△I where a and b are constants determined from prior testing of similar sensors. Prior testing of similar sensors was performed using in vitro bench tests, and the constants were derived in this experiment using conventional computer plotting techniques.

[0410] Linear and non-linear equations were derived for the relationship between the change in sensitivity and the change in impedance at time t since the last calibration. The percent change in sensitivity at time t is expressed as: △S = % change in sensitivity = (sensitivity at time t - Sc) / Sc*100% where Sc is the sensitivity determined in the calibration, and t is the time since the last calibration.

[0411] The change in sensor impedance from time t is expressed as: △I = % impedance change = (impedance at time t - Ic) / Ic*100% where Ic is the calibration impedance.

[0412] Conventional computer plotting software was used to calculate linear correlation to plot the change in sensitivity (y-axis) versus the change in impedance (x-axis) based on the in vitro test data and determine the best linear fit of the data. In this experiment, the best linear fit was determined to be: △S=-4.807*△I-0.742 This best linear fit was used as the linear sensitivity vs. impedance correlation.

[0413] Change in Sensitivity with Change in Impedance Conventional computer plotting software was used to calculate the non-linear correlation to plot ΔS / ΔI (y-axis) versus log(t). The best linear fit of the plotted data was then determined using the plotting software to yield ΔS / ΔI=a*log(t)+b, from which the change in sensitivity can be derived: ΔS = (a*log(t)+b)*ΔI. In this experiment, the plotting software yielded ΔS / ΔI=-1.175*log(t)-1.233. From this equation, ΔS can be derived: ΔS=(-1.175*log(t)-1.233)*ΔI.

[0414] A linear or non-linear correlation can then be applied to correct for changes in sensor sensitivity using the sensor impedance measurements.

[0415] Figure 50 shows a process 5000 for correcting the estimated glucose values ​​used in this experiment. Process 5000 begins at block 5002 with sensor insertion at time t = 0. Process 5000 then splits into two branches.

[0416] The first branch begins at block 5004 where sensor data is generated using a continuous glucose sensor in the form of a current or count. The first branch continues to decision block 5006 where process 5000 determines whether calibration is required. In this experiment, it is determined that calibration is required at 1 hour and 24 hours after sensor insertion. If it is determined that calibration is not required, the first branch essentially ends. On the other hand, if it is determined that calibration is required, process 5000 proceeds to obtain a baseline measurement using a finger stick meter at blocks 5008 and 5010. Sensor data corresponding to the baseline measurement (at time Tc) is determined at block 5012 and is used with the baseline data to calculate the sensor sensitivity (Sc) at block 5014.

[0417] The second branch of the process 5000 begins with measuring the impedance of the sensor in block 5016. The impedance is measured using the peak current technique discussed in this application with reference to FIG. 13. Next, in decision block 5018, it is determined whether calibration is needed. In this experiment, this is the same decision made in decision block 5006. If it is determined that calibration is needed, then the impedance measured in block 5016 is flagged and stored as impedance Ic (impedance calibration) in block 5020. If calibration is not needed, the process 5000 skips block 5020 and determines the change in impedance from the impedance flagged as impedance Ic in block 5022, which in this experiment is the difference between the impedance measured in block 5016 and impedance Ic. A compensation algorithm (using a linear or non-linear compensation algorithm) is then used in block 5024 to calculate a change in sensitivity in block 5026 based on the change in impedance determined in block 5022.

[0418] The first and second branches of process 5000 meet at block 5028 to determine a modified sensitivity that is subsequently used to convert the sensor data in units of current or counts to an estimated glucose value in units of glucose concentration.

[0419] The process 5000 is repeated for each sensor data point that is to be converted to an estimated glucose value.

[0420] In this experiment, the improvement of sensor performance by impedance compensation is demonstrated below. The sensor was tested using a continuous glucose sensor placed in sheep serum for 2 days. The sensor sensitivity was calculated at 1 hour and 24 hours. In FIG. 51, the sensor sensitivity obtained at the time of calibration (1 hour or 24 hours) was used for the entire day without any correction. In FIG. 52, the sensitivity was corrected using linear impedance correction according to the process shown in FIG. 50. In FIG. 53, the sensitivity was adjusted by a non-linear sensitivity-impedance correlation according to the process shown in FIG. 50. Both linear and non-linear corrections demonstrated improved sensor performance, but the non-linear correction produced better results.

[0421] Some embodiments disclosed herein apply stimulation signals continuously or repeatedly during successive sensor sessions, or use and extrapolate information from the output associated with the stimulation, such as using peak current measurements, EIS, etc. Although certain electrochemical analysis techniques are described herein, it is understood that many other techniques can be used instead to detect the sensor properties described herein, such as voltammetry, chronopotentiometry, current or potential step techniques, differential chronopotentiometry, oxygen absorption rate measurements, potential / current sweep or pulse methods, etc.

[0422] While this disclosure has been illustrated and described in detail in the drawings and the foregoing description, such drawings and description are considered to be illustrative or exemplary and not restrictive. This disclosure is not limited to the disclosed embodiments. Modifications to the disclosed embodiments can be understood and effected by those skilled in the art, from a study of the drawings, the disclosure, and the appended claims, in practicing the claimed disclosure.

[0423] All references cited herein are incorporated herein in their entirety. To the extent that publications and patents or patent applications incorporated by reference contradict any disclosure contained in this specification, it is intended that this specification supersede and / or supersede any such conflicting material.

[0424] Unless otherwise specified, all terms (including technical and scientific terms) are to be given their ordinary and customary meanings to those skilled in the art, and are not to be limited to any special or customized meaning unless expressly so specified herein. It should be noted that the use of a particular term, when describing a feature or aspect of this disclosure, should not be taken to mean that the term is redefined in a restrictive manner herein to include characteristics specific to the feature or aspect of this disclosure with which the term is associated. Terms and expressions used in this application, and variations thereof, should be construed as open ended as opposed to restrictive, unless expressly stated, particularly in the appended claims. For the foregoing examples, the term "including" should be read to mean "including, without limitation," "including but not limited to," and the like; "comprising" as used herein is synonymous with "including," "containing," and "characterized by," and is inclusive or open-ended, and does not exclude additional, unrecited elements or method steps; the term "having" should be interpreted as "having at least;" the term "including" should be interpreted as "including but not limited to;" the term "embodiment" is a term used to provide illustrative examples of the items, and is not an exhaustive or limiting list; "known," "normal," "standard," "standard" and "standard" are examples of the items used herein. The adjectives "preferably," "preferred," "desired," "preferable," and words of similar import should not be construed as limiting the described items to those items available in a given period or at a given time, but instead should be read to include known, normal, or standard technology available or known at any time now or in the future; and the words "preferably," "preferred," "desired," "preferable," and similar imports should not be understood to imply that a feature is critical, essential, or critical to the structure or function of the invention, but instead are intended merely to highlight alternative or additional features that may or may not be utilized in a particular embodiment of the invention.Similarly, a group of items joined by the conjunction "and" should not be read as requiring one of the items to be present in each and every group, but rather as "and / or" unless expressly stated. Similarly, a group of items joined by the conjunction "or" should not be read as requiring mutual exclusivity among the group, but rather as "and / or" unless expressly stated.

[0425] Where a range of values ​​is provided, it is to be understood that the upper and lower limits, and each value intervening between the upper and lower limits of the range, are encompassed within the scope of the embodiments.

[0426] With respect to the use herein of substantially plural and / or singular terms, those skilled in the art may translate from plural to singular and / or from singular to plural as appropriate to the context and / or application. Various singular / plural permutations may be expressly stated herein for clarity. The indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain means are recited in mutually different dependent claims does not indicate that a combination of these means cannot be used to advantage. Any reference signs in the claims shall not be interpreted as limiting the scope.

[0427] It will be further understood by those skilled in the art that when a specific number of introduced claim recitations is intended, such intent is clearly recited in the claim and, absent such recitation, such intent does not exist. For example, as an aid to understanding, the following appended claims may include the use of introductory phrases such as "at least one" and "one or more" to introduce claim recitations. However, the use of such phrases should not be construed as meaning that the introduction of a claim recitation with the indefinite article "a" or "an" limits any particular claim that includes a claim recitation so introduced to embodiments that include only such recitations, even when the same claim includes the introductory phrase "one or more" or "at least one" and an indefinite article such as "a" or "an" (e.g., "a" and / or "an" are typically interpreted to mean "at least one" or "one or more"); the same applies to the use of definite articles used to introduce claim recitations. In addition, even if a particular number of recitations in an introduced claim is explicitly recited, those skilled in the art will understand that such recitation should typically be interpreted as meaning "at least" the recited number (e.g., the bare recitation of "two" without other modifications typically means "at least two" or "two or more"). Furthermore, in cases where a term similar to "at least one of A, B, and C, etc." is used, such a configuration is generally intended to be the meaning that a skilled artisan would understand the term (e.g., "a system having at least one of A, B, and C" includes, but is not limited to, systems having only A, only B, only C, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.).In instances where a term similar to "at least one of A, B, or C, etc." is used, such a configuration is generally intended to be the meaning that one of ordinary skill in the art would understand the term (e.g., "a system having at least one of A, B, or C" includes, but is not limited to, systems having only A, only B, only C, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). Additionally, one of ordinary skill in the art will understand that any substantially disjunctive word and / or phrase may indicate two or more alternative terms, but should be understood to anticipate the possibility of including one of the terms, either of the terms, or both of the terms, whether in the description, claims, or drawings. For example, the phrase "A or B" is understood to include the possibilities of "A" or "B" or "A and B."

[0428] All numbers expressing quantities of ingredients, reaction conditions, and so forth used in the specification should be understood to be modified in all instances by the term "about." Accordingly, unless expressly indicated to the contrary, the numerical parameters set forth in this specification are approximations that may vary depending upon the desired attributes sought to be obtained. At the very least, and without attempting to limit the application of the doctrine of equivalents to the scope of any application claiming priority to this application, each numerical parameter should be construed in light of the number of significant digits and ordinary rounding approaches.

[0429] Furthermore, although the foregoing has been described in some detail by way of illustration and example for purposes of clarity and understanding, it will be apparent to one skilled in the art that certain changes and modifications may be practiced. Therefore, the descriptions and examples should not be construed as limiting the scope of the invention to the specific embodiments and examples described herein, but rather will cover all modifications and variations that are within the true scope and spirit of the invention.

Claims

1. 1. A system for calibrating a glucose sensor including a membrane, comprising: means for generating sensor data using the glucose sensor; means for determining, in an electronic device, a sensitivity and baseline value of the glucose sensor as a function of time by applying a priori information of the glucose sensor, the sensitivity and baseline values ​​being a function of time of the glucose sensor; means for calibrating the sensor data based at least in part on the determined sensitivity value and the baseline value; Including, The a priori information is - information about the sensor material, including the material of the sensor membrane, the thickness of the sensor membrane, and the permeability of the membrane to the chemical species of the sensor membrane; and - information regarding the process conditions under which the sensor was manufactured, including the temperature at which the sensor film was cured and the amount of time the sensor was immersed in a particular coating solution; Including, the system.

2. The system of claim 1 , wherein calibrating the sensor data is performed at discrete time intervals over a sensor session.

3. The system of claim 2 , wherein the discrete time intervals are five minute intervals.

4. 3. The system of claim 2, wherein the discrete time intervals are 30 second intervals.

5. The system of claim 1 , wherein the step of determining the sensitivity value and the baseline value is performed for substantially an entire sensor session.

6. The system of claim 1 , wherein the a priori information is associated with at least one predetermined sensitivity value that is associated with a predetermined time after a sensor session begins.

7. The system of claim 6 , wherein the a priori information is associated with at least one predetermined baseline value associated with a predetermined time after a sensor session begins.

8. The system of claim 1 , wherein the a priori information is associated with a predetermined sensitivity function that uses time as an input.

9. The system of claim 8 , wherein the a priori information is associated with a predetermined baseline function that uses time as an input.

10. The system of claim 8 or 9, wherein the time corresponds to a time after a sensor session has started.

11. 10. The system of claim 8 or 9, wherein the time corresponds to a time since manufacture.

12. The system of claim 1 , wherein the sensitivity value of the glucose sensor is also a function of at least one other parameter.

13. 13. The system of claim 12, wherein the at least one other parameter is selected from the group consisting of temperature, pH, level or duration of hydration, curing conditions, glucose concentration of a fluid surrounding the glucose sensor during activation of the sensor, and combinations thereof.

14. The system of claim 1 , wherein calibrating the sensor data is performed without the use of reference blood glucose data.

15. a transcutaneous glucose sensor having a membrane, the transcutaneous glucose sensor configured to generate sensor data when at least partially inserted into a host's body; a processor configured to receive the sensor data from the transcutaneous glucose sensor, - determining a sensitivity value and a baseline value for the transcutaneous glucose sensor by applying a priori information of the transcutaneous glucose sensor; calibrating the sensor data based at least in part on the determined sensitivity values ​​and the baseline values; A processor configured to: Equipped with The a priori information is - information about the sensor material, including the material of the sensor membrane, the thickness of the sensor membrane, and the permeability of the membrane to the chemical species of the sensor membrane; and - information regarding the process conditions under which the sensor was manufactured, including the temperature at which the sensor film was cured and the amount of time the sensor was immersed in a particular coating solution; 1. A glucose detection system comprising:

Images