Systems and methods providing sensitive and specific alerts

Abstract

This document provides systems and methods for providing sensitive and specific alarms indicating blood glucose status. In one embodiment, the method for transitioning between states related to a subject's blood glucose status includes: evaluating sensor data from a continuous glucose sensor and activating an alarm state based on sensor data that meets activation transition criteria associated with a hypoglycemic or hyperglycemic state; providing an output associated with the activated alarm state, indicating a hypoglycemic or hyperglycemic state; transitioning from an activated state to a confirmed state over a period of time in response to at least one of the user-confirmed alarm state or data indicating that the subject's glucose is trending towards normal; actively monitoring data related to the subject's hypoglycemic or hyperglycemic state over a period of time in the confirmed state; and transitioning from the confirmed state to at least one of an inactive state or an activated state in response to data that meets predetermined criteria related to the subject's hypoglycemic or hyperglycemic state.

Description

[0001] This application is a divisional application of Chinese Patent Application No. 202211637276.1, filed on October 16, 2013, entitled "System and Method for Providing Sensitive and Specific Alarms".

[0002] Incorporated by referencing relevant applications This application claims the benefits of the following applications: U.S. Application 13 / 742841, filed January 16, 2013; U.S. Application 13 / 742,694, filed January 16, 2013; and U.S. Provisional Application 61 / 720,286, filed October 30, 2012, the disclosures of which are incorporated herein by reference in their entirety and are expressly incorporated as part of this application. Technical Field

[0003] This system and method relate to processing analyte sensor data from a continuous analyte sensor. More specifically, this system and method relate to providing accurate predictive alerts to the user. Background Technology

[0004] Diabetes mellitus is a disorder in which the pancreas fails to produce enough insulin (type 1 or insulin-dependent) and / or is ineffective (type 2 or non-insulin-dependent). In a diabetic state, patients experience high blood sugar, which can cause numerous physiological disturbances associated with small blood vessel damage, such as kidney failure, skin ulcers, or bleeding into the vitreous humor of the eye. Hypoglycemic reactions (hypoglycemia) can be caused by an unintentional overdose of insulin, or by excessive exercise or insufficient food intake following normal use of insulin or hypoglycemic agents.

[0005] Typically, people with diabetes carry a self-monitoring blood glucose (SMBG) device, which usually requires an uncomfortable finger prick to obtain a blood sample for measurement. Due to the discomfort and inconvenience associated with the finger prick, people with diabetes usually only measure their glucose levels two to four times a day. Unfortunately, the time intervals between measurements may be too long, causing people with diabetes to detect hypoglycemia or hyperglycemia too late, which can sometimes cause dangerous side effects. Based on traditional methods, not only is it impossible for people with diabetes to obtain timely SMBG values, but they may also not know that their blood glucose levels are rising (higher) or falling (lower). This can therefore hinder a diabetic patient from making a decision to receive insulin therapy.

[0006] Another instrument some diabetic patients use to monitor their blood sugar is a continuous analyte sensor. Continuous analyte sensors typically consist of a sensor placed subcutaneously, transdermally (e.g., through the skin), or within a blood vessel. The sensor measures the concentration of a given analyte in the body and generates a signal that is transmitted to an electronic device connected to the sensor.

[0007] One of the main benefits discovered with continuous analyte sensors is that these instruments can have alarms to warn diabetic users before a hyperglycemic or hypoglycemic event occurs. Furthermore, it is desirable that these instruments be accurate so that users do not become desensitized to the alarms and ignore them due to inconvenience. This disclosure addresses these needs. Summary of the Invention

[0008] This system and method relate to processing analyte sensor data. This document provides a system and method that allows users to receive warnings or alerts indicating blood glucose status in a more accurate and useful manner. Therefore, users can have an improved user experience using this system and method.

[0009] In a first aspect—which may be independently combined with any aspect or embodiment given herein—a method is provided for activating a hypoglycemic indicator based on continuous glucose sensor data, the method comprising: evaluating sensor data using a first function to determine whether a real-time glucose value meets one or more first criteria; evaluating sensor data using a second function to determine whether a predicted glucose value meets one or more second criteria; activating the hypoglycemic indicator if one or more first criteria or one or more second criteria are met; and providing an output based on the activated hypoglycemic indicator. In one embodiment of the first aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—evaluating sensor data using the first function to determine whether a real-time glucose value meets one or more first criteria includes determining whether the real-time glucose value exceeds a glucose threshold. In one embodiment of the first aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—evaluating sensor data using the first function to determine whether a real-time glucose value meets one or more first criteria further includes determining whether the magnitude or direction of the rate of change meets a rate of change criterion. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—a first function is used to evaluate sensor data to determine whether a real-time glucose value meets one or more first criteria, including assessing the static risk of a basic real-time glucose value. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—a second function is used to evaluate sensor data to determine whether a predicted glucose value meets one or more second criteria, including assessing the dynamic risk of the predicted glucose value. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—the second function includes an artificial neural network model that utilizes at least one of exercise, stress, disease, or surgery to determine the predicted glucose value. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—the second function utilizes a first-order autoregressive model to determine the predicted glucose value. In one embodiment of the first aspect—which is generally applicable (i.e., independently in combination with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—the first-order autoregressive model includes a parameter α and wherein α is recursively estimated whenever a sensor data point is received.In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—the first-order autoregressive model includes a forgetting factor, a prediction range, and a prediction threshold, which are adjusted to provide no more than one additional alert per week based on retrospective analysis, which compares the use of the first function and the second function together with the first function alone. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—evaluating sensor data using both the first and second functions allows for an increase in the warning time of hypoglycemia warnings without substantially increasing the number of warnings, compared to evaluating sensor data using the first function without evaluating sensor data using the second function. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—the second function includes a Kalman filter to determine the predicted glucose value using an estimate of the rate of change of blood glucose as input. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—one or more first criteria include a first threshold configured to be user-settable. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—one or more second criteria include a second threshold that is not user-configurable. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—the second function includes a prediction range that is not user-configurable. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—one or more second criteria include a second threshold adaptively set by the processor module based on the first threshold. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the first aspect—the second function includes a prediction range adaptively set by the processor module based on the first threshold.In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—the hypoglycemia indicator includes an indication having a specific set of indications associated with whether the hypoglycemia indicator is activated based on a first function satisfying one or more criteria or whether the hypoglycemia indicator is activated based on a second function satisfying one or more criteria. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—the output includes at least one auditory, tactile, or visual output, and wherein said output is distinguishable, and / or said output selectively provides information based on whether the hypoglycemia indicator is activated based on a first function satisfying one or more criteria or whether the hypoglycemia indicator is activated based on a second function satisfying one or more criteria. In one embodiment of the first aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the first aspect—providing output includes transmitting a message to an insulin delivery device, which includes indications associated with at least one of the following: a) pausing insulin delivery, b) initiating a hypoglycemia and / or hyperglycemia minimization algorithm, c) controlling insulin delivery, or d) information related to the hypoglycemia indicator.

[0010] In a second aspect—which may be independently combined with any aspect or embodiment given herein—a system for processing data is provided, the system comprising: a continuous analyte sensor configured for implantation within the body; and sensor electronics configured to receive and process sensor data output from the sensor, the sensor electronics including a processor configured to: evaluate the sensor data using a first function to determine whether a real-time glucose value meets one or more first criteria; evaluate the sensor data using a second function to determine whether a predicted glucose value meets one or more second criteria; activate a hypoglycemia indicator if one or more first criteria or one or more second criteria are met; and provide an output based on the activated hypoglycemia indicator. In one embodiment of the second aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—evaluating the sensor data using the first function to determine whether a real-time glucose value meets one or more first criteria includes determining whether the real-time glucose value exceeds a glucose threshold. In one embodiment of the second aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—evaluating the sensor data using the first function to determine whether a real-time glucose value meets one or more first criteria further includes determining whether the magnitude or direction of the rate of change meets a rate of change criterion. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—a first function is used to evaluate sensor data to determine whether a real-time glucose value meets one or more first criteria, including assessing the static risk of a basic real-time glucose value. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—a second function is used to evaluate sensor data to determine whether a predicted glucose value meets one or more second criteria, including assessing the dynamic risk of the predicted glucose value. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the second function includes an artificial neural network model that utilizes at least one of exercise, stress, disease, or surgery to determine the predicted glucose value. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the second function utilizes a first-order autoregressive model to determine the predicted glucose value.In one implementation of the second aspect—which is generally applicable (i.e., independently in combination with any aspect or implementation given herein), and particularly applicable to any other implementation of the second aspect—the first-order autoregressive model includes a parameter α, wherein α is recursively estimated each time a sensor data point is received. In one implementation of the second aspect—which is generally applicable (i.e., independently in combination with any aspect or implementation given herein), and particularly applicable to any other implementation of the second aspect—the first-order autoregressive model includes a forgetting factor, a prediction range, and a prediction threshold, which are adjusted to provide no more than one additional alert per week based on retrospective analysis that compares the use of the first and second functions together with the first function alone. In one implementation of the second aspect—which is generally applicable (i.e., independently in combination with any aspect or implementation given herein), and particularly applicable to any other implementation of the second aspect—using both the first and second functions to evaluate sensor data allows for an increase in the warning time for hypoglycemia warnings without substantially increasing the number of warnings, compared to using the first function to evaluate sensor data without using the second function. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the second function includes Kalman filtering to determine a predicted glucose value using an estimate of the rate of change of blood glucose as input. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—one or more first criteria include a first threshold configured to be user-settable. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—one or more second criteria include a second threshold that is not user-settable. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the second function includes a prediction range that is not user-settable. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—one or more second criteria include a second threshold adaptively set by a processor module based on the first threshold. In one implementation of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the second aspect—the second function includes a prediction range adaptively set by the processor module based on a first threshold.In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the hypoglycemia indicator includes an indication having a specific set of indications associated with whether the hypoglycemia indicator is activated based on a first function satisfying one or more criteria or whether the hypoglycemia indicator is activated based on a second function satisfying one or more criteria. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—the output includes at least one auditory, tactile, or visual output, and wherein said output is distinguishable, and / or said output selectively provides information based on whether the hypoglycemia indicator is activated based on a first function satisfying one or more criteria or whether the hypoglycemia indicator is activated based on a second function satisfying one or more criteria. In one embodiment of the second aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the second aspect—providing output includes transmitting a message to an insulin delivery device, which includes indications associated with at least one of the following: a) pausing insulin delivery, b) initiating a hypoglycemia and / or hyperglycemia minimization algorithm, c) controlling insulin delivery, or d) information related to the hypoglycemia indicator.

[0011] In a third aspect—which can be independently combined with any aspect or implementation described herein—a method is provided for transitioning between states related to a subject's blood glucose status, the method comprising: evaluating sensor data from a continuous glucose sensor and activating an alarm state based on sensor data that meets one or more activation transition criteria related to a hypoglycemic or hyperglycemic state; providing an output associated with the activation of the alarm state, wherein the output indicates a hypoglycemic or hyperglycemic state; transitioning from an activated state to a confirmed state over a period of time in response to at least one of the user's confirmation of the alarm state or data indicating that the subject's glucose is trending toward normal blood glucose; actively monitoring data related to the subject's hypoglycemic or hyperglycemic state over a period of time in the confirmed state; and transitioning from the confirmed state to at least one of an inactive state or an activated state in response to data related to the subject's hypoglycemic or hyperglycemic state that meets one or more predetermined criteria. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from an active state to a confirmed state includes a transition from an active state to a confirmed state based on data indicating that the subject's glucose is trending toward normal blood glucose levels, wherein the data is selected from a) sensor data indicating changes in glucose trends or b) insulin information related to condition correction. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from an active state to a confirmed state includes a transition from an active state to a confirmed state based on user confirmation, wherein the data is selected from a) user confirmation of an alarm on a user interface or b) user input of insulin information or c) user input of dietary information. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—active monitoring includes monitoring at least one of sensor data, sensor diagnostic information, dietary information, insulin information, or event information. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from a confirmed state includes a transition from a confirmed state to an inactive state based on sensor data that no longer meets one or more criteria associated with a hypoglycemic or hyperglycemic state. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from a confirmed state includes a transition from a confirmed state to an inactive state based on sensor data that meets one or more inactive transition criteria, wherein the inactive criteria differ from one or more active transition criteria associated with a hypoglycemic or hyperglycemic state.In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from the confirmed state includes a transition from the confirmed state to the inactive state based on insulin data and / or dietary information. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the transition from the confirmed state includes a transition from the confirmed state to the active state based on the fulfillment of one or more activation transition criteria associated with a hypoglycemic or hyperglycemic state and based on the expiration of a predetermined time period. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—after receiving confirmation of the alarm status from the user and determining that data indicates the subject's glucose is trending towards normal blood glucose, the method further includes a transition from the confirmed state to the active state during the active monitoring period based on the subject's glucose trending away from normal blood glucose. In one embodiment of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the third aspect—the method further includes selectively outputting information related to the state transition. In one implementation of the third aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the third aspect—the output associated with transitioning to an active state differs from the output associated with transitioning from a confirmed state to an inactive state.

[0012] In a fourth aspect—which can be independently combined with any aspect or implementation described herein—a system for processing data is provided, the system comprising: a continuous analyte sensor configured to be implanted in the body; and sensor electronics configured to receive and process sensor data output from the sensor, the sensor electronics including a processor configured to: evaluate sensor data from the continuous glucose sensor and activate an alarm state based on sensor data that meets one or more activation transition criteria associated with a hypoglycemic or hyperglycemic state; provide an output associated with the activation of the alarm state, wherein the output indicates a hypoglycemic or hyperglycemic state; transition from an activated state to a confirmed state over a period of time in response to at least one data indicating that the subject's glucose is trending toward normal blood glucose; actively monitor data associated with the subject's hypoglycemic or hyperglycemic state over a period of time in the confirmed state; and transition from the confirmed state to at least one of an inactive state or an activated state in response to data associated with the subject's hypoglycemic or hyperglycemic state that meets one or more predetermined criteria. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the transition from an active state to a confirmed state includes a transition from an active state to a confirmed state based on data indicating that the subject's glucose is trending toward normal blood glucose levels, wherein the data is selected from a) sensor data indicating changes in glucose trends or b) insulin information related to correction of the condition. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the transition from an active state to a confirmed state includes a transition from an active state to a confirmed state based on user confirmation, wherein the data is selected from a) user confirmation of an alarm on a user interface or b) user input of insulin information or c) user input of dietary information. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—active monitoring includes monitoring at least one of sensor data, sensor diagnostic information, dietary information, insulin information, or event information. In one implementation of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or implementation given herein), and particularly applicable to any other implementation of the fourth aspect—the transition from the confirmed state includes a transition from the confirmed state to the inactive state based on sensor data that no longer meets one or more criteria associated with a hypoglycemic or hyperglycemic state.In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the transition from a confirmed state includes a transition from a confirmed state to an inactive state based on sensor data satisfying one or more inactive transition criteria, wherein the inactive criteria differ from one or more active transition criteria associated with a hypoglycemic or hyperglycemic state. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the transition from a confirmed state includes a transition from a confirmed state to an inactive state based on insulin data and / or dietary information. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the transition from a confirmed state includes a transition from a confirmed state to an active state based on the satisfaction of one or more active transition criteria associated with a hypoglycemic or hyperglycemic state and based on an expiration of a predetermined time period. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—after receiving confirmation of the alarm status from the user and determining that data indicates the subject's glucose is trending towards normal blood sugar, the system further includes transitioning from a confirmed state to an active state during the active monitoring period based on the subject's glucose trending away from normal blood sugar. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the system further includes selectively outputting information related to the state transition. In one embodiment of the fourth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fourth aspect—the output associated with transitioning to an active state differs from the output associated with transitioning from a confirmed state to an inactive state.

[0013] In the fifth aspect—which can be independently combined with any aspect or implementation described herein—a method is provided for determining when to re-alarm a user after the user has acknowledged a first alarm. This method includes: initially activating an alarm state based on one or more criteria met, said one or more criteria being based on data related to a hypoglycemic or hyperglycemic state; transitioning to an acknowledgment state during a predetermined active monitoring period in response to user acknowledgment or indication that the user's glucose is trending towards normal; actively monitoring data related to the user's hypoglycemic or hyperglycemic state by a processor module during the active monitoring period; and re-activating the first alarm state during the acknowledgment period initiated by data related to the user's hypoglycemic or hyperglycemic state that meets one or more second criteria. In one implementation of the fifth aspect—which is generally applicable (i.e., independently combined with any aspect or implementation described herein), and particularly applicable to any other implementation of the fifth aspect—the second one or more criteria differ from the first one or more criteria. In one implementation of the fifth aspect—which is generally applicable (i.e., independently combined with any aspect or implementation described herein), and particularly applicable to any other implementation of the fifth aspect—the method further includes providing a first output related to the initial activation and providing a second output related to the reactivation. In one embodiment of the fifth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fifth aspect—the first output and the second output differ. In one embodiment of the fifth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fifth aspect—the second criterion includes indicating a state where the subject's glucose tends towards normal blood glucose levels and further includes indicating a state where the subject's glucose tends towards moving away from normal blood glucose levels after tending towards normal blood glucose levels during the active monitoring period. In one embodiment of the fifth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the fifth aspect—one or more second criteria associated with reactivation include changes in real-time glucose values ​​compared to real-time glucose values ​​associated with initial activation.

[0014] In the sixth aspect—which can be independently combined with any aspect or embodiment given herein—a system for processing data is provided, the system comprising: a continuous analyte sensor configured to be implanted in the body; and sensor electronics configured to receive and process sensor data output by the sensor, the sensor electronics including a processor configured to: initially activate an alarm state based on one or more criteria satisfied, the one or more criteria being based on data related to hypoglycemia or hyperglycemia; transition to a confirmed state during a predetermined active monitoring period in response to at least one piece of data indicating that the subject's glucose is trending toward normal; actively monitor data related to the subject's hypoglycemia or hyperglycemia during the active monitoring period; and reactivate the first alarm state during the confirmed period initiated by data related to the subject's hypoglycemia or hyperglycemia satisfied by one or more second criteria. In one embodiment of the sixth aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), particularly applicable to any other embodiment of the sixth aspect—the second or more criteria differ from the first or more criteria. In one embodiment of the sixth aspect—which is generally applicable (i.e., can be independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the sixth aspect—the system further includes providing a first output associated with initial activation and providing a second output associated with reactivation. In one embodiment of the sixth aspect—which is generally applicable (i.e., can be independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the sixth aspect—the first output and the second output are different. In one embodiment of the sixth aspect—which is generally applicable (i.e., can be independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the sixth aspect—the second criterion includes indicating a state where the subject's glucose tends towards normal blood glucose levels and further includes indicating a state where the subject's glucose tends towards deviating from normal blood glucose levels after tending towards normal blood glucose levels during active monitoring time. In one embodiment of the sixth aspect—which is generally applicable (i.e., can be independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the sixth aspect—one or more second criteria associated with reactivation include changes in real-time glucose values ​​compared to real-time glucose values ​​associated with initial activation.

[0015] In the seventh aspect—which can be independently combined with any aspect or implementation described herein—a method for avoiding unnecessary hyperglycemia alarms is provided, the method comprising: initially activating a first alarm state based on one or more first criteria associated with a hyperglycemic state; waiting for a period of time before providing output related to the first alarm state; during the waiting period, actively monitoring data related to the subject's hyperglycemic state by a processor module; and after the waiting period, providing output related to the first alarm state based on data related to the subject's hyperglycemic state that meets one or more second criteria. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the seventh aspect—active monitoring includes determining the average glucose within a time window. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the seventh aspect—active monitoring includes determining the magnitude and / or direction of the rate of change. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the seventh aspect—active monitoring includes determining the magnitude and / or direction of the rate of acceleration. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—active monitoring includes assessing insulin information. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—active monitoring includes assessing dietary information or timing. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—the waiting time period is user-selectable. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—the method further includes not providing output related to the first alarm state after the waiting time period based on data related to the subject's hyperglycemic status that does not meet one or more second criteria. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—one or more first criteria and one or more second criteria are the same. In one implementation of the seventh aspect—which is generally applicable (i.e., independently in combination with any aspect or implementation given herein), and particularly applicable to any other implementation of the seventh aspect—one or more first criteria and one or more second criteria differ.In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—the method further includes transitioning from a first alarm state to an inactive alarm state based on data related to the subject's hyperglycemic status that meets one or more third criteria. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—one or more first criteria and one or more third criteria are the same. In one embodiment of the seventh aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the seventh aspect—one or more first criteria and one or more third criteria are different.

[0016] In the eighth aspect—which may be independently combined with any aspect or embodiment given herein—a system for processing data is provided, the system comprising: a continuous analyte sensor configured for implantation; and sensor electronics configured to receive and process sensor data output by the sensor, the sensor electronics including a processor configured to: initially activate a first alarm state based on one or more first criteria associated with a hyperglycemic state; wait for a time period before providing output associated with the first alarm state; actively monitor data associated with the subject's hyperglycemic state during the waiting time period; and, after the waiting time period, provide output associated with the first alarm state based on data associated with the subject's hyperglycemic state satisfying one or more second criteria. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—active monitoring includes determining average glucose within a time window. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently combined with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—active monitoring includes determining the magnitude and / or direction of the rate of change. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—active monitoring includes determining the magnitude and / or direction of the acceleration rate. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—active monitoring includes assessing insulin information. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—active monitoring includes assessing dietary information or timing. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—the waiting time period is user-selectable. In one embodiment of the eighth aspect—which is generally applicable (i.e., independently in conjunction with any aspect or embodiment given herein), and particularly applicable to any other embodiment of the eighth aspect—the system further includes not providing output related to the first alarm state after the waiting time period based on data related to the subject's hyperglycemic status that does not meet one or more second criteria. In one implementation of the eighth aspect—which is generally applicable (i.e., independently in combination with any aspect or implementation given herein), and particularly applicable to any other implementation of the eighth aspect—one or more first criteria and one or more second criteria are identical.In one embodiment of the eighth aspect—which is generally applicable (i.e., can be independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the eighth aspect—one or more first criteria and one or more second criteria differ. In one embodiment of the eighth aspect—which is generally applicable (i.e., can be independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the eighth aspect—the system further includes a transition from a first alarm state to an inactive alarm state based on data related to the subject's hyperglycemic status that meets one or more third criteria. In one embodiment of the eighth aspect—which is generally applicable (i.e., can be independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the eighth aspect—one or more first criteria and one or more third criteria are the same. In one embodiment of the eighth aspect—which is generally applicable (i.e., can be independently combined with any aspect or implementation described herein), and particularly applicable to any other embodiment of the eighth aspect—one or more first criteria and one or more third criteria differ.

[0017] Any feature of the embodiments of the first, second, third, fourth, fifth, sixth, seventh, or eighth aspects may be applied to all aspects and embodiments given herein. Furthermore, any feature of the embodiments of the first, second, third, fourth, fifth, sixth, seventh, or eighth aspects may be independently combined with, in any manner, with other embodiments described herein, either partially or entirely; for example, one, two, three, or more embodiments may be combined wholly or partially. Further, any feature of the embodiments of the first, second, third, fourth, fifth, sixth, seventh, or eighth aspects may be optional for other aspects or embodiments. Any aspect or embodiment of the method may be performed by any system or apparatus of any aspect or embodiment, and any aspect or embodiment of the system may be configured to perform the method of any aspect or embodiment. Attached Figure Description

[0018] Details of this disclosure regarding its structure and operation can be understood by studying the accompanying drawings, in which the same reference numerals refer to the same parts. The drawings are not necessarily to scale, but are intended to illustrate the principles of this disclosure.

[0019] Figure 1 This is a schematic diagram of a continuous analyzer sensor system that is attached to the subject and communicates with multiple instance devices.

[0020] Figure 2 It is a block diagram, and its illustration is similar to... Figure 1 Electronic devices connected to the sensor system.

[0021] Figure 3A This diagram illustrates one implementation method, in which... Figure 1 The receiver displays a digital representation of the estimated analyte values ​​on its user interface.

[0022] Figure 3B This diagram illustrates one implementation method, in which... Figure 1 The receiver displays estimated glucose values ​​and one-hour historical data trends on its user interface.

[0023] Figure 3C This diagram illustrates one implementation method, in which... Figure 1 The receiver displays estimated glucose levels and three-hour historical trend data on its user interface.

[0024] Figure 3D This diagram illustrates one implementation method, in which... Figure 1 The receiver displays estimated glucose levels and nine-hour historical trend data on its user interface.

[0025] Figure 4A , 4B Figure 4C is a diagram of the receiver's liquid crystal display, which shows an implementation of the screen display.

[0026] Figure 4D It is a screenshot of a smartphone, illustrating one implementation of an alarm that indicates the user's blood sugar is dropping and will soon be in a low range.

[0027] Figure 4E This is a screenshot from a smartphone, showing one implementation of a graphical representation of blood glucose trends.

[0028] Figure 4F This is one implementation of the blood glucose trend arrow.

[0029] Figure 5 It is a graphical representation of a continuous trace of glucose values ​​measured over a time range according to embodiments of the present disclosure.

[0030] Figure 6 It is a flowchart illustrating the process of dynamically and intelligently providing predictive alarms / alarms according to embodiments of this disclosure.

[0031] Figure 7 It is a graphical representation of a continuous trace of glucose values ​​measured over a time range according to embodiments of the present disclosure.

[0032] Figure 8 It is a flowchart illustrating the process of dynamically and intelligently monitoring the status after an alarm / alarm is triggered, according to an embodiment of this disclosure.

[0033] Figure 9 It is a flowchart illustrating the process of determining state changes according to an embodiment of this disclosure.

[0034] Figure 10 It is a flowchart illustrating the process of determining whether the reactivation conditions are met according to the embodiments of this disclosure.

[0035] Figure 11 It is a diagram of a state diagram that shows the transitions from various states according to embodiments of this disclosure.

[0036] Figure 12-16 This is an example diagram showing the estimated glucose value (“EGV”) and, according to an embodiment of this disclosure, when an alarm is expected to be provided for the EGV. Detailed Implementation

[0037] The following detailed description of the present embodiment is illustrated with reference to the accompanying drawings. In the drawings, reference numerals denote elements of the present embodiment. These reference numerals are reproduced below in conjunction with the discussion of the corresponding features of the drawings.

[0038] Sensor systems and users Figure 1 An example system 100 is depicted according to some implementation methods. System 100 includes a continuous analyte sensor system 8, which includes sensor electronics 12 and a continuous analyte sensor 10. System 100 may include other devices and / or sensors, such as a drug delivery pump 2 and a glucose meter 4. The continuous analyte sensor 10 may be physically connected to the sensor electronics 12 and may be integrated with (e.g., non-detachably connected to) or detachably connected to the continuous analyte sensor 10. The sensor electronics 12, the drug delivery pump 2, and / or the glucose meter 4 may be coupled to one or more devices, such as display devices 14, 16, 18, and / or 20.

[0039] In some implementations, system 100 may include a cloud-based analyte processor 490 configured to analyze analyte data (and / or other patient-related data) provided via network 406 (e.g., via wired, wireless, or a combination thereof) from sensor system 8 and other devices, such as display devices 14-20 and the like, which are relevant to the subject (also referred to as the patient) and configured to generate reports providing high-level information, such as statistics, about the measured analytes over a given time period. A full discussion of the use of cloud-based analyte processing systems can be found in U.S. Patent Application 61 / 655,991, entitled “Cloud-Based Processing of Analyte Data,” filed June 5, 2012, which is incorporated herein by reference in its entirety.

[0040] In some implementations, sensor electronics 12 may include electronic circuitry associated with measuring and processing data generated by the continuous analyte sensor 10. The generated continuous analyte sensor data may include algorithms that can be used to process and calibrate the continuous analyte sensor data, although these algorithms may also be provided in other ways. Sensor electronics 12 may include hardware, firmware, software, or a combination thereof to provide measurements of analyte levels via a continuous analyte sensor, such as a continuous glucose sensor. Further implementations of sensor electronics 12 are described below regarding... Figure 2 Describe it.

[0041] As used herein, the term "sensor data" is a broad term and is given its common and customary meaning to those skilled in the art (and is not limited to a specific or prescribed meaning), and further, without limitation, refers to any data relating to a sensor, such as a continuous analyte sensor. Sensor data includes raw data streams, or simply data streams, of analog or digital signals (or other signals received from another sensor) relating to the analyte being measured from an analyte sensor, as well as calibrated and / or filtered raw data. In one instance, sensor data includes "counting" digital data converted from analog signals (e.g., volts or amperes) by an A / D converter, and includes one or more data points representing glucose concentration. Therefore, the terms "sensor data point" and "data point" generally refer to a digital representation of sensor data at a specific time. The term broadly includes multiple time-interval data points from a sensor, such as a basic continuous glucose sensor, encompassing various measurements acquired over time intervals ranging from fractions of a second to, for example, 1, 2, or 5 minutes or longer. In another instance, sensor data includes integrated digital values ​​representing one or more data points averaged over a period of time. In some instances, sensor data may include calibrated data, smoothed data, filtered data, transformed data, and / or any other sensor-related data.

[0042] As described, sensor electronics 12 can connect (e.g., wirelessly, etc.) to one or more devices, such as display devices 14, 16, 18, and / or 20. Display devices 14, 16, 18, and / or 20 can be configured to present information (and / or alarms), such as sensor electronics 12 transmitting sensor information for display on display devices 14, 16, 18, and / or 20.

[0043] Display devices may include relatively small keychain-like display devices 14, relatively large handheld display devices 16, mobile phones 18 (e.g., smartphones, tablets, etc.), computers 20 and / or any other user devices configured to present at least information (e.g., medication delivery information, separate self-monitoring glucose readings, heart rate monitors, calorie intake monitors, etc.).

[0044] In some implementations, the relatively small keychain-like display device 14 may include a watch, strap, necklace, pendant, piece of jewelry, adhesive patch, pager, keychain ornament, plastic card (e.g., credit card), identity (ID) card, and / or the like. This small display device 14 may include a relatively small display (e.g., smaller than the large display device 16) and may be configured to display certain types of displayable sensor information, such as numerical values ​​and arrows.

[0045] In some implementations, the relatively large handheld display device 16 may include a handheld receiver device, a PDA, and / or the like. This large display device may include a relatively large display (e.g., larger than the small display device 14) and may be configured to display information, such as a graphical representation of continuous sensor data, including current and historical sensor data output by the sensor system 8.

[0046] In some implementations, the continuous analyte sensor 10 includes a sensor for detecting and / or measuring the analyte, and the continuous analyte sensor 10 can be configured as a non-invasive device, subcutaneous device, percutaneous device, and / or intravascular device for continuous detection and / or measurement of the analyte. In some implementations, the continuous analyte sensor 10 can analyze a variety of intermittent blood samples, although other analytes may also be used.

[0047] In some implementations, the continuous analyte sensor 10 may include a glucose sensor configured to measure glucose in the blood, employing one or more measurement techniques, such as enzymatic, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, iontophoresis, radiometric, immunochemical, and similar techniques. In implementations where the continuous analyte sensor 10 includes a glucose sensor, the glucose sensor may include any device capable of measuring glucose concentration and may employ a variety of techniques to measure glucose, including invasive, minimally invasive, and non-invasive sensing techniques (e.g., fluorescence monitoring), to provide data indicating glucose concentration in the subject, such as a data stream. The data stream may be a raw data signal that is converted into a calibrated and / or filtered data stream for providing glucose values ​​to the subject, such as a user, patient, or caregiver (e.g., a parent, relative, guardian, teacher, doctor, nurse, or any other individual interested in the subject's health). Furthermore, the continuous analyte sensor 10 may be implanted as at least one of the following types of sensors: implantable glucose sensor, percutaneous glucose sensor, implanted in the subject's blood vessel or via extracorporeal circulation, subcutaneous sensor, re-implantable subcutaneous sensor, and intravascular sensor.

[0048] Although this document refers to some implementations including a continuous analyte sensor 10 containing a glucose sensor, the continuous analyte sensor 10 may also include other types of analyte sensors. Furthermore, while some implementations refer to a glucose sensor as an implantable glucose sensor, other types of devices capable of detecting glucose concentration and providing an output signal representing glucose concentration may also be used. Further, although this specification refers to glucose as an analyte to be measured, processed, etc., other analytes may also be used, including, for example, ketone bodies (e.g., acetone, acetoacetic acid and β-hydroxybutyrate, lactate, etc.), glucagon, acetyl-CoA, triglycerides, fatty acids, intermediates in the citric acid cycle, choline, insulin, cortisol, testosterone, etc.

[0049] Figure 2 An example of a sensor electronics device 12 performing according to some examples is described. Sensor electronics device 12 may include sensor electronics configured to process sensor information, such as sensor data, and to generate transformed sensor data and displayable sensor information, for example, via a processor module. For example, the processor module may transform sensor data into one or more of the following: filtered sensor data (e.g., one or more filtered analyte concentration values), raw sensor data, corrected sensor data (e.g., one or more corrected analyte concentration values), rate of change information, trend information, acceleration / deceleration rate information, sensor diagnostic information, location information, alarm / alarm information, calibration information, sensor data smoothing and / or filtering algorithms, and / or similar information.

[0050] In some instances, processor module 214 is configured to perform most (if not all) of the data processing. Processor module 214 may be integrated with sensor electronics 12 and / or may be remotely located, for example, within one or more of devices 14, 16, 18, and / or 20 and / or cloud 490. In some instances, processor module 214 may include multiple smaller sub-components or sub-modules. For example, processor module 214 may include an alarm module (not shown) or a prediction module (not shown), or any other suitable module for effectively processing the data. When processor module 214 comprises multiple sub-modules, the sub-modules may be located within processor module 214, including within sensor electronics 12 or other related devices (e.g., 14, 16, 18, 20, and / or 490). For example, in some instances, processor module 214 may be at least partially located within cloud-based analytics processor 490 or elsewhere in network 406.

[0051] In some implementations, processor module 214 can be configured to calibrate sensor data, and data storage memory 220 can store the calibrated sensor data points as transformed sensor data. Furthermore, in some implementations, processor module 214 can be configured to wirelessly receive calibration information from display devices such as devices 14, 16, 18, and / or 20 to calibrate sensor data from sensor 12. Further, processor module 214 can be configured to perform additional algorithmic processing on sensor data (e.g., calibrated and / or filtered data and / or other sensor information), and data storage memory 220 can be configured to store transformed sensor data and / or sensor diagnostic information related to this algorithm.

[0052] In some implementations, sensor electronics 12 may include an application-specific integrated circuit (ASIC) 205 coupled to user interface 222. ASIC 205 may further include a voltage regulator 210, a telemetry module 232 for transmitting data from sensor electronics 12 to one or more devices such as devices 14, 16, 18, and / or 20, and / or other components for signal processing and data storage (e.g., processor module 214 and data storage memory 220). Although Figure 2 The ASIC 205 is described, but other types of circuitry may also be used, including field-programmable gate arrays (FPGAs), one or more microprocessors configured to provide some (if not all) of the processing performed by the sensor electronics 12, analog circuitry, digital circuitry, or combinations thereof.

[0053] exist Figure 2 In the depicted example, voltage regulator 210 is coupled to continuous analyte sensor 10, such as a glucose sensor, to generate sensor data for the analyte. Voltage regulator 210 can also provide voltage to continuous analyte sensor 10 via data line 212 to apply a bias voltage to the sensor for measuring values ​​(e.g., current, etc.) indicating the concentration of the analyte (also referred to as the analog portion of the sensor). Depending on the number of operating electrodes of continuous analyte sensor 10, voltage regulator 210 may have one or more channels.

[0054] In some implementations, regulator 210 may include a resistor that converts the current value from sensor 10 into a voltage value. In some implementations, a current-to-frequency converter (not shown) may also be configured to continuously integrate the measured current value from sensor 10 using, for example, a charge counting device. In some implementations, an analog-to-digital converter (not shown) may digitize the analog signal from sensor 10 into a so-called "count" to allow processing by processor module 214. The resulting count may be directly correlated with the current measured by regulator 210, which may be directly correlated with the level of an analyte in the sample, such as glucose level.

[0055] Telemetry module 232 may be operatively connected to processor module 214 and may provide hardware, firmware, and / or software enabling wireless communication between sensor electronics 12 and one or more other devices, such as display devices, processors, network access devices, etc. Various wireless radio technologies that may be implemented in telemetry module 232 include Bluetooth, Bluetooth Low Energy, ANT, ANT+, ZigBee, IEEE 802.11, IEEE 802.16, cellular wireless access technologies, radio frequency (RF), infrared (IR), paging network communication, magnetic induction, satellite data communication, spread spectrum communication, frequency hopping communication, near-field communication, and / or similar technologies. In some implementations, telemetry module 232 includes a Bluetooth chip, although Bluetooth technology may also be implemented in a combination of telemetry module 232 and processor module 214.

[0056] Processor module 214 can control the processing performed by sensor electronics 12. For example, processor module 214 can be configured to process data from the sensor (e.g., counting), filter the data, correct the data, perform fail-safe checks, and / or similar processing.

[0057] In some implementations, processor module 214 may include a digital filter, such as an infinite impulse response (IIR) or finite impulse response (FIR) filter. This digital filter smooths the raw data stream received from sensor 10. Typically, the digital filter is programmed to filter data sampled at predetermined time intervals (also known as the sampling rate). In some implementations, such as when regulator 210 is configured to measure an analyte (e.g., glucose and / or the like) at separate time intervals, these time intervals determine the sampling rate of the digital filter. In some implementations, regulator 210 may be configured for continuous analyte measurement, for example, using a current-to-frequency converter. In these current-to-frequency converter implementations, processor module 214 may be programmed to request digital values ​​from the integrator of the current-to-frequency converter at predetermined time intervals (acquisition times). Due to the continuity of current measurement, these digital values ​​obtained by processor module 214 from the integrator can be averaged over the acquisition time. Therefore, the acquisition time can be determined by the sampling rate of the digital filter.

[0058] Processor module 214 may further include a data generator (not shown) configured to generate data packets for transmission to devices, such as display devices 14, 16, 18, and / or 20. Further, processor module 214 may generate data packets transmitted to these external sources via telemetry module 232. In some implementations, the data packets may be customized for each display device as described, and / or may include any useful data, such as timestamps, displayable sensor information, converted sensor data, identification codes of the sensor and / or sensor electronics 12, raw data, filtered data, corrected data, rate of change information, trend information, error detection or correction, and / or similar data.

[0059] Processor module 214 may also include program memory 216 and other memory 218. Processor module 214 may be coupled to a communication interface, such as communication port 238, and a power source, such as battery 234. Furthermore, battery 234 may be further coupled to battery charger and / or regulator 236 to provide power to sensor electronics 12 and / or to charge battery 234.

[0060] Program memory 216 can be executed as semi-static memory for storing data such as identifiers of the coupled sensor 10 (e.g., sensor identifier (ID)) and for storing code (also referred to as program code) to configure ASIC 205 to perform one or more operations / functions described herein. For example, program code may configure processor module 214 to process data streams or perform counting, filtering, calibration, fault-safe checks, etc.

[0061] Memory 218 can also be used to store information. For example, processor module 214, which includes memory 218, can be used as a cache memory for the system, providing temporary storage for the most recent sensor data received from the sensor. In some implementations, the memory may include memory storage components such as read-only memory (ROM), random access memory (RAM), dynamic RAM, static RAM, non-static RAM, erasable programmable read-only memory (EEPROM), rewritable ROM, flash memory, etc.

[0062] Data storage memory 220 can be coupled to processor module 214 and can be configured to store various sensor information. In some implementations, data storage memory 220 stores one or more days of continuous analyte sensor data. For example, data storage memory can store 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20 and / or 30 days (or more) of continuous analyte sensor data received from sensor 10. The stored sensor information may include one or more of the following: timestamps, raw sensor data (one or more raw analyte concentration values), calibrated data, filtered data, converted sensor data and / or any other displayable sensor information, calibration information (e.g., reference BG values ​​and / or previous calibration information), sensor diagnostic information, etc.

[0063] User interface 222 may include various interfaces, such as one or more buttons 224, a liquid crystal display (LCD) 226, an oscillator 228, a voice changer (e.g., a speaker) 230, a backlight (not shown), and / or the like. The components constituting user interface 222 provide control for interaction with a user (e.g., the person in question). One or more buttons 224 may allow, for example, toggling, menu selection, option selection, status selection, responding yes / no to on-screen questions, a "turn off" function (e.g., for alarms), a "confirm" function (e.g., for alarms), reset, and / or similar operations. LCD 226 may provide the user with, for example, visual data output. Voice changer 230 (e.g., a speaker) may provide sound signals in response to the triggering of certain alarms, such as presenting and / or predicting hyperglycemia and hypoglycemia conditions. In some implementations, the sound signals may be distinguished by tone, volume, duty cycle, mode, duration, and / or the like. In some implementations, the sound signal can be configured to be muted (e.g., confirmed or turned off) by pressing one or more buttons 224 on the sensor electronics 12 and / or by using a button or selection signal on a display device (e.g., a keychain ornament, a mobile phone, and / or the like) to the sensor electronics 12.

[0064] Although about Figure 2 Sound and vibration alarms are described, but other alarm mechanisms may also be used. For example, in some implementations, tactile alarms are provided, including push-poke mechanisms configured to "poke" or physically contact the patient in response to one or more alarm conditions.

[0065] Battery 234 can be operatively connected to processor module 214 (and possibly other components of sensor electronics 12) and provide the necessary power to sensor electronics 12. In some implementations, the battery is a lithium manganese dioxide battery; however, any battery of suitable size and power can be used (e.g., AAA, nickel-cadmium, zinc-carbon, alkaline, lithium, nickel-metal hydride, lithium-ion, zinc-air, zinc-mercury oxide, silver-zinc, or sealed). In some implementations, the battery is rechargeable. In some implementations, multiple batteries can be used to power the system. In other implementations, for example, the receiver can be powered via an inductive connection.

[0066] The battery charger and / or regulator 236 may be configured to receive energy from internal and / or external chargers. In some implementations, the battery regulator (or balancer) 236 regulates the charging process by discharging excess charging current to allow all batteries or battery packs in the sensor electronics 12 to be fully charged without overcharging others. In some implementations, the battery 234 (or multiple batteries) is configured to be charged via an inductive and / or wireless charging pad, although any other charging and / or power supply mechanism may also be used.

[0067] One or more communication ports 238, also known as external connectors, may be provided to allow communication with other devices, such as a PC communication (com) port, to enable system communication separate from or integrated with the sensor electronics 12. The communication ports may include, for example, a serial (e.g., Universal Serial Bus or "USB") communication port and allow communication with another computer system (e.g., a PC, personal digital assistant, or "PDA" server or the like). In some implementations, the sensor electronics 12 may be able to transmit historical data to a PC or other computing device (e.g., the analyte processor disclosed herein) for retrospective analysis by patients and / or physicians.

[0068] In some continuous analyte sensor systems, the skin portion of the sensor electronics can be simplified to minimize the complexity and / or size of the on-skin electronics, for example, by providing only raw, corrected, and / or filtered data to a display device configured to run additional algorithms required for correcting and displaying the sensor data. However, the sensor electronics 12 (e.g., via processor module 214) can execute intended algorithms for generating transformed sensor data and / or displayable sensor information, including, for example, performing algorithms such as: assessing the clinical acceptability of reference and / or sensor data; optimizing correction based on corrected data incorporating standard assessments; assessing correction quality; comparing estimated analyte values ​​with time-corresponding measured analyte values; analyzing changes in estimated analyte values; assessing the stability of the sensor and / or sensor data; detecting signal artifacts (noise); replacing signal artifacts; determining the rate of change and / or trend of the sensor data; performing dynamic and intelligent analyte value estimation; performing diagnostics on the sensor and / or sensor data; setting operating modes; and assessing data anomalies and / or similar operations.

[0069] Although Figure 2 The display shows separate data storage and program storage, and various configurations are also available. For example, one or more storage devices can be used to provide storage space to support the data processing and storage requirements of the sensor electronics 12.

[0070] Now see Figures 3A to 3D A more detailed schematic diagram of the handheld receiver 16 is shown. The handheld receiver 16 may include systems required for receiving, processing, and displaying sensor data from an analyte sensor, such as those described elsewhere herein. Specifically, the handheld receiver 16 may be, for example, a pager-sized device and includes a user interface with multiple buttons 242 and a liquid crystal display (LCD) screen 244, and may include backlighting. In some instances, the user interface may include a keyboard, speaker, and oscillator.

[0071] In some instances, users can switch using a toggle button on the handheld receiver. Figures 3A to 3D Some or all of the screens are displayed. In some instances, the user is able to interactively select the type of output displayed on their user interface. In some instances, the sensor output may be selectively configurable.

[0072] In some instances, the analyte value is displayed on a monitor, such as that of the receiving medical device. In other instances, prompts or messages may be displayed on the display device to convey information to the user, such as reference to abnormal values, requests for reference to analyte values, treatment recommendations, deviations between measured and estimated analyte values, provision of predictive alarms / alarms, monitoring of blood glucose alarm status after triggering an alarm / alarm, identification of status changes, etc. Furthermore, prompts may be displayed to guide the user in performing calibration or troubleshooting.

[0073] Figures 4A to 4C The illustration shows some additional visual displays that can be provided on user interface 222. While these visual displays may include the same or similar output as displayed on handheld device 16 in FIG. 3, the visual displays of FIG. 4 can be provided on any suitable user interface 222, such as those on devices 14, 16, 18, and 20. In some instances, LCD 226 is a touch-activated screen, and each selection can be made by the user, for example, from a menu on the screen. Buttons are provided for, for example, toggling, menu selection, option selection, mode selection, and reset. In some alternative implementations, a microphone may be provided to allow voice-activated control.

[0074] Figure 4A This is a graphical representation of an LCD screen 226 that displays continuous and single-point glucose information in the form of a trend graph 184 and individual numerical values ​​186. The trend graph shows upper and lower boundaries 182, representing the target range within which the user should maintain their glucose levels. Preferably, the receiver is configured such that these boundaries 182 can be configured or customized by a user, such as the user or observer. By providing visual boundaries 182, combined with continuous analyte values ​​over time (e.g., trend graph 184), compared to only a single point (e.g., a single numerical value 186), the user can better learn how to control their analyte concentration (e.g., a person with diabetes can better learn how to control their glucose concentration). Although Figure 4A A trend chart for 1 hour can be displayed (e.g., described using a 1-hour time range 188), but multiple time ranges, such as 3 hours, 9 hours, 1 day, etc., can be presented on screen 226.

[0075] Figure 4B This is a diagram of an LCD screen 226 displaying a low alarm screen, which is displayed in response to the subject's analyte concentration falling below a lower boundary (see boundary 182). In this example screen, for example, the subject's glucose concentration drops to 55 mg / dL, which is below... Figure 4AThe lower limit is set. Arrow 190 indicates the direction of the analyte trend, such as indicating that glucose concentration is continuing to decrease. Label 192 ("Low") helps to immediately and clearly warn the person that their glucose concentration has dropped below the preset limit and what is considered a clinically safe value.

[0076] In comparison, Figure 4C This is an illustration of an LCD screen 226 displaying a high alarm screen, which is activated in response to the user's analyte concentration rising above the upper limit (see boundary 182). In this example screen, the user's glucose concentration has risen to 200 mg / dL, exceeding the user's set boundary, thus triggering the high boundary screen. Arrow 190 represents the direction of the analyte trend, for example, indicating that the glucose concentration is continuing to rise. The label 192 ("High") helps to immediately and clearly warn the user, for example, that his / her glucose concentration is above the preset limit and what is considered a clinically safe value.

[0077] While this document describes several example screens, various screens may be provided to illustrate any information described in the provided embodiments, as well as other information. Users may switch between these screens, and / or the screens may be automatically displayed in response to programming, for example, in the handheld receiver 16, and may be accompanied by another type of alarm (e.g., sound or touch).

[0078] For example, Figure 4D It is a screenshot of a smartphone 18 display, illustrating one implementation of an alarm that indicates the user's blood sugar is dropping and will soon be in a low range. Figure 4E It is a screenshot of a smartphone 18 display, illustrating one implementation of a blood glucose trend graph. Figure 4F This is one implementation of the blood glucose trend arrow.

[0079] In some instances, processor module 214 can provide a predictive alarm on the smartphone 18 display or user interface 222 when a severe hypoglycemic event is anticipated to occur in the near future. For example, the predictive alarm could be triggered when a severe hypoglycemic event is predicted to occur within 5 minutes, 10 minutes, 15 minutes, 20 minutes, 30 minutes, etc. See also Figure 4DArrow 300 can be displayed on the trend screen, pointing to a BG value 302 indicating a severe hypoglycemic event, such as 55 mg / dL. Arrow 300 can change color as blood glucose transitions from normal to hypoglycemia to emphasize the expected change in glucose level. Furthermore, arrow 300 can be made vivid to flash, emphasizing the severity of the alarm. The display can show text 304, such as "becoming lower." This predictive alarm can be configured to precede (take precedence over) any mode or application currently in use by smartphone 18 when processor module 214 determines that a severe hypoglycemic event is predicted to occur. In other words, the alarm can interrupt any current alarms on smartphone user interface 222.

[0080] In these embodiments, processor module 214 can be programmed with blood glucose values ​​corresponding to a threshold below which the user is considered hypoglycemic. As processor module 214 receives multiple EGV values ​​as input at time intervals, it processes each input by comparing it to the programmed value and also to previously received EGV values. If the user's blood glucose shows a downward trend and approaches the programmed value, processor module 214 outputs an alarm, for example... Figure 4D The smartphone user interface 222 is shown. Therefore, the user receives an early warning of a possible hypoglycemic event so that he or she can take appropriate measures to avoid it.

[0081] In various other embodiments, processor module 214 may change the color of user interface 222 to reflect the user's current blood glucose level. For example, the user's EGV may be displayed on the screen as a number, as a trend graph, a horizontal bar chart, etc. The text and / or background on user interface 222 may change as the user's current blood glucose level transitions from one state to another. For example, if the user's blood glucose is within a healthy range, the text / background may display a first color, such as green, and if the user's blood glucose is low or high, a second color, such as red, may be displayed. Optionally, the first color may be used for the healthy range, the second color for low, and the third color for high. Further, when in the low or high range, the intensity of the color may be increased as the user's blood glucose becomes increasingly lower or higher. The text / background may also blink, with the blinking frequency increasing as the user's blood glucose becomes increasingly lower or higher.

[0082] In these embodiments, processor module 214 can be programmed with blood glucose values ​​corresponding to low and high threshold BG values. As processor module 214 receives multiple EGVs as input at time intervals, it processes each input by comparing it with a programmed value. If the user's blood glucose value exceeds one of the thresholds, processor module 214 outputs an alarm to smartphone user interface 222 in the form of a change in the color of text and / or background. If the user's blood glucose value continues to become increasingly lower or higher, processor module 214 generates additional output, such as increasing the intensity of the color and / or causing the text / background to flash. These additional outputs may be generated in response to processor module 214 comparing the input EGV with additionally programmed thresholds.

[0083] In various other embodiments, processor module 214 may use illustrative symbols and / or alarm symbols reflecting real-time data. For example, if a user's blood sugar drops, an icon on smartphone 18 may display an image of the user's blood sugar trend, such as using actual data points from EGV. The input-processing-output of this embodiment is substantially the same as in the previous embodiments.

[0084] Extremely low blood sugar can cause a person to lose consciousness. Therefore, in some current embodiments, the processor module 214 may enter an emergency response indication mode upon the occurrence of a predetermined level or event that could indicate loss of consciousness (low blood sugar level, failure to press the button after an alarm, etc.). This mode may include false alarms from others near the alarm user. For example, the smartphone user interface 222 may display step-by-step instructions on how to assist an unconscious user, such as administering glucose tablets or other forms of sugar, calling an ambulance, etc.

[0085] In these embodiments, processor module 214 may receive input from CGM, which is the user's EGV. Processor module 214 may process the input by comparing it with one or more thresholds and determine that the user's blood glucose is low. Processor module 214 may generate an output in the form of an alarm. If the user does not respond to the alarm by pressing a button or an area on the touchscreen user interface 222, processor module 214 may determine, for example, that the user may be unconscious and generate another output in the form of an emergency response indication mode as described herein.

[0086] In various other implementations, processor module 214 can provide a distinguishable comparison between visually high / low thresholds and alarm thresholds. For example, processor module 214 can be programmed with low and high blood glucose thresholds. These thresholds can be displayed as horizontal lines on a blood glucose trend graph on user interface 222, which the user should strive to avoid. Typically, exceeding either line may trigger an alarm. However, too many alarms can annoy the user and may reduce patient compliance. Therefore, in some instances, the boundaries of the visually high / low target ranges displayed on the graph may differ from the boundaries that trigger alarms. For example, the boundaries that trigger alarms may be wider than the visual target range threshold boundaries on user interface 222, and the boundaries that trigger alarms may be hidden and invisible. This configuration gives the user a buffer to exceed either visual boundary without triggering an alarm. Alternatively, the boundaries that trigger alarms may be visible but can be distinguished from the target range boundaries. Examples of visual distinction may include different colors, flashing versus static, solid versus dashed lines, different line thicknesses, alarm icons adjacent to alarm lines, and so on. In some instances, high / low target boundaries may always be displayed, but alarm boundaries may be displayed or not, depending on user settings, modes (e.g., silent), thresholds, etc.

[0087] In various other embodiments, the user interface of the processor module 214 may be the first thing the user sees when the user activates the user interface 222 of the smartphone. For example, many smartphones 18 automatically put the user interface 222 to sleep (e.g., into sleep mode) after a predetermined amount of time without detected activity. This measure saves battery power. To reactivate the user interface 222, the user must press a button on the smartphone 18. In some embodiments, the first thing the user sees when the user activates the user interface 222 is the user interface of the processor module 214. In these embodiments, as input, the processor module 214 receives a notification that the user interface 222 has entered sleep mode, followed by a subsequent notification that the user interface 222 has been reactivated. The processor module 214 can process these inputs and, as output, produce the display of the user interface 222 of the processor module 214 on the smartphone 18.

[0088] In various other embodiments, the trend graph displayed by processor module 214 is color-coded. For example, see... Figure 4EIf within the target range, the color of Figure 400 (trend line 402 or background 404) can be green; if outside the target range by ±10%, it can be yellow; if outside the target range by ±15%, it can be orange; and if outside the target range by ±20%, it can be red. Similarly, the trend arrow 406 can be color-coded, and the angle at which the trend arrow 406 is directed can correspond to the actual rate of change of the user's glucose, for example, a more horizontal arrow indicates a low rate of change, while a steeply sloping arrow indicates a high rate of change. In these embodiments, the processor module 214 can receive continuous EGV as input from the sensor system 8. In some instances, the rate of change is calculated by the sensor system 8 and sent to the processor module 214 for display (e.g., determining how to display and the final display), although the processor module 214 can also perform the rate of change calculation. The rate of change based on a linear or nonlinear function is applied to a window of the most recent sensor data. In some instances, the rate of change calculation includes calculating at least two point-to-point rate of change calculations, and wherein the rate of change calculation also includes adaptively selecting a filter to apply to the point-to-point rate of change calculation based at least in part on a determined noise level. Processor module 214 can output these values ​​as data points on trend graph 400 on user interface 222, and also update the value 408 displayed in the small box containing the user's recent EGV. If the user's blood sugar is decreasing, processor module 214 outputs this information by pointing arrow 406 downwards; if the user's blood sugar is increasing, processor module 214 outputs this information by pointing arrow 406 upwards. In some instances, the trend arrow is located at the end of the trend graph (e.g., in a separate small box / area).

[0089] In some implementations, the size of the value 408 displayed in the small box containing the user's most recent EGV can vary depending on how far the user is from their target area. For example, the number can become larger as the user's glucose level moves further away from the target area. This expansion can be in one direction or either direction, meaning that if it is outside the target range in either direction, the EGV displayed on the trend graph may become larger, or if it is outside the target range on the lower side, it may simply become larger (e.g., hypodermic injection). The same applies to trend arrow 406. See also Figure 4F Trend arrow 406 can be drawn large enough to fit EGV 408 within arrow 406. Figure 4F The layout of the trend arrows 406 / EGV 408 can be adopted independently of the previous implementation, in which the size of the trend arrows 406 / EGV 408 changes dynamically with the user's glucose level.

[0090] In various other embodiments related to Figure 4, instead of using a hard threshold for transitioning from one color to the next, the display can show a gradient trend graph. That is, instead of transitioning directly from green to yellow once the user's glucose reaches a threshold, such as ±10% outside the target range, the display gradually transitions from green to yellow as the user's glucose moves away from the target range toward the established threshold. Thus, within ±5% outside the target range, the display shows a color between green and yellow, and as the user's glucose moves through ±6%, ±7%, ±8%, etc. outside the target range, the color gradually becomes more yellow.

[0091] Predictive alarms / alarms In some instances, the sensor outputs a signal in the form of current; however, any output signal from any measurement technique can be used for the predictive alarms / alarms described herein. Generally, a conversion function is applied to the sensor signal to produce a user output that the user understands as representing the concentration of the analyte in his or her bloodstream. Such a suitable conversion function can take into account many variables, such as sensitivity (slope), baseline (intercept), offset, temperature correction, factory source information, or other corrections or adjustments to the data. After applying a suitable conversion function, the user may see an output similar to, for example… Figures 3A to 3D The output displayed in the image.

[0092] Within the scope of preventing the consequences of diabetes, the goal is to prevent hypoglycemic and / or hyperglycemic events rather than simply generating an alarm when such events occur. For example, generating an alarm indicating that a hypoglycemic event will occur within 20 minutes without intervention would allow the individual or patient to consume and absorb carbohydrates in a timely manner.

[0093] Now see Figure 5 This shows an example of a continuous trace of glucose values, measured over a time range, with a threshold indicator x1 and a prediction indicator x2 placed on it. Figure 5 The glucose trace presented here is a graph that compares glucose levels, such as mg / dl, over a time range, such as 24 hours.

[0094] As shown, in some implementations, three thresholds or limits are used to monitor glucose levels: TV1, TV2, and TV. P TV1 can be set by the user and typically defines the upper limit or upper glucose level that the user can operate on before the monitor alarms. Similarly, TV2 can be set by the user and typically defines the lower limit or lower glucose level that the user can operate on before the monitor alarms. PThis is a prediction threshold, such as a threshold to which a predicted value is compared. It should be understood that while the illustrated implementation takes a threshold into account, a threshold range or other criteria (such as blood glucose status) may optionally be used.

[0095] As shown, TV P Specific glucose levels were not provided. This is because TV... P It can be a fixed or permanent value set during factory setup, rather than being user-defined. In some instances, TV... P This can represent dangerously low glucose levels, such as indicating a severe hypoglycemic event. In some instances, TV... P It can represent a value of 55 mg / dL or around that.

[0096] In some instances, TV P It can be adaptively determined based on TV2. For example, if the user sets TV2 to a relatively high value, such as 90 mg / dL, the algorithm or function can determine TV. P It should be set at 65 mg / dL. Conversely, if the user sets TV2 to a relatively low value, such as 70 mg / dL, the algorithm or function can determine TV. P The pH should be set at 55 mg / dL. Additionally or optionally, the predicted pH range can be preset or adaptively selected based on another prediction-related parameter, such as the selected TV2 and / or TV. P .

[0097] See also Figure 5 The diagram displays two indicators, x1 and x2. Indicator x1 is generally considered a threshold indicator and is configured to alert the user that a first threshold, such as TV2, has been met. In some instances, threshold indicator x1 is used to notify the user promptly or approximately promptly when a threshold has been met or exceeded (e.g., considering processing delays). An example of using threshold indicator x1 is when a user wants to be notified that their glucose level has reached a certain high or low value. When it is determined that the user's glucose level meets a predetermined user-defined value, the user can be notified via an alarm or alert. Additional rate of change conditions can be added (e.g., TV1 and a rate of change increasing at >0.5 mg / dL / min, or TV2 and a rate of change decreasing at >0.5 mg / dL / min).

[0098] The x2 indicator can generally be viewed as a predictive indicator and can be configured to warn users of a second threshold, such as TV. P The prediction is expected to be met within a predetermined or predefined time range or time scope PH. In some instances, the predictive indicator x2 is used to promptly notify the user that the predicted threshold has been exceeded or met within a predefined time range, such as 20 minutes.

[0099] Figure 5 One advantage of the glucose monitoring curve displayed is that the prediction can be paired with user-set threshold alarms, where an alarm is triggered for either the threshold or the prediction that is met first. This is achieved using pre-tuned prediction algorithm parameters (e.g., dependence on simulated and measured data, time-sensitive weighting of past data, prediction range pH, and prediction threshold TV). P It can improve or optimize the warning time given to patients before a serious event occurs, and can minimize the number of noise / false alarms heard by the user.

[0100] In some instances, the prediction parameter (TV) P The parameters (PH) can be invisible to the user and are preset or fixed. In some implementations, the prediction parameters are determined by the sensor manufacturer using various historical data, such as user history data, population history data, and specific sensor history data.

[0101] See back Figure 3A This displays a user's glucose trace over a day. The visible trace, as can be understood, typically falls within a shaded or boundary area 246. This area is generally referred to as the "target area" and urges the user to "stay between lines." This area is usually visible to the user on their monitor or a device with a viewable screen and can be used to quickly check how the user's glucose levels look over a specific period. Another example of this boundary area is shown... Figure 4A It is shown in the figure and marked as 182.

[0102] In some instances, the shaded "target area" differs from the areas defined by TV1 and TV2, which are referred to below as the "alarm boundaries." In some instances, TV1 and TV2 are invisible to the user but are internal values ​​used by appropriate algorithms or functions, for example, to warn the user when needed.

[0103] In other implementations, the alarm boundaries can be visible to the user. It should be understood that numerical values, images or icons, or simple labels such as "high" and "low" can be associated with TV1 and TV2 respectively. Furthermore, for example, TV... P It can be displayed as a simple alarm icon.

[0104] In some instances, predictive alarms can be activated using a simple on / off button. Additionally, in some instances, one or more general settings can be used for prediction, such as "sensitive," "normal," and "not annoying," to meet the needs of different users. For example, a sensitive predictive setting could be set to pH=30 minutes and TV... P =70 mg / dL; normal predictive settings can be set to pH = 20 min and TVP = 55 mg / dL; and a non-disgusting predictive setting can be set to pH = 10 min and TV P = 55 mg / dL.

[0105] It should be noted that although this disclosure focuses on the prediction range and criteria associated with predictions under low glucose (hypoglycemia), all the principles applied to hypoglycemia alarms / alarms can be performed for high glucose (hyperglycemia), as will be understood by those skilled in the art.

[0106] Figure 6 This is flowchart 500, which illustrates the process of dynamically and intelligently providing predictive alarms / alarms according to embodiments of this disclosure. As explained above, providing predictive alarms / alarms is highly desirable because it can minimize and / or prevent the number of hypoglycemic and / or hyperglycemic events experienced by the user.

[0107] In block 510, processor module 214 may be configured to receive sensor data (e.g., a data stream), including sensor data points from sensor 10 at one or more time intervals. In some embodiments, the sensor data points may be averaged, smoothed, and / or filtered using filters such as finite impulse response (FIR) or infinite impulse response (IIR) filters.

[0108] At block 520, processor module 214 can be configured to evaluate sensor data using a first set of indications or criteria. The first set of indications or criteria may include any algorithm suitable for determining whether a data point meets one or more predetermined criteria associated with hypoglycemia or hyperglycemia. Such predetermined criteria may be input by the user, for example, using a menu to input various alarm thresholds. Optionally, the predetermined criteria may be set according to factory settings and may be fixed so that the user cannot change, for example, the alarm thresholds. In some instances, the predetermined thresholds exist, for example, in a lookup table and may depend on other parameters, such as time of day (e.g., more sensitive or less sensitive at night), patient history, etc. In other embodiments, more complex algorithms may be used to define the user's current blood glucose status rather than static thresholds, such as static risk or dynamic risk models, where criteria are defined based on these complex algorithms or indications (e.g., gradients, yes / no indications, percentages, odds, etc.).

[0109] In some instances, sensor data includes real-time glucose values ​​(e.g., blood glucose (BG)), corrected glucose values ​​(e.g., estimated glucose (EGV)), glucose rate of change, direction of glucose change, acceleration or deceleration of glucose, insulin information, event information, historical trend analysis results, etc. Therefore, in some instances, processor module 214 can be configured to evaluate the sensor data using a first function to determine whether the real-time glucose value meets one or more predetermined criteria.

[0110] In some instances, the standard may be considered to include at least one component. For example, the standard may represent a single value or an absolute value. In some instances, the standard may include two or more components. For example, a threshold may represent a range of values. Optionally, the threshold may represent a single value associated with a time component. In other embodiments, the threshold may represent a single value associated with direction or directional rate.

[0111] As mentioned above, one or more criteria can be user-defined first thresholds. For example, a user might decide that they need to be alerted whenever their blood glucose reading drops to 70 mg / dL or 70 mg / dL at a negative rate of change (indicating a positive decrease in glucose levels). In other cases, the user might decide that 70 mg / dL may be too low a reading and that an alert should be issued whenever their glucose drops below 80 mg / dL. As those skilled in the art will understand, there is a good balance between excessively frequent alerts, such as annoying alarms, and sufficient alerts when there is a real event. Therefore, users can be allowed some input regarding how often they should be alerted based on their selection of a first threshold. The higher the selected value, the more sensitive the alarm will be to trigger; for example, the user will be alerted more frequently.

[0112] In some instances, the first threshold may be a user-configurable value or an indication that a qualitatively sensitive threshold has been exceeded. This qualitatively sensitive indication may include settings such as "sensitive," "normal," or "non-annoying," as described above. For example, the analyte monitoring system 8 may detect numerous alarms (e.g., >2 per day) and ask the user questions to determine whether settings should be adjusted to avoid unnecessary annoyance. For instance, if the number of alarms is higher than usual (e.g., twice the usual amount), the system 8 may suggest that the user change the qualitatively sensitive indication from sensitive to normal or from normal to non-annoying.

[0113] In some instances, the first set of indicators or criteria may include any algorithm suitable for determining whether a data point meets or exceeds a predetermined threshold. In some implementations, the first set of indicators or criteria may include a more complex assessment of blood glucose status, including, for example, other parameters such as the magnitude and / or direction of glucose change rate, the rate of acceleration / deceleration of glucose, insulin and / or dietary consumption. In some instances, the first set of indicators or criteria may employ risk measurements, such as static risk and / or dynamic risk models, for continuous glucose monitoring data to generate and determine whether a specific criterion has been met. In some instances, the algorithm performs its calculations on uncalibrated sensor data, then transforms the results into calibrated data and compares them with criteria and / or thresholds. Performing some or all of the algorithmic processing on uncalibrated data may be advantageous in reducing the adverse effects of calibration-related errors or biases; however, in some instances, the algorithmic processing may be performed on calibrated data.

[0114] In other implementations, more sophisticated algorithms can be used to define the user’s predicted blood glucose status rather than static thresholds, such as static or dynamic risk models, where the inputs may include real-time or predicted values, and where the criteria may be defined based on the outputs of these sophisticated algorithms (gradients, yes / no indicators, percentages, odds, etc.).

[0115] In block 530, processor module 214 can be configured to evaluate sensor data using a second set of indications or standards. In some instances, the first set of indications in block 520 differs from the second set of indications in block 530.

[0116] Similar to box 520, in box 530, the second set of indicators or criteria may include any one or more algorithms suitable for evaluating sensor data to determine whether a blood glucose state (high or low blood glucose) has been predicted, for example by determining whether a data point exceeds a predetermined threshold more than predicted or displayed. Suitable algorithms include those based on multinomial and autoregressive models, Kalman filter (KF) based algorithms, artificial neural networks, statistical and digital logic algorithms, and machine learning.

[0117] In some instances, the second set of indicators can utilize artificial neural networks, which, if available, can consider other relevant information such as food, exercise, stress, illness, or surgery, to predict future glucose levels. This structure can include three layers: the first layer collects the input, the hidden layers transform the input using polynomial or nonlinear functions such as squaring, sigmoid functions, or thresholding, and the third layer combines the outputs of the hidden layers into an output value or prediction. Neurons can be fully connected and feedforward. A useful artificial neural network implementation is described by W. A. ​​Sandham, D. Nikoletou, DJ Hamilton, K. Patterson, A. Japp, and C. MacGregor in "Blood glucose prediction for diabetes therapy using a current artificial neural network." IX European Signal Processing Conference (EUSIPCO), Rhodes, Described in 1998, pp. 673-676. Each neuron in the second and third layers takes weights from the output of the previous layer as input. These weights are tuned through a process called training to give the best prediction. That is, the neural network starts with an initial guess and uses training and test data sets with known inputs and outputs to find the best possible weights to give the best output. Once the training process is complete, the network can be used to predict the output of any new data. The network input information can be the current glucose measurement and its timestamp, along with a limited number of previous glucose samples from the CGM system. The NNM (Neural Network Model) can consider glucose measurements up to 20 minutes before the current time. Because the sampling rate differs between one CGM system and another, the number of NNM inputs can vary for each dataset. The network output can be a glucose prediction for a given time range.

[0118] In some instances, the second set of indicators can be used using autoregressive models (e.g., first-, second-, or third-order) to predict future glucose values. A useful first-order autoregressive model implementation is described by G. Sparacino, F. Zanderigo, S. Corazza, A. Maran, A. Facchinetti, and C. Cobelli in "Glucose Concentration can be Predicted. Ahead in Time From Continuous Glucose Monitoring Sensor Time-Series." Biomedical Engineering. IEEE Transactions 011. 2007, vol. 54, pp. Described in sections 931-937. The algorithm predicts future glucose values ​​by taking the current glucose value y(n) and multiplying the previous glucose value y(n-1) by a coefficient α. α will be a value slightly greater than 1 when glucose rises and a decimal less than 1 when glucose falls. In this algorithm, the model parameter (α) can be recursively estimated (e.g., updated every 5 minutes to account for glucose kinetics) to minimize the sum of the squared residuals of all pairs of predicted and current glucose values. Whenever a new sensor data point is received (e.g., every 5 minutes), the estimated value of α is updated (e.g., using weighted least squares regression). Prediction residuals (e.g., forgetting factor, prediction range, and prediction threshold) can be pre-adjusted to optimize alarm timing for the patient before a severe hyperglycemic event (e.g., 55 mg / dL) occurs and to minimize the number of annoying / false alarms heard by the patient. For example, since glucose orientation changes over time, a forgetting factor μ can be used to more heavily weight recent data with values ​​between 0 and 1. The prediction range and / or prediction threshold can be pre-determined by the system and / or the user, as described in more detail elsewhere in this document.

[0119] In some instances, the first-order autoregressive model includes a forgetting factor, prediction range, and prediction threshold, which are adjusted to provide additional alerts no more than once per week. This is based on retrospective analysis, which compares the first and second functions together with only the first function. In some instances, this is achieved at least in part by monitoring the user's qualitative sensitivity indicators and prompts or encouraging the user to select appropriate settings.

[0120] In some instances, the second set of indicators can be used with a Kalman filter (an optimized estimation method) to predict future glucose values. The Kalman filter balances the probability of a measured glucose change being caused by sensor noise with the actual glucose change to obtain the most probable estimate of glucose (and its first and second derivatives). A useful way to implement a Kalman filter is described by Palerm, C. and Bequette, W. in “Hypoglycemia Detection and Prediction Using Continuous Glucose Monitoring—A Study on Hypoglycemic ClampData,” J Diabetes Sci Technol. 2007 September; 1(5): 624–629. The state is the blood glucose concentration ( g k ), its rate of change ( d k For example, the rate of change of speed and the rate of change of velocity (e.g., velocity) f k (e.g., acceleration). The latter is assumed to vary randomly, influenced by input noise. w k (with covariance matrix) Q Driven by ) and describing the changes in the process. It is assumed that the sensor's glucose measurement contains noise, which is caused by υ k (with covariance matrix) R Description. Prediction model parameters (e.g., Q / R ratio, prediction range, and prediction threshold) can be extracted and adjusted, and / or user-selectable adjustments.

[0121] In some instances, the second set of instructions or functions may optionally include mechanisms to incorporate user input, such as insulin, exercise, diet, stress, disease, patient history information (e.g., patterns or trends), etc.

[0122] In some instances, sensor data includes real-time glucose values ​​(e.g., blood glucose levels (BGs)), corrected glucose values ​​(e.g., estimated glucose levels (EGVs)), the rate of change of glucose values, the direction of change of glucose values, acceleration or deceleration of glucose values, insulin information, event information, historical trend analysis results, etc. Therefore, in some instances, processor module 214 can be configured to evaluate the sensor data using a second function to determine whether the predicted glucose value meets one or more predictive alarm criteria.

[0123] The second predetermined criterion may include a single value, a range of values, a direction associated with the value, a rate of directional change, etc. In some instances, the second criterion includes a predetermined threshold, which is a fixed value set as part of the factory settings. For example, it may be desirable to have a second threshold that the user cannot manipulate or change due to its importance associated with the threshold. For example, in some instances, the second predetermined threshold may represent a value indicating a severe hypoglycemic event, such as 55 mg / dL.

[0124] In some instances, the second predetermined criterion is determined based on the probability of hypoglycemia within a specific time frame, which may take into account the rate of change of sensor data, the direction in which sensor data is traveling, the current glucose level, the past history of glucose changes, insulin information, dietary information, exercise information, etc. One example of applying additional criteria to processor module 214 is if a prediction of hypoglycemia in the near future is impossible or unpromising when the current glucose level is 200 mg / dL; therefore, restrictions can be imposed on glucose values, rates of change, etc. Another example of using dietary information is if the user indicates they have recently eaten, in which case glucose levels may change rapidly, and it is unnecessary to warn them at this time.

[0125] In some instances, the second predetermined criterion may be based at least in part or adapted to the first predetermined threshold. For example, a suitable set of algorithms or functions may be based on, for example, a prediction range and / or a second predetermined threshold based on the first predetermined threshold.

[0126] In some instances, the second set of instructions is entirely predictive, meaning that the instructions use past and / or current data to determine whether a user will meet or exceed a second predetermined threshold within a predetermined time frame or range. This predetermined frame or range is preferably long enough for the user to take action to avoid the predicted event. For example, the predetermined range may be at least 15 minutes in some instances, at least 20 minutes in others, and at least 30 minutes in some implementations.

[0127] In some instances, a predetermined time range or scope may have additional capacity, or additional information may be considered when setting the time range. For example, as is known to those skilled in the art, some continuous glucose monitoring systems sense glucose in tissue fluid rather than blood (e.g., capillary blood traditionally used to obtain finger measurements). Therefore, there may be a time interval between the measured value and the actual blood glucose value, such as 5 minutes or more. In some instances, the time interval may be as little as 0 minutes, while in others it may be as long as 15 minutes or more. The time interval can also be variable, which can be represented by the standard deviation in the measurement, for example, 10 minutes for a 5-minute interval.

[0128] As sensors become more accurate through improved sensor design, inaccuracies due to time intervals can become a more significant contributor to overall error. While not wishing to be bound by any particular theory, it is assumed that physiological function, the time since sensor insertion, glucose state (high or low), rate of glucose change, understanding of filtering practices, and other variables can influence the amount of time interval the sensor will experience. Therefore, in some instances, time interval adjustment algorithms or indicator sets can be used to determine or combine preset time ranges or time categories. In some instances, time interval adjustment indicator sets utilize near-time predictions to predict and display estimated glucose values ​​at future time points, such as from approximately 2.5 minutes to approximately 15 minutes, depending on additional information, such as that obtained through algorithms or alternative detection methods. Information that can influence the prediction range includes: the time since sensor insertion, glucose state or rate of change, what the average time interval for each individual is, and adaptive learning algorithms applying unique settings, etc. Therefore, in some instances, time interval adjustment algorithms can use one or more of the following variables as inputs: the time since sensor insertion, glucose state or rate of change, and prior user information. In some instances, additional sensors can be implemented to directly or indirectly measure the time interval. One example sensor can use impedance data between a spatially separated working electrode and a reference electrode, or impedance data between an electrode in the sensing area and a second electrode on the skin surface.

[0129] In some instances, boxes 520 and / or 530 may be iteratively repeated, where any new data is accepted, including data from sensors, another medical device (such as an insulin delivery device), and / or data from the user or owner (using user input) before the hypoglycemia indicator is activated.

[0130] In block 540, processor module 214 can be configured to activate a hypoglycemia indicator if a first criterion is met or if a second criterion is predicted to be met. In some instances, processor module 214 is configured to determine which set of indicators meets its criteria (e.g., 520 or 530, where x1 or x2 is indicated). For example, in block 520, the first set of indicators could determine that a first threshold was met at 8:47 pm, using real-time data. In the same instance, in block 530, the second set of indicators could determine that a second threshold is predicted to be met at 9:00 pm, using a 20-minute prediction range. Then, in this instance, if both the first and second assessments meet their criteria, either or both can be used in subsequent processing.

[0131] As described above, a hypoglycemia indicator may include a single indicator that indicates whether any function promptly determines or predicts that a threshold has been exceeded first. In some instances, a hypoglycemia indicator includes a flag with a specific set of indicators associated with it, depending on whether a first assessment using a first criterion (x1) or a second assessment using a second criterion (x2) is met. Responses to the first or second assessment may also be differentiated and / or output differently in the processing. In some instances, where both assessments show that the criterion is met, there is a unique processing and output indicating that both the actual criterion and the prediction-based criterion are met. Optionally, rules may be provided to determine which criterion is met.

[0132] For example, in some instances, the x1 and x2 designations can be distinguished on the user display or user interface 222. In some instances, the user interface 222 can be configured to display both simultaneous values ​​or two criteria (x1 and x2), such as “x1 at 76 mg / dL” and “x2 predicted to be 55 mg / dL within 20 minutes”.

[0133] In block 550, processor module 214 can be configured to provide an alarm or alert if it is ensured that a threshold is met or predicted to be met—as indicated by a hypoglycemia indicator. In some instances, processor module 214 provides an alarm or alert based on the earliest threshold that is met or predicted to be met. In some instances, providing an alarm based on the earliest checked or predicted threshold is desirable because this gives the user more time to avoid actual or predicted hypoglycemia and / or hyperglycemia events.

[0134] In addition to sending any of the aforementioned sensor data to the insulin delivery device, in some instances, processor module 214 also sends messages to the insulin delivery device, including at least one of the following: a) pausing insulin delivery (e.g., configured to pause basal or rapid glucose delivery in the insulin delivery device in response to sensor data that meets predetermined criteria), b) initiating a hypoglycemia and / or hyperglycemia minimization algorithm (e.g., configured to control the user's blood glucose to a target range using an automated insulin delivery device), c) controlling insulin delivery in response to it, or d) alarm-related information (e.g., a hypoglycemia indicator). Sensor data and / or messages may be sent directly to the dedicated insulin delivery device or indirectly via a controller, such as in a smartphone or via the cloud. Some implementations provide necessary sensor data for use in a hypoglycemia avoidance system to be sent to the insulin delivery device; for example, a hypoglycemia indicator may include information that can be used to determine when to pause or reduce basal insulin delivery and for how long (e.g., predicted glucose values ​​and predicted ranges, rate of glucose change, static risk, dynamic risk), or may send a practical indication (e.g., pausing basal input for 20 minutes).

[0135] In some instances, alarms and / or alarm settings can be manipulated by the user or one or more thresholds can be set (e.g., approximately 60-100 mg / dL). In some instances, the user can disable / enable the first and / or second functions. In some instances, a single message indicating an "early warning" alarm can be provided to the user. In some instances, particularly in closed-loop or semi-closed-loop systems, the settings are configured based on a control algorithm that can be programmed or downloaded by the user.

[0136] In some instances, the alarm may be selected from the group consisting of: sound, touch, vision, and / or data transmission (e.g., to a remote monitoring point) to subsequent sound, touch, and / or visual output to user interface 222. For example, if an alarm is activated for a child, data may be transmitted to the parent's smartphone (and the same or different alarm information may be provided). In such an instance, the child may receive an audible alarm while the parent receives a detailed glucose record. In some instances, an alarm is provided to the user if a first function determines that a first threshold is met first. In other implementations, an alarm is provided to the user if a second function predicts that a second threshold will be met first.

[0137] In some instances, predictive alarms may suppress threshold alarms to provide a more meaningful sense of urgency. For example, if a predictive alarm is associated with a reading of 55 mg / dL with a 20-minute prediction range, and a threshold alarm is associated with a reading of 70 mg / dL, then the predictive alarm indicator might be configured to control the alarm screen and output.

[0138] In some instances, predictive alarm sounds / screens convey a more meaningful sense of urgency than settable threshold alarms. The rationale for having different sounds / screens for alarms and alerts is to help users understand the difference between the two thresholds; for example, one might be a routine alarm while the other indicates a serious event. In some instances, an alarm is delivered to the user regardless of which alarm is met first or predicted to be met first. It should be understood that these predictive alarms and threshold alarms differ from targeted alarms, which can be used to "keep the user on two lines," as mentioned above. Figure 3B and 4A Description. In some instances, information from the results of the first and second indicator groups can be combined and delivered separately on the user interface, regardless of whether either group meets a certain criterion. For example, a combined alarm screen might include a predictive alarm (e.g., the victim is below an alarm threshold (TV2) and it is predicted that they will fall below a predicted threshold TV within 20 minutes). P The indication, or the victim's condition is below the alarm threshold (TV2) but is predicted not to fall below the predicted threshold TV within 20 minutes. P (Instructions). In some instances, additional information, such as treatment recommendations or requests for additional information, may be displayed.

[0139] Once a hypoglycemia alarm or alert has been triggered, processor module 214 can continue post-alarm monitoring, as described in detail below. Additionally or optionally, in systems that include remote monitoring (e.g., remote electronic devices receiving and tracking users, such as a patient's mobile phone or a caregiver's personal computer), once a hypoglycemia alarm is triggered and sent to a remote device, improvements to communication, data transmission, etc., can be initiated to allow caregivers or patients to monitor the patient more closely during hypoglycemia or a predicted hypoglycemic event. Improvements to communication or data transmission may include more frequent push-pull of sensor data from a remote device, additional input, or transmission to a remote device. Requests for additional information may originate from a remote device, and processor module 214 can be configured to receive and process requests for additional information directly from a remote device.

[0140] Monitoring after alarm Another common problem with continuous glucose monitors is that once an alarm or alert is triggered, the user can remain at or near the trigger threshold, and repeated or redundant alarms can continue to be triggered afterward. Such alarms can be annoying to the user, causing them to pause (e.g., briefly stop) the monitor's alarm / alarm feature. A brief pause based solely on time (e.g., not additional sensor data) may be undesirable, as there may be situations where additional alarms are requested after the initial trigger. If the user has paused the alarm / alarm feature, and the pause or "brief stop" feature is based solely on time, the user may not be aware that they are at risk of approaching a hypoglycemic and / or hyperglycemic event during the pause or brief stop period.

[0141] Now see Figure 7 This is an example of a continuous trace of glucose values ​​measured over a time range, showing different scenarios that could trigger repeated alarms / alarms. Similar to... Figure 5 ,exist Figure 7 In this example, a continuous glucose trace can be represented as a graph with glucose levels, for example, mg / dL, compared over a time range, such as 24 hours.

[0142] As shown, three thresholds or limits are used in glucose monitoring: TV1, TV2, and TV. P These have already been discussed above. Figure 5 The discussion is ongoing.

[0143] exist Figure 7 In the middle, at the first and second time points TA 1S and TA 1E It defines a region or interval. Time point TA 1S This is typically referred to as the time when the first alarm begins or is triggered. It is assumed that the user already knows the time point TA.1S The police were called shortly afterward, so the time point TA 1E This is typically referred to as the time it takes for the user's blood glucose status to change. As shown, in TA 1S Users can first provide an alarm when the threshold TV2 has been met. In TA 1S After the initial alarm, the user's glucose level usually hovers or oscillates around the blood glucose level that triggered the initial alarm. During this period, no further alarms should occur, but after this, a further alarm should occur.

[0144] In some traditional systems, users are warned again whenever their glucose levels exceed thresholds such as TV1 and TV2. This can be extremely annoying and may even affect the user's response to the initial alert. 1S Then turn off the sensor alarm. Furthermore, if the user uses a simple time-based pause to turn off or even “pause” one or more threshold alarms (e.g., at 80 mg / dl), the user may not be aware that at TA... 1E And then things got even worse.

[0145] See also Figure 7 It can be achieved through the first and / or second glucose threshold TV Z1 and TV Z2 Further define the region or range of glucose value oscillation. First glucose threshold (TV) Z1 This can typically be referred to as the upper limit of a region or interval of values ​​that oscillate around, for example, TV2. The second glucose threshold, TV... Z2 This can typically be referred to as the lower bound of a region or interval of values ​​oscillating around, for example, TV2. In some instances, TV... Z1 and TV Z2 Together, a buffer zone is provided, also known as the boundary zone above and below the threshold that would trigger the initial alarm (e.g., + / -5, 10, 15% or + / -5, 10, 15, 20 mg / dL), in which repeated alarms are rarely (if any) provided to the user. In other words, in TA... 1S After an initial alarm is provided to the user and the user confirms the alarm, the user may not be alerted again until TA. 1E Later, for example, when the glucose value leaves the buffer (e.g., a change in blood glucose status). In some instances, TV Z1 and TV Z2 It may have asymmetrical boundaries and / or no lower boundary (no TV). Z2 At least for reasons suggested by glucose trace scenarios, smarter and more dynamic decision-making can be beneficial so that users are not over-alarmed in certain situations (e.g., in TA). 1S and TA 1E(between), but in other situations it is fully alerted (e.g., in TA). 1E (Afterwards). In this paper, this buffer is also referred to as the edge region or Δ, which can be bidirectional and / or asymmetric, depending on the situation.

[0146] Figure 8 This is a flowchart illustrating, for example, the process 700 of dynamically and intelligently monitoring an individual's blood glucose status after an alarm state is activated based on sensor data that meets one or more activation transition criteria associated with a hypoglycemic or hyperglycemic state, as described in block 540. Furthermore, post-alarm monitoring can be applied to any alarm or alert, whether it evaluates multiple criteria as in blocks 520 and 530, or simply compares against a single threshold (e.g., a hypoglycemic or hyperglycemic threshold). In some instances, an alarm or alert is merely an identifier or label for that alarm or alert; for example, some hyperglycemic alarm conditions include a waiting time (e.g., 0 to 120 minutes), such as a user-configurable / enabled "Enable wait before first alarm" option. When this input is enabled, the waiting time is applied before the first warning to the user, as described in more detail elsewhere herein, unless active monitoring determines a worse condition.

[0147] In some instances, triggering an alarm or alert involves providing an alarm or alert to the user. Typically, the processor module may be configured to provide output associated with the activation of the alarm state, where the output indicates a hypoglycemic or hyperglycemic condition. In some instances, post-alarm monitoring will begin regardless of whether the user has acknowledged the alarm, where the acknowledgment changes the post-alarm state, as described elsewhere herein. In some instances, post-alarm monitoring begins with a transition from an active state to an acknowledged state.

[0148] The criteria for triggering an alarm, also known as the transition from an inactive state to an active state or the activation criterion, can be any criterion or threshold related to hypoglycemia or hyperglycemia. Activation criteria may include glucose levels, predicted glucose levels, rates of glucose change (direction and / or magnitude), rates of glucose acceleration (direction and / or magnitude), static risk models, dynamic risk models, etc. Additionally or optionally, the activation criteria disclosed for the transition from inactive state 1030 to active state 1055 can initiate dynamic and intelligent monitoring of the subject's blood glucose status after activating an alarm state related to hypoglycemia or hyperglycemia.

[0149] In block 710, processor module 214 can be configured to trigger an alarm / alarm to a user (e.g., in...). Figure 6The system actively monitors data related to the subject's hypoglycemia or hyperglycemia for a period of time after confirmation by the user (as described in the relevant text) and / or the user. In some instances, the user or subject has already confirmed the initial alarm. As described in more detail elsewhere in this document, confirmation alarms may include user interaction with the system (“user action”), such as button or menu selection, or other data input (e.g., input of dietary or insulin information). Additionally or optionally, confirmation alarms may be based on monitoring data related to blood glucose status, including sensor data, insulin data, and dietary data. In some instances, the confirmation may originate from a remote monitor, such as the patient's mobile phone. Confirmation alarms can transition processor module 214 from an active state to a confirmed state, during which active monitoring can be performed.

[0150] In some instances, processor module 214 actively monitors data related to the subject's hypoglycemia or hyperglycemia for a period of time in an acknowledged state, or continues monitoring in the acknowledged state until the state transitions to inactivity (e.g., based on inactivity criteria). In some instances, processor module 214 monitors sensor data or other data received after an alarm is activated. For example, the data may include information such as: sensor data (e.g., glucose levels, trends, distance between peak and trough values ​​indicating meal times, etc.), sensor diagnostic information (noise indicators), dietary information (e.g., calorie intake and intake time), insulin information, or other event information. In some instances, actively monitored data includes determining the average glucose level within a time window, the magnitude and / or direction of the rate of glucose change, or the magnitude and / or direction of the rate of glucose acceleration.

[0151] In some instances, processor module 214 tracks how quickly and / or frequently user-acknowledged alarms are triggered to determine further action. For example, patterns in the timing of user acknowledgments, information about the type of alarm triggered, and / or the eventual recovery of the situation can be evaluated, and future acknowledgments or alarms can be improved upon based on this. Typically, however, processor module 214 may process alarms based on the assumption that user-acknowledged alarms / alarms indicate that the user is aware of the situation. Therefore, re-alarm conditions can differ, and may be more stringent in some instances, as described herein.

[0152] In some instances, sensor data is actively monitored during the confirmation period, also known as the active monitoring period. Suitable time periods include, for example, 20 minutes, 40 minutes, 60 minutes, etc., and can be user-configurable. This time period can begin or commence at the first data point after an alarm or alert triggering data point. In some instances, the user is allowed to select this time period. In some instances, sensor data monitoring continues after an alarm until the alarm is stopped, as described in more detail elsewhere in this document.

[0153] In block 720, processor module 214 is configured to transition from an acknowledged state to at least one inactive or active state in response to data related to a subject's hypoglycemia or hyperglycemia condition that meets one or more predetermined criteria. Processor module 214 may determine whether a state change has occurred based on sensor data or other data related to the subject's blood glucose status. In some instances, a state change may represent a change in the subject's blood glucose status.

[0154] In some instances, a state change can be a positive event, such as an indication that the user has safely avoided an anticipated hypoglycemic event, which can trigger a confirmation state to recognize the recovered sub-state and / or inactive state, as described in more detail elsewhere in this document. In other implementations, a state change can be a negative event, such as an indication that the user is further declining toward a hypoglycemic event during a specific time period (e.g., 20 minutes) (e.g., acceleration / deceleration analysis of sensor data), confirming or not confirming a recovered sub-state. In a second instance of a negative event discussed here, even when the user has confirmed the alarm, it may be desirable to reactivate the alarm. This second alarm or alert may be valuable in alerting the user that his condition is deteriorating—what actions he has taken after the first alarm (if any) are insufficient to avoid or improve his glucose reading. In some instances, a state change indicating such a negative event can bypass the confirmation state (described elsewhere in this document) and cause the system to transition back to an active alarm state (also known as a re-alarm or reactivation). However, if the system is already in an active alarm state (e.g., without user confirmation), the resulting output can be progressively increased, such as becoming louder and more frequent, and / or may lead to contact with 911, caregivers, etc.

[0155] In some instances, processor module 214 can be configured to determine if no change has occurred in the last x minutes, such as 15, 30, 45, or 60 minutes, as described in the description of conditions associated with remaining in a specific state (e.g., an active state) and / or reactivation conditions, as described in more detail elsewhere in this document. For example, in cases where the data point hovers around a threshold that triggers an initial alarm (e.g., for a low rate of change), there may be no change. It should be understood that in such cases, an alarm or alert would be triggered in a conventional system whenever the data point exceeds the threshold in an undesirable direction, such as below a threshold for a hypoglycemic event or above a threshold for a hyperglycemic event. This is generally considered annoying and may be the reason why users take action to turn off their alarms or become desensitized to alarms. In these cases, intelligent post-alarm monitoring algorithms can avoid these annoying alarms (e.g., as long as the user remains in the same state and / or the reactivation conditions have not been met, as described in more detail elsewhere in this document).

[0156] As used herein, a user can be in different monitoring "states," allowing the processor module 214 to detect when the user transitions from one state to another. Examples of such states include, for example, "active," "inactive," and "confirmed."

[0157] An active state is defined here as an alarm state triggered by conditions where data is monitored to determine whether the alarm state should be changed to "acknowledged" or "inactive". An active state can be transitioned from an inactive and / or acknowledged state by comparing data against different criteria (sensor, glucose, insulin, user-provided information, etc.) for each state. Re-alarms and / or reactivation of alarms can occur in this state, progressively increasing the initial alarm level, even before acknowledgment.

[0158] An inactive state is defined here as an alarm state triggered by conditions where data assessment indicates that the blood glucose level is in the safe or target zone. An inactive state can be transitioned from an active and / or confirmed state by comparing data against different criteria (sensor, insulin, user-provided information, etc.) for each state.

[0159] A confirmed state is defined herein as an alarm state triggered by conditions in which the user has acknowledged the alarm and / or alert and no additional alarms / alarms are provided during a predetermined period unless certain reactivation criteria are met (where the reactivation criteria differ from and / or are more stringent than the initial activation criteria, e.g., as described in more detail elsewhere in this document). A confirmed state may be entered from an active state based on user interaction and / or indications of recovery from a high or low blood glucose condition. It should be understood that the terminology used herein for states and / or functions is merely descriptive and may be used with other names, provided the functionality remains substantially the same.

[0160] In some instances, users can confirm the alarm by pressing a button, selecting a menu screen, etc. In other instances, when the system enters the confirmation state, a timer is set for a predetermined period of time, after which the system can automatically transition to active and / or inactive based on the selected status.

[0161] In some instances, the confirmed state assessment is based on data related to the individual's blood glucose status (including glucose, dietary, and / or insulin information) and can be transitioned to a confirmed state based on an assessment instructing the individual to recover from the blood glucose status that triggered the alarm. The individual's recovery can also be considered a confirmed sub-state, and can be determined separately for each state transition by comparing data against different criteria (sensor, insulin, user-provided information, etc.). During the confirmed state and / or a confirmed and recovered sub-state, alarms can be paused for a certain period.

[0162] In some instances, the alarm remains active if the user does not acknowledge it. In other instances, if the user does not acknowledge the alarm after X minutes (e.g., 5, 10, 15 minutes), the alarm can be progressively escalated by outputting to a secondary display device, a remote monitor, or emergency contact. Similarly, if the situation worsens during active monitoring (e.g., in an acknowledged state), the alarm can be progressively escalated by outputting to a secondary display device, a remote monitor, or emergency contact.

[0163] Below Figure 11 The discussion provides examples of transitions from one state to another. It should be understood that using different states to describe a user's or incident's blood glucose status can be used to track the user's condition after an alarm or alert has been triggered. Therefore, certain trends, such as an improvement in the user's blood glucose level or simply hovering near a minor alarm, may be sufficient grounds to discontinue alerting the user. However, other trends, such as a rapid decline in the user's blood glucose level or even hovering near a severe alarm, may be grounds to continue warning or alerting the user. Many appropriate responses (e.g., related to reactivating proactive alarms) may exist, for example, in lookup tables, depending on the state transition and other known information.

[0164] Now see Figure 11 A state diagram 1000 is provided, showing the transitions between the following states: Activation (A) 1010, Determined (K) 1020, and Inactive (I) 1030. These states and their transitions can be described as follows: After the alarm is activated, the processor module transitions from an inactive state (1030) (1055) to an active state (1010). Activation conditions include various criteria or thresholds, hereinafter referred to as "activation criteria" or "activation conditions," which can be used to detect actual or future hyperglycemia or hypoglycemia, as described in more detail elsewhere in this document (e.g., Figure 6 The alarm conditions are met, as indicated by reference numeral 1055 in the attached figure. In addition to sensor data (glucose value, derivative (rate of change), and its second derivative (acceleration)), alarm conditions may involve other data, including insulin data and / or user input (dietary information, exercise information, etc.). Furthermore, the implementation methods discussed herein regarding reactivation can be applied here (e.g., analysis of static and / or dynamic risks). However, regardless of the method used to detect the activation of hypoglycemia or hyperglycemia, the disclosed active monitoring and state transition methods can be applied.

[0165] In some instances, certain activation conditions based on hyperglycemic status may be applied to the transition from an inactive to an activated state: glucose levels exceeding a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); average glucose levels over a predetermined time period (e.g., 10, 20, 30, 40 minutes) exceeding a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); current glucose levels exceeding a predetermined threshold plus a predetermined boundary (e.g., 25, 50, 75 mg / dL); and / or a rate of glucose change exceeding a predetermined threshold (e.g., -0.5 mg / dL).

[0166] In some instances, based on a hypoglycemic condition, the following activation conditions may be applied to the transition from an inactive to an activated state: glucose levels are below a predetermined threshold (e.g., 80, 60, 60 mg / dL); glucose levels are below a predetermined threshold and the rate of change is less than a predetermined rate (1.0 mg / dL / min, e.g., no rapid rise); or glucose levels are expected to fall below a second threshold (e.g., 55 mg / dL) within a predicted range (e.g., 10, 15, 20 minutes).

[0167] In some instances, the transition from the active state (1010) to the confirmed state (1080) is based on data or confirmation criteria indicating that the subject's glucose is trending toward normal blood sugar levels. The data may include sensor data indicating changes in glucose trends and / or insulin information related to the correction of the condition.

[0168] For example, once an alarm indicating high or low blood glucose is activated, sensor data indicating changes in glucose trends may include assessments of sensor data whereby the direction and / or magnitude of glucose levels, the rate of glucose change, or acceleration / deceleration (e.g., change in direction or trend) of glucose levels indicating a trend back to normal blood glucose levels is sufficient to automatically transition from activation to a confirmed state (with or without user interaction using a user interface). A trend back to normal blood glucose levels may also indicate or trigger a “recovery” substate, which is a state within the confirmed state indicating a trend back to normal blood glucose levels. In one instance, following the activation transition associated with a hyperglycemic condition, the state transition to the confirmed but recovering substate may be based on confirmation criteria or conditions, such as a glucose level (or the average glucose level over a predetermined time period) being less than a predetermined threshold minus a Δ (e.g., 10, 15, 20 mg / dL), and in some instances, a rate of change trending towards normal blood glucose levels (e.g., a decrease faster than approximately 1 mg / dL / min). Similarly, following the activation transition associated with a hypoglycemic state, the state transition to a confirmed but regressive substate can be based on confirmation criteria or conditions, such as glucose levels (or average glucose levels over a predetermined period) approximately a predetermined threshold plus a Δ (e.g., 10, 15, 20 mg / dL), and in some instances, a rate of change trending toward conditions of normal glucose levels (e.g., rising faster than approximately 1 mg / dL / min).

[0169] In some instances, the transition from the active state (1010) to the confirmed state (1075) is based on confirmation criteria instructing the user to confirm the alarm on the user interface, user-entered insulin information, and / or user-entered dietary information. For example, the user may tap a touchscreen “button” to confirm the alarm. Additionally or optionally, when the continuous glucose sensor is operatively connected to an insulin delivery device (including an associated remote programmer), changes related to the basal or rapid delivery curve or volume can be considered user input, particularly when such changes are caused by user interaction. Similarly, when the continuous glucose sensor is operatively connected to (or integrated with) an electronic device capable of receiving dietary information (e.g., carbohydrates and consumption time), such dietary information can be considered user input. In some instances, user input may originate from another electronic device (e.g., via remote monitoring from the patient’s smartphone).

[0170] In some instances, the transition from the confirmed state (1020) to the inactive state (1085) is based on sensor data that no longer meets one or more activation transition criteria associated with hypoglycemia or hyperglycemia, and / or on one or more inactive transition criteria (e.g., which may differ from one or more activation transition criteria associated with hypoglycemia or hyperglycemia (e.g., associated with an initial alarm)). Additionally or optionally, the transition from the confirmed state (1080) to the inactive state may be based on insulin data and / or dietary information. In some instances, the transition (1090) includes a time factor, for example, after a confirmed (active monitoring) period that can be fixed and / or user-configurable.

[0171] The deactivation criteria used to move from confirmation to deactivation can be similar to or the same as the deactivation-to-activation criteria (except when one or more of the first or second criteria (see, for example)). Figure 6 (This is not met and / or used for transitioning from activation to inactivation to a confirmed state). In some instances, based on hyperglycemic status, certain of the following inactivation criteria or conditions may be applied to the transition from a confirmed to an inactivation state: the average glucose level during a predetermined time period (e.g., 10, 20, 30, 40 minutes) is less than a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); a predetermined Δ is subtracted (e.g., 10, 15, 20 mg / dL); the confirmed time period has elapsed and the glucose level is less than the predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); and / or the rate of glucose change decreases at a predetermined rate.

[0172] Optionally, the inactivation criteria for transitioning from confirmed to inactive states can differ (e.g., be more stringent). In addition to sensor data, other data, including insulin data and / or user input, can be considered for state transition criteria. In some instances, based on a hypoglycemic condition, the following inactivation criteria or conditions may be applied to the transition from confirmed to inactive states, which may include determining whether glucose levels have increased more than a predetermined amount (e.g., 10, 15, 20 mg / dL, which may indicate some user action) and whether glucose levels have risen above a threshold condition (e.g., from a first function). The inactivation conditions for real-time alarms (first function) and predictive alarms (second function) can be the same or different. In some instances, a positive rate of glucose change exceeding a certain value (the increasing rate of rise) can be used as a condition, as this rate of rise can be considered to indicate that the user should take some preventative action. In some instances, the rate of rise condition can be combined with a glucose threshold condition. In some instances, the glucose rate of rise condition is 0.25, 0.5, or 1 mg / dL / min.

[0173] In some instances, where a hypoglycemic condition has been activated and actively monitored after user confirmation, the status can transition from the confirmed state to the inactive state before the confirmation period expires. This is based on inactive criteria or data indicating a significant improvement in the hypoglycemic condition, such as determining that the subject's glucose level is greater than a lower threshold plus a Δ and the predicted glucose level of the predicted range is greater than a predetermined limit (which may be the same as or different from the initially activated predicted range or threshold).

[0174] In some instances, the transition from a confirmed state (1020) to an activated state (1060) is based on the fulfillment of one or more activation criteria associated with a hypoglycemic or hyperglycemic condition and an expiration date based on a predetermined time period. The activation criteria for transitioning from confirmed to activated can be any of the same conditions as the initial alarm or can be different (e.g., based on a glucose trend moving away from normal blood glucose). In addition to sensor data, other data, including insulin data and / or user input, can be considered for state transition criteria. Furthermore, the implementation methods discussed herein regarding reactivation are applicable here.

[0175] In some instances, the reactivation criteria for transitioning from a confirmed state (1020) to an active state (1065) may differ from the initial criteria for transitioning to the active state. For example, even when the user has already confirmed the alarm (during active monitoring and / or confirmation time), a deteriorating condition may indicate the need for a re-alarm based on one or more criteria indicating a worsening condition, such as sensor data indicating a deteriorating blood glucose condition, for example, a second function that meets one or more criteria, and / or sensor data indicating a further trend of glucose away from normal blood glucose levels.

[0176] In some instances, based on a hypoglycemic condition where the confirmation time has expired, the following reactivation criteria or conditions may be applied to the transition from confirmation to activation: glucose levels are below a predetermined threshold (e.g., 80, 60, 60 mg / dL); glucose levels are below a predetermined threshold and the rate of change is less than a predetermined rate (e.g., 1.0 mg / dL / min, e.g., no rapid rise); or glucose levels are predicted to fall below a second threshold (e.g., 55 mg / dL) within a predicted range (e.g., 10, 15, 20 minutes). However, in some instances, reactivation criteria indicating a worsening hypoglycemic condition may cause a transition from confirmation to activation (reactivation or re-alarm) before the confirmation time expires, based on one or more re-alarm criteria (e.g., criteria different from the initial alarm), some of which are described in more detail elsewhere in this document. In some instances, if a second function meets a second criterion (at 530), an alarm activated by a first function (at 520) may trigger a re-alarm or reactivation (transition to activation before the time expires) during the confirmation time.

[0177] In some instances, activation criteria related to hyperglycemic status, transitioning from confirmation to activation, may include determining whether glucose levels exceed predetermined thresholds (e.g., 160, 180, 200, 220 mg / dL); whether the average glucose level over a predetermined time period (e.g., 10, 20, 30, 40 minutes) is greater than predetermined thresholds (e.g., 160, 180, 200, 220 mg / dL); whether the current glucose level is approximately a predetermined threshold plus a predetermined boundary (e.g., 25, 50, 75 mg / dL); and / or whether the rate of glucose change is greater than a predetermined threshold (e.g., -0.5 mg / dL per minute).

[0178] In some instances, the transition from a confirmed state (1020) to an active state (1065), also known as reactivation, occurs after confirmed data indicates that the subject's glucose is trending toward normal blood glucose—also known as a recovering sub-state—and subsequently trending away from normal blood glucose during active monitoring (based on one or more criteria). In other words, a rebound can occur after an alarm is activated when a user's blood glucose status initially trended toward normal blood glucose (recovery) but subsequently trended back to high or low blood glucose. Therefore, reactivation of the first alarm state during the confirmed period can be made using data related to the subject's hypoglycemia or hyperglycemia status that meets one or more criteria for a rebound (e.g., reactivation). Typically, the criteria for transitioning from recovery to activation can be any of the same conditions as the initial alarm or can be different (e.g., glucose value, rate of change, or an accelerating counter-trend). In some instances, the one or more rebound criteria include conditions indicating that the subject's glucose is trending toward normal blood glucose (e.g., a recovering sub-state) and subsequent conditions indicating that the subject's glucose trend is moving away from normal blood glucose during active monitoring. In addition to sensor data, other data, including insulin data and / or user input, can be considered for state transition criteria. Furthermore, the implementation methods discussed herein regarding reactivation can be applied here.

[0179] In some instances, the transition from the active state (1010) to the inactive state (1070) can be any of the same conditions as the initial alarm (except when ( Figure 6 (The state enters an inactive state when one or more of the first or second criteria are not met.) Optionally, the inactivation criteria for transitioning from activation to inactivation can be different (e.g., different thresholds). In addition to sensor data, other data, including insulin data and / or user input, can be considered for state transition criteria. In one instance, the transition from activation to inactivation is based on alarm conditions that are no longer met, such as EGV and ΔEGV, as described in more detail elsewhere in this document.

[0180] In some instances, based on a hyperglycemic state, certain inactivation criteria or conditions may be applied to the transition from an activated to an inactivated state: the mean glucose level during a predetermined time period (e.g., 10, 15, 30, 45 minutes) is less than a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL) minus a predetermined Δ (e.g., 10, 15, 20 mg / dL); the glucose level during a predetermined time period (e.g., 10, 15, 30, 45 minutes) is below a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL) minus a predetermined Δ (e.g., 10, 15, 20 mg / dL); the mean glucose level during the predetermined time period is less than a predetermined threshold; and / or the rate of glucose change decreases at a rate greater than a predetermined rate (e.g., 1 mg / dL / min) (negative).

[0181] In some instances, based on a hyperglycemic state, certain of the following inactivation criteria or conditions may be applied to the transition from an activated to an inactivated state: the mean glucose level during a predetermined time period (e.g., 10, 15, 30, 45 minutes) is greater than a predetermined threshold (e.g., 60, 70, 80 mg / dL) plus a predetermined Δ (e.g., 10, 15, 20 mg / dL); the glucose level during a predetermined time period (e.g., 10, 15, 30, 45 minutes) is greater than a predetermined threshold (e.g., 60, 70, 80 mg / dL) plus a predetermined Δ (e.g., 10, 15, 20 mg / dL); the mean glucose level during the predetermined time period is greater than a predetermined threshold; and / or / or the rate of glucose change increases (positively) at a rate greater than a predetermined rate (e.g., 1 mg / dL / min).

[0182] For example, in some instances, the criteria for transitioning from a confirmed state based on user confirmation data (e.g., pressing a button) to an inactive or active state (e.g., reactivation) differ from the criteria for transitioning from a confirmed state based on detected user action (e.g., recovery detected by monitoring data). This confirmation based on detected user action is discussed in more detail below in the discussion of confirming sub-state recovery. For instance, to transition from a confirmed state based on user confirmation data to an inactive state, a first criterion can be applied (e.g., a criterion that allows oscillations near a threshold without inactivation, or in other words, a criterion that ensures the user does not merely oscillate near a threshold). In contrast, to transition from a confirmed state based on detected user action (a recovered sub-state) to an inactive state, a second criterion can be applied (e.g., this second criterion does not consider oscillations but rather considers data confirming a successful recovery to normal blood glucose levels).

[0183] As mentioned above, the conditions or criteria for reactivation can be more stringent than the initial activation. Similarly, the reactivation criteria for transitioning from a confirmed state based on user confirmation data (e.g., pressing a button) to an activated state can differ from those for transitioning from a confirmed state based on detected user action (e.g., detection of recovery via monitoring data). In some instances, the reactivation criteria for transitioning from confirmation (based on user confirmation) to activation can be time-based or may include a stringent second set of criteria (e.g., a change greater than 200 mg / dL). In contrast, the reactivation criteria for transitioning from confirmation (based on detected user action, recovery sub-state) to activation may include monitoring based on sensor data indicating a counter-trend or deterioration of the rebound condition. Exemplary deterioration may include or may be based on a sharp change in glucose away from the target, and may be based on glucose level (g), the amount of glucose change (Δg), the rate of glucose change (Δg / t), acceleration, or a combination thereof.

[0184] See back Figure 8 In block 730, processor module 214 may be configured to process responses to alarm state changes or transitions. In some instances, processor module 214 may include indications or criteria on how to handle an appropriate response to a state change. In some instances, such responses may be stored in, for example, a lookup table. In some instances, the output associated with a transition to an active state differs from the output associated with a transition from an acknowledged state to an active state and / or differs from the output associated with a transition from an acknowledged state to an inactive state.

[0185] It should be understood that, depending on the state change, one or more possible and / or appropriate responses may exist. For example, if sensor data indicates that the user has stabilized their glucose levels and is sufficiently recovering from a hyperglycemic event (e.g., an active-to-inactive transition), a specific type of alarm indicating a positive state change or a kudos may be provided. Kudos may have specific sounds, such as specific pitches or harmonics. In some instances, users may be able to customize or select alarm flags, similar to those available to smartphone users today.

[0186] In other implementations, a specific type of alarm may be provided to indicate a negative state change or a warning (e.g., a transition back to an active state). This warning may also have a specific sound, and in some cases, may be customizable by the user. In some instances, the severity of the warning may be reflected in the sound of the warning itself; for example, the warning may be louder or more intense, or it may sound like a recognizable sound of grief. In some instances, a re-alarm or subsequent alarm may differ from the initial alarm.

[0187] The state transition is described in more detail elsewhere in this document; however, it should be understood that, based on pre-set or user-selectable options, various indicators may be provided in an auditory, gustatory, or visual manner, and / or communication may be possible via data transmission.

[0188] In block 740, processor module 214 is optionally configured to output output information related to state transitions, such as providing an alarm, alert, or commendation if a state or state change is confirmed to be detected. In some instances, the output associated with transitioning to an active state differs from the output associated with transitioning from an acknowledged state to an inactive state and / or an active (reactivation) state. In some instances, alarms, alerts, or commendations may be time-limited, for example, if the user has acknowledged the alarm. This can prevent the user from receiving alarms / alerts / commendations in very rare cases, such as when the alarm characteristic is a reactivation condition, as see [reference]. Figure 9 More detailed description.

[0189] There are use cases where immediate notification of an alarm may not be advantageous. For example, postprandial blood glucose deviations are normal for people with diabetes, even those injecting insulin, or for people without diabetes. In some cases, alerting a user about a blood glucose deviation they are already aware of can lead to disappointment and desensitization to the alarm. Waiting and determining whether the subject's blood glucose deviation is recovering normally before an alarm may be preferred. Therefore, in some instances, the processor module may be configured to provide output related to a first alarm state after a predetermined waiting period, wherein the output is based on data related to a subject's hyperglycemic state meeting one or more second criteria after the predetermined waiting period. Thus, it will be understood that the processor module 214 may be configured not to provide output related to a first alarm state after the waiting period based on data related to a subject's hyperglycemic state not meeting one or more second criteria, thereby allowing the state to transition to an active state and back to an inactive state without an alarm and / or otherwise providing output to the user. In these embodiments, the one or more first criteria and the one or more second criteria may be the same or different, and the waiting period may be fixed or user-selectable. For example, one or more second criteria may include determining whether: the glucose level exceeds a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); whether the average glucose level over a predetermined time period (e.g., 10, 20, 30, 40 minutes) is greater than a predetermined threshold (e.g., 160, 180, 200, 220 mg / dL); whether the current glucose level is greater than a predetermined threshold plus a predetermined boundary (e.g., 25, 50, 75 mg / dL); the estimated time to the threshold (e.g., based on the rate of glucose change) and / or whether the rate of glucose change is greater than a predetermined threshold (e.g., -0.5 mg / dL per minute).

[0190] Figure 9 This is flowchart 800, which illustrates an example process for determining when to re-alarm and / or determine alarm activation after a state transition from acknowledgement to activation, also known as reactivation conditions, according to embodiments of this disclosure. In some instances, to avoid unnecessary "flickering" re-alarms, compared to the initial triggering conditions for activating the alarm state, see, for example, [reference needed]. Figure 6 The described reactivation conditions have different criteria for entering the activated alarm state (providing a re-alarm output).

[0191] In block 810, in one example implementation, processor module 214 may be configured to determine whether one or more sensor data points are within a predetermined or predefined interval during an alarm triggering period following a predetermined time period. Such a predetermined or predefined interval may include a region above and / or below a threshold that triggers the initial alarm provided to the user. For example, see back. Figure 7 TV Z1 and TV Z2 A buffer or envelope range above and below the threshold can be provided. In some instances, the predefined range is a range that defines the area around the value that triggers the first alarm. For example, the predefined range could be 10 mg / dL above and / or 10 mg / dL below the value that triggers the first alarm.

[0192] In some instances, data points within a predefined interval indicate no state change. In such implementations, no further action is taken. In implementations utilizing state transitions, it is assumed that the user has acknowledged the alarm / alarm (or based on data analysis), the acknowledged alarm state remains, and no transition to inactive or active status occurs. In contrast, if the sensor data point moves outside the interval, the state remains acknowledged and can be actively monitored (e.g., monitoring recovery or deterioration based on a criterion indicating that the user's glucose level is trending towards or away from normal blood sugar), or based on a second (more stringent than the first) or one more criterion, if the sensor data moves towards normal blood sugar, an inactive transition is detected, or an active transition is detected (e.g., based on one or more additional reactivation criteria, if the sensor data moves away from normal blood sugar, reactivation is detected). For example, if the user transitions from an acknowledged state to an active state, a new alarm can be triggered.

[0193] In block 820, processor module 214 may also optionally be configured to determine whether the user has acknowledged the alarm, as described in more detail elsewhere herein, thereby not providing an alarm or alert to the user for a set time period, such as 30 minutes. In some instances, processor module 214 may be configured to determine whether the user has taken some type of action independently of the sensor during a predetermined time period following the alarm. For example, the user may have eaten or increased insulin since the initial alarm, which could result in a significant change in glucose levels. The user may or may not output this type of information to the sensor. However, certain types of changes, such as insulin updates, may have identifiable patterns of user action that processor module 214 can determine as changes in glucose levels, such as changes in glucose levels, direction, rate of change, or acceleration / deceleration.

[0194] In block 830, processor module 214 can be configured to determine whether a reactivation condition has been met via one or more sensor data points. As used herein, a "reactivation condition" refers to a condition that causes an alarm or alert to be provided to a user after an initial alarm, during a "confirmation period" (e.g., a transition from a confirmed state to an active state) in which the user does not expect to receive additional unnecessary alarms. Generally, a reactivation condition is less likely to involve an alarm related to an event considered dangerous to the user's health, such as a severe hypoglycemic event. In some instances, a reactivation condition may be considered a condition or event that causes the subject or user to change from a confirmed state to an active state, as per [reference to...]. Figure 11 Discussed.

[0195] Furthermore, because reactivation can be unpredictable or a response to attempts to suppress it, the output associated with reactivation can be more explicit or different from other alerts, for example, by displaying illustrative information about the actions taken or the problem and / or by enhancing the alert, as those skilled in the art will understand.

[0196] Figure 10 This is flowchart 900, which illustrates an example process for determining whether reactivation conditions are met according to an exemplary embodiment of this disclosure. In block 910, processor module 214 can be configured to determine whether one or more sensor data points are outside a predetermined or predefined range. For example, as... Figure 7 As shown, there may be areas or intervals where data points appear to oscillate after an alarm has been provided to the user. Once it is determined that a data point falls outside this predefined interval, it may indicate that the user's glucose level is shifting toward an unwanted value or following an unwanted trend (e.g., a trend away from normal blood sugar levels).

[0197] In block 920, processor module 214 may be configured to determine the direction and / or rate of data movement or tendency, depending on whether the data point has a high or low value, for example, whether it is within a high or low threshold range indicating a hyperglycemic event or a hypoglycemic event, respectively. Furthermore, in some instances, processor module 214 may be further configured to consider user input and / or insulin information as variables for assessing whether reactivation conditions are met.

[0198] As can be understood, having reactivation conditions can also be considered a smart acknowledgment condition because it limits the number of alarms that can be provided to the user after the first alarm. This can help ensure that the user is not bothered by a series of additional alarms issued after the first alarm, but allows for a smarter and safer approach than a simple time-based pause. In some instances, the data points that can be considered as reactivation conditions can be predetermined or fixed through factory settings. Such reactivation conditions can be stored, for example, in a lookup table.

[0199] In some instances, reactivation aims to alert the user again when their actions are insufficient or when they return to a dangerous range. For example, after a low-threshold alarm, the user might eat something, which initially increases their glucose levels, but not enough to cause them to drop again. In this case, a reactivation might be desirable, but it should be distinguished from an oscillating / annoying alarm. This can be achieved, for example, by setting a standard where glucose rises a certain distance above the low threshold and then falls below it, or by increasing the rate of change to, for example, above 1 mg / dL / min before turning negative.

[0200] In some instances, after the initial alarm is sent to the user, the user can pause additional alarms by confirming the initial alarm. For example, when the user or the person involved confirms the initial alarm, additional alarms are paused for a period of time, except for re-alarm conditions. In other words, after a threshold alarm or predictive alarm, the user will not immediately hear another threshold alarm or predictive alarm, but will wait for a period of time, such as 30 minutes. In some instances, the pause time is user-configurable. In some instances, the user can have a default value of 30 minutes and a maximum value of, for example, 2 hours for safety.

[0201] In some instances, a set of indicators or algorithms can be used after an alarm to detect the transition from a hypoglycemic and / or hyperglycemic event to a recovery sub-state, so that a re-alarm only occurs if another event occurs (e.g., an increase of 5 mg / dL above the threshold, an upward slope greater than 1 mg / dL / min, etc.). In such an implementation, this may help prevent disturbing situations of stable glucose oscillating around 80 mg / dL and prevent repeated threshold alarms, as described above.

[0202] In some instances, an alarm may be triggered if the user remains in the same condition, for example, if their blood glucose level is below 70 mg / dL for an extended period, because prolonged hypoglycemia can be considered a health risk.

[0203] Example Some examples are provided below. When referring to a “low” threshold, it should be understood that the low threshold generally refers to a low glucose value that indicates a hypoglycemic event. Similarly, when referring to a “high” threshold, it should be understood that the high threshold generally refers to a high glucose value that indicates a hyperglycemic event. As used herein, the glucose value can be an estimated glucose value (EGV) or any known type of glucose indicator or sensor data.

[0204] It should be understood that the following embodiments can be performed according to the flowcharts described above. The embodiments relate to specific implementation methods and provide for a deeper understanding of how to operate the methods of this disclosure. The embodiments should not be construed as limiting, but rather as general guidance on how certain aspects can be performed.

[0205] Example 1: Low threshold alarm In this embodiment, the processor module may be configured to receive sensor data (510) and evaluate the sensor data using a first function (520) to determine whether the "real-time" glucose value has exceeded a first threshold (80 mg / dL), and to evaluate the sensor data using a second function (530) to determine whether the predicted glucose value will exceed a second threshold (55 mg / dL) within 15 minutes (pH). The processor module may be configured to actively monitor data related to blood glucose status (710) after the hypoglycemia indicator has been triggered based on the first or second function to determine whether a state change transition should occur. In this embodiment, there are three inactivation conditions (transition from activation or confirmation to inactivation): 1) whether the glucose value increases by more than a predetermined amount (15 mg / dL) and the glucose value is higher than the first threshold (80 mg / dL) after the hypoglycemia indicator is activated; 2) the rate of glucose change increases at a value greater than a predetermined value (1 mg / dL / min); and 3) the confirmation period ends (30 minutes).

[0206] Option A - Glucose levels above the lower threshold In this scheme, a hypoglycemic indicator is triggered based on a first function (the user's glucose level is below 80 mg / dL), and an alarm is issued to the user via a user interface (visually and audibly). The user confirms the alarm by pressing a button, and the state transitions to a confirmed state. In the confirmed state, as the glucose level drops to 65 mg / dL, then begins to rise back to 75 mg / dL, but then begins to fall again, the processor module actively monitors the user's blood glucose status. In this embodiment, the confirmed state is maintained for 30 minutes, during which the user will not receive any additional alarms. During active monitoring, the processor module determines that the user's glucose is recovering (rising back to 75 mg / dL), which triggers a confirmed recovery sub-state. The user is not alarmed until his / her glucose reaches the reactivation or deactivation condition. In the confirmed state—the recovery sub-state—the criteria for re-alarming can include any condition indicating a glucose "rebound." Advantageously, the recovery sub-state utilizes the fact that some user interaction has been detected, thus it can be assumed that the user is attempting to manage the situation themselves, and an alarm should only be issued if the user's action is insufficient and a rebound is detected (e.g., a predetermined trend reversal or glucose deterioration). In this scheme, the state transitions back to activation when glucose begins to decrease again (i.e., based on the reactivation condition), and the user is alerted again.

[0207] Option B - Glucose levels oscillate around the threshold. In this scheme, a hypoglycemic indicator is triggered based on a first function (the user's glucose level drops to 80 mg / dL), and the user is alerted via the user interface (visually and audibly). However, this time, the user's glucose level remains in the range of 70-90 mg / dL, fluctuating above and below 80 mg / dL several times. Once the user acknowledges the alarm, the acknowledged state is maintained, and the user will not receive any further alarms until the acknowledgment period ends. In other words, as the user's glucose level oscillates above and below 80 mg / dL, the state does not transition back and forth from active to inactive, and this back-and-forth transition does not trigger a re-alarm. Advantageously, this avoids annoying alarms by allowing some buffering and preventing the alarm from flashing on and off. If the user does not acknowledge, there may be a re-alarm every 5 minutes (maintaining the active state until user acknowledgment). However, it is important to note that this alarm is not a traditional "threshold" alarm. Instead, it is an "early warning" alarm, meaning that it is reasonable for the user to be alerted to the condition. After the acknowledgment period has elapsed, the state will transition to active or inactive, depending on whether the alarm condition is still met. The status transitions to active and an alarm can be displayed only if the EGV remains below the threshold of 80 mg / dL (e.g., if the confirmation time is 30 minutes and the EGV is 75 mg / dL after 30 minutes, the user will receive an alarm). If the EGV is 85 mg / dL, the status transitions to inactive, and the user will not receive an alarm (unless the predicted rate of decline is sufficient to trigger an alarm).

[0208] Option C – Predict by exceeding a threshold In this scenario, the user's glucose level is 100 mg / dL and is predicted to reach 55 mg / dL within 15 minutes, as determined by the second function (530). An activation state is triggered and an alarm is displayed to the user (as per box 550). Once the user confirms, the confirmed state remains until the glucose level drops to 80 mg / dL (the lower threshold) within 10 minutes, at which point the state remains confirmed as the user's glucose level remains near 80 mg / dL for 20 minutes. Importantly, no alarm is triggered when 80 mg / dL is reached (the confirmed state remains), and the state transitions back to active after 30 minutes if the threshold condition is met.

[0209] Example 2: Actionable alarms related to hyperglycemia Advantageously, the user can be alerted only when user action is likely necessary. In one such embodiment, the user can still be alerted as long as the glucose level exceeds an alarm threshold. However, the user may also choose to enable a wait time under certain conditions, such as when correcting for hyperglycemic shifts using known dietary intake. In some instances, regardless of whether a wait time is enabled, annoying recurring alarms when glucose levels oscillate around a threshold can be minimized or eliminated, while ensuring that the user is alerted if the glucose level exceeds the threshold twice for potentially irrelevant reasons. For example, the user may be alerted once after a glucose level rises due to carbohydrate intake (hyperglycemic alarm), or no alarm may be triggered at all. Then, the user ingests insufficient insulin, causing the glucose level to drop slightly, but then rise again.

[0210] Figure 12 This is an example graph showing that the average EGV is high in the last T minutes. Specifically, at the beginning of T minutes, the user's glucose level exceeds a predetermined hyperglycemia threshold. The processor module transitions to an active state 700 based on a (first) activation criterion of a glucose level exceeding the first predetermined hyperglycemia threshold level of 180 mg / dL, thereby triggering dynamic and intelligent monitoring of the user's blood glucose status. However, no alarm is triggered because the user has enabled a predetermined waiting period (T) of 60 minutes before the high alarm. During the 60 minutes, the processor module actively monitors the user's blood glucose status using one or more hyperglycemia (second) criteria based on the average glucose level during the predetermined waiting period. After this, if the glucose level is determined to be greater than the predetermined threshold level (180 mg / dL), an alarm-related output is generated at 1200. In this case, the active state is initiated based on the first criterion (glucose level exceeding the threshold), but no alarm is provided to the user until the second criterion is met (based on the average glucose level exceeding the threshold during the waiting period).

[0211] Figure 13 This is an example graph showing a high average glucose level that is rapidly decreasing, therefore no alarm is issued to the user or incident because although the average glucose level is high, the glucose level is decreasing rapidly. Specifically, the user's glucose level exceeds a predetermined hyperglycemic threshold at the beginning of minute T. The processor module transitions to an active state 700 based on a (first) activation criterion of a glucose level exceeding a first predetermined hyperglycemic threshold level of 180 mg / dL, thereby triggering dynamic and intelligent monitoring of the incident's blood glucose status. However, no alarm is triggered because the user has enabled a predetermined waiting time period (T) of 60 minutes before the high alarm. During the 60 minutes, the processor module actively monitors the incident's blood glucose status using one or more hyperglycemic (second) criteria based on the rate of change of glucose level, which is determined to be decreasing at a rate greater than a predetermined value (1 mg / dL / min), resulting in no output at 1300. In this case, the active state is initiated based on the first criterion (glucose level exceeding the threshold), but the processor module transitions to an inactive state at 1300.

[0212] In some instances, activation conditions may include time standards or time components, meaning that the user's glucose levels must consistently exceed a threshold over a predetermined time period, such as 60 minutes. Applied to Figure 13 In the example shown, the processor module can continue to use one or more hyperglycemia criteria, actively monitoring the user's blood glucose status based on average glucose levels and the rate of change in glucose levels. In this scenario, at 1300, the 60-minute average glucose level is found to be higher than a predetermined threshold; however, the rate of change in glucose is determined to be decreasing at a rate greater than a predetermined value (1 mg / dL / min), resulting in no output at 1300. Therefore, in this case, the system does not enter an active state and no output is provided to the user.

[0213] Figure 14This is an example graph showing a high glucose level, followed by a rapid rate of decline, but then a stable glucose level still above a predetermined threshold. Specifically, at the beginning of minute T, the user's glucose level exceeds a predetermined hyperglycemic threshold. The processor module transitions to activation state 700 based on a (first) activation criterion of a glucose level exceeding the first predetermined hyperglycemic threshold level of 180 mg / dL, thereby triggering dynamic and intelligent monitoring of the user's blood glucose status. However, no alarm is triggered because the user has enabled a predetermined waiting period (T) of 60 minutes before the high alarm. During the 60 minutes, the processor module actively monitors the user's blood glucose status using one or more hyperglycemic (second) criteria, based on the glucose level and the rate of change of the glucose level. At the end of the waiting period, it is determined that the level is still above the predetermined threshold and has a rate of decline that is not faster than the predetermined rate, resulting in an output at 1400. In this case, as indicated by the figure at 1400, although the glucose level has decreased for a period of time, it is stable and high (e.g., above the threshold) at the end of the waiting period, and no alarm should be triggered on the user or the person experiencing the high glucose level.

[0214] Figure 15 This is an example diagram showing that the user's glucose level exceeded the threshold plus the boundary even after the waiting time had elapsed. Specifically, the user's glucose level exceeded a predetermined hyperglycemic threshold at the beginning of minute T. The processor module transitioned to an active state 700 based on a (first) activation criterion of a glucose level exceeding the first predetermined hyperglycemic threshold level of 180 mg / dL, thereby triggering dynamic and intelligent monitoring of the user's blood glucose status. However, no alarm was triggered because the user activated a predetermined waiting time period (T) of 60 minutes before the high alarm. During the 60 minutes, the processor module actively monitored the user's blood glucose status using one or more hyperglycemic (second) criteria based on the glucose level plus the boundary (180 mg / dL + 50 mg / dL = 230 mg / dL), which occurred before the end of the waiting time, resulting in an output of 1500 before the 60 minutes had elapsed. In this case, since the glucose level had risen to a level that should have exceeded the non-alarm waiting time, the user or the person involved should have been alarmed at 1500.

[0215] Figure 16 It is an example diagram, which is displayed in a format similar to... Figure 14In this scenario, after the user confirms the alarm, although the glucose level decreases for a period of time, it remains stable and high (e.g., above a threshold) at the end of the waiting period. After the alarm is output at 1600, the user confirms the alarm at 1610, transitioning to a confirmed state with a 60-minute active monitoring period. Specifically, the user's glucose level exceeds a predetermined hyperglycemia threshold shortly after the user confirms the alarm, but is not low enough to trigger an inactive alarm (because the inactive condition includes the threshold plus a Δ standard). Therefore, the processor module continues to actively monitor the user's blood glucose status based on the average glucose level, which determines the average glucose level somewhere above the predetermined threshold at the end of the 60 minutes. In this case, since the average EGV remains high after 1620 during the confirmation period, the user or the person being alerted again or provided with a follow-up alarm.

[0216] The above schemes assume that "enable wait before alarm" is enabled. However, if the wait time is not enabled and the user selects a confirmation time period of 60 minutes, any of these schemes will work similarly. In this case, the user will receive a first alarm after the first criterion is met (at the beginning of the time period T), and then another alarm after the second criterion is met, for example at 1200, 1400, 1500, or 1600 (but not at other times in between when the user's glucose level oscillates around a predetermined threshold).

[0217] This embodiment further describes how the processor module actively monitors the user's blood glucose status after the user confirms a hyperglycemia alarm (e.g., triggered by the user's glucose level meeting one or more first or second criteria). After the hyperglycemia alarm is output, the user confirms the alarm, thus it can be assumed that the user is observing or taking some action. The processor module transitions to the confirmation state. The processor module then actively monitors the user's blood glucose status such that if the user's average glucose level since the alarm activation is less than (180 mg / dL) minus Δ (15 mg / dL); the user's actual (real-time) glucose level is less than a threshold minus Δ (165 mg / dL); the average glucose level since the alarm activation is less than a predetermined threshold and the confirmation time has elapsed; or the user's glucose level decreases at a rate faster than a predetermined rate (e.g., 1.0 mg / dL / min) and the glucose level is below the threshold, the processor module receives or confirms predetermined dietary information; or receives or confirms predetermined insulin delivery information, then the processor module transitions to inactivity based on those inactivation criteria.

[0218] This disclosure presents the systems and methods for performing the processing of analyte sensor data in such full, clear, concise, and precise terms the best way to consider the systems and methods for doing so, so that any person skilled in the art can practice them. However, from what has been discussed above, these systems and methods are readily available for modification and alternative structures, and they are fully equivalent.

[0219] For example, it should be further understood that the execution and / or implementation of all methods and processes can be carried out by any suitable device or system, whether local or remote. Furthermore, any combination of devices or systems can be used to execute these methods and processes. Additionally, in some instances, the methods and processes described herein can be placed in a non-transitory computer-readable storage medium, which includes code that enables the operation of this disclosure when executed by at least one processor.

[0220] Furthermore, while the foregoing descriptions have been presented in detail through illustrations and examples for clarity and understanding, it will be apparent to those skilled in the art that certain changes and modifications can be made. The functions, steps, and / or actions of the methods described in the embodiments of the invention described herein need not be performed in any particular order. Furthermore, while elements of the invention may be described or claimed in the singular, the plural form is also contemplated unless expressly stated to be limited to the singular. Therefore, this specification and embodiments should not be construed as limiting the scope of this disclosure to the specific embodiments and examples described herein, but rather encompass all modifications and substitutions within the true scope and spirit of this disclosure.

Claims

1. A method for activating a blood glucose indicator, characterized in that, The method includes: The first function is used to evaluate the selected glucose sensor data to determine whether the real-time glucose value meets or exceeds the user-set real-time threshold. The same selected glucose sensor data is evaluated using a second function different from the first function to determine whether the predicted glucose value meets or exceeds a prediction threshold within a fixed prediction range; wherein the user-configurable real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; and When the predicted glucose value meets or exceeds the prediction threshold within the fixed prediction range, a second output including a second blood glucose indicator is provided.

2. The method according to claim 1, characterized in that, The fixed prediction range cannot be set by the user.

3. The method according to claim 1, characterized in that, The fixed prediction range is preset.

4. The method according to claim 3, characterized in that, The selected glucose sensor data represents the value obtained by averaging one or more data points over a predetermined time period.

5. The method according to claim 1, characterized in that, The user-configurable real-time threshold defines the user's upper limit for glucose; The prediction threshold defines the user's upper limit for glucose; and The first blood glucose indicator and the second blood glucose indicator are hyperglycemia indicators.

6. The method according to claim 1, characterized in that, The user-configurable real-time threshold defines the user's lower limit for glucose. The prediction threshold defines the user's lower glucose limit; and The first blood glucose indicator and the second blood glucose indicator are hypoglycemia indicators.

7. A non-transitory computer-readable medium for storing instructions, which, when executed by a processor, cause the processor to perform a method for activating a blood glucose indicator, the method comprising: The first function is used to evaluate the selected glucose sensor data to determine whether the real-time glucose value meets or exceeds the user-set real-time threshold. The same selected glucose sensor data is evaluated using a second function different from the first function to determine whether the predicted glucose value meets or exceeds a prediction threshold within a fixed prediction range; wherein the user-configurable real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; as well as When the predicted glucose value meets or exceeds the prediction threshold within the fixed prediction range, a second output including a second blood glucose indicator is provided.

8. The non-transitory computer-readable medium according to claim 7, characterized in that, The fixed prediction range is 20 minutes or less.

9. The non-transitory computer-readable medium according to claim 7, characterized in that, The fixed prediction range cannot be set by the user.

10. The non-transitory computer-readable medium according to claim 7, characterized in that, The fixed prediction range is preset.

11. The non-transitory computer-readable medium according to claim 10, characterized in that, The fixed prediction range is preset at the factory.

12. The non-transitory computer-readable medium according to claim 10, characterized in that, The user-configurable real-time threshold defines the user's upper limit for glucose; The prediction threshold defines the user's upper limit for glucose; and The first blood glucose indicator and the second blood glucose indicator are hyperglycemia indicators.

13. The non-transitory computer-readable medium according to claim 10, characterized in that, The user-configurable real-time threshold defines the user's lower limit for glucose. The prediction threshold defines the user's lower glucose limit; and The first blood glucose indicator and the second blood glucose indicator are hypoglycemia indicators.

14. A system for activating a blood glucose indicator, characterized in that, The system includes: A memory configured to store glucose sensor data from a transdermal glucose sensor; and A processor, coupled to the memory, is configured to: The first function is used to evaluate the selected glucose sensor data to determine whether the real-time glucose value meets or exceeds the user-set real-time threshold. The same selected glucose sensor data is evaluated using a second function different from the first function to determine whether the predicted glucose value meets or exceeds a prediction threshold within a fixed prediction range; wherein the user-configurable real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; and When the predicted glucose value meets or exceeds the prediction threshold within the fixed prediction range, a second output including a second blood glucose indicator is provided.

15. The system according to claim 14, characterized in that, The fixed prediction range is 20 minutes or less.

16. The system according to claim 14, characterized in that, The fixed prediction range cannot be set by the user.

17. The system according to claim 14, characterized in that, The fixed prediction range is preset at the factory.

18. The system according to claim 14, characterized in that, The user-configurable real-time threshold defines the user's upper limit for glucose; The prediction threshold defines the user's upper limit for glucose; and The first blood glucose indicator and the second blood glucose indicator are hyperglycemia indicators.

19. The system according to claim 14, characterized in that, The user-configurable real-time threshold defines the user's lower limit for glucose. The prediction threshold defines the user's lower glucose limit; and The first blood glucose indicator and the second blood glucose indicator are hypoglycemia indicators.

20. A method for activating a blood glucose indicator, characterized in that, The method includes: The selected glucose sensor data is evaluated to determine whether the real-time glucose value meets or exceeds a user-set real-time threshold, wherein the selected glucose sensor data includes values ​​representing the average of one or more data points over a predetermined time period. The same selected glucose sensor data is evaluated to determine whether the predicted glucose value meets or exceeds a prediction threshold within a preset fixed prediction range, wherein the user-set real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; and When the predicted glucose value meets or exceeds the prediction threshold within the preset fixed prediction range, a second output including a second blood glucose indicator is provided.

21. A non-transitory computer-readable medium for storing instructions, which, when executed by a processor, cause the processor to perform a method for activating a blood glucose indicator, the method comprising: The selected glucose sensor data is evaluated to determine whether the real-time glucose value meets or exceeds a user-set real-time threshold, wherein the selected glucose sensor data includes values ​​representing the average of one or more data points over a predetermined time period. The same selected glucose sensor data is evaluated to determine whether the predicted glucose value meets or exceeds a prediction threshold within a preset fixed prediction range, wherein the user-set real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; as well as When the predicted glucose value meets or exceeds the prediction threshold within the preset fixed prediction range, a second output including a second blood glucose indicator is provided.

22. A system for activating a blood glucose indicator, characterized in that, The system includes: A memory configured to store glucose sensor data from a transdermal glucose sensor; and A processor, coupled to the memory, is configured to: The selected glucose sensor data is evaluated to determine whether the real-time glucose value meets or exceeds a user-set real-time threshold, wherein the selected glucose sensor data includes values ​​representing the average of one or more data points over a predetermined time period. The same selected glucose sensor data is evaluated to determine whether the predicted glucose value meets or exceeds a prediction threshold within a preset fixed prediction range, wherein the user-set real-time threshold is different from the prediction threshold. When the user-configurable real-time threshold is met or exceeded, a first output including a first blood glucose indicator is provided; and When the predicted glucose value meets or exceeds the prediction threshold within the preset fixed prediction range, a second output including a second blood glucose indicator is provided.

23. A method for activating a hypoglycemia indicator based on continuous glucose sensor data, the method comprising: The first function is used to evaluate sensor data to determine whether the real-time glucose value meets one or more first criteria; The second function is used to evaluate the sensor data to determine whether the predicted glucose value meets one or more second criteria; If one or more of the first criteria or one or more of the second criteria are met, then the hypoglycemia indicator is activated; and Output is provided based on the activated hypoglycemia indicator.

24. The method of claim 23, wherein evaluating sensor data using a first function to determine whether a real-time glucose value meets one or more first criteria includes determining whether the real-time glucose value exceeds a glucose threshold.

25. The method of claim 23 or 24, wherein the sensor data is evaluated using a first function to determine whether the real-time glucose value meets one or more first criteria, further comprising determining whether the magnitude of the rate of change or the direction of the rate of change meets the rate of change criteria.

26. A system for processing data, the system comprising: A continuous analyte sensor, configured to be implanted in the body; and A sensor electronics device configured to receive and process sensor data output by the sensor, the sensor electronics device including a processor configured to: The first function is used to evaluate sensor data to determine whether the real-time glucose value meets one or more first criteria; The second function is used to evaluate the sensor data to determine whether the predicted glucose value meets one or more second criteria; If one or more of the first criteria or one or more of the second criteria are met, then the hypoglycemia indicator is activated; and Output is provided based on the activated hypoglycemia indicator.

27. The system of claim 26, wherein evaluating sensor data using a first function to determine whether a real-time glucose value meets one or more first criteria includes determining whether the real-time glucose value exceeds a glucose threshold.

28. The system of claim 26 or 27, wherein the sensor data is evaluated using a first function to determine whether the real-time glucose value meets one or more first criteria, further comprising determining whether the magnitude of the rate of change or the direction of the rate of change meets the rate of change criteria.

29. A method for transitioning between states related to an individual's blood glucose status, the method comprising: Evaluate sensor data from a continuous glucose sensor and activate alarm states based on sensor data that meets one or more activation transition criteria associated with hypoglycemia or hyperglycemia. Provides outputs related to the activation alarm state, wherein the outputs indicate a hypoglycemic or hyperglycemic condition; The system responds to at least one of the user's confirmation of the alarm status or the data indicating that the user's glucose level is trending toward normal blood sugar levels, and transitions from an active state to a confirmed state over a period of time. During the confirmed state, data related to the individual's hypoglycemia or hyperglycemia is actively monitored for a period of time. and In response to data related to the subject's hypoglycemia or hyperglycemia that meets one or more predetermined criteria, the subject transitions from the confirmed state to at least one of the inactive or active states.

30. The method of claim 29, wherein the transition from the active state to the confirmed state comprises transitioning from the active state to the confirmed state based on data indicating that the subject's glucose is trending toward normal blood glucose levels, wherein the data is selected from a) sensor data indicating changes in glucose trends or b) insulin information related to correction of the condition.

31. The method of claim 29 or 30, wherein the transition from the active state to the confirmed state comprises a transition from the active state to the confirmed state based on user confirmation, wherein the data is selected from a) the user confirming an alarm in the user interface or b) the user entering insulin information or c) the user entering dietary information.

32. A system for processing data, the system comprising: A continuous analyte sensor, configured to be implanted in the body; and A sensor electronics device configured to receive and process sensor data output by the sensor, the sensor electronics device including a processor configured to: Evaluate sensor data from a continuous glucose sensor and activate alarm states based on sensor data that meets one or more activation transition criteria associated with hypoglycemia or hyperglycemia. Provides outputs related to the activation alarm state, wherein the outputs indicate a hypoglycemic or hyperglycemic condition; In response to at least one of the user confirming the alarm status or indicating that the user's glucose is trending toward normal blood sugar levels, the system transitions from an active state to a confirmed state over a period of time. During the confirmed state, data related to the individual's hypoglycemia or hyperglycemia is actively monitored for a period of time. and In response to data related to the subject's hypoglycemia or hyperglycemia that meets one or more predetermined criteria, the subject transitions from the confirmed state to at least one of the inactive or active states.

33. The system of claim 32, wherein the transition from the active state to the confirmed state comprises a transition from the active state to the confirmed state based on data indicating that the subject's glucose is trending toward normal blood glucose levels, wherein the data is selected from a) sensor data indicating changes in glucose trends or b) insulin information related to correction of the condition.

34. The system of claim 32 or 33, wherein the transition from the active state to the confirmed state includes a transition from the active state to the confirmed state based on user confirmation, wherein the data is selected from a) the user confirming an alarm in the user interface or b) the user entering insulin information or c) the user entering dietary information.

35. A method for determining when to issue a second alarm to a user after the user has acknowledged the first alarm, the method comprising: The alarm state is initially activated based on one or more criteria that are met based on data related to hypoglycemia or hyperglycemia. In response to at least one piece of data indicating that the user's glucose level is trending toward normal blood glucose, the system transitions to a confirmed state for a predetermined period of active monitoring time. During the active monitoring period, the processor module actively monitors data related to the subject's hypoglycemia or hyperglycemia status. and The first alarm state is reactivated during the confirmation period when data related to the subject's hypoglycemia or hyperglycemia condition meets one or more second criteria.

36. The method of claim 35, wherein the one or more second criteria are different from the one or more first criteria.

37. The method of claim 35 or 36, further comprising providing a first output associated with the initial activation and providing a second output associated with the reactivation.

38. A system for processing data, the system comprising: A continuous analyte sensor, configured to be implanted in the body; and A sensor electronics device configured to receive and process sensor data output by the sensor, the sensor electronics device including a processor configured to: The alarm state is initially activated based on one or more criteria that are met based on data related to hypoglycemia or hyperglycemia. In response to at least one piece of data indicating that the user's glucose level is trending toward normal blood glucose, the system transitions to a confirmed state for a predetermined period of active monitoring time. During the active monitoring period, data related to the subject's hypoglycemia or hyperglycemia status are actively monitored; and The first alarm state is reactivated during the confirmation period when data related to the subject's hypoglycemia or hyperglycemia condition meets one or more second criteria.

39. The system of claim 38, wherein the one or more second standards are different from the one or more first standards.

40. The system of claim 38 or 39, further comprising providing a first output associated with the initial activation and providing a second output associated with the reactivation.

Images