Data Stream Bridging for Sensor Migration

Abstract

Data stream bridging for sensor transition is described. A first data stream of glucose measurements is received from a first glucose sensor worn by a user. An end event of the first glucose sensor is detected when generation and / or communication of the first glucose measurements via the first data stream stops. A second data stream of glucose measurements is then received from a second glucose sensor worn by the user that replaces the first glucose sensor. During a warm-up period of the second glucose sensor, an estimated glucose value is output for the user based on both the first data stream of glucose measurements received from the first glucose sensor prior to the end event and the second data stream of glucose measurements received from the second glucose sensor.

Description

[Technical field]

[0001] Related Applications This application claims the benefit of U.S. Provisional Patent Application No. 63 / 189,429, filed May 17, 2021, entitled "Data-Stream Bridging for Sensor Transitions," and U.S. Provisional Patent Application No. 63 / 231,502, filed August 10, 2021, entitled "Data-Stream Bridging for Sensor Transitions," the disclosures of each of which are incorporated by reference in their entireties herein. [Background technology]

[0002] Diabetes is a metabolic condition that affects hundreds of millions of people. For these people, monitoring blood glucose levels and regulating those levels within an acceptable range is important not only to mitigate long-term problems such as heart disease and vision loss, but also to avoid the effects of high and low glucose. Because blood glucose levels are almost constantly changing over time and in response to daily occurrences such as diet or exercise, maintaining the levels within an acceptable range can be difficult. Advances in medical technology have allowed the development of various systems for monitoring blood glucose, including continuous glucose monitoring (CGM) systems that measure and record glucose concentrations in virtually real time. CGM systems are an important tool for users of these systems to ensure that measured glucose values ​​are within an acceptable range.

[0003] Many glucose monitoring systems utilize wearable devices that include a sensor that can be inserted into a user's skin to monitor glucose. Often, these sensors are disposable and designed to operate for a predetermined amount of time (e.g., 10 days), after which the sensor must be replaced with a new sensor. When a new sensor is inserted into a user's skin, the sensor may require a significant amount of time (e.g., 2 hours) to "warm up" before the sensor is able to consistently produce accurate measurements. During this period, i.e., the warm-up period, glucose measurements produced using the newly inserted glucose sensor may not be accurate and / or may not be consistently accurate.

[0004] Due to the unreliability of glucose sensors during the warm-up period, many conventional glucose monitoring systems simply refrain from displaying any type of glucose data during this period. However, the lack of available glucose data during the transition period between sensors can result in glucose alerts and alarms being disabled, which can lead to health risks for users who rely on glucose data to make important decisions, such as when to administer insulin. As a result, users may resort to using painful fingersticks during the transition period to monitor their glucose levels. Relying on fingersticks can also be inconvenient or uncomfortable and can limit the activities that those users can perform during the warm-up period. Summary of the Invention [Means for solving the problem]

[0005] To overcome these problems, data stream bridging for sensor transition is described. A first data stream of glucose measurements is received from a first glucose sensor worn by a user. The first data stream corresponds to a first period during which the wearable glucose monitoring device generates glucose measurements and streams (e.g., communicates) those measurements to a computing device associated with the user. An end event of the first glucose sensor is detected when the generation and / or communication of the first glucose measurements via the first data stream stops. Then, a second data stream of glucose measurements is received from a second glucose sensor worn by the user that replaces the first glucose sensor. During a warm-up period of the second glucose sensor, an estimated glucose value is output for the user based on both the first data stream of glucose measurements received from the first glucose sensor prior to the end event and the second data stream of glucose measurements received from the second glucose sensor.

[0006] This Summary introduces a selection of concepts in a simplified form that are further described below in the Detailed Description. As such, this Summary is not intended to identify essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0007] The detailed description is given with reference to the accompanying drawings. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is an illustration of an environment in an example implementation operable to use the techniques described herein. [Diagram 2] 1 illustrates an example of a wearable glucose monitoring device. [Diagram 3] 1 illustrates an example system for generating estimated glucose values ​​during warm-up of a glucose sensor. [Figure 4] 1 illustrates an example implementation of a user interface that displays a plot of a user's glucose over time, including estimated glucose values ​​for a warm-up period of a glucose sensor. [Diagram 5] 1 illustrates an example scenario in which a bridging system generates estimated glucose values ​​during a transition period between sensors. [Figure 6] 1 illustrates an example implementation of a user interface that displays a plot of a user's glucose over time, including during transition periods between glucose sensors. [Figure 7] An example scenario is shown in which the bridging system retroactively generates an estimated glucose value to replace a glucose measurement at the end of a first sensor's data stream. [Figure 8] 1 illustrates a procedure in an example implementation in which an estimated glucose value is output based on both a first stream of glucose measurements received from a first glucose sensor prior to a termination event and a second data stream of glucose measurements received from a second glucose sensor that replaces the first glucose sensor. [Figure 9] 1 illustrates a procedure in an example implementation in which a glucose measurement associated with a first glucose sensor is retroactively updated based on a glucose measurement received from a second glucose sensor that replaces the first glucose sensor. [Figure 10] FIG. 13 illustrates a procedure in an example implementation in which the warm-up period for a new glucose sensor ends when a glucose measurement received from the new glucose sensor meets an accuracy threshold. [Figure 11] An example of a system including various components of an exemplary device that may be implemented as any type of computing device described and / or utilized with reference to Figures 1-10 for implementing embodiments of the technology described in this specification is shown. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] overview Data stream bridging for sensor transition is described. In accordance with the described techniques, a bridging system is configured to "bridge the gap" between sensors, such as when a user replaces a first glucose sensor with a second glucose sensor. Doing so enables the system to output the user's estimated glucose values ​​during a sensor transition period that corresponds to the period between when the old sensor expires (e.g., a period after the old sensor is no longer used to report data for the user and / or a period close to when the old sensor is no longer used to report data for the user (e.g., 1 hour ago, 2 hours ago, 12 hours ago, etc.)) and when the new sensor is properly warmed up and calibrated. The bridging system may also be utilized to reduce the length of the new sensor's warm-up period. As discussed in more detail throughout, the sensor warm-up period may be any amount of time (e.g., substantially no time, 0-1 hour, 1-2 hours, more than 2 hours, etc.).

[0010] In one or more implementations, the bridging system receives a first data stream of glucose measurements from a first glucose sensor worn by a user. The first data stream corresponds to a first time period during which the wearable glucose monitoring device generates glucose measurements and streams (e.g., communicates) those measurements to a computing device associated with the user. The bridging system detects or determines an expiration event for the first glucose sensor corresponding to a time when generation and / or communication of the first glucose measurements via the first data stream stops and / or will stop (e.g., a "grace period" before actual sensor expiration).

[0011] The first sensor may be disposable and may be designed to operate for a predetermined time (e.g., 10 days) or until the occurrence of some event is detected (e.g., signal quality degrades below a quality threshold, the variance of glucose measurements exceeds a variance threshold, or a chemical reaction occurs). Thus, in some cases, the bridging system detects an end event when a timer associated with the first glucose sensor expires, indicating that a predetermined amount of time has elapsed. Alternatively or additionally, the end event may be detected when the user physically removes the first glucose sensor. For example, the first glucose sensor may be subcutaneously inserted into the user's skin and the end event may be detected when the first glucose sensor is removed from the user's skin, which results in a loss of connectivity between the first sensor and the computing device.

[0012] Regardless of how the termination event is detected or determined, the sensor bridging system is configured to "bridge the gap" during a transition period between sensors (e.g., can provide data while a second sensor intended to replace the first sensor is warming up). This transition period between sensors may be a period during which multiple sensors (e.g., two of them) are implanted in the user's body at the same time and / or a period during which only one sensor (or no sensors) is implanted in the user's body. In one example, this transition period can extend from expiration of the first sensor to insertion of the second sensor. In another example, this transition period can extend from a grace period before expiration of the first sensor to insertion of the second sensor. In another example, the transition period can extend from expiration of the first sensor to after warming up of the second sensor. In another example, the transition period can extend from a grace period before expiration of the first sensor to after warming up of the second sensor. Any desired amount of time for the transition period may be used in accordance with the concepts discussed throughout. For example, the transition period or a portion of the transition period may be a predetermined amount of time. In another example, the transition period or a portion of the transition period may be a patient-specific amount of time determined based on one or more data streams from one or more sensors and / or other data inputs.

[0013] In particular, the transition period between sensors may include both a first period beginning when an end event of the first glucose sensor is detected and ending when the second sensor is activated (e.g., inserted into the user's body), and a second period during which the second glucose sensor warms up. In particular, during the first period, no glucose measurements are received, but during the second period, i.e., the warm-up period, glucose measurements generated using the newly inserted glucose sensor may not be accurate and / or may not be consistently accurate. Due to the fact that measurements generated by the glucose sensor during the warm-up period may not accurately estimate the person's actual glucose level for a significant portion of that period, most conventional systems simply refrain from outputting the user's estimated glucose value until the warm-up period ends. However, preventing the output of glucose data during the transition between sensors may result in frustration for users who closely monitor their glucose levels and may pose a health risk for some users who are unable to see glucose data for longer periods during the transition between sensors.

[0014] To solve this problem, the bridging system described herein is configured to generate and output a set of estimated glucose measurements during the transition period between sensors. To do so, the bridging system generates an estimated glucose value for the user based on both a first glucose measurement received from a first glucose sensor prior to the termination event and a second glucose measurement received from a second glucose sensor replacing the first glucose sensor. By way of example and not limitation, the bridging system may provide the first glucose measurement and the second glucose measurement, or data representing those measurements (e.g., feature vectors), as inputs to one or more machine learning models. Such machine learning models may be configured to predict the user's current glucose level based on the user's past glucose measurements from a previous period (e.g., the first glucose measurement) and based on the user's relatively recent but potentially inappropriate measurements (e.g., the second glucose measurement). In some cases, the models may also leverage other data streams to form a more accurate glucose prediction, such as by leveraging temperature data, activity data, food logging data, and the like. By generating and outputting estimated glucose measurements, the bridging system may "bridge" at least some of the gaps in producing accurate glucose measurements that occur during transition periods. In this way, a user can view estimated glucose values, as well as other glucose-related information (e.g., alerts and visualizations), in an uninterrupted manner during transition periods between sensors.

[0015] In one or more implementations, the bridging system is configured to reduce the amount of time that the warm-up period of the new glucose sensor lasts. To shorten the warm-up period of the new glucose sensor, the bridging system determines whether a new data stream of glucose measurements received from the new glucose sensor meets an accuracy threshold based at least in part on a previous data stream of glucose measurements received from the previous glucose sensor prior to a termination event of the previous glucose sensor. And, when the accuracy threshold is met, the warm-up period of the new glucose sensor ends. In particular, by using the accuracy of the glucose measurements received from the new sensor to determine the warm-up time, the warm-up time can be shortened for a predetermined amount of time. For example, the warm-up time can be shortened from a first predetermined amount of time (e.g., 2 hours, 30 minutes, etc.) to an amount of time less than the first predetermined amount of time (e.g., less than 2 hours, less than 30 minutes, etc., respectively). This allows the user to take action to manage the glucose level sooner than if the predetermined amount of time were used for the warm-up period. Furthermore, by presenting accurate glucose sooner, potentially dangerous events with respect to the user's health can be avoided. This would also enable people who rely on wearable glucose monitoring devices to reduce the use of finger sticks, thereby eliminating a potentially painful and uncomfortable activity from their lives.

[0016] In particular, this approach can also personalize the warm-up period for a user, since different users may have different warm-up accuracy thresholds based on their confidence in the accuracy of the first and second data streams. For example, measuring a large amount of sensor noise in the first data stream may indicate gradual sensor degradation. In this case, the system may have a lower confidence in the first sensor data, which may result in a stricter accuracy threshold. Similarly, if a user has a strong reaction on their skin upon insertion of the second glucose sensor, resulting in more noise in the readings, this may also increase the accuracy threshold.

[0017] In the following description, an exemplary environment in which the techniques described herein can be used is first described. Then, implementation details and example procedures that can be performed in the exemplary environment, as well as in other environments, are described. The execution of the exemplary procedures is not limited to the exemplary environment, and the exemplary environment is not limited to the execution of the exemplary procedures.

[0018] Example Environment FIG. 1 is an illustration of an environment 100 in an example implementation operable to employ data stream bridging for sensor migration as described herein. The illustrated environment 100 includes a person 102 shown wearing a wearable glucose monitoring device 104, examples of which include a wearable glucose monitoring device 104(a) and a wearable glucose monitoring device 104(b) having first and second glucose sensors, respectively. The illustrated environment 100 also includes a computing device 106, which is shown to have a bridging system 108. The wearable glucose monitoring device 104 and the computing device 106 are communicatively coupled, including via a network 110. Some or all of the components of the environment 100 shown in FIG. 1 may be configured to be disposable (e.g., configured for one-time use by a user) or reusable (e.g., configured for use by a user for multiple analyte monitoring sessions), or none of the components may be so configured.

[0019] Alternatively or additionally, the wearable glucose monitoring device 104 and the computing device 106 may be communicatively coupled in other manners, such as using one or more wireless communication protocols or techniques. By way of example, the wearable glucose monitoring device 104 and the computing device 106 may communicate with each other using one or more of Bluetooth (e.g., a Bluetooth Low Energy link), near-field communication (NFC), 5G, etc.

[0020] In accordance with the described technology, the wearable glucose monitoring device 104 is configured to provide a measurement of glucose for the person 102. Although a wearable glucose monitoring device is discussed herein, it should be understood that data stream bridging may be used for sensor transfer of other devices capable of providing glucose measurements, such as non-wearable glucose devices (such as blood glucose meters requiring a finger prick), patches, etc. Additionally or alternatively, data stream bridging may be used in connection with the transfer of sensors that generate values ​​other than glucose values, such as temperature, other analytes (e.g., lactate, sodium, insulin, etc.), and heart rate, to name a few. However, in implementations that include a wearable glucose monitoring device 104, the wearable glucose monitoring device may be configured with a glucose sensor that detects analytes indicative of glucose for the person 102 to enable the generation of a glucose measurement, as described above and below.

[0021] In one or more implementations, the wearable glucose monitoring device 104 is a continuous glucose monitoring (CGM) system. As used herein, the term "continuous" used in connection with glucose monitoring may refer to the ability of a device to generate measurements substantially continuously, such that the device can be configured to generate glucose measurements at time intervals (e.g., every hour, every 30 minutes, every 5 minutes, every minute, every 30 seconds, etc.) in response to establishing a communication coupling with a different device (e.g., when the computing device 106 establishes a wireless connection with the wearable glucose monitoring device 104 to retrieve one or more of the measurements). This functionality, along with further aspects of the configuration of the wearable glucose monitoring device 104, are described in more detail in connection with FIG. 2.

[0022] Additionally, the wearable glucose monitoring device 104 transmits glucose measurements to the computing device 106, such as via a wireless connection. The wearable glucose monitoring device 104 can communicate the measurements in real time, for example, as the measurements are generated using a glucose sensor. Alternatively or additionally, the wearable glucose monitoring device 104 can communicate the glucose measurements to the computing device 106 at set time intervals. For example, the wearable glucose monitoring device 104 can be configured to communicate glucose measurements to the computing device 106 (as the measurements are generated) every 5 minutes. Certainly, the time intervals at which the glucose measurements are communicated may differ from the above example without departing from the spirit or scope of the described technology. The measurements may be communicated by the wearable glucose monitoring device 104 to the computing device 106 according to other bases in accordance with the described techniques, such as based on a request from the computing device 106. In any event, the computing device 106 can maintain the glucose measurements of the person 102 at least temporarily, for example, in a computer-readable storage medium of the computing device 106.

[0023] Computing device 106 may be configured in a variety of ways without departing from the spirit or scope of the described technology. By way of non-limiting example, computing device 106 may be configured as a mobile device (e.g., a mobile phone, a wearable device, or a tablet device), a desktop computer, or a laptop computer, to name a few form factors. In one or more implementations, computing device 106 may be configured as a dedicated device associated with a glucose monitoring platform (not shown) that has the functionality to, for example, obtain glucose measurements from wearable glucose monitoring device 104, perform various calculations related to the glucose measurements, display information related to the glucose measurements and the glucose monitoring platform (not shown), communicate glucose measurements to the glucose monitoring platform, etc.

[0024] Furthermore, computing device 106 may represent two or more devices in accordance with the described techniques. In one or more scenarios, for example, computing device 106 may correspond to both a wearable device (e.g., a smart watch) and a mobile phone. In such a scenario, both of these devices may be capable of performing at least some of the same operations, such as to receive glucose measurements from wearable glucose monitoring device 104, communicate those measurements over network 110 to a glucose monitoring platform, display information related to the glucose measurements, etc. Alternatively or additionally, different devices may have different capabilities that other devices do not have or that are limited to the devices specified through computing instructions.

[0025] In a scenario in which the computing device 106 corresponds to a separate smartwatch and a mobile phone, for example, the smartwatch may be configured with various sensors and capabilities for measuring various physiological markers (e.g., heart rate, heart rate variability, respiration, blood flow velocity, etc.) and activities (e.g., walking or other movement) of the person 102. In this scenario, the mobile phone may not be configured with these sensors and capabilities or may include a limited amount of the capabilities - although in other scenarios, the mobile phone may be capable of providing the same functionality. Continuing with this particular scenario, the mobile phone may have capabilities that the smartwatch does not have, such as a camera for capturing images associated with glucose monitoring, and an amount of computing resources (e.g., battery and processing speed) that allows the mobile phone to more efficiently perform calculations related to glucose measurements. Even in scenarios in which the smartwatch is capable of performing such calculations, the computing instructions may limit the execution of those calculations to the mobile phone to efficiently utilize available resources without burdening both devices. To this extent, the computing device 106 may be configured in a different manner and represent a different number of devices than described herein without departing from the spirit and scope of the described technology.

[0026] In accordance with the described techniques, the bridging system 108 may be configured to "bridge the gap" during a transition between sensors, such as when the person 102 replaces a first glucose sensor with a second glucose sensor, and / or may include a transition between sensors where a first glucose sensor and a second glucose sensor are simultaneously implanted within the person 102 for an amount of time. In the illustrated environment 100, the person 102 is shown in three stages, including a first stage 112, a second stage 114, and a third stage 116. Furthermore, the chronological order of the stages may correspond to the first stage 112, followed by the second stage 114, followed by the third stage 116. In the first stage 112, the person 102 is shown wearing a wearable glucose monitoring device 104(a) including a first glucose sensor. In the second stage 114, the person 102 is shown without wearing a glucose monitoring device. This may correspond to a period of time after person 102 removes wearable glucose monitoring device 104(a) but before person 102 applies wearable glucose monitoring device 104(b), for example, when person 102 switches from using a first sensor of wearable glucose monitoring device 104(a) to using a second sensor of wearable glucose monitoring device 104(b). In a third phase 116, person 102 is shown wearing wearable glucose monitoring device 104(b) including a second glucose sensor, as described above. In alternative embodiments, there may be additional and / or different phases (e.g., phases where person 102 wears wearable glucose monitoring device 104(a) and wearable glucose monitoring device 104(b) simultaneously). In any embodiment, the bridging system 108 may be used to help the person 102 be accurate or "warmed up" to "bridge the gap" between a first data stream associated with the wearable glucose monitoring device 104(a) and a second data stream associated with the wearable glucose monitoring device 104(b).

[0027] In one or more implementations, the wearable glucose monitoring device 104(a) and the wearable glucose monitoring device 104(b) can share one or more components, such as a common transmitter, processor, or computer-readable storage device, to name a few. In such implementations, the shared components can be reusable, such that the reusable components can be combined with one or more disposable components, such as glucose sensors, that are disposed of after a period of time. The disposable components can be designed to operate, for example, for a predetermined amount of time (e.g., 10 days, 15 days, etc.) or until the occurrence of some event is detected (e.g., signal quality degrades below a quality threshold, variance of glucose measurements exceeds a variance threshold, or a chemical reaction occurs). In this manner, the wearable glucose monitoring device 104(b) can use the same transmitter as the wearable glucose monitoring device 104(a) to transmit glucose measurements to the computing device 106.

[0028] Alternatively, the wearable glucose monitoring device 104(a) and the wearable glucose monitoring device 104(b) may not share any components and may be completely separate devices. In such implementations, the wearable glucose monitoring device 104(a) and the wearable glucose monitoring device 104(b) may have the same design and be configured in the same format, but they may be manufactured as separate devices to correspond to a first physical device and a separate second physical device. For example, the wearable glucose monitoring device 104(a) may include a first glucose sensor and a first transmitter, and the wearable glucose monitoring device 104(b) may include a second glucose sensor and a second transmitter that are separate from the first glucose sensor and the first transmitter but have the same configuration as the first glucose sensor and the transmitter. Alternatively or additionally, the wearable glucose monitoring device 104(a) and the wearable glucose monitoring device 104(b) may be separate devices of different configurations, e.g., different model numbers, having different types of components from each other. In such a scenario, separate devices having different designs and configurations may nevertheless be configured to communicate glucose measurements to the computing device 106.

[0029] The illustrated environment 100 also shows a first data stream 118, a second data stream 120, a transition period 122, an end event 124, and a warm-up period 126. In accordance with the described techniques, the first data stream 118 includes a first glucose measurement 128, which is generated by the wearable glucose monitoring device 104(a) using its respective glucose sensor (e.g., a first glucose sensor) and communicated to the computing device 106 as part of the first data stream 118. Similarly, the second data stream 120 includes second glucose measurements 130(a), 130(b), which are generated by the wearable glucose monitoring device 104(b) using its respective glucose sensor (e.g., a second glucose sensor) and communicated to the computing device 106 as part of the first data stream 118. Here, the second glucose measurement 130(a) corresponds to a portion of a measurement in the second data stream 120 generated by the wearable glucose monitoring device 104(b) during the warm-up period 126. In contrast, the second glucose measurement 130(b) corresponds to a portion of the measurements in the second data stream 120 generated by the wearable glucose monitoring device 104(b) after the warm-up period 126, e.g., a subsequent portion of the glucose measurements in the second data stream 120.

[0030] In the environment 100, the first data stream 118 and the second data stream 120 are depicted as blocks that visually convey the time period to which the streams correspond. By way of example, the first data stream 118 corresponds to a first time period during which the wearable glucose monitoring device 104(a) generates first glucose readings 128 and streams (e.g., communicates) those readings to the computing device 106. The termination event 124 corresponds to a time (e.g., a point in time) at which the generation and / or communication of the first glucose readings 128 via the first data stream 118 ends. Various events that may cause or be detected as the termination event 124 are described in more detail below.

[0031] Moreover, the illustrated environment 100 does not include a “block” between the termination event 124 that terminates the first data stream 118 and the portion of the second data stream 120 that corresponds to the warm-up period 126. This indicates that, in one or more implementations, no glucose measurements are generated by the glucose monitoring device and communicated to the computing device 106 during the period between the termination event 124 and the warm-up period 126 of the wearable glucose monitoring device 104(b). Indeed, this is consistent with the depiction of the person 102 in the second stage 114, where the person 102 is depicted without wearing the glucose monitoring device, for example, after the person 102 removes the wearable glucose monitoring device 104(a) and before the person 102 applies the wearable glucose monitoring device 104(b). However, in some implementations, there may be no "interruption" in the generation and streaming of glucose measurements, but the glucose measurements may come from different sources, such as when wearable glucose monitoring device 104(b) is applied while person 102 is still wearing wearable glucose monitoring device 104(a). In this case, bridging system 108 may receive sensor measurements from both wearable glucose monitoring device 104(a) and wearable glucose monitoring device 104(b) simultaneously. In other words, the cool-down period of a first glucose sensor may overlap with the warm-up period of a second glucose sensor. An estimated glucose value may be determined in this scenario based on real-time glucose measurements received from two different sensors that are worn simultaneously.

[0032] The second data stream 120 corresponds to a second time period that follows the first time period. This second time period also corresponds to when the wearable glucose monitoring device 104(b) generates second glucose measurements 130(a), 130(b) and streams (e.g., communicates) those measurements to the computing device 106.

[0033] The first glucose reading 128 and the second glucose reading 130(a), 130(b) are shown separately from the first and second data streams 118, 120 to visually convey how the glucose readings of the person 102 may change over time, even though the first glucose reading 128 and the second glucose reading 130(a), 130(b) are included in the first and second data streams 118, 120, respectively. By showing the first glucose reading 128 and the second glucose reading 130(a), 130(b) separately as dots, the illustrated environment 100 shows how the glucose readings of the person 102 may change over time. The illustrated "dots" may also more closely represent how the computing device 106 outputs (e.g., displays) the glucose of the person 102, for example, as a glucose trace.

[0034] Notably, the second glucose reading 130(a) corresponding to the warm-up period 126 (e.g., the warm-up period of the second glucose sensor) of the wearable glucose monitoring device 104(b) is shown to have more variability than the first glucose reading 128 and the second glucose reading 130(b) generated and communicated after the warm-up period 126. This represents that in one or more scenarios, a glucose sensor may require a certain amount of time (e.g., 2 hours, 30 minutes, etc.) before it can be used to consistently generate accurate readings. During this period, i.e., during the warm-up period 126, glucose readings generated using a newly inserted glucose sensor may not be accurate and / or may not be consistently accurate.

[0035] This may occur because some glucose sensors are designed to undergo one or more chemical reactions in response to insertion into a body (e.g., of person 102). The one or more chemical reactions may enable the sensor to detect an analyte in the body indicative of glucose. One example of such a chemical reaction is a substance that coats the glucose sensor that dissolves when inserted into the body of person 102 over warm-up period 126. Alternatively or additionally, the glucose sensor may need to be filled with fluid from the body of person 102 during warm-up period 126. Indeed, the glucose sensor may not be able to generate, or may not be trusted to generate, accurate and / or consistently accurate measurements of glucose of person 102 during the warm-up period for different reasons without departing from the spirit or scope of the described technology. In any event, measurements generated by wearable glucose monitoring device 104(b) during warm-up period 126 may not accurately estimate the actual glucose level of person 102 for a significant portion of that period, if at all.

[0036] In general, the bridging system 108 is configured to generate estimated glucose values ​​for periods during which the glucose monitoring device may generate inappropriate glucose measurements, e.g., measurements that fail to accurately estimate the person's actual glucose level. In one or more implementations, for example, the bridging system 108 is configured to generate and output a set of estimated glucose values ​​132 during the warm-up period 126 for a glucose sensor (the second glucose sensor described above) of the wearable glucose monitoring device 104(b). The bridging system 108 may generate the estimated glucose value 132 for the person 102 based on a first glucose measurement 128 in the first data stream 118 and a second glucose measurement 130(a) in the second data stream 120 that correspond to the warm-up period 126.

[0037] By way of example and not limitation, the bridging system 108 may provide the first glucose measurement 128 and the second glucose measurement 130(a), or data representative of those measurements (e.g., feature vectors), as inputs to one or more machine learning models. Such machine learning models may be configured to predict the user's current glucose level based on the user's past glucose measurements from a prior period (e.g., the first glucose measurement 128) and based on the user's more recent but potentially inappropriate measurements (e.g., the second glucose measurement 130(a)).

[0038] By generating the estimated glucose value 132, the bridging system 108 may "bridge" at least a portion of the gap in generating accurate glucose measurements that occurs during the transition period 122. In one or more implementations, the bridging system 108 may also be configured to generate an estimated glucose value for the person 102 during the second stage 114, e.g., while the person is not wearing a glucose monitoring device and corresponding to the period between the end event 124 and the start of the warm-up period 126. The bridging system 108 may be configured to generate an estimated glucose value for the person 102 during this period based solely on the first glucose measurement 128, such as by using one or more machine learning models configured to predict the person's glucose based on past glucose measurements. To this end, the bridging system 108 may generate an estimated glucose value for the person 102 while the person 102 is not wearing a glucose monitoring device (e.g., during the second stage 114) based on the first glucose measurement 128. In contrast, the bridging system 108 may generate an estimated glucose value 132 for the person 102 during the warm-up period 126 based on both the first glucose measurement 128 and the second glucose measurement 130(a) generated during the warm-up period 126. Additionally or alternatively, the bridging system 108 may leverage other data streams to accurately generate the estimated glucose value 132, such as by leveraging temperature data, activity data, food logging data, etc.

[0039] In one or more implementations, the computing device 106 (e.g., an application associated with the computing device 106) may effectively replace the second glucose measurement 130(a) with the estimated glucose value 132, for example, for decision support and display. By using the first glucose measurement 128 along with the second glucose measurement 130(a), the bridging system 108 generates a more accurate trace of the person 102's actual glucose (e.g., the estimated glucose value 132) than is represented by the second glucose measurement 130(a). With a more accurate trace, the person 102 or the person's 102's health care provider may rely on the estimated glucose value 132 during the warm-up period 126 to guide treatment decisions such as whether and what to eat, whether to administer insulin, or whether to contact a health care provider to alleviate a serious health condition, to name a few. For example, consider the following description of FIG. 2 in the context of continuously measuring glucose with a continuous glucose monitoring system and obtaining data describing such measurements.

[0040] 2 illustrates an example implementation 200 of a wearable glucose monitoring device in more detail. In particular, the illustrated example 200 includes a top view and a corresponding side view of the wearable glucose monitoring device 104. It should be understood from the following description that the wearable glucose monitoring device 104 may be modified in various ways in implementation without departing from the spirit or scope of the described techniques. As discussed above, for example, data stream bridging for sensor migration may be used in conjunction with other types of devices for glucose monitoring, such as non-wearable devices (e.g., blood glucose meters requiring finger pricks), patches, etc.

[0041] In this example 200, the wearable glucose monitoring device 104 is illustrated to include a glucose sensor 202 and a sensor module 204. Here, the glucose sensor 202 is shown in a side view inserted, for example, subcutaneously into the skin 206 of the person 102. The sensor module 204 is illustrated as a dashed rectangle in a top view. The wearable glucose monitoring device 104 also includes a transmitter 208 in the illustrated example 200. The use of a dashed rectangle for the sensor module 204 indicates that the sensor module may be contained or otherwise implemented within the housing of the transmitter 208. In this example 200, the wearable glucose monitoring device 104 further includes an adhesive pad 210 and an attachment mechanism 212.

[0042] In operation, the glucose sensor 202, adhesive pad 210, and attachment mechanism 212 may be assembled to form an application assembly configured to be applied to the skin 206 such that the sensor 202 is inserted subcutaneously as depicted. In such a scenario, the transmitter 208 may be attached to the assembly after application to the skin 206 via the attachment mechanism 212. Alternatively, the transmitter 208 may be incorporated as part of the application assembly such that the glucose sensor 202, adhesive pad 210, attachment mechanism 212, and transmitter 208 (with sensor module 204) can be applied to the skin 206 all at once. In one or more implementations, this application assembly is applied to the skin 206 using a separate sensor applicator (not shown). Unlike the finger prick required by conventional blood glucose meters, user-initiated application of the wearable glucose monitoring device 104 is substantially painless and does not require the drawing of blood. Additionally, the automatic sensor applicator generally allows the person 102 to implant the glucose sensor 202 subcutaneously in the skin 206 without the assistance of a clinician or healthcare provider.

[0043] The application assembly can also be removed by peeling the adhesive pad 210 from the skin 206. It should be understood that the wearable glucose monitoring device 104 and its various components as illustrated are simply one example form factor, and that the wearable glucose monitoring device 104 and its components can have different form factors without departing from the spirit or scope of the presently described techniques.

[0044] In operation, the glucose sensor 202 is communicatively coupled to the sensor module 204 via at least one communication channel, which may be a wireless or wired connection. Communication from the glucose sensor 202 to the sensor module 204 or from the sensor module 204 to the glucose sensor 202 may be performed actively or passively, and these communications may be continuous (e.g., analog) or discrete (e.g., digital).

[0045] The glucose sensor 202 may be a device, molecule, and / or chemical that changes or causes a change in response to an event that is at least partially independent of the glucose sensor 202. The sensor module 204 is implemented to receive an indication of a change to or caused by the glucose sensor 202. For example, the glucose sensor 202 may include glucose oxidase that reacts with glucose and oxygen to form hydrogen peroxide that is electrochemically detectable by the sensor module 204, which may include electrodes. In this example, the glucose sensor 202 may be configured as or include a glucose sensor configured to detect an analyte in blood or interstitial fluid that is indicative of a glucose level using one or more measurement techniques. In one or more embodiments, the glucose sensor 202 may also be configured to detect analytes in blood or interstitial fluid that are indicative of other markers, such as lactate levels, which may improve accuracy in identifying or predicting glucose-based events. Additionally or alternatively, the wearable glucose monitoring device 104 may include additional sensors to the glucose sensor 202 for detecting those analytes indicative of other markers.

[0046] In another example, the glucose sensor 202 (or additional, not shown, sensors of the wearable glucose monitoring device 104) can include first and second conductors, and the sensor module 204 can electrically detect a change in electrical potential across the first and second conductors of the glucose sensor 202. In this example, the sensor module 204 and the glucose sensor 202 are configured as a thermocouple, such that the change in electrical potential corresponds to a change in temperature. In some examples, the sensor module 204 and the glucose sensor 202 are configured to detect a single analyte, e.g., glucose. In other examples, the sensor module 204 and the glucose sensor 202 are configured to detect multiple analytes, e.g., sodium, potassium, carbon dioxide, and glucose. Alternatively or additionally, the wearable glucose monitoring device 104 includes multiple sensors for detecting not only one or more analytes (e.g., sodium, potassium, carbon dioxide, glucose, and insulin), but also one or more environmental conditions (e.g., temperature). Thus, the sensor module 204 and glucose sensor 202 (and any additional sensors) may detect the presence of one or more analytes, the absence of one or more analytes, and / or a change in one or more environmental conditions.

[0047] In one or more embodiments, the sensor module 204 can include a processor and memory (not shown). The sensor module 204 can utilize the processor to generate a glucose reading 214 based on communication with the glucose sensor 202 exhibiting the above-mentioned changes. The glucose reading 214 can correspond to a glucose reading of any of the data streams discussed in connection with FIG. 1, such as at least a portion of the first glucose reading 128 of the first data stream 118 or at least a portion of the second glucose reading 130(a), 130(b) of the second data stream 120. Indeed, the glucose reading 214 can correspond to other data streams without departing from the spirit or scope of the present technology. In any case, based on the above-mentioned communication from the glucose sensor 202, the sensor module 204 is further configured to generate a communicable data package including at least one glucose reading 214. In one or more implementations, the sensor module 204 can configure these packages to include additional data, including, by way of example and not limitation, supplemental sensor information 216. Such supplemental sensor information may include a sensor identifier, a sensor status, a temperature corresponding to the glucose measurement 214, other analyte measurements corresponding to the glucose measurement 214, etc. It should be understood that the supplemental sensor information 216 may include a variety of data that supplements at least one glucose measurement 214 without departing from the spirit or scope of the presently described technology.

[0048] In implementations in which the wearable glucose monitoring device 104 is configured for wireless transmission, the transmitter 208 can wirelessly transmit the glucose readings 214 and / or the supplemental sensor information 216 as a data stream to a computing device. Alternatively or additionally, the sensor module 204 can buffer the glucose readings 214 and / or the supplemental sensor information 216 (e.g., in a memory of the sensor module 204 and / or in another physical computer-readable storage medium of the wearable glucose monitoring device 104) and later have the transmitter 208 transmit the buffered glucose readings 214 and / or the supplemental sensor information 216 at various intervals, such as time intervals (every second, every 30 seconds, every minute, every 5 minutes, every hour, etc.), storage intervals (when the buffered glucose readings 214 and / or supplemental sensor information 216 reach a threshold amount of data or number of readings), etc.

[0049] Having considered an example environment and an example wearable glucose monitoring device, we now consider a description of some example technical details for data stream bridging for sensor migration according to one or more implementations.

[0050] Data Stream Bridging for Sensor Migration 3 illustrates an example system 300 for generating an estimated glucose value for a warm-up period of a glucose sensor. The illustrated example 300 includes a bridging system 108.

[0051] In this example 300, the bridging system 108 is shown receiving as inputs a first data stream 118 and a second data stream 120. The second data stream 120 may be received subsequent to and / or simultaneously with the first data stream 118. The bridging system 108 is also shown outputting an estimated glucose value 132. As shown, the bridging system 108 includes a stream handler 302 and a value prediction engine 304. The stream handler 302 is shown to include an end detection engine 306 and a data accuracy module 308. The value prediction engine 304 is shown to include prediction logic 310. These components are configured to perform various aspects that enable the bridging system 108 to "bridge the gap" for transitions between sensors, as described above and below. Although the bridging system 108 is illustrated with these various components, it should be understood that in implementations, the bridging system 108 may include fewer, more, and / or different components without departing from the spirit or scope of the described technology.

[0052] In one or more implementations, the termination detection engine 306 is configured to detect a glucose sensor termination event. In the context of FIG. 1, for example, the termination detection engine 306 is configured to detect the termination event 124. As described above, the termination event 124 may correspond to a time at which the generation and / or communication of the first glucose reading 128 by the wearable glucose monitoring device 104(a) as part of the first data stream 118 ends or is scheduled to end (e.g., a grace period before termination). The termination detection engine 306 may be configured to detect different types of termination events according to the described techniques.

[0053] By way of example, a glucose sensor may be configured for use over a limited, predetermined period of time, e.g., 7 days, 10 days, or 30 days, just to name a few. In such implementations, each glucose sensor may be associated with a timer that begins, for example, when the glucose sensor is inserted into a user's skin or otherwise deployed to measure the user's glucose. The termination detection engine 306 may maintain the timer and / or detect when the timer expires. In implementations in which the timer counts down (decays), for example, the termination detection engine 306 may detect when the timer reaches zero or drops to some other threshold indicating the end of the respective sensor's life. In contrast, in implementations in which the timer counts up (increases), the termination detection engine 306 may detect when the timer reaches a threshold (e.g., 10 days) indicating the end of the respective sensor's life.

[0054] Alternatively or additionally, the termination detection engine 306 may detect an termination event corresponding to a change in conditions other than identifying an expiration timer value representative of the "life" of the glucose sensor. By way of example, the termination detection engine 306 may detect an termination event based on a loss of connectivity (e.g., communication or physical) with the glucose sensor. For example, the termination detection engine 306 may detect that an termination event has occurred when no glucose measurements are received from the glucose sensor after a threshold amount of time, when no signal (e.g., a "heartbeat") is received from the glucose sensor, when a signal indicating termination of connectivity with the glucose sensor is received, when end-of-life chemistry is detected for the glucose sensor, or when chemistry is detected for a new sensor, just to name a few.

[0055] As described above, the glucose sensors of the wearable glucose monitoring device 104 may be inserted subcutaneously into the skin of the person 102. In one or more implementations, the termination detection engine 306 may be configured to detect a termination event when those glucose sensors are removed from the user's skin, e.g., pulled out from the skin of the person 102. In accordance with the described technology, the termination detection engine 306 is configured to detect a termination event 124 indicating when the generation and / or communication of glucose measurements of the first data stream 118 ends. Alternatively or additionally, the termination event may be initiated by a user interaction with a user interface (e.g., the user interface 316). For example, a user may be able to select a control to terminate a session with a first glucose sensor and initiate a new session with a second glucose sensor. It should be understood that the termination detection engine 306 may detect different types of termination events without departing from the spirit or scope of the described technology.

[0056] In general, the value prediction engine 304 is configured to generate the estimated glucose value 132 using actual glucose measurements generated using a glucose sensor. For example, the value prediction engine 304 may generate the estimated glucose value 132 using measurements from one or more of the first data stream 118 or the second data stream 120. Additionally, the value prediction engine 304 may generate the estimated glucose value 132 to replace one or more glucose measurements in the first data stream 118 or the second data stream 120. In the context of FIG. 1, for example, the value prediction engine 304 generates the estimated glucose value 132 to replace a glucose measurement in the second data stream 120, specifically to replace the second glucose measurement 130(a) corresponding to the warm-up period 126 of the glucose sensor of the wearable glucose monitoring device 104(b). The value prediction engine 304 may also be used to generate estimated glucose values ​​for other periods, as described in connection with FIGS. 5-7. These periods may include, for example, periods when the user is not wearing any sensor, and retroactively, at the end of sensor use (or connectivity), e.g., at the end of the sensor's life. Additionally or alternatively, the value prediction engine 304 may leverage other data streams (not shown) to accurately generate the estimated glucose value, such as by leveraging temperature data, activity data, food logging data, etc.

[0057] When generating the estimated glucose values ​​132 for the warm-up period 126 of the glucose sensor, the value prediction engine 304 may generate those values ​​based on the first glucose reading 128 and also based on the second glucose reading 130(a) corresponding to the warm-up period 126. To generate the estimated glucose value 132 from the first glucose reading 128 and the second glucose reading 130(a), the value prediction engine 304 uses a prediction logic 310. In operation, the prediction logic 310 may be configured as executable code (e.g., "binary") capable of receiving those glucose readings as inputs, processing them according to an encoded algorithm, formula, set of rules, and / or model, and determining and outputting the estimated glucose value 132. The prediction logic 310 may be implemented based on various algorithms, formulas, rules, and / or models without departing from the spirit or scope of the described technology.

[0058] In one or more implementations, for example, the prediction logic 310 may determine the estimated glucose value 132 using one or more weighting techniques, such as a weighted average. For example, the prediction logic 310 may associate a relatively smaller weight with the second glucose measurement 130(a) toward the beginning of the warm-up period 126 and a relatively larger weight with the second glucose measurement 130(a) closer to the end of the warm-up period 126. In one or more implementations, the prediction logic 310 may associate a gradually increasing weight with the second glucose measurement 130(a) over time. As an example, at the beginning of the warm-up period 126, the prediction logic 310 may assign a weight of 0.9 to the first glucose measurement 128 and a weight of 0.1 to the second glucose measurement 130(a), and then calculate the estimated glucose value 132 as the sum of the weighted first and second glucose measurements. In contrast, towards the end of the warm-up period 126, the prediction logic 310 may assign a weight of 0.1 to the first glucose measurement 128 and a weight of 0.9 to the second glucose measurement 130(a), and then calculate an estimated glucose value 132 as the weighted sum of the first and second glucose measurements.

[0059] As part of determining the estimated glucose value 132, the prediction logic 310 may also associate weights with the first glucose measurement 128 or a prediction derived from those measurements. For example, the prediction logic 310 may include one or more machine learning models trained on historical data of a population of users to predict the person's glucose over a future period (e.g., the coming 30 minutes, 2 hours, or all day). Such machine learning models may predict the glucose of the person 102 in response to receiving one or more sequences (e.g., a trace) of glucose measurements as input (e.g., the last 30 minutes, last 6 hours, or last day of glucose measurements of the person 102). In such implementations, the prediction logic 310 may determine the estimated glucose value 132 based on a glucose value predicted for the warm-up period 126 (e.g., predicted from the first glucose measurement 128) and also based on a second glucose measurement 130(a) corresponding to the warm-up period 126.

[0060] For example, while the prediction logic 310 associates a progressively greater weight with the second glucose measurement 130(a) over time, the prediction logic 310 may also associate a progressively lesser weight with the glucose value predicted for the warm-up period 126. In this manner, the predicted glucose value may have a greater influence on the estimated glucose value 132 toward the beginning of the warm-up period 126, and the second glucose measurement 130(a) may have a greater influence on the estimated glucose value 132 toward the end of the warm-up period 126. The weight associated with the predicted glucose value may be said to "decay," whereas the weight associated with the second glucose measurement 130(a) may be said to "rise" over time. In one or more implementations, the prediction logic 310 may determine the estimated glucose value 132 by calculating a weighted average of the predicted glucose value and the second glucose measurement 130(a).

[0061] The prediction logic 310 may also determine the estimated glucose value 132 based on the stream supplement 312. The stream supplement 312 is shown in dashed lines to indicate that its use is optional, e.g., in one or more implementations, it may not be used to determine the estimated glucose value 132. Nonetheless, in implementations in which the stream supplement 312 is used to determine the estimated glucose value 132, the stream supplement 312 may include data describing one or more aspects of the person 102, one or more sensors used to obtain measurements, the first or second data streams 118, 120 from which measurements are communicated to the computing device 106, and / or other factors that may affect the accuracy or reliability of measurements generated by those sensors and / or received by the computing device 106.

[0062] By way of example, aspects of the person 102 that the stream supplement 312 may describe include the person's 102's temperature, the food the person 102 ate (e.g., how much, when, type of food, and / or macros), the exercise performed by the person 102 (e.g., how long, intensity, type, and / or performance metrics), the person's 102's sleep (e.g., duration, quality, and / or when), analytes measured for the person 102 in addition to glucose (e.g., insulin, lactate, sodium, potassium, and / or carbon dioxide), the person's physiological markers (e.g., heart rate, heart rate variability (HRV), blood pressure, and / or blood oxygen level), the person's 102's stress (e.g., identification of stressful events), the person's demographics (e.g., age, gender, location, and genetic markers), and the person's identified or diagnosed health conditions (e.g., type 1 diabetes, type 2 diabetes, pregnancy, injury, and / or surgery), just to name a few.

[0063] Exemplary aspects of one or more sensors that the stream supplement 312 may describe include the "state" of the sensor, which refers to a value that the sensor is designed to determine and associate with one or more of its measurements. Such a state indicates whether the sensor is operating "normally" or outside normal conditions when the respective measurement is generated. Other exemplary aspects of a sensor that the stream supplement 312 may describe include, for example, the manufacturing lot of the sensor, a unique identifier for the sensor, the model number of the sensor, factory calibration information for the sensor, the remaining time the sensor will be used to generate measurements, the relative accuracy of the sensor, the relative variability of the sensor, information regarding errors detected for the sensor, a measure of the quality of the information generated by the sensor, and / or information regarding the interface of the sensor with other components, to name just a few.

[0064] Exemplary aspects of the data stream that the stream supplement 312 may describe include connectivity issues (e.g., loss of connectivity) between the wearable glucose monitoring device 104(a) and the computing device 106, connectivity issues between the wearable glucose monitoring device 104(b) and the computing device 106, corruption detected for one or more portions of the data stream, quality metrics associated with one or more portions of the data stream (e.g., describing the quality of corresponding packets and / or communications of those packets), and delays in one or more portions of the data stream, to name just a few. Indeed, these are merely examples of aspects that may be described by the stream supplement 312, and various other aspects that affect the accuracy of measurements generated by sensors and / or received by the computing device 106 may be described without departing from the spirit or scope of the described technology.

[0065] As described above, the prediction logic 310 may be implemented based on various algorithms, formulas, rules, and / or models. In implementations in which the prediction logic 310 is implemented based on or otherwise includes one or more machine learning models, for example, at least one of the models may be “generic” to a population. Such models may be trained using training data describing users of the user population and are trained to generate glucose predictions for various users of the user population. Generally, a model that is “generic” to a population, when used in operation to generate predictions for a particular end user, has not been further trained for a particular end user, e.g., through transfer learning and / or further training using training data for the particular end user. Alternatively or additionally, at least one of the models used by the prediction logic 310 may be specific to an end user. Such a model may be trained using historical data describing the particular end user that is trained to generate glucose predictions. In one or more implementations, such a “user-specific” model may initially be “general” to the population and then further trained using transfer learning and / or training data that describes a particular user.

[0066] In one or more implementations, the machine learning model of the prediction logic 310 may receive as input data from the first data stream 118 (e.g., one or more of the first glucose measurements 128), data from the second data stream 120 (e.g., one or more of the second glucose measurements 130(a)), and the stream supplement 312, which describe various aspects of the person 102 and / or aspects of the context in which the prediction logic 310 generates the estimated glucose value 132. The machine learning model can then determine the estimated glucose value 132 by applying weights to the inputs (e.g., feature vectors representing the aforementioned data). The weights may correspond to an underlying model determined using a user population and / or historical data of the person 102. In such implementations, the machine learning model is configured to output data (e.g., feature vectors) indicative of the estimated glucose value 132. The output data (e.g., feature vectors) may be processed to obtain the estimated glucose value 132. The value prediction engine 304 can use prediction logic 310 to generate estimated glucose values ​​132 during and until the warm-up period 126 ends. In one or more implementations, the machine learning model can output a probability that a user's glucose reading will be above a particular glucose threshold (e.g., above 180 mg / dL) during a transition between sensors, indicating a risk of hyperglycemia, and / or a probability that a user's glucose reading will be below a particular glucose threshold (e.g., below 70 mg / dL) during a transition between sensors, indicating a risk of hypoglycemia.

[0067] In this example 300, the data accuracy module 308 determines when the warm-up period ends. In one or more implementations, the data accuracy module 308 determines the end of the warm-up period based on a predetermined amount of time (e.g., substantially no time required to warm up, less than an hour, such as 30 minutes, 2 hours, more than 2 hours, etc.) after detecting that the sensor is inserted into the person 102. The predetermined amount of time may be set based on numerous real-world controlled tests that show that the glucose sensor meets a threshold amount of accuracy after the predetermined amount of time in the absence of any defects. In the context of FIG. 1, for example, the data accuracy module 308 may determine that the warm-up period 126 ends a predetermined amount of time after detecting that the glucose sensor of the wearable glucose monitoring device 104 is inserted into the person 102. Alternatively or additionally, the data accuracy module 308 may determine the end of the warm-up period based on a determination that the data in the second data stream 120 accurately describes the glucose of the person 102, e.g., the accuracy of the second data stream 120 meets an accuracy threshold.

[0068] In one or more implementations, for example, the prediction logic 310 may determine the accuracy of the data of the second data stream 120 based on the glucose prediction generated by the prediction logic 310 (e.g., using the first data stream 118) and based on the data of the second data stream 120. By way of example, the data accuracy module 308 can compare the data of the second data stream 120 to the prediction and determine the accuracy of the data of the second data stream 120 based in part on the comparison. For example, the data accuracy module 308 can compare the prediction and the data of the second data stream 120 to identify differences therebetween. Based on these differences, the data accuracy module 308 can further determine when the data of the second data stream 120 remains within the range of the glucose prediction generated by the prediction logic 310 for a threshold amount of time, at a threshold frequency, and / or for a threshold number of measurements.

[0069] Alternatively or additionally, the data accuracy module 308 can calculate one or more metrics regarding the data of the second data stream 120 and determine the accuracy of the second data stream 120 based on the metrics. For example, the data accuracy module 308 can calculate the variability of the second glucose measurements 130(a) of the second data stream 120. In this scenario, when the data accuracy module 308 determines that the calculated variability meets the accuracy threshold, the data accuracy module 308 can also determine that the warm-up period 126 has ended. In one or more implementations, the data accuracy module 308 can generate a measure of accuracy based on one or more of the techniques described above and / or using other techniques. It should be appreciated that the data accuracy module 308 can determine the accuracy of the data stream generated by and received from the sensor in a variety of ways without departing from the spirit or scope of the described techniques.

[0070] By using the accuracy of the data of the second data stream 120 to determine the warm-up period 126, the warm-up period 126 may be shortened for a predetermined amount of time. For example, the warm-up period 126 may be shortened from a first predetermined amount of time (e.g., 2 hours, 30 minutes, etc.) to an amount of time less than the first predetermined amount of time (e.g., less than 2 hours, less than 30 minutes, etc., respectively). Using a glucose prediction for the warm-up period 126 based on the first data stream 118 may also shorten the warm-up period 126 in relation to the predetermined amount of time. In doing so, the computing device 106 may present the glucose of the person 102 to them (or another user associated with the person 102) sooner than if the predetermined amount of time were used for the warm-up period 126. This allows the person 102, a health care provider, or another user to take steps to manage the glucose level of the person 102 sooner than if the predetermined amount of time were used for the warm-up period 126. By presenting accurate glucose sooner, potentially dangerous events regarding the health of the person 102 may be avoided. This also allows people who rely on the wearable glucose monitoring device 104 to reduce finger sticks compared to when a predetermined amount of time is used, eliminating a potentially painful and uncomfortable activity from their lives and improving their quality of life.

[0071] The illustrated example 300 also includes a display module 314, which is shown to receive the estimated glucose value 132 as an input. In general, the display module 314 is configured to generate a user interface 316 that displays a glucose presentation 318. In one or more implementations, the user interface 316 is displayed via a display device of the computing device 106, where the user interface 316 may be displayed within an application. Such an application may correspond to a glucose monitoring platform (e.g., a provider of the wearable glucose monitoring device 104) and may run on the computing device 106, such as in the background and / or in response to selection of an icon via a display device of the computing device 106. Furthermore, such an application may correspond to a web application that interacts with the glucose monitoring platform over the network 110 to provide various functions.

[0072] The glucose presentation 318 may be configured in various ways to display one or more of the first glucose reading 128, the estimated glucose value 132, and / or the second glucose reading 130(a), 130(b). For example, the glucose presentation 318 may be configured as one or more glucose traces plotted over time. In one or more implementations, the glucose presentation 318 may also be configured to display the second glucose reading 130(a) simultaneously with the estimated glucose value 132. By displaying both, the glucose presentation 318 visually conveys the accuracy (or inaccuracy) of the actual reading generated during the warm-up period 126 and / or the difference between the actual reading and the estimated glucose value 132. Alternatively or additionally, the display module 314 may determine not to display the glucose reading or the estimated glucose value in various scenarios, such as when the difference between the estimated glucose value 132 and the second glucose reading 130(a) exceeds a threshold. Consider the following description of FIG. 4 in the context of displaying a user interface that includes a glucose presentation.

[0073] FIG. 4 shows an example implementation 400 of a user interface that displays a plot of a user's glucose over time, including estimated glucose values ​​for a warm-up period of a glucose sensor.

[0074] Example 400 illustrates an example of a computing device 106, where the computing device includes a display device 402 that is shown to output a user interface 404 for display in four temporally ordered stages 406-412, including a first stage 406, a second stage 408, a third stage 410, and a fourth stage 412. It should be appreciated that the user interface 404 may correspond to the user interface 316. Thus, the display module 314 may cause the output of the user interface 404 for display.

[0075] In the first stage 406, the user interface 404 includes a graph plotting the first glucose readings 128 over time. This may correspond to one example of a glucose presentation 318. Here, the user interface 404 also includes a key 414 identifying a first symbol 416 and a second symbol 418. In this example 400, the first symbol 416 represents a glucose reading generated by a glucose sensor, and the second symbol 418 represents an estimated glucose value estimated by the prediction logic 310, for example, based on the glucose readings generated by the glucose sensor. In particular, the graph presented via the user interface 404 in the first stage 406 displays only the glucose readings represented by the multiple first symbols 416 plotted on the graph. Furthermore, the graph shown in the first stage 406 does not display the estimated glucose value, which is indicated by the absence of the second symbol 418 on the graph.

[0076] The user interface 404 also includes a current glucose element 420. Generally, the current glucose element 420 displays a glucose measurement or an estimated glucose value corresponding to the current time, such as the most recent glucose measurement generated or the most recent estimated glucose value. However, when the user is not wearing a glucose sensor, the current glucose element 420 may display one or more symbols (e.g., "--" or "N / A") indicating that no measurement or estimate is available for display in the current glucose element 420. Alternatively, the current glucose element 420 may display an estimated glucose value when the user is not wearing a glucose sensor.

[0077] In this example 400, the second stage 408 corresponds to a time point subsequent to an earlier time point corresponding to the first stage 406. In the second stage 408, the user interface 404 is shown displaying the plurality of first glucose readings 128 displayed in the first stage 406. However, the graph does not show a symbol between the time 422 and the current time, indicated by the text "NOW" along the x-axis of the graph. This may indicate that the corresponding user is not wearing a glucose sensor, such as after removing the first sensor but before applying the second subsequent sensor. In the context of FIG. 1, for example, the user interface as displayed in the second stage 408 may correspond to the person 102 in the second stage 114, for example, after removing the wearable glucose monitoring device 104(a) but before applying the wearable glucose monitoring device 104(b). In the second stage 408, the current glucose element 420 does not display a value indicative of the corresponding user's glucose. Instead, the current glucose element 420 displays "--" in the second stage 408, which may indicate, as described above, that the user is not currently wearing a sensor or that no glucose measurements are being received by the computing device 106 from the sensor.

[0078] The third stage 410 corresponds to a time point subsequent to the time point corresponding to the second stage 408. In the third stage 410, the user interface 404 is shown to display a number of first glucose readings 128 displayed in the first and second stages 406, 408, but fewer readings than in either preceding stage. In this example 400, no glucose symbols are shown between the time 422 and the subsequent time 424. In accordance with the described technique, this gap between the symbols may correspond to the period during sensor wearing, i.e., after the first sensor is removed and before the second sensor is applied. In addition, the user interface 404 is shown to display a number of estimated glucose values ​​132 in the third stage 410. This may indicate that the corresponding user is wearing a second sensor (e.g., a sensor of the wearable glucose monitoring device 104(b)), but the glucose values ​​are estimated and displayed, rather than displaying readings generated by and received from the second sensor. This third stage 410 may correspond to a point during the second sensor warm-up period 126 at which, for example, an estimated glucose value 132 is displayed in place of the second glucose measurement 130(a) because the second glucose measurement 130(a) does not meet an accuracy threshold.

[0079] In the third stage 410, the current glucose element 420 displays the glucose value with an additional information indicator "*". In accordance with the described technology, the additional information indicator in the current glucose element 420 can function as a visual "warning" or "alert" indicating that there is relevant information related to the value or symbol displayed in the current glucose element 420. Here, the additional information indicator corresponds to a warning reminding the user that the estimated glucose value 132 is estimated and may not correspond to the user's actual glucose in some circumstances, such as when the user's actual glucose changes rapidly or due to events not considered by the prediction logic 310. The current glucose element 420 may include additional information indicators in connection with displaying different relevant information without departing from the spirit or scope of the described technology.

[0080] In the third stage 410, the user interface 404 also displays a time remaining indicator 426. The time remaining indicator 426 may indicate an actual amount of time remaining in the warm-up period 126, such as when the warm-up period 126 is configured to last a predetermined amount of time, e.g., 2 hours, 30 minutes, etc. Alternatively or additionally, the time remaining indicator 426 may indicate an estimated amount of time remaining in the warm-up period 126, such as when the data accuracy module 308 determines the end of the warm-up period 126 based on the accuracy of the glucose measurements generated by and received from each sensor that meet an accuracy threshold. The time remaining indicator 426 may include a numeric value indicating the amount of time remaining and / or a non-numeric visual element indicating the amount of time remaining. The time remaining indicator 426 may indicate the actual or estimated amount of time remaining in a variety of ways, including visually, audibly, and / or tactilely, without departing from the spirit or scope of the described technology.

[0081] A fourth stage 412 corresponds to a time point after the time point corresponding to the third stage 410. In the fourth stage 412, the user interface 404 is shown displaying the multiple estimated glucose values ​​132 displayed in the third stage 410, along with additional values ​​of the estimated glucose values ​​132. To the left of time 424, no symbols are displayed on the graph of the user interface 404 in the fourth stage 412. This gap in symbols and the leftward shift of the estimated glucose values ​​132 across stages 408-412 indicates the passage of time. In the fourth stage 412, the user interface 404 also displays a second glucose measurement 130(b), which in the context of FIG. 1 corresponds to a period after the warm-up period 126. In accordance with the described techniques, the display module 314 can display the second glucose measurement 130(b) after the glucose sensor has been deployed for a warm-up period 126, which continues until a predetermined amount of time has elapsed or in response to a determination by the data accuracy module 308 that the accuracy of the measurement meets an accuracy threshold.

[0082] FIG. 5 illustrates an example scenario 500 in which a bridging system generates estimated glucose values ​​during a transition period between sensors.

[0083] The illustrated example 500 includes the person 102 shown in the first, second, and third stages 112-116 along with the computing device 106 and the bridging system 108. The illustrated example 500 also shows the first glucose reading 128 in the first data stream 118 and the second glucose readings 130(a), 130(b) in the second data stream 120. In this example 500, the bridging system 108 is also shown outputting an estimated glucose value 132 for the warm-up period 126, similar to FIG.

[0084] However, in contrast to FIG. 1 , the bridging system 108 is shown in this example 500 to output a second set of estimated glucose values ​​502. In this example 500, the second set of estimated glucose values ​​502 corresponds to the second stage 114, which is a period during which the person 102 does not wear a glucose sensor. This may correspond to a time when the user transitions between sensors, for example, from a first sensor (e.g., of the wearable glucose monitoring device 104(a)) to a second sensor (e.g., of the wearable glucose monitoring device 104(b)). Stated differently, the bridging system 108 may generate estimated glucose values ​​for a period whose beginning corresponds to the removal of the wearable glucose monitoring device 104(a) from the person 102 and whose ending corresponds to the deployment of the wearable glucose monitoring device 104(b). In contrast to the estimated glucose values ​​132, the second set of estimated glucose values ​​502 may be generated based solely on the first glucose measurement 128. This is because the sensor may not have been deployed to the person 102 for the period of time corresponding to determining the second set of estimated glucose values ​​502. For example, the glucose sensor of the wearable glucose monitoring device 104(b) may not have been deployed yet when the bridging system 108 determines and causes output of the second set of estimated glucose values ​​502.

[0085] According to the described techniques, the second set of estimated glucose values ​​502 may correspond to a glucose prediction generated by the prediction logic 310 based on the first glucose measurement 128 of the first data stream 118. In contrast to the estimated glucose values ​​132, the prediction logic 310 may determine the second set of estimated glucose values ​​502 without using the second glucose measurement 130(a). In fact, the prediction logic 310 may determine the second set of estimated glucose values ​​502 before the second glucose measurement 130(a) is generated, for example, because the wearable glucose monitoring device 104(a) may not yet be deployed when the second set of estimated glucose values ​​502 is determined. In this scenario, the prediction logic 310 may generate the glucose prediction used as the second set of estimated glucose values ​​502 using one or more machine learning models, such as one or more general models configured to generate predictions for users of a user population and / or one or more user-specific models configured to generate predictions for a particular user. Consider the following description of FIG. 6 in the context of displaying a set of estimated glucose values ​​during a transition between sensors.

[0086] FIG. 6 shows an example implementation 600 of a user interface that displays a plot of a user's glucose over time, including during transition periods between glucose sensors.

[0087] The illustrated example 600 shows another example of a computing device having a display device 402. In this example 600, the display device 402 is shown outputting a user interface 404 in four temporally ordered stages 602-608, including a first stage 602, a second stage 604, a third stage 606, and a fourth stage 608. The user interface 404 may correspond to the user interface 316, and the display module 314 may cause the output of the user interface 404. However, in contrast to the example shown in FIG. 4, the stages 602-608 may plot a corresponding user's glucose as usage transitions between glucose sensors, including when the user is not wearing a glucose sensor.

[0088] In this example 600, the first stage 602 is similar to the first stage 406 shown in Figure 4. In the first stage 602, for example, the user interface 404 includes a graph that plots the first glucose measurement 128 over time. The presentation of the graph with glucose plotted may correspond to another example of a glucose presentation 318. Notably, the graph presented via the user interface 404 in the first stage 602 displays only glucose measurements and does not display estimated glucose values.

[0089] The second stage 604 in this example 600 corresponds to a time subsequent to an earlier time corresponding to the first stage 602. Similar to the second stage 408 of FIG. 4, the user interface 404 is shown displaying the plurality of first glucose readings 128 in the second stage 604 that were previously displayed in the first stage 602. Here, the user interface 404 is shown displaying the plurality of first glucose readings 128 up to a time 610. The time 610 may correspond to an end event 124 when the person 102 removes the wearable glucose monitoring device 104(a) and its respective glucose sensor and stops generating and transmitting glucose readings.

[0090] However, in contrast to the second stage 408 of Figure 4, the user interface 404 is shown to display a second set 502 of estimated glucose values ​​between time 610 and a current time, indicated by the text "NOW" along the x-axis of the graph. This is in contrast to the second stage 408 of Figure 4, as the second stage 408 of Figure 4 does not show a symbol during this same time period. Thus, stages 602-608 illustrate a different scenario than stages 406-412 of Figure 4, i.e., stages 602-608 illustrate a scenario in which the bridging system 108 is configured to predict glucose values ​​for periods when the user is not wearing a sensor and to cause a display of those values.

[0091] As the user interface 404 in the second stage 604 displays such predicted values, the current glucose element 420 displays the glucose value with an additional information indicator "*" in the second stage 604. In this scenario, the additional information indicator can function as a visual "warning" or "alert" that there is relevant information related to the value or symbol displayed in the current glucose element 420. Here, the additional information indicator corresponds to a warning reminding the user that the second set of estimated glucose values ​​502 are estimated and may not correspond to the user's actual glucose in some circumstances, such as when the user's actual glucose changes rapidly or due to events not considered by the prediction logic 310.

[0092] The third and fourth stages 606, 608 are similar to the third and fourth stages 410, 412 of Figure 4, except that rather than displaying no measurements or values ​​while the user transitions between sensors, the third and fourth stages 606, 608 display the second set of estimated glucose values ​​502. This may allow the user or healthcare provider to continue making treatment decisions during periods when the user is not wearing the glucose sensor, e.g., transitioning between sensors. Consider the following discussion of Figure 7 in the context of retrospectively updating at least one glucose sensor value of a data stream of glucose measurements.

[0093] FIG. 7 illustrates an example scenario 700 in which the bridging system retroactively generates an estimated glucose value to replace a glucose measurement at the end of a first sensor's data stream.

[0094] The illustrated example 700 includes the person 102 shown in the first, second, and third stages 112-116 along with the computing device 106 and the bridging system 108. The illustrated example 700 also shows a first glucose reading 128 in the first data stream 118 and a second glucose reading 130(a), 130(b) in the second data stream 120. In this example 700, the bridging system 108 is also shown outputting an estimated glucose value 132 for the warm-up period 126, similar to FIG.

[0095] However, in contrast to FIG. 1, the bridging system 108 is shown to output a retrospective glucose value 702 in this example 700. The transition period 704 in this example 700 also spans a different amount of time than the transition period 122 of FIG. 1, which begins at the end event 124 and lasts until the end of the warm-up period 126. In contrast, the transition period 704 begins at the beginning of the cool-down period 706 and lasts until the end of the warm-up period 126, as shown. This cool-down period 706 may correspond to a predetermined amount of time at the end of the sensor's life. Alternatively, the data accuracy module 308 (or a different module configured to calculate a confidence in the accuracy of the glucose measurements) can determine the start of the cool-down period 706 based on the accuracy of the first glucose measurement 128, such that the cool-down period 706 is determined to start when the first glucose measurement 128 does not meet a threshold accuracy, for example, based on an unexpected fluctuation in the sensor signal. Regardless of whether the cool-down period 706 corresponds to a predetermined amount of time or is determined based on the accuracy of the measurements, the bridging system 108 may be configured to retroactively update the first glucose measurement 128 that corresponds to the cool-down period 706.

[0096] In one or more implementations, the bridging system 108 can determine the retrospective glucose value 702 based on the first glucose reading 128 generated during the cool-down period 706 and based on one or more of the second glucose readings 130(a), 130(b). In one or more scenarios, the data accuracy module 308 may not determine that the first glucose reading 128 does not meet the threshold accuracy until the computing device 106 receives the second glucose readings 130(a), 130(b). In fact, the bridging system 108 may determine the retrospective glucose value 702 as the glucose sensor of the wearable glucose monitoring device 104(b) generates glucose values ​​and communicates them to the computing device 106. Thus, in one or more implementations, the prediction logic 310 may be configured to retrospectively generate glucose values ​​given, for example, as input measurements preceding the prediction (e.g., one or more of the first glucose measurements 128 before the cool-down period 706), measurements corresponding to the same time as the prediction (e.g., the first glucose measurement 128 corresponding to the cool-down period 706), and measurements following the prediction (e.g., one or more of the second glucose measurements 130(a), 130(b)). The prediction logic 310 may generate the retrospective glucose value 702 for the cool-down period 706 using one or more machine learning models and / or weighted averages.

[0097] Having described exemplary details of a technique for data stream bridging for sensor migration, we will now consider some example procedures to illustrate additional aspects of the technique.

[0098] Exemplary Procedure This section describes example procedures for data stream bridging for sensor migration. Aspects of the procedures may be implemented in hardware, firmware, software, or combinations thereof. The procedures are illustrated as a set of blocks that specify operations to be performed by one or more devices, and are not necessarily limited to the order shown for performing the operations by each block. In at least some implementations, the procedures are performed by a bridging system, such as bridging system 108, utilizing a stream handler 302 and a value prediction engine 304.

[0099] FIG. 8 shows a procedure 800 in an example implementation in which an estimated glucose value is output based on both a first stream of glucose measurements received from a first glucose sensor prior to a termination event and a second data stream of glucose measurements received from a second glucose sensor that replaces the first glucose sensor or is intended to replace the first glucose sensor at some point.

[0100] A first data stream of glucose measurements is received from a first glucose sensor worn by a user (block 802). As an example, the bridging system 108 receives a first data stream 118 of first glucose measurements 128 from a first glucose sensor of a wearable glucose monitoring device 104(a) worn by the person 102.

[0101] A first glucose sensor termination event is detected (block 804). As an example, the termination detection engine 306 detects a first glucose sensor termination event 124. As described above, the termination event 124 corresponds to a time when the generation and / or communication of a first glucose reading 128 by the wearable glucose monitoring device 104(a) as part of the first data stream 118 ends or will end. In different implementations, the termination detection engine 306 may be configured to detect different types of termination events.

[0102] By way of example, a glucose sensor may be configured for use over a limited, predetermined period of time, e.g., 7 days, 10 days, or 30 days, just to name a few. In such implementations, each glucose sensor may be associated with a timer that begins, for example, when the glucose sensor is inserted into a user or otherwise deployed to measure the user's glucose. The end detection engine 306 may maintain the timer and / or detect when the timer expires. For example, if the timer counts down (decays), the end detection engine 306 may detect when the timer reaches 0 or drops to some other threshold indicating the end of the respective sensor's life. In contrast, if the timer counts up (increases), the end detection engine 306 may detect when the timer reaches a threshold (e.g., 10 days) indicating the end of the respective sensor's life.

[0103] Alternatively or additionally, the termination detection engine 306 may detect an termination event corresponding to a change in a condition other than identifying an expiration timer value for the "life" of the glucose sensor. As an example, the termination detection engine 306 may detect an termination event based on a loss of connectivity (e.g., communication or physical) with the glucose sensor. For example, the termination detection engine 306 may detect that an termination event has occurred when no glucose reading is received from the glucose sensor after a threshold amount of time, when no signal (e.g., a "heartbeat") is received from the glucose sensor, when a signal indicating termination of connectivity with the glucose sensor is received, when an end-of-life chemistry is detected for the glucose sensor, or when a chemistry is detected for a new sensor, just to name a few. As described above, the glucose sensor of the wearable glucose monitoring device 104 may be inserted subcutaneously into the skin of the person 102. In one or more implementations, the termination detection engine 306 may be configured to detect an termination event when the glucose sensor is removed from the user's skin, e.g., pulled out of the skin of the person 102.

[0104] A second data stream of glucose measurements is received from a second glucose sensor worn by the user that replaces (e.g., immediately or eventually) the first glucose sensor (block 806). By way of example, the bridging system 108 receives a second data stream 120 of glucose measurements 130(a) from a second glucose sensor of a wearable glucose monitoring device 104(a) worn by the person 102. The second data stream 120 and the first data stream 118 may be received subsequent to one another and / or simultaneously with one another.

[0105] During the warm-up period of the second glucose sensor, an estimated glucose value for the user is output based on both the first data stream of glucose measurements received from the first glucose sensor (e.g., prior to the termination event) and the second data stream of glucose measurements received from the second glucose sensor (block 808). As an example, during the warm-up period 126 of the second glucose sensor of the wearable glucose monitoring device 104(b), the value prediction engine 304 outputs an estimated glucose value 132 for the person 102 based on both the first data stream 118 of the first glucose measurements 128 received from the first glucose sensor prior to the termination event 124 and the second data stream 120 of the glucose measurements 130(a) received from the second glucose sensor.

[0106] In one or more implementations, the prediction logic 310 may determine the estimated glucose value 132 using one or more weighting techniques, such as weighted averaging. For example, the prediction logic 310 may associate a relatively small weight with a second glucose measurement 130(a) corresponding to the warm-up period 126 toward the beginning of the warm-up period 126, and the prediction logic 310 may associate a relatively large weight with a second glucose measurement 130(a) corresponding to the warm-up period 126 closer to the end of the warm-up period 126. In one or more implementations, the prediction logic 310 may associate gradually increasing weights with the second glucose measurements 130(a) corresponding to the warm-up period 126 over time.

[0107] FIG. 9 shows a procedure 900 in an example implementation in which a glucose measurement associated with a first glucose sensor is retroactively updated based on a glucose measurement received from a second glucose sensor that replaces the first glucose sensor.

[0108] A first data stream of glucose measurements is received from a first glucose sensor worn by a user (block 902). As an example, the bridging system 108 receives a first data stream 118 of first glucose measurements 128 from a first glucose sensor of a wearable glucose monitoring device 104(a) worn by the person 102.

[0109] A second data stream of glucose measurements is received from a second glucose sensor worn by the user that replaces the first glucose sensor (block 904). As an example, the bridging system 108 receives a second data stream 120 of glucose measurements 130(a) from a second glucose sensor of a wearable glucose monitoring device 104(b) worn by the person 102 that replaces the first glucose sensor.

[0110] At least one glucose measurement of the first data stream of glucose measurements is retroactively updated based on a second data stream of glucose measurements received from a second glucose sensor (block 906). By way of example, the transition period between sensors may include a cool-down period 706 corresponding to a predetermined amount of time at the end of the life of the sensor. Alternatively, the data accuracy module 308 may determine the start of the cool-down period 706 based on the accuracy of the first glucose measurement 128, such that the cool-down period 706 is determined to begin when the first glucose measurement 128 does not meet a threshold accuracy. Regardless of whether the cool-down period 706 corresponds to a predetermined amount of time or is determined based on the accuracy of the measurement, the bridging system 108 may be configured to retroactively update at least one of the first glucose measurements 128 that corresponds to the cool-down period 706.

[0111] The bridging system 108 can determine the retrospective glucose value 702 based on the first glucose reading 128 generated during the cool-down period 706 and based on one or more of the second glucose readings 130(a), 130(b). In one or more scenarios, the data accuracy module 308 may not determine that the first glucose reading 128 does not meet the threshold accuracy until the computing device 106 receives the second glucose readings 130(a), 130(b). In fact, the bridging system 108 may determine the retrospective glucose value 702 as the glucose sensor of the wearable glucose monitoring device 104(b) generates glucose values ​​and communicates them to the computing device 106.

[0112] Thus, in one or more implementations, the prediction logic 310 may be configured to retrospectively generate glucose values ​​given, for example, as input measurements preceding the prediction (e.g., one or more of the first glucose measurements 128 before the cool-down period 706), measurements corresponding to the same time as the prediction (e.g., the first glucose measurement 128 corresponding to the cool-down period 706), and measurements following the prediction (e.g., one or more of the second glucose measurements 130(a), 130(b)). The prediction logic 310 may generate the retrospective glucose value 702 for the cool-down period 706 using one or more machine learning models and / or weighted averages.

[0113] FIG. 10 illustrates a procedure 1000 in an example implementation in which the warm-up period for a new glucose sensor ends when a glucose measurement received from the new glucose sensor meets an accuracy threshold.

[0114] A warm-up period is initiated for a new glucose sensor worn by a user to replace a previous glucose sensor (block 1002). During the warm-up period of the new glucose sensor, a new data stream of glucose measurements is received from the new glucose sensor (block 1004), and it is determined whether the new data stream of glucose measurements received from the new glucose sensor meets an accuracy threshold based at least in part on a previous data stream of glucose measurements received from the previous glucose sensor prior to an end event of the previous glucose sensor (block 1006). For example, the prediction logic 310 may determine an accuracy of the data of the second data stream 120 based on a glucose prediction generated by the prediction logic 310 (e.g., using the first data stream 118) and based on the data of the second data stream 120. The data accuracy module 308 can compare the data of the second data stream 120 to the prediction and determine an accuracy of the data of the second data stream 120 based in part on the comparison.

[0115] For example, the data accuracy module 308 can compare the predictions to the data of the second data stream 120 to identify those differences. Based on those differences, the data accuracy module 308 can further determine when the data of the second data stream 120 remains within the range of the glucose prediction generated by the prediction logic 310 over a threshold amount of time, at a threshold frequency, and / or over a threshold number of measurements. Alternatively or additionally, the data accuracy module 308 can calculate one or more metrics regarding the data of the second data stream 120 and determine the accuracy of the second data stream 120 based on the metrics. For example, the data accuracy module 308 can calculate the variability of the second glucose measurements 130(a) of the second data stream 120.

[0116] The warm-up period of the new glucose sensor ends when the accuracy threshold is met (block 1008). For example, the data accuracy module 308 may determine that the warm-up period 126 ends when the data accuracy module 308 determines that the calculated variability meets the accuracy threshold. In particular, by using the accuracy of the data of the second data stream 120 to determine the warm-up period 126, the warm-up period 126 may be shortened for a predetermined amount of time. The systems and procedures described above may be used in various combinations to implement different methods without departing from the spirit or scope of the described technology. Indeed, at least one method may include different aspects of the described procedures and / or systems according to the described technology.

[0117] Having described example procedures according to one or more implementations, we now turn to example systems and devices that can be utilized to implement the various techniques described herein.

[0118] Exemplary Systems and Devices 11 illustrates, generally at 1100, an example system including an example computing device 1102 that represents one or more computing systems and / or devices capable of implementing various techniques described herein. This is illustrated through the inclusion of a bridge system 108. The computing device 1102 may be, for example, a server of a service provider, a device associated with a client (e.g., a client device), an on-chip system, and / or any other suitable computing device or computing system.

[0119] Computing device 1102 as shown includes a processing system 1104, one or more computer-readable media 1106, and one or more I / O interfaces 1108 communicatively coupled to each other. Although not shown, computing device 1102 may further include a system bus or other data and command transfer system coupling various components together. The system bus may include any one or combination of different bus structures, such as a memory bus or memory controller, a peripheral bus, a universal serial bus, and / or a processor or local bus utilizing any of a variety of bus architectures. Various other examples, such as control and data lines, are also contemplated.

[0120] The processing system 1104 represents functionality for performing one or more operations using hardware. Thus, the processing system 1104 is illustrated as including hardware elements 1110, which may be configured as a processor, functional blocks, etc. This may include implementations in hardware as application specific integrated circuits or other logic devices formed using one or more semiconductors. The hardware elements 1110 are not limited by the materials from which they are formed or the processing mechanisms employed therein. For example, a processor may be comprised of semiconductors and / or transistors (e.g., electronic integrated circuits (ICs)). In such a context, processor-executable instructions may be electronically executable instructions.

[0121] The computer readable medium 1106 is illustrated as including memory / storage 1112. The memory / storage 1112 represents memory / storage capacity associated with one or more computer readable media. The memory / storage component 1112 may include volatile media (such as random access memory (RAM)) and / or non-volatile media (such as read only memory (ROM), flash memory, optical disks, magnetic disks, etc.). The memory / storage component 1112 may include fixed media (e.g., RAM, ROM, fixed hard drives, etc.) as well as removable media (e.g., flash memory, removable hard drives, optical disks, etc.). The computer readable medium 1106 may be configured in a variety of other ways, as further described below.

[0122] Input / output interface 1108 represents functionality that allows a user to input commands and information into computing device 1102 and that allows information to be presented to a user and / or other components or devices using various input / output devices. Examples of input devices include keyboards, cursor control devices (e.g., a mouse), microphones, scanners, touch capabilities (e.g., capacitive or other sensors configured to detect physical contact), cameras (e.g., employing visible wavelengths or non-visible wavelengths such as infrared frequencies to recognize movements as non-touch gestures), and the like. Examples of output devices include display devices (e.g., a monitor or projector), speakers, printers, network cards, haptic response devices, and the like. Thus, computing device 1102 can be configured in a variety of ways to support user interaction, as described further below.

[0123] Various techniques may be described herein in the general context of software, hardware elements, or program modules. Generally, such modules include routines, programs, objects, elements, components, data structures, etc. that perform particular tasks or implement particular abstract data types. As used herein, the terms "module," "function," and "component" generally refer to software, firmware, hardware, or combinations thereof. Aspects of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.

[0124] An implementation of the described modules and techniques may be stored on or transmitted across some form of computer-readable media. Computer-readable media may include a variety of media that may be accessed by computing device 1102. By way of example, and not limitation, computer-readable media may include "computer-readable storage media" and "computer-readable signal media."

[0125] A "computer-readable storage medium" may refer to a medium and / or device that allows for persistent and / or non-transient storage of information, as opposed to merely a signal transmission, carrier wave, or signal itself. Thus, a computer-readable storage medium refers to a non-signal-bearing medium. A computer-readable storage medium includes hardware, such as volatile and non-volatile, removable and non-removable media, and / or storage devices implemented in a manner or technology suitable for storage of information, such as computer-readable instructions, data structures, program modules, logic elements / circuits, or other data. Examples of computer-readable storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, hard disk, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device, or other storage device, tangible media, or any article suitable for storing the desired information and that can be accessed by a computer.

[0126] A "computer-readable signal medium" may refer to a signal-bearing medium configured to transmit instructions to the hardware of the computing device 1102, such as via a network. Signal media may typically embody computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave, data signal, or other transport mechanism. Signal media also includes any information delivery media. The term "modulated data signal" means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media include wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media.

[0127] As previously described, the hardware elements 1110 and computer-readable media 1106 represent modules, programmable device logic, and / or fixed device logic implemented in hardware form that may be employed in some embodiments to implement at least some aspects of the techniques described herein, such as to execute one or more instructions. The hardware may include integrated circuits or components of on-chip systems, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs), and other implementations in silicon or other hardware. In this context, the hardware may operate as a processing device that executes program tasks defined by instructions and / or logic embodied by the hardware, as well as hardware utilized to store instructions for execution, such as computer-readable storage media previously described.

[0128] Combinations of the foregoing may be used to implement various techniques described herein. Thus, software, hardware, or executable modules may be implemented as one or more instructions and / or logic embodied on some form of computer-readable storage medium and / or by one or more hardware elements 1110. Computing device 1102 may be configured to implement specific instructions and / or functions corresponding to software and / or hardware modules. Thus, implementation aspects of modules executable as software by computing device 1102 may be achieved at least partially in hardware, for example, through the use of computer-readable storage medium and / or hardware elements 1110 of processing system 1104. The instructions and / or functions may be executable / operable by one or more articles of manufacture (e.g., one or more computing devices 1102 and / or processing system 1104) to implement the techniques, modules, and examples described herein.

[0129] The techniques described herein may be supported by various configurations of computing device 1102 and are not limited to the specific examples of the techniques described herein. This functionality may also be implemented in whole or in part through the use of a distributed system, such as on a "cloud" 1114 via platform 1116, as described below.

[0130] The cloud 1114 includes and / or represents a platform 1116 for resources 1118. The platform 1116 abstracts the underlying functionality of the hardware (e.g., servers) and software resources of the cloud 1114. The resources 1118 may include applications and / or data that may be utilized while computer processing is running on a server that is remote from the computing device 1102. The resources 1118 may also include services provided over the Internet and / or over a subscriber network, such as a cellular network or a Wi-Fi network.

[0131] The platform 1116 can abstract resources and functionality for connecting the computing device 1102 with other computing devices. The platform 1116 can also serve to abstract the scaling of resources to provide a level of scale corresponding to the demand faced by the resources 1118 implemented via the platform 1116. Thus, in an embodiment of interconnected devices, implementation aspects of the functionality described herein can be distributed throughout the system 1100. For example, functionality can be implemented partially on the computing device 1102 as well as via the platform 1116 that abstracts the functionality of the cloud 1114.

[0132] conclusion Although the systems and techniques have been described in language specific to structural features and / or methodological acts, it should be understood that the systems and techniques defined in the appended claims are not necessarily limited to the particular features or acts described. Rather, the particular features and acts are disclosed as exemplary forms for implementing the claimed subject matter. [Explanation of symbols]

[0133] 100 Environment in an Exemplary Implementation 102 people 104 Wearable Glucose Monitoring Devices 106 Computing Devices 108 Bridging System 110 Network 112 First Stage 114 Second Stage 116 Third Stage 118 First Data Stream 120 Second Data Stream 122 Transition Period 124 Ending Event 126 Warm-up Period 128 First Glucose Reading 130 Second Glucose Reading 132 Estimated Glucose Level 202 Glucose Sensor 204 Sensor Module 206 Skin 208 Transmitter 210 Adhesive Pad 212 Attachment mechanism 214 Glucose Measurements 216 Supplemental Sensor Information 302 Stream Handler 304 Value Prediction Engine 306 Exit Detection Engine 308 Data Accuracy Module 310 Prediction Logic 312 Stream Supplement 314 Display Module 316 User Interface 318 Glucose Presentation 402 Display Devices 404 User Interface 406 First Stage 407 steps 408 Second Stage 409 steps 410 Third Stage 411 steps 412 Fourth Stage 414 Identifying Key 416 First Symbol 418 Second Symbol 420 Current glucose elements 422 hours 424 hours 426 Time Remaining Indicator 502 2nd set 502 Estimated Glucose Level 602 First Stage 603 steps 604 Second Stage 605 steps 606 Third Stage 607 steps 608 Fourth Stage 610 hours 702 Retrospective glucose values 704 Transition Period 706 Cooldown Period 800 Procedure in an Exemplary Implementation 802 Block 804 Block 806 Block 808 Block 900 Procedure in an Exemplary Implementation 902 Block 904 Block 906 Block 1000 Procedure in an exemplary implementation 1002 Block 1004 Block 1006 Block 1008 Block 1100 System 1102 Computing Devices 1104 Processing System 1106 Computer-readable medium 1108 Interface 1110 Hardware elements 1112 Storage device 1114 Cloud 1116 Platform 1118 resources

Claims

1. Receiving a first data stream of glucose measurement values from a first glucose sensor worn by a user; Detecting an end event of the first glucose sensor; Receiving a second data stream of glucose measurement values from a second glucose sensor worn by the user, which replaces the first glucose sensor; Outputting an estimated glucose value for the user based on both the first data stream of glucose measurement values received from the first glucose sensor before the end event of the first glucose sensor and the second data stream of glucose measurement values received from the second glucose sensor during a warm-up period for the second glucose sensor; A method comprising the above.

2. The outputting further comprises: Predicting the user's glucose value during the warm-up period based on the first data stream of glucose measurement values received from the first glucose sensor before the end event of the first glucose sensor; Outputting the estimated glucose value based on both the predicted glucose value and the second data stream of glucose measurement values received from the second glucose sensor; The method according to claim 1, further comprising the above.

3. The method according to claim 2, wherein the predicting is further based on at least one of food logging data or activity data.

4. The method according to claim 2 or 3, wherein the outputting further comprises determining a weighted average between the predicted glucose value and the second data stream of glucose measurement values received from the second glucose sensor.

5. The method according to claim 2 or 3, further comprising outputting an estimated glucose value based on the second data stream of glucose measurement values received in real time from the second glucose sensor worn by the user after the warm-up period for the second glucose sensor has ended.

6. The method according to claim 1 or 2, wherein the warm-up period for the second glucose sensor comprises a predetermined amount of time.

7. Determining the accuracy of the glucose measurement values of the second data stream; When the accuracy of the glucose measurement value of the second data stream satisfies an accuracy threshold, ending the warm-up period of the second glucose sensor; The method according to claim 1 or 2, further comprising.

8. The method according to claim 7, wherein the accuracy of the glucose measurement value of the second data stream is based on a comparison of the glucose measurement value of the second data stream with a predicted glucose value determined based on the first data stream of glucose measurement values received before the end event.

9. Predicting the glucose value of the user based on the first data stream of glucose measurement values received from the first glucose sensor before the end event; Outputting the predicted glucose value over a period after the end event and before the second glucose sensor is worn by the user; The method according to claim 1 or 2, further comprising.

10. The method according to claim 1 or 2, wherein detecting the end event of the first glucose sensor includes detecting the end event when a timer associated with the first glucose sensor expires.

11. The method according to claim 1 or 2, wherein detecting the end event of the first glucose sensor includes detecting the end event based on a loss of connectivity with the first glucose sensor or based on user interaction with a user interface.

12. The method according to claim 1 or 2, wherein the first glucose sensor is subcutaneously inserted into the skin of the user, and the end event is detected in response to the first glucose sensor being removed from the skin of the user.

13. The method according to claim 1 or 2, wherein the outputting includes displaying the estimated glucose value on a user interface of a computing device communicatively coupled to the second glucose sensor.

14. The method according to claim 1 or 2, wherein the first glucose sensor and the second glucose sensor are disposable continuous glucose monitoring (CGM) sensors.

15. Receiving a first data stream of glucose measurement values from a first glucose sensor worn by a user; Receiving a second data stream of glucose measurement values from a second glucose sensor worn by the user, which replaces the first glucose sensor; Retrospectively updating at least one glucose measurement value in the first data stream of glucose measurement values based on the second data stream of glucose measurement values received from the second glucose sensor; A method comprising the above.

16. Starting a warm-up period for a new glucose sensor worn by a user, which replaces a previous glucose sensor; Receiving a new data stream of glucose measurement values from the new glucose sensor during the warm-up period of the new glucose sensor; Determining whether the new data stream of glucose measurement values received from the new glucose sensor meets an accuracy threshold based at least in part on a previous data stream of glucose measurement values received from the previous glucose sensor before an end event of the previous glucose sensor; Ending the warm-up period of the new glucose sensor when the accuracy threshold is met; A method comprising the above.

17. Outputting an estimated glucose value of the user based on a first data stream of glucose measurement values received from a first glucose sensor worn by the user; Outputting the estimated glucose value of the user based on the first data stream of glucose measurement values received from the first glucose sensor and a second data stream of glucose measurement values received from a second glucose sensor during a warm-up period of the second glucose sensor, which replaces the first glucose sensor; Outputting the estimated glucose value of the user based on the second data stream of glucose measurement values received from the second glucose sensor after the warm-up period of the second glucose sensor ends; A method comprising the above.

18. Outputting a predicted glucose value of the user based on the first data stream of glucose measurements received from the first glucose sensor before the end event, after the end event associated with the first glucose sensor and before the warm-up period of the second glucose sensor begins, the method according to claim 17, further comprising.

19. The method according to claim 17 or 18, further comprising preventing output of the predicted glucose value of the user after an end event associated with the first glucose sensor and before a warm-up period of the second glucose sensor begins.

20. The method according to claim 17 or 18, wherein the first glucose sensor and the second glucose sensor are disposable continuous glucose monitoring (CGM) sensors.

21. One or more processors; A memory, Receiving a first data stream of glucose measurements from a first glucose sensor worn by a user; Detecting an end event of the first glucose sensor; Receiving a second data stream of glucose measurements from a second glucose sensor worn by the user, which replaces the first glucose sensor; Outputting an estimated glucose value for the user based on both the first data stream of glucose measurements received from the first glucose sensor before the end event and the second data stream of glucose measurements received from the second glucose sensor during a warm-up period for the second glucose sensor; A memory storing computer-readable instructions executable by the one or more processors to perform operations including; A system comprising.

22. One or more processors; A memory, Receiving a first data stream of glucose measurements from a first glucose sensor worn by a user; Receiving a second data stream of glucose measurements from a second glucose sensor worn by the user, which replaces the first glucose sensor; Based on the second data stream of glucose measurement values received from the second glucose sensor, retroactively updating at least one glucose measurement value in the first data stream of glucose measurement values; A memory storing computer-readable instructions executable by the one or more processors to perform operations including; A system comprising. **Claim 23** One or more processors; A memory that Starts a warm-up period for a new glucose sensor worn by a user that replaces a previous glucose sensor; Receives a new data stream of glucose measurement values from the new glucose sensor during the warm-up period of the new glucose sensor; Determines whether the new data stream of glucose measurement values received from the new glucose sensor meets an accuracy threshold based at least in part on a previous data stream of glucose measurement values received from the previous glucose sensor prior to an end event of the previous glucose sensor; Ends the warm-up period of the new glucose sensor when the accuracy threshold is met; A memory storing computer-readable instructions executable by the one or more processors to perform operations including; A system comprising. **Claim 24** One or more processors; A memory that Outputs an estimated glucose value of a user based on a first data stream of glucose measurement values received from a first glucose sensor worn by the user; Outputs the estimated glucose value of the user based on the first data stream of glucose measurement values received from the first glucose sensor and a second data stream of glucose measurement values received from a second glucose sensor during a warm-up period of the second glucose sensor that replaces the first glucose sensor; Outputs the estimated glucose value of the user based on the second data stream of glucose measurement values received from the second glucose sensor after the warm-up period of the second glucose sensor has ended; A memory storing computer-readable instructions executable by the one or more processors to perform operations including A system comprising **Claim 25** Receiving a first data stream of glucose measurements from a first glucose sensor worn by a user; Detecting an end event of the first glucose sensor; Receiving a second data stream of glucose measurements from a second glucose sensor worn by the user to replace the first glucose sensor; Outputting an estimated glucose value for the user based on both the first data stream of glucose measurements received from the first glucose sensor before the end event and the second data stream of glucose measurements received from the second glucose sensor during a warm-up period for the second glucose sensor; A non-transitory computer-readable medium storing instructions executable by one or more processors to perform operations including **Claim 26** Receiving a first data stream of glucose measurements from a first glucose sensor worn by a user; Receiving a second data stream of glucose measurements from a second glucose sensor worn by the user to replace the first glucose sensor; Retrospectively updating at least one glucose measurement in the first data stream of glucose measurements based on the second data stream of glucose measurements received from the second glucose sensor; A non-transitory computer-readable medium storing instructions executable by one or more processors to perform operations including **Claim 27** Starting a warm-up period for a new glucose sensor worn by a user to replace a previous glucose sensor; Receiving a new data stream of glucose measurements from the new glucose sensor during the warm-up period of the new glucose sensor Determining whether a new data stream of glucose measurement values received from the new glucose sensor meets an accuracy threshold based at least in part on a previous data stream of glucose measurement values received from the previous glucose sensor prior to an end event of the previous glucose sensor; When the accuracy threshold is met, ending the warm-up period of the new glucose sensor; A non-transitory computer-readable medium storing instructions executable by one or more processors to perform operations including the above. **Claim 28** Outputting an estimated glucose value of a user based on a first data stream of glucose measurement values received from a first glucose sensor worn by the user; During a warm-up period of a second glucose sensor replacing the first glucose sensor, outputting the estimated glucose value of the user based on the first data stream of glucose measurement values received from the first glucose sensor and a second data stream of glucose measurement values received from the second glucose sensor; After the warm-up period of the second glucose sensor ends, outputting the estimated glucose value of the user based on the second data stream of glucose measurement values received from the second glucose sensor; A non-transitory computer-readable medium storing instructions executable by one or more processors to perform operations including the above.

Images