System and method for detecting a missed bolus dose

Abstract

Methods, systems, and apparatuses are provided for dynamically adjusting the sensitivity of a computer-implemented detector for detecting missed insulin boluses. The disclosed methods, systems, and apparatuses receive blood glucose measurements and insulin dosing information for a user and analyze the received information to detect whether the user likely ingested food without taking an insulin bolus. The sensitivity level of the detector can be dynamically adjusted based on contextual information about the user to adjust for the user's context, environment, and / or conditions. If the detector detects a likely missed bolus, the detector can generate a user notification.

Description

Technical Field

[0001] This disclosure relates to apparatus and methods for detecting missed insulin bolus doses. More particularly, this disclosure relates to apparatus and methods for detecting when a person with diabetes may have missed an insulin bolus to make up for a meal event, and alerting the person so that they can take action to prevent or mitigate any resulting hyperglycemic fluctuations. Background Technology

[0002] People with diabetes typically have a reduced ability to regulate their own blood sugar levels. If blood sugar levels drop too low, they enter a dangerous state called hypoglycemia. If their blood sugar levels rise too high, they enter another dangerous state called hyperglycemia. Therefore, people with diabetes strive to keep their blood sugar levels within a target range by taking insulin (to lower blood sugar levels) or by eating food and / or taking glucagon (to raise blood sugar levels). Insulin can be administered in various forms, including by injection and / or using a pump. For example, insulin can be administered as a single, discrete dose (e.g., a bolus), or as a steady drip or a series of micro-pumps infused over time, over several minutes or hours, using a pump. Too much insulin will lower blood sugar levels too much, causing hypoglycemia. Too little insulin will raise blood sugar levels too much, causing hyperglycemia. Therefore, people with diabetes must take the right amount of insulin at the right time.

[0003] When people with diabetes eat, their blood sugar levels are typically expected to rise, sometimes to dangerously high levels, unless compensated for by insulin boluses. Therefore, some people with diabetes require multiple daily insulin injections (multiple boluses) to regulate their blood sugar levels. These boluses are usually taken immediately before, simultaneously with, or immediately after eating. However, the psychological burden of having to remember to take an insulin bolus every time someone eats, snacks, or drinks a carbonated beverage can be significant. If someone with diabetes forgets to take an insulin bolus to offset the food intake (e.g., misses a bolus), their blood sugar levels can rise high enough to cause hyperglycemia. Severe hyperglycemia may require immediate emergency treatment. Even less severe but chronic hyperglycemia can have long-term negative health effects.

[0004] Therefore, there is a need for methods and devices to detect when a person with diabetes has missed an insulin bolus. Such methods and devices should preferably notify a person with diabetes of a missed bolus at a sufficiently early time so that the person can take insulin and prevent or mitigate any hyperglycemic episodes. Summary of the Invention

[0005] According to exemplary embodiments of the present disclosure, a method for detecting missed insulin boluses is provided. The method includes: receiving at a computing device at at least one first signal representing a plurality of blood glucose measurements and insulin dosing information of the user, the insulin dosing information including the administration time of at least one insulin bolus; analyzing the plurality of blood glucose measurements and the insulin dosing information using a detector of the computing device configured to a first sensitivity level to determine whether the at least one first signal indicates that the user missed an insulin bolus after a meal event; generating a user notification when the detector determines that the at least one first signal indicates that the user missed an insulin bolus; receiving context information about the user at the computing device; reconfiguring the detector, at least based on the context information, to detect missed boluses according to a second sensitivity level different from the first sensitivity level; receiving at the computing device at at least one second signal representing additional blood glucose measurements and additional insulin dosing information of the user; and analyzing the additional blood glucose measurements and the additional insulin dosing information using a detector of the computing device configured to a second sensitivity level to determine whether the at least one second signal indicates that the user missed an insulin bolus after a meal event. Attached Figure Description

[0006] The above and other features and advantages of this disclosure, as well as ways of implementing them, will become more apparent and better understood by referring to the following description of embodiments of the invention, taken in conjunction with the accompanying drawings, wherein:

[0007] Figure 1 A system according to some embodiments is described for detecting a possible missed insulin bolus dose and notifying the user of this.

[0008] Figure 2 An exemplary process for detecting missed insulin bolus doses, according to some embodiments, is described.

[0009] Figure 3 Illustrative graphs according to some embodiments are depicted, showing the trade-offs between different levels of sensitivity for detecting missed injections.

[0010] Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Figure 9 and Figure 10 Each describes an exemplary process, according to some embodiments, for receiving context information and reconfiguring the sensitivity level of the detector.

[0011] In the various views, corresponding reference numerals denote corresponding parts. The examples set forth herein illustrate exemplary embodiments of the invention, and these examples should not be construed as limiting the scope of the invention in any way. Detailed Implementation

[0012] As used herein, the terms “logic,” “control logic,” “application,” “process,” “method,” “algorithm,” and “instruction” can include software and / or firmware that executes on one or more programmable processors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), hardwired logic, or combinations thereof. Therefore, various logics can be implemented in any suitable manner according to embodiments, and will still be based on the embodiments disclosed herein.

[0013] Figure 1 A system 100, according to some embodiments, is described for detecting a possible missed injection and notifying the user of this. System 100 includes a computing device 110 that wirelessly communicates with a blood glucose sensing device 120, a drug delivery device 140, and a sensor 130. The computing device 110 can also communicate with a server 160 via a network 150.

[0014] Computing device 110 exemplarily includes mobile devices such as smartphones. Alternatively, any suitable computing device may be used, including but not limited to laptop, desktop, tablet, or server computers. Computing device 110 includes a processor 112, memory 116, a display / user interface (UI) 118, and communication devices 119.

[0015] Processor 112 includes at least one processor that executes software and / or firmware stored in memory 116 of computing device 110. The software / firmware code contains instructions that, when executed by processor 112, cause processor 112 to perform the functions described herein. Such instructions exemplify detectors and / or detector logic 114 operable to perform the functions described in further detail below. Memory 114 is any suitable computer-readable medium accessible by processor 112. Memory 114 may be a single storage device or multiple storage devices, may be located internally or externally to processor 112, and may include volatile and non-volatile media. Exemplary memory 114 includes random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic storage devices, optical disk storage, or any other suitable medium configured to store data and accessible by processor 112.

[0016] The computing device 110 includes a display / user interface 118 that communicates with the processor 112 and is operable to provide user input data to the system and to receive and display data, information, and prompts generated by the system. The user interface 118 includes at least one input device for receiving user input and providing it to the system. In the illustrated embodiment, the user interface 118 is a graphical user interface (GUI) including a touchscreen display operable to display data and receive user input. The touchscreen display allows the user to interact with presented information, menus, buttons, and other data to receive information from the system and provide user input to the system. Alternatively, a keyboard, keypad, microphone, mouse pointer, or other suitable user input device may be provided.

[0017] The computing device 110 also includes a communication device 119 that allows the computing device 110 to establish wired and / or wireless communication links with other devices. The communication device 119 may include one or more wireless antennas and / or signal processing circuitry for transmitting and receiving wireless communications, and / or one or more ports for receiving physical cables used for transmitting and receiving data. Using the communication device 119, the computing device 110 can establish one or more short-range communication links, including one or more of communication links 101 with the blood glucose sensing device 120, communication links 102 with the sensor 130, and communication links 103 with the drug delivery device 140. Such short-range communication links can utilize any known wired or wireless communication technology or protocol, including but not limited to radio frequency communication (e.g., Wi-Fi, Bluetooth, Bluetooth Low Energy (BLE), Near Field Communication (NFC), RFID, etc.), infrared transmission, microwave transmission, and optical wave transmission. Such short-range communication links can be unidirectional links (e.g., a data stream only from the blood glucose sensor 120, sensor 130, and / or device 140 to the computing device 110) or bidirectional links (e.g., bidirectional data streams). Communication device 119 can also allow computing device 110 to establish a remote communication link with server 160 via network 150 and communication links 104 and 105. Server 160 may be located remotely from computing device 110, for example, in another building, another city, or even in another country or continent. Network 150 may include any cellular or data network suitable for potentially relaying information from computing device 110 to and / or from server 160 via one or more intermediate nodes or switches. Examples of suitable networks 150 include cellular networks, metropolitan area networks (MANs), wide area networks (WANs), and the Internet.

[0018] The blood glucose sensor 120 exemplarily includes any sensor suitable for measuring blood glucose levels in a person with diabetes, such as a blood glucose monitor (BGM), a continuous glucose monitor (CGM), and / or a flash glucose monitor (FGM). The blood glucose sensor 120 includes processing circuitry 122, a blood glucose sensor 124, and a communication device 126. The processing circuitry 122 may include any processing circuitry that receives and processes data signals and outputs results as a result in the form of one or more electrical signals. The processing circuitry 122 may include a processor (similar to processor 112), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), hardwired logic, or a combination thereof. The blood glucose sensor 124 includes any sensor capable of extracting and / or analyzing analytes (e.g., blood or interstitial fluid) from the body of a person with diabetes to measure and / or record that person's blood glucose levels. The communication device 126 allows the blood glucose sensor 120 to communicate with a computing device 110 via a communication link 101 and relays the measured blood glucose levels to the computing device 110.

[0019] Sensor 130 exemplarily includes any sensor configured to measure at least one of a person's physiological or medical parameters, the person's geographic or physical location, and the person's movement, and to transmit the measured information to computing device 110. Sensor 130 may be a wearable and / or portable sensor configured to be worn, attached, or carried by a person with diabetes. Examples of wearable sensors include smartwatches (e.g., Apple Watch®, Fitbit®, etc.), heart rate monitors, cardiac monitors, etc. Sensor 130 may also be an implantable sensor implanted within the human body. In yet another embodiment, sensor 130 may be neither wearable nor implantable, but may be configured to observe a person with diabetes. For example, sensor 130 may be a pressure and / or motion sensor placed on top, under, or inside a person's bed and configured to record the time the person sleeps in bed and information about the person's sleep quality. Sensor 130 may also include a camera placed in a person's home, office, car, or other location and configured to observe the person's behavior, presence, and / or appearance. Sensor 130 includes processing circuitry 132, one or more sensors 134 configured to measure the aforementioned reference information about a person, and communication device 136. Processing circuitry 132 may include any of the possible types of processing circuitry described above. Communication device 136 allows sensor 130 to communicate with computing device 110 via communication link 102 and relays measured information to computing device 110.

[0020] Drug delivery device 140 exemplarily includes any device configured to deliver a dose of insulin to a person with diabetes, measure and / or record the time and amount of the delivered dose, and transmit that information to computing device 110. The term "insulin" refers to one or more therapeutic agents, including insulin, insulin analogs (such as lispro insulin or glargine insulin), and insulin derivatives. Such a device is operated by a patient, caregiver, or healthcare professional in a manner generally as described herein to deliver insulin to a person. Insulin delivered by device 140 may be accompanied by one or more adjuvants. Drug delivery device 140 may be configured as a reusable device, capable of being refilled with insulin once its stored insulin is depleted, or it may be configured as a disposable device designed to be discarded and replaced once its stored insulin is depleted. Drug delivery device 140 includes processing circuitry 142, a dose detection sensor 144, and a communication device 146. Processing circuitry 142 may include any possible type of processing circuitry described above. Dose detection sensor 144 may include any suitable sensor for detecting and / or recording the time and amount of the delivered dose. The communication device 146 allows the drug delivery device 140 to communicate with the computing device 110 via the communication link 103.

[0021] Server 160 exemplarily includes any computing device configured to receive information about persons with diabetes from computing device 110 via network 150, process said information, and optionally send responses, notifications, or instructions to computing device 110 in response to said information. Server 160 includes processing circuitry 162, memory 164, and communication device 166. Processing circuitry 162 may include any of the possible types of processing circuitry described above. Processing circuitry 162 may execute software and / or firmware stored in memory 164 of server 160. The software / firmware code contains instructions that, when executed by processing circuitry 162, perform the functions described herein. Memory 164 may also be configured to store information about one or more persons with diabetes, such as biographical information and / or medical information (e.g., insulin dosing records, medical history, etc.). Information received from or sent to computing device 110 may also be stored in memory 164. Memory 164 may include any of the possible types of memory described above. Communication device 166 allows server 160 to communicate with computing device 110 via communication link 105, network 150 and communication link 104.

[0022] In some embodiments, system 100 can be modified by omitting one or more of the glucose sensing device 120, sensor 130, and drug delivery device 140. For example, instead of using the connected glucose sensing device 120 as shown in the figure, a user of system 100 can instead use other methods (e.g., using an unconnected glucose sensing device, such as a BGM) to measure or estimate his / her own blood glucose level, and then manually input the measured blood glucose level and measurement time into computing device 110. As another example, instead of using the connected drug delivery device 140 as shown in the figure, a user of system 100 can instead manually inject himself / herself using an unconnected delivery device (e.g., a syringe), and then manually input the time and amount of insulin dose taken. In some embodiments, computing device 110 can operate solely based on glucose measurement and insulin dosing information, and therefore the type of information provided by sensor 130 may not be required. Even if computing device 110 does use this additional information, the user can manually input the information directly into computing device 110. In another embodiment, computing device 110 can operate without being connected to any network 150 and / or server 160.

[0023] In other embodiments, system 100 can be modified by adding components. For example, server 160 can be configured as multiple networked servers 160 collaboratively processing information. Such a networked server configuration can be referred to as a server "cloud" performing the functions described herein. One or more servers 160 can communicate with multiple computing devices 110 via network 150, and each computing device 110 can optionally be connected to one or more glucose sensing devices 120, one or more sensors 130, and one or more drug delivery devices 140.

[0024] This document describes system 100 as implementing detector 114 within processor 110. As described in further detail below, detector 114 is configured to detect one or more indications that a user has missed an insulin bolus or needs an bolus. Detector 114 may be configured to detect such indications of missed insulin doses using different sensitivity levels, as described herein. Detector 114 may include software instructions and / or logic stored in memory 116 and executed by processor 112 to implement the functions described herein. However, in other embodiments, detector 114 may take other forms. For example, the functions performed by detector 114 may be implemented at least in part by dedicated hardware and / or firmware, such as a separate dedicated processor and / or processing circuitry. In some embodiments, the functions performed by detector 114 may also be implemented, in whole or in part, on server 160, glucose sensing device 120, sensor 130, and / or drug delivery device 140.

[0025] Figure 2 An exemplary process 200 for detecting a missed insulin bolus dose, according to some embodiments, is described. For ease of explanation, Figure 2 (as well as Figures 4 to 10 The steps described in ( ) are described below as being implemented on computing device 110. However, it should be understood that the following description regarding Figure 2 and Figures 4 to 10 Any of the steps described can be implemented on the computing device 110, server 160, glucose sensing device 120, sensor 130, and / or drug delivery device 140. In some embodiments, some or all of the above-described devices can cooperate to implement process 200; for example, one device can perform some steps while another device performs other steps, or one device can perform some steps by means of intermediate calculations and / or data provided by another device.

[0026] Process 200 begins at step 202, where computing device 110 receives at least one first signal representing multiple blood glucose measurements and insulin dosing information for a person with diabetes (also referred to herein as the "user"). Blood glucose measurements can be received from blood glucose sensing device 120 via communication link 101 or can be manually entered by the user. Insulin dosing information includes the time of administration of at least one insulin bolus administered by the user, and optionally may also include the amount of insulin taken. Insulin dosing information can be received from drug delivery device 140 via communication link 103 or can be manually entered by the user. After receiving this information, process 200 branches to step 204.

[0027] In step 204, the computing device 110 uses a detector (e.g., detector 114) to analyze multiple blood glucose measurements and insulin dosing information to determine whether the at least one first signal indicates that the user missed an insulin bolus after a meal event. As used herein, “missed bolus” can include situations where a diabetic user who requires insulin at a mealtime has ingested food (this food intake is also referred to herein as a “meal event”) without administering an insulin bolus to compensate for an increase in blood glucose levels caused by or expected to be caused by the ingested food. A “meal event” or “food” can include any type of food, beverage, or meal expected to cause an increase in the user’s blood glucose levels, including but not limited to breakfast, lunch, dinner, any snacks, and / or any beverages.

[0028] The detector can be configured to detect missed injections based on different sensitivity levels. Figure 3An illustrative graph is plotted to illustrate the trade-offs between different levels of sensitivity for detecting missed boluses. When the detector is configured to detect missed boluses at a low sensitivity level, it will only output a possible missed bolus indication if it is relatively certain that the user's blood glucose and insulin dosing information indicates that the user has missed a bolus. This means the detector may exhibit a relatively low false positive rate; for example, the frequency or probability of the detector indicating that the user has missed a bolus when the user has not actually missed one may be relatively low. However, a low sensitivity level may also mean that the detector is more likely to exhibit a relatively high false negative rate; for example, the frequency or probability of the detector indicating that the user has not missed a bolus when the user has actually missed one may be relatively high. On the other hand, when the detector is configured to detect missed boluses at a high sensitivity level, it will output a possible missed bolus indication, even if it is relatively less certain that the user has missed a bolus. This means that the high sensitivity detector may exhibit a relatively high false positive rate and a relatively low false negative rate compared to the low sensitivity detector.

[0029] Different sensitivity levels can lead to different advantages and disadvantages. For example, a low-sensitivity detector is less likely to bother its user with false alarms, thus reducing alarm fatigue and the possibility of the user stopping using the detector altogether. On the other hand, a low-sensitivity detector may not be able to detect missed boluses completely. Even if a low-sensitivity detector does detect a missed bolus, it may only detect it after a relatively long period of time. This means that by the time the user receives a notification about a missed bolus, their blood sugar level may have already risen to an undesirable high level, and it may be too late for the user to effectively prevent or mitigate the high blood sugar fluctuations.

[0030] Conversely, a high-sensitivity detector may lead to more false alarms, increasing the likelihood that users will become overwhelmed and stop using the detector altogether. However, a high-sensitivity detector will be more likely to detect missed boluses, and / or detect them early enough to allow users to prevent or mitigate high blood sugar fluctuations. Therefore, there is no single sensitivity level suitable for all users or all situations. The optimal sensitivity level can vary depending on user preferences or the user's changing context, environment, and / or conditions.

[0031] like Figure 3As shown, the detector can be configured to use a continuous range of sensitivity settings to detect missed pushes. Changing the sensitivity settings on the detector can include adjusting one or more numerical thresholds in the detector's pattern detection algorithm, or adding, modifying, or deleting one or more steps in the detector's algorithm. In some cases, changing the sensitivity settings can include switching from one type of algorithm to a completely different type of algorithm. In some embodiments, the detector can be configured to use a discrete plurality of sensitivity settings to detect missed pushes, such as two (e.g., "high" and "low"), three (e.g., "high", "medium", and "low"), four, five, or more discrete sensitivity levels. For illustration, several exemplary methods for detecting missed pushes and for adjusting the sensitivity of the methods are discussed in more detail below.

[0032] Back Figure 2 In step 206, when the detector determines that the at least one first signal indicates that the user has missed an insulin bolus, the computing device 110 generates a user notification. This notification may take the form of any mechanism that might attract the user's attention, such as an audible tone or message, a light indicator, a pop-up message on the user interface of the computing device 110, a telephone call, a text message, an email, a tactile indicator, etc. In some embodiments, the computing device 110 may also record in its memory that it sent a notification to the user (or transmitted such a notification to the server 106) to aid in later analysis of the frequency, timing, and / or circumstances of previous missed bolus notifications, or the user's response to missed bolus notifications. The user notification may also take the form of notification to users other than the user who may have missed the bolus (e.g., to a friend, family member, or care provider of the person with diabetes). Such notification to certain other users may replace or supplement the notification to the user with diabetes. In some embodiments, if the detector determines that the at least one first signal does not indicate that the user has missed an insulin bolus, the computing device 110 may prevent the generation of a user notification.

[0033] In step 208, the computing device 110 receives context information about the user and reconfigures the detector based on the received context information to detect missed injections according to a second sensitivity level (different from the first sensitivity level). See below for reference. Figures 4 to 10 Examples of different types of contextual information are discussed in further detail. A second sensitivity level can be more sensitive or less sensitive than the first sensitivity level.

[0034] In step 210, the computing device 110 receives at least one second signal representing additional blood glucose measurement and additional insulin dosing information on behalf of the user. This additional blood glucose measurement and additional insulin dosing information may be acquired, recorded, and / or received at a later time than the blood glucose measurement and insulin dosing information received in step 202.

[0035] In step 212, the computing device 110 uses a detector configured with a second sensitivity level to analyze additional blood glucose measurements and additional insulin dosing information to determine whether the at least one second signal indicates that the user missed an insulin bolus after a meal event. This analysis is performed using the second sensitivity level.

[0036] In this way, process 200 allows computing device 110 to dynamically change the sensitivity level of its detector based on user preferences, context, environment, and / or conditions determined such as received context information. This ability to dynamically change the sensitivity level allows process 200 to adapt to user preferences, context, environment, and / or conditions, and apply the most appropriate sensitivity level in each case. Process 200 is therefore an improvement on previously known omission injection detection algorithms that use a single sensitivity level regardless of user preferences, context, environment, and / or conditions.

[0037] Figure 4 An exemplary process 400, according to some embodiments, is described for implementing step 208 of process 200, for example, for receiving contextual information about a user and reconfiguring the sensitivity level of a detector based on that contextual information. Process 400 begins at step 402, where computing device 110 receives manual user input instructing the detector to use a different sensitivity level. This manual input may instruct the user to simply increase or decrease the sensitivity level, or may specify a precise sensitivity level to switch to. In step 404, computing device reconfigures the detector to use a sensitivity level different from the first sensitivity level. In this way, process 400 allows the user to manually specify their preference for what sensitivity level to use.

[0038] Figure 5An exemplary process 500 for implementing step 208 in process 200, according to some embodiments, is depicted. Process 500 begins at step 502, where computing device 110 receives user feedback indicating that a detector has detected one or more erroneous missed insulin boluses. An erroneous missed insulin bolus may include a situation where the detector has indicated that the user has missed an insulin bolus when the user has not actually missed an insulin bolus. In other words, an erroneous missed insulin bolus may include a situation where one or more previously delivered missed bolus notifications were delivered incorrectly. This feedback may be received in a variety of different ways. For example, each notification of a possible missed bolus delivered in step 206 of process 200 may be accompanied by an inquiry to the user regarding whether the notification is correct (i.e., the user did miss a bolus) or incorrect (i.e., the user did not miss a bolus). The user feedback received in step 502 may include one or more user responses to such inquiries, and / or summary data derived from such user responses (e.g., "50% of the missed bolus notifications delivered last week were incorrect"). In some embodiments, such user feedback may be received in response to survey questions managed separately from any individual missed bolus notification. The survey questions may ask the user to indicate his / her perception of the accuracy of the missed bolus detector, the degree to which he / she is annoyed or fatigued by excessive alerts, or the user's level of concern regarding the consequences of a missed insulin bolus. In yet another embodiment, the user feedback received in step 502 may be collected from users other than the user currently operating computing device 110. For example, user feedback may be received by server(s)106 from users operating other computing devices 110. In some embodiments, these users may represent the general user group using system 100, or may be selected or filtered to represent specific users currently operating computing device 110 (e.g., similar in biographical details, medical history, geographic location, culture, etc.).

[0039] In step 504, computing device 110 reconfigures itself based on the user feedback to use a different sensitivity level (higher or lower). For example, if the user feedback received in step 502 indicates that the user operating computing device 110 (or the group of users communicating with server(s) 106) is annoyed or fatigued by too many alarms, computing device 110 may reconfigure the detector to use a lower sensitivity level. Conversely, if the user feedback indicates that one or more users would prefer a higher sensitivity setting (e.g., if one or more users indicate they are concerned about potentially missing bets, or are more likely to miss bets), computing device 110 may reconfigure the detector to use a higher sensitivity level.

[0040] Figure 6Another exemplary process 600 for implementing step 208 in process 200, according to some embodiments, is depicted. Process 600 begins at step 602, where computing device 110 receives and analyzes user blood glucose responses from previously missed insulin boluses. These blood glucose responses may be collected or recorded in response to a missed bolus notification output by a missed bolus detector, or in response to user confirmation that he / she missed a bolus. Blood glucose responses may be received from blood glucose sensing device 120, manually entered by the user, or received via other means. These blood glucose responses may be analyzed to determine how severe and / or how quickly the user's blood glucose response to a missed insulin bolus is.

[0041] In step 604, the computing device 110 reconfigures the detector to use different sensitivity levels (higher or lower) based on the analysis of these blood glucose responses. If the user's blood glucose response indicates that the user's blood glucose level rises by a relatively large amount or rises relatively quickly after the missed bolus, the health consequences of the missed bolus may be relatively severe for that particular user. The computing device 110 may accordingly reconfigure the detector to use a higher sensitivity level. Conversely, if the user's blood glucose response indicates that the user's blood glucose level rises by a relatively small or moderate amount, or tends to rise relatively slowly, the health consequences of the missed bolus may be relatively mild for that particular user. The computing device 110 may accordingly reconfigure the detector to use a lower sensitivity level, thereby avoiding unnecessary annoyance or warning to the user.

[0042] Figure 7 Another exemplary process 700 for implementing step 208 in process 200, according to some embodiments, is depicted. Process 700 begins at step 702, where computing device 110 receives the current midday time. The current time can be determined using an onboard clock included within or attached to computing device 110, or can be received from a source external to computing device 110 (e.g., via network 105). In step 704, computing device 110 reconfigures the detector based on the midday time to use a different sensitivity level (higher or lower). Adjusting the sensitivity level based on the midday time is based on the understanding that the rate and / or severity of blood glucose fluctuations due to missed bolus injections can vary depending on the time of day. For example, breakfast tends to contain higher carbohydrates and a higher glycemic index, while lunch and dinner tend to contain proteins and fats that slow carbohydrate absorption. Therefore, blood glucose fluctuations due to missed bolus injections at breakfast time may result in more severe and / or faster hyperglycemic fluctuations compared to missed bolus injections at lunch or dinner time. Therefore, the computing device 110 can configure the detector to use a higher sensitivity level during the morning hours (e.g., 7 a.m. to 11 a.m.) and a lower sensitivity level during other times of the day.

[0043] In some embodiments, computing device 110 and / or server 106 can analyze the timing of a patient's past blood glucose fluctuations to infer one or more times of the day when the user is most likely to miss a bolus. For example, by analyzing a user's blood glucose levels over a period of several weeks or months, computing device 110 and / or server 106 can determine that the user typically does not remember to take boluses at regular mealtimes but tends to eat sugary snacks around 3 or 4 p.m. and forget to take boluses. Therefore, computing device 110 can configure the detector to use a higher sensitivity level during the 3 or 4 p.m. time period to compensate for the observed user habits and tendencies.

[0044] In some embodiments, process 700 can be generalized in addition to reconfiguring the detector based on the current daytime. For example, process 700 can be modified to reconfigure the detector based on the midday of the week or based on the current date, replacing or supplementing the current daytime. The current date can be received from an onboard calendar function included in or attached to computing device 110, or from a source external to computing device 110 (e.g., via network 105). The midday of the week may be relevant because users may be more likely to forget to place bets on certain days of the week compared to other days. For example, users may have relatively little difficulty remembering to place bets during weekdays but may tend to forget to place bets during weekends—therefore, computing device 110 can configure the detector to use a higher sensitivity level on weekends and a lower sensitivity level during weekdays. Similarly, users may be more likely to forget to place bets during national holidays (e.g., Christmas)—therefore, if the current date corresponds to a national holiday, festival, or event, computing device 110 can configure the detector to use a higher sensitivity level.

[0045] Figure 8Another exemplary process 800 for implementing step 208 in process 200, according to some embodiments, is depicted. Process 800 begins at step 802, wherein computing device 110 receives at least one of physiological data and accelerometer data belonging to a user from sensor 130. The physiological data may indicate, or be derived from, the user's heart rate, blood pressure, blood oxygen saturation level (SpO2), muscle oxygen saturation level (SmO2), electrocardiogram (ECG), respiratory rate, and body temperature, while the accelerometer data may indicate or be derived from the user's movement. Computing device 110 may draw numerous conclusions about the user's context, environment, or conditions based on the physiological and / or accelerometer data. For example, if the user exhibits an elevated heart rate and / or is engaged in strenuous exercise, the user may be exercising. If the user exhibits an elevated heart rate and / or blood pressure without strenuous exercise, the user may be experiencing stress. If the user's body temperature is elevated, the user may be ill. If the user exhibits consistent or regular movement without an elevated heart rate and / or blood pressure, the user may be walking or driving. If a user exhibits a low heart rate and little or no movement, the user may be asleep.

[0046] In step 804, the computing device 110 reconfigures the detector to use different sensitivity levels (higher or lower) based on the received physiological and / or accelerometer data. For example, if the user's physiological and / or accelerometer data indicates that the user is exercising, the computing device 110 may reconfigure the detector to use a lower sensitivity level, or even deactivate the detector entirely. If the user's physiological data indicates that the user may be ill or stressed, the detector may appropriately increase or decrease its sensitivity level. If the user's motion / accelerometer data indicates that the user may be walking or driving, the computing device 110 may reconfigure the detector to use a lower sensitivity level.

[0047] Figure 9Another exemplary process 900 is depicted for implementing step 208 in process 200, according to some embodiments. Process 900 begins at step 902, where computing device 110 receives input instructing a user to prepare for exercise. This input may be received by a mobile application that helps diabetic patients prepare for and perform exercise safely while mitigating or minimizing blood sugar fluctuations. Additional details of such an application are further described in the following reference, the disclosure of which is expressly incorporated herein by reference in its entirety: U.S. Provisional Patent No. 62 / 827,350, filed April 1, 2019, entitled “METHODS AND APPARATUS FOR INSULIN DOSING GUIDANCE AND DECISIONSUPPORT FOR DIABETIC PATIENT EXERCISE”.

[0048] When a user is preparing for exercise, the user may intentionally limit or omit his / her insulin dose because he / she expects the upcoming exercise to lower his / her blood sugar level. Therefore, in step 904, if the input indicates that the user is preparing for exercise in the near future, the computing device 110 reconfigures the detector to use a lower sensitivity level (or disables the detector completely).

[0049] Figure 10 Another exemplary process 1000 for implementing step 208 in process 200, according to some embodiments, is depicted. Process 1000 begins at step 1002, wherein computing device 110 receives input indicating the user's current physical and / or geographic location. In some embodiments, a GPS locator installed within computing device 110 or sensor 130 may be used to determine the user's current physical and / or geographic location. Additionally or alternatively, the user's location may be determined by using communication means in computing device 110 to detect wireless signals from one or more wireless access points and / or cellular nodes and comparing the received signals, the received signal strength, and / or the direction of the received signals with a database listing the locations of multiple wireless access points and / or cellular nodes.

[0050] In step 1004, the computing device 110 reconfigures the detector based on the user's location to use a different sensitivity level (higher or lower). This can be achieved in various ways.

[0051] For example, if the user's location corresponds to a location where the user might eat food, such as a restaurant, cafeteria, or the user's kitchen or living room, the detector can be configured to increase its sensitivity level. To make this determination, computing device 110 can be configured to access a database storing information about different locations (locally stored on device 110 or remotely stored on server 106). The database can designate certain types of locations (e.g., restaurants, cafeterias, homes) as areas where the user is likely to consume food, and other types of locations (e.g., parks / wilderness, highways, laboratories, factories) as areas where the user is unlikely to consume food. In some embodiments, computing device can be configured to compare the user's GPS coordinates with the database to determine whether the user is likely to consume food at their current location.

[0052] In some embodiments, while computing device 110 may know the coordinates of the user's current location, it may not have access to contextual and / or geospatial data indicating what type of location those coordinates correspond to (e.g., what address the coordinates correspond to, or whether the coordinates correspond to a restaurant, office building, or the user's home, etc.). This could be because the data is unavailable; computing device 110 and / or server 106 do not have sufficient processing power or memory to store, access, or process such contextual data; or because the computing device's access to that data is restricted to protect the user's privacy. Even in these cases, computing device 110 and / or server 106 can still adjust the detector's sensitivity level by tracking the locations where the user has eaten food. If the user's current location corresponds to a location where the user has eaten food (possibly more than once), computing device 110 can be configured to increase the sensitivity level. In some embodiments, device 110 and / or server 106 can also track the locations where other users have eaten food at the user's current location. If the user's current location corresponds to a location where the user has not eaten food but many other users have eaten food, device 110 can still be configured to increase the detector's sensitivity level.

[0053] In other embodiments, computing device 110 may use the user's location to determine whether the user is moving or traveling. For example, if the user's GPS location is constantly changing, computing device 110 may infer that the user is driving, walking, or otherwise traveling. In these cases, computing device 110 may use a lower sensitivity level for the detector because the user is unlikely to be eating while moving or traveling.

[0054] Exemplary method for detecting missed injections

[0055] Several exemplary and illustrative methods for detecting missed boluses based on blood glucose measurements and insulin dosing information will now be discussed.

[0056] Blood glucose elevation threshold method

[0057] An exemplary method for detecting missed insulin boluses, referred to herein as the "glucose elevation threshold" method, determines that a user may have missed an insulin bolus if the following conditions are met:

[0058] (i) The user's blood glucose level within the predetermined blood glucose consideration time window (T) at the current time. G (e.g., within 5 to 10 minutes) increases exceeding the maximum permissible blood glucose increase threshold (ΔG). max (e.g., 20 to 60 mg / dL); and

[0059] (ii) The user's scheduled time period (T) at the current time B For example, if no insulin bolus was administered within 2 hours.

[0060] The parameter ΔG can be adjusted. max T G and T B This can be used to make the blood glucose threshold method more or less sensitive. For example, increasing the maximum permissible blood glucose threshold (ΔG) max This will reduce sensitivity, and thus reduce ΔG. max It will increase sensitivity. Increasing blood glucose levels should be considered in relation to the time window (T). G This will increase sensitivity and decrease T. G This will reduce sensitivity. Increasing the injection time window (T) will reduce sensitivity. B This will reduce sensitivity, and thus reduce T. B It will increase sensitivity.

[0061] Rate of Change ("ROC") Threshold Method

[0062] Another exemplary method for detecting missed insulin boluses, referred to herein as the "glucose ROC threshold" method, determines that a user may have missed an insulin bolus if the following conditions are met:

[0063] (i) The user's blood glucose level shows a value greater than the maximum permissible rate of change of blood glucose (ROC). max For example, the rate of change (ROC) of 2 mg / dL / hr G );as well as

[0064] (ii) The user's scheduled time period (T) at the current time B For example, if no insulin bolus was administered within 2 hours.

[0065] ROC GThe ROC can be provided or calculated by some commercially available continuous glucose monitors (CGMs), such as the G6 CGM sensor manufactured and sold by DexcomM. For example, if a glucose sensor records three consecutive glucose readings, each no more than 5 minutes apart from its nearest neighbor, the ROC can be calculated by dividing the difference between the last and first glucose readings by the amount of time elapsed between the first and last glucose readings. G You can also use the method for calculating ROC. G Other methods or devices. Because the user retains control over whether to take action on a missed injection notification, and no insulin is infused without the user's knowledge or consent, the methods or devices used do not need to be as precise or specific as those used in fully automated insulin infusion systems (e.g., automated insulin pumps).

[0066] The ROC parameter can be adjusted. max and T B This can be used to make the blood glucose ROC threshold method more or less sensitive. For example, increasing the ROC... max It will reduce sensitivity, and thus reduce ROC. max This will increase sensitivity. Increasing the injection time window (T) will help. B This will reduce sensitivity, and thus reduce T. B It will increase sensitivity.

[0067] Absolute blood glucose level threshold method

[0068] Another exemplary method for detecting missed insulin boluses, referred to herein as the "absolute blood glucose level threshold" method, determines that a user may have missed an insulin bolus if the following conditions are met:

[0069] (i) The user's blood glucose level exceeds the absolute blood glucose level threshold (G). max (e.g., 180 mg / dL); and

[0070] (ii) During the pre-determined betting consideration period (T) B For example, if no insulin bolus was administered within 2 hours.

[0071] Parameter G can be adjusted max and T B This can be used to make the absolute blood glucose level threshold method more or less sensitive. For example, increasing the maximum permissible blood glucose threshold (G) max This will reduce sensitivity, and reduce G. max This will increase sensitivity. Increasing the injection time window (T) will also improve sensitivity. B This will reduce sensitivity, and thus reduce T. B It will increase sensitivity.

[0072] In some embodiments, reconfiguring the detector to detect missed boluses based on a second sensitivity level different from the first sensitivity level may also include switching between methods. For example, in some embodiments, the glucose elevation threshold method and the glucose ROC threshold method may be considered "high" or "medium" sensitivity methods. When switching to a "low" sensitivity level, the detector may switch the algorithm entirely to the absolute glucose level threshold method.

[0073] The terms “first,” “second,” “third,” etc., used in both the specification and the claims, are intended to distinguish similar elements and not necessarily to describe a sequence or chronological order. It should be understood that such terms are interchangeable where appropriate (unless explicitly disclosed otherwise), and the embodiments of this disclosure described herein can operate in other orders and / or arrangements than those described or shown herein.

[0074] While the invention has been described with exemplary design, further modifications can be made to the invention within the spirit and scope of this disclosure. Therefore, this application is intended to cover any variations, uses, or modifications of the invention using its general principles. Furthermore, this application is intended to cover such deviations from this disclosure within the scope of known or conventional practice in the field to which this invention pertains.

[0075] This disclosure describes various aspects, including but not limited to the following:

[0076] 1. A method for detecting missed insulin boluses, comprising:

[0077] At a computing device, at least one first signal is received representing multiple blood glucose measurements of a user and insulin dosing information of the user, the insulin dosing information including at least one insulin bolus administration time.

[0078] The detector of the computing device configured with a first sensitivity level is used to analyze the plurality of blood glucose measurements and the insulin dosing information to determine whether the at least one first signal indicates that the user missed an insulin bolus after a meal event.

[0079] A user notification is generated when the detector determines that at least one first signal indicates that the user has missed an insulin bolus.

[0080] Receive context information about the user at the computing device;

[0081] The detector is reconfigured at least based on the context information to detect missed injections according to a second sensitivity level that is different from the first sensitivity level;

[0082] At the computing device, at least one second signal representing additional blood glucose measurement and additional insulin dosing information of the user is received; and

[0083] The detector of the computing device configured with a second sensitivity level is used to analyze the additional blood glucose measurement and the additional insulin dosing information to determine whether the at least one second signal indicates that the user missed an insulin bolus after a meal event.

[0084] 2. The method according to aspect 1, wherein when the detector is configured according to a second sensitivity level, the detector exhibits a different false positive rate than when the detector is configured according to a first sensitivity level.

[0085] 3. The method according to any one of aspects 1 to 2, wherein the context information includes manual user input instructing the detector to use a sensitivity level different from a first sensitivity level.

[0086] 4. The method according to any one of aspects 1 to 3, wherein the context information includes manual user input indicating that the detector has detected an erroneous missed insulin bolus.

[0087] 5. The method according to any one of aspects 1 to 4, wherein the context information includes the current daytime.

[0088] 6. The method according to any one of aspects 1 to 5, wherein the context information includes data measured by wearable sensors worn by the user, wherein the data includes at least one of physiological data and accelerometer data.

[0089] 7. The method according to aspect 6, wherein the physiological data includes at least one of the user's heart rate, the user's blood pressure, the user's electrocardiogram (ECG), the user's blood oxygen saturation level (SpO2), the user's muscle oxygen saturation level (SmO2), the user's respiratory rate, and body temperature.

[0090] 8. The method according to any one of aspects 6 to 7, wherein the context information indicates a user's exercise event.

[0091] 9. The method according to any one of aspects 6 to 8, wherein the context information indicates the user's sleep events.

[0092] 10. The method according to any one of aspects 1 to 9, wherein the context information includes the user's geographical location.

[0093] 11. The method according to aspect 10, wherein the context information further includes data from a database indicating whether the user previously ingested food at the geographic location.

[0094] 12. The method according to any one of aspects 10 to 11, wherein the context information further includes data from a database indicating whether other users have previously ingested food at the geographic location.

[0095] 13. The method according to any one of aspects 10 to 12, wherein the context information further includes data that designates the user's current location as a location where people may ingest food.

[0096] 14. The method according to any one of aspects 1 to 13, wherein the context information includes data indicating whether the user is driving.

[0097] 15. The method according to any one of aspects 1 to 14, wherein the plurality of blood glucose measurements are recorded by at least one of a blood glucose monitor (BGM), a continuous glucose monitor (CGM), and a flash glucose monitor (FGM).

[0098] 16. The method according to any one of aspects 1 to 15, wherein the first sensitivity level is more sensitive to missed injections than the second sensitivity level.

[0099] 17. The method according to aspect 16, wherein when the detector is configured to detect a missed injection based on a first sensitivity level, the detector is configured to detect a missed injection if: the user's blood glucose level increases by more than a maximum permissible blood glucose increase threshold within a predetermined blood glucose consideration time window at the current time, and the user does not administer an insulin injection within the predetermined injection consideration time period at the current time.

[0100] 18. The method according to aspect 17, wherein reconfiguring the detector to detect missed injections according to a second sensitivity level includes changing at least one of the following settings: (i) increasing the maximum allowable increase threshold for blood glucose, (ii) shortening the blood glucose consideration time window, and (iii) extending the injection consideration time period.

[0101] 19. The method according to aspect 17, wherein reconfiguring the detector to detect missed injections according to a second sensitivity level includes reconfiguring the detector to detect missed injections when: the user's blood glucose level exceeds a maximum blood glucose threshold, and the user has not administered an insulin injection during a second predetermined injection consideration period at the current time.

[0102] 20. A mobile device for detecting missed insulin boluses, the device comprising:

[0103] A communication device configured to establish one or more communication links with at least one of a blood glucose sensing device, a wearable sensor, and a drug delivery device;

[0104] Memory that stores computer-executable instructions; and

[0105] A processor configured to execute the instructions to implement the method according to any one of aspects 1 to 19.

[0106] 21. The mobile device according to aspect 20, wherein the communication device is further configured to establish one or more communication links with a remote server via a network.

[0107] 22. A non-transitory computer-readable medium storing computer-executable instructions that, when executed by one or more processors, are operable to cause the one or more processors to perform the method according to any one of aspects 1 to 19.

Claims

1. A method for detecting a missed insulin bolus, comprising: receiving, at a computing device, at least one first signal representative of a plurality of blood glucose measurements of a user and insulin dosing information of the user, the insulin dosing information including an administration time of at least one insulin bolus; analyzing, with a detector of the computing device configured at a first sensitivity level, the plurality of blood glucose measurements and the insulin dosing information to determine whether the at least one first signal indicates that the user missed an insulin bolus after a meal event; generating a user notification when the detector determines that the at least one first signal indicates that the user missed an insulin bolus; receiving, at the computing device, context information about the user; reconfiguring the detector based at least on the context information to detect a missed bolus according to a second sensitivity level that is different from the first sensitivity level, wherein the second sensitivity level is dynamically determined according to a user's preferences, context, environment, and / or conditions determined from the received context information; receiving, at the computing device, at least one second signal representative of additional blood glucose measurements of the user and additional insulin dosing information; and analyzing, with the detector of the computing device configured at the second sensitivity level, the additional blood glucose measurements and the additional insulin dosing information to determine whether the at least one second signal indicates that the user missed an insulin bolus after a meal event.

2. The method of claim 1, wherein, The detector exhibits a different false positive rate when configured according to the second sensitivity level than when configured according to the first sensitivity level.

3. The method of any one of claims 1-2, wherein, The context information includes a manual user input indicating that the detector use a sensitivity level that is different from the first sensitivity level.

4. The method of claim 1, wherein, The context information includes a manual user input indicating that the detector detected a false missed insulin bolus.

5. The method of claim 1, wherein, The context information includes a current time of day.

6. The method of claim 1, wherein, The context information includes data measured by a wearable sensor worn by the user, wherein the data includes at least one of physiological data and accelerometer data.

7. The method of claim 6, wherein, The physiological data includes at least one of a heart rate of the user, a blood pressure of the user, an electrocardiogram (ECG) of the user, a blood oxygen saturation level (Sp02) of the user, a muscle oxygen saturation level (Sm02) of the user, a respiration rate of the user, and a body temperature of the user.

8. The method of claim 6, wherein, The context information indicates an exercise event of the user.

9. The method of claim 1, wherein, The context information indicates a sleep event of the user.

10. The method of claim 1, wherein, The context information includes a geographic location of the user.

11. The method of claim 10, wherein, The context information further includes data from a database indicating whether the user previously consumed food at the geographic location.

12. The method of claim 10, wherein, The context information further includes data from a database indicating whether other users previously consumed food at the geographic location.

13. The method of claim 10, wherein, The context information further includes data designating the current location of the user as a location where people are likely to consume food.

14. The method of claim 10, wherein, The context information includes data indicating whether the user is driving.

15. The method of claim 1, wherein, The plurality of blood glucose measurements are recorded by at least one of a blood glucose meter, a continuous blood glucose monitor, and a flash blood glucose monitor.

16. The method of claim 1, wherein, The first sensitivity level is more sensitive to missed boluses than the second sensitivity level.

17. The method of claim 16, wherein, When the detector is configured to detect missed boluses according to the first sensitivity level, the detector is configured to detect a missed bolus when a blood glucose level of the user increases by more than a maximum allowed blood glucose increase threshold within a predetermined blood glucose consideration time window at a current time and the user has not taken an insulin bolus within a predetermined bolus consideration time period at the current time.

18. The method of claim 17, wherein, Reconfiguring the detector to detect missed boluses according to the second sensitivity level includes changing at least one of the following settings: (i) increasing the maximum allowed blood glucose increase threshold, (ii) shortening the blood glucose consideration time window, and (iii) lengthening the bolus consideration time period.

19. The method of claim 17, wherein, Reconfiguring the detector to detect missed boluses according to the second sensitivity level includes reconfiguring the detector to detect a missed bolus when a blood glucose level of the user exceeds a maximum blood glucose threshold and the user has not taken an insulin bolus within a second predetermined bolus consideration time period at the current time.

20. A mobile device for detecting missed insulin boluses, the device comprising: a communication device configured to establish one or more communication links with at least one of a blood glucose sensing device, a wearable sensor, and a medication delivery device; a memory storing computer executable instructions; and a processor configured to execute the instructions to implement the method of any one of claims 1-19.

21. The mobile device of claim 20, wherein, The communication device is further configured to establish one or more communication links with a remote server via a network.

22. A non-transitory computer-readable medium storing computer executable instructions that, when executed by one or more processors, are operable to cause the one or more processors to implement the method of any one of claims 1-19.

Images