Adaptive Decision Support System

Abstract

Certain aspects of the present disclosure relate to a method for configuring an application with one or more application features. The method includes receiving a request to configure an application for use by a user. The method further includes identifying a goal for the user and identifying classification information associated with the user, where the classification information includes at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication regimen information. The method further includes selecting a group of users based on one or more affinities between the user and the group of users. The method further includes identifying one or more application features based on the user's goal and a correlation between each of a plurality of application features and the goal. The method further includes configuring the application with the one or more application features.

Description

Technical Field

[0001] Incorporation by reference to related applications This application claims the benefit of U.S. Provisional Patent Application No. 63 / 016,784, filed Apr. 28, 2020. The foregoing application is hereby incorporated by reference in its entirety and made a part hereof as if fully set forth herein.

Background Art

[0002] This application generally relates to medical devices such as analyte sensors and to systems and methods for using such devices to provide decision support guidance to patients, caregivers, healthcare providers or other users to assist in improving a patient's health.

[0003] Description of related art Diabetes is a metabolic state related to the production or use of insulin by the body. Insulin is a hormone that enables the body to use glucose for energy or store glucose as fat.

[0004] When a person eats a meal containing carbohydrates, the food is processed by the digestive system and glucose is produced in the person's blood. Blood glucose can be used for energy or stored as fat. The body normally maintains blood glucose levels within a range that provides sufficient energy to support the body's functions and avoids problems that can occur when glucose levels are too high or too low. Regulation of blood glucose levels depends on the production and use of insulin, which regulates the movement of blood glucose into cells.

[0005] Blood glucose levels can rise above the normal range if the body does not produce enough insulin or cannot effectively use the insulin that is present. A condition where blood glucose levels are higher than normal is called "hyperglycemia." Chronic hyperglycemia can lead to many health problems, including cardiovascular disease, cataracts and other eye diseases, neuropathy, and kidney damage. Hyperglycemia can also lead to acute problems, such as diabetic ketoacidosis (a condition in which the body becomes excessively acidic due to the presence of glucose in the blood and ketones produced when the body cannot utilize glucose). A condition where blood glucose levels are lower than normal is called "hypoglycemia." Severe hypoglycemia can lead to an acute crisis that can result in seizures or death.

[0006] Diabetic patients can receive insulin to manage their blood sugar levels. Insulin can be administered, for example, through manual injection using a needle. Wearable insulin pumps are also available. Diet and exercise also affect blood sugar levels.

[0007] Diabetes is sometimes referred to as "type 1" and "type 2." Typically, people with type 1 diabetes can use insulin when it is present, but their bodies cannot produce enough insulin due to problems with the insulin-producing beta cells in the pancreas. People with type 2 diabetes can produce some insulin, but they are "insulin-resistant" due to decreased sensitivity to insulin. As a result, even when insulin is present in the body, it is not used sufficiently by the patient's body to effectively regulate blood glucose levels.

[0008] Managing diabetes can present complex challenges for patients, clinicians, and caregivers, as numerous factors can influence a patient's glucose levels and glucose tendencies. To better manage this condition and assist patients, various diabetes intervention software applications (hereinafter referred to as "applications") have been developed by various providers. However, these applications, which generally run on patients' mobile devices, suffer from very high dropout rates during the first few days, weeks, or months of use. Such high dropout rates are not unique to diabetes intervention applications; they are common to most health-related applications. These health-related applications may include not only intervention applications to assist in the treatment of various diseases, but also applications that help improve a patient's overall health, such as weight loss applications.

[0009] This background is provided to introduce a concise context for the following summary and detailed explanation. This background is not intended to help determine the scope of the claimed subject matter, nor should it be considered to limit the claimed subject matter to implementations that address any or all of the shortcomings or problems presented above. [Overview of the project]

[0010] One embodiment is a system comprising a memory circuit and a processor configured to receive requests for configuring an application for use by a user, the application being at least partially resident on the computing device for managing sensor data generated by a glucose monitoring system associated with the user, the application being at least partially resident on the computing device for managing sensor data generated by a glucose monitoring system associated with the user, the application being to receive, identify the user's goals, identify classification information associated with the user, the classification information being at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information, select a group of users from a pool of users, the selection being based on one or more similarities between the user and the group of users with respect to the identified classification information, select one or more application functions from a plurality of application functions, the identification being based on the user's goals and the correlation between each of the plurality of application functions in a dataset associated with the group of users and the goals, and automatically configure an application using one or more application functions.

[0011] In the above system, the processor being configured to identify a goal includes being configured to receive user input regarding what the user intends to achieve with respect to the user's diabetes, and to convert the user input into a goal based on one or more defined guidelines. In the above system, the processor being configured to convert user input into a goal based on one or more defined guidelines includes being configured to categorize the user into a category based on the guidelines and user-related information, wherein the user-related information includes classification information, and the guidelines indicate the goal of the category, and to select the user's goal based on the categorization.

[0012] In the above system, the processor being configured to identify a goal includes receiving user input regarding what the user intends to achieve with respect to the user's diabetes, and converting the user input into a goal based on information related to the group of users. In the above system, the information related to the group of users includes one or more glucose-related metrics for the group of users. In the above system, the correlation between each of the multiple application functions and the goal includes the correlation between each of the multiple application functions and the achievement of the goal. In the above system, selecting a group of users is further based on the program result metric of each user in the pool of users with respect to the goal. In the above system, the program result metric of the selected group of users is above a threshold program result metric related to the achievement of the goal, and the threshold program result metric related to the achievement of the goal represents a defined minimum amount of positive progress toward achieving the goal.

[0013] In the system described above, the correlation between each of the multiple application features and the goal is based on the number of users who used the feature within a selected group of users, and the behavioral engagement of those users with respect to the feature. In the system described above, the behavioral engagement of each of those users with respect to the application feature is represented by the Behavioral Engagement Metric (BEM) for each of those users with respect to the application feature, the BEM being based on the interaction between each of those users and the feature, the interaction including at least one of the following: how often each of those users interacts with the application feature, how often each of those users ignores the guidance generated by the application feature, the average amount of time each of those users spends interacting with the application feature, or how faithful each of those users' behaviors is to the guidance generated by the application feature. In the system described above, each of one or more application features has a correlation with the goal that exceeds a correlation threshold.

[0014] In the system described above, the processor is further configured to receive a plurality of inputs, the plurality of inputs including a first input containing a glucose measurement related to the user generated by a glucose monitoring system, and a second input indicating the user's behavior with respect to one or more application functions; calculate program result metrics related to a goal based on at least the first input, the program result metrics indicating the degree to which the user has achieved the goal; calculate one or more behavioral engagement metrics (BEMs) for one or more application functions based on the second input, such that a separate BEM is calculated for each of the one or more application functions; identify one or more users in a selected group of users or pool of users who have BEMs similar to the one or more calculated BEMs; identify new application functions not included in one or more functions, based on the fact that the function is associated with a BEM that exceeds a threshold for at least one of the one or more users; and reconfigure the application with the new application functions based on at least one of the one or more BEMs and program result metrics.

[0015] In the above system, each of the one or more BEMs is based on the interaction between the user and a corresponding application function from among one or more application functions, the interaction including at least one of the following: how often the user interacts with the corresponding application function, how often the user ignores the guidance generated by the corresponding application function, the average amount of time the user spends interacting with the corresponding application function, or how faithful the user's actions are to the guidance generated by the corresponding application function. In the above system, the processor being configured to reconfigure the application with a new application function includes the processor being configured to identify a low-performing application function from among one or more application functions that has a corresponding BEM below a threshold, and to replace the low-performing application function from among one or more application functions with a new application function.

[0016] In the system described above, the processor being configured to reconfigure the application with new application functions includes being configured to identify that one or more application functions with lower performance are relevant to the goal and that the program result metrics are below a threshold. In the system described above, each of the one or more application functions includes a function configuration.

[0017] Another aspect is a method for configuring an application using one or more application functions, the method comprising: receiving a request to configure an application for use by a user, the application being at least partially resident on a computing device for managing sensor data generated by a glucose monitoring system associated with the user; identifying the user's goals; identifying classification information associated with the user, the classification information including at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information; selecting a group of users from a pool of users, the selection being based on one or more similarities between the user and the group of users with respect to the identified classification information; identifying one or more application functions from a plurality of application functions, the identification being based on the user's goals and the correlation between each of the plurality of application functions and the goals in a dataset associated with the group of users; and configuring an application using one or more application functions.

[0018] In the above method, identifying a goal further includes receiving user input regarding what the user intends to achieve with respect to their diabetes, and converting the user input into a goal based on one or more defined guidelines. In the above method, the conversion further includes categorizing the user into a category based on the guidelines and user-related information, where the user-related information includes classification information and the guidelines indicate the category's goal, and selecting the user's goal based on the categorization. In the above method, identifying a goal further includes receiving user input regarding what the user intends to achieve with respect to their diabetes, and converting the user input into a goal based on information related to a group of users.

[0019] In the above method, information related to a group of users includes one or more glucose-related metrics for that group of users. In the above method, the correlation between each of the multiple application features and the goal includes the correlation between each of the multiple application features and the achievement of the goal. In the above method, selecting a group of users is further based on the program result metric of each user in the pool of users with respect to the goal. In the above method, the program result metric of the selected group of users is above the threshold program result metric related to the achievement of the goal, and the threshold program result metric related to the achievement of the goal represents the defined minimum amount of positive progress toward achieving the goal.

[0020] In the above method, the correlation between each of the multiple application features and the goal is based on the number of users in a selected group of users who used the feature, and the behavioral engagement of those users with respect to the feature. In the above method, the behavioral engagement of each of those users with respect to the application feature is represented by the behavioral engagement metrics (BEM) of each of those users with respect to the application feature, the BEM being based on the interaction between each of those users and the feature, the interaction including at least one of the following: how often each of those users interacts with the application feature, how often each of those users ignores the guidance generated by the application feature, the average amount of time each of those users spends interacting with the application feature, or how faithful each of those users' behaviors is to the guidance generated by the application feature. In the above method, each of one or more application features has a correlation with the goal that exceeds the correlation threshold.

[0021] The above method further includes receiving multiple inputs, the multiple inputs including a first input containing a glucose measurement related to a user generated by a glucose monitoring system, and a second input indicating the user's behavior with respect to one or more application features; calculating program outcome metrics related to a goal based on at least the first input, the program outcome metrics indicating the degree to which the user has achieved the goal; calculating one or more behavioral engagement metrics (BEMs) for one or more application features based on the second input, such that a separate BEM is calculated for each of the one or more application features; identifying one or more users in a selected group of users or pool of users who have BEMs similar to the one or more BEMs calculated; identifying new application features not included in one or more features, on the basis that the feature is associated with a BEM above a threshold for at least one of the one or more users; and reconfiguring the application with the new application features based on at least one of the one or more BEMs and program outcome metrics.

[0022] In the above method, each BEM of one or more BEMs is based on an interaction between a user and a corresponding application function from one or more application functions, the interaction including at least one of the following: how often the user interacts with the corresponding application function, how often the user ignores the guidance generated by the corresponding application function, the average amount of time the user spends interacting with the corresponding application function, or how faithful the user's behavior is to the guidance generated by the corresponding application function. In the above method, reconfiguring the application with a new application function includes identifying a low-performing application function from one or more application functions that has a corresponding BEM below a threshold, and replacing the low-performing application function from one or more application functions with a new application function.

[0023] In the method described above, reconfiguring the application with new application features includes identifying that one or more underperforming application features are relevant to the goal, and identifying that the program result metrics are below a threshold. In the method described above, each of the one or more application features includes feature settings.

[0024] Another aspect is a non-temporary computer-readable medium storing instructions that, when executed by a processor, cause a computing system to perform a method for configuring an application using one or more application functions, wherein the method includes receiving a request to configure an application for use by a user, the application being at least partially resident on the computing device for managing sensor data generated by a glucose monitoring system associated with the user, identifying the user's goals, identifying classification information associated with the user, the classification information including at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information, selecting a group of users from a pool of users, the selection being based on one or more similarities between the user and the group of users with respect to the identified classification information, identifying one or more application functions from a plurality of application functions, the identification being based on the user's goals and the correlation between each of the plurality of application functions and the goals in a dataset associated with the group of users, and configuring an application using one or more application functions.

[0025] In the above medium, identifying a goal further includes receiving user input regarding what the user intends to achieve with respect to the user's diabetes, and converting the user input into a goal based on one or more defined guidelines. In the above medium, converting includes categorizing the user into a category based on the guidelines and information related to the user, where the information related to the user includes classification information and the guidelines indicate category goals, and further includes selecting the user's goal based on the categorization. In the above medium, identifying a goal further includes receiving user input regarding what the user intends to achieve with respect to the user's diabetes, and converting the user input into a goal based on information related to the user's group. In the above medium, the information related to the user's group includes one or more glucose-related metrics of the user's group.

[0026] In the above medium, the correlation between each of the plurality of application functions and the goal includes the correlation between each of the plurality of application functions and the achievement of the goal. In the above medium, selecting the user's group is further based on each program result metric in the pool of users regarding the goal. In the above medium, the program result metrics of the selected user's group exceed the threshold program result metrics related to the achievement of the goal, and the threshold program result metrics related to the achievement of the goal indicate the defined minimum amount of positive progress towards the achievement of the goal.

[0027] In the above medium, the correlation between each of the application features and the objective is based on the number of users who used the feature and the behavioral engagement of those users with respect to the feature. In the above medium, the behavioral engagement of each of those users with respect to the application feature is represented by the Behavioral Engagement Metric (BEM) for each of those users with respect to the application feature, the BEM being based on each of those users' interactions with the feature, the interactions including at least one of the following: how often each of those users interacts with the application feature, how often each of those users ignores the guidance generated by the application feature, the average amount of time each of those users spends interacting with the application feature, or how faithful each of those users' behaviors are to the guidance generated by the application feature.

[0028] In the above medium, each of the one or more application functions has a correlation with a target that exceeds a correlation threshold. In the above medium, the method comprises receiving a plurality of inputs, the plurality of inputs including a first input including glucose measurement values related to the user generated by a glucose monitoring system, and a second input indicating the user's behavior regarding the one or more application functions, calculating a program result metric related to the target based at least on the first input, the program result metric indicating the degree to which the user has achieved the target, calculating, based on the second input, one or more behavior engagement metrics (BEMs) for each of the one or more application functions such that a separate BEM is calculated for each of the one or more application functions, identifying one or more users having a BEM similar to the calculated one or more BEMs in a selected group of users or a pool of users, identifying a new application function not included in the one or more functions based on the function being related to a BEM that exceeds a threshold for at least one of the one or more users, and reconfiguring the application using the new application function based on at least one of the one or more BEMs and the program result metric.

[0029] In the above medium, each BEM of the one or more BEMs is based on the user's interaction with the corresponding application function among the one or more application functions, and the interaction includes at least one of the frequency with which the user interacts with the corresponding application function, the frequency with which the user ignores the guidance generated by the corresponding application function, the average amount of time spent by the user interacting with the corresponding application function, or the degree to which the user's behavior is faithful to the guidance generated by the corresponding application function.

[0030] In the above medium, reconfiguring an application with new application functionality includes identifying one or more underperforming application functionality that has a corresponding BEM below a threshold, and replacing one or more underperforming application functionality with new application functionality. In the above medium, reconfiguring an application with new application functionality includes identifying one or more underperforming application functionality that is relevant to the goal, and identifying that the program result metrics are below a threshold. In the above medium, each of the one or more application functionality includes functionality settings. [Brief explanation of the drawing]

[0031] [Figure 1A] This specification discloses several embodiments of an exemplary decision support system for selecting and adapting the application configuration of a diabetes intervention application that provides decision support guidance to a user. [Figure 1B] A glucose monitoring system, as shown in Figure 1A, according to several embodiments disclosed herein, is shown in more detail, along with several mobile devices. [Figure 2] This specification shows exemplary inputs and exemplary metrics calculated based on the inputs for use by the decision support system shown in Figure 1A, according to several embodiments disclosed herein. [Figure 3] This is a flowchart illustrating the operations performed by a system, such as the decision support system shown in Figure 1A, according to some embodiments disclosed herein. [Figure 4A] This figure shows how the application configuration of an application is first selected and then sequentially fitted using a data model, according to some embodiments disclosed herein. [Figure 4B]Several datasets used in the operation of the data model in Figure 4A, according to some embodiments disclosed herein, are shown. [Figure 5A] This figure shows how the application configuration of an application is first selected and then sequentially fitted using a data model, according to some embodiments disclosed herein. [Figure 5B] Several datasets used in the operation of the data model in Figure 5A, according to some embodiments disclosed herein, are shown. [Figure 6] This is a block diagram illustrating a computing device configured to perform the operation shown in Figure 3, according to a particular embodiment disclosed herein. [Modes for carrying out the invention]

[0032] In certain embodiments, the applications described herein provide guidance that can help patients, caregivers, healthcare providers, or other users improve their lifestyle or clinical / patient outcomes by addressing a variety of challenges, such as nocturnal glucose control (e.g., reducing the incidence of hypoglycemic events or hyperglycemic fluctuations), in-meal and post-meal glucose control (e.g., using historical information and trends to improve blood glucose control), hyperglycemia correction (e.g., increasing time in the target zone while avoiding hypoglycemic events due to overcorrection), hypoglycemia management (e.g., managing hypoglycemia while avoiding "rebound" hyperglycemia), exercise, and / or other health factors. In certain embodiments, the applications may further be configured to have an optimization tool that learns the patient's physiological functions and behaviors and calculates guidance that helps the patient identify optimal or desirable therapeutic parameters, such as basal insulin requirements, insulin-to-carbohydrate ratios, correction factors, and / or changes in insulin sensitivity due to exercise.

[0033] The application may help patients respond to problems in real time by, for example, predicting hypoglycemic or hyperglycemic events or trends, providing recommendations for actions to address occurring or potential hypoglycemic or hyperglycemic events or trends, and / or monitoring the patient's physiological and / or behavioral responses to various events in real time. This type of calculated guidance and support may reduce the cognitive load on the user.

[0034] Physiological sensors, such as continuous glucose monitors, can provide useful data that users can use to manage glucose levels; however, this data may require considerable processing to develop effective strategies for glucose management. The sheer volume of data, as well as the correlations between data types, trends, events, and outcomes, can far exceed human processing capabilities. This is particularly significant when decisions about treatment or responses to physiological conditions are made in real time. Integrating real-time or recent data with historical data and patterns can provide useful guidance for real-time treatment decision-making. Technical tools can process this information to provide decision-supporting guidance calculated to be useful for a particular patient in a specific condition or situation at a particular time.

[0035] When used regularly, applications can be a very powerful tool for managing a patient's diabetic condition. However, the continued use of an application is directly correlated with whether or not the application's configuration helps the patient consistently achieve their goals. Certain existing applications have a technical problem in that they cannot adjust or personalize the application's configuration based on a patient's specific goals and behaviors. Therefore, since each patient's goals and behaviors can differ from those of other patients, a static configuration that continuously provides all patients with the same set of features presents a technical problem that reduces the likelihood of such applications being used by patients. As a result of this technical problem, these applications experience high churn rates.

[0036] For example, a patient might download an application to their mobile device with the intention of achieving a specific goal. However, after using it for only a few days, the patient might determine that the functions presented by the application are not very important to them in that they do not help them achieve any of those goals. As a result, the patient quickly loses interest in the application and stops using it altogether. In another example, a patient might initially use an application regularly, but as their condition, goals, and behaviors change over time, they may begin to feel that the application is no longer important and is useless to them, ultimately leading to a loss of interest in the application.

[0037] Accordingly, certain embodiments described herein provide technical solutions to the technical problems described above by providing decision support systems and methods for automatically configuring and adapting application configurations based on information about a specific user and / or information about a pool of users, such as a layered group of users similar to a user in one or more embodiments. In certain embodiments, application configuration as used herein refers to a set of application functions and their associated settings.

[0038] In certain embodiments, information relating to a particular user includes the user's goals, interests, abilities, demographic information, disease progression, medication information, and / or at least one of a plurality of inputs and metrics as further described below. In certain embodiments, the plurality of inputs and metrics include outcome and behavioral metrics. In certain embodiments, information relating to a particular user may include real-time information, historical information, and / or trends. In certain embodiments, information relating to a stratified group of users includes at least one of the user's goals, interests, abilities, demographic information, disease progression information, medication plan information ("medication information"), a plurality of inputs and metrics relating to such user, and / or application configuration of the user's application. In certain embodiments, information relating to a stratified group of users may include real-time information, historical information, and / or trends.

[0039] For example, after a user downloads an application, in certain embodiments, the decision support system may first act to identify specific information about the user. In certain embodiments, the information about the user may include the user's disease progression, medication information, demographic information, goals, interests, and / or abilities. In certain embodiments, the decision support system may then identify stratified groups of users from a pool of users in the user database based on one or more stratification factors. In certain embodiments, one or more stratification factors may include the user's disease progression, medication information, demographic information, goals, interests, and / or abilities. Information related to the stratified group may provide indicators that can be useful to the user, such as which application configuration is most helpful in achieving the user's goals. Therefore, in certain embodiments, based on the information related to the user and / or the information related to the stratified group, the decision support system may select a specific application configuration and automatically configure the application using the selected application configuration. Subsequently, in certain embodiments, the decision support system may periodically monitor and evaluate one or more behavioral and outcome metrics related to the user's use of the application and its selected application configuration. In certain embodiments, the decision support system then automatically adapts the application configuration based on the user's behavioral metrics and / or information about other users in the user database. In certain embodiments, other users are, for example, users who meet the same goals and exhibit similar behavioral metrics regarding the use of the same application configuration. In certain embodiments, the decision support system continuously adapts the application configuration.

[0040] As described above, in certain embodiments, the application utilizes input from one or more physiological sensors, such as one or more analyte sensors, to provide important and effective guidance. An example of an analyte sensor described herein is a glucose monitoring sensor that measures the concentration of glucose and / or the concentration of substances indicating the concentration or presence of glucose and / or another analyte in the user's body. In some embodiments, the glucose monitoring sensor is a continuous glucose monitoring device, such as a subcutaneous, transdermal, non-invasive, intraocular, and / or intravascular (e.g., intravenous) device. In some embodiments, the device can analyze multiple intermittent blood samples. The glucose monitoring sensor can utilize any method of glucose measurement, including enzymatic, chemical, physical, electrochemical, optical, photochemical, fluorescence-based, spectrophotometric, spectroscopy (e.g., absorption spectroscopy, Raman spectroscopy, etc.), polarization, calorimetry, ionophoresis, and radiation.

[0041] Glucose monitoring sensors can provide a data stream indicating the concentration of an analyte in a host using any known detection method, including invasive, minimally invasive, and non-invasive sensing techniques. The data stream is typically a raw data signal used to provide a useful value of the analyte to a user who may be using the sensor, such as a patient or healthcare professional (HCP, e.g., physician, internist, nurse, or caregiver).

[0042] In some embodiments, the glucose monitoring sensor is an implantable sensor, as described with reference to U.S. Patent No. 6,001,067 and U.S. Patent Publication No. 2011 / 0027127-A1. In some embodiments, the glucose monitoring sensor is a transdermal sensor, as described with reference to U.S. Patent Publication No. 2006 / 0020187-A1. In yet another embodiment, the glucose monitoring sensor is a dual-electrode analyte sensor, as described with reference to U.S. Patent Publication No. 2009 / 0137887-A1. In yet another embodiment, the glucose monitoring sensor is configured to be implanted intravascularly or extracorporeally in a host, such as the sensor described in U.S. Patent Publication No. 2007 / 0027385-A1. These patents and publications are incorporated herein by reference in their entirety.

[0043] While many of the following descriptions and examples are associated with glucose monitoring sensors capable of measuring glucose concentrations in a host, the systems and methods of the embodiments described herein can be used in combination with any type of analyte sensor for any measurable analyte. Furthermore, while certain embodiments described herein are described in relation to diabetes intervention applications, the systems and methods of the embodiments described herein may be used in combination with any health-related applications provided to a user to improve the user's health. For example, a health-related application may assist a user in treating a specific disease, or simply assist in improving the health of a user who has not necessarily been diagnosed with a disease.

[0044] Exemplary System Figure 1A shows an exemplary decision support system 100 (also referred to as the "health monitoring system") for selecting and adapting the application configuration of a diabetes intervention application ("application") 106 that provides decision support guidance to a user 102 (hereinafter referred to as the "user") in a particular embodiment. In a particular embodiment, the user may be a patient or a caregiver of a patient. In the embodiments described herein, for the sake of simplification only, the user is assumed to be a patient, but is not limited to that. In a particular embodiment, the system 100 includes the user, a glucose monitoring system 104, a mobile device 107 running the application 106, a decision support engine 112, and a user database 110. While the system described above is a glucose monitoring system, it should be noted that this is exemplary, and one or more additional or alternative analytes may be monitored and used according to the embodiments provided herein. In some embodiments, the analyte monitoring system (e.g., glucose monitoring system 104) is configured to measure at least one analyte selected from the group consisting of albumin, alkaline phosphatase, alanine transaminase, aspartate aminotransferase, bilirubin, blood urea nitrogen, calcium, carbon dioxide, chloride, creatinine, glucose, gamma-glutamyl transpeptidase, hematocrit, lactate, lactate dehydrogenase, magnesium, oxygen, pH, phosphorus, potassium, sodium, total protein, uric acid, metabolic markers, and drugs.

[0045] As used herein, the term “analyte” is a broad term and should be given its ordinary and customary meaning to those skilled in the art (and not limited to any special or customized meaning), but refers to a substance or chemical component in a biological sample (e.g., body fluids including blood, serum, plasma, interstitial fluid, cerebrospinal fluid, lymph, ocular fluid, saliva, oral fluid, urine, excrement, or exudate). Analytes may include naturally occurring substances, artificial substances, metabolites, and / or reaction products. In some embodiments, analytes for measurement by sensing area, apparatus, and method are albumin, alkaline phosphatase, alanine transaminase, aspartate aminotransferase, bilirubin, blood urea nitrogen, calcium, CO2, chloride, creatinine, glucose, gamma-glutamyl transpeptidase, hematocrit, lactate, lactate dehydrogenase, magnesium, oxygen, pH, phosphorus, potassium, sodium, total protein, uric acid, metabolic markers, and drugs. However, other analytes are also expected and not limited to, acetaminophen, dopamine, ephedrine, terbutaline, ascorbic acid, uric acid, oxygen, d-amino acid oxidase, plasma amine oxidase, xanthine oxidase, NADPH oxidase, alcohol oxidase, alcohol dehydrogenase, pyruvate dehydrogenase, diol, Ros, NO, bilirubin, cholesterol, triglycerides, gentisic acid, ibuprofen, L-dopa, methyldopa, salicylic acid, tetracycline, trazamide, tolbutamide, acarboxyprothrombin, acylcarnitine; adenine phosphoribosyltransfer Gelase; Adenosine deaminase; Albumin; Alpha-fetoprotein; Amino acid profile (Arginine (Krebs cycle), Histidine / Urocanic acid, Homocysteine, Phenylalanine / Tyrosine, Tryptophan); Andrenostenedione; Antipyrine; Arabinitol enantiomer; Arginase; Benzoyl kugonin (Cocaine); Biotinidase; Biopterin; C-reactive protein; Carnitine; Carnosinase; CD4; Ceruloplasmin; Chenodeoxycholic acid; Chloroquine; Cholesterol; Cholinesterase; Conjugated 1-β-hydroxycholic acid; Cortisol; Creatine kinase;Creatine kinase MM isozymes; cyclosporine A; d-penicillamine; deethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetylator polymorphisms); alcohol dehydrogenase; alpha-1-antitrypsin; cystic fibrosis; Duchenne / Becker muscular dystrophy; glucose-6-phosphate dehydrogenase; hemoglobin A; hemoglobin S; hemoglobin C; hemoglobin D; hemoglobin E; hemoglobin F; D-Punjab; Becker Thalassemia, Hepatitis B virus, HCMV, HIV-1, HTLV-1, Leber's hereditary optic neuropathy, MCAD, RNA, PKU, Plasmodium vivax, sex differentiation, 21-deoxycortisol); desbutylhalofantrin; dihydrotheridine reductase; diphtheria / tetanus antitoxin; erythrocyte arginase; erythrocyte protoporphyrin; esterase D; fatty acids / acylglycine; free β-human chorionic gonadotropin; free erythrocyte porphyrin; free cytoplasmic spermane Loxin (FT4); free triiodothyronine (FT3); fumarylacetase; galactose / gal-1-phosphate; galactose-1-phosphate uridyltransferase; gentamicin; glucose-6-phosphate dehydrogenase; glutathione; glutathione peroxidase; glycolic acid; glycosylated hemoglobin; halofantrin; hemoglobin variant; hexosaminidase A; human erythrocyte carbonic anhydrase I; 17-alpha-hydro Xyprogesterone; Hypoxanthine phosphoribosyltransferase; Immunoactivating trypsin; Lactic acid; Lead; Lipoprotein ((a), B / A-1, β); Lysozyme; Mefloquine; Netylmycin; Phenobarbiton; Phenyloin; Phytanic / pristanic acid; Progesterone; Prolactin; Prolidase; Purine nucleoside phosphorylase; Quinine; Reverse triiodothyronine (rT3); Selenium; Serum pancreatic lipase; Disomicin; Somatomedin C;Specific antibodies (adenovirus, antinuclear antibody, anti-zeta antibody, arbovirus, Aueszky's disease virus, dengue fever virus, guinea pig, tapeworm, amoeba histolytica, enterovirus, giardia lamblia, Helicobacter pylori, hepatitis B virus, herpesvirus, HIV-1, IgE (atopic disease), influenza virus, Donovan leishmania, leptospirosis, measles / mumps / rubella, Mycoplasma leprae, myoglobin, circumcised fimbriae, parainfluenza virus, Plasmodium falciparum, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus) This includes rickettsia (scrub typhus), schistosomiasis mansoni, Toxoplasma protozoa, Treponema pallidum, Trypanosoma cruz / Langer's, vesicular stomatitis virus, Wuchereria bancrofti, yellow fever virus); specific antigens (hepatitis B virus, HIV-1); succinylacetone; sulfadoxine; theophylline; thyrotropin (TSH); thyroxine (T4); thyroxine-binding globulin; trace elements; transferrin; UDP-galactose-4-epimerase; urea; europorphyringen I synthesizerase; vitamin A; leukocytes; and zinc protoporphyrin. Naturally occurring salts, sugars, proteins, fats, vitamins, and hormones in blood or interstitial fluid may also constitute the analyte in certain embodiments. Analytes may be naturally occurring in biological fluids, such as metabolites, hormones, antigens, and antibodies. Alternatively, the analytes may include, for example, contrast agents for imaging, radioisotopes, chemical agents, fluorocarbon-based synthetic blood, or, but are not limited to, insulin; ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, Ritalin, Silate, Preludin, Didrex, Prestate, Boranil, Sandrex, Pregin); depressants (barbituates, tranquilizers such as methaquan, barium, Librium, Miltown, Serax, Ecuanil, and tranxen); hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin);Drugs or pharmaceutical compositions containing narcotics (heroin, codeine, morphine, opium, meperidine, percocet, percodan, tausonex, fentanyl, darvon, talwin, romotyl); synthetic narcotics (fentanyl, meperidine, amphetamine, methamphetamine, and phencyclidine analogs, e.g., ecstasy); anabolic steroids; and nicotine can be introduced into the body. Metabolites of drugs and pharmaceutical compositions are also considered analytes. Analytes such as neurochemicals and other chemicals produced in the body, such as ascorbic acid, uric acid, dopamine, norepinephrine, 3-methoxytyramine (3MT), 3,4-dihydroxytyramine, 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), histamine, advanced glycation end products (AGEs), and 5-hydroxyindoleacetic acid (FHIAA), can also be analyzed.

[0046] One aspect of a preferred embodiment provides an analytic spectroscopy monitoring system for in vivo continuous analytic spectroscopy (e.g., albumin, alkaline phosphatase, alanine transaminase, aspartate aminotransferase, bilirubin, blood urea nitrogen, calcium, carbon dioxide, chloride, creatinine, glucose, gamma-glutamyl transpeptidase, hematocrit, lactate, lactate dehydrogenase, magnesium, oxygen, pH, phosphorus, potassium, sodium, total protein, uric acid, metabolic markers, drugs, various minerals, various metabolites, etc.) that can be operably coupled to a catheter to measure the concentration of an analytic spectroscopy in the bloodstream of a host. In some embodiments, the system includes an analytic spectroscopy sensor that extends only a short distance into the bloodstream (e.g., out of the catheter) without substantially occluding the catheter or the bloodstream of the host. The catheter can be fluid-coupled to additional IV and diagnostic devices such as a saline bag, an automated blood pressure monitor, or a blood chemistry monitor. In some embodiments, a blood sample can be removed from the host via the sensor system as described elsewhere in this specification. In one embodiment, the analyte sensor is a glucose sensor, and medical staff monitor the host's glucose level. In other embodiments described elsewhere in this specification, the analyte sensor is located in or on the catheter itself, such as the in vivo portion of the catheter. In yet another embodiment, the analyte sensor is located entirely in and / or on a fluid coupler that is then fluidically coupled to the catheter or other vascular access device, as described elsewhere in this specification.

[0047] In certain embodiments, the glucose monitoring system 104 includes a sensor electronic module and a glucose sensor that measures the concentration of blood glucose and / or the concentration of substances indicating the concentration or presence of glucose and / or other analytes in the user's body. In certain embodiments, the glucose sensor is configured to perform measurements continuously. The sensor electronic module transmits the blood glucose measurements to a mobile device 107 for use by application 106. In some embodiments, the sensor electronic module transmits the glucose measurements to the mobile device 107 via a wireless connection (e.g., Bluetooth connection). In certain embodiments, the mobile device 107 is a smartphone. However, in certain embodiments, the mobile device 107 may instead be any other type of computing device, such as a laptop computer, a smartwatch, a tablet, or any other computing device capable of running application 106.

[0048] In certain embodiments, the decision support engine 112 refers to a set of software instructions having one or more software modules, including a data analysis module (DAM) 113 as well as an adaptive application configuration module (AACM) 115. In some embodiments, the decision support engine 112 runs entirely on one or more computing devices in a private or public cloud. In such embodiments, the application 106 communicates with the decision support engine 112 over a network (e.g., the internet). In some other embodiments, the decision support engine 112 runs partially on one or more local devices, such as a mobile device 107, and partially on one or more computing devices in a private or public cloud. In some other embodiments, the decision support engine 112 runs entirely on one or more local devices, such as a mobile device 107.

[0049] As described below, in certain embodiments, the AACM 115 is configured to adapt the application configuration 108 of application 106 using information about users stored in user profiles 116 and / or information about pools of users, such as layered groups of users, whose profiles may be stored in user database 110. In certain embodiments, the DAM 113 is configured to process a set of inputs received from application 106 and calculate several metrics, such as behavioral metrics 128 and result metrics 130, which may then be stored by application 106 in user profiles 116. The behavioral metrics 128 and result metrics 130 may also be used by application 106, such as by various functions of application 106, to provide real-time guidance to the user. Various data points of user profiles 116, including behavioral metrics 128 and result metrics 130, are described in more detail below.

[0050] In certain embodiments, application 106 provides guidance to the user based on a adaptive application configuration 108. In certain embodiments, application configuration 108 includes or refers to a set of functions 1 to N. In certain embodiments, the set of functions may be selected from a larger pool of possible functions, where N is intended to indicate that application configuration 108 may have any number of functions. In certain embodiments, one or more of functions 1 to N are configured to provide the user with some form of guidance to help the user make more informed decisions. In certain embodiments, to provide the user with effective and important on-time guidance, the functions may take as input information about the user stored in the user profile 116, and / or information about the pool of users stored in the user profile of such user in the user database 110. In certain embodiments, the functions may interact with the user through various means such as text, email, notifications (e.g., push notifications), phone calls, and / or other forms of communication such as displaying content (e.g., graphs, trends, charts, etc.) on the user interface of application 106. In certain embodiments, a function also includes settings that define how the function operates. In certain embodiments, changing the settings of a function results in changing how the function operates, such as the way the function provides guidance and interacts with the user. For example, changing the settings of a function may result in changing one or more of the following: frequency, content, timing, the form of guidance provided by the function, or the time frame of the guidance (e.g., the guidance may be real-time or retrospective / reflective).

[0051] For example, in a particular embodiment, functions 1 to N of the application configuration 108 may include several different exercise management functions. As an example, functions 1, 2, 3, and 4 of the application configuration 108 may correspond to exercise management function 1 (e.g., a reminder function to remind the user to exercise at a specific time), exercise management function 2 (e.g., an exercise recommendation function that recommends various types of exercise based on information about the user), exercise management function 3 (e.g., a function that calculates how long the user should exercise to reach a specific blood glucose level), and exercise management function 4 (e.g., a function that recommends walking routes / hiking trails, etc.).

[0052] Furthermore, in certain embodiments, the exercise management function may be configured using several settings. For example, exercise management function 1 may be configured using one of five settings (e.g., settings A, B, C, D, and E), each setting configuring exercise management function 1 to provide guidance to the user using one or more different aspects such as frequency, content, timing, form, and time frame. For example, setting A may be the least challenging or interactive (e.g., fewer reminders, different content), while setting E may be the most interactive. Thus, in certain embodiments, the application configuration 108 may include a combination of multiple functions and function settings ("FFSC"), each function and function setting combination corresponding to a specific function (e.g., exercise management function) having a specific setting (e.g., setting A). Defining functions at this level of granularity can help identify which exercise management functions and settings are useful and which are not for the user to meet their goals. For example, as will be further explained in relation to Figure 3, user behavior metrics 128 may indicate that exercise management function 1 with setting A is not useful to the user, while information on stratified user groups may indicate that a particular user who found exercise management function 1 with setting A to be unuseful instead found exercise management function 1 with setting B to be very useful in helping them achieve their goals. In such an example, AACM 115 may replace exercise management function 1 with setting A with exercise management function 1 with setting B. In contrast, if only a single exercise management function with a single setting is defined, it may become more difficult to identify what the user finds truly useful about that function and which aspects of the function the user does not want to use. Additional exemplary functions are described in more detail below.

[0053] As described above, in certain embodiments, application 106 is configured to take information about the user as input and store that information in the user's user profile 116. For example, application 106 may acquire the user's demographic information 118, disease progression information 120, and / or medication information 122 and record them in the user profile 116. In certain embodiments, demographic information 118 may include one or more of the user's age, BMI (body mass index), ethnicity, sex, etc. In certain embodiments, disease progression information 120 may include information about the user's disease, such as whether the user has type I diabetes, type II diabetes, is pre-diabetic, or has gestational diabetes. In certain embodiments, information about the user's disease may also include the length of time since diagnosis, the level of diabetes control, the level of adherence to diabetes management treatment, predicted pancreatic function, other types of diagnoses (e.g., heart disease, obesity) or health measures (e.g., heart rate, exercise, stress, sleep, etc.). In certain embodiments, the medication plan information 122 may include information about the amount and type of insulin or non-insulin-based antidiabetic drugs and / or non-diabetic drugs to be taken by the user.

[0054] In certain embodiments, application 106 may obtain demographic information 118, disease progression information 120, and / or medication information 122 from the user in the form of user input or from other sources. In certain embodiments, if some of this information changes, application 106 may receive updates from the user or other sources. In certain embodiments, the user profile, including user profile 116, is stored in a user database 110, which is accessible to application 106 and decision support engine 112 through one or more networks (not shown). In some embodiments, the user database 110 refers to a storage server that may operate on a public or private cloud.

[0055] In certain embodiments, in addition to the user's demographic information 118, disease progression information 120, and / or medication information 122, application 106 acquires an additional set of inputs 127, which are also utilized by functions 1 to N to provide guidance to the user. In certain embodiments, such inputs 127 are acquired sequentially. In certain embodiments, application 108 receives inputs 127 (including, for example, inputs 210 and 220 in Figure 2) through user inputs and / or a plurality of other sources of information, including glucose monitoring system 104, other applications running on mobile device 107, and / or one or more other sensors and devices. In certain embodiments, such sensors and devices include, but are not limited to, one or more of the following: insulin pump, other types of analyte sensors, sensors or devices provided by mobile device 107 or other user accessories (e.g., smartwatch) (e.g., accelerometer, camera, global positioning system (GPS), heart rate monitor, etc.), or any other sensors or devices that provide relevant information about the user.

[0056] In certain embodiments, application 106 further uses at least a portion of input 127 to obtain several metrics, such as behavioral metrics 128 and result metrics 130. As further described in relation to Figure 2, in some embodiments, application 106 sends at least a portion of input 127 to DAM 113 for processing, and DAM 113 generates several metrics based on this. In certain embodiments, behavioral metrics 128 and result metrics 130 may then be used by application 106 as inputs to functions 1 to N for providing guidance to the user. In certain embodiments, behavioral metrics 128 and result metrics 130 may also be stored in user profile 116, which may be used by AACM 115 to adapt application configuration 108.

[0057] In certain embodiments, behavioral metrics 128 are a set of metrics that can indicate user behavior and habits, such as one or more of the following: user behavior or interaction with the application, user behavior regarding disease treatment, health-related behavior, etc. Behavioral metrics 128 may include real-time metrics, historical metrics, and / or trends. Outcome metrics 130 may, in at least some cases, provide a general indication of the user's health or condition, such as one or more of the following: the user's physiological or psychological state (e.g., stress level, well-being, etc.), user health or condition-related trends, or how far or near the user is to achieving their own goals based on their health or condition. In certain embodiments, outcome metrics 130 may directly correlate with behavioral metrics 128, i.e., outcome metrics 130 may be a result of user behavior or patterns of behavior. Examples of outcome metrics 130 include one or more of the following: metabolic rate, glucose levels and trends, user health or illness, etc. In certain embodiments, the result metrics 130 may include real-time metrics, historical metrics, and / or trends.

[0058] In certain embodiments, the user profile 116 also includes the user's goals 132, interests 134, and abilities 136, which, in certain embodiments, are obtained from the user by the application 106 in the form of user input, as described in relation to Figure 3, or are generated by the application 106 based on information provided by the user. In certain embodiments, the goals 132 relate to what the user intends to achieve in terms of improving their health, such as weight loss, food choices, medication reminders, A1c reduction, medication planning, sleep, time within target range, and reduction of hypoglycemic events, as well as the user's psychological state. In certain embodiments, the goals 132 may be dynamic, i.e., the user may change the goals over time. In some embodiments, the goals 132 are measurable metrics such that the decision support engine 112 can determine whether the goals 132 have been met based on the outcome metrics 130.

[0059] Furthermore, in certain embodiments, the decision support engine 112 can use objective 132 to track or determine the effectiveness of the application configuration 108. For example, if the user is achieving or making progress toward achieving objective 132, the decision support engine 112 may consider this an indication that the functionality of the application configuration 108 is assisting the user to at least some extent. Conversely, if the user is not achieving or making progress toward achieving objective 132, this may indicate that the application configuration 108 can be improved. In certain embodiments, interest 134 refers to the user's interests, such as the user's interest in one or more different activities, foods, amounts of sleep, or a particular type of treatment and / or medication. In certain embodiments, capability 136 refers to the user's capabilities, such as the user's ability to perform one or more activities from a possible pool of activities that could be recommended to the user in the form of guidance. Examples include one or more physical disabilities, inability to eat certain foods, inability to engage in certain types of activities, or inability to engage in certain activities at certain times but being able to at other times.

[0060] In certain embodiments, the user profile 116 is dynamic because at least some of the information stored in the user profile 116 can be modified over time, and / or new information can be added to the user profile 116 by the application 106. In certain embodiments, the AACM 115 is configured to adapt the application configuration 108 based on the dynamic user profile 116 and / or the dynamic user profiles of the pool of users in the user database 110 in order to provide guidance to help the user achieve objective 132. Adapting the application configuration 108 may include changing (e.g., adding or deleting) a set of functions 1 to N, and / or changing the settings of one or more of functions 1 to N. In other words, in certain embodiments, each function is adaptable because the settings of the functions can be adapted.

[0061] While not limited to this list, some exemplary functions may include one or more of the following: goal setting and identification functions, reward functions, reporting functions, behavioral intervention functions, medication reminder functions, blood glucose effect estimation functions, and educational functions.

[0062] As will be further described below, in certain embodiments, the goal-setting and identification function may guide the user by defining a set of goals that the user can expect to achieve with the help of application 106. In certain embodiments, the reward function may provide the user with rewards to encourage healthy behaviors.

[0063] In certain embodiments, the reporting function may provide reports to the user in various forms. For example, the reporting function may be configured to report blood glucose fluctuations to the user in a way that allows the user to isolate the fluctuations and consider the causes of such fluctuations. Reports may be provided to the user in various forms as those skilled in the art will understand. For example, the reporting function may display a graph of the user's glucose measurement trends, highlighting the two blood glucose fluctuations, along with the text, "Today you had two high-glucose events, the first lasting 55 minutes and the second lasting 30 minutes." In one example, the reporting function may provide an afternoon report that provides information about the user's average blood glucose levels to date, shows the average for the remainder of the day needed to stay on schedule toward the user's goal, and recommends a hypoglycemic load or physical activity. In another example, the reporting function may provide an overnight summary, including daily, weekly, and monthly average blood glucose levels, as well as an estimated A1c. The summary informs the user what their average blood glucose levels need to be the following day to achieve their A1c goal. In a particular embodiment, the summary is sent to the user at 8 p.m. and pushed again first thing in the morning if the user has not yet read it.

[0064] In certain embodiments, the reporting function may focus on providing the user with educational opportunities. In certain embodiments, educational opportunities identify the impact of behaviors (e.g., physical activity, diet, medication adherence, and sleep) on blood glucose. Educational opportunities may be pushed to the user in the form of notifications and / or recorded on a glucose monitoring curve for timely review by the user. For example, educational opportunities may be visually displayed on the curve and may include behaviors and glucose responses to inform the user about which behaviors caused which glucose responses.

[0065] An example of a positive teaching opportunity may be provided when the user eats a high-glycemic load breakfast that raises blood glucose levels around 10 a.m. In a particular embodiment, based on input from the user's accelerometer, the reporting function may then determine that the user went for a 30-minute walk and that this brought the user's blood glucose back to its target range. In a particular embodiment, the reporting function may mark this event as a teaching opportunity (e.g., by displaying a star on the CGM curve) and send the user a notification stating, "Nice walk, Sharon! Your blood glucose is now back to the value you wanted after a 30-minute walk."

[0066] In certain embodiments, the behavioral intervention feature includes any functionality that works to change the user's behavior, such as by encouraging the user to engage in a particular behavior or to refrain from engaging in a particular behavior. For example, the behavioral feature may be configured to send a push notification to the user to encourage the user to engage in a particular behavior. The behavioral feature may also be capable of determining, for example, based on the user's past behavior, that the user is about to engage in a particular behavior and sending a push notification to the user to refrain from engaging in such behavior. In some embodiments, the push notification to the user may be based on information about the user (e.g., goal 132, outcome metrics 130, behavioral metrics 128, etc.) and / or information about a stratified group of users. For example, the content of the push notification (e.g., the type or specific content of the behavior recommended by the push notification) may be based on data about the types of behavior that have led other users in the stratified group to achieve the same goal.

[0067] For example, a certain type of behavioral intervention function may include exercise management, and for that purpose, one or more exercise management functions may be provided. One example is an exercise management function that encourages the user to exercise via push notification when it is determined that the user's glucose level is high or that the user's lack of exercise is at a level that could raise the user's glucose level. In certain embodiments, the exercise management function may be configured to receive and analyze data from an accelerometer, a global positioning system (GPS), a heart rate monitoring sensor, a glucose monitoring system 104, and / or other types of sensors and devices in order to provide the user with more effective and tailored guidance. For example, by receiving information from one or more of these sensors and devices, the exercise management function may be able to make decisions regarding whether one or more users have actually engaged in exercise, how long the user should exercise to ensure that the user's blood glucose returns to a normal range, and what walking route the user should take.

[0068] In certain embodiments, another type of behavioral intervention function may be accompanied by diet management, which may provide one or more diet management functions. For example, a diet management function may act as a virtual nutritionist to provide the user with guidance on one or more of the following: when to eat, what to eat, how much to eat, etc. In certain embodiments, a diet management function may provide personalized diet recommendations based on one or more of the user's real-time status (e.g., real-time blood glucose measurements), the user's physical response to a particular meal, etc. In certain embodiments, a diet management function may also assist the user with meal preparation and / or shopping and / or allow the user to input information about meals consumed by the user in order to understand nutritional value, etc. In certain embodiments, a diet management function may further do one or more of the following: make menu and ingredient alternative suggestions at restaurants, or suggest healthy restaurants and grocery stores within a particular geographical area. In certain embodiments, a diet management function may provide notifications based on information about the user's diet information. For example, if a user eats a meal and their blood glucose does not subsequently decrease to the target range (e.g., the next pre-meal peak (2 hours after the glucose rise from the first meal) is 180 mg / dL or higher), an emergency alert may be issued to the user to exercise immediately. However, if the user's last meal had a pre-meal peak of less than 180 mg / dL and pre-meal glucose was within the range (80-130), the meal management function may randomize whether or not the user receives an alert after the next meal (e.g., reducing the likelihood of an alert by 33%). It should be noted that the exercise management function and meal management function described above are just two examples of behavioral intervention functions.

[0069] In certain embodiments, the medication management function may provide the user with notifications about one or more of the following: when the user needs to take medication, what type of medication the user should take (e.g., oral medication for type II diabetes and insulin injections for type I diabetes), and what dose or amount. The medication management function may provide such notifications based on specific information of the user, such as one or more of the user's disease (e.g., type I or type II), current outcome metrics (e.g., current blood levels and metrics). In certain embodiments, the medication management function may also track how well the user is following the medication schedule, automatically order medication for the user before the user runs out of medication, and / or provide information about the medication itself (e.g., educating the user about the effects and benefits of the medication). For example, the medication management function may ask the patient whether they have taken their medication. If the user answers "yes" three times in a row, the medication management function may randomize the probability of asking the user the following day and assign it 33%. If the user does not answer "yes" three times in a row, the medication management function may send the user a reminder to take their medication the following day.

[0070] In certain embodiments, the medication management function may automatically communicate with an insulin delivery device, such as a pump, to deliver the correct dose of insulin to the device based on the user's current result metrics 130. For example, the medication management function may consider the user's current glucose level and metrics to determine the exact amount of long-lasting insulin to be administered. In certain embodiments, the medication management function may then signal the medication delivery device to deliver that amount of insulin. In certain embodiments, the medication management function may take into account information about a stratified group of users when providing guidance to the user (e.g., when calculating the amount of insulin to be administered).

[0071] Other behavioral management features may include comorbidity management features, obesity and weight management features, and gender-specific features. Comorbidity management features may include cardiovascular health management features (e.g., providing lactate-guided exercise sessions for cardiovascular health by replacing typical laboratory-specific cardiopulmonary metrics), liver health management features (e.g., providing insights into the user's fasting lactate levels and lactate clearance rate), and features for managing glucose for users with chronic kidney disease (e.g., if A1C is not a good indicator of blood glucose status, patients with kidney disease can receive glucose-derived metrics).

[0072] The obesity and weight management features may provide (1) "fatmax" exercise-guided training for weight management / weight loss through the use of lactate and / or glycerol measurements, (2) diet-specific guidance for those interested in the ketogenic diet, and (3) post-exercise nutrition-specific guidance guided by lactate measurement for users with type 1 diabetes, type 2 diabetes, and / or users interested in health and wellness.

[0073] Gender-specific features include (1) a function that provides glucose management guidance related to changes in the menstrual cycle using sensor data from temperature data, and (2) a function that provides glucose management guidance to male users in order to lower maximum glucose concentration and extend lifespan.

[0074] In certain embodiments, the blood glucose impact estimation function may use the camera of the mobile device 107 to scan the menu and convert each meal item into an estimated blood glucose impact metric. In some embodiments, the blood glucose impact estimation function displays the blood glucose impact metrics superimposed on the menu items. In some embodiments, the blood glucose impact metrics are based on data from user profiles of a stratified group of users. In some embodiments, the blood glucose impact estimation function may use different coloring (e.g., green for healthy items, red for unhealthy items) to highlight different menu items based on how healthy the item is.

[0075] In certain embodiments, the educational function educates the user about their condition and how they can improve their health. For example, the educational function educates the user about the potential effects they may experience if they adopt a particular lifestyle. In certain embodiments, to determine the potential effects, the educational function may consider the effects experienced by other users in a stratified group who have adopted the same lifestyle. For example, the educational function may tell the user, "By following this program, patients similar to you have been able to lose 3 pounds per month or lower their A1C by 5%." In certain embodiments, the educational function may educate the user about the causes and effects of potential behaviors, such as based on the effects experienced by users in a stratified group. In another example, the educational function educates the user about topics that are selected and tailored based on the user's goals 132, interests 134 and / or abilities 136.

[0076] Figure 1B shows the glucose monitoring system 104 in more detail. Figure 1B also shows several mobile devices 107a, 107b, 107c, and 107d. Note that the mobile device 107 in Figure 1A may be any one of the mobile devices 107a, 107b, 107c, or 107d. In other words, any one of the mobile devices 107a, 107b, 107c, or 107d may be configured to perform application 106. The glucose monitoring system 104 may be communicatively coupled to the mobile devices 107a, 107b, 107c, and / or 107d.

[0077] In general and as an example, the glucose monitoring system 104 may be implemented as an encapsulated microcontroller that performs sensor measurements, generates analyte data (e.g., by calculating values ​​for continuous glucose monitoring data), and engages in wireless communication (e.g., Bluetooth and / or other wireless protocols) to transmit data to remote devices such as mobile devices 107a, 107b, 107c and / or 107d. Paragraphs

[0137] to

[0140] of U.S. Patent Application No. 2019 / 0336053 and Figures 3A, 3B and 4 further describe skin-borne sensor assemblies that may be used in connection with the glucose monitoring system 104 in certain embodiments. Paragraphs

[0137] to

[0140] of U.S. Patent Application No. 2019 / 0336053 and Figures 3A, 3B and 4 are incorporated herein by reference.

[0078] In certain embodiments, the glucose monitoring system 104 includes an analyte sensor electronic module 138 and a glucose sensor 140 associated with the analyte sensor electronic module 138. In certain embodiments, the analyte sensor electronic module 138 includes electronic circuitry related to the measurement and processing of analyte sensor data or information, and includes algorithms related to the processing and / or calibration of analyte sensor data / information. The analyte sensor electronic module 138 may be physically / mechanically connected to the glucose sensor 140 and may be integrated with the glucose sensor 140 (i.e., non-removably mounted) or removablely mountable.

[0079] The analyte sensor electronic module 138 may also be electrically coupled to the glucose sensor 140 so that its components can be electromechanically coupled to one another. The analyte sensor electronic module 138 may include hardware, firmware, and / or software that enables measurement and / or estimation of analyte levels in the user via the glucose sensor 140 (which may be / include a glucose sensor, for example). For example, the analyte sensor electronic module 138 may include one or more potentiostats, a power supply for supplying power to the glucose sensor 140, other components useful for signal processing and data storage, and a telemetry module for transmitting data from the sensor electronic module to one or more display devices. The electronic circuitry can be mounted on a printed circuit board (PCB) or platform within the glucose monitoring system 104 and can take various forms. For example, the electronic circuitry can take the form of an integrated circuit (IC), such as an application-specific integrated circuit (ASIC), a microcontroller, a processor, and / or a state machine.

[0080] The analyte sensor electronic module 138 may include sensor electronic circuits configured to process sensor information such as sensor data and generate converted sensor data and displayable sensor information. Examples of systems and methods for processing sensor analyte data are described in more detail herein, U.S. Patents Nos. 7,310,544 and 6,931,327, and U.S. Patent Publications 2005 / 0043598, 2007 / 0032706, 2007 / 0016381, 2008 / 0033254, 2005 / 0203360, 2005 / 0154271, 2005 / 0192557, 2006 / 0222566, 2007 / 0203966 and 2007 / 0208245, all of which are incorporated herein by reference in their entirety.

[0081] The glucose sensor 140 is configured to measure the concentration or level of an analyte in the user 102. The term "analyte" is further defined by paragraph

[0117] of U.S. Patent Application No. 2019 / 0336053, which is incorporated herein by reference. In some embodiments, the glucose sensor 140 includes a continuous glucose sensor such as a subcutaneous, transdermal (e.g., through the skin) or intravascular device. In some embodiments, the glucose sensor 140 can analyze multiple intermittent blood samples. The glucose sensor 140 can use any method of glucose measurement, including enzymatic, chemical, physical, electrochemical, spectrophotometric, polarization, calorimetry, ionophoresis, radiometric, immunochemical, and the like. Additional details relating to continuous glucose sensors are provided in paragraphs

[0072] -

[0076] of U.S. Patent Application No. 13 / 827,577. Paragraphs

[0072] to

[0076] of U.S. Patent Application No. 13 / 827,577 are incorporated herein by reference.

[0082] Referring further to Figure 1B, the mobile devices 107a, 107b, 107c, and / or 107d may be configured to display (and / or alarm) displayable sensor information that can be transmitted by the sensor electronic module 138 (for example, in customized data packages transmitted to a display device based on their respective preferences). Each of the mobile devices 107a, 107b, 107c, and / or 107d may each include a display such as a touchscreen display 109a, 109b, 109c, and / or 109d for displaying a graphical user interface of application 106 for presenting sensor information and / or analyte data to user 102 and / or receiving input from user 102. In certain embodiments, the mobile devices may include other types of user interfaces, such as a voice user interface, instead of or in addition to the touchscreen display, for communicating sensor information to user 102 of the mobile devices and / or receiving user input. In certain embodiments, one, some, or all of the mobile devices 107a, 107b, 107c, and / or 107d may be configured to display or otherwise communicate sensor information without any additional foresight processing required for calibration and / or real-time display of sensor data, as communicated from the sensor electronic module 138 (for example, in data packages sent to their respective display devices).

[0083] The multiple mobile devices 107a, 107b, 107c and / or 107d depicted in Figure 1B may include custom or proprietary display devices, such as the analyte display device 107b, which are specifically designed to display certain types of displayable sensor information (e.g., numerical values ​​and / or arrows in certain embodiments) related to analyte data received from the sensor electronic module 138. In certain embodiments, one of the multiple mobile devices 107a, 107b, 107c and / or 107d may include a smartphone, such as the mobile phone 107c, based on Android, iOS, or another operating system configured to display a graphical representation of continuous sensor data (e.g., including current and / or historical data).

[0084] Figure 2 provides a more detailed diagram of exemplary inputs and exemplary metrics determined based on said inputs according to a particular embodiment. Figure 2 shows exemplary inputs 210 and 220 on the left, application 106 and DAM 113 in the center, and result metrics 130 and behavioral metrics 128 on the right. Application 106 acquires inputs 210 and 220, which are part of input 127, through one or more channels (e.g., manual user input, sensors, other applications running on the mobile device 107). In a particular embodiment, inputs 210 and 220 may be used by functions 1 to N of application 106 to provide guidance to the user. Inputs 210 and 220 may also be further processed by DAM 113 to output multiple metrics such as result metrics 130 and behavioral metrics 128, which may also be used by functions 1 to N of application 106 to provide guidance to the user.

[0085] As described above, in certain embodiments, the result metrics 130 and behavior metrics 128 are also used by the AACM 115 to adapt the application configuration 108. In one example, input 210 is used by the DAM 113 to output the result metrics 130, and input 220 is used to output the behavior metrics 128. However, in other examples, either input 210 or 220 may be used to calculate either the result metrics 130 or the behavior metrics 128. In certain embodiments, each metric may correspond to one or more values, such as discrete numerical values, ranges, or qualitative values ​​(high / medium / low or stable / unstable).

[0086] In certain embodiments, starting with input 210, food consumption information may include information about one or more meals, snacks and / or beverages, such as size, contents (carbohydrates, fats, proteins, etc.), order of consumption, and time of consumption. In certain embodiments, food consumption may be provided by the user through manual input, by providing a photograph through an application configured to recognize the type and quantity of food, and / or by scanning a barcode or menu. In various examples, the size of a meal may be manually entered as one or more of calories, quantity ("three cookies"), menu item ("Cheese Royale"), and / or food exchange (one fruit, one dairy product). In some examples, a meal may be entered along with the user's typical items or combinations for this time or content (e.g., breakfast on a workday at home, brunch on a weekend at a restaurant). In some examples, meal information may be received via a convenient user interface provided by application 106.

[0087] In certain embodiments, activity information may also be provided as input. Activity information may be provided, for example, by an accelerometer sensor on a wearable device such as a wristwatch, fitness tracker, and / or patch. In certain embodiments, activity information may also be provided through manual user input.

[0088] In certain embodiments, patient statistics may also be provided, such as one or more of the following: age, height, weight, body mass index, body composition (e.g., body fat percentage), height, build, or other information. In certain embodiments, patient statistics are provided through a user interface, by interfacing with an electronic information source such as an electronic medical record, and / or from the measuring device. In certain embodiments, the measuring device includes, for example, one or more of the following: a Bluetooth-enabled wireless scale and / or a camera, which can communicate with the mobile device 107 to provide patient data.

[0089] In certain embodiments, inputs related to patient insulin delivery may be received via wireless connection in a smart pen, via user input, and / or from an insulin pump. Insulin delivery information may include one or more of the following: insulin amount, delivery time, etc. Other parameters, such as insulin action time or duration of insulin action, may also be received as input.

[0090] In certain embodiments, the input may be received from a sensor such as a physiological sensor capable of detecting one or more of the following: heart rate, respiration, oxygen saturation, or body temperature (for example, to detect illness). In certain embodiments, the electromagnetic sensor may also detect a low-power RF field emitted from an object or a tool touching or near an object, which may provide information about the patient's activity or location. An example of information that may be received from the sensor is the user's blood glucose level.

[0091] In certain embodiments, blood glucose information may be provided as input, for example, through a glucose monitoring system 104. In certain embodiments, blood glucose information may be received from one or more sensors that measure peripheral neuropathy using tactile responses, such as by using the tactile function of a smartphone or professional device.

[0092] In certain embodiments, time may be provided as an input such as a time or time from a real-time clock.

[0093] User input through a user interface, such as the user interface of the mobile device 107, may include any other types of input that the user may provide to the application 106, such as the other types of input 210 described above. For example, in certain embodiments, user input may include one or more of the following: mental state or stressor information, delivery of treatment such as the use of glucagon to stimulate hepatic glycogen release in response to hypoglycemia, recommended basal rate or insulin-to-carbohydrate ratio (e.g., received from a clinician), recorded activity (e.g., intensity, duration and time of completion or start). In certain embodiments, user input may indicate drug intake (e.g., type and dosage of drug and timing of drug intake).

[0094] As described above, in certain embodiments, the DAM 113 determines or calculates user result metrics 130 based on the input 210. An exemplary list of result metrics 130 is shown in Figure 2.

[0095] In certain embodiments, the metabolic rate is a metric that represents or may include the basal metabolic rate (e.g., energy consumed at rest) and / or activity metabolism, such as energy consumed by activity, such as exercise or strenuous activity. In some examples, the basal metabolic rate and activity metabolism may be tracked as separate outcome metrics. In certain embodiments, the metabolic rate may be calculated by the DAM 113 based on one or more of the inputs 210, such as one or more of activity information, sensor inputs, time, and user inputs.

[0096] In certain embodiments, the activity level metric may indicate the user's level of activity. In certain embodiments, the activity level metric is determined based on input from, for example, an activity sensor or other physiological sensor. In certain embodiments, the activity level metric may be calculated by the DAM 113 based on one or more inputs 210, such as one or more of activity information, sensor input, time, user input, etc.

[0097] In certain embodiments, insulin sensitivity metrics may be determined using historical data, real-time data, or a combination thereof, and may be based on one or more inputs 210, such as one or more of food consumption information, blood glucose information, insulin delivery information, and resulting glucose levels. In certain embodiments, insulin onboard metrics may be determined using insulin delivery information and / or known or learned (e.g., from patient data) insulin time-acting profiles, which may take into account both basal metabolic rate (e.g., insulin refresh to maintain bodily function) and insulin use driven by activity or food consumption.

[0098] In certain embodiments, the dietary status metric may indicate the state in which a user is related to food consumption. For example, dietary status may indicate whether a user is fasting, pre-meal, eating, post-meal response, or stable. In certain embodiments, dietary status may also indicate onboard nutrition, such as a meal, snack, or beverage consumed, and may be determined from food consumption information, meal timing information, and / or digestibility information, which may correlate with the type, quantity, and / or order of food (e.g., which food / beverage was eaten first).

[0099] In certain embodiments, health and illness metrics may be determined based on one or more of the following: user input (e.g., pregnancy information or known illness information), physiological sensors (e.g., temperature), activity sensors, or a combination thereof. In certain embodiments, based on the values ​​of the health and illness metrics, the user's state may be defined as one or more of the following: healthy, sick, resting, or fatigued.

[0100] In certain embodiments, glucose level metrics may be determined from sensor information (e.g., blood glucose information obtained from a glucose monitoring system 104). In some examples, glucose level metrics may also be determined based on historical information about glucose levels in a particular situation, such as a given combination of food consumption, insulin, and / or activity. In certain embodiments, blood glucose trends may be determined based on glucose levels over a particular period of time.

[0101] In certain embodiments, the outcome metrics may also include disease stages, such as for patients with type II diabetes. Exemplary disease stages for patients with type II diabetes may include the prediabetic stage, the oral treatment stage, and the basal insulin treatment stage. In certain embodiments, the degree of glycemic control (not shown) may also be determined as an outcome metric, and may be based on, for example, one or more of glucose levels, glucose level fluctuations, or insulin administration patterns.

[0102] In certain embodiments, clinical metrics generally indicate the clinical state in which the user is in relation to one or more of the user's conditions, such as diabetes. For example, in the case of diabetes, clinical metrics may be determined based on blood glucose measurements including one or more of the following: A1c, A1c trend, time within target range, time elapsed below threshold, time elapsed above threshold, and / or other metrics derived from blood glucose levels. In certain embodiments, clinical metrics may include one or more of the following: estimated A1c, blood glucose variability, hypoglycemia, and / or health indicators (magnitude of time outside the target range).

[0103] In certain embodiments, the program result is a metric calculated to determine how successful the user is in meeting their defined goals, as will be further described below. For example, DAM113 may take one or more of the important result metrics described above as input and then determine how far or close the user is to achieving the corresponding goal, which may also be defined in terms of specific metrics. In certain embodiments, whether or not a result metric is important depends on the user's goal. For example, if the user's goal is to maintain an A1c of 5%, one or more metrics such as glucose level, glucose trend, and clinical metrics may be used to determine whether the user is meeting that goal. In certain embodiments, DAM113 may also use input 210 to determine whether or not the user is meeting that goal. For example, if the user's goal is to lose 5 pounds, user input related to the user's weight may be used to determine a program result metric indicating whether or not the user has lost 5 pounds. In one example, the program result metric may be quantified using a percentage. For example, if the user in the above example loses 3 pounds, the user's program result metric with respect to that goal may be 60% (3 / 5). In certain embodiments, the DAM 113 may also evaluate trends with respect to the input 210 and result metrics 130 to determine the user's progress toward achieving the goal (i.e., whether the user is moving in the right direction) with respect to achieving the goal, even if the user has not yet achieved the goal.

[0104] Figure 2 also shows behavioral metrics determined based on input 220. In certain embodiments, input 220 includes user input, such as input through a user interface. In certain embodiments, user input, or its absence, may indicate the user's level of interest in a particular FFSC. The level of interest in an FFSC may indicate how useful the user considers the FFSC to be in achieving their goals. For example, low user interest in a particular FFSC may be recorded if the user ignores certain reminders of the FFSC, fails to input information requested by the FFSC, or clearly indicates that the user does not like certain forms of guidance or interaction provided by the FFSC. In certain embodiments, user input may also indicate drug intake (e.g., type and dosage of drug, and / or timing of drug intake). In certain embodiments, input 220 also includes calendar information, such as availability or activity information received from a calendar application on a computer or smartphone running on the mobile device 107. In certain embodiments, input 220 also includes activity information as described above. In certain embodiments, input 220 also includes information about the user's location (e.g., GPS data) and / or time (e.g., from a real-time clock). Inputs from sensors and inputs regarding the user's food consumption are described above. In certain embodiments, additional inputs are also possible.

[0105] As described above, in certain embodiments, the DAM 113 determines or calculates user behavior metrics 128 based on the input 210. An exemplary list of behavior metrics 128 is shown in Figure 2. Behavioral metrics include behavioral engagement metrics and / or one or more metrics related to one or more of the following: eating habits, disease treatment adherence, medication type and adherence, healthcare use, exercise plan, behavioral status, etc.

[0106] In certain embodiments, behavioral engagement metrics (BEM) indicate the level of engagement with a user's interaction with a particular FFSC in application 106. In certain embodiments, the BEM may be calculated for each FFSC in application configuration 108, as further described below. In certain embodiments, the BEM associated with a particular FFSC may be calculated based on the user's interaction with that feature, indicated by user inputs provided as part of input 220, and / or other inputs such as time and calendar (e.g., to calculate the frequency of interactions, etc.). For example, the BEM for a particular FFSC may be calculated based on one or more pieces of information such as how often the user interacts with the FFSC, how often the user ignores guidance generated by the FFSC, and the average time the user spends interacting with the FFSC. In some embodiments, data points related to the frequency of user interaction with the FFSC or other time-related data may be calculated based on an interaction log that holds records of one or more pieces of information such as the time each time the user interacts with a feature, and the amount of time the user spends interacting with a feature. In certain embodiments, the BEM calculated for FFSC may also take into account the consistency of the user's behavior toward FFSC. In certain embodiments, the BEM calculated for FFSC may also take into account the reliability of the user's behavior toward FFSC. The reliability of the behavior may relate to the level of confidence or conviction in the user's behavior toward FFSC.

[0107] In certain embodiments, BEM may also be based on other behavioral metrics, such as one or more of the following: dietary habits, medication adherence, blood glucose data, etc., because these other behavioral metrics may indicate how engaged the user is with application 106. As an example, metrics related to the user's dietary habits may be considered to determine the user's BEM for the dietary management FFSC. For example, if the user is not consuming the type of diet recommended by the dietary management FFSC, which may be determined based on food consumption information, the user's BEM for the dietary management FFSC may be lowered. In certain embodiments, the FFSC is the dietary management FFSC. X The BEM for this may be calculated as follows: BEM FFSCX =(W1)(User Interaction)+(W2)(MH)+···

[0108] In the above formula, W1 is the user's FFSC X W2 is a defined weight that can be assigned to interactions with the user. W2 is a defined weight that can be assigned to the user's eating habits metric. FFSC X In the above formula for calculating the BEM metric for , additional and / or alternative metrics may be considered or included. The BEM for other FFSCs may similarly be calculated not only based on the user's interaction with the FFSC, but also, in certain embodiments, based on other relevant behavioral metrics.

[0109] In certain embodiments, dietary habits are measured by one or more metrics based on the content and timing of the user's meals. For example, if the dietary habits metric is on a scale of 0 to 1, in one example, the better / healthier the user's dietary habits metric will be, the higher it will be toward 1. Also in this example, the more faithful the user's food consumption is to a certain time schedule, the closer the user's dietary habits metric will be toward 1. In certain embodiments, disease management and compliance are measured by one or more metrics indicating how committed the user is to managing their disease. In certain embodiments, the disease management and compliance metrics are calculated based on one or more of the user's diet or food consumption, exercise plan, medication adherence, etc. In certain embodiments, medication adherence is measured by one or more metrics indicating how committed the user is to their medication plan. In certain embodiments, medication adherence metrics are calculated based on one or more of the following: when the user takes the medication (e.g., whether the user takes it on time or according to schedule), the type of medication (e.g., whether the user takes the correct type of medication), and the dosage of the medication (e.g., whether the user takes the correct dosage).

[0110] In certain embodiments, healthcare use is measured by one or more metrics indicating how often a user visits a pharmacy or healthcare professional. In certain embodiments, healthcare use metrics are calculated based on information about the user's prescriptions, pharmacy visits, visits to a doctor's office, etc., and one or more user inputs, including one or more of location inputs, calendar inputs, etc. In certain embodiments, exercise planning is measured by one or more metrics indicating one or more of the types of activities a user engages in, how intense the activities are, and how often the user engages in such activities. In certain embodiments, exercise planning metrics may be calculated based on one or more activity sensors, calendar inputs, user inputs, etc. In certain embodiments, a user's behavioral state refers to one or more metrics for measuring the user's current behavior (e.g., sleep state (e.g., sleep, rest, or wakefulness, which can be inferred from activity sensors, calendar, or other information), appetite (e.g., which can be inferred from eating patterns), etc.).

[0111] Figure 3 is a flowchart illustrating exemplary operations 300 performed by a system (e.g., system 100) to select an application configuration and adapt it based on information about users and / or information about a pool of users, such as a layered group of users similar to a user in one or more embodiments. Operations 300 are described below with reference to Figures 1 and 2 and their components.

[0112] In step 302, operation 300 begins by identifying the user's goals in application 106. Step 302 may be performed by application 106 in some embodiments. In one example, application 106 may identify the user's goals 132 during an initial setup process, such as when the user first downloads the application. In another example, application 106 may identify the user's goals 132 ("goals") at a later point in time. In such an example, the user may decide to use application 106 for a period of time and then interact with application 106 to set some goals. As described above, in certain embodiments, application 106 may use a Goal Setting and Identification Function (OSIF) configured to identify the user's goals through some kind of interaction with the user.

[0113] In certain embodiments, by identifying the user's goals, application 106 can determine what is truly important to the user and how the user defines success. In some embodiments, OSIF can ask the user to identify their goals by providing the user with a user interface that allows the user to input their goals. The user interface may include, for example, a drop-down menu with several potential goals that the user can select. For example, the drop-down menu may include options such as a) "I want to lose weight," b) "I want to sleep better at night," c) "I want to better manage my nighttime blood sugar levels," d) "I want to improve my food choices," e) "I want to optimize my medication plan," f) "I want to improve my time within target limits." As those skilled in the art will understand, a variety of other interactive user interface options may be used to obtain user input.

[0114] In some embodiments, OSIF may obtain the user's goals by interacting with the user. In one example, OSIF may interact with the user using a chatbot. In another example, OSIF may interact with the user through voice by asking the user what their goals are. In such embodiments, the user's responses may be processed and recorded as text using speech recognition capabilities. The text may then be processed using natural language processing capabilities to derive one or more goals. In some examples, OSIF may ask the user a question that directly asks about their goals, such as "What are your goals?" In other examples, OSIF may engage in a more casual conversation with the user and infer the user's goals from the user's responses. For example, OSIF may start a conversation with the user by saying something like, "How would your diabetes management look in a perfect world?" or "How would you rate what you're doing to live with diabetes?" In such examples, OSIF infers the goals from the user's responses.

[0115] In some cases, user input regarding their goals may be very specific and quantifiable. For example, a user might define a goal as "lower my A1c by 2 points." In other cases, OSIF may allow users to define goals in a less specific and more qualitative form to facilitate the goal identification process. For example, OSIF may provide users with options such as "I want to strive for 'normal' glucose control" or "I want to strive for challenging glucose control." In one example, the user may select the first option in response, in which case OSIF may consider a set of guidelines, in which application 106 may be configured to translate "normal" glucose control into a specific and quantifiable goal. For example, if OSIF determines that the user is pregnant, "normal" glucose control may be translated into a 5% A1c. In another example, OSIF may translate "normal" glucose control into a specific goal by considering information about a stratified group of users. In such an example, this translation may be performed after a stratified group of users has been selected in step 304.

[0116] In some embodiments, a user's input regarding their goals may be relative to a stratified group of users. For example, a user might select or state, "I want to be in the top third of my peers, for example, with regard to glucose control, health, weight, etc." In such an example, OSIF may similarly translate the statement into specific, quantifiable goals defined in the form of metrics, based on reviewing information about the stratified group of users after the stratified group of users has been selected in step 304. For example, OSIF might translate "being in the top third of my peers" into a specific set of metrics (e.g., a 6% A1c, a weight of 180 pounds, etc.) and a set of possible treatments or action plans. A set of possible treatments may include (a) reducing carbohydrate intake by 25% and exercising for an additional 20 minutes three times a week, (b) taking additional or different medications (e.g., seeing a doctor to add an additional daily oral medication), or (c) following a specific diet plan that reduces carbohydrate intake below a certain amount per day. In certain embodiments, OSIF may then present a set of possible treatments and allow the user to select one of them, clearly defined by quantifiable metrics that the user is expected to achieve, for example, to be in the top third of their peers. In certain embodiments, OSIF may also allow the user to fine-tune a treatment after they have made their selection.

[0117] It should be noted that OSIF may follow a similar process even in scenarios where non-specific user input is not defined for a stratified group of users. Examples of non-specific user input that is not relative to a stratified group include "normal glucose control," "good health," and "healthy range." After receiving such user input, OSIF may consider information about the stratified group of users and translate such non-specific user input into a specific set of metrics and a set of possible treatments.

[0118] In certain embodiments, OSIF may determine a set of specific goals (in the form of metrics) that the user should strive to achieve, based on a set of user-related inputs (e.g., input 127). In certain embodiments, these inputs may already be available during the setup process of application 106. In such embodiments, the goals 132 set during step 302 are based on such inputs (e.g., for a user who was using application 106 before using OSIF). In certain embodiments, such inputs may not be available during the setup process. Therefore, an initial set of goals may be set for the user, but if inputs become available for the user during the user's use of application 106, OSIF may modify the initial goals based on the user's own inputs. For example, initially, during the setup process, the user's A1C level may be X%, and OSIF may suggest a goal of 0.9X%. However, after the user uses application 106, OSIF may determine that the user's A1C level is actually 0.85X%, and therefore adjust the goal to 0.8X%. In another example, when a user uses application 106, OSIF may determine that the user's A1C level is actually 1.2X%, and therefore 0.9X% may be unattainable for the user, and adjust the target to X%. In yet another example, when a user uses application 106, OSIF may determine that the user's A1C level is X%, but similar users within a stratified group of users have an average A1C level of 0.7X%. In such an example, OSIF may adjust the user's initial target of 0.9X% to 0.8X%.

[0119] In some embodiments, the goals may be categorized into short-term and long-term goals. An example of a short-term goal is "to cut the dosage and cost by half." An example of a long-term goal is "to eliminate the medication completely."

[0120] In some embodiments, the OSIF further identifies the user's interests 134 and abilities 136 through the same types of interactions described above. Identifying the user's interests 134 and abilities 136 is advantageous because the application configuration 108 can be configured and adapted by the AACM 115 based not only on the user's goals 132 but also on the user's interests 134 and abilities 136. For example, if the user indicates that they are disabled and unable to walk, the exercise management function of the application 106 may be configured to suggest exercises or activities to the user that do not require the user to walk or run. In an example, the OSIF may ask the user about their interests and abilities based on the FFSC available through the application 106. For example, in certain embodiments, whether the user is interested in playing chess may not be important because the application 106 does not provide a function from which it can benefit. However, asking whether the user likes swimming or hiking may be more important because, in certain embodiments, the application 106 actually provides an exercise management function from which it can benefit from the user's response about whether they like swimming.

[0121] In step 304, operation 300 continues by identifying a stratified group of similar users from the user database based on information related to the user. Step 304 may be performed by AACM 115 in some embodiments. For example, AACM 115 may retrieve a user profile 116 from the user database 110 and select a stratified group of users from the user database 110 based on the information in the user profile 116. In certain embodiments, AACM 115 selects a stratified group of users based on one or more similarities between the information in the user profile 116 and the user profiles in the pool of users in the user database 110. For example, AACM 115 may use one or more similarities or stratification factors for stratification, one or more stratification factors including at least one of the user's disease progression information, medication information, demographic information, goals, achievement of corresponding goals, interests, abilities, behavioral metrics, outcome metrics, or a combination thereof. However, for users who have just started using application 106, behavioral metrics and outcome metrics may not yet be available.

[0122] In certain embodiments, additional stratification factors are also possible. For example, stratification may be performed based on the user's interaction with application 106 during the setting process (e.g., during the goal setting process). For example, stratification may be performed based on the user's likelihood of focusing on one or more of the following: physical activity, overall interaction with application 106, etc. In one example, application 106 may categorize users based on three categories: negative, passively open, and actively open. A negative user may be one who refuses to accept that they have diabetes and does not want to hear about it or take action about it. A passively open user may be one who accepts that they have diabetes and is open to information, but may not be serious about taking all the steps and means necessary to improve their health. An actively open user may be one who accepts that they have diabetes, is eager to get more information, and is willing to do whatever it takes to improve their health. In certain embodiments, application 106 may categorize a user into one of these categories based on the user's interaction with application 106 during the configuration process (for example, by asking questions designed to determine which category the user belongs to).

[0123] It should be noted that in certain embodiments, goal achievement may be used as a stratification factor. In certain embodiments, how a user has performed with respect to achieving a particular goal is indicated by a corresponding program result metric calculated for that particular goal and stored in the user's profile. In certain embodiments, goal achievement may be defined by a threshold. For example, a 100% program result metric for a particular goal may indicate that a user in the pool of users has fully achieved that goal, while a 90% program result metric may indicate that a user has made positive progress toward achieving the goal by achieving at least 90% of the goal (i.e., losing 9 pounds instead of 10 pounds). With respect to stratification, in embodiments where goal achievement is used as a stratification factor, a threshold program result metric may be defined. For example, the stratification factor may be at least, for example, a 70% program result metric. In that case, users who have made at least 70% progress toward achieving the goal would be included in the stratified group of users.

[0124] Examples of user information that may be used to select stratified groups include: 1) a 62-year-old woman with a 3-year history of diabetes receiving only basal insulin; 2) a 55-year-old man receiving basal and rapid-acting bolus therapy; 3) a 70-year-old man with heart disease receiving metformin; and 4) a 40-year-old woman actively managing diabetes with diet and exercise.

[0125] Various methods and techniques may be used to stratify the user database 110 based on one or more stratification factors (e.g., disease progression, medication information, demographic information, goals, achievement of corresponding goals, interests, abilities, behavioral metrics, outcome metrics, or a combination thereof). In certain embodiments, AACM 115 may use one of various data filtering techniques to filter the broad user database 110 based on one or more stratification factors. For example, if a user has type 1 diabetes, AACM 115 may filter all user profiles in the user database 110 that also have type 1 diabetes. In this case, the stratified group of users would include all such users. However, in certain embodiments, if additional stratification factors are used for stratification, additional filtering may be performed to further narrow the group of users within the stratified group. For example, if the stratification factors include disease progression and demographic information, and the user is a male with type 1 diabetes, AACM 115 may filter all user profiles of all male users in the user database 110 that also have type 1 diabetes.

[0126] In certain embodiments, AACM115 may use machine learning algorithms to stratify the user database 110. For example, an unsupervised learning algorithm may be used to cluster all user profiles in the user database 110 and determine which cluster a user profile 116 belongs to. Unsupervised learning is a type of machine learning algorithm used to derive inferences from a dataset consisting of labeled, unresponsive input data. As those skilled in the art will understand, in addition to unsupervised learning algorithms focused on clustering analysis, other types of unsupervised learning algorithms may be used.

[0127] In certain embodiments, a supervised learning algorithm may be used instead. Supervised learning is a machine learning task that learns a function that maps inputs to outputs based on exemplary input-output pairs. In certain embodiments, using a supervised learning algorithm, AACM115 may be configured to classify user profiles 116 by determining which class or stratified group a user belongs to, based on a machine learning model trained using a labeled dataset. In certain embodiments, the labeled data already includes various classes of users that are classified based on one or more characteristics, such as disease progression. For example, in certain embodiments, one class of users includes all user profiles with type 1 diabetes, and another class of users includes all user profiles with type 2 diabetes. In such an example, if the disease progression information 120 of user profile 116 indicates that the user has type 1 diabetes, AACM115 selects that class as a stratified group of users about the user.

[0128] In step 306, operation 300 continues by configuring the application (e.g., automatically) based on user-related information, including the user's goals, and / or information related to a stratified group of users. Configuring the application automatically means configuring the application without user action or involvement. For example, the application may be automatically configured using a specific number of FFSCs selected by AACM115 from more FFSCs available in the application binary file downloaded to the user device 107.

[0129] Step 306 may be performed by AACM115 in some embodiments. For example, AACM115 may configure application 106 using application configuration 108 based on user profile 116 and / or user profiles of a stratified group of users. In certain embodiments, AACM115 may analyze which one or more FFSCs within the stratified group have the strongest correlation with the achievement of the user's goal 132. In certain embodiments, AACM115 then configure application 106 using at least a subset of such FFSCs, resulting in application configuration 108. In certain embodiments, the subset may include FFSCs having correlation scores above a certain correlation threshold. As described above, FFSC refers to a specific combination of function and function setting, such as exercise management function 2 having setting 3. In certain embodiments, step 306 may be performed by application 106 itself. For example, AACM115 may present a subset of FFSC to application 106, which then automatically configures itself using the subset of FFSC, resulting in application configuration 108.

[0130] In one example, the user may be pre-diabetic and 77 years old. The user may also have goals such as losing 5 pounds and eating less carbohydrates and more protein. The user's interests may include hiking, walking, swimming, eating lean meat, Italian cuisine, and disliking fish. In such an example, stratified user groups may be selected based on the user's disease type and / or age. AACM115 may then determine which FFSCs are most strongly correlated with losing 5 pounds and eating less carbohydrates. For example, AACM115 may determine that among several, exercise management function, reporting function, reward function, and diet management function are most strongly correlated with achieving the above goals for users within the stratified group. That is, in such an example, the profiles of users within the stratified group who achieved those goals may most frequently show their use of one or more of these FFSCs and a high BEM associated with such FFSCs. Note that in this example, the achievement of these goals is used as a stratification factor.

[0131] Those skilled in the art will recognize various types of operations or algorithms that can be used to find correlations between the achievement of a user's goals and the FFSCs used by users within a stratified group that are most strongly associated or correlated with the achievement of those goals. Examples of algorithms that may be used include one or more of the following: linear correlation algorithms (e.g., Pearson's correlation coefficient), nonlinear correlation algorithms (e.g., distance correlation, maximum information coefficient, etc.), and various types of machine learning algorithms.

[0132] In certain embodiments, a linear correlation algorithm may be used to determine the correlation between two variables, including the correlation between achieving a specific goal (second variable) and a specific FFSC (first variable). In certain embodiments, the underlying dataset on which the correlation algorithm is performed includes data points from user profiles of several users (e.g., stratified groups of users). In certain embodiments, the data points include the FFSCs each user utilizes, the user's goals, and whether the user achieved those goals. Using such a dataset, in some embodiments, AACM115 may perform multiple analyses, each focusing on the correlation between one of a possible number of FFSCs and one of the user's goals. Other types of analyses may be performed so that the correlation between the user's goals and multiple FFSCs can be determined. Instead of the correlation algorithm as described above, in certain embodiments, a machine learning model may be used to identify features that strongly correlate with the achievement of the user's goals. Various examples of algorithms or data models used to perform step 306 are illustrated with reference to Figures 4A-5B.

[0133] Once a pool of FFSCs positively correlated with achieving the user's goals has been determined, in a particular embodiment, AACM115 may then select all of these FFSCs to constitute application 106 and reduce to application configuration 108. In a particular embodiment, AACM115 may select a subset of FFSCs from the pool of FFSCs that are very strongly correlated with achieving the user's goals, for example, by applying a correlation threshold. As an example, a pool of FFSCs may be determined first, each FFSC having a specific correlation score, where the correlation score indicates the correlation between the FFSC and achieving the user's goals. AACM115 may then select FFSCs that have correlation scores above a threshold. Using the selected FFSCs, AACM115 may constitute application 106 and reduce to application configuration 108.

[0134] In step 308, operation 300 continues by evaluating user behavior metrics and / or outcome metrics. In some embodiments, step 308 is performed by DAM 113. For example, if application 106 is configured using application configuration 108 and a user begins using application 106, DAM 113 may receive input 127 from application 106, based on which DAM 113 determines outcome metrics 130 and behavior metrics 128. In some embodiments, outcome metrics 130 may be determined to evaluate whether the user is meeting or at least improving their goals. For example, program outcome metrics may be calculated for each of the user's goals. Program metrics may indicate, for example, how close the user is to meeting their goal, using a percentage.

[0135] As described above, in certain embodiments, the behavioral metrics 128 include BEMs, where each BEM represents the user's engagement with a different FFSC. In certain embodiments, as described above, when calculating the BEM for a particular FFSC, the DAM 113 also considers other behavioral metrics that are important to the FFSC. Thus, the BEM may indicate whether the application configuration 108 needs to be adapted to be more effective in helping the user achieve their goals.

[0136] As described above, in certain embodiments, the result metrics 130 and behavioral metrics 128 are calculated continuously in real time by the DAM 113 and stored in the user profile 116. The timing associated with all these metrics may also be stored (for example, using timestamps) so that trends can be calculated over time.

[0137] In step 310, operation 300 continues by adapting the application configuration based on information about at least some user behavior and / or result metrics, and / or information about at least some user behavior and / or result metrics in the user database (e.g., users in a stratified group). Step 310 may be performed by AACM 115. For example, in a particular embodiment, AACM 115 may adapt the application configuration 108 based on information about users who exhibit similar behavior metrics in relation to the use of all or at least some of the FFSCs of the application configuration 108. Automatically reconfiguring an application means reconfiguring the application without user action or involvement. For example, an application may be automatically reconfigured by disabling certain FFSCs that were used when application 106 was previously configured, and enabling certain new FFSCs on application 106. In some embodiments, AACM 115 analyzes a dataset related to the stratified group of users selected in step 304 and identifies users who exhibited similar behavior metrics in relation to the use of some or all of the FFSCs of application 106. In other words, AACM115 further stratifies already stratified user groups based on the similarity of behavioral metrics associated with using some or all of the FFSCs in application 106. This further stratification allows AACM115 to predict how a user will behave with respect to FFSCs they have not yet used. This is because if a user and users in this further stratified group exhibit similar behavior with respect to the first set of FFSCs, it is highly likely that the user will exhibit similar behavior with respect to a second set of FFSCs they have not yet used, as exhibited by users in the further stratified group.

[0138] To illustrate this with an example, the application configuration 108 may include exercise management function 1 by setting 2, exercise management function 3 by setting 1, sleep management function 2 by setting 5, reporting function 1 by setting 3, etc. The BEM of users associated with such FFSC may be 20%, 90%, 68%, and 80%, respectively.

[0139] The analysis by AACM115 may then show that users within the stratified groups selected in step 304 who met their goals and exhibited similar behavioral engagement metrics for their precise FFSCs showed a BEM of over 90% for motor management function 1 under setting 5. Therefore, AACM115 may decide to swap motor management function 1 under setting 2 with motor management function 1 under setting 5, but leave the remaining FFSCs unchanged.

[0140] In certain embodiments, the decision to maintain or replace a particular FFSC may depend on whether the FFSC's BEM falls below a certain threshold and / or whether the user is meeting their respective goals. In the example above, the user may have two goals: to lose 5 pounds and to sleep uninterrupted at night. Based on the program outcome metrics for each goal, AACM115 may determine that the user is not meeting the first goal but is meeting the second goal. Therefore, in certain embodiments, AACM115 modifies the exercise management function 1 by setting 2 to help the user meet their goals, because the BEM of the exercise management function 1 by setting 2 is below 70% (e.g., a configured threshold) and the user is not meeting their respective goal of losing 5 pounds. However, in certain embodiments, although the behavioral engagement metrics of the sleep management function 2 by setting 5 are below 70%, AACM115 may decide to maintain the FFSC because the user is meeting their goals. In some embodiments, AACM115 may be configured to replace FFSC with a BEM below a configured threshold, even if the user is meeting the corresponding goals. For example, AACM115 may be configured to replace the sleep management function 2 by setting 5 with a BEM below 70%, even though the user is meeting sleep-related goals.

[0141] In certain embodiments, steps 308 and 310 are performed frequently when real-time or more recent information (e.g., behavioral metrics 128 and outcome metrics 130) is acquired or generated about the user. As a result, in certain embodiments, AACM 15 frequently reconfigures application 106 with FFCS that are more likely to engage the user and thereby help the user achieve their goals. In certain embodiments, if the user changes their goals, AACM 115 may re-execute at least some of the steps of operation 300. For example, AACM 115 may re-execute steps 304 and 306 and replace some of the user's existing FFCS related to goals the user is no longer interested in with new FFSCs related to the user's new goals. In certain embodiments, which FFSCs are related to the user's new goals may be determined in step 306, where AACM 15 identifies FFCS that are highly correlated with the user's goal achievement by evaluating information related to users in a stratified group of users who were able to achieve the same goals, or at least had the same goals.

[0142] In certain embodiments, one of various data models may be used to perform steps 302-310 of operation 300. One exemplary data model is described in relation to Figures 4A-4B. A second exemplary data model is described in relation to Figures 5A-5B. As those skilled in the art will understand, there are various techniques for adapting the application configuration 108 to better assist the user in achieving their goals, avoid stagnation and boredom in the user's behavior toward application 106, and thereby reduce the application drop-off rate. Therefore, the data models described in relation to Figures 4A-5B are merely illustrative.

[0143] Figure 4A shows how an application configuration 108 is initially selected and then adapted using a data model that includes algorithm 450 and model 451, according to a particular embodiment. Figure 4A is illustrated with reference to Figure 4B, which shows some of the datasets used in the operation of the data model in Figure 4A. As described above, during step 302 of operation 300 in Figure 3, the user's goals 132, interests 134 and / or abilities 136 are identified and stored in the user profile 116, which already includes the user's demographic information 118, disease progression 120, and medication information 122. In the example of Figure 4A, at this stage, the user is a new user who has not yet used the functions of application 106, so information about the user's behavior and outcome metrics has not yet been obtained. For the same reason, the user profile 116 does not contain any application configuration information. That is, application 106 is not yet configured at this point to make any application configuration information available. As a result, in the example in Figure 4A, all user-specific information that may be available in step 302 of operation 300 is the user's demographic information 118, disease progression information 120, medication information 122, goals 132, interests 134 and / or abilities 136.

[0144] Using the available information, by performing step 304 of operation 300, AACM 115 stratifies the user database 110, which contains a pool of user profiles, and identifies a group of stratified users similar to the user in a particular embodiment. As those skilled in the art will understand, one of a variety of data filtering techniques or algorithms may be used when stratifying the user database 110. In the example in Figure 4A, AACM 115 stratifies the user database 110 based on a set of stratification factors, which include the user's demographic information 118, disease progression information 120, medication information 122, goals 132, interests 134 and / or abilities 136. For example, the user may be a 77-year-old male with type 1 diabetes who is injecting insulin. The user's goal is to lose 5 pounds, and his interests include running, swimming, and cycling, but he is unable to exercise on weekends. The user may have more than one goal, but for simplicity, algorithms 450 and 451 will be described herein with respect to one goal.

[0145] When the user database 110 is stratified, AACM 115 identifies user group 2 as a stratified group of users, all of whom are male, between 67 and 87 years old, have type 1 diabetes, and inject insulin. Users in user group 2 also all have the goal of losing weight and have achieved that goal. In other words, in the example in Figures 4A-4B, users in user group 2 are selected not only because they have the same goal as the user, but also because they have achieved that goal. As explained above, whether a user has achieved their goal can be indicated by a program outcome metric related to the user's goal. Therefore, in this example, the user's goal achievement is used as a stratification factor. As discussed, goal achievement may also be defined against a threshold program outcome metric. In such an example, the threshold program outcome metric may be used as a stratification factor so that all users with a program outcome metric smaller than a certain threshold (e.g., 70%, 80%, 100%, etc.) are then dropped from the stratified group of users. AACM115 may also take into account user interests 134 and abilities 136 when stratifying the user database 110. For example, all users in user group 2 may have the same interests and abilities.

[0146] In some embodiments, when selecting stratified user groups, AACM115 may be configured to define ranges around each stratification factor. For example, AACM115 may be configured to define a broad user group 2 by including all users whose goal is to lose weight (any amount) and who were able to achieve that goal. In some other examples, AACM115 may be configured to define a narrower user group 2 by including only users whose goal is to lose weight within a specific range from 5 pounds (e.g., any amount in the range of 4-6 pounds or 3-7 pounds) or exactly 5 pounds and who were able to achieve that goal. Regarding other stratification factors, for example, a user is 77 years old, but in a particular embodiment, user group 2 may include all users in the age range of 67-87 years old. In the case of interests and abilities, as an example, in a particular embodiment, AACM115 may also include users whose interests are walking and weightlifting, as well as users who have some time constraints regarding when they can or cannot engage in exercise.

[0147] In some embodiments, stratification may be performed by filtering the datasets available to all users in the user database 110 based on one or more of the stratification factors discussed above. An example of a dataset containing information about all users in the user database 110 is shown as dataset 480 in Figure 4B. As illustrated, each row in dataset 480 belongs to a different user, while each column corresponds to a different data point provided by the user's profile. For example, the columns include the user's goals, demographic information, disease progression, medication information, interests, abilities, and FFSC. While a single column is used for the user's goals, interests, and abilities, it should be noted that each of these columns may represent multiple columns. For example, dataset 480 may include columns related to multiple goals, with each goal-related column corresponding to a specific possible goal that the user could choose. In the example in Figure 4B, the recorded value for each goal corresponds to a program result metric associated with that goal, indicating whether the user met that goal or not. "1" indicates that the user met the goal, and "0" indicates the opposite. For example, dataset 480 indicates that user 1 has achieved their goal. If no value is recorded for a particular goal, it means that the user did not select that goal themselves.

[0148] As discussed, it should be noted that program result metrics may be defined as percentages. Consequently, in certain embodiments, percentages may be recorded in the dataset to indicate program result metrics. In certain other embodiments, 0 and 1 may still be used for simplification by using threshold program result metrics. For example, in certain embodiments, "1" may be recorded in the dataset for any user having a program result metric greater than a certain percentage (e.g., 70%). The dataset 480 also includes a set of FFSCs. For each FFSC, the BEM associated with that FFSC is recorded. For example, for FFSC1 for user 1, a BEM of 90% is recorded, indicating that user 1 is very engaged with FFSC1. As illustrated, no value is recorded for FFSC2 for user 1, indicating that user 1 has not yet used that FFSC.

[0149] Figure 4B also shows dataset 440, which is the result of stratifying dataset 480 based on the stratification factors described above. For example, users 2 and 3 are excluded from dataset 440 because user 2 failed to meet its own goals, while user 3 did not even have the same goals as user 102.

[0150] In a particular embodiment, once a dataset 440 is defined, the AACM 115 then uses an algorithm 450 that takes at least a portion of the dataset 440 as input and outputs a ranked list of FFSCs based on their correlation to the achievement of user goals. How strongly an FFSC correlates with the achievement of user goals may be determined based on several factors, including the number of users in group 2 of users who used the FFSC, and the level of engagement such users with the FFSC (e.g., the corresponding BEM). As an example, the more users who used a particular FFSC, and / or the higher the level of engagement toward the FFSC, the more likely it is that a user (i.e., user 102) will be engaged by that FFSC and thereby achieve their goals. In a simplified example, the correlation (C) of each FFSC with the achievement of user goals may be defined as follows: C=(W1)*(N)+(W2)*(Average BEM)

[0151] In the above formula, W1 is a defined weight, N is the number of users using FFSC, W2 is another defined weight, and the mean BEM is the average of the BEMs of all users in dataset 440 for the corresponding FFSC. Using the above formula, C1 to Cn are calculated for FFSC1 to FFSCN in dataset 440. n The FFSCs are then ranked from highest to lowest so that the FFSCs with the highest correlation can be identified. Thus, algorithm 450 outputs a list of FFSCs ranked based on their correlation to the achievement of the user's goal, which is shown as output 460 in Figure 4A. AACM115 may then select some FFSCs from the list to include in the application configuration 108. For example, AACM115 may select the top 10, or any other defined number, of FFSCs from the list.

[0152] In the examples shown with respect to Figures 4A and 4B, user goal achievement was used as a stratification factor; however, it should be noted that in certain other embodiments, such a method is not essential. For example, in certain embodiments, stratification may be performed based on user goals, rather than goal achievement, such that the stratified group of users includes users who all have the same goal as selected by the user. In such embodiments, the stratified group of users would include users who behaved differently with respect to that goal. In such embodiments, as those skilled in the art will understand, the correlation algorithm may still be used to find the FFSC that has the highest correlation to goal achievement, which may be defined using threshold program results as described above. In yet another certain embodiment, stratification may be performed based on factors unrelated to the user's goal, such as user demographic information or disease progression. Similarly, in such embodiments, the stratified group of users would include users who do not even have the same goal as the user. In such embodiments, as those skilled in the art will recognize, a correlation algorithm may still be used to find the FFSC that has the highest correlation with achieving the goal which may be defined using the threshold program results as described above.

[0153] Once application 106 is configured using application configuration 108, the user begins using the selected FFSC. Over time, as described above, application 106 receives input 127, which includes inputs 210 and 220, and DAM 113 uses this to calculate behavioral metrics 128 and outcome metrics 130. As part of the behavioral metrics 128, a BEM is calculated for each of the FFSCs in application configuration 108. Based on the user's BEM, AACM 115 can further adapt application configuration 108 using model 451 (which may also be referred to as model 451). For example, AACM 115 may create a dataset 442 which includes the user's BEM data 490 and BEM data 441, which is part of a dataset 440 related to users in user group 2. In certain embodiments, by evaluating the similarity between a user's BEM data 490 and other users' BEM data 441, Model 451 can recommend FFSCs that are most likely to engage the user and thereby help the user achieve their goals and improve their health.

[0154] In the example in Figure 4A, Model 451 is a memory-based, user-user collaborative filtering recommendation model (e.g., a type of machine learning model) configured to recommend items (FFSCs in this example) to a user based on the similarity between the user's behavior and the behavior of a pool of users in the dataset. For example, based on the similarity between a user's BEM data 490 and the BEM data of other users in dataset 441, Model 451 may recommend one or more FFSCs that the user has not yet used and that are likely to help the user achieve their goals. The memory-based user-user collaborative filtering algorithm is built on the assumption that a user who behaves similarly for one item (e.g., FFSC1) will also behave similarly for another item (e.g., FFSC3).

[0155] For example, in dataset 442, a user has a BEM of 93% for FFSC1, 18% for FFSC2, and 25% for FFSC N. In this example, model 451 may determine that the user is more likely to behave similarly towards users 1 and 4 than towards user 8. Similarly, in a particular embodiment, model 451 may identify other users who have behaved similarly with respect to the same FFSCs used by user 102, and then determine a set of FFSCs that are more likely to engage the user and thereby help them achieve their goals. This determined set of FFSCs is shown as output 455 in Figure 4A. In a particular embodiment, AACM 115 then uses this new set of FFSCs to reconstruct application configuration 108, resulting in application configuration 408. For example, AACM 115 may remove some of the FFSCs that user 102 did not engage with much and replace them with some of the FFSCs from output 445.

[0156] By adapting application configuration 108 and replacing one or more FFSCs that were not useful to the user with one or more FFSCs from this new set, AACM115 can increase the likelihood that the user will achieve their goals.

[0157] The training of Model 451 may be performed by one or more processors or computing systems communicating data with AACM 115 or the decision support engine 112. Model 451 may be a new model initialized with random weights and parameters, or it may be partially or fully pre-trained (e.g., based on previous training rounds). Model 451 may be trained using an algorithm such as a message passing algorithm. Model 451 may be continuously fine-tuned or retrained as Model 451 continues to provide outputs 445 (e.g., feature recommendations) and the dataset 442 is updated with behavior metrics 128 and outcome metrics 130 for specific users corresponding to the outputs 445. Note that in the example in Figure 4A, data filtering is used to obtain the dataset 440 belonging to group 2 of stratified users, but one of various other techniques may be used instead. For example, in some embodiments, an unsupervised machine learning algorithm may be used to determine which user cluster a user belongs to. For example, an unsupervised machine learning algorithm may be trained over a period of time on a dataset relating to the entire pool of users in a user database 110. The trained unsupervised machine learning algorithm may then be configured to cluster users by taking as input a vector about users containing information about all of the stratification factors described above. As a result of this process, clusters of users (e.g., group 2 of users) are identified, and a dataset (e.g., dataset 440) containing information about the users in that cluster is obtained. As those skilled in the art will recognize, other types of algorithms may also be used to select stratified groups of users and obtain corresponding datasets.

[0158] Figure 5A shows how the application configuration 108 is initially selected and then fitted using a data model that includes machine learning models 550 and 451. Figure 5A is illustrated with reference to Figure 5B, which shows some of the datasets used in the operation of the data model in Figure 5A. As described above, during step 302 in Figure 3, the user's goals 132, interests 134 and / or abilities 136 are identified and stored in the user profile 116, which already includes the user's demographic information 118, disease progression 120 and medication information 122. Similar to Figure 4A, in the example of Figure 5A, all user-specific information that may be available at this stage is the user's demographic information 118, disease progression information 120, medication information 122, goals 132, interests 134 and / or abilities 136.

[0159] In the embodiment shown in Figure 5A, by performing step 304 of operation 300, the AACM 115 stratifies the user database 110 based on the user's demographic information 118, disease progression information 120, and / or medication information 122. As described above, one of various methods may be used for this stratification. The result obtained from the stratification is, for example, a group of users 2, which includes a pool of users similar to the user with respect to demographic information, disease progression information, and medication information.

[0160] In a particular embodiment, after identifying a group of users 2 as a stratified group, the AACM 115 uses user-related information as input to a machine learning model 550, which is specifically trained for the group of users 2, in order to determine an initial set of FFSCs for the application configuration 108. In a particular embodiment, the machine learning model 550 may be trained using one of various datasets. Figure 5B shows a dataset 540 for all users in the group of users 2 who have chosen the same goals as the users and have also achieved those goals. The dataset 540 shows each user's interests and abilities, as well as the BEMs of these users related to the FFSCs that these users interacted with. The machine learning model 550 (i.e., F(X)) is trained by providing information about each user's interests and abilities (e.g., each row portion referred to as X) as input and obtaining a predicted set of BEMs ("PBEM") (e.g., Y') for FFSCs 1-N as output. The predicted set of BEM (e.g., Y') may then be compared to the actual set of BEM users (i.e., each row portion referred to as Y) to determine the error of the machine learning model 550. The machine learning model 550 may then be tuned using a training algorithm based on the error to configure the model to make predictions with higher accuracy. Training of the machine learning model 550 may be performed by one or more processors or computing systems communicating data with AACM 115 or the decision support engine 112.

[0161] For example, during the training process, a vector 542 containing information about user 1's interests and abilities is supplied to the machine learning model 550 as "X". The machine learning model 550 then outputs a vector 544 as "Y'". The actual BEM (i.e., "Y") for user 1 shown in dataset 540 is compared to vector 544 to determine the error, and the machine learning model 550 is trained based on this.

[0162] Similarly, for a user (i.e., user 102), vector 546 may be supplied to a machine learning model 550, which then outputs vector 548 containing a set of PBEMs corresponding to FFSC1-N. In a particular embodiment, AACM115 may then select a subset of FFSC1-N in vector 548 based on the PBEMs. For example, AACM115 may select the FFSCs that have the highest PBEMs in each category. As an example, if there are 10 different FFSCs (e.g., FFSC1-10) in vector 548 relating to exercise management, AACM115 may select the two with the highest PBEMs. In a particular embodiment, AACM115 then configures application 106 using application configuration 108 containing the selected subset of FFSCs.

[0163] In the example in Figure 5B, it should be noted that the input to the machine learning model 550 includes only information about the user's interests and abilities, because the dataset is already stratified with other types of information such as demographic information, disease progression, medication information, goals, and goal achievement. However, in other examples, additional inputs may be included in the dataset used to train the machine learning model 550. Similarly, such additional inputs may be included when information related to user 102 is supplied to the model 550 to obtain a set of PBEMs.

[0164] Once application 106 is configured using application configuration 108, the user begins using the corresponding FFSC over time, during which behavioral metrics 128 and outcome metrics 130 are acquired. As part of the behavioral metrics 128, a BEM is calculated for each of the FFSCs in application configuration 108. At this stage, AACM 115 possesses information about the user's behavior, which AACM 115 can use to further adapt application configuration 108. For example, as described above with respect to Figure 4B, AACM 115 may generate a dataset 442 that AACM 115 can use as input to recommendation model 451. Recommendation model 451 then generates an output 560 corresponding to a set of FFSCs that are likely to engage the user and thereby help the user achieve their goals. As described above, AACM 115 then uses this new set of FFSCs to reconstruct application configuration 108, resulting in application configuration 508.

[0165] Figure 6 is a block diagram of a computer 600 configured to selectively and sequentially adapt the application configuration of an application run by either the computer 600 or another computer communicating with the computer 600, according to a particular embodiment disclosed herein. Although depicted as a single physical device, in embodiments the computer 600 may be implemented using virtual devices and / or across several devices, such as in a cloud environment. As shown, the computer 600 includes a processor 605, memory 610, storage device 615, network interface 625, and one or more I / O interfaces 620. In the illustrated embodiment, the processor 605 retrieves and executes programming instructions stored in memory 610 and further stores and retrieves application data residing in storage device 615. The processor 605 generally represents a single CPU and / or GPU, multiple CPUs and / or GPUs, a single CPU and / or GPU with multiple processing cores, etc. Memory 610 is generally included as representing random access memory. The storage device 615 may be any combination of disk drives, flash-based storage devices, etc., and may include fixed and / or removable storage devices such as fixed disk drives, removable memory cards, caches, optical storage devices, network-attached storage (NAS), or storage area networks (SAN).

[0166] In some embodiments, input / output (I / O) devices 635 (such as a keyboard or monitor) may be connected via an I / O interface 620. Furthermore, the computing device 600 may be communicatively coupled to one or more other devices and components, such as a user database 110, via a network interface 625. In certain embodiments, the computing device 600 is communicatively coupled to other devices via a network, which may include the Internet and local networks. The network may include wired connections, wireless connections, or a combination of wired and wireless connections. As shown in the figure, the processor 605, memory 610, storage device 615, network interface 625, and I / O interface 620 are communicatively coupled by one or more interconnections 630. In certain embodiments, the computing device 600 represents a mobile device 107 associated with a user. In certain embodiments, as discussed above, the mobile device 107 may include the user's laptop, computer, and smartphone. In another embodiment, the computing device 600 is a server running in a cloud environment.

[0167] In the illustrated embodiment, the storage device 615 includes a user profile 116. The memory 610 includes a decision support engine 112 which itself includes an AACM 115 and a DAM 113. The decision support engine 112 is executed by the computing device 600 to perform operations 304-310 of operation 300 in Figure 3. The AACM 115 is configured with or includes any algorithms (e.g., algorithms 450 and 451) necessary for the operation of the data model described in relation to Figures 4A-5B.

[0168] Accordingly, certain embodiments described herein provide technical solutions to technical problems in the art of personalizing or adapting the configuration of diabetes or health-related software applications (e.g., mobile applications) to help improve a user's health or manage a disease. Automatically reconfiguring an application by changing a set of functions and function settings (e.g., making some available and others unavailable) based on the user's own information and information related to a specific set of users similar to the user in one or more embodiments is a technical improvement to (1) how health-related software applications operate, and (2) how health monitoring systems, including glucose monitoring systems and health-related software applications, operate. For example, as described above, in certain embodiments, an application configuration having a set of FFSCs may be initially selected for a user based on the user's own information (e.g., goals, interests, abilities, demographic information, disease progression, medication information, etc.) and / or information related to a specific set of users similar to the user in one or more embodiments. In certain embodiments, this specific set of users is users who have / had and / or achieved the same goals as the user. In certain embodiments, information relating to a particular set of users includes user behavior information related to FFSC used by the user, and / or whether the user was able to achieve the same goals. Furthermore, in certain embodiments, the initial application configuration may be frequently reset based on (1) the user's own behavior information and information relating to the user's health and / or performance toward the user's goals, and (2) users who have shown similar behavior toward those same goals and / or health-related performance toward the same goals.

[0169] In certain embodiments, the technical field of personalizing or adapting the configuration of diabetes or health-related software applications is improved by configuring the application in an initial configuration using the information described above, and by frequently reconfiguring the application based on the additional information described above and the techniques described herein. Applications configured based on the embodiments described herein are likely to engage users and thereby help improve their health. It should be noted that, with respect to health-related, particularly disease management applications, personalizing the application based on the information and techniques / algorithms described herein can make a significant difference to the user's life and health, to the extent that in some cases the application's guidance may help save the user's life. Therefore, the difference between an application that is not personalized or not effectively personalized and an application that is personalized based on the embodiments described herein may be the difference between a user who stops using the application without engaging and thereby may suffer a deterioration in their health, and a user who engages with the application and is able to effectively manage their disease and improve their health through the personalized guidance and functions provided to them.

[0170] Each of these non-limiting examples can stand on its own or can be combined with one or more of the other examples in various permutations or combinations. The above detailed description includes references to the accompanying drawings, which form part of the detailed description. The drawings illustrate, as examples, specific embodiments in which the present invention may be carried out. These embodiments are also referred to herein as “examples.” Such examples may include elements in addition to those illustrated or described. However, the inventors also envision examples comprising only the illustrated or described elements. Furthermore, the inventors also envision examples using any combination or permutation of the illustrated or described elements (or one or more embodiments thereof) with respect to a particular example (or one or more embodiments thereof) or to other examples (or one or more embodiments thereof) illustrated or described herein.

[0171] In the event of any conflict between the usage described herein and any document incorporated herein by reference, the usage described herein shall prevail.

[0172] In this specification, as is common in patent documents, the term “a or an” is used to include one or more, independently of any other examples or uses of “at least one” or “one or more.” In this specification, the term “or” is used to refer to non-exclusive or, unless otherwise specified, such that “A or B” includes “A exists but B does not,” “B exists but A does not,” and “A and B.” In this specification, the terms “including” and “in which” are used as plain English equivalents of the terms “comprising” and “wherein,” respectively. Furthermore, in the following claims, the terms “including” and “comprising” are open-ended, meaning that a system, apparatus, article, composition, formulation or process containing elements in addition to the elements described after such terms in a claim is still considered to be within the scope of that claim. Furthermore, in the following claims, terms such as "first," "second," and "third" are used merely as labels and are not intended to impose numerical requirements on the subject.

[0173] Geometric terms such as "parallel," "perpendicular," "round," or "square" are not intended to require absolute mathematical precision unless the context indicates otherwise. Instead, such geometric terms allow for manufacturing variability or equivalent functionality. For example, if an element is described as "round" or "generally round," components that are not perfectly circular (e.g., slightly oval or polygonal) are still included in this description.

[0174] Examples of the methods described herein can be implemented at least partially by machine or computer. Some examples may include computer-readable or machine-readable media coded with instructions that can be operated to configure an electronic device to perform the methods described above. Implementations of such methods may include code such as microcode, assembly language code, or high-level language code. Such code may include computer-readable instructions for performing various methods. The code may form part of a computer program product. Furthermore, in one example, the code may be tangibly stored, for example, during execution or at other times, on one or more volatile, non-temporary, or non-volatile tangible computer-readable media. Examples of these tangible computer-readable media may include, but are not limited to, hard disks, removable magnetic disks, removable optical disks (e.g., compact disks and digital video disks), magnetic cassettes, memory cards or sticks, random access memory (RAM), and read-only memory (ROM).

[0175] The above description is illustrative and not intended to be limiting. For example, the examples (or one or more of them) described above may be used in combination with one another. Other embodiments may be used, for example, by those skilled in the art when considering the above description. The abstract is provided in accordance with 37 CFR §1.72(b) to enable the reader to quickly confirm the nature of the technical disclosure. The abstract is submitted with the understanding that it is not to be used to interpret or limit the scope or meaning of the claims. Also, in the above detailed description, various functions may be grouped together in order to simplify the disclosure. This should not be interpreted as meaning that any disclosed function not claimed is essential to any claim. Rather, the subject matter of the invention may not be present in all functions of a particular disclosed embodiment. Accordingly, the following claims are incorporated herein as examples or embodiments, and each claim is assumed to stand as a separate embodiment in itself, and such embodiments are assumed to be combined with one another in various combinations or permutations. The scope of the invention should be determined by reference to the appended claims, along with the full scope of equivalents to which such claims are granted.

Claims

1. It is a system, Memory circuit and It is a processor, Receiving a request to configure an application for use by a user, the application being at least partially resident on a computing device for managing sensor data generated by a glucose monitoring system associated with the user, Identifying the user's goals, Identifying classification information related to the user, wherein the classification information includes at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information. The selection of a group of users similar to the user from a pool of user profiles in a user database, wherein the pool of user profiles includes the classification information and multiple application functions associated with each user in the user database, and the selection is based on one or more similarities between the user and the group of users regarding the identified classification information, and each program result metric in the pool of user profiles regarding the objective, wherein the program result metric indicates whether each user in the user database has met the objective. Identifying one or more application functions from the plurality of application functions, wherein the identification is based on the user's goals and the correlation between each of the plurality of application functions and the goals in a dataset related to the user group, the dataset being stored in the user database and including the classification information and the plurality of application functions related to the user group, A system comprising a processor configured to automatically configure the application using one or more of the aforementioned application functions.

2. The fact that the processor is configured to identify the target means that the processor, The system receives user input regarding the user's diabetes and what the user intends to achieve. The system according to claim 1, comprising being configured to convert the user input into the target based on one or more defined guidelines.

3. The processor is configured to convert the user input to the target based on one or more defined guidelines, Categorizing the user into categories based on the aforementioned guidelines and information related to the user, The information relating to the user includes the classification information, The aforementioned guidelines indicate the objectives of the aforementioned categories, and categorize them. The system according to claim 2, comprising being configured to select the user's goal based on the categorization.

4. The fact that the processor is configured to identify the target means that the processor, Receiving user input regarding what the user intends to achieve with respect to the user's diabetes, The system according to claim 1, comprising being configured to convert the user input into the target based on information relating to the group of users.

5. The system according to claim 4, wherein the information relating to the group of users includes one or more glucose-related metrics for the group of users.

6. The system according to claim 1, wherein the correlation between each of the plurality of application functions and the objective includes the correlation between each of the plurality of application functions and the achievement of the objective.

7. The program result metrics of the selected group of users exceed the threshold program result metrics related to achieving the goal. The system according to claim 1, wherein the threshold program result metric related to the achievement of the objective represents a defined minimum amount of positive progress toward the achievement of the objective.

8. The system according to claim 1, wherein the correlation between each of the plurality of application functions and the goal is based on the number of users in the selected group of users who used the function, and the behavioral engagement of that number of users with respect to the function.

9. The behavioral engagement of each of the aforementioned number of users with respect to the application function is represented by the Behavioral Engagement Metric (BEM) for each of the aforementioned number of users with respect to the application function, and the BEM is based on the interaction between each of the aforementioned number of users and the function, and the interaction is The frequency with which each of the aforementioned number of users interacts with the application function, The frequency with which each of the aforementioned number of users ignores the guidance generated by the application function, The average amount of time each of the aforementioned number of users spends interacting with the application features, or The system according to claim 8, comprising at least one of the following: how faithful the actions of each of the aforementioned number of users are to the guidance generated by the application function.

10. The system according to claim 1, wherein each of the one or more application functions has a correlation with the target that exceeds a correlation threshold.

11. The aforementioned processor, Receiving multiple inputs, wherein the multiple inputs are A first input, which includes a glucose measurement value related to the user, generated by the glucose monitoring system, and Receiving includes a second input indicating the user's actions with respect to one or more application functions, The calculation of program result metrics related to the objective, based at least on the first input, wherein the program result metrics indicate the degree to which the user has achieved the objective. Based on the second input, one or more behavioral engagement metrics (BEMs) for the one or more application functions are calculated such that a separate BEM is calculated for each of the one or more application functions. Identifying one or more users in the selected group of users or the pool of user profiles that have a BEM similar to the one or more BEMs calculated above, Identifying new application functions not included in the aforementioned one or more functions based on the fact that the function is associated with a BEM that exceeds a threshold for at least one of the aforementioned one or more users, The system according to claim 1, further configured to reconfigure the application using the new application functionality based on at least one of the one or more BEMs and the program result metrics.

12. Each of the one or more BEMs is based on an interaction between the user and a corresponding application function among the one or more application functions, and the interaction is The frequency with which the user interacts with the corresponding application function, The frequency with which the user ignores the guidance generated by the corresponding application function, The average amount of time the user spends interacting with the corresponding application function, or The system according to claim 11, comprising at least one of the following: how faithful the user's actions are to the guidance generated by the corresponding application function.

13. The fact that the processor is configured to reconfigure the application using the new application functionality means that the processor Identifying the underperforming application function among the one or more application functions that has a corresponding BEM below a threshold, The system according to claim 11, further comprising being configured to replace the low-performing application function among the one or more application functions with the new application function.

14. The fact that the processor is configured to reconfigure the application using the new application functionality means that the processor Identifying that the low-performing application function among the one or more application functions is related to the objective, The system according to claim 13, comprising being configured to identify that the program result metric is below a threshold.

15. The system according to claim 1, wherein each of the one or more application functions includes function settings.

16. A method for configuring an application using one or more application functions executed by a computer, Receiving a request to configure the application for use by the user, the application being at least partially resident on the computing device for managing sensor data generated by a glucose monitoring system associated with the user, Identifying the user's goals, Identifying classification information related to the user, wherein the classification information includes at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information. The selection of a group of users similar to the user from a pool of user profiles in a user database, wherein the pool of user profiles includes the classification information and multiple application functions associated with each user in the user database, and the selection is based on one or more similarities between the user and the group of users regarding the identified classification information, and each program result metric in the pool of user profiles regarding the objective, wherein the program result metric indicates whether each user in the user database has met the objective. Identifying one or more application functions from the plurality of application functions, wherein the identification is based on the user's goals and the correlation between each of the plurality of application functions and the goals in a dataset related to the user group, the dataset being stored in the user database and including the classification information and the plurality of application functions related to the user group, A method comprising configuring the application using one or more of the aforementioned application functions.

17. Identifying the aforementioned target means Receiving user input regarding what the user intends to achieve with respect to the user's diabetes, The method according to claim 16, further comprising converting the user input to the target based on one or more defined guidelines.

18. The aforementioned conversion is performed by Categorizing the user into categories based on the aforementioned guidelines and information related to the user, The information relating to the user includes classification information, The aforementioned guidelines indicate the objectives of the aforementioned categories, and categorize them. The method according to claim 17, further comprising selecting the user's goal based on the categorization.

19. Identifying the aforementioned target means Receiving user input regarding what the user intends to achieve with respect to the user's diabetes, The method according to claim 16, comprising converting the user input into the target based on information relating to the group of users.

20. The method according to claim 19, wherein the information relating to the group of users includes one or more glucose-related metrics for the group of users.

21. The method according to claim 16, wherein the correlation between each of the plurality of application functions and the objective includes the correlation between each of the plurality of application functions and the achievement of the objective.

22. The program result metrics of the selected group of users exceed the threshold program result metrics related to achieving the goal. The method according to claim 16, wherein the threshold program result metric related to the achievement of the objective indicates a defined minimum amount of positive progress toward the achievement of the objective.

23. The method according to claim 16, wherein the correlation between each of the plurality of application functions and the goal is based on the number of users in the selected group of users who used the function and the behavioral engagement of that number of users with respect to the function.

24. The behavioral engagement of each of the aforementioned number of users with respect to the application function is represented by the Behavioral Engagement Metric (BEM) for each of the aforementioned number of users with respect to the application function, and the BEM is based on the interaction between each of the aforementioned number of users and the function, and the interaction is The frequency with which each of the aforementioned number of users interacts with the application function, The frequency with which each of the aforementioned number of users ignores the guidance generated by the application function, The average amount of time each of the aforementioned number of users spends interacting with the application features, or The method according to claim 23, comprising at least one of the following: how faithful the actions of each of the aforementioned number of users are to the guidance generated by the application function.

25. The method according to claim 16, wherein each of the one or more application functions has a correlation with the target that exceeds a correlation threshold.

26. Receiving multiple inputs, wherein the multiple inputs are A first input, which includes a glucose measurement value related to the user, generated by the glucose monitoring system, and Receiving includes a second input indicating the user's actions with respect to one or more application functions, The calculation of program result metrics related to the objective, based at least on the first input, wherein the program result metrics indicate the degree to which the user has achieved the objective. Based on the second input, one or more behavioral engagement metrics (BEMs) for the one or more application functions are calculated such that a separate BEM is calculated for each of the one or more application functions. Identifying one or more users in the selected group of users or the pool of user profiles that have a BEM similar to the one or more BEMs calculated above, Identifying new application functions not included in the aforementioned one or more functions based on the fact that the function is associated with a BEM that exceeds a threshold for at least one of the aforementioned one or more users, The method according to claim 16, further comprising reconfiguring the application using the new application functionality based on at least one of the one or more BEMs and the program result metrics.

27. Each of the one or more BEMs is based on an interaction between the user and a corresponding application function among the one or more application functions, and the interaction is The frequency with which the user interacts with the corresponding application function, The frequency with which the user ignores the guidance generated by the corresponding application function, The average amount of time the user spends interacting with the corresponding application function, or The method according to claim 26, comprising at least one of the following: how faithful the user's actions are to the guidance generated by the corresponding application function.

28. Reconfiguring the application using the new application functionality means Identifying the underperforming application function among the one or more application functions that has a corresponding BEM below a threshold, The method according to claim 26, comprising replacing the low-performing application function among the one or more application functions with the new application function.

29. Reconfiguring the application using the new application functionality means Identifying that the low-performing application function among the one or more application functions is related to the objective, The method according to claim 28, comprising identifying that the program result metric is below a threshold.

30. The method according to claim 16, wherein each of the one or more application functions includes function settings.

31. A non-temporary computer-readable medium storing instructions that, when executed by a processor, cause a computing system to execute a method for configuring an application using one or more application functions, wherein the method is Receiving a request to configure the application for use by the user, the application being at least partially resident on the computing device for managing sensor data generated by a glucose monitoring system associated with the user, Identifying the user's goals, Identifying classification information related to the user, wherein the classification information includes at least one of the user's goals, interests, abilities, demographic information, disease progression information, or medication plan information. The selection of a group of users similar to the user from a pool of user profiles in a user database, wherein the pool of user profiles includes the classification information and multiple application functions associated with each user in the user database, and the selection is based on one or more similarities between the user and the group of users regarding the identified classification information, and each program result metric in the pool of user profiles regarding the objective, wherein the program result metric indicates whether each user in the user database has met the objective. Identifying one or more application functions from the plurality of application functions, wherein the identification is based on the user's goals and the correlation between each of the plurality of application functions and the goals in a dataset related to the user group, the dataset being stored in the user database and including the classification information and the plurality of application functions related to the user group, A non-temporary computer-readable medium, which includes configuring the application using one or more of the aforementioned application functions.

32. Identifying the aforementioned target means Receiving user input regarding what the user intends to achieve with respect to the user's diabetes, A non-temporary computer-readable medium according to claim 31, further comprising converting the user input to the target based on one or more defined guidelines.

33. The aforementioned conversion is performed by Categorizing the user into categories based on the aforementioned guidelines and information related to the user, The information relating to the user includes classification information, The aforementioned guidelines indicate the objectives of the aforementioned categories, and categorize them. The non-temporary computer-readable medium according to claim 32, further comprising selecting the user's objective based on the categorization.

34. Identifying the aforementioned target means Receiving user input regarding what the user intends to achieve with respect to the user's diabetes, The non-temporary computer-readable medium according to claim 31, further comprising converting the user input to the target based on information relating to the group of users.

35. The non-temporary computer-readable medium according to claim 34, wherein the information relating to the group of users includes one or more glucose-related metrics for the group of users.

36. The non-temporary computer-readable medium according to claim 31, wherein the correlation between each of the plurality of application functions and the objective includes the correlation between each of the plurality of application functions and the achievement of the objective.

37. The program result metrics of the selected group of users exceed the threshold program result metrics related to achieving the goal. The non-temporary computer-readable medium according to claim 31, wherein the threshold program result metrics related to the achievement of the objective indicate a defined minimum amount of positive progress toward the achievement of the objective.

38. The non-temporary computer-readable medium according to claim 31, wherein the correlation between each of the plurality of application functions and the objective is based on the number of users in the selected group of users who used the function and the behavioral engagement of that number of users with respect to the function.

39. The behavioral engagement of each of the aforementioned number of users with respect to the application function is represented by the Behavioral Engagement Metric (BEM) for each of the aforementioned number of users with respect to the application function, and the BEM is based on the interaction between each of the aforementioned number of users and the function, and the interaction is The frequency with which each of the aforementioned number of users interacts with the application function, The frequency with which each of the aforementioned number of users ignores the guidance generated by the application function, The average amount of time each of the aforementioned number of users spends interacting with the application features, or The non-temporary computer-readable medium according to claim 38, comprising at least one of the following: how faithfully each user's actions are to the guidance generated by the application function of the aforementioned number of users.

40. The non-temporary computer-readable medium according to claim 31, wherein each of the one or more application functions has a correlation with the target that exceeds a correlation threshold.

41. The aforementioned method, Receiving multiple inputs, wherein the multiple inputs are A first input, which includes a glucose measurement value related to the user, generated by the glucose monitoring system, and Receiving includes a second input indicating the user's actions with respect to one or more application functions, The calculation of program result metrics related to the objective, based at least on the first input, wherein the program result metrics indicate the degree to which the user has achieved the objective. Based on the second input, one or more behavioral engagement metrics (BEMs) for the one or more application functions are calculated such that a separate BEM is calculated for each of the one or more application functions. Identifying one or more users in the selected group of users or the pool of user profiles that have a BEM similar to the one or more BEMs calculated above, Identifying new application functions not included in the aforementioned one or more functions based on the fact that the function is associated with a BEM that exceeds a threshold for at least one of the aforementioned one or more users, The non-temporary computer-readable medium according to claim 31, further comprising reconfiguring the application using the new application functionality based on at least one of the one or more BEMs and the program result metrics.

42. Each of the one or more BEMs is based on an interaction between the user and a corresponding application function among the one or more application functions, and the interaction is The frequency with which the user interacts with the corresponding application function, The frequency with which the user ignores the guidance generated by the corresponding application function, The average amount of time the user spends interacting with the corresponding application function, or The non-temporary computer-readable medium according to claim 41, comprising at least one of the following: how faithful the user's actions are to the guidance generated by the corresponding application function.

43. Reconfiguring the application using the new application functionality means Identifying the underperforming application function among the one or more application functions that has a corresponding BEM below a threshold, A non-temporary computer-readable medium according to claim 41, comprising replacing the low-performing application function among the one or more application functions with the new application function.

44. Reconfiguring the application using the new application functionality means Identifying that the low-performing application function among the one or more application functions is related to the objective, The non-temporary computer-readable medium according to claim 43, comprising identifying that the program result metric is below a threshold.

45. The non-temporary computer-readable medium according to claim 31, wherein each of the one or more application functions includes function settings.

Images