Systems and methods for analyzing, interpreting, and acting on continuous glucose monitoring data

Abstract

To provide a method and system for optimizing a glucose state of a user via a mobile application.SOLUTION: The method includes: receiving, using a continuous glucose monitoring (CGM) device, a glucose value of a user; determining a time-in-range (TIR) value; determining a TIR state; receiving a glucose variability (GV) value; determining a GV state; determining an onset state based on the TIR state and the GV state; determining that an onset state corresponds to a non-ideal state; generating an optimization path to reach an ideal state based on one or more account vectors, such as addressing self-managed behavior including meals, activities, and medication use; and providing the optimization path directly to a patient. The optimization path is based on computer detection and classification of important events of interest over time.SELECTED DRAWING: Figure 5A

Description

[Technical Field]

[0001] (Related Applications) This application relates to 1) U.S. Provisional Application No. 63 / 135,818, filed January 11, 2021 2) U.S. Provisional Application No. 62 / 992,385, filed March 20, 2020; and ) A continuation of U.S. Provisional Application No. 62 / 992,409, filed March 20, 2020 and claims the benefit of priority to US Pat. No. 6,229,999, each of which is incorporated herein by reference in its entirety. Use.

[0002] The present disclosure generally relates to a method for collecting data to generate optimized routes and improve a user's health. In some embodiments, the acquiring and processing may be performed in particular by a mobile application. The present invention relates to optimizing a user's glucose status via [Background technology]

[0003] Rising medical costs are limiting users' access to appropriate care. Care companies increase the workload of providers and limit interaction between doctors and users. Treatment of urinary diseases often does not provide sufficient data to effectively provide treatment options. relies on sporadic measurements (e.g., glucose measurements) that are often For example, changes are recommended based on only a few measurements and are used alone. Therefore, any medical, dietary, and / or lifestyle recommendations made as a result of a given measurement may be misleading. The variation in the IL is limited given the sparse data received through sporadic measurements. There are.

[0004] The present disclosure is directed to addressing one or more of the above problems. The introduction provided is intended to generally set out the context of the disclosure. Unless otherwise indicated, the material in this section is not intended to be prior art to the claims of this application. , which are prior art or an indication of prior art by their inclusion in this section. It is not an admission of this. Summary of the Invention

[0005] The present disclosure is directed to a computer-implemented method for managing a user's glucose status. , receiving a user's blood glucose level using a continuous glucose monitoring (CGM) device; and determining a time-in-range (TIR) ​​value for the user's blood glucose level, the TIR value being The determination is based on the amount of time the user's blood glucose level is within a threshold band over a reference period. determining a TIR state based on the TIR value; and determining a TIR state based on at least the user's blood glucose level. and receiving a blood glucose variation (GV) value based on the GV value, the GV value being determined by a standard deviation or coefficient of variation (C V), and CV takes into account the standard deviation of blood glucose levels over a reference period. and determining a GV status based on the GV value. determining a starting condition based on the TIR condition and the GV condition; and determining whether the starting condition is a non-ideal condition. determining whether the state corresponds to an ideal state based on one or more account vectors; generating an optimized path to reach the target state, the optimized path comprising one or generating an optimized path, the path including one or more adjustments of a plurality of account vectors; and providing the information to the user.

[0006] The threshold band may be between approximately 70 mg / dL and 180 mg / dL, and the reference period The CV value is the standard deviation of blood glucose levels over a 24-hour period. The TIR condition may be determined by dividing by the TIR value. A good TIR state may be a binary state selected from one of the TI The GV condition may correspond to a TIR value greater than the R threshold. A good GV state may be a binary state selected from one of the GV states. The account vector may correspond to a GV value greater than a threshold. One or more of the following parameters may be used: intake, physical activity, psychosocial parameters, or social determinants parameters. The account vector may contain one or more categorized based on severity score. The optimized pathway may include blood glucose levels based on multiple CGM events. The user attributes can be social attributes, medical attributes, user preferences, metabolic attributes, or user preferences. The optimized route is selected from one or more of the following conditions: The improvement may include an increase in a condition-improving habit and / or a decrease in one or more condition-deteriorating habits.

[0007] The present disclosure is directed to a computer-implemented method for managing a user's glucose status. Generate multiple optimization profiles to reach an ideal state from a non-ideal state. The ideal state is a good time-in-range (TIR) ​​state and good blood glucose variability (GV ) condition, and a non-ideal condition corresponds to at least one of a bad TIR condition or a bad GV condition. Also includes one that generates and displays the current TIR status based on the user's blood glucose TIR value. determining a state of the blood glucose level of the user within a threshold band over a reference period, The current TIR status is determined based on the amount of time the R state and the GV value associated with the user's blood glucose level. determining the current GV status based on the standard deviation (S D) or coefficient of variation (CV), which takes into account the standard deviation of blood glucose levels over the reference period. and determining, indicative of fluctuations in the user's blood glucose level, one or more accounts of the user. receiving a vector; and one or more account vectors, a TIR state, and a CV. identifying one of the optimization profiles based on the state; and identifying an optimized route based on the profile, the optimized route comprising one or more Identifying and optimizing a path for a user, including adjusting one or more of the multiple accounting vectors. and providing the information to the user.

[0008] The optimization profiles receive an account vector as input and a machine learning model configured to output one or more adjustments to the count vector; The multiple optimization profiles may be generated by: by associating one or more adjustments of the Each of the multiple optimization profiles may be further generated based on a potential TIR condition. , which may correspond to a potential GV state, and one or more potential account vectors. One or more user attributes may be received, and one of the optimization profiles may include one Alternatively, the CV value may be determined based on a plurality of user attributes. may be determined by dividing by the average blood glucose level over the reference period.

[0009] The present disclosure is also directed to a system for managing a user's blood glucose level, the system comprising: a memory for storing processor-readable instructions; and a processor for accessing the memory and executing the processor-readable instructions. and a processor configured to execute the processor-readable instructions. When run, it uses a component that penetrates the user's skin to obtain glucose readings. A user may use a continuous glucose monitoring (CGM) device configured to electronically receiving the user's blood glucose level and determining the user's time-in-range (TIR) ​​value the TIR value is determined when the user's blood glucose level is within a threshold band for a reference period. The threshold range is approximately between 70 mg / dL and 180 mg / dL. The reference period is 24 hours. The TIR state is determined based on the TIR value. determining whether the TIR condition is a good TIR condition or a bad TIR condition; and receiving a blood glucose variability (GV) value based on at least the user's blood glucose level. wherein the GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is , taking into account the standard deviation of blood glucose levels over a reference period, a received and determining a GV state based on the GV value, wherein the GV state is a good GV state. determining whether the TIR condition is a good GV condition or a bad GV condition; determining a starting state based on the detecting CGM events based on the user's blood glucose levels; and Based on one or more of the following: GM classification or severity and CGM event trace shape characterization In this case, characterizing CGM events, multiparameter CGM classification is Characterize the event, including the blood glucose level at the start, severity, and end of the CGM event. and one or more account vectors and a reasoning based on characterizing the CGM event. generating an optimized path to reach the desired state, the optimized path being one or more generating an optimized path including one or more adjustments of a plurality of account vectors; Providing the optimized route to the user includes providing the optimized route to the user. and providing context-based instructions to a user based on the context-based instructions. The processor may be configured as follows. [Brief explanation of the drawings]

[0010] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate examples of the present disclosure. , together with the detailed description, serve to explain the principles of the present disclosure. [Figure 1] FIG. 1 is a schematic diagram of a health management system according to one embodiment of the present disclosure. [Figure 2] FIG. 2 is a schematic diagram of a portion of the health care system of FIG. [Figure 3A] FIG. 3A is a schematic diagram of another portion of the health care system of FIG. [Figure 3B] FIG. 3B is a schematic diagram of training an exemplary machine learning model, according to one embodiment of the present disclosure. [Figure 4A] FIG. 4A is a continuous glucose monitoring (CGM) chart according to one embodiment of the present disclosure. [Figure 4B] FIG. 4B is a continuous glucose monitoring (CGM) report according to one embodiment of the present disclosure. [Figure 5A] FIG. 5A is a flowchart of a health management method according to one embodiment of the present disclosure. [Figure 5B]FIG. 5B is a flowchart of an exemplary health management method according to another embodiment of the present disclosure. [Figure 6A] FIG. 6A is a graph of patient status according to another embodiment of the present disclosure. [Figure 6B] FIG. 6B is a graph of patient condition over time correlated with changes in patient condition according to another embodiment of the present disclosure. [Figure 6C] FIG. 6C shows three graphs of patient condition over time according to another embodiment of the present disclosure. [Figure 6D] FIG. 6D shows a graph of standard deviation and a graph of coefficient of variation according to another embodiment of the present disclosure. [Figure 6E] FIG. 6E illustrates a condition distribution for multiple patients according to another embodiment of the present disclosure. [Figure 6F] FIG. 6F illustrates a change in the distribution of conditions for multiple patients according to another embodiment of the present disclosure. [Figure 6G] FIG. 6G shows the change in glucose levels over time and the corresponding first derivative graph according to another embodiment of the present disclosure. [Figure 7A] FIG. 7A shows average glucose, a range chart, and fluctuations in glucose and range charts according to another embodiment of the present disclosure. [Figure 7B] FIG. 7B shows a continuous glucose monitoring activation time and variability chart according to another embodiment of the present disclosure. [Figure 8A] FIG. 8A shows a continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8B] FIG. 8B shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8C] FIG. 8C shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8D] FIG. 8D shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 9] FIG. 9 is another flowchart of a health management method according to one embodiment of the present disclosure. [Figure 10A]FIG. 10A is a visualization of a continuous glucose monitoring event according to one embodiment of the present disclosure. [Figure 10B] FIG. 10B is a visualization of a meal monitoring event according to one embodiment of the present disclosure. [Figure 11] FIG. 11 is a visualization of severity counts according to one embodiment of the present disclosure. [Figure 12] FIG. 12 is an automatic alert generation chart according to one embodiment of the present disclosure. [Figure 13A] FIG. 13A is a screenshot of an example message according to one embodiment of the present disclosure. [Figure 13B] FIG. 13B is another screenshot of an example message according to one embodiment of the present disclosure. [Figure 13C] FIG. 13C is another screenshot of an exemplary message according to an embodiment of the present disclosure. [Figure 13D] FIG. 13D is a glucose computer according to one embodiment of the present disclosure. [Figure 14] FIG. 14 is a simplified functional block diagram of a computer that may be configured, for example, as a host server, to function as a healthcare provider decision-making server, according to an embodiment of the present disclosure.

[0011] The appendix is ​​provided with this specification and contains a description using examples of the present disclosure, including experimental results. . DETAILED DESCRIPTION OF THE INVENTION

[0012] Reference will now be made in detail to the embodiments of the present disclosure, which are illustrated in the accompanying drawings. The same reference numbers are used throughout the drawings to refer to the same or like parts. .

[0013] In the discussion below, relative terms such as "about," "substantially," "approximately," and the like are used to is used to indicate a possible variation of ±10% in the numerical values ​​given. The foregoing descriptions are merely exemplary in nature and may not be construed as limiting the scope of the present subject matter, or as limiting the scope of applications and / or features of such embodiments. Please note that the present specification is not intended to limit the scope and use of the present invention. Any implementation described herein should not be construed as preferred or advantageous over other implementations. Rather, as mentioned above, the term "exemplary" should not be construed as meaning "ideal." The term "comprises" is used in the sense of an example or "is an example" rather than "includes." "," "include," "having," "with," and any variations thereof are non-exclusive. are used synonymously to indicate or describe other inclusions. A process, method, article, or apparatus using such terms is not intended to be a representation of its steps, structures, or does not include only elements not expressly described or such processes, methods, It may include other steps, structures, or elements inherent to the article or apparatus. The terms "first," "second," etc. in this document do not denote any order, quantity, or importance. It is not intended to be limiting unless expressly stated otherwise and is used to distinguish one element from another. The use of the singular terms "a" and "an" herein does not denote a limitation of quantity but This indicates the presence of at least one of the items listed.

[0014] (Healthcare and Computing Environments) FIG. 1 is a block diagram of a health management system 100 according to one embodiment of the present disclosure. Mobile devices, computers, medical devices, or electronic networks such as the Internet any other electronic device configured to access the network 32. A user 8 (e.g., a patient, consumer, etc.) with a device 19 may use mobile health (mH health) application 1 or otherwise accessible In some embodiments, network 32 may be a cellular network. , Wi-Fi, LAN, WAN, Bluetooth (registered trademark), Near Field Communication (N FC), or other suitable form of network communication. A plurality of electronic devices 19 are configured to access an electronic network 32. Users 8 may use various devices (e.g., mobile phones, tablets, and laptop computers) A single application linked to multiple electronic devices 19 (through one or more of the computers) The mHealth application 1 may be accessed through the account. Also included are mobile health devices, desktop computers or workstations, Laptop computers, mobile handsets, personal digital assistants ( PDAs), mobile phones, network devices, cameras, smartphones, smartwatches, Enhanced General Packet Radio Service (EGPRS) mobile phones, media players, navigation systems devices, game consoles, set-top boxes, biosensing devices with communication functions a smartphone, a smart TV, or any combination thereof, or at least one processor processor, local memory, display (e.g., monitor or touch screen display) (b), one or more user input devices, and a network communication interface. The computing device may include, but is not limited to, other types of computing devices. The electronic device 19 includes a display monitor, a keyboard, a touchpad, an accelerometer, a jack Iroscope, mouse, touch screen, camera, projector, touch panel, pointing device scrolling devices, buttons, switches, motion sensors, Input / output devices such as audio sensors, pressure sensors, heat sensors, and / or microphones The electronic device 19 may also include any type or combination of (e.g., Wi-Fi, Radio Frequency (RF), Infrared (IR), Blue any suitable wired or wireless communication (via tooth, near field communication, or any other suitable means) They may communicate with each other by wire means.

[0015] The mHealth application 1 may communicate with other entities or entities to send or receive information. In some embodiments, the mHealth application 1 may be used, for example, for exercise tracking (e.g., step tracking) applications and / or or one or more applications associated with the user 8, such as other health-related applications. mHealth application 1 may communicate with other applications. Import data from applications and create treatment plans for users For example, an mHealth application may analyze and use the data from other applications. You may import activity tracking data from your application and use that data to Glucose data collected before exercise and use of mHealth application 1 for user 8 The mHealth application 1 may also identify patterns between other From mobile health applications, for example, blood pressure, BMI, A1C, exercise type, exercise Activity time, exercise distance, calories burned, total number of steps, exercise date, start and end time of exercise, and sleep Any other suitable data may be imported. The application 1 may also be used in conjunction with other mobile devices, e.g., those with social or interactive features. Export data to other mobile applications, including A healthcare provider 7, such as a doctor, may prescribe the application. mHealth applications1 may not require a prescription and may be available, for example, on a computer. Accessible without a prescription from a digital distribution platform for computer software It could also be a commercially available consumer application that can be used to The application 1 may be tailored to a particular user 8 and may be administered by a pharmacy 9 or other authorized entity. It may also be activated directly by the user by visiting the User 8 enters an access code that authorizes access to mHealth application 1. The user 8 may receive the medicine from the mHealth support system 25 and / or The application trainer24 teaches the patient how to use mHealth applications1 mHealth applications1 may be functionally It may include various forms of programming 28, such as machine learning programming algorithms 26. The user's treatment plan may be dispensed by a pharmacy 9 (e.g., drugs, devices, and The pharmacy 9 may also provide a prescription for the user's 8 healthcare treatment plan. Following compliance-based approval, the prescribed product / treatment may be re-prescribed. For example, communication from application 1 via network 32 and various servers 29 The use of drugs or other medical products / treatments may also be received by the pharmacy 9 by mail. , transmitted to the manufacturer 37 via the network 32 and used by the user 8 The quantity of the medical product or treatment may be communicated to the manufacturer 37. This information may be used by the manufacturer 37 to It can help assess demand for a product or treatment and plan supply. and receiving a report based on the user information received by application 1, The user's treatment plan may be updated based on information from the user's electronic medical record (EMR). 14 was also received by mHealth application 1, application Based on user information, which may include electronically transmitted user feedback regarding The health care provider 7 may, for example, provide a doctor, Any appropriate medical professional, including nurses, educators, social workers, MAs, PAs, etc. It may also be a provider.

[0016] FIG. 2 is a schematic diagram of additional embodiments of system 100. For example, system 1 00 transmits the decision model stored in the decision model database 270 via the network 32. The retrieved decision model may be accessed by the mobile device 2. 15, tablet device 220, computer 225 (e.g., laptop or desktop) kiosk 230 (e.g., kiosk with medical and / or prescription information, drug at a medical center, clinic, or hospital), and / or any other facility connected to network 32. for display and / or processing by one or more electronic devices 19, such as It can be used.

[0017] In the embodiment shown in FIG. 2, a mobile device 215, a tablet 220, and a computer Each computer 225 may communicate with any server 29 and / or one or more GPS satellites 255. and obtain and report location information, such as GPS data, via the network 32. It may be equipped with or include, for example, a GPS receiver for detecting the location of a vehicle.

[0018] Mobile device 215, tablet device 220, computer 225, and / or Each of the electronic devices 19, including the kiosk 230, communicates with the server via the network 32. It may be configured to send and receive data (e.g., clinical information) to and from 29 systems. Each of the devices 19 receives information such as clinical data from a server 29 via a network 32. The server 29 may receive the clinical data server 240, the algorithm server 245, may include a user interface (UI) server 250, and / or any other suitable server. The electronic device 19 is in data communication with the UI server 250 via the network 32. Each server may include a decision model database 270. Each server may have access to the memory, processor, and / or For example, the clinical data server 240 may include a provider database. a process configured to retrieve clinical data from the patient's electronic medical record and / or The algorithm server 245 may have a database containing various algorithms. and a processor configured to process the clinical data. 50 is configured to receive and process user 8 input, such as clinical decision-making preferences. The satellite 255 is configured to transmit and receive information between the server 29 and the device 19. It is possible.

[0019] The clinical data server 240 communicates with the UI server 220 via the network 32 or indirectly. 50, may receive clinical data, such as data about the user, from electronic device 19. The clinical data server 240 may store information in a memory, such as a computer-readable memory. do.

[0020] The clinical data server 240 also includes an algorithm server 245 and / or external servers, etc. The server 29 may be in communication with one or more other servers, such as a server 29a, 29b, 29c, 29d, 29e, 29f, 29g, 29h, 29i, 29j, 29m ... The data may include data about the user's health and / or the user's health history. The algorithm server 240 may include data from other users. The algorithm server 245 may also include machine learning and / or other suitable algorithms. It may also communicate with other external servers and may be updated as desired. The algorithm server 245 may provide new algorithms, more powerful programming, and / or The clinical data server 240 and / or algorithms may be updated with more data. The model server 245 processes the information and stores the data in the model database 270 for processing. In one embodiment, the algorithm server(s) 245 may simply Get pattern definitions in a convenient format and use models, e.g., Markov models, Gaussian models , Bayesian, PCA (Principal Component Analysis), multivariate linear or nonlinear regression, and / or linear discriminant analysis Classification models such as functions, nonlinear discriminant functions, composite discriminant functions, and random forest algorithms By using the model, we can predict several time steps into the future and then make decisions based on that prediction. It optimizes results, detects transitions between patterns, abstracts data, extracts information, Inferring higher-level knowledge, combining high-level and low-level information, and providing user Understand clinical behavior and interpret multi-temporal (e.g., different time scales) data and related Infer from information, use variable order Markov models, and / or gradient-based and curve-level Clustering algorithms such as smoothing algorithms and k-means clustering Its use can reduce noise over time.

[0021] a clinical data server 240, an algorithm server 245, and a UI server 250; Each server in the Server 29 system is a web server, an application server, a This includes, but is not limited to, proxy servers, network servers, or server farms. Each server in the system of servers 29 may represent any of a variety of types of servers. The server may communicate with the device 19 or any other computing device, for example, via a network 32. other computing devices, including but not limited to a computing device (not shown) It may be implemented using any general-purpose computer capable of providing data to the device. Such a general-purpose computer consists of a processor and memory for executing and storing instructions. The memory may include, but is not limited to, a server device having a floppy disk drive. magnetic storage, including disks, hard disks, or magnetic tapes; solid state Solid-state drive (SSD) or flash memory, semiconductor storage, optical disk Any software embodied in a physical storage medium, such as a hard disk storage or optical disk storage. Contains any type of random access memory (RAM) or read-only memory (ROM) Software may include one or more applications and operating systems. The hardware may include a processor, memory, and a graphical UI display. Each server may also include, but is not limited to, a clustered display. configured to function together in a shared computing environment or server farm Multiple processors and multiple shared or separate memory components That's fine.

[0022] FIG. 3A is a block diagram of system 100 showing additional details of electronic device 19 and server 29. The electronic device 19 and the server 29 each include a processor 301-1 The processor may include one or more processors, such as processors 301-1 and 304-1. and 304-1 are respectively a central processing unit, a microprocessor, a general-purpose processor, and a special-purpose processor. The electronic device 19 and the sub-device 10 may be a processor or any device that executes instructions. Server 29 also includes memory 301-2 for storing one or more software modules. The memory 301-2 and 304- 2. Any hard drive, CD, DVD, flash memory, RAM, ROM, etc. The memory 301-2 may be implemented using a computer-readable storage medium. 301-1. The memory 304-2 may be executed by the processor 304-1. may be stored.

[0023] The electronic device 19 may further include one or more UIs. may enable one or more interfaces for presenting information to the user 8, such as The UI may be web-based, such as a web page, or may be a standalone application. The UI can also be used for data entry and user feedback. The device may be configured to accept information about the user 8. The information may be entered manually or automatically. (or the user's caregiver) can keep track of when medications are taken or what food and drink the user 8 is taking. The electronic device 19 may also input information such as whether the person has ingested something. It may also include an interface for receiving information from a test device (not shown) or test equipment. The devices may include, for example, blood glucose meters, heart rate monitors, weight scales, blood pressure cuffs, etc. The device 19 also includes a camera, a microphone, and other devices for collecting feedback from the user 8. The system may also include one or more sensors (not shown), such as a microphone, or an accelerometer. In an embodiment, the device includes a blood glucose monitor for reading and automatically reporting the user's blood glucose levels. It may also include a value measurer.

[0024] The electronic device 19 may also include a presentation layer. The layers are web browsers, applications, messaging interfaces (e.g., The electronic device 19 may be a Through the presentation layer, notifications, alerts, reading material, references, guides, and reminders The presentation layer may present the user 8 with a message or suggestion. Related articles, reminders to buy medications, tutorials on the topic Realistic (e.g., tutorials on carbohydrates), testimonials from others with similar symptoms, And / or, one or more goals (e.g., carb counting goals) may be presented. The presentation layer can also contain tutorials (e.g., user guides or instructional videos). and / or to communicate information between the healthcare provider and the user 8, e.g., the patient. Communication between a healthcare provider and a user 8, e.g., a patient, may be electronic. via child messaging (e.g., email or SMS), voice, or real-time video One or more of these items may be included in a treatment plan or update, as described below. The presentation layer also allows the user to view the results of the treatment plan. It may also be used to receive feedback from

[0025] System 100 also includes one or more databases, such as database 302. The database 302 may be a relational database or an object-oriented database. may be implemented using any database technology known to those skilled in the art, such as The database 302 may store data 302-1. The data 302-1 may be used to perform inferences. The data may include a knowledge base, statistical models, and / or user information for performing the analysis. 2-1, or a part thereof, may alternatively or simultaneously be transmitted to a server 29 or an electronic device 19. It may be stored.

[0026] The system 100 may be used to, for example, address a user's health care, maintain a user's finances, and monitoring and tracking a user's nutrition and / or sleep. In some implementations of the system 100, any received data may be used for , enhance data security against unauthorized access, HIPAA privacy, and / or or in encrypted form to comply with other legal, healthcare, financial, or other regulations. It may be stored in a database.

[0027] With respect to any server or server system 29 depicted in system 100, the server or The server system may include one or more databases. A database is any type of data that can be used to store any type of data. For example, the database 302 may store and store information about each prescription in a treatment plan. Contains information related to the user's treatment plan, including timing and dosage associated with medications. Stores data received by or processed by the server 29, including The database 302 may also include a literacy level of the user 8 associated with each of a plurality of prescription drugs. It may also store information related to the user 8, including the bell.

[0028] As further disclosed herein, one or more components of the disclosed subject matter may be implemented using a machine learning model. 3 illustrates an exemplary training module 310 for training multiple machine learning models. The modules are used to train each of the machine learning models disclosed herein. and / or a single training module 310 may train two or more machine learning models. It will be appreciated that the present invention may also be used to

[0029] As shown in FIG. 3B, training data 312 is associated with the machine learning model being trained. This may include one or more of the stage inputs 314 and known results 318. The input 314 may be received from a healthcare provider 7, one or more servers 29, an electronic device 19, an EMR, or the like. 14. Output from a step (e.g., flowchart 500 of FIG. 5A or flowchart 500 of FIG. 9) One or more outputs from steps from chart 900, time-in-range (TIR) ​​values, hypertension Glucose range (TAR) value, hypoglycemia range (TBR) value, severity score, continuous glucose monitoring (CG The known results 318 may be from any applicable source, including classification, etc. , for machine learning models generated based on supervised or semi-supervised training. Unsupervised machine learning models may not be trained using known outcomes318. The known outputs 318 are similar or different to the stage inputs 314 that do not have corresponding known outputs. or may include known or desired outputs for future inputs in the same category.

[0030] Training data 312 and training algorithm 320 are used to process training data 312 into a training algorithm. The training component 330 may apply the training data to the training system 320 to generate a machine learning model. According to one implementation, the training component 330 may A comparison result 316 is provided that compares the previous output, and the previous result is applied to the machine learning model. The comparison results 316 may be used to update the corresponding machine learning model. The training algorithm 320 may be used by the training component 330. Convolutional Neural Networks (DNN), Convolutional Neural Networks (CNN), Fully Convolutional Networks (FCN) and Recurrent Neural Networks (RCN) ) and other deep learning networks, Bayesian networks, and graphical models. probabilistic models such as eigenvalues, and / or discriminative models such as decision trees and maximum margin methods. Machine learning networks and / or models may be utilized, including but not limited to: stomach.

[0031] (health condition) Diabetes mellitus (commonly called diabetes) occurs when a patient's body does not produce insulin at all or not enough. Inability to produce insulin or inability to use the insulin produced (insulin Phosphorus resistance), a chronic, persistent metabolic disease (or condition) that reduces the amount of glucose in a patient's blood. The three identifiable types of diabetes are prediabetes, type 1 diabetes, and and type 2 diabetes. Prediabetes is when blood sugar levels are high but not high enough to be considered type 2 diabetes. Type 2 diabetes is a chronic condition that affects the way the body processes blood sugar. Finally, type 1 diabetes occurs when the pancreas produces little or no insulin. It is a chronic condition.

[0032] Generally, diabetes is diagnosed in several ways. Diagnosing diabetes involves determining the severity of the condition. Tests may need to be repeated on multiple days to confirm a positive diagnosis of type 2 diabetes. Some health parameters used by a doctor or other appropriate health care provider when confirming a diagnosis of disease The meter measures blood glycosylated hemoglobin (A1C), fasting blood glucose (FPG), and oral glucose tolerance. Generally, a healthcare provider will provide a diagnosis of diabetes. The patient's A1C value is of interest to help determine the cause of the disease. Glycated hemoglobin is the average blood glucose level over a three-month period. A form of hemoglobin that is primarily measured to determine plasma glucose levels. and / or other appropriate health care providers. Health parameters may include weight, age, Nutritional intake, physical activity, cholesterol levels, triglyceride levels, obesity, tobacco use, and family history include.

[0033] Once the diagnosis of diabetes type has been confirmed by a physician or other appropriate healthcare provider, the patient If you are diagnosed with diabetes, you may be treated to manage your diabetes. The patient being tracked or monitored may be taking measures to improve diet, exercise, oral medications, and / or insulin therapy. Complications can be treated by combining blood sugar control through Regular testing of blood glucose levels is also required for some patients. Depending on the duration, mHealth applications1 may be used to manage the patient's condition. Oral medications typically reduce the production of glucose by the liver. , including tablets taken by mouth to make muscles more sensitive to insulin. For example, if the diabetes is more severe, additional medications, including injections, may be used to treat the patient's diabetes. Additional medications may be needed. Basal insulin injections are used to provide background insulin. Also known as sulin, it is administered by a healthcare provider to maintain consistent fasting blood sugar levels. During fasting, the patient's body produces glutamate in the blood to provide energy to the cells. This helps control blood sugar levels and ensures cells get the energy they need. Injections of basal insulin are required to allow the blood to absorb glucose. Basal insulin is usually taken once or twice daily, depending on the type of insulin. Basal insulin acts for a relatively long period of time, so it is often used in combination with long-acting or intermediate insulin. In contrast, bolus insulin is a fast-acting For example, bolus insulin can be used to control blood glucose levels after a meal. In some instances, a doctor or healthcare provider may administer the drug at mealtimes to reduce side effects. When creating a treatment plan to manage a patient's diabetes, a doctor may consider, for example, Basal-bolus medication, which involves giving several injections Create a regimen. A basal-bolus regimen may include injections with each meal. , which attempts to closely mimic how the body releases insulin in non-diabetic people. The basal-bolus regimen is applicable to people with type 1 and type 2 diabetes. In addition to a basal-bolus regimen requiring insulin injections, The treatment plan may be supplemented with the use of prescribed oral medications. Patient adherence to the treatment plan depends on the patient's For example, if a patient has been diagnosed with diabetes for more than six months, For example, if you have a very specific treatment regimen for achieving healthy or desirable blood glucose levels, Ultimately, the use of these drug types of treatment should be adhered to by the patient. Weekly patterns can be important in managing diabetes. Section 1 may recommend a treatment plan to help the patient manage their diabetes.

[0034] (Exemplary Method) Diabetes is a condition in which a patient is unable to keep their glucose levels within normal or recommended target ranges. It is a chronic condition that causes such fluctuating blood glucose levels (i.e., blood sugar levels that are higher than normal or recommended levels) to worsen. Sporadic blood glucose monitoring (B GM) is difficult to develop meaningful insights from only a few intermittent measurements per week. identifies patterns and any underlying causes of these patterns (e.g., B based on dietary type) This may not be the basis for understanding the GM increase.

[0035] Continuous glucose monitoring (CGM) is the monitoring of regular glucose readings (e.g., a user's blood glucose level). automatically collected through wearable sensors (e.g., subcutaneous sensors) that provide It offers the possibility of high density data (e.g., data collected every 5 minutes or less). The CGM allows a user 8 or other entity (e.g., a healthcare provider 7) to monitor the health of the individual at any time of day. To be more aware of the user's blood glucose level at any given time, the glucose data may be continuously updated (e.g., 5 minute or less) or semi-continuous (e.g., more than every 5 minutes) readouts to the user or other Such data may be provided to entities to improve diabetes care. This may allow the healthcare provider 7 to more optimally tailor a treatment plan for the user 8 .

[0036] A CGM monitor is a continuous current monitor that includes any sensor configuration that provides an output signal indicative of the concentration of an analyte. The CGM monitor may be an analyte sensor system based on bodily fluids (e.g., interstitial fluid). For example, the concentration of an analyte may be sensed to determine glucose levels. For example, raw data streams, filtering filtered, smoothed, and / or otherwise transformed sensor data The output signal, which may be in the form of sensor data, such as The signal may be transmitted to a receiver that may be connected to the sensor or may be remote from the sensor. Depending on the implementation, the CGM monitor may include a transcutaneous glucose sensor, a subcutaneous glucose sensor, Continuous refillable subcutaneous glucose sensors, continuous intravascular glucose sensors, etc. The CGM monitor may be inserted onto the abdomen of the user 8 and penetrate the skin of the user 8. A compact medical system having one or more sensors, including a small cannula through which the sensor passes. An adhesive patch may hold the monitor in place. The sensor may Glucose measurements may be sensed in the interstitial fluid on a continuous or semi-continuous basis.

[0037] The transmitter allows the CGM monitor to wirelessly transmit glucose readings to a monitoring device. The monitoring device may be connected to a sensor to enable the CGM It may be a dedicated monitoring device or a third-party device, It may be an electronic device 19 or any other applicable device. The monitoring device provides one or more functions in addition to CGM monitoring. It may be a dedicated monitoring device or electronic device 19. Other software may collect glucose readings and related data via the monitoring device. The monitoring device may be used to facilitate analysis and / or display of the data. Used to analyze and / or view data associated with glucose measurements Alternatively or additionally, the CGM monitor may The CGM monitor and / or external device may include a display for viewing the data. based on glucose data (e.g., if blood sugar levels are too high or too low, or configured to generate and / or provide a warning if the It may be possible.

[0038] By using CGM data, the time in range (TIR) ​​value is calculated based on the time the user The blood glucose level of the user may be determined based on the amount of time the user's blood glucose level is within the threshold band over a reference period. The threshold band may be pre-determined, user specific, or dynamically determined. This may also be done.

[0039] The threshold band may be a predetermined value based on, for example, a cohort of patients. The lifestyles, habits, and medical test results of each patient in the group are used to determine pre-determined values. For example, one or more patient cohorts may be used to determine patient liabilities. The threshold band may be determined based on lifestyle, habits, demographics, etc., and may be one or more The threshold bands may be generated for each of the cohorts. For example, the optimal result (e.g., a preferred A1C value) based on The machine learning model may be generated using a training module 310. The model may be trained using the blood glucose levels of a cohort of patients as stage input 314, The corresponding A1C value may be received as a known result 318. The training machine learning model may As input, data (e.g., A1C values) for a cohort of patients may be received, and the patients The blood glucose threshold bands (i.e., upper and lower glucose limits) of the cohort Alternatively, threshold bands may be used for the general population so that they are not cohort specific. This may be a predetermined value. According to an implementation, the TIR threshold band is approximately 70 mg / dL and approximately 180 mg / dL. The TIR value is the time when user 8's blood glucose level is between baseline and The amount of time that the TIR threshold band between the two points may be within the TIR threshold band. The reference period may be 24 hours, but more granular changes in TIR values ​​may occur over a 24 hour period. The determination may be based on reducing the reference period to less than 2 hours, and a wider change may be based on reducing the reference period to less than 2 hours. It will be appreciated that the determination may be based on increasing reference periods longer than four hours. .

[0040] The user-specific threshold band may be determined based on attributes for the user 8. The attributes may include: The threshold may be a user history, a medical history, demographics, etc. According to one implementation, the user-specific threshold may be: may be generated using a machine learning model trained using the training module 310 The machine learning model may receive updated attributes based on the user's 8 comparison results. 316 component to retrain itself using the updated attributes. In one embodiment, the machine learning model may be updated based on changes in the weight of the user 8. The comparison result 316 component updates the previously provided threshold bands based on the weight received. Therefore, a user-specific threshold band may be a change in the attributes provided to the user. The user may change the user's password from time to time based on one or more attributes of the user. The threshold bands may be based on the user, cohort of users, external conditions, environmental conditions, updated recommendations, etc. The determination may be based on a change in one or more associated attributes.

[0041] As applied herein, a user vector (e.g., a patient vector) is a vector of a given patient. Any line associated with or that may be associated with or that may be changed by a person It may be a movement, activity, item (e.g., consumable), service, parameter, or value. The user vector may be a TIR or GV state of the user 8, as further disclosed herein. By way of example, the patient vector may be modified to improve the patient's health, medication, food intake characteristics, etc. , exercise values, psychosocial parameters, social determinant parameters, etc. may include:

[0042] As applied herein, a user attribute (e.g., a patient attribute) is a user attribute associated with a patient. Patient attributes can be easily compared to the patient vector. For example, the patient may be Attributes include social attributes, medical history or condition, patient preferences, metabolic attributes, patient demographics, etc. That's fine too.

[0043] FIG. 4A shows an example CGM-based blood glucose trace 402 for user 8. The time period shown through the display may be a full day (i.e., 24 hours). The blood glucose level of User 8 was measured for a portion of the day except during the TAR period 402A and the TBR period 402B. Regarding the threshold range represented by the upper threshold 404A and the lower threshold 404B, The user 8 may have a TIR during the day or at the end of the day. The CGM data may then be provided to the user 8. The user 8 may be provided with a current blood glucose level of the user 8 and / or a blood glucose level associated with the user 8. The user may also be notified of trends.

[0044] FIG. 4B illustrates an exemplary CGM-based report that may be provided to a user 8 or healthcare provider 7. 406. The report shows the Ambulatory Glucose Profile ( AGP) format, and can include not only graph data but also a number of metrics (e.g., The report may include glucose statistics and goals 408, A GP profile 410, daily glucose profile 412, time range 414, etc. However, most people with diabetes have a tendency to have blood sugar changes. It may be impossible to interpret such CGM data and / or AGP information. Similarly, healthcare providers7 need to discuss multiple patient needs and consider CGM monitoring and / or Or interpret the data provided via AGP information to optimize blood glucose levels, even temporarily. The technology disclosed herein provides tracking of essential parameters and manages the health of users 8. do.

[0045] According to implementations disclosed herein, the CGM data may be stored in a manner similar to that described further herein. may be used to recommend changes based on one or more patient vectors, such as CGM events (e.g., changes in CGM state, portions of CGM traces, etc.) are used by diabetes self-diagnosis. Defined as identifiable areas of the CGM trace that correlate with self-management activity (DSMA). As further applied herein, a CGM trace may identify a CGM trend. It may be used to monitor the drug status or may be a CGM trend. These may be changes or additions, changes or additions to foods, changes or additions to exercise, etc. , may drive automated coaching to the user 8. Similarly, CGM-based results (e.g., The results of automated coaching and / or DSMA-based glucose signatures will be used in future DSMA studies. Provide coaching and / or tailored specific decision support for healthcare providers It may be provided to Donor 7.

[0046] According to an implementation, the Detection, Notification, Classification, and Engagement (DICE) framework Detect various diabetes-related events from the M trace and optimize the process through one or more visualizations. Inform the healthcare provider 7 and / or user 8 about progress along the pathway and provide guidance for further interventions Classify detected events into one or more classifications and / or two-dimensional CGM quadrant starting states. and / or an overview of techniques for engaging and guiding patients toward improved outcomes The technology associated with the DICE framework is based on metabolic data, lifestyle data, and Integrate data from multiple disciplines, including clinical, socioeconomic, and other data, to improve patient care. The automated CGM event detection and classification techniques disclosed herein increase accuracy and By reducing errors, it is possible to improve the quality of care. Automated coaching based on the patient's needs can be tailored to the scale of any patient needing care and / or support. The visualizations provided herein are based on high-density CG. M-data and other applicable data in easy-to-use charts, graphs, and / or other visualizations By focusing on a specific area, the data burden on the user 8 and / or healthcare provider 7 is reduced. FIG. 5A illustrates a method for providing an optimized pathway for improving the glucose status of a user 8. 500. At 502, a blood glucose level of a user 8 may be received. As such, blood glucose levels may be provided continuously or semi-continuously by a CGM monitor. The blood glucose level may be received by a component of the CGM monitor itself, or by an electronic device. 19, mHealth application 1, one or more servers 29, etc. The blood glucose level may be received by one or more remote components from a CGM monitor. May be provided automatically to multiple components or pushed at the time of blood glucose collection Alternatively, the CGM monitor may ping to transmit one or more collected blood glucose values. This may be done.

[0047] In one example, the user 8 may wear a CGM monitor on their body. A CGM monitor may collect blood glucose readings every 5 minutes. Network connection, Local Area Network connection, Wide Area Network connection, WiFi Connected to the user's mobile device (via a USB cable, Bluetooth connection, etc.) According to a first exemplary implementation, the CGM monitor may: (e.g., every 5 minutes), automatically transmits blood glucose readings to the user's mobile device Alternatively or additionally, one or more blood glucose measurements may be and / or The mobile device or another component of the device transmits one or more blood glucose readings. When a CGM monitor requests that a blood glucose reading be sent, it stores one or more blood glucose readings. That's fine.

[0048] At 504 of FIG. 5A, a time-in-range (TIR) ​​value associated with a blood glucose measurement is In-range glucose values ​​are determined by the amount of time that a blood glucose measurement is within a predetermined range. Ratio of in-range to out-of-range blood glucose readings, count of in-range and out-of-range blood glucose readings The TIR value may correspond to when the user's 8 blood glucose level is in range and when the user 8 blood glucose level is out of range. As shown in FIG. 4A, the blood glucose level is When it is between the threshold 404A and the lower threshold 404B, it may be considered to be within the range. The patient's TIR value is the time the patient's blood glucose level is between 70 mg / dL and 180 mg / dL. The upper threshold 404A may be 180 mg / dL, and the lower threshold 404B may be 180 mg / dL, so that the amount of The threshold 404B may be 70 mg / dL.

[0049] The TIR value determined at 504 in FIG. 5A is the time it takes for the user's 8 blood glucose levels to change over a reference period. The reference period may be a single 24-hour day. The reference period may be the same as the reference period (e.g., for user 8). Thus, it may be predetermined (e.g., pre-programmed) by the healthcare provider 7. , or may be dynamically determined based on one or more factors. may be a patient vector, a patient attribute, a current or previous TIR state, etc.

[0050] According to one implementation, the TIR value may be for a reference period, or may be for multiple reference periods. It may be a TIR value associated with a patient over a period of time. For example, the TIR of a user The R value may be determined for each day for a total of 10 days. TIR values ​​from each day for 10 days indicates that the TIR associated with user 8 over a 10-day period is the combined TIR value. may be combined using any applicable technique (e.g., averaging) .

[0051] According to one implementation, the TIR value is used to exclude or flag outliers in blood glucose levels. are filtered to be weighted lower than non-lagged blood glucose measurements. In one example, a blood glucose reading of 65 mg / dL during a first measurement may be A second measurement taken 5 minutes after the first may increase to 200 mg / dL. A third reading taken five minutes after the second may show a blood glucose level of 68 mg / dL. Density-based techniques (e.g., k-nearest neighbors, local outlier factor, isolation forest, etc.) Use filters, subspace, correlation-based, and / or tensor-based for high-dimensional data Filters that use outlier detection, filters that use one-class support vector machines, Replicator neural network, autoencoder, variational autoencoder, long Filters using short-term memory neural networks, filters using Bayesian networks Filters, filters using hidden Markov models (HMMs), cluster analysis-based decoupling Filters using value detection, association rules and deviations from frequent itemsets ,filters using fuzzy logic-based outlier detection,filters using ensemble techniques. Filters, feature bagging, score normalization and filters using different sources of diversity, Filters such as filters using convolutional LSTMs with mixtures of recursive principal component analyzers The data may contain outliers and / or blood pressure data that may be misread and may be insignificant outliers. One such filtering technique may be used to identify glucose readings. Or multiple may also be used in conjunction with the machine learning models disclosed herein. Depending on the implementation, the TIR value associated with the user 8 may be determined by one or more such filters. As further disclosed, blood glucose measurements filtered through a filter may be taken into account. Additionally, such filtering may be used to filter out outliers, outlier data, and / or irregular measurements. This may prevent the provision of an optimized route, which may be impaired as a result.

[0052] At 506 in FIG. 5A, the TIR status of user 8 is one or more can be determined based on multiple TIR values. The TIR status is associated only with the TIR value. The condition may be due to one or more other factors (e.g., the frequency of glucose measurements). based on factors such as the degree of glucose in the blood, the quality of the glucose reading, other perceived readings, and patient-based factors For simplicity, in this disclosure, we will assume that the binary state of the TIR is based solely on the TIR value. However, the T The IR state may be a multidimensional state based on the TIR value and one or more other factors. As applied herein, good TIR conditions (e.g., first TIR condition) corresponds to a TIR ratio greater than the TIR cutoff, and poor TIR condition ( For example, the second TIR state) corresponds to a TIR ratio less than the TIR cutoff.

[0053] Figure 6A shows a chart 600 of TIR status for a number of different patients. As disclosed, the chart includes four quadrants based on the TIR ratio and the GV ratio. The condition is based on the TIR axis, which corresponds to the Y axis in chart 600. The TIR ratio is: The TIR ratio is the percentage of time over a baseline period that a patient's blood glucose levels are within the threshold band. Alternatively, the TIR ratio is the TIR ratio calculated (e.g., averaged) over multiple reference periods. the percentage of time over the reference periods where the patient's blood glucose level is within threshold bands for multiple reference periods, such that may be.

[0054] The TIR ratio value is specified as the cutoff for good versus poor TIR status. Chart 600 of FIG. 6A shows that a TIR ratio greater than 0.5 (e.g., for user 8) A threshold band for blood glucose levels to exceed 50% of the time or for calculated measurements A TIR ratio below 0.5 is considered to be in a good TIR state (e.g., within , the user's blood glucose level exceeds 50% of the time, or the calculated measurement A threshold of 0.5 is used to consider poor TIR conditions (if outside the threshold band for TIR). The cutoff may be predetermined or may be dynamically determined. The predetermined cutoff may be based on medical criteria or may be determined by cohort or user criteria. The dynamically determined cutoff may be specified by the healthcare provider 7 to the user 8. , may be based on a cohort or a given user, or may be determined by a machine learning model. The machine learning model takes as input the patient vector, patient attributes, and past patient The user 8 or You may output a cutoff specifically for a cohort. Therefore, the cutoff is to the value considered optimal for the corresponding user or cohort on which the data was based. It may be adjusted.

[0055] As shown in Chart 600, having a TIR value above the cutoff of 0.5 Patients were considered to have a good TIR status and had a TIR below the cutoff of 0.5. Patients with values ​​below 0.05 are considered to have poor TIR status. It is understood that the number of patients with good or bad TIR status will vary accordingly. For example, if the TIR ratio were adjusted to 0.9 instead of 0.5, most The patient will have a poor TIR status.

[0056] At 508 in FIG. 5A, blood glucose variability associated with blood glucose measurements for a given user 8 is Blood glucose variability (GV) values ​​are determined. Glucose variability measures the amount of change in glucose over a period of time. The GV value is calculated using the standard deviation (SD) of the mean mean GV. ) value, coefficient of variation (CV), or any other applicable measure of variation.

[0057] SD may be collected over a period of time (e.g., 1 hour, 1 day, or any other applicable period). (i) It can be a measure of the amount of fluctuation or variability in a set of glucose values. It may indicate that the course values ​​tend to be close to the mean value of the set of glucose values. The SD of glucose values ​​may indicate that the values ​​are spread over a wider range. The SD of the glucose values ​​can be calculated as shown in Equation 1. It is possible.

number

[0058] CV can be a standardized measure of the variability of a probability or frequency distribution. The CV of the values ​​was calculated by determining the ratio of the standard deviation of the blood glucose values ​​to the mean value of the blood glucose values. CV can indicate the degree of variation associated with the average blood glucose value over a period of time. CV can be calculated as shown in Equation 2.

number

[0059] As mentioned above, the GV value may be an SD value or a CV value. The type of V value (e.g., SD value, CV value, etc.) may be based on the user 8 or based on current or past patient vectors, patient attributes, or other information related to the user8. According to another implementation, the type of GV value may be determined by the healthcare provider 7 based on: or based on one or more inputs, such as patient vectors, patient attributes, or past analysis, determined by a machine learning model that is configured to output the optimal type of V value. Good too.

[0060] At 510 of FIG. 5A, the GV status of user 8 is one or more The GV status can be determined based on multiple GV values. or one or more other factors (e.g., frequency of glucose measurements, group may be based on the quality of the course measurements, other perceived measurements, patient-based factors, etc. For simplicity, this disclosure will consider the binary state of GV based solely on the GV value (i.e., good However, the GV state is different from the GV value. It will be appreciated that the present invention may be a multi-dimensional state based on one or more other factors. As applied to the specification, a good GV condition (e.g., a first GV condition) is a GV cut. A bad GV state (e.g., the second GV state) corresponds to a GV value greater than OFF. Corresponds to GV values ​​smaller than the cutoff.

[0061] FIG. 6A shows a chart 600 of GV status for several different patients. As shown, the chart includes four quadrants based on the TIR and GV ratios. The GV axis corresponds to the X axis in Chart 600. GV is the Alternatively, GV may be the SD or CV associated with the patient's blood glucose level. Multiple values ​​are used so that V is a value calculated (e.g., averaged) over multiple time periods. The SD or CV associated with the patient's blood glucose levels over a period of time.

[0062] The GV value may be specified as a cutoff for good versus poor GV conditions. For the 6A chart 600, a GV value above 0.8 is considered to be in good GV condition, and 0.8 is A cutoff of 0.8 is included so that GV values ​​below are considered to be in poor GV status. The cutoff may be predetermined or may be dynamically determined. The cutoff may be based on medical standards or based on the medical care provided by the cohort or user. Dynamically determined cutoffs may be assigned by the cohort or a given user. The data may be based on a user or determined by a machine learning model. The inputs are the patient vector, patient attributes, and the TIR or GV values ​​or changes of the past patients. etc., and may be cut specifically regarding the user or cohort with which the input is associated. Therefore, the cutoff is determined by the corresponding It may be adjusted to a value that is considered optimal for the user or cohort.

[0063] As shown in Chart 600, patients with GV values ​​above the cutoff of 0.8 Those who were considered to have a good GV status were those with a GV value below the cutoff of 0.8. Patients with a GV score of 0.01 or lower are considered to have poor GV status. If the cutoff is shifted, It will be appreciated that the number of patients with good or bad GV status will vary depending on the For example, if the GV value is adjusted to 0.9 instead of 0.8, the cutoff is 0.8. Compared with when using a 24-well plateau, more patients will have poor GV outcomes. According to the study, the optimal cutoff value for distinguishing between good and bad conditions is 0.7. It is also possible.

[0064] As shown in FIG. 6A, the TIR cutoff value (i.e., in the example shown in FIG. 6A) is 0.5) and the GV cutoff value (i.e., 0.8 in the implementation shown in Figure 6A). Four quadrants are created based on the Y-axis (TIR ratio) and X-axis (GV value) divided according to the Patients in the upper left quadrant 602 are in good TIR status (i.e., TIR cutoff for patients in poor GV status (i.e., below the cutoff GV) and poor GV status (i.e., below the cutoff GV). This state corresponds to a first characterization (i.e., good) corresponding to a TIR state and a second The characterization (i.e., poor) is considered a good-bad (GB) condition corresponding to the GV condition. Patients in the lower left quadrant 604 have poor TIR status (i.e., below the TIR cutoff). corresponding to patients within a poor GV condition (i.e., lower than the cutoff GV) This condition may be considered a poor-poor (BB) condition. Patients in the lower right quadrant 606: Poor TIR status (i.e., below the TIR cutoff) and good GV status (i.e., This condition corresponds to patients within the poor-good range (B Patients in the lower left quadrant 604 may be considered to be in a poor TIR state (i.e., below the TIR cutoff) and poor GV status (i.e., lower than the cutoff GV This condition can be considered a poor-poor (BB) condition. Each of patients 02, 604, and 606 was in at least one of TIR and GV states. Patients with non-ideal conditions, such as one who is in a non-optimal condition (e.g., a "poor" condition) Patients in the upper right quadrant 606 may be considered to have a good TIR condition (i.e., TIR cardiovascular disease). Patients in good GV status (i.e., above the cutoff GV) and in good GV status (i.e., higher than the cutoff GV) This condition can be considered a good-good (GG) condition. The researcher aims to ensure that both the TIR and GV conditions are optimal (e.g., "good" conditions). This patient can be considered to be in an ideal condition.

[0065] As shown at 512 in FIG. 5A, the starting state for a given patient is the patient's TIR state. As shown in FIG. 6A, the starting state of user 8 is The TIR and GV states may correspond to quadrants into which they are classified. For example, as shown in FIG. 6A, As such, user 610 has a TIR ratio that does not meet the TIR cutoff and is therefore Good TIR condition and GV higher than the GV cutoff, therefore good GV Therefore, as determined in 514 of FIG. 5A, the In the illustrated embodiment, the non-ideal starting bad-good condition is represented by the lower right quadrant 606. As further disclosed herein, there may be a non-ideal starting failure of the user 610. A good overall state may be the state of the user 610 at a point in time, and may be a state of It may change.

[0066] As determined at 514 of FIG. 5A, the non-ideal starting conditions may affect the diabetic management of user 8. For example, non-ideal starting conditions may indicate a low TIR and / or a non-optimal Therefore, a non-ideal starting condition may indicate that the user 8 condition is unreasonable. The corresponding user's 8 diabetes management needs to be adjusted so that the desired state can be changed. It may also be necessary.

[0067] According to one implementation, as described herein and shown in FIG. 6A, a two-dimensional frame Work is produced through a new data integration extract, transform, and load (ETL) process. The process may be implemented in a system. either by the computer, the electronic device 19, and / or any other applicable component. The extracted data may be analyzed using one or more of the following methods: A production data set that includes a machine learning model and can determine a starting state (e.g., at 512 in FIG. 5A). may be converted and / or exported to a database.

[0068] Thus, a macro view of user 8's state-based data (e.g., starting state) would be: It can be represented by two orthogonal parameters: the TIR state and the GV state. As disclosed in, the corresponding state (e.g., as shown in FIG. 6A) is The radiological health status can be assessed and visualized and reported to users, healthcare providers, etc. The state-based data may be used in conjunction with other methods (e.g., as further disclosed herein) to optimize the may be used to provide overall glucose health recommendations (via the metabolic pathway) .

[0069] At 516 of FIG. 5A, an optimized path can be generated to reach the ideal state. The optimized pathway may be one or more adjustments of one or more patient vectors, and may be non- Ideal condition (i.e., good-poor, poor-poor, or poor-good condition), patient vector , and / or patient attributes. The optimized pathway may be determined based on medication, food intake characteristics, Physical activity, psychosocial parameters, and / or social determinants parameters, but The adjustment may be, but is not limited to, one or more patient vectors.

[0070] Medication adjustments may be provided based on the user's 8 current medications or may be provided by the user. Adjustments may be made based on new medications that may be available. Adjustments may be made by adjusting the dosage of the medication. This allows you to change the time or frequency at which a medication is taken by adding or removing medication. By doing so, the environment associated with the drug (e.g., the type of food consumed with the drug) can be For example, the specific intake that the user 8 is currently taking may be changed. The medication intake may be adjusted to a higher dose.

[0071] Adjustments for food intake characteristics may be made for one or more foods, food groups, food types, food intake times, and dietary intakes. Change, remove, add, or otherwise alter product pairings, food and drug pairings, etc. For example, after ingesting food, the patient's blood glucose level may be adjusted to a threshold value. Based on patient attributes indicating an increase above the threshold, the patient may be monitored for a period of time during which their current blood glucose level is low. The patient may be provided with a warning to avoid consuming the food.

[0072] Exercise value adjustments can be made by changing one or more exercises, exercise type, exercise duration, exercise time, etc. This may include changing, removing, adding, or otherwise modifying the patient's A given patient's GV may be more stable if they exercise at the beginning of the day; Therefore, adjustments may be made to prioritize exercising in the morning.

[0073] Psychosocial and / or social determinant parameters may also be adjusted, or May be modified to reflect changes in meditation schedule or type, social activities, interactions, and / or the same To change, remove, add, or otherwise modify the duration or frequency of It may include.

[0074] At 516, an optimized route can be generated using a machine learning model. The machine learning model may be trained to take as input: Patient vector, starting state (e.g., TIR state, GV state, good-bad state, bad-bad state) and one or more of the following: a patient's condition, a poor-good condition, etc.; a patient attribute; and a CGM characteristic. The machine learning model may generate an output of an optimized path based on such input. As disclosed herein, an optimized pathway may involve adjusting one or more patient vectors. may be.

[0075] At 518 in FIG. 5A, the optimized pathway is delivered directly to the patient (e.g., via an mHealth app). (e.g., to a user 8 via an application 1 using an electronic device 19), The optimized pathway may be provided to one or more patients, a provider, or both. The patient vector may be a summary of changes to the patient vector, or an automatic adjustment of one or more patient vectors. Alternatively, the control may be gradually increased or decreased based on one or more actions, timing, levels, values, etc. The optimized pathways provided over time may be provided to a corresponding one or more patient vectors. The patient vectors may be modified based on the patient's current state. As an example, when the patient's blood glucose level is at the lower end of the threshold band, the change to the patient vector is When a patient's CGM monitor records such blood glucose levels, A mobile device alert may be provided. The mobile device alert may be provided by a user who has Provides instructions to users that they should consume the food within a given time period based on the warning. Good too.

[0076] Optimized routes can also be scheduled periodically (e.g., daily, hourly, weekly, etc.) or based on a trigger. The predetermined time may be based on a change based on the optimized route. For example, an optimized pathway that modifies a patient's meal schedule may use alerts during mealtimes. As another example, an optimized route of administration for a patient's medication may be provided using Alerts may be provided during medication times.

[0077] The frequency, manner, and / or mode of providing the optimized path may be determined by the successful implementation of the optimized path. The habit index may be based on one or more key behaviors or variables associated with The Habit Index may be determined for a patient or cohort of patients with similar attributes of: It may be a classification of the patient's behavior, specifying habits (e.g., frequent communication, frequent Non-busy communication, technical communication, telephone communication, Human communication, graphic communication, time communication It may be a value or score, or it may be a suitable adjustment to the optimization path. The patient's actions may be directed to any other applicable designation.

[0078] The habit index is based on three factors: frequency-based factors, time-cue-based factors, and context-cue-based factors. The habit index may be determined based on habits or preferences, including factors such as: can be provided by the habit index, and used to provide the patient with an optimized path In one embodiment, the habit index may be calculated based on the number of times the user 8 has minimal communication. Preferences, and preferences for any communication that takes place through mHealth applications. Therefore, the patient vector changes via the optimized pathway may be It may be provided to the user once via an mHealth application. The Habit Index then provides patients with an optimized pathway in a personalized way based on their individual behavioral preferences. It may be used to provide.

[0079] Figure 5B shows a flowchart 540 of an example implementation based on CGM. Step 512 in corresponds to step 512 in FIG. 5A and, as disclosed herein, determining a starting state for a given patient based on the patient's TIR and GV states; At 520, a determination is made as to whether the starting condition determined at 512 is an ideal condition. If the starting condition is an ideal condition, then at 522 the CGM monitor may persist in running the CGM if the starting conditions are not ideal (i.e., non-ideal conditions, such as good-bad, bad-bad, or bad-good conditions), at 524 One or more attributes of the non-ideal conditions may be determined. The attributes of the non-ideal conditions may include TIR or The value may be a GV value, a TIR or a change in GV, etc. The patient vector can be determined by the user 8 and the healthcare provider. 7, via electronic device 19, via server 29, or may be obtained via any other applicable method.

[0080] At 528, an optimized path is determined to transition the patient from a non-ideal state to an ideal state. Reaching an intermediate non-ideal state is part of reaching an ideal state. It will be appreciated that, for example, a bad-bad (i.e., a bad TIR condition and Patients with a non-ideal starting condition (poor GV condition) may experience a worsening of the condition before reaching a good-good condition. First, the patient may be provided with an optimized pathway to transition from good to poor or poor to good. The machine learning model can be generated using a TIR state or value, a GV state or value, and one or more patient vectors. , based on inputs including one or more of one or more patient characteristics, CGM events, etc. The optimized path may be output, including the change in one or more patient vectors. The optimized pathway can then be provided to the patient. The optimized pathway can be calculated based on the probability that the patient will follow the optimized pathway. To enhance the patient's performance, the patient's performance may be provided based on a habit index associated with the patient. and / or providing an optimized route at 530. After that, at 522, the CGM monitor continues the CGM, and flowchart 540 continues. By starting at 512 based on sustaining CGM at 522 Flowchart 540 may be repeated iteratively for a given patient or patients. Occurs for any applicable time period, either predetermined or dynamically determined, for a cohort You may do so.

[0081] FIG. 6B shows charts 612 and 614. The first chart 612 shows several Shows multiple statuses of a given patient over the course of a month. For example, the initial status of a patient shown by 613A The condition is good-good condition (i.e., good TIR condition and good GV condition), and 6 The subsequent states after the initial state indicated by 13B are bad-good states (i.e., bad T IR status and good GV status). Chart 612 shows the various statuses of the patient over the course of several months. Each state (e.g., 613A, 613B, etc.) is a representative state for that period. For example, the initial state indicated by 613A is the average of all states during the first month. It may be an average value, or the same day of the month is used for each month shown in chart 612. The chart 614 in FIG. 6B is similar to the chart 612. However, the two-dimensional state-based quadrants show the corresponding TIR and GV states. When associated with states, chart 614 is a secondary chart that allows the viewer to see the distribution of states. Chart 612 and / or chart 614 show the original state-based quadrants. To enable a User 8 or a healthcare provider to better understand the User 8's condition status, It may be provided to the donor 7.

[0082] FIG. 6C shows charts 616, 618, and 620, each with various amounts of data. Chart 616 has the most data with 15 months of CGM-based status information. Chart 618 shows eight months of CGM-based status information, and chart 620 shows Shows 3 months of CGM-based status information. More data points allow viewers to Understanding a patient's blood glucose-based history more holistically than with a few data points It can be made possible.

[0083] FIG. 6D shows charts 621A and 621B showing GV values ​​for a given patient over a 15-month period. Chart 621A shows the standard deviation (SD) of the blood glucose readings, and Table 621B shows the coefficient of variation (CV) of the blood glucose readings. The type of GV (e.g., SD, CV, etc.) may change the GV state at a given time. For example, 621C of Chart 621A shows the SD-based GV value of the 9th measurement. As shown, 621C corresponds to a poor GV condition. However, Chart 6 The CV-based GV value of the same corresponding 9th measurement represented by 621D in 21B. corresponds to a good GV condition. The type of GV applied (e.g., SD, CV, etc.) is Based on one or more factors, such as patient vector, historical glucose information, and patient characteristics. The selection may be made as follows, but is not limited to these.

[0084] Figure 6E shows a chart of various conditions for each of multiple patients, represented by anonymized patient IDs. For example, patient 624 (i.e., patient ID 42799) is shown in bar 6. 26A, a large number of missing conditions (e.g., due to missing CGM data); A number of good-good conditions (i.e., good TIR conditions and Good GV condition), a number of poor-good conditions represented by bar 626C, and bar 62 A given cohort of patients may have multiple poor-poor conditions represented by 6D. The medical institution or healthcare provider 7 monitoring the patient may be provided with the chart 622 on a regular basis. By reviewing the visual changes in state shown in the graph 622, the viewer can Trends in state changes for all or a subset of users who implement the technology disclosed in this specification The direction may be easily determined.

[0085] The medical institution or healthcare provider 7 may also use the chart 628 and / or diagram 6 of FIG. 30. A medical institution or healthcare provider 7 may use the chart 628 to For example, the chart 628 can be viewed by The participant counted the number of patients whose status changed from good to poor to poor to good (i.e., 1 It may be possible to determine that 1) is more common than the opposite direction (i.e., 9). The new data can be used to refine the network layers (e.g., weights, etc.) and provide improved optimized routes. (to provide a better understanding of the habit index) and to update machine learning algorithms (e.g., to improve how optimization paths are provided / implemented (e.g., based on changes made to the may be used.

[0086] Similarly, the diagram 630 may be used by a medical institution or healthcare provider 7 to The trend of the change in the status of the group can be reviewed by using the diagram 630. This allows viewers to quickly see trends in status changes and over multiple time periods. For example, a viewer who has been provided with diagram 630 may compare such trends. The number of status changes from poor-good to good-good (i.e., 20) is easily calculated. The chart 628 and diagram 630 may be compared to the change from the previous month. is expressed as a number of stat changes, whereas stat changes are expressed as percentages of change, etc. It will be understood that it may be represented in any applicable manner.

[0087] As disclosed herein, the optimized path generated in 516 of FIG. Or it may be generated based on multiple patient vectors. 6 shows a chart 632 of a patient's blood glucose readings 634 (collected using a CGM monitor) A filter or other smoothing mechanism is used to generate the corresponding trend line 636. The chart 638 may be a smoothed gradient of the blood glucose readings 634 or the chart 632. The first derivative 640 of the chart representing the rate of change of the trend line 636 is shown. Both the and the machine learning model can process such vectors and their associated attributes (e.g. For example, the time of day, duration of exercise, food type, medication type and / or administration of each given vector. The motion vector 642, the food vector 643, and the characteristics of the vector (such as the amount of 44, a patient vector including a medicine vector 646 and another food vector 648. The blood glucose reading 634 is calculated mechanically along with the first derivative 640 of the patient's blood glucose reading 634. The blood glucose level may be used as an input to a learning model, or may be measured 634 or a primary blood glucose level. Any of the derivatives 640 may be used individually. The output optimization pathway provided by the present invention may be based on the patient's blood glucose measurement 634, a patient vector (e.g., Exercise vector 642, food vector 644, medicine vector 646, and another food vector 6 48), and / or may be based on the first derivative 640.

[0088] As shown in Figure 4B, the AGP report provides not only graph data but also a number of metrics. These metrics may be used by patients and healthcare providers. All 10 scales can be numerous and difficult to understand for both the presenter and the presenter. One of the AGP reports is that the degree is not necessary because it does not provide unique, independent information. The components of this metric are determined by other measures and are therefore not intended to be limiting. The technique described is based in part on minimizing the number of metrics. B shows the mean glucose 704, the hyperglycemic range (TAR) and TIR 706, and the fluctuation of glucose 710, hypoglycemic range (TBR), percentage of CGM activation time 722, and / or glucose The standard deviation and variation of the principal component values ​​(PCVs) 724 are used to determine whether the techniques disclosed herein Average glucose 704 indicates the average glucose and glucose control. The blood glucose level may be based on a GMI (Glucose Measurement Index), which is a predictive indication of blood glucose levels. R and TIR706 are very high TAR indications (TAR_VH), high TAR indications ( The glucose excursion 710 may be based on an indication of glucose saturation (TAR_H), TAR_H, and / or TIR. The TBR712 can be calculated based on the standard deviation of the course and PCV. , very low TBR (TBR_VL), and TBR indications.

[0089] According to one implementation of the disclosed subject matter, one or more CGM events are associated with a patient's blood glucose level. The classification may be based on the one or more CGM events associated with each of the events. The severity score may be based at least on the severity score and / or may be related to the patient's blood glucose level. The method may be based on one or more characteristics of the curves. 5A shows an exemplary classification of CGM events. The CGM event may be based in part on one or more classified CGM events. The severity or other characteristics of the disease may be fed into a machine learning model, and the optimized pathway may be one or The output may be based at least in part on the plurality of classified CGM events. As such, the severity score may show the presence of a sharp peak, or the presence of a high severity score. The frequency may indicate a high amount of fluctuation in the patient's blood glucose levels. The report, for example, suggests that TIR may not reveal unhealthy fluctuations in a patient's blood sugar levels, making it more likely that a patient has high T This can be particularly useful if you have IR.

[0090] Applying CGM events to determine an optimized path can be achieved by using a CGM trace (e.g., Detection of events from a series of glucose readings and classification of the events into one or more classes The classification may include classification based on severity score and / or glucose count. The severity score may include a category. The severity score may be calculated using time and shape characteristics of the CGM trace. It may be determined.

[0091] Severity scores and / or CGM events are determined with respect to individual variations in CGM data. and may be part of the microview of the CGM. Events can be real-time coaching or behavioral outputs (e.g., related to medication, diet, exercise, etc.). Therefore, the techniques disclosed herein may be used for The macro view of CGM data (e.g., using state data as illustrated in Figure 5A) and micro-views of CGM data (e.g., severity scores and / or CGM events) to provide both real-time feedback and overall well-being. Provide both feedback and improvements.

[0092] FIG. 8A illustrates a chart 800 having a CGM trace 802 that includes a CGM event 802A. FIG. 8B shows a chart 800 with a CGM trace 806 including a CGM event 806A. 4. CGM events 802A may be detected based on one or more mathematical methods. In the example provided in FIG. 8A, a CGM event 802A has three parameters: Clinical significance, including starting glucose, severity, and ending glucose (b, s, d) can be classified based on a multi-parameter CGM classification.

[0093] Parameter b corresponds to the glucose category at or near the start of a given CGM event. Glucose category b can be very high (e.g., +2), high (e.g., + 1), in range (e.g., 0), low (e.g., -1), or very low (e.g., -2 In the example of CGM event 802A, trace 802 at the beginning of CGM trace 802 as it curves towards the peak of CGM event 802A. As shown via trace 802, which is within the threshold range indicated by 803, The blood glucose level indicated by the CGM source 802 is within the threshold range 803 at the start of the CGM event 802A. , b corresponds to 0. In the example of CGM event 806A in FIG. 8B, trace The start of CGM event 806A when 806 curves towards the peak of CGM event 806A As shown via trace 806, which is sometimes within threshold range 805, Thus, the blood glucose level indicated is within the threshold range 805 at the start of CGM event 806A, b corresponds to 0.

[0094] The parameter s is the relationship between the height of the curve of the CGM event and the time the curve remains above the target. The severity score s may correspond to a severity score that encompasses both the height of the curve of the CGM event and the severity score s. and a value (e.g., 0 to 9) indicating the duration the curve remains above the target. The severity score may be calculated based on the height of the CGM curve and the corresponding trace that is outside the threshold range. Any applicable technology that provides a severity score based on the combination of the duration of the As a simplified example, the value associated with the height of the curve may be calculated via may be multiplied by a value associated with the duration of the trace that falls outside the threshold range. One or both of the duration values ​​may be greater than 1. According to one implementation: Height and duration are used to determine whether the severity score is based more on one of height or duration. As shown, different weights may be assigned. Higher severity scores indicate a greater likelihood of achieving the goal. A higher combination of height and duration may indicate a lower severity score. may indicate a lower combination of height and duration above the target. , a lower severity score may be preferable to a higher severity score.

[0095] In the example of CGM event 802A, the parameter s is associated with CGM event 802A. The height of the curve and the duration that the trace 802 is outside the threshold range 803 are used to determine the This corresponds to a severity score of 6. In the example of CGM event 806A, the parameter The data are the height of the curve associated with the CGM event 806A and the time it takes for the trace 806 to fall within a threshold range. 805, corresponds to a severity score of 2, determined based on the duration outside the 8A and 8B, the height of the curve of the CGM event 802A and the target threshold range The duration of time is determined by the height of the curve of the CGM event 806A and the duration of time outside the target threshold range. is greater than the time. Therefore, when compared to CGM event 806A (i.e., 2), , the severity score for CGM event 802A is higher (i.e., 6).

[0096] The parameter e corresponds to the glucose category at or near the end of a given CGM event. Glucose category b can be very high (e.g., +2), high (e.g., + 1), in range (e.g., 0), low (e.g., -1), or very low (e.g., -2 In the example of CGM event 802A, trace 802 8 at the end of CGM trace 802 when it flattens after the peak of CGM event 802A. As shown via trace 802, which is approximately outside the threshold range indicated by . The blood glucose level indicated by trace 802 is in range 80 at the end of CGM trace 802A. 3, so e corresponds to 1. In the example of CGM event 806A in FIG. 8B, When the race 806 flattens and changes direction below the threshold range 805, the CGM event 806 As shown via trace 806, which is within the threshold range 805 at the end of trace A, If the blood glucose level indicated by 806 falls below the threshold range 805 at the end of the CGM event 806A, Since it is in , e corresponds to -1.

[0097] According to another implementation, the CGM events are characterized using one or more other techniques. For example, CGM events may be characterized based on the severity score and shape of the CGM event. Figures 8C and 8D show the severity score and CGM event trace shape. As disclosed herein, the severity scale is used to characterize CGM events. Cores are based on the height of the CGM trace and the duration of the trace that is outside the threshold glucose range. The shape of the CGM event may be calculated based on, for example, broad category, high category, The categories may be classified in any applicable way, such as a normal category. The goal may be based on the start, peak, and end of a given CGM event, The parameters associated with classifying a race as a given category are predetermined. or may be determined based on a given patient, multiple CGM tracings, etc. For example, the ratio of the area outlined by a given CGM trace to the height of the trace is , may be used to classify CGM traces.

[0098] FIG. 8C shows a CGM trace 812, a threshold range 803, and three CGM events 812A, 8 shows a chart 810 with CGM events 812B, 812C, and 812D. The first CGM event 812A is , with a severity score of 8 and a CGM trace shape characterized as wide. The second CGM event 812B was characterized as having a severity score of 5 and a CG The third CGM event 812C has a severity score of 0 and a normal M trace shape. The third CGM event 812C has a CGM trace shape characterized as The severity score is calculated based on whether the trace 812 at the peak of the CGM event 812C is within the threshold range 813. Since it is in , it is 0.

[0099] Depending on the implementation, the CGM trace shape may also be characterized as short. Additionally, the machine learning model can, for example, generate CGM traces based on past CGM trace shapes. The machine learning model may be used to identify the shape of the updated CGM trace. For example, updated glucose values ​​may be updated based on the severity of the optimized pathway. Score, calculated by the CGM monitor after being provided based on CGM trace shape, etc. The updated glucose values ​​may include the impact of the optimized route on the user. The updated glucose values ​​are provided to the machine learning model to update the model. This may be used to generate an updated CGM trace. For example, If the user does not improve their state, the machine learning model will It may be updated to improve the output.

[0100] FIG. 8D shows a CGM trace 822, a threshold range 823, and two CGM events 822A and 822B. The first CGM event 822A shows a chart 820 with a number of overlaps of 9 and 822B. The second CGM has a severity score and a CGM trace shape that is characterized as high. Event 822B had a severity score of 8 and a CGM trace pattern characterized as normal. It has the following characteristics.

[0101] One or more clinically significant CGM events for a given user8 are classified into CGM classifications (e.g., For example, b, s, e in Figures 8A and 8B, or the severity scores and shapes in Figures 8C and 8D. characterization, or any other applicable characterization). Optimization pathways (e.g., via automated coaching messages, DSMA, etc.) are tailored to the CGM situation. Further based on the characterization of the elephant, it may be transmitted to user 8.

[0102] FIG. 9 includes a flowchart 900 for one implementation of the disclosed subject matter. In this case, multiple optimization profiles are generated to reach the ideal state from the non-ideal state. Multiple optimization profiles may not be patient-specific, but may be patient-based. Each of the above-described methods may be generated for a combination of vector and patient attribute. As such, multiple optimization profiles were trained using machine learning models as provided in Figure 5B. Multiple optimization profiles can be generated using the output of a machine learning model. may be provided as a set of criteria, may be based on a cohort of historical patients, and may be derived from non-ideal conditions. It may further be based on successful or unsuccessful attempts to reach the ideal state.

[0103] The multiple optimization profiles each include one or more patient attributes and / or patient vectors. For example, for a given set of patient vectors and patient attributes, ,A particular optimization profile can be used to handle possible ,nonideal conditions (e.g., good-bad, bad-bad). , bad-good, etc.).

[0104] At 904, the TIR status of a given patient is determined, and at 906, the TIR status of a given patient is determined. With the techniques shown, the GV status of a given patient can be determined. One or more patient vectors and one or more patient attributes of the patient may be received. The vectors and / or patient attributes may be provided by a given patient and / or by a healthcare provider. Advantageously, via electronic device 19, via server 29, or any other applicable means. The data may be acquired through stages.

[0105] At 910, an optimization profile is identified based on the patient vector and the patient attributes. An optimization profile may include a limited number of optimization paths, each of which , may correspond to a predetermined combination of TIR and GV states. The file is in a good-bad starting state, a bad-bad starting state, and a bad-good starting state. Therefore, a predetermined optimization profile may include an optimization path related to patient attributes. and vectors, and a limited number of optimizations based on the patient's starting state. It may also include a profile.

[0106] At 912, an optimization pathway is determined from the optimization profile based on the TIR status and The optimized pathway can be identified based on the patient's vectors and attributes. For example, during the first iteration, a given The patient's optimization profile is based on the patient's attributes and vectors at the time of the first iteration. The TIR status and GV status of the patient during the first iteration (e.g., good- However, a first optimized path may be identified based on the second During recovery, the predetermined patient vectors are the same (i.e., the same optimization profile is identified). Even if the system is in a state where the optimization is not possible, different optimization paths may be implemented based on the change in state (e.g., from bad to good state). At 914, the technique disclosed herein may be used to determine the optimal A route of administration may be provided to a given patient and / or healthcare provider 7.

[0107] Steps 502-517 of FIG. 5A, 512-530 of FIG. 5B, and 902-91 of FIG. 9 4 is depicted in a particular order, the principles of the present disclosure are not limited to the order depicted therein.

[0108] FIG. 10A is a diagram 1002 including a chart 1002 of CGM events by time and day. Such a diagram or other visual output may be used to visualize the CGM journal for a given period. To make the need easier to understand, The diagram 1000 may be provided to a healthcare provider 7 or a user 8 via the application 1. contains journey days as the Y-axis and time as the X-axis. 000 and easily identify patterns over a given day, time, and / or number of days or times. It may be judged as follows.

[0109] FIG. 10B shows a chart 1012 of total carbohydrates ingested by time of day and meal type. 10 shows a diagram 1010 including the diagram. Such a diagram or other visual output may be To more easily understand your eating habits over time, you can use an application (e.g., mHealth The diagram can be provided to a healthcare provider 7 or user 8 via an application 1). The diagram 1010 includes total carbohydrate count as the Y-axis and time as the X-axis. The pattern 1010 is received and displayed on a predetermined day, time, and / or number of days or times. For example, the user may easily determine the types of meals consumed during the day, and You can easily see the calories associated with a meal type.

[0110] FIG. 11 illustrates a first patient, as shown via chart 1102, and a second patient, as shown via chart 1103. Chart 1 shows the severity score counts for the second patient as shown through 4. Each bar in chart 102 and chart 1104 represents the number of patients with a given severity score for each first patient and In general, such a distribution represents the number of times that the first patient's CGM data was shown in the second patient's CGM data. A higher count with a lower severity score is preferred as it may indicate better diabetes control. A medical institution or health care provider7 may provide one or more Even if the system receives multiple patient distribution charts (e.g., chart 1102 and chart 1104), Optimally, a scatter plot may be used to monitor the progress of an entire patient population. The distribution chart can be used to show the distribution of patients for specific patient groups (e.g., based on time of treatment, based on medical team, etc.). Patient attributes, patient vectors, etc. may be generated and trends of specific patient groups may be analyzed. Alternatively, trends between multiple patient groups may be compared.

[0111] Figure 12 shows a CGM-based approach to provide coaching to patients based on CGM data. 12 shows a diagram 1200 of an implementation of the C The attributes can be provided to the GM message generator 1212. trends 1204 (e.g., CGM trends, CGM event data, etc.), carbohydrate information information 1206, activity information 1208, insulin information 1210, etc., or a combination thereof. Based on the attributes, the message level may be determined. For example, the message level can be 1222, where urgent action is required (e.g., For example, the user should take insulin or consume carbohydrates to avoid harm. ), an alert level 1224 where action may be required immediately but is not urgent (e.g., user (If you are taking this medicine, you should carefully monitor your glucose levels) or an informational message is provided. The default value may be 1226 (e.g., no action by the user is required).

[0112] The CGM message generator 1212 generates a patient vector (e.g., attributes 1202-1210 ) may be used to provide an optimized route based on the 9 may be applied at 914. For example, the optimized route may be applied at 914 ... The message may be determined based at least in part on the CGM message generator 1212. and may be provided to a corresponding patient.

[0113] FIG. 13A shows an example message provided using the CGM message generator 1212. The message 1302 is provided via the electronic device 19 of the user 8. In the example provided in FIG. 13A, the message may be a message of the behavior level 1222. As shown, the exemplary message 130 2 says, "Action Required: Hey Charlie, your glucose is high and rising rapidly Go to your insulin computer and schedule your insulin dose so you can get back in range. Go and get it." Message 1302 is sent with high importance. 1300 to the user 8 via the mobile phone 1300. 9 provides audible, tactile, and visual warnings in addition to the message 1302. may result.

[0114] FIG. 13B shows a CGM message provided using a CGM message generator 1212 via a mobile phone 1300. 13 shows another exemplary message 1314 provided. Message 1314 is a warning level 1 224 message and may not include urgent action. Message 1314 reads, "No action required: Charlie, your glucose is on target, but , has risen slightly. No action is required at this time. Additional information such as blood glucose levels 1312 may also be provided via the mobile phone 1300 and associated messages. The hologram may be provided near page 1314.

[0115] FIG. 13C shows a CGM message provided using a CGM message generator 1212 via a mobile phone 1300. 13 shows another exemplary message 1322 provided by the advisory board. This is a bell 1226 message containing general advice for patients. As shown, message 1322 also includes carbohydrate information, blood glucose levels, insulin doses, etc. , other patient vectors.

[0116] Therefore, as shown through the examples in Figures 13A-13C, machine learning-driven autonomy Dynamic user coaching or CGM feedback may be provided to the user 8. For example, The system and method are intended to address critical situations such as hypoglycemia or extreme hyperglycemia, where critical action is required. This can be used to provide a warning when a less serious glucose reading occurs. Notification messages may also be provided. Insulin administration support may be provided via an optimized pathway. As part of the patient vector correction, it provides correction insulin based on glucose trends. According to one implementation, insulin dosage may be adjusted based on glucose trends (e.g., C The current blood glucose level may also be a patient vector that may be adjusted based on the current blood glucose level (GM event). This may be a factor when determining insulin adjustment amounts. For example, bolus insulin , a period of time (e.g., 30 minutes) may be required to provide the intended results, and accordingly Therefore, a trend forecast at the end of that period may be more useful than the current blood glucose level alone. ,trends can be an important component.

[0117] According to one implementation of the disclosed subject matter, an insulin computer may be provided. The insulin computer may, for example, generate behavioral output including the amount of insulin to administer at a given time. A context computer receives one or more factors as input to provide a The insulin computer may be part of a CGM monitor, or may be external to the CGM monitor (e.g., one or more electronic devices 19 The external insulin computer may be connected to an electronic network 32 or the like. It may be connected to a CGM monitor via a wired or wireless connection.

[0118] An insulin computer is software that runs on a CGM monitor or external device. For example, the insulin computer may be an mHeal The insulin computer may be part of one or more It may receive a number of complex inputs and provide a behavioral output, which may include insulin whether insulin is needed, how much insulin is needed, and how much glucose is needed whether or not, how much glucose is needed, and food intake is required whether, how much food intake is needed, whether exercise is needed, and how much How much exercise is required? Including, but not limited to, one or more rows The insulin computer's functions include: , may change based on the user's condition. For example, the behavioral output may be to optimize the user's blood glucose level. In this embodiment, the CGM trainer may be adjusted to maintain a safe and effective range. The input can be used to determine the optimal blood glucose level.

[0119] The insulin computer may receive CGM trends as input. As disclosed in detail, CGM trends include CGM traces, CGM events, etc. Alternatively, it may be based on a CGM trace, CGM events, etc. May be based on changes in two or more glucose measurements over a period of time. CGM Trends may be based on glucose measurements provided by a CGM device. The trend is corrected or updated as additional glucose measurements are taken. Historical CGM trends may also be used as input. This may also be done.

[0120] The insulin computer may receive meal information as input. The user input can be used to capture the content (e.g., images, videos, etc.) or examples of food before it is consumed. representative foods (e.g., a picture of pizza found online to represent a food eaten) Provide the insulin to the computer in any applicable manner, such as by entering The content may be entered using the electronic device 19 or may be generated by the user. The food intake of the user 8 may be received from a resource such as an application that tracks the user's food intake. The information is collected at a point in time (e.g., when a computer is used to determine a behavioral output). ) for user 8, may include insulin carbohydrate ratio, or insulin A computer may calculate the insulin carbohydrate ratio. The behavioral output based on the dietary information of a user 8 is compared with that of another user who has the same dietary information on a given day. The impact of food intake may be individualized for each user, so that it may vary from user to user. The insulin computer will calculate one or more insulin doses based on the meal information and / or insulin carbohydrate ratio. The behavioral outputs may be coordinated. Past dietary information may also be used as an input. stomach.

[0121] The insulin computer may also receive exercise (e.g., any activity) information as input. Exercise information can be generated by user input (e.g., past or planned exercise) or by For example, an exercise or health tracker (from an electronic device 19) may monitor one or more of the CGMs. by any applicable method, such as by multiple components, one or more sensors, etc. The exercise information may be provided to the insulin computer by the following: calorie information, heart rate information, The insulin calculation may include the duration of exercise, exercise intensity, physical strain, etc. The data is used to calculate the behavioral output based on the exercise information of a user 8 in relation to another user who has the same exercise information on a given day. The effect of exercise on users 8 may be individualized so that it may vary by user. The sling computer may adjust one or more behavioral outputs based on the movement information. Past athletic information may also be used as an input.

[0122] The insulin computer receives information about previous insulin doses as input. As discussed further herein with reference to FIG. 13D, The data may be based on one or more factors associated with the user (e.g., diet, exercise, individual physical condition, etc.). Taking into account factors such as body characteristics, CGM trends, and past data, For example, if the user 8 decides to administer additional insulin, the behavioral output may be determined based on the time of day. When determining whether to administer insulin, the half-life of the administered insulin should be considered. The computer determines how much insulin is still in the user's body from the previous dose. It may also be determined whether

[0123] The insulin computer may receive as input information about current blood glucose levels. Additionally, the insulin computer may use CGM trends (e.g., the user's body temperature) as input. The patient may receive information about the current blood glucose level and / or the rate of change of glucose in the blood. M Trend is a function that allows the insulin computer to monitor the blood glucose trends (e.g., increase or decrease) of the user's blood glucose levels. Such information may enable the determination of the magnitude of the change (e.g., increase, decrease, stability, etc.) and the rate of change. Based on this, the insulin computer may adjust one or more behavioral outputs. The blood glucose level may also be used as an input.

[0124] The insulin computer receives as input the user's sensitivity to insulin. The sensitivity to insulin may be based on a predetermined value or may be determined by measuring the insulin sensitivity. This may be based on past data received from a sleep computer, CGM monitor, etc. According to one implementation, the sensitivity adjusts over time based on the user's 8 insulin usage. Therefore, the insulin computer may periodically monitor the patient's sensitivity to insulin. It may be updated periodically or whenever a user's behavioral output is calculated.

[0125] The insulin computer may receive as input the user's 8 history of hypoglycemia. When the glucose computer provides a behavioral output, it is necessary to determine the relationship between the hypoglycemic event and the calculation of the behavioral output. The insulin computer may also consider periods of low The severity of the glycemic event may also be taken into account. As an example, the history of the user 8 may be used to calculate the behavioral output. If the user has a hypoglycemic event within the past two days, or if the user has a blood glucose level greater than 4% for three consecutive days If you experience a hypoglycemic event, the insulin computer recommendation will be may be more conservative than in the case of

[0126] FIG. 13D illustrates an exemplary insulin computer 100 according to one implementation of the disclosed subject matter. 330. As shown, of the four time zones provided in FIG. 13D One time zone is the time when the behavioral output provided by the insulin computer 1330 , the indicators may be modified based on the time zone at the time of providing the behavioral output. Each time zone may be provided as an input to the user computer 1330. The first time may be determined based on the last time the patient received insulin (e.g., a bolus injection). Zone 1334 is the time from the time of the meal bolus 1332 to the time of the meal bolus. The second zone 1336 may be within two hours of the meal bolus 1332. The third zone 1338 is for a period of 2 to 4 hours from the mealtime bolus 1332. The fourth zone 1340 may be greater than four hours from the meal bolus 1332. Each of the four zones may be divided into four zones, as shown in Figure 13D. Including Intake On Board (IOB), Correction Factor (CF), Insulin Carbohydrate Ratio (ICR), etc. If in the first zone 1334, the IOB, CGM token, CF adjusted based on trend and ICR administration can all be considered. If the patient is within 336 days, IOB and CF may not be considered, but ICR administration may be considered. If within the third dime zone 1338, IOB, CF without CGM adjustment If in the fourth zone 1340, IOB, CGM, and ICR administration may be considered. Trend-adjusted CF and ICR dosing may all be considered.

[0127] Therefore, based on the factors discussed herein, the insulin computer whether insulin is needed, how much insulin is needed, and glucose whether glucose is needed, how much glucose is needed, and whether food intake is needed. whether or not a diet is required, how much food intake is required, and whether or not exercise is required , how much exercise is needed, etc., or a combination thereof. The insulin computer may provide behavioral outputs such as, but not limited to, As a result of one or more factors, such as the different histories of each patient, the first user with the input personalization, so that a user may receive different behavioral outputs than a second user with similar inputs. The contextual behavior output may be provided.

[0128] According to one implementation, the one or more behavioral outputs are part of an insulin computer. or determined using a machine learning model associated with the insulin computer. The machine learning model may be based on known good outputs and / or The past behavioral output provided by the CGM trainer and the past behavioral output provided by the CGM trainer A supervised model trained to provide behavioral outputs based on the corresponding changes in the trend. For example, the machine learning model may be one or more of the following, as discussed herein: The machine learning mode may be configured to provide a behavioral output based on multiple inputs. may receive updated CGM trends after providing the behavioral output. Dell analyzes CGM trends and updates models (e.g., weighted) based on CGM trends. (updating neural networks, layers, etc.) and the future performance provided by the machine learning model. The machine learning model may improve future action outputs (e.g., actions provided to the user). output and subsequent CGM trends of the user) with respect to an individual or to one or more Multiple users based on feedback (i.e., CGM trends) from multiple users The model may be updated with respect to

[0129] FIG. 14 illustrates a system that may be configured to host a server to act as a decision server for a healthcare provider, for example. 14 is a simplified functional block diagram of a computer that can be configured as shown in FIG. denotes a network or host computer platform 1400. Those skilled in the art will appreciate that Familiar with the construction, programming, and general operation of such computer equipment and as a result, the drawings are intended to be self-explanatory.

[0130] For example, a platform such as the server 1400 may be configured to provide data for packet data communication. The platform may also include a data communication interface 1460. a central processing unit (CPU) 1420 in the form of one or more processors for executing The platform typically includes an internal communication bus 1410, a program storage and the like processed by the platform, such as ROM 1430 and RAM 1440. It includes data storage for various data files that are stored and / or communicated. hardware elements of the device, operating systems, and programming languages, such as is conventional in nature and those skilled in the art are presumed to be fully familiar with it. The server 1400 also includes a keyboard, mouse, touch screen, monitor, and display. and a communication port 1450 for connecting to an input / output device such as a Of course, various server functions may be implemented using multiple parallel servers to distribute the processing load. Alternatively, the server may be implemented in a distributed manner across multiple platforms. Implemented by appropriate programming on two computer hardware platforms This may also be done.

[0131] As described herein, the present disclosure may be implemented in software, hardware, firmware, or any combination thereof. and / or that the entities shown in the figures may be implemented in many different embodiments. It will be apparent to those skilled in the relevant arts. Any actual software code with expert control is not limited to the detailed description. Thus, the examples are not intended to be limiting given the level of detail provided herein, and variations and modifications of the examples are within the scope of the present invention. Aspects of the subject matter described are described with the understanding that various modifications and variations are possible. A form of executable code and / or associated data carried or embodied on a type of medium "Storage" type media can be considered as "products" or "articles of manufacture" of any kind. Any or all of the tangible memory, such as computers, processors, etc., or various semiconductors memory, tape drives, disk drives, and other related modules. may provide non-transitory storage for software programming at any time. Portions of the Software may, at times, be transmitted over the Internet or various other telecommunications networks. For example, such communication may be via, for example, a mobile communication network. from the management server or host computer to the server's computer platform, and / or from a server to a mobile device from one computer or processor It may also be possible to load the software into another computer or processor. Therefore, another type of media that may involve software elements is the physical interface between local devices. through the Internet, through wired and optical fixed line networks, and through various air links. This includes light waves, radio waves and electromagnetic waves, such as those used over wires, cables, or other devices that carry such waves. Physical elements, such as wireless links, optical links, etc., may also be considered as media carrying the software. As used herein, "storage" refers only to non-transitory, tangible "storage" media. Unless otherwise specified, terms such as computer or machine "readable medium" shall not be construed as limiting the scope of the invention. "Instruction" refers to any medium involved in providing instructions to a processor.

[0132] Both the foregoing general description and the following detailed description are exemplary and explanatory only. It is to be understood that the invention is not limited to the disclosed embodiments as set forth in the claims.

[0133] Other embodiments of the present disclosure will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It will be apparent to those skilled in the art that the specification and examples are to be considered as exemplary only and that no part of the invention is to be construed as limiting the scope of the invention. It is intended that the true scope and spirit of the invention be indicated by the following claims.

[0134] As is apparent from the figures, text, and examples presented above, various embodiments are possible. However, it is not limited to these.

[0135] 1. A computer-implemented method for managing a user's glucose status, comprising: receiving a blood glucose level of said user using a continuous glucose monitoring (CGM) device; And, determining a time-in-range (TIR) ​​value for the user's blood glucose level, is based on the amount of time the user's blood glucose level is within a threshold band over a reference period. The determining; determining a TIR state based on the TIR value; receiving a blood glucose variability (GV) value based on at least the blood glucose level of the user, The GV value is one of the standard deviation or coefficient of variation (CV), and the CV is the a time series that indicates the fluctuation of the user's blood glucose level, taking into account a standard deviation of the blood glucose level over a sub-period; and determining a GV state based on the GV value; determining a starting state based on the TIR state and the GV state; determining that the starting condition corresponds to a non-ideal condition; to arrive at an ideal state based on one or more user vectors and the starting state. generating an optimized route, the optimized route being said generating including one or more adjustments of vectors; providing the optimized route to the user; A method comprising:

[0136] 2. An embodiment in which the threshold band is between approximately 70 mg / dL and 180 mg / dL The method described in 1.

[0137] 3. The method of embodiment 1, wherein the reference period is 24 hours.

[0138] 4. The CV value is the standard deviation of the blood glucose levels over the reference period. 2. The method of embodiment 1, wherein the mean value is determined by dividing by the mean value.

[0139] 5. The TIR condition is selected from one of a good TIR condition or a bad TIR condition. 2. The method of claim 1, wherein the state is a binary state.

[0140] 6. The good TIR condition corresponds to a TIR value greater than the TIR cutoff. The method according to embodiment 5.

[0141] 7. The GV condition is selected from one of a good GV condition or a bad GV condition. 2. The method of claim 1, wherein the state is a binary state.

[0142] 8. An embodiment in which the good GV status corresponds to a GV value greater than a GV cutoff. 7. The method described in Item 7.

[0143] 9. The user vector may be a combination of medication, food intake, exercise values, psychosocial parameters, or social 2. The method of embodiment 1, comprising one or more of the following quantitative determinant parameters:

[0144] 10. Classifying one or more CGM events based on the user's blood glucose level. wherein said classifying includes classifying a severity associated with each of said one or more CGM events. the classifying is based at least on a score, and the classified one or more Cs. generating the optimized route further based on a GM event. The method described below.

[0145] 11. The optimized route is further based on user attributes, and the user attributes are social or one of: a medical attribute, a user preference, a metabolic attribute, or a user demographic 2. The method of embodiment 1, wherein the method is selected from a plurality of

[0146] 12. The optimized pathway may include increasing one or more condition-improving habits and / or 2. The method of embodiment 1, comprising a reduction in multiple condition-exacerbating habits.

[0147] 13. A computer-implemented method for managing a user's glucose status, comprising: Receive multiple optimization profiles to reach an ideal state from a non-ideal state The ideal state is a good time-in-range (TIR) ​​state and good blood glucose variability. The non-ideal conditions correspond to a bad TIR condition or a bad GV condition. said receiving including at least one of: determining a current TIR state based on a TIR value of the user's blood glucose level, The TIR value is based on the amount of time the user's blood glucose level is within a threshold band over a reference period. The current TIR status is based on whether the TIR status is good or bad. the determining step being one; determining a current GV state based on a GV value associated with the user's blood glucose level; wherein the GV value indicates the standard deviation (SD) or coefficient of variation (CV) of the blood glucose level; The CV is the user's blood glucose level, taking into account the standard deviation of the blood glucose levels over the reference period. the determining indicates a change in value; receiving one or more user vectors for said users; determining the optimal user vector based on the one or more user vectors and one or more user attributes; Identifying one of the optimization profiles; The optimized profile is determined based on the identified optimization profile, the TIR condition, and the GV condition. identifying an optimized path, the optimized path being a path that is based on the one or more user vectors; said identifying including one or more adjustments of the rule; providing the optimized route to the user; A method comprising:

[0148] 14. Each of the plurality of optimization profiles includes a plurality of user vectors and a plurality of user 14. The method of claim 13, including different combinations with attributes.

[0149] 15. The plurality of optimization profiles are each associated with a plurality of optimization paths; and each of the plurality of optimized paths is one of a potential TIR state or a potential GV state. 15. The method of embodiment 14, wherein the method is identified based on one or more of:

[0150] 16. The machine learning model receives as input the optimization profile, the TIR state, and and the GV status; and outputting the optimized route.

[0151] 17. Receiving one or more user attributes and processing the one or more user attributes. and identifying one of the optimization profiles based on the 14. The method of embodiment 13.

[0152] 18. The CV value is the standard deviation of the blood glucose levels over the reference period. 14. The method of embodiment 13, wherein the mean value of the mean value of the saturation level is determined by dividing the mean value of the saturation level by the mean value of the saturation level.

[0153] 19. A system for managing a user's blood glucose level, comprising: a memory for storing processor-readable instructions; a processor configured to access the memory and execute the processor-readable instructions; The processor and Including, The processor-readable instructions, when executed by the processor, and obtaining glucose levels using a component that penetrates the user's skin. a continuous glucose monitoring (CGM) device configured to: receiving the glucose level electronically; determining a time-in-range (TIR) ​​value for the user's blood glucose level, The value is based on the amount of time the user's blood glucose level is within a threshold band over a reference period. , the threshold range is between approximately 70 mg / dL and 180 mg / dL, and the reference period is 24 hours; determining a TIR status based on the TIR value, wherein the TIR status is good; said determining being selected from a good TIR condition or a bad TIR condition; receiving a blood glucose variability (GV) value based on at least the blood glucose level of the user; The GV value is one of a standard deviation or a coefficient of variation (CV), and the CV is and indicating the variability of the user's blood glucose levels, taking into account the standard deviation of the blood glucose levels over a reference period. said receiving; determining a GV state based on the GV value, and the GV state is a good GV state; determining that the condition is one of a GV condition or a poor GV condition; determining a starting state based on the TIR state and the GV state; determining that the starting condition corresponds to a non-ideal condition; Detecting a CGM event based on the user's blood glucose level; Multiparameter CGM classification or severity and CGM event trace shape characterization characterizing the CGM event based on one or more of the multi-part The parameter CGM classification was based on the blood glucose level at the start of the CGM event, the severity, and the CGM event. said characterizing including the blood glucose level at the end of said period; Based on one or more account vectors and the characterization of the CGM event. generating an optimized path to reach the ideal state based on the optimized path; The path includes one or more adjustments of the one or more account vectors. And, providing the optimized route to the user; 20. A system, comprising: a processor configured to perform a method comprising:

[0154] 20. Providing the optimized route to the user includes: 20. The system of claim 19, further comprising providing contextual text-based instructions to the user. Tem.

[0155] Additional embodiments include the following:

[0156] 1. A system for providing glucose trend-based behavioral output, comprising: Outputting multiple glucose readings based on analyzing body fluids over a period of time a continuous glucose monitoring (CGM) device configured to: a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: The multiple data output by the CGM device and / or stored in the memory Determine CGM trends based on changes in the number of glucose measurements, taking at least one action based on the CGM trend and at least one additional factor; Determine the strength, the process is configured to provide the at least one behavioral output to a user. And, Including, the system.

[0157] 2. The CGM device calculates a subsequent glucose monitoring result based on the body fluids after the period of time. and outputting the subsequent glucose measurement. and determining an updated CGM trend based on the source measurements. The system of embodiment 1.

[0158] 3. The at least one behavioral output indicates whether and how much insulin is needed. How much insulin is needed, whether glucose is needed, and how much glucose is needed Whether a course is required, whether food intake is required, and how much food intake is required whether or how much exercise is required The system according to embodiment 1, which corresponds to at least one behavior category selected from Tem.

[0159] 4. The at least one behavioral output category is based on the type of the one additional factor. 4. The system of embodiment 3, wherein the system is selected based on the

[0160] 5. The system of embodiment 1, wherein the at least one additional factor includes dietary information. .

[0161] 6. The system of embodiment 5, wherein the meal information includes an insulin-carbohydrate ratio.

[0162] 7. The system of embodiment 1, wherein the at least one additional factor includes athletic information. .

[0163] 8. The exercise information may include calorie information, heart rate information, exercise duration, exercise intensity, or body weight. 8. The system of embodiment 7, which may include at least one of: a physical load;

[0164] 9. The at least one additional factor includes information regarding previous insulin doses. 2. The system of embodiment 1.

[0165] 10. The system of embodiment 1, wherein the at least one additional factor comprises blood glucose levels. .

[0166] 11. In embodiments wherein the at least one additional factor includes information regarding a history of hypoglycemia. 1. The system described in 1.

[0167] 12. The occurrence of hypoglycemia within the threshold time triggers a behavioral output in the insulin recommended behavior category, The behavior in the insulin recommended behavior category without the hypoglycemic episode within a threshold time. 12. The system of embodiment 11, which makes the force more conservative compared to the force.

[0168] 13. The processor may further include a processor for combining one or more past behavioral outputs with a past CGM trend. and a corresponding change in the at least one behavioral output. 2. The system of claim 1, comprising a machine learning model.

[0169] 14. The at least one behavioral output is provided by the CGM monitor, electronic device, or app. 10. The method of claim 1, wherein the user is provided with at least one of the following applications: The system described.

[0170] 15. Computer-Implemented Method for Providing Glucose Trend-Based Behavioral Output There was, The CG analyzes body fluids over a period of time from a continuous glucose monitor (CGM) device. receiving a plurality of glucose measurements based on the M device; based on a change in the plurality of glucose measurements output by the CGM device; , determining CGM trends; determining at least one behavioral output based on the CGM trend; providing said at least one behavioral output to a user; and A method comprising:

[0171] 16. The at least one behavioral output indicates whether and how much insulin is needed. How much insulin is needed, how much glucose is needed, and how much is needed Whether glucose is needed, whether food intake is needed, and how much food intake whether exercise is needed or how much exercise is needed 16. The method of claim 15, wherein the at least one behavioral category is selected from the following: How to do it.

[0172] 17. The CGM device, after the period of time, calculates a subsequent glucose monitoring plan based on the body fluid. and further configured to output a subsequent glucose measurement based on the subsequent glucose measurement. 16. The method of embodiment 15, further comprising determining an updated CGM trend. Law.

[0173] 18. Receiving the updated CGM trend at the processor; determining at least one updated behavioral output based on the CGM trend; and providing the at least one updated behavioral output to a user. 18. The method of claim 17.

[0174] 19. A system for providing glucose trend-based behavioral output, comprising: Outputting multiple glucose readings based on analyzing body fluids over a period of time A continuous glucose monitoring (CGM) device configured as follows: The CGM device accesses the bodily fluid through the user's skin, and The continuous glucose monitor is configured to obtain glucose measurements on a sub-minute basis. Continuously Generating Membrane (CGM) devices and a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: The multiple data output by the CGM device and / or stored in the memory determining a CGM trend based on a change in the number of glucose measurements, CGM Trends is a CGM trend that maps the glucose measurements over a period of time. The CGM trend is determined using a CGM event or severity score. the determining is further based on at least one of: receiving at least one additional factor, said at least one additional factor comprising: , dietary information, exercise information, carbohydrate insulin ratio, information on previous insulin doses, said receiving including one or more of information regarding blood glucose levels and history of hypoglycemia; and, Based on the CGM trend and at least one additional factor, insulin is needed. whether or not blood is pumped, how much insulin is needed, and how much glucose is needed. whether, how much glucose is needed, whether food intake is needed, How much food intake is needed, whether exercise is needed, and how much exercise is needed. identifying at least one behavioral category selected from which an action is required; at least one CGM trend based on the CGM trend and the at least one additional factor; determining a behavioral output, wherein the at least one behavioral output is from one identified behavioral category, and the at least one behavioral output is one or more Based on the number of past behavioral outputs and corresponding changes in past CGM trends, The behavioral output is determined using a machine learning model that is configured to output a single behavioral output. said determining; a graphical user interface based on the at least one identified behavioral category; Generate a GUI interface, providing the at least one behavioral output to the user via the generated GUI; And, After providing the at least one behavioral output to the user, an updated CGM trend is receiving a trend update for the at least one row of the CGM; the receiving is based on a glucose measurement after providing a dynamic output to the user. and, updating the machine learning model based on the updated CGM trends; Including, the system.

[0175] 20. Providing the updated CGM trends as input to the insulin computer. and providing, by the insulin computer, the updated CGM trend. determining at least one updated behavioral output based on the at least one 20. The system of claim 19, further comprising: Tem.

[0176] Additional embodiments include the following:

[0177] 1. A system for managing a user's glucose status, comprising: Outputting multiple glucose readings based on analyzing body fluids over a period of time a continuous glucose monitoring (CGM) device configured to: a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: generating a CGM trace based on the plurality of glucose measurements over the period of time; And, determining a severity score for the CGM tracing, the severity score being a the height of the CGM tracing and the duration that the CGM tracing remains above the target value the identifying being based on time; identifying a starting state based on the severity score, wherein the starting state is said identifying being indicative of a glucose health status of the user; To arrive at an ideal state based on one or more user vectors and a starting state, generating an optimized route, the optimized route being a route that is optimized based on the one or more user vectors; said generating including one or more adjustments of torque; providing the optimized route to the user; Including, the system.

[0178] 2. Identifying a starting parameter, the starting parameter being compared to a target range. and a scaled value determined based on the starting point of the CGM trace. generating the optimized route further based on the identifying and the starting parameters; 2. The system of embodiment 1, further comprising:

[0179] 3. The starting parameters are very high parameters, high parameters, and range parameters. In an embodiment, the parameter is selected from one of a low parameter, a low parameter, and a very low parameter. 2. The system according to claim 2.

[0180] 4. Identifying an end parameter, the end parameter being compared to a target range. and a scaled value determined based on the end point of the CGM trace. generating the optimized route further based on the identifying and the end parameter; 2. The system of embodiment 1, further comprising:

[0181] 5. The severity score is calculated by multiplying the height of the CGM tracing by the distance the CGM tracing travels to the eye. 2. The system of claim 1, wherein the measured value is determined by multiplying the duration above a reference value by the measured value.

[0182] 6. The height of the CGM trace is given a first weight, and the CGM trace is The duration above the target value is assigned a second weight different from the first weight. 6. The system of embodiment 5.

[0183] 7. A lower severity score, when compared with a higher severity score, indicates a greater likelihood of the ideal condition. 2. The system of claim 1, wherein the state corresponds to a starting state closer to the state of the system.

[0184] 8. The user vector may be a combination of medication, food intake, exercise values, psychosocial parameters, or social 2. The system of claim 1, further comprising one or more of the following determinant parameters:

[0185] 9. The optimization path is selected from an optimization profile, the optimization profile being , wherein the severity score and one or more user characteristics are determined based on the severity score. The system described in

[0186] 10. Determining a time-in-range (TIR) ​​value of the CGM trace, wherein the T The IR value is calculated based on the amount of time the CGM trace is within a threshold band over a reference period. the determining step; determining a TIR state based on the TIR value; receiving a blood glucose variability (GV) value based on at least the CGM trace; The GV value is one of a standard deviation or a coefficient of variation (CV), and the CV is The standard deviation of the glucose measurements over a reference period is taken into account. said receiving indicating a variation; determining a GV state based on the GV value; determining the starting state further based on the TIR state and the GV state; 2. The system of embodiment 1, further comprising:

[0187] 11. A computer-implemented method for managing a user's glucose status, comprising: A continuous glucose monitoring (CGM) device collects data about the user's glucose levels over a period of time. receiving a radiometric measurement; generating a CGM trace based on the received glucose measurements; determining a severity score for the CGM tracing, the severity score being a the height of the CGM tracing and the duration that the CGM tracing remains above the target value the identifying being based on time; Identifying a CGM trace shape of the CGM trace, The trace shape is based on at least one of the height or width of the CGM trace. To determine identifying a starting state based on the severity score and the CGM trace shape; wherein the starting state is indicative of the user's glucose health state; to arrive at an ideal state based on one or more user vectors and the starting state. generating an optimized route, the optimized route being said generating including one or more adjustments of vectors; providing the optimized route to the user; A method comprising:

[0188] 12. The CGM trace shape includes wide, narrow, short, and tall shapes. 12. The method of embodiment 11, which is one of:

[0189] 13. The CGM trace shape is based on the CGM trace. 13. The method of claim 12, wherein the method is identified by a machine learning model configured to output How to post.

[0190] 14. The machine learning model generates a CGM trace shape based on past CGM trace shapes. 14. The method of embodiment 13, wherein the method can be configured to output a state.

[0191] 15. A system for managing a user's glucose status, comprising: Outputting multiple glucose readings based on analyzing body fluids over a period of time A continuous glucose monitoring (CGM) device configured as follows: The CGM device accesses the bodily fluid through the user's skin, and the CGM device continues to the continuous glucose monitor configured to obtain glucose measurements in increments of (CGM) devices and a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: generating a CGM trace that maps said glucose measurements over a period of time; And, determining a severity score for the CGM tracing, the severity score being a the height of the CGM tracing and the duration that the CGM tracing remains above the target value the identifying being based on time; Using a machine learning model to identify a CGM trace shape of said CGM trace. The CGM trace shape is at least one of the height or width of the CGM trace. said identifying being based on one of identifying a starting state based on the severity score and the CGM trace shape; wherein the starting state is indicative of the user's glucose health state; to arrive at an ideal state based on one or more user vectors and the starting state. generating an optimized route, the optimized route being said generating including one or more adjustments of vectors; Generate a graphical user interface (GUI) based on the optimized path. And, providing the at least one optimized path to a user via the generated GUI; And, receiving an updated CGM trace after providing the optimized route to the user; and the updated CGM trace provided the optimized route to the user. said receiving being based on a subsequent glucose measurement; updating the machine learning model based on the updated CGM trace; Including, the system.

[0192] 16. Identifying a starting parameter, wherein the starting parameter is relative to a target range. is a scaled value determined based on the starting point of the CGM trace, compared to generating the optimized route further based on the identifying and the starting parameters; 16. The system of embodiment 15, further comprising:

[0193] 17. The starting parameter may be a very high parameter, a high parameter, a range parameter, a low parameter, a low parameter, and a very low parameter, 17. The system according to claim 16.

[0194] 18. Identifying an end parameter, wherein the end parameter is relative to a target range. is a scaled value determined based on the end points of the CGM trace, compared to generating the optimized route further based on the identifying and the end parameter; 16. The system of embodiment 15, further comprising:

[0195] 19. The CGM trace shape includes wide, narrow, short, and tall shapes. The system described in embodiment 15, which is one of the above.

[0196] 20. The user vector may be a combination of medication, food intake, exercise values, psychosocial parameters, or social 16. The system of embodiment 15, comprising one or more of the following social determinant parameters:

Claims

1. 1. A system for providing a glucose trend-based behavioral output for treatment of a glucose-based condition, comprising: a continuous glucose monitoring (CGM) device configured to output a plurality of glucose measurements based on analyzing a bodily fluid over a period of time; a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: determining a CGM trend based on changes in the plurality of glucose measurements output by the CGM device and / or stored in the memory, wherein the CGM trend is determined using a CGM trace mapping the glucose measurements over a period of time, and the CGM trend is further based on at least one of a CGM event or a severity score; determining at least one behavioral output based on the CGM trend and at least one additional factor, wherein the at least one behavioral output includes insulin dosage information; providing said at least one behavioral output to a user; and generating an optimized path to reach an ideal state based on one or more user vectors and the CGM trend, the optimized path including one or more adjustments to the one or more user vectors, the one or more adjustments including medication adjustments, food intake adjustments, and exercise value adjustments; providing the optimized pathway to the user via a graphical user interface (GUI) for treating the glucose-based condition with adjustment of the one or more user vectors; receiving an updated CGM trace over a second time period by sensing an updated glucose concentration in a bodily fluid obtained over the second time period using the CGM device; generating an updated optimized path to reach the ideal state based on the updated CGM trace, the updated optimized path including insulin uptake information; providing the updated optimized route to the user via the GUI, the optimized route being provided as an output of a machine learning model, the output of the machine learning model including a change in a user vector based on machine learning inputs including the severity score of the CGM trace, the machine learning inputs further including user attributes, the user attributes including medical attributes, user preferences, metabolic attributes, and user demographics, the optimized route further based on a habit index score of the user, the habit index score of the user being determined based on a cohort of users having one or more user attributes in common with the user; the processor configured to: Including, the system.

2. 2. The system of claim 1, wherein the CGM device is further configured to output subsequent glucose measurements based on the bodily fluid after the period of time, and the processor is further configured to determine an updated CGM trend based on the subsequent glucose measurements.

3. 2. The system of claim 1, wherein the insulin intake information corresponds to at least one behavioral category selected from whether insulin is needed, how much insulin is needed, whether glucose is needed, how much glucose is needed, whether food intake is needed, how much food intake is needed, whether exercise is needed, or how much exercise is needed.

4. The system of claim 3 , wherein the at least one behavioral category is selected based on the type of the one additional factor.

5. The system of claim 1 , wherein the at least one additional factor includes dietary information.

6. The system of claim 5 , wherein the meal information includes an insulin-carbohydrate ratio.

7. The system of claim 1 , wherein the at least one additional factor includes athletic information.

8. The system of claim 7 , wherein the exercise information can include at least one of calorie information, heart rate information, exercise duration, exercise intensity, or physical strain.

9. The system of claim 1 , wherein the at least one additional factor includes information regarding previous insulin doses.

10. The system of claim 1 , wherein the at least one additional factor includes blood glucose level.

11. The system of claim 1 , wherein the at least one additional factor includes information regarding a history of hypoglycemia.

12. 12. The system of claim 11, wherein an occurrence of hypoglycemia within a threshold time causes a behavioral output of an insulin recommended behavioral category to be more conservative compared to the behavioral output of the insulin recommended behavioral category without the occurrence of hypoglycemia within the threshold time.

13. 10. The system of claim 1, wherein the processor includes a machine learning model configured to output the at least one behavioral output based on one or more past behavioral outputs and corresponding changes in past CGM trends.

14. The system of claim 1 , wherein the at least one behavioral output is provided to the user using at least one of a CGM monitor, an electronic device, or an application.

15. 1. A system for providing a glucose trend-based behavioral output for treatment of a glucose-based condition, comprising: a continuous glucose monitoring (CGM) device configured to output a plurality of glucose measurements based on analyzing a bodily fluid over a period of time, the CGM device accessing the bodily fluid through a user's skin, the CGM device configured to obtain glucose measurements every five minutes or less; a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: determining a CGM trend based on changes in the plurality of glucose measurements output by the CGM device and / or stored in the memory, wherein the CGM trend is determined using a CGM trace mapping the glucose measurements over a period of time, and the CGM trend is further based on at least one of a CGM event or a severity score; receiving at least one additional factor, the at least one additional factor including one or more of meal information, exercise information, carbohydrate-insulin ratio, information about previous insulin doses, blood glucose levels, and information about a history of hypoglycemia; determining at least one behavioral category selected from whether insulin is needed, how much insulin is needed, whether glucose is needed, how much glucose is needed, whether food intake is needed, how much food intake is needed, whether exercise is needed, and how much exercise is needed based on the CGM trend and at least one additional factor; determining at least one behavioral output based on the CGM trend and the at least one additional factor as an input to a machine learning model, the at least one behavioral output being from the at least one identified behavioral category, the at least one behavioral output being determined using a machine learning model configured to output the at least one behavioral output based on one or more past behavioral outputs and corresponding changes in past CGM trends, the at least one behavioral output including insulin intake information and at least one of dietary adjustments and exercise adjustments; generating a graphical user interface (GUI) based on the at least one identified behavioral category; providing the at least one behavioral output to the user via the generated GUI; receiving an updated CGM trend after providing the at least one behavioral output to the user, the updated CGM trend being based on glucose measurements obtained after providing the at least one behavioral output to the user; updating the machine learning model based on the updated CGM trends; generating an optimized path to an ideal state based on one or more user vectors and the CGM trend, the optimized path including one or more adjustments to the one or more user vectors, the one or more adjustments including medication adjustments, food intake adjustments, and exercise value adjustments; providing the optimized pathway to the user via a graphical user interface (GUI) for treating the glucose-based condition with adjustment of the one or more user vectors, wherein the optimized pathway is provided as an output of a machine learning model, the output of the machine learning model including changes in user vectors based on machine learning inputs including the severity score of the CGM trace, the machine learning inputs further including user attributes, the user attributes including medical attributes, user preferences, metabolic attributes, and user demographics, the optimized pathway further based on a habit index score of the user, the habit index score of the user determined based on a cohort of users having one or more user attributes in common with the user; receiving an updated CGM trace over a second time period by sensing an updated glucose concentration in a bodily fluid obtained over the second time period using the CGM device; generating an updated optimized path to reach the ideal state based on the updated CGM trace, the updated optimized path including insulin uptake information; providing the updated optimized route to the user via the GUI; the processor configured to: Including, the system.

16. providing the updated CGM trend as an input to an insulin computer; determining, by the insulin computer, at least one updated behavioral output based on the updated CGM trend; and providing the at least one updated behavioral output to a user.

17. The processor further comprises: receiving updated additional factors for the user; Identifying at least one updated behavior category; determining at least one updated behavioral output and the updated additional factor based on the updated CGM trend, wherein the at least one updated behavioral output is determined from the identified and updated at least one behavioral category; providing the at least one updated behavioral output to the user via the generated GUI. The system of claim 15 configured to:

18. 16. The system of claim 15, wherein the CGM device is configured to obtain bodily fluid using a skin-piercing component and to determine the user's glucose level by sensing a concentration of a target component in the bodily fluid.

19. 1. A system for providing a glucose trend-based behavioral output for treatment of a glucose-based condition, comprising: a continuous glucose monitoring (CGM) device configured to output a plurality of glucose measurements based on analyzing a bodily fluid over a period of time, the CGM device accessing the bodily fluid through a user's skin, the CGM device configured to obtain glucose measurements every five minutes or less; a memory configured to store the plurality of glucose measurements; 1. A processor, comprising: determining a CGM trend based on changes in the plurality of glucose measurements output by the CGM device and / or stored in the memory, wherein the CGM trend is determined using a CGM trace mapping the glucose measurements over a period of time, and the CGM trend is further based on at least one of a CGM event or a severity score; receiving at least one additional factor, the at least one additional factor including one or more of meal information, exercise information, carbohydrate-insulin ratio, information about previous insulin doses, blood glucose levels, and information about a history of hypoglycemia; determining at least one behavioral category selected from whether insulin is needed, how much insulin is needed, whether glucose is needed, how much glucose is needed, whether food intake is needed, how much food intake is needed, whether exercise is needed, and how much exercise is needed based on the CGM trend and at least one additional factor; determining at least one behavioral output based on the CGM trend and the at least one additional factor provided as an input to a machine learning model, the at least one behavioral output being from the at least one identified behavioral category, and the at least one behavioral output being determined using a machine learning model configured to output the at least one behavioral output based on one or more past behavioral outputs and corresponding changes in past CGM trends; generating a graphical user interface (GUI) based on the at least one identified behavioral category; providing the at least one behavioral output to the user via the generated GUI; receiving an updated CGM trend after providing the at least one behavioral output to the user, the updated CGM trend being based on glucose measurements after providing the at least one behavioral output to the user; updating the machine learning model based on the updated CGM trends; generating an updated optimized path to an ideal state based on the updated CGM trace, the optimized path including one or more adjustments to the one or more user vectors, the one or more adjustments including medication adjustments, food intake adjustments, and exercise value adjustments; providing the optimized pathway to the user via a graphical user interface (GUI) for treating the glucose-based condition by adjusting the one or more user vectors, wherein the optimized pathway is provided as an output of a machine learning model, the output of the machine learning model including changes in user vectors based on machine learning inputs including the severity score of the CGM trace, the machine learning inputs further including user attributes, the user attributes including medical attributes, user preferences, metabolic attributes, and user demographics, the optimized pathway further based on a habit index score of the user, the habit index score of the user determined based on a cohort of users having one or more user attributes in common with the user; receiving an updated CGM trace over a second time period by sensing an updated glucose concentration in a bodily fluid obtained over the second time period using the CGM device; generating an updated optimized path to reach the ideal state based on the updated CGM trace, the updated optimized path including insulin uptake information; providing the updated optimized route to the user via the GUI; the processor configured to: Including, the system.

Images