Systems and methods for analyzing, interpreting, and addressing continuous glucose monitoring data.

Abstract

To improve user health, we provide a method and system for optimizing a user's glucose state via a mobile application. [Solution] The health management method manages glucose status by using a continuous glucose monitoring (CGM) device to receive the user's blood glucose level 502, determine the time in range (TIR) ​​value 504, determine the TIR state 506, receive the glucose fluctuation (GV) value 508, determine the GV state 510, determine the starting state based on the TIR state and GV state 512, determine that the starting state corresponds to a non-ideal state 514, and generate an optimized path 516 to reach an ideal state based on one or more account vectors, such as addressing self-management behaviors including diet, activity, and medication use.

Description

Technical Field

[0001] (Related Applications) This application is a continuation of 1) U.S. Provisional Application No. 63 / 135,818, filed on January 11, 2021 , 2) U.S. Provisional Application No. 62 / 992,385, filed on March 20, 2020, and 3 ) U.S. Provisional Application No. 62 / 992,409, filed on March 20, 2020, and claims the benefit of priority, each of which is hereby incorporated by reference in its entirety for use herein.

[0002] This disclosure generally relates to obtaining and processing data to generate an optimized route and improve a user's health, and in some embodiments, particularly to optimizing a user's glucose status via a mobile application.

Background Art

[0003] Increased healthcare costs are limiting users' access to appropriate care. At the same time, healthcare companies are increasing the workload of providers and limiting the interaction between physicians and users. Diabetes treatment often relies on sporadic measurements (e.g., glucose measurements) that do not provide sufficient data to effectively offer treatment options. Such measurements are often used alone, such that any medical, dietary, and / or lifestyle changes recommended as a result of a given measurement are limited considering the sparse data received via sporadic measurements.

[0004] This disclosure is directed to addressing one or more of the above problems. Provided herein The introductory section is intended to provide a general context for this disclosure. Unless otherwise indicated, the materials described in this section are not prior art of the claims of this application. By including it in this section, it is clear that it is prior art or an indication of prior art. This does not constitute an endorsement. [Overview of the Initiative]

[0005] This disclosure pertains to a computer implementation method for managing a user's glucose state. , receiving the user's blood glucose levels using a continuous glucose monitoring (CGM) device. This involves determining the user's time in range (TIR) ​​value for blood glucose levels, and the TIR value is determined by the user The determination is based on the amount of time over a reference period when the blood glucose level is within the threshold range. Furthermore, the TIR state is determined based on the TIR value, and at least the user's blood glucose level is used. This involves receiving the blood glucose fluctuation (GV) value, where the GV value is the standard deviation or coefficient of variation (C) V) is one of the options, and CV takes into account the standard deviation of blood glucose levels over the reference period. The system shows fluctuations in blood glucose levels, and determines the GV state based on the GV value. The start state is determined based on the TIR state and GV state, and if the start state is not ideal, Determining whether it corresponds to a state and ideally based on one or more account vectors To reach a target state, an optimized path is generated, and the optimized path is one or The process of generating and optimizing paths involves adjusting one or more account vectors. This includes providing it to the user.

[0006] The threshold range may be approximately between 70 mg / dL and 180 mg / dL, and the reference period This can be 24 hours. The CV value is the standard deviation of blood glucose levels compared to the mean of blood glucose levels during the reference period. The TIR state may be determined by dividing by a certain factor. It may also be a binary state selected from one of the R states. A good TIR state is TI It may also accommodate TIR values ​​greater than the R threshold. The GV state is either a good GV state or a bad GV state. It may also be a binary state selected from one of the GV states. A good GV state is GV It may also accommodate GV values ​​greater than the threshold. Account vectors include blood glucose levels, medications, and food intake. One or more of the following: kinetic values, psychosocial parameters, or social determinant parameters It may include numbers. The account vector is one or that is classified based on the severity score. It may include blood glucose levels based on multiple CGM events. The optimization pathway is further based on user attributes. The user attributes include social attributes, medical attributes, user preferences, metabolic attributes, or One or more of the demographics of the system are selected. The optimal path is one or more states This may include an increase in habits that improve one's condition, and / or a decrease in one or more habits that worsen one's condition.

[0007] This disclosure pertains to a computer implementation method for managing a user's glucose state. This generates multiple optimization profiles to reach an ideal state from a non-ideal state. The ideal state is a good time-in-time (TIR) ​​state and good blood glucose fluctuation (GV) state. ) corresponds to the state, and the non-ideal state is one of the poor TIR state or poor GV state, less than Both include generating and based on the user's blood glucose TIR value, the current TIR state This involves determining the state, and the TIR value is determined when the user's blood glucose level stays within the threshold range over a reference period. based on the amount of time therein, and the current TIR state is a good TIR state or a bad TI R state, one of which is determined, and the current GV state associated with the user's blood glucose value is determined based on the current GV value, where the GV value represents the standard deviation (S D) or coefficient of variation (CV) of the blood glucose value, and the CV takes into account the standard deviation of the blood glucose value over a reference period and indicates the variation of the user's blood glucose value, determining, receiving one or more account vectors of the user, and based on the one or more account vectors, TIR state, and CV state, identifying one of the optimization profiles, and identifying an optimization path based on the identified optimization profile, where the optimization path includes one or more adjustments to one or more of the account vectors and providing the optimization path to the user.

[0008] The plurality of optimization profiles may be generated by a machine learning model configured to receive an account vector as input and output one or more adjustments to the received account vector. The plurality of optimization profiles may be further generated by associating one or more adjustments to the received account vector with one or more TIR states or GV states. Each of the plurality of optimization profiles may correspond to a potential TIR state, 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 be further identified based on the one or more user attributes. The CV value may be determined by dividing the standard deviation of the blood glucose value by the average value of the blood glucose value over the reference period. ​​​​​

[0009] The present disclosure also relates to a system for managing a user's blood glucose level, the system comprising a memory storing processor-readable instructions and a processor configured to access the memory and execute the processor-readable instructions. When executed by the processor, the processor-readable instructions cause the system to electronically receive the user's blood glucose level using a continuous glucose monitoring (CGM) device configured to obtain glucose values using a component that penetrates the user's skin, determine a time in range (TIR) value for the user's blood glucose level, where 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 threshold band being between approximately 70 mg / dL and 180 mg / dL, and the reference period being 24 hours, determine a TIR state based on the TIR value, where the TIR state is selected from a good TIR state or a poor TIR state, receive a glucose variability (GV) value based at least on the user's blood glucose level, where the GV value is one of a standard deviation or a coefficient of variation (CV), and the CV indicates the variability of the user's blood glucose level considering the standard deviation of the blood glucose levels over the reference period, determine a GV state based on the GV value, where the GV state is one of a good GV state or a poor GV state, determine a start state based on the TIR state and the GV state, determine that the start state corresponds to a non-ideal state, detect a CGM event based on the user's blood glucose level, and characterize one or more of a multi-parameter CGM classification or severity and a CGM event trace shape. ロセッサ可読命令を格納するメモリと、メモリにアクセスし、プロセッサ可読命令を実行 するように構成されているプロセッサとを含む。プロセッサ可読命令は、プロセッサによ って実行されると、ユーザの皮膚に貫通するコンポーネントを使用してグルコース値を取 得するように構成されている持続血糖モニタリング(CGM)デバイスを使用して、ユー ザの血糖値を電子的に受信することと、ユーザの血糖値の範囲内時間(TIR)値を判定 することであって、TIR値は、ユーザの血糖値が基準期間にわたって閾値帯内にある時 間の量に基づいており、閾値帯は、およそ70mg / dLと180mg / dLとの間であ り、基準期間は、24時間である、判定することと、TIR値に基づいてTIR状態を判 定することであって、TIR状態は、良好なTIR状態又は不良なTIR状態から選択さ れる、判定することと、少なくともユーザの血糖値に基づく血糖変動(GV)値を受信す ることであって、GV値は、標準偏差又は変動係数(CV)のうちの1つであり、CVは 、基準期間にわたる血糖値の標準偏差を考慮して、ユーザの血糖値の変動を示す、受信す<0​​​​​​​​​​The multi-parameter CGM classification is used to characterize CGM events. Characterizing the elephant, including its initial blood glucose level, severity, and blood glucose level at the end of the CGM event. Based on one or more account vectors and characterizing CGM events, To reach the desired state, an optimized path is generated, and the optimized path is one or This involves generating and optimizing paths, which include adjusting one or more account vectors. This includes providing the user with an optimized path. Providing the user with an optimized path includes the Based on this, we will implement methods that include providing context-based instructions to the user. The processor may be configured as follows. [Brief explanation of the drawing]

[0010] The accompanying drawings incorporated herein and constituting a part thereof illustrate embodiments of this disclosure. This is used to explain the principles of this disclosure, along with a detailed explanation. [Figure 1] Figure 1 is a schematic diagram of a health management system according to one embodiment of the present disclosure. [Figure 2] Figure 2 is a schematic diagram of a portion of the health management system shown in Figure 1. [Figure 3A] Figure 3A is a schematic diagram of another part of the health management system shown in Figure 1. [Figure 3B] Figure 3B is a schematic diagram illustrating the training of an exemplary machine learning model according to one embodiment of the present disclosure. [Figure 4A] Figure 4A is a continuous glucose monitoring (CGM) chart according to one embodiment of the present disclosure. [Figure 4B] Figure 4B shows a continuous glucose monitoring (CGM) report according to one embodiment of the present disclosure. [Figure 5A] Figure 5A is a flowchart of a health management method according to one embodiment of the present disclosure. [Figure 5B]Figure 5B is a flowchart illustrating an exemplary health management method according to another embodiment of the present disclosure. [Figure 6A] Figure 6A is a graph of the patient's condition according to another embodiment of the present disclosure. [Figure 6B] Figure 6B shows a graph of patient condition changes over time, correlated with patient condition changes, according to another embodiment of the present disclosure. [Figure 6C] Figure 6C shows three graphs of patient status over time, according to another embodiment of the present disclosure. [Figure 6D] Figure 6D shows graphs of the standard deviation and coefficient of variation according to another embodiment of the present disclosure. [Figure 6E] Figure 6E shows the status distribution of multiple patients according to another embodiment of the present disclosure. [Figure 6F] Figure 6F shows the changes in the status distribution of multiple patients according to another embodiment of the present disclosure. [Figure 6G] Figure 6G shows the change in glucose value over time and the corresponding first derivative graph in another embodiment of the present disclosure. [Figure 7A] Figure 7A shows the average glucose, a range chart, and the fluctuations of the glucose and range chart according to another embodiment of the present disclosure. [Figure 7B] Figure 7B shows a continuous glucose monitoring activation time and variability chart according to another embodiment of the present disclosure. [Figure 8A] Figure 8A shows a continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8B] Figure 8B shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8C] Figure 8C shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 8D] Figure 8D shows another continuous glucose monitoring chart according to another embodiment of the present disclosure. [Figure 9] Figure 9 is another flowchart of a health management method according to one embodiment of the present disclosure. [Figure 10A]Figure 10A shows a visualization of a continuous glucose monitoring event according to one embodiment of the present disclosure. [Figure 10B] Figure 10B shows a visualization of meal monitoring events according to one embodiment of the present disclosure. [Figure 11] Figure 11 shows a visualization of the severity count according to one embodiment of the present disclosure. [Figure 12] Figure 12 shows an automated warning generation chart according to one embodiment of the present disclosure. [Figure 13A] Figure 13A is a screenshot of an exemplary message from one embodiment of the present disclosure. [Figure 13B] Figure 13B is another screenshot of an exemplary message from one embodiment of the present disclosure. [Figure 13C] Figure 13C is another screenshot of an exemplary message according to an embodiment of the present disclosure. [Figure 13D] Figure 13D shows a glucose computer according to one embodiment of the present disclosure. [Figure 14] Figure 14 is a simplified functional block diagram of a computer that may be configured, for example, as a host server to function as a decision server for a healthcare provider, according to an embodiment of the present disclosure.

[0011] The appendix is ​​provided with this specification and includes a description using examples of this disclosure, including experimental results. . [Modes for carrying out the invention]

[0012] Next, embodiments of the present disclosure, as shown in the attached drawings, will be referenced in detail. Reference numbers are used throughout the drawing to refer to the same or similar parts. .

[0013] In the discussion below, relative terms such as "about," "effectively," and "approximately" are not used. This is used to indicate a possible variation of ±10% in the given value. The above explanation is essentially just an example, and does not apply to the embodiments described herein, or to the application of such embodiments. Please note that this is not intended to restrict use. This specification is an example. Any implementation described in this document is interpreted as being preferable or advantageous to other implementations. It should not be. Rather, as mentioned above, the term “exemplary” means “ideal.” It is used not as "na" but to mean "example" or "one example." The term "comprise" is used in a different sense. "Include", "Have", "Together", and any variations thereof are nonexclusive. It is used synonymously to indicate or describe other inclusions. Processes, methods, articles, or apparatus that use such terminology, their steps, structure, or It does not include only the elements, but does not explicitly state, or such processes or methods. This specification may include other steps, structures, or elements specific to the article or apparatus. The terms "first," "second," etc., used in this context do not indicate any arbitrary order, quantity, or importance. Furthermore, this specification is used to distinguish one element from another. The singular forms "a" and "an" in the terminology do not indicate a limitation of quantity, but rather refer to a reference. This indicates the existence of at least one of the items being considered.

[0014] (Healthcare and computing environments) Figure 1 is a block diagram of a health management system 100 according to one embodiment of the present disclosure. Yes. Mobile devices, computers, medical devices, or electronic networks such as the internet. Any other electronic device, such as any other electronic device configured to access the twerk 32, A user 8 (e.g., a patient, consumer, etc.) with device 19 can access mobile health (mH ealth) can communicate with application 1, or access it in other ways. It is possible. In some embodiments, network 32 is a mobile phone network. Wi-Fi, LAN, WAN, Bluetooth (registered trademark), Near Field Wireless Communication (N) Even if wireless or wired links are included, such as FC, or other appropriate forms of network communication. Good. Multiple electronic devices 19 are configured to access the electronic network 32. It is also possible that User 8 may use (for example, a mobile phone, tablet, and laptop computer) A single A linked to multiple electronic devices 19 (via one or more of the computers) The count may be used to access mHealth application 1. Electronic device 19 Also, mobile health devices, desktop computers or workstations, Laptop computers, mobile handsets, personal digital assistants ( PDAs, mobile phones, network equipment, cameras, smartphones, smartwatches, Extended General Packet Radio Service (EGPRS) mobile phones, media players, navigation systems Communication devices, game consoles, set-top boxes, biosensing devices with communication capabilities A smart TV, or any combination thereof, or at least one processor Socket, local memory, display (e.g., monitor or touchscreen display) (i) Having one or more user input devices and a network communication interface This may include, but is not limited to, other types of computing devices. The electronic device 19 includes a display monitor, keyboard, touchpad, accelerometer, and Iroscope, mouse, touchscreen, camera, projector, touch panel, point Turning devices, scrolling devices, buttons, switches, motion sensors, Input / output devices such as audio sensors, pressure sensors, thermal sensors, and / or microphones It may include any type or combination of the elements. The electronic device 19 also sends and receives information. To trust, (for example, Wi-Fi, radio frequency (RF), infrared (IR), Blue (via any suitable wired or wireless means) via any suitable wired or wireless means (near-field communication, or any other suitable means) They may communicate with each other by means of a line.

[0015] The mHealth application 1 sends and receives information from other entities or It can communicate with the network. In some embodiments, the mHealth application 1 For example, exercise tracking (e.g., step tracking) applications and / or This includes one or more applications associated with user 8, such as other health-related applications. It may communicate with other applications. mHealth application 1 may communicate with other applications You can import data from the system and create a treatment plan for user 8. Sometimes it can be analyzed and used. For example, mHealth application 1 is another application You can also import activity tracking data from the application and use that data. User 8's exercise and glucose collected before using mHealth application 1. You may also identify patterns between values. mHealth application 1 also has other From a mobile health application, for example, blood pressure, BMI, A1C, exercise type, exercise Exercise time, distance, calories burned, total steps, exercise days, start and end times of exercise, and sleep You may import any other suitable data. (mHealth application) N1 also includes, for example, other mobile healthcare devices with social or interactive features. Export data to other mobile applications, including the S application. This is also acceptable. Healthcare providers such as doctors 7 may prescribe this application. However, The mHealth application 1 does not require a prescription, for example, a computer Accessible without prescription from a digital distribution platform for user software. It could also be a commercially available consumer application. Form 1 may be adapted to a specific user 8, and to a pharmacy 9 or other approved entity. It may also be activated directly by user 8 by visiting the site. For example, - The 8 will enter an access code to authorize access to mHealth application 1. You may receive it from a pharmacy. User 8 can use the mHealth support system 25 and / or The application trainer 24 uses the mHealth application 1. You may receive training on how to do this. mHealth Application 1 is a machine It may include various forms of programming 28, such as machine learning programming algorithms 26. The user's treatment plan may be dispensed by pharmacy 9, (e.g., drugs, devices, and This may include prescriptions for (and / or treatment). Pharmacy 9 may include User 8's healthcare treatment plan. After compliance-based approval, it may be possible to re-prescribe the prescribed product / treatment. Approval is, For example, communication from application 1 via network 32 and various servers 29 By trust, it may be received by pharmacy 9. The use of drugs or other medical products / treatments is also It is transmitted to the manufacturer 37 via network 32 and used by user 8. The manufacturer 37 may be notified of the quantity of the medical product or treatment. This information may be used by the manufacturer 37 to inform the medical This can help assess the demand for a product or treatment and plan its supply. Healthcare provider 7 also You may receive a report based on user information received by Application 1, The user's treatment plan may be updated based on the information of the user's electronic medical record (EMR). 14 is also received by mHealth application 1, application Based on user information, which may include electronically transmitted user 8 feedback regarding The information can be automatically updated via network 32. The healthcare provider 7, for example, a doctor, Any appropriate medical care, including specialists, nurses, educators, social workers, MAs, PAs, etc. They can also be providers.

[0016] Figure 2 is a schematic diagram of an additional embodiment of System 100. For example, System 1 00 is stored in the decision model database 270 via the network 32. You may access the decision model. The retrieved decision model is available on mobile device 2 15. Tablet devices 220, computers 225 (e.g., laptops or desktops) (Cut top), kiosk 230 (e.g., kiosks with medical and / or prescription information, drug) (at a station, clinic, or hospital) and / or any connected to network 32 Used for display and / or processing by one or more electronic devices 19, such as a device. It can be used.

[0017] In the embodiment shown in Figure 2, a mobile device 215, a tablet 220, and a computer are used. Each of the users 225 is connected to any server 29 and / or one or more GPS satellites 255. In between, location information, such as GPS data, is acquired and reported via network 32. For this purpose, it may be equipped with, or may include, a GPS receiver.

[0018] Mobile device 215, tablet device 220, computer 225, and / or Each of the electronic devices 19, including the kiosk 230, connects to a server via the network 32. It may be configured to send and receive data (e.g., clinical information) to and from the 29 systems. Each of the devices 19 receives information such as clinical data from the server 29 via the network 32. Server 29 can receive the clinical data server 240, algorithm server 245, and - May include the interface (UI) server 250 and / or any other suitable server. The electronic device 19 communicates data with the UI server 250 via the network 32. It may include a user interface. Each server has a decision model database 270. It is possible to access and search for decision models. Each server has memory, processor, and / Or it may include a database. For example, the clinical data server 240 may include a database of providers. A process configured to retrieve clinical data from the patient's and / or electronic medical records. It may have a database containing various algorithms. The algorithm server 245 is a database containing various algorithms. It may have a processor configured to process clinical data. UI Server 2 50 is configured to receive and process user input such as clinical decision-making preferences. Satellite 255 is configured to send and receive information between server 29 and device 19. It is possible.

[0019] The clinical data server 240 accesses the UI server 2 via network 32, or indirectly. Through 50, clinical data such as user data can be received from the electronic device 19. The clinical data server 240 can store information in memory such as computer-readable memory. ru.

[0020] The clinical data server 240 also includes the algorithm server 245 and / or external servers, etc. It may communicate with one or more other servers. Server 29 is in accordance with the provider's preferences. This may include data on the user and / or the user's health history. Furthermore, clinical data service The 240 may contain data from other users. The algorithm server 245 is a machine This may include machine learning and / or other appropriate algorithms. Algorithm server 245 is Furthermore, it may communicate with other external servers and may be updated as desired. For example, A Lugorhythm Server 245 features new algorithms, more powerful programming, and / or This may be updated with more data. Clinical data server 240 and / or algorithm The Zoom Server 245 processes information and stores the data in the Model Database 270 for processing. It may be transmitted. In one embodiment, the algorithm server 245 (one or more) is simple Obtain the pattern definition in a specific format and create a model, such as a Markov model or Gaussian model. Bayesian regression, PCA (principal component analysis), multivariate linear or nonlinear regression, and / or linear discriminant analysis. Classification functions such as nonlinear discriminant functions, composite discriminant functions, and random forest algorithms. By using Dell, we can predict several time steps in the future and base our predictions on them. Then optimize the results, detect transitions between patterns, obtain abstract data, and extract information. It infers higher-level knowledge, combines high-level and low-level information, and the user 8. Understand clinical behavior and multi-time (e.g., different time scales) data and related information. Inferences are made from the information, using variable-order Markov models and / or gradient-based and curve-based models. Clustering algorithms such as smoothing algorithms and k-means clustering. By using it, noise can be reduced over time.

[0021] Including 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, application server, and This includes, but is not limited to, Roxy servers, network servers, or server farms. This may represent any various types of servers. Each server in the server 29 system For example, via network 32, device 19 or any other computing This includes, but is not limited to, other computing devices (not shown). It can also be implemented using any general-purpose computer that can provide data to the chair. Yes. Such a general-purpose computer has a processor and memory for executing and storing instructions. This may include, but is not limited to, server devices having memory, floppy disks. - Magnetic storage including disks, hard disks, or magnetic tapes, solid-state storage Semiconductor storage such as hard disks (SSDs) or flash memory, optical disks Any physical storage medium such as disk storage or magneto-optical disk storage Includes random access memory (RAM) or read-only memory (ROM) of the following types. That's fine. The software is one or more applications and operating systems. It may include a system. Hardware includes a processor, memory, and a graphical UI. This may include, but is not limited to, displays. Each server may also be, for example, clustered. Configured to function together within a computing environment or server farm. It has multiple processors and multiple shared or separate memory components. That's fine.

[0022] Figure 3A shows additional details of the system 100, including the electronic device 19 and the server 29. This is another expression of the part. Electronic device 19 and server 29 each have processor 301-1 It may include one or more processors, such as processor 301-1 and 304-1. 304-1 are, respectively, a central processing unit, a microprocessor, a general-purpose processor, and a specialized processor. This may be a processor or any device that executes instructions. Electronic device 19 and - The 29 also stores one or more software modules, memory 301-2 and may include one or more memories such as 304-2. 2 is any hard drive, CD, DVD, flash memory, RAM, ROM, etc. It may be implemented using a computer-readable storage medium. Memory 301-2 is a processor Module 301-3 can be stored, which can be executed by 301-1. Similarly, Mori 304-2 can be executed by processor 304-1, module 304-3 It can store.

[0023] The electronic device 19 may further include one or more UIs. The UI is a planning or intermediary. This may enable one or more interfaces for presenting information such as input to user 8. The UI may be web-based, such as a web page, or a standalone application. It may also be a reference. The UI also includes data entry and user feedback. It may be configured to accept information about user 8. The information may be entered manually, or it may be entered automatically. In one embodiment, user 8 (Or the user's caregiver) should know when the medication was taken, or what kind of food and drink user 8 consumed. Information such as whether a substance was ingested may also be entered. The electronic device 19 is also a testing device ( It may include an interface for receiving information from (not shown) or testing equipment. The devices may include, for example, blood glucose meters, heart rate monitors, scales, blood pressure cuffs, etc. Device 19 also includes a camera and microphone for collecting feedback from user 8. It may include one or more sensors (not shown), such as a lophone or accelerometer. In this embodiment, the device reads the user's blood glucose level and automatically reports it. It may include a value measuring instrument.

[0024] The electronic device 19 may also include a presentation layer. The layers include web browsers, applications, and messaging interfaces (e.g., electronics). This may also be a text message, instant message, SMS, etc. Electronic device 19 is Through the presentation layer, notifications, warnings, reading material, references, guides, and reminders Alternatively, the proposal may be presented to User 8. For example, the presentation layer may present to User 8. Related articles, reminders to purchase medication, and tutorials on topics. Real-world examples (e.g., tutorials on carbohydrates), testimonies from others with similar symptoms, And / or, one or more goals (e.g., carbohydrate counting goals) may be presented. The presentation layer also includes tutorials (e.g., user guides or explanatory videos). Information such as the above may be presented, and / or the relationship between the healthcare provider and the user 8, for example, the patient. Communication between them may be enabled. Communication between the healthcare provider and the user 8, for example, the patient, may be enabled by electricity. via child messages (e.g., email or SMS), voice, or real-time video It may be done. One or more of these items may be included in the treatment plan or further as described below. The presentation layer may also be based on the user. It may be used to receive feedback.

[0025] System 100 also includes one or more databases, such as database 302. To obtain. Database 302 is related to database technology or object-oriented database. It may be implemented using any database technology known to those skilled in the art, such as [specific technology name]. Database 302 may store data 302-1. Data 302-1 is used for inference. This may include a knowledge base, statistical models, and / or user information for conducting the task. Data 30 2-1, or a portion thereof, alternatively or simultaneously, to the server 29 or electronic device 19 It may be stored.

[0026] System 100 can, for example, address the user's healthcare needs and maintain the user's finances. A wide range of applications, including monitoring and tracking the user's nutrition and / or sleep. It can be used for. In some implementations of system 100, any received data , enhance data security against unauthorized access, HIPAA privacy, and / Or in an encrypted format 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 described in system 100, the server or The server system may include one or more databases. In one embodiment, A database can be used to store any type of data. It may be a store or recording medium. For example, database 302 contains each prescription of the treatment plan. Includes information related to the user's treatment plan, including the timing and dosage associated with the medication. The data received by or processed by server 29 is stored. It may also be possible. Database 302 also contains the literacy codes of user 8 associated with each of the multiple prescription drugs. Information related to user 8, including bells, may be stored.

[0028] One or more components of the disclosed subject matter, as further disclosed herein This may be implemented using a machine learning model. Figure 3B shows one disclosed herein. Alternatively, an exemplary training module 310 for training multiple machine learning models is shown. The module is used to train each of the machine learning models disclosed herein. Also, a single training module 310 can train two or more machine learning models. It should be understood that it may be used for that purpose.

[0029] As shown in Figure 3B, the training data 312 is related to the machine learning model being trained. The stage input 314 and the known result 318 may include one or more of these. Input 314 is a healthcare provider 7, one or more servers 29, an electronic device 19, and an EMR. 14. Output from the step (for example, flowchart 500 in Figure 5A or flowchart 9) One or more outputs from steps of Chart 900, time in range (TIR) ​​values, hyperglycemia Total glucose level (TAR), total blood glucose level (TBR), severity score, continuous glucose monitoring (CG) M) may be from any applicable source including classifications, etc. Known result 318 is , included for machine learning models generated based on supervised or semi-supervised training That's also good. Unsupervised machine learning models may not be trained using known results 318. The known output 318 is similar to the stage input 314 which does not have a corresponding known output. This may include known or desired outputs for future inputs in the same category.

[0030] The training data 312 and the training algorithm 320 are used to train the training data 312. Applied to M320 and provided to training component 330 capable of generating machine learning models This is also good. According to one implementation, the training component 330 is the corresponding machine learning model Comparison result 316 is provided to compare with the previous output, and the previous result is applied to the machine learning model. You may retrain it. Comparison result 316 shows that in order to update the corresponding machine learning model, It may be used by the training component 330. The training algorithm 320 is - Unstructured neural networks (DNN), Convolutional neural networks (CNN), Fully Convolutional Networks (FCNs) and Recurrent Neural Networks (RCNs) ) and other deep learning networks, Bayesian networks and graphical models Probability models such as , and / or discriminative models such as decision trees and maximum margin methods. You may also use machine learning networks and / or models, including but not limited to those mentioned above. stomach.

[0031] (Health condition) Diabetes mellitus (commonly known as diabetes) is a condition in which the patient's body does not produce enough or any insulin. Insulin cannot be produced, or the insulin that is produced cannot be used (insulin Phosphorus resistance is a chronic, persistent metabolic disorder (or condition) in which the glucose in the patient's blood is It causes an increase in values. The three identifiable types of diabetes are prediabetes, type 1 diabetes, This includes type 2 diabetes. Pre-diabetes are people with high blood sugar levels, but not necessarily type 2 diabetes. The blood sugar level is not high. Type 2 diabetes is a chronic condition that affects how the body processes blood sugar. This is a typical condition. Finally, type 1 diabetes is a condition in which the pancreas produces little to no insulin. It is a chronic condition.

[0032] Generally, diabetes is diagnosed in several ways. Diagnosing diabetes involves... It may be necessary to repeat the tests over several days to confirm a positive diagnosis of the type. Diabetes Several health parameters used by a doctor or other appropriate healthcare provider when confirming a diagnosis of an illness. The meter measures glycated hemoglobin (A1C) levels, fasting blood glucose (FPG) levels, and oral glucose tolerance. This includes load testing and / or random blood glucose testing. Generally, healthcare providers diagnose diabetes. To support this, we are interested in the patient's A1C levels. Glycated hemoglobin is measured over a 3-month period. A form of hemoglobin that is primarily measured to determine plasma glucose concentration, and by physicians and It may be used by and / or other appropriate healthcare providers. Health parameters include weight, age, Nutritional intake, exercise activity, cholesterol levels, triglyceride levels, obesity, tobacco use, and family history include.

[0033] Once the diagnosis of the type of diabetes is confirmed by a doctor or other appropriate healthcare provider, the patient You may receive treatment to manage diabetes. Diabetes can be detected by a doctor or other healthcare provider. Patients being tracked or monitored may receive dietary, exercise, oral medication, and / or insulin therapy. It can be treated by combining it with controlling blood sugar through other means. Complications Regular checkups are also necessary for some patients. Depending on the period, the mHealth application 1 is used to manage the patient's condition. A treatment plan may be proposed. Oral medications typically reduce glucose production by the liver. This includes tablets taken orally to make muscles more sensitive to insulin. For example, if the diabetes is more severe, additional treatments, including injections, may be used to treat the patient's diabetes. Additional medication may be necessary. Basal insulin injections are used to treat background insulin. Also known as sulin, it is used by healthcare providers to keep fasting blood sugar levels constant. It can be used. When fasting, the patient's body releases glucose into the bloodstream to supply energy to the cells. It releases blood sugar steadily. Therefore, it controls blood sugar levels and allows cells to use energy. In order to enable glucose uptake, basal insulin injections are necessary. Basal insulin is administered once or twice a day, depending on the type of insulin. Basal insulin acts for a relatively long period, so it is administered as a supplement to long-acting insulin or medium-acting insulin. It is thought to be inter-acting insulin. In contrast, bolus (additional) insulin is rapidly It can be used to control blood glucose levels after a meal. For example, bolus insulin can be used to control blood glucose levels after a meal. It may be administered especially with meals to roll it out. In some cases, a physician or healthcare provider may When creating a treatment plan to manage a patient's diabetes, a doctor might, for example, consider the patient's condition throughout the day. Basal bolus medication, which involves administering several injections. Prepare a regimen. Basal bolus regimens may include injections with each meal. This attempts to closely mimic how the body of a non-diabetic person releases insulin. Yes, basal bolus regimens are applicable to people with type 1 and type 2 diabetes. It may be possible. In addition to basal bolus regimens requiring insulin injections, The treatment plan can be reinforced by the use of prescribed oral medications. Patient adherence to the treatment plan is important for the patient This can be important in managing the disease state. For example, if a patient has been diagnosed with diabetes for more than 6 months... If this occurs, for example, in order to obtain healthy or desirable blood glucose levels, very specific treatment methods... The patient should adhere to the following. Ultimately, the treatment of those drug types Weekly patterns can be important in managing diabetes. Option 1 may recommend a treatment plan to help patients manage their diabetes.

[0034] (Example method) Diabetes is a condition in which a patient is unable to keep their glucose levels within a normal or recommended target range. It is a chronic disease that causes such fluctuations in blood glucose levels (i.e., normal or recommended). Blood glucose levels outside the target range can lead to serious health complications. Sporadic blood glucose monitoring (B In GM, it is difficult to gain meaningful insights, and the measurements are intermittent, only a few times a week. This includes patterns, and any underlying causes of these patterns (e.g., B based on diet type). Understanding how to determine a rise in GM (Glycemic Genesis) may not be fundamental.

[0035] Continuous glucose monitoring (CGM) monitors regular glucose levels (for example, user 8's blood glucose). The data is automatically collected through wearable sensors (e.g., subcutaneous sensors) that provide the value. This offers the potential for high-density data (e.g., data based on collection frequencies of 5 minutes or less). CGM allows users 8 or other entities (e.g., healthcare provider 7) to monitor the system at any time of day. To make the user's blood glucose level more recognizable in real time, sequential glucose data (for example, 5 Readouts (at intervals of less than a minute) or semi-continuous (for example, longer than every 5 minutes) are available to user 8 or other By providing this data to entities, diabetes care can be improved. This could enable healthcare provider 7 to more optimally adjust the treatment plan for user 8.

[0036] The CGM monitor is a continuous spectroscopy monitor that includes any sensor configuration that provides an output signal indicating the concentration of the analyte. It could be an analyte sensor system. The CGM monitor is based on body fluids (e.g., interstitial fluid). For example, the concentration of the analyte may be sensed to determine the glucose level. Body fluids are... It may be accessed through the skin. For example, raw data stream, filtering. Processed data, smoothed data, and / or otherwise transformed sensor data Output signals, which may take the form of sensor data, are transmitted via wired or wireless connection to the CGM (Continuously Variable Mechanism). It may be connected to a NITA and transmitted to a receiver that may be local or remote from the sensor. Depending on the implementation, the CGM monitor may include a transcutaneous glucose sensor, a subcutaneous glucose sensor, Sustained-release refillable subcutaneous glucose sensors, sustained-release intravascular glucose sensors, etc. It may include the following. The CGM monitor is inserted onto user 8's abdomen and through user 8's skin. A compact medical system having one or more sensors, including a small cannula for insertion. It may also be a sticky patch. The adhesive patch may hold the monitor in place. The sensor is connected Interstitial fluid glucose levels may be detected continuously or semi-continuously.

[0037] The transmitter wirelessly transmits glucose measurements from the CGM monitor to the monitoring device. To enable this, it may be connected to a sensor. The monitoring device is CGM. It may be a dedicated monitoring device for the monitor, or it may be 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. A dedicated monitoring device or electronic device 19 may also be used. Other software, via monitoring devices, collects glucose measurements and related data. It may be used to facilitate the analysis and / or display of the data. The monitoring device is Used to analyze and / or view data associated with glucose measurements. Alternatively or additionally, a CGM monitor can measure glucose levels and / or related data. A display for viewing data may be included. CGM monitor and / or external device Based on glucose data (for example, if blood glucose levels are too high or too low, or If it shows an undesirable trend, it is configured to generate and / or provide a warning. It's fine if you do that.

[0038] By using CGM data, the Time Intake (TIR) ​​value is obtained, the TIR value is used It may be determined that the amount of time the blood glucose level in the 8 is within the threshold range over a reference period is the basis for determining this. The threshold range may be determined in advance, may be user-specific, or may be determined dynamically. It's okay.

[0039] The threshold range may be a value predetermined based on, for example, a cohort of patients. Each patient's lifestyle, habits, and medical test results within the group are determined based on pre-determined values. It may be used to determine the patient's life. For example, a cohort of one or more patients may be used to determine the patient's life. The determination may be based on lifestyle, customs, demographics, etc., and there may be one or more threshold ranges. The threshold range may be generated for each of the cohorts. The threshold range is derived from the analysis of blood glucose levels over a certain period of time. The determination may be based on the optimal result (e.g., a preferred A1C value) derived from the above. The machine learning model may be generated using the training module 310. The trainer may use the patient cohort's blood glucose levels as stage input 314. The corresponding A1C value may be received as a known result 318. The training machine learning model is As input, patient cohort data (e.g., A1C values) may be received, and that patient The threshold range of blood glucose levels in the cohort (i.e., having an upper limit and a lower limit for glucose) Alternatively, the output may be (ru). Alternatively, the threshold range may be the general population so that it is not cohort-specific. It may be a predetermined value. Depending on the implementation, the TIR threshold range is approximately 70 mg. The TIR value is between / dL and approximately 180 mg / dL. User 8's blood glucose level is the baseline. It may also be the amount of time within the intermediate TIR threshold range. Depending on the implementation form disclosed The reference period may be 24 hours, but finer-grained changes in TIR values ​​may be measured over 24 hours. The determination may also be based on reducing the base period which is less than an hour, and a broader change is 2 It should be understood that the determination may be based on increasing the reference period to one longer than 4 hours. .

[0040] User-specific threshold ranges can be determined based on attributes of user 8. These attributes include disease. This could be history, physical history, demographics, etc. According to one implementation, the user-specific threshold is: It may be generated using a machine learning model trained using training module 310. The machine learning model may receive updated attributes based on user 8, and the comparison results Through the 316 components, by using updated attributes, it can retrain itself. It may be refined. As one example, the change in user 8's weight is updated by the machine learning model. Based on the weight, the previously provided threshold range is updated, and the comparison results of 316 components are updated. This may also be a change in the attributes provided to the user. Therefore, the user-specific threshold range is It may be changed from time to time based on one or more attributes of the 8. Similarly, it is dynamically determined. The threshold range is related to user 8, user cohort, external conditions, environmental conditions, updated recommendations, etc. The determination may be based on changes in one or more related attributes.

[0041] As applied herein, a user vector (e.g., a patient vector) is a given patient Any row associated with, or potentially associated with, a person, and which may be changed. It may be a movement, activity, item (e.g., consumables), service, parameter, or value. The user vector is the TIR state or GV state of user 8, as further disclosed herein. It may be modified to improve the patient vector, drug, food intake characteristics. One or more of the following: motion values, psychosocial parameters, social determinant parameters, etc. It may include.

[0042] As applied herein, user attributes (e.g., patient attributes) are associated with a patient. It may be an attribute or characteristic. Patient attributes are easily compared to patient vectors. It may be something that cannot be modified or changed. As an example, patient Attributes include social attributes, medical history or condition, patient preferences, metabolic attributes, patient demographics, etc. But that's fine.

[0043] Figure 4A shows an exemplary CGM-based blood glucose trace 402 for user 8. The period shown can be a full day (i.e., 24 hours). As shown, -Za8 blood glucose levels are measured for a portion of the day, excluding the TAR period 402A and TBR period 402B. In relation to this, it must be within the threshold range represented by the upper threshold 404A and the lower threshold 404B. This may result in a TIR. User 8 can have such graphs during the day or at the end of the day. Display may be provided. Therefore, CGM data may be provided to user 8. User 8 is associated with User 8's current blood glucose level and / or User 8's current blood glucose level. You may notify us of the trend.

[0044] Figure 4B shows an exemplary CGM-based report that may be provided to user 8 or healthcare provider 7. It shows 406. The report is Ambulatory Glucose Profile ( It can be in AGP format, and not only graph data but also numerous metrics (for example, 1 May include 0 metrics). The report may include glucose statistics and target 408, A GP profile 410, daily glucose profile 412, time range 414, etc. It may be included. However, most diabetic patients will not be affected by changes in blood sugar levels. It may not be possible to interpret such CGM data and / or AGP information. Similarly, healthcare provider 7 requires discussions with multiple patients regarding CGM monitoring and / Alternatively, it may be possible to interpret the data provided via AGP information and temporarily optimize blood glucose levels. The technology disclosed herein provides tracking of essential parameters and manages the health of user 8. do.

[0045] According to the implementations disclosed herein, CGM data is further disclosed herein. It may be used to recommend changes based on one or more patient vectors. CGM events (e.g., changes in CGM status, partial CGM trace, etc.) are signs of diabetes. Defined as an identifiable region of CGM traces that correlates with Self-Management Activity (DSMA). This is also acceptable. As further applied herein, CGM tracing identifies CGM trends. It may be used for, or it may be a CGM trend. DSMA is a drug This may include changes or additions, changes or additions to food, changes or additions to exercise, etc. CGM is User 8 may be driven to automate coaching. Similarly, CGM-based results (e.g., The results of automated coaching and / or glucose characteristics based on DSMA will be used in future DSMA This drives coaching and / or coordinated specific decision-making support for healthcare. It may be provided to the provider 7.

[0046] Depending on the implementation, the Detection, Notification, Classification, and Engagement (DICE) framework is used in CG. Various diabetes-related events are detected from M-traces, and optimization processes are optimized through one or more visualizations. Notify healthcare provider 7 and / or user 8 of the progress along the route and for additional interventions. The detected events are classified into one or more classifications and / or two-dimensional CGM quadrant start states. and / or outlines techniques for engaging with and guiding patients toward improved outcomes. The technologies associated with the DICE framework include metabolic data, lifestyle data. Integrating data from multiple fields, such as patient care, socioeconomic data, and clinical data, To improve accuracy. The automated CGM event detection and classification techniques disclosed herein improve accuracy, Reducing errors makes it possible to improve the quality of care. Various quantitative methodologies Automated coaching based on this system is suitable for any patient who needs care and / or support. This enables improved accessibility and reachability. The visualizations provided herein are high-density computer graphics (CG). Easy-to-use charts, graphs, and / or other visualizations of M data and other applicable data. By focusing on data conversion, the data burden on user 8 and / or healthcare provider 7 is reduced. Figure 5A shows a method for providing an optimization pathway to improve the glucose state of user 8. This indicates 500. At 502, the blood glucose level of user 8 may be received. Disclosed herein Blood glucose levels may be provided continuously or semi-continuously by a CGM monitor. Blood glucose levels may be received by a component of the CGM monitor itself, or by an electronic device. S19, mHealth application 1, one or more local servers 29, etc. Alternatively, it may be received by a remote component. Blood glucose levels are obtained from one or more CGM monitors. It may be automatically provided to multiple components, and may also be pushed when blood glucose levels are collected. Often, or CGM monitors, ping to transmit one or more collected blood glucose levels. It may also be used.

[0047] As one embodiment, user 8 may attach the CGM monitor to their body, and the CGM monitor The CGM monitor may collect blood glucose measurements every 5 minutes. Network connection, local area network connection, wide area network connection, WiFi (via Bluetooth connection, etc.) Connects to User 8's mobile device Alternatively, according to the first exemplary implementation, the CGM monitor, each time a measurement is collected, (For example, every 5 minutes) Blood glucose readings are automatically sent to user 8's mobile device. Alternatively or additionally, one or more blood glucose measurements may be used as multiple measurements. It is sent to user 8's mobile device as a group, and / or, The 8 mobile device or another component sends one or more blood glucose measurements. When requiring confirmation, the CGM monitor stores one or more blood glucose measurement values. That's fine.

[0048] In Figure 5A, at 504, the time-in-range (TIR) ​​value associated with the blood glucose measurement is shown. The determination is made. The range of glucose values ​​is the amount of time during which the measured blood glucose level is within a predetermined range. The ratio of blood glucose measurements within the range to those outside the range, and the ratio of blood glucose measurements within and outside the range. It may also be possible to accommodate things like the T. The TIR value is when user 8's blood glucose level is within the range, and user It is possible to distinguish when the blood glucose level of 8 is outside the range. As shown in Figure 4A, the blood glucose level is on the upper side. When it is within the range of threshold 404A and lower threshold 404B, it can be considered to be within the range. The patient's TIR value is the time it takes for the patient's blood glucose level to be between 70 mg / dL and 180 mg / dL. To accommodate the amount, the upper threshold 404A may be 180 mg / dL, and the lower threshold The threshold 404B may be 70 mg / dL.

[0049] The TIR value determined at 504 in Figure 5A is the blood glucose level of user 8 over the reference period. It may also be based on the amount of time that falls within a threshold range. The reference period is a single 24-hour day. It may be the same, or it may be a different reference period. The reference period is (for example, for user 8) Therefore, it may be determined in advance by the healthcare provider 7 (e.g., pre-programmed). Alternatively, it may be determined dynamically based on one or more factors. This could be a patient vector, patient attributes, current or previous TIR status, etc.

[0050] In one implementation, the TIR value may be for the base period, or for multiple base periods. This could also be the TIR value associated with the patient over a period of time. For example, the TI of user 8 The R value may be determined for each day of the total 10 days. TIR values ​​from each day of the 10 days The TIR associated with user 8 over a 10-day period is calculated using the combined TIR value. As such, they may be combined using any applicable technique (e.g., averaging). .

[0051] In one implementation, the TIR value is used to exclude abnormal blood glucose levels, or to filter out abnormal values. The blood glucose measurements that cannot be lag are filtered to be weighted lower than those that cannot be lag-based. It is also possible. For example, if a blood glucose level of 65 mg / dL is measured during the first measurement, The level may increase to 200 mg / dL in the second measurement, which is taken 5 minutes after the first measurement. The third measurement, taken 5 minutes after the second measurement, may show a blood glucose level of 68 mg / dL. Density-based techniques (e.g., k-nearest neighbors, local outlier factor analysis, separated forests, etc.) Filters, subspaces, correlation-based, and / or tensor-based high-dimensional data to use Filters that use outlier detection, filters that use one-class support vector machines, Replicator neural network, autoencoder, variational autoencoder, long... Filters that use short-term memory neural networks, filters that use Bayesian networks Filters, filters using Hidden Markov Models (HMMs), and cluster analysis-based outliers. Filters that use value detection, related rules, and filters that use deviations from sets of frequent items. Filters that use fuzzy logic-based outlier detection, and filters that use ensemble techniques. Filters, feature bagging, score normalization, and filters using diverse sources, Filters using convolutional LSTM with a mixture of fractional principal component analyzers, etc. "Ta" can be misread and may be an outlier of no importance, such as an abnormal value and / or blood. It may be used to identify measurements of sugar levels. One such filtering technique. Alternatively, multiple models may also be used in conjunction with the machine learning models disclosed herein. According to the configuration, the TIR value associated with user 8 is such as one or more files It is possible to consider blood glucose measurements filtered through the LUTA. Therefore, such filtering is used to identify outliers, outlier data, and / or irregular measurements. This can prevent the provision of an optimized path that would be compromised as a result.

[0052] In Figure 5A, at 506, the TIR state of user 8 is one of the states associated with user 8. This can be determined based on multiple TIR values. The TIR state is associated only with the TIR value. The condition may be such that, or one or more other factors (e.g., frequency of glucose measurement) Based on factors such as the degree, the quality of glucose measurements, other perceived measurements, and patient-based factors. It may be. For simplicity, in this disclosure, the binary state of TIR based only on the TIR value is ( That is, we will discuss good TIR conditions and poor TIR conditions. However, T The IR state may be a multidimensional state based on the TIR value and one or more other factors. Please understand that a good TIR condition (e.g., first TIR state) corresponds to a TIR ratio greater than the TIR cutoff, and a bad TIR state ( For example, the second TIR state corresponds to a TIR ratio smaller than the TIR cutoff.

[0053] Figure 6A shows Chart 600 of TIR status for several different patients. Further details are provided herein. As disclosed, the chart includes four quadrants based on the TIR ratio and the GV ratio. The status is based on the TIR axis corresponding to the Y axis in Chart 600. The TIR ratio is: This is the percentage of time during a reference period that a patient's blood glucose level is within the threshold range. Alternatively, it can be expressed as the TIR ratio. The TIR ratio is a value calculated (e.g., averaged) over multiple reference periods. To that end, the proportion of time over a reference period in which the patient's blood glucose level is within the threshold range of multiple reference periods. That's fine.

[0054] The TIR ratio value is specified as the cutoff between good TIR conditions and poor TIR conditions. This is also good. Chart 600 in Figure 6A shows that a TIR ratio greater than 0.5 is (for example, user 8) A threshold range for blood glucose levels exceeding 50% of the time, or for calculated measurements. (If within the range) it is considered a good TIR condition, and a TIR ratio below 0.5 is (for example) Regarding user 8's blood glucose level exceeding 50% of the time, or the calculated measurement... Regarding this, if it is outside the threshold band, it is considered to be a poor TIR state, so the value is 0.5. Includes the cutoff value. The cutoff may be determined in advance or dynamically. The predetermined cutoff may be based on medical standards, or on a cohort or yu - May be designated by healthcare provider 7 for the 8. The dynamically determined cutoff is This may be based on a cohort or a specified user 8, or determined by a machine learning model. It may be determined. The machine learning model takes patient vectors, patient attributes, and past patient data as input. TIR values ​​or GV values ​​or changes may be received, and the input is associated with user 8 or A cutoff may be output specifically for the cohort. Therefore, the cutoff is the input data The value that the data was based on is considered optimal for the corresponding user 8 or cohort. It may be adjusted.

[0055] As shown in Chart 600, it has a TIR value that exceeds the cutoff of 0.5. The patient is considered to have a good TIR status, with a TIR below the cutoff of 0.5. Patients with a value are considered to have a poor TIR condition. Accordingly, it is understood that the number of patients with good or poor TIR status will change. It should be understood. For example, if the TIR ratio is adjusted from 0.5 to 0.9, most The patient will likely develop a poor TIR (Thorough Intravascular Reaction) condition.

[0056] In Figure 5A, at 508, the blood glucose fluctuation associated with the blood glucose measurement of a predetermined user 8. The (GV) value is determined. Blood glucose fluctuations are measured by the amount of change in glucose over a certain period of time. Furthermore, by utilizing fluctuations in glucose levels, diabetes management can be improved. The GV value is the standard deviation (SD). This may be the ) value, coefficient of variation (CV), or any other applicable measure of variation.

[0057] SDs are collected over (for example, an hour, a day, or any other applicable period) (t) It can be a measure of the amount of variation or variability in a set of glucose values. A low SD is a measure of glucose This may indicate that the course value tends to be close to the mean of the set of glucose values. This may indicate that the value is spread over a wider range. The standard deviation of glucose values ​​is Gluco It can be the square root of the variance of the glucose value. The standard deviation (SD) of the glucose value is calculated as shown in Equation 1. It is possible.

number

[0058] CV can be a standardized measure of the variability of a probability distribution or frequency distribution. (Patient's blood glucose) The coefficient of variation (CV) of a value is calculated by determining the ratio of the standard deviation of blood glucose levels to the mean blood glucose levels. It is possible. CV can indicate the degree of variability related to the average blood glucose level over a certain period. The coefficient of variation (CV) can be calculated as shown in Equation 2.

number

[0059] As mentioned above, the GV value may be the SD value or the CV value. According to one implementation, G The type of V value (e.g., SD value, CV value, etc.) may be based on user 8, or This is based on current or past patient vectors, patient attributes, or other information related to user 8. It may be based on. According to another implementation, the type of GV value is determined by the healthcare provider 7. Alternatively, based on one or more inputs such as patient vectors, patient attributes, and historical analysis, G Determined by a machine learning model configured to output the optimal type of V value. That's good too.

[0060] In Figure 5A, at 510, the GV status of user 8 is one or associated with user 8. It can be determined based on multiple GV values. The GV state is determined based solely on the GV value. There may be, or one or more other factors (e.g., frequency of glucose measurements, glucose This may also be based on factors such as the quality of course measurements, other perceived measurements, and patient-based factors. For simplicity, this disclosure does not describe a binary state of GV based solely on the GV value (i.e., good). We will discuss GV state and poor GV state. However, GV state is related to GV value and It will be understood that this may be a multidimensional 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 Corresponding to GV values ​​greater than "off," a bad GV state (e.g., a second GV state) corresponds to GV It supports GV values ​​smaller than the cutoff value.

[0061] Figure 6A shows chart 600 of the GV status of several different patients. Disclosed herein The chart includes four quadrants based on the TIR ratio and GV ratio. The GV state is... It is based on the GV axis, which corresponds to the X axis in Chart 600. GV is over a certain period of time. The standard deviation (SD) or central variability (CV) may be associated with the patient's blood glucose level. Alternatively, the GV may be G V is a value calculated (e.g., averaged) over multiple periods. This may be SD or CV associated with the patient's blood glucose levels over the period.

[0062] The GV value may be specified as a cutoff between a good GV state and a poor GV state. Chart 600 of 6A is considered to have a good GV state with a GV value above 0.8. A cutoff of 0.8 is included so that a GV value below this level is considered to indicate a poor GV state. The turn-off can be determined in advance or dynamically. The cutoff may be based on medical standards, or on the provision of medical care to a cohort or user 8. Person 7 may specify. The dynamically determined cutoff is a cohort or a predetermined user. It may be based on the user 8 or may be determined by a machine learning model. The machine learning mo del may receive, as input, a patient vector, patient attributes, the TIR value or GV value of a past patient, or a change or the like, and may output a particular cut-off for the user 8 or cohort associated with the input. Thus, the cut-off may be adjusted to a value that is considered optimal for the corresponding user 8 or cohort based on the input data.

[0063] As shown in the chart 600, patients having a GV value exceeding a cut-off of 0.8 are considered to have a good GV state, and patients having a GV value below the cut-off of 0.8 are considered to be in a poor GV state. It will be understood that when the cut-off is shifted, the number of patients having a good or poor GV state will change accordingly. For example, if the GV value is adjusted to 0.9 instead of 0.8, more patients will be in a poor GV state compared to when the cut-off is 0.8. According to one implementation form, the optimal cut-off value for distinguishing between a good state and a poor state may be 0.7.

[0064] As shown in FIG. 6A, four quadrants are formed based on the Y-axis (TIR ratio) and X-axis (GV value) divided based on the TIR cut-off value (i.e., 0.5 in the embodiment shown in FIG. 6A) and the GV cut-off value (i.e., 0.8 in the implementation example shown in FIG. 6A). Patients in the upper left quadrant 602 are patients within a good TIR state (i.e., exceeding the TIR cut-off value) and a poor GV state (i.e., lower than the cut-off GV). Corresponds. In this state, the first characterization (i.e., good) corresponds to the TIR state, and the second The characterization (i.e., "bad") can be considered to correspond to the good-bad (GB) state, which is the GV state. Obtain. Patients in the lower left quadrant 604 have a poor TIR condition (i.e., below the TIR cutoff). Addressing patients with a rotational GV (and poor GV) condition (i.e., lower than the cutoff GV). This condition can be considered a poor-poor (BB) condition. Patient in quadrant 606 in the lower right is Poor TIR conditions (i.e., below TIR cutoff) and good GV conditions (i.e.) This corresponds to patients whose GV is within the range of the cutoff GV. This condition is poor-good (B -G) This can be considered a condition. Patients in quadrant 604 in the lower left quadrant are in a poor TIR state (i.e., (Below TIR cutoff) and poor GV condition (i.e., lower than cutoff GV) This corresponds to patients in the quadrant 6. This condition can be considered a poor-poor (BB) condition. Patients 02, 604, and 606, respectively, are in at least one of the following states: TIR state or GV state. Patients with non-ideal conditions such that one is in a non-optimal state (e.g., a "poor" state) and It may be considered. Patients in the upper right quadrant 606 have a good TIR status (i.e., TIR ka Patients with a GV level above the cutoff and within a good GV state (i.e., higher than the cutoff GV) This corresponds to the person. This condition can be considered a good-good (GG) condition. Patient in quadrant 608 The user aims to ensure that both the TIR state and the GV state are in an optimal state (e.g., a "good" state). This could be considered a patient in an ideal state.

[0065] As shown in Figure 5A, 512, the starting state of a given patient is the patient's TIR state. and based on the GV state. As shown in Figure 6A, the starting state of user 8 is The TIR state and GV state can correspond to the quadrant in which they are classified. For example, as shown in Figure 6A Therefore, user 610 has a TIR ratio that does not satisfy the TIR cutoff, and thus, The TIR is in good condition, the GV is higher than the GV cutoff, and therefore the GV is good. It is possible that it is in that state. Therefore, as determined in 514 of Figure 5A, as shown in Figure 6A In this embodiment, the poor-good state of the non-ideal start is represented by quadrant 606 in the lower right. Possible. As further disclosed herein, a non-ideal start failure of user 610 -The overall good condition may be the condition of user 610 at a specific point in time, and may change over time. It's okay to change.

[0066] As determined in Figure 5A, 514, the non-ideal start state is the diabetes management of user 8. This can indicate that the logic is not optimal. For example, a non-ideal starting state has a low TIR and / or optimal It may show an unsuitable GV. Therefore, the non-ideal starting state is when the state of user 8 is unreasonable. To enable a transition from a hypothetical state to an ideal state, adjustments to the diabetes management of the corresponding user 8 are necessary. It can be considered essential.

[0067] According to one implementation configuration, as described herein and shown in Figure 6A, a two-dimensional frame Work is produced through a new data integration extract, transform, and write (ETL) process. It may be implemented in the system. The process is captured by the CGM monitor, and the CGM monitor By any of the following: T, electronic device 19, and / or any other applicable component The analyzed CGM data may be extracted. The extracted data may be processed by one or more machines. A production system that includes a machine learning model and can determine the starting state (for example, in 512 in Figure 5A). It may be converted into and / or written out to a database.

[0068] Therefore, the macro view of the state-based data of user 8 (e.g., the starting state) can be represented by two orthogonal parameters, the TIR state and the GV state. As disclosed in this specification (e.g., as shown in FIG. 6A), the corresponding state is visualized to evaluate the overall glucose health state and can be reported to user 8, medical provider 7, etc. The state-based data may be used (e.g., via an optimization pathway, as further disclosed in this specification) to provide overall glucose health recommendations.

[0069] In 516 of FIG. 5A, an optimization pathway can be generated to reach an ideal state. The optimization pathway may be one or more adjustments of one or more patient vectors and may be determined based on a non-ideal state (i.e., good-bad, bad-bad, or bad-good state), patient vector, and / or patient attributes. The optimization pathway may include adjustments of one or more patient vectors including, but not limited to, medications, food intake characteristics, exercise values, psychosocial parameters, and / or social determinant parameters.

[0070] The adjustment of the medication may be provided based on the user 8's current medication or may be provided based on a new medication that the user 8 may be provided. The adjustment may change the time or frequency of medication intake by adjusting the dosage of the medication, adding or eliminating the medication, or changing the environment associated with the medication (e.g., the type of food taken with the medication). This may be done by making changes, for example, the specific intake that user 8 is currently taking. The dosage of the medication may be adjusted to a higher dose.

[0071] Adjustment of food intake characteristics involves one or more foods, food groups, food types, food intake times, and meals. Change, eliminate, add, or otherwise modify product pairings, food and drug pairings, etc. This may include modifying the method. For example, after ingesting food, the patient's blood glucose level may reach a threshold. Based on patient attributes that indicate an increase beyond a certain range, patients are able to maintain low blood glucose levels during periods when their current blood glucose levels are low. They may be given a warning to consume the food.

[0072] Adjusting exercise values ​​involves changing one or more exercises, exercise types, exercise duration, exercise time, etc. This may include amending, removing, adding, or otherwise modifying. For example, If a person exercises at the beginning of the day, the GV of a given patient may be more stable, however Therefore, adjustments may be made to prioritize exercising in the morning.

[0073] Psychosocial parameters and / or social determinant parameters are also adjusted, or Meditation schedules or types, social activities, interactions, and / or the same may be modified. To change, eliminate, add, or otherwise modify the duration or frequency It may be included.

[0074] In 516, the optimized path can be generated using a machine learning model, as shown in Figure 3B. The machine learning model may be trained in this manner. The machine learning model takes the following as input: Patient vector, starting state (e.g., TIR state, GV state, good-bad state, bad-bad state) The patient receives one or more of the following information: condition (e.g., poor-good condition), patient attributes, and CGM characteristics. It is also possible that a machine learning model generates an optimized path output based on such an input. Good. As disclosed herein, the optimization pathway involves adjusting one or more patient vectors. That's fine.

[0075] In Figure 5A, at 518, the optimized pathway is directly to the patient (e.g., via the mHealth app). Provided to user 8 via communication 1 using electronic device 19, etc., medical It may be provided to provider 7, or to both. The optimized pathway is for one or more patients This may be an outline of changes to patient vectors, or an automatic adjustment of one or more patient vectors. It is also fine to do so gradually based on one or more actions, timings, levels, values, etc. The optimization pathway provided gradually may be one or more corresponding patient vectors. Provided based on the TOR, which may result in changes to one or more patient vectors. As an example, when the patient's blood glucose level is at the lower end of the threshold range, the change to the patient vector is food If this involves ingesting, when the patient's CGM monitor records such blood glucose levels, Mobile device warnings may be provided. Mobile device warnings are intended to alert users to potential threats. Provides instructions to users that they should consume food within a specified period based on the notice. That's good too.

[0076] The optimization path can also be performed periodically (e.g., daily, hourly, weekly, etc.) or trigger-based. It may be provided as is, and the predetermined time is based on changes based on the optimized path. For example, an optimization pathway that modifies a patient's meal schedule uses alerts during meal times. It may be provided using... Another embodiment is an optimization pathway that modifies the patient's medication, The medication may be provided with a warning during the medication time.

[0077] The frequency, method, and / or form of providing the optimization path is the successful implementation of the optimization path. It may be based on the main behavior or variable associated with it. The habit index may be one or more. The determination may be made for patients or cohorts of patients having similar attributes. The habit index is, This could also be a classification of patient behavior, or a specification of habits (e.g., frequent communication, frequent Infrequent communication, technical communication, telephone communication, Human communication, graphic communication, communication of time It may be a value or score, or it may be an appropriate adjustment of the optimization path. This may be any other applicable designation that provides instructions for the patient's actions.

[0078] The habit index is based on frequency-based factors, time-queue-based factors, and context-queue-based factors. The decision may be based on habits or preferences, including factors such as the following. The habit index is the optimization path It can be used to provide patients with an optimized pathway, as can be provided by a habit index. It may be done. As one example, the habit index is calculated by determining that user 8 engages in minimal communication. Preferences, and any communication conducted through the mHealth application. It may be shown that... Therefore, the change in the patient vector via the optimization path is... It may be provided to user 8 once via the mHealth application. The habit index then optimizes the patient's behavioral preferences in an individualized way. It may be used to provide.

[0079] Figure 5B shows a flowchart 540 of an exemplary implementation based on CGM. Step 512 corresponds to step 512 in Figure 5A, and as disclosed herein, This includes determining the initial state of a given patient based on the patient's TIR status and GV status. In 520, regarding whether the starting state determined in 512 is an ideal state, A determination is made. If the starting state is an ideal state, the CGM monitor at 522 It may be possible to continue running CGM if the starting state is not ideal (i.e., In 524, (non-ideal conditions such as good-bad, bad-bad, or bad-good conditions) One or more attributes of a non-ideal state can be determined. The attributes of a non-ideal state are determined by TIR or This may be the GV value, TIR, or change in GV. In 526, the patient is associated The patient vector can be identified. The patient vector is identified by user 8, and the healthcare provider 7 may be provided by, via electronic device 19, via server 29, or This may be obtained through any other applicable method.

[0080] In 528, the optimization pathway is used to transition the patient from a non-ideal state to an ideal state. It can be generated. Reaching an intermediate, non-ideal state is part of reaching the ideal state. It should be understood that this is possible. For example, a bad-bad (i.e., a bad TIR condition and Patients with an initiation of a poor GV state (non-ideal condition) may reach a good-good state before... First, the patient may be provided with an optimized pathway to transition between a good-to-poor state or a poor-to-good state. The machine learning model uses TIR states or values, GV states or values, and one or more patient vectors. Based on inputs including one or more patient characteristics, CGM events, etc. The system may output an optimized path that includes changes in one or more patient vectors. 530 odor The optimized pathway can then be provided to the patient. The optimized pathway is the probability that the patient will follow the optimized pathway. To enhance this, it may be provided based on a habit index associated with the patient. In addition to providing an optimized path, and / or providing an optimized path in 530 After that, in 522, the CGM monitor continues the CGM, and flowchart 540 is Based on continuing CGM at 522, by starting at 512 And it can be repeated itself. Flowchart 540 shows a given patient or patient Regarding the cohort, occurrences occur within any applicable period, which is predetermined or dynamically determined. You may do so.

[0081] Figure 6B shows charts 612 and 614. The first chart, 612, is several This shows multiple states of a given patient during a month. For example, the initial state of a patient shown by 613A. The condition is good-good (i.e., good TIR condition and good GV condition), 6 The subsequent state after the initial state shown by 13B is a bad-good state (i.e., a bad T The patient is in an IR state and a good GV state. Chart 612 shows the patient's various conditions over several months. It shows the state. Each state (e.g., 613A, 613B, etc.) is representative of the state during that period. It is possible. For example, the initial state shown by 613A is the average of all states during the first month. It can be an average value, or the same day of the month can be used for each month shown in Chart 612. The sea urchin may be in the condition specified on the day. Chart 614 in Figure 6B is the same as Chart 612. The same state is shown. However, the two-dimensional state-based quadrant corresponds to the TIR state and GV state. When relating to states, chart 614 is secondary, allowing viewers to see the distribution of states. This shows the quadrant based on the original state. Charts 612 and / or 614 are for viewers. To enable a better understanding of the status of the 8, the user 8 or medical advisor It may be provided to the provider 7.

[0082] Figure 6C shows charts 616, 618, and 620, each containing varying amounts of data. Chart 616 contains the most data with 15 months of CGM-based status information. Includes. Chart 618 shows 8-month CGM-based status information, and Chart 620 shows, This shows CGM-based status information for 3 months. More data points would allow viewers to see more Rather than relying on a small number of data points, it is better to understand the patient's overall blood glucose-based history. It can be made possible.

[0083] Figure 6D shows charts 621A and 621B illustrating the GV values ​​of a given patient over a 15-month period. Each of these is shown. Chart 621A shows the standard deviation (SD) of blood glucose measurement values, and Chart Section 621B shows the coefficient of variation (CV) of blood glucose measurements. As shown, applicable The type of GV (e.g., SD, CV, etc.) can change the GV state at a given time. For example, chart 621A, 621C, shows the SD-based GV value for the 9th measurement. As shown, 621C corresponds to a bad GV condition. However, Chart 6 The CV-based GV value of the same corresponding 9th measurement, represented by 21B's 621D. This corresponds to a good GV condition. The applicable GV type (e.g., SD, CV, etc.) is Based on one or more factors such as patient vectors, historical glucose information, and patient characteristics. You may choose from these options, but you are not limited to them.

[0084] Figure 6E shows the various conditions of multiple patients, each represented by an anonymized patient ID. Show chart 622. For example, patient 624 (i.e., patient ID 42799) is bar 6 Numerous missing states represented by 26A (for example, due to missing CGM data), A number of good-good states (i.e., good TIR state and) represented by bar 626B Good GV condition), numerous bad-good conditions represented by bar 626C, and bar 62 A number of poor--poor conditions represented by 6D may be present. A given cohort of patients is monitored. The healthcare institution or healthcare provider 7 that monitors the patient may periodically provide Chart 622. By reviewing the visual changes in the state shown in section 622, viewers can understand the main point. The trend of state changes for all or a subset of users implementing the technologies disclosed in the document. The direction can be easily determined.

[0085] Medical institutions or healthcare providers 7 also refer to Chart 628 and / or Diagram 6 in Figure 6F. 30 may be provided. Healthcare institutions or healthcare providers 7 use Chart 628 and patient population The trends in changes in status can be reviewed. For example, Chart 628 is provided for viewing. The person counts the patient's status as it changes from good-poor to poor-good (i.e., 1 It is possible to determine that 1) is greater than the opposite direction (i.e., 9). The data (for example, the network layer, weights, etc.) is improved and the optimized path is proposed. Used in updated machine learning algorithms (for the purpose of providing), (for example, in addition to the habit index) To improve how optimization paths are provided / implemented (based on changes made, etc.) It may be used.

[0086] Similarly, diagram 630 is used by medical institutions or healthcare providers 7 and patient groups The trends in changes in the group's status can be reviewed. By using diagram 630... This allows viewers to quickly see trends in status changes over multiple periods. It would be good to compare such trends. For example, viewers who were provided with Diagram 630 This easily calculates the number of status changes (i.e., 20) from poor-good to good-good. You may compare this to the change from the previous month. (Chart 628 and Diagram 630) This is indicated by the number of status changes, but status changes can also be expressed using percentages of change, etc. It should be understood that it may be expressed in any applicable way.

[0087] As disclosed herein, the optimization path generated in 516 of Figure 5A is one Alternatively, it may be generated based on multiple patient vectors. Figure 6G shows the data over a day (for example, Chart 632 shows 634 patient blood glucose measurements (collected using a CGM monitor). A filter or other smoothing mechanism is used to generate the corresponding trend line 636. This is also good. Chart 638 is a smoothed slope of blood glucose measurement 634 or Chart 632. The first derivative 640 of the chart representing the rate of change of the direction line 636 is shown. Charts 632 and 638. Both of these are examples of machine learning models that use vectors and their associated attributes (e.g.) For example, the time of each predetermined vector, the duration of the movement, the food type, the drug type, and / or administration. Motion vector 642, food vector 6 44, including patient vectors including drug vector 646 and another food vector 648. The blood glucose measurement 634, along with the first derivative 640 of the patient's blood glucose measurement 634, is mechanically... It may be used as input to a training model, or as a patient's blood glucose measurement 634 or primary Any of the derivatives 640 may be used individually. Therefore, the machine learning model The output optimization pathway provided is a patient blood glucose measurement 634, patient vector (for example, Motion vector 642, food vector 644, drug vector 646, and another food vector 6 48), and / or based on the first derivative 640.

[0088] As shown in Figure 4B, the AGP report includes not only graph data but also numerous metrics. It may include metrics (for example, 10 metrics). These metrics are used to inform patients and healthcare providers. For both the giver and the receiver, this can be numerous and difficult to understand. All 10 scales The degree is not necessary because it does not provide unique and independent information, as is the case with one of the AGP reports. The components of the metric are determined by other measures, and are therefore not disclosed herein. The technique described is partially based on minimizing the number of metrics. (Figures 7A and 77) B represents an average glucose level of 704, a hyperglycemic range (TAR) and a TIR of 706, and glucose variability. 710, hypoglycemia range (TBR), percentage of CGM activation time 722, and / or glucose Using the variation in standard deviation and principal component value (PCV) 724, the techniques disclosed herein This demonstrates that it can be implemented. Average glucose 704 is average glucose and glucose management. It may also be based on an index (GMI), which is a predicted representation of blood glucose levels. R and TIR706 indicate very high TAR (TAR_VH) and high TAR ( This may be based on the TAR_H and / or TIR indications. The glucose variation 710 is based on the glucose It is obtained based on the course standard deviation and PCV. TBR712 is a low TBR (TBR_L) This can be obtained based on a very low TBR (TBR_VL) and the TBR display.

[0089] According to one implementation of the subject matter disclosed, one or more CGM events correspond to the patient's blood glucose level. Classification may be based on the following: The classification is associated with each of one or more CGM events. It may be based at least on a severity score and / or on the patient's blood glucose level. It may be based on one or more characteristics of the curves. Figures 8A, 8B, and 8C are shown. An illustrative classification of CGM events is shown. The optimization path generated in 516 in Figure 5A is 1 It may be partially based on one or more classified CGM events. For example, CGM events The severity or other characteristics may be provided to the machine learning model, and the optimization path may be one or Output may be based at least partially on multiple classified CGM events. One example Therefore, the severity score may show the presence of a sharp peak, or a high severity score The frequency may indicate a large fluctuation in the patient's blood glucose level. Such CGM-based event information For example, the report indicates that TIR does not show an unhealthy amount of fluctuation in the patient's blood glucose level, therefore the patient has high T This can be particularly useful if you have an IR (Information Resource).

[0090] Applying CGM events to determine the optimal path is equivalent to CGM tracing (for example, Detection of events from a series of glucose value measurements, and classification of events into one or more classes. This may include similar classifications. Classification is based on severity score and / or glucose Categories may be included. Severity scores are calculated using the time and shape characteristics of the CGM trace. It may be judged.

[0091] Severity scores and / or CGM events are determined in relation to individual variations in CGM data. It may also be part of the microview of the CGM. Severity score and / or CGM The events involve real-time coaching or behavioral output (e.g., regarding medication, diet, exercise, etc.). It may be used for current coaching. Therefore, the technology disclosed herein This is a macro view of CGM data (for example, using state data as explained in Figure 5A). (to use) and a microview of CGM data (e.g., severity score and / or CGM) It provides both (using events) and real-time feedback and overall health We will provide both feedback and suggestions for improvement.

[0092] Figure 8A shows chart 800 having CGM trace 802 including CGM event 802A. Figure 8B shows chart 80 having CGM trace 806 including CGM event 806A. 4 is shown. CGM event 802A may be detected based on one or more mathematical methods. i. In the embodiment provided in Figure 8A, CGM event 802A has three parameters: Clinical significance of factors such as initial glucose, severity, and final glucose (b, s, d) It can be classified based on a multi-parameter CGM classification.

[0093] Parameter b is the glucose category at or near the start of a given CGM event. It is possible to correspond. Glucose category b is very high (e.g., +2), high (e.g., + 1) within the range (e.g., 0), low (e.g., -1), or very low (e.g., -2) ) may be scales such as. In the example of CGM event 802A, trace 802 When it curves toward the peak of CGM event 802A, at the start of CGM trace 802 As shown via trace 802, which is within the threshold range indicated by 803, The blood glucose level indicated by -802 is within the threshold range 803 at the start of CGM event 802A. Since it is located there, b corresponds to 0. In the embodiment of CGM event 806A in Figure 8B, trace When 806 curves toward the peak of CGM event 806A, it indicates the onset of CGM event 806A. Sometimes, as indicated by trace 806 which is within the threshold range 805, trace 806 Therefore, the blood glucose level shown is within the threshold range of 805 at the start of CGM event 806A. b corresponds to 0.

[0094] The parameter s is the height of the curve for a CGM event and the time the curve remains above the target. It can correspond to a severity score that encompasses both. The severity score s is the height of the curve for CGM events. This is expressed as a value (for example, between 0 and 9) that indicates the duration for which the curve remains above the target. It may be used. The severity score is determined by the height of the CGM curve and the corresponding traces that are outside the threshold range. Any applicable technique that provides a severity score based on the combination of duration It may be calculated via [a specific method]. In a simplified embodiment, the value associated with the height of the curve. This may be multiplied by a value associated with the duration of traces that are outside the threshold range. Height and One or both of the duration values ​​may be greater than 1. According to one implementation, Height and duration are based more on the severity score being greater on either height or duration. As shown, different weights may be assigned. A higher severity score indicates a higher target. A higher combination of height and duration may be indicated. A lower severity score. This may indicate a lower combination of height and duration exceeding the target. A lower severity score may be preferable to a higher severity score.

[0095] In the example of CGM event 802A, parameter s is associated with CGM event 802A. The determination is based on the height of the curve and the duration for which trace 802 is outside the threshold range 803. This corresponds to a severity score of 6. In the example of CGM event 806A, the parameters The 's' represents the height of the curve associated with CGM event 806A, and trace 806 is within the threshold range. This corresponds to a severity score of 2, which is determined based on the duration outside of 805. As shown in Figures 8A and 8B, the height of the curve for CGM event 802A and its position outside the target threshold range are significant. The duration of the time is determined by the height of the curve for CGM event 806A and the duration of the time outside the target threshold range. It is greater than time. Therefore, when compared to CGM event 806A (i.e., 2) The severity score for CGM event 802A is higher (i.e., 6).

[0096] Parameter e is the glucose category at or near the end of a given CGM event. It is possible to correspond. Glucose category b is very high (e.g., +2), high (e.g., + 1) within the range (e.g., 0), low (e.g., -1), or very low (e.g., -2) ) may be scales such as. In the example of CGM event 802A, trace 802 When it flattens after the peak of CGM event 802A, at the end of CGM trace 8 As shown via trace 802, which is approximately outside the threshold range indicated by 03, The blood glucose level indicated by Trace 802 is within range 80 at the end of CGM Trace 802A. Since it is higher than 3, e corresponds to 1. In the embodiment of CGM event 806A in Figure 8B, When race 806 flattens and changes direction below the threshold range 805, CGM event 806 As shown via trace 806 which is within threshold range 805 at the end of A, trace The blood glucose level indicated by 806 is below the threshold range 805 at the end of CGM event 806A. Since it is located at -1, e corresponds to -1.

[0097] According to another implementation, 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. It may be added. Figures 8C and 8D show the severity score and CGM event trace shape. This illustrates an exemplary CGM event characterized as such. As disclosed herein, the severity of The core is based on the height of the CGM trace and the duration of the trace that are outside the threshold glucose range. It may be calculated accordingly. The shape of a CGM event is, for example, a broad category, a high category, And may be classified in any applicable way, such as normal categories. Gori may be based on the start, peak, and end of a predetermined CGM event, and CGM The parameters associated with classifying races into predetermined categories are determined in advance. This may be done by means of a designated patient, or the determination may be made based on multiple CGM traces, etc. For example, the ratio of the area outlined by a given CGM trace to the height of the trace is It may also be used to classify CGM traces.

[0098] Figure 8C shows the CGM trace 812, the threshold range 803, and the three CGM events 812A. Chart 810 is shown, which includes 812B and 812C. The first CGM event 812A is It has a severity score of 8 and a CGM trace shape characterized as being broad. The second CGM event, 812B, was characterized by a severity score of 5 and high CG. It has an M trace shape. The third CGM event 812C has a severity score of 0 and is normal. It has a CGM trace shape that is characterized as a common feature. The third CGM event 812C The severity score is determined by whether the trace 812 at the peak of CGM event 812C falls within the threshold range 813. Since it is located at [location], the value is 0.

[0099] Depending on the implementation, the CGM trace shape may also be characterized as short. Furthermore, machine learning models can, for example, perform CGM tracing based on past CGM trace shapes. It may be used to identify the shape. The machine learning model is updated with the CGM trace. It may be updated based on the severity of the optimized pathway. For example, the updated glucose value may be based on the severity of the optimized pathway. After being provided based on the score, CGM trace shape, etc., it is calculated by the CGM monitor. It is acceptable. The updated glucose value may include the impact of the optimized pathway on the user. Good. The updated glucose values ​​are provided to the machine learning model to update the model. It may be used to generate updated CGM traces. For example, if the optimized path is If the user's state is not improved, the machine learning model will not improve in subsequent or future iterations. It may be updated to improve the output.

[0100] Figure 8D shows the CGM trace 822, the threshold range 823, and the two CGM events 822A. Chart 820 shows the first CGM event 822A having a weight of 9. It has a severity score and a CGM trace shape characterized as high. Second CGM Event 822B was characterized by a severity score of 8 and a CGM trace pattern that was normal. It has a certain condition.

[0101] One or more clinically significant CGM events relating to a specified user 8 are classified as CGM events (e.g., For example, b, s, and e in Figures 8A and 8B, or the severity scores and shapes in Figures 8C and 8D They may be classified using characterization (or any other applicable characterization). Then, The optimization path (e.g., via automated coaching messages, DSMA, etc.) is CGM Based on further characterization of the elephant, it may be sent to user 8.

[0102] Figure 9 includes a flowchart 900 for one implementation of the disclosed subject. 902 In this process, multiple optimization profiles are generated to reach the ideal state from a non-ideal state. It is possible. Multiple optimization profiles may not be patient-specific, but multiple patient profiles Each combination of counsel and patient attributes may be generated as disclosed herein. As shown in Figure 5B, multiple optimization profiles are trained using machine learning models. It can be generated using Dell. Multiple optimization profiles are used for the output of the machine learning model and It may be provided as such, and may be based on a cohort of past patients, and may be from a non-ideal state. The process of reaching the ideal state may also rely on successful or unsuccessful attempts.

[0103] Each of the multiple optimization profiles is based on one or more patient attributes and / or patient vectors. It may be associated with a predetermined set of patient vectors and patient attributes. For example, with respect to a predetermined set of patient vectors and patient attributes. A specific optimization profile can handle possible non-ideal states (e.g., good-bad, bad-bad). They may be generated for each (e.g., poor - good).

[0104] In 904, the TIR status of a predetermined patient is determined, and in 906, the information disclosed herein is The technique shown allows for the determination of the GV status of a given patient. In 908, the given patient One or more patient vectors and one or more patient attributes of a person may be received. The patient and / or patient attributes may be provided by the healthcare provider 7 by the specified patient. Often, via electronic device 19, via server 29, or any other applicable means It may be obtained via a step.

[0105] In 910, an optimized profile is identified based on patient vectors and patient attributes. Obtain. The optimization profile may contain a limited number of optimization paths, and each optimization path is It may correspond to a predetermined combination of TIR state and GV state. For example, the optimization profile The file is in a good-bad starting state, a bad-bad starting state, and a bad-good starting state. It may include an optimization pathway related to patient attributes. Therefore, a given optimization profile may include a patient attribute And may be identified based on vectors, and a limited number of optimizations based on the patient's initial state. It may include a profile.

[0106] In 912, the optimization pathway is determined from the optimization profile based on the TIR status of a given patient and The GV status can be identified. The optimal pathway is determined based on the vectors and attributes of all patients. Even if it is the same, it may differ for the same patient. For example, during the first repetition, The patient's optimized profile is based on the patient's attributes and vectors at the time of the first iteration. The patient's TIR status and GV status during the first iteration (e.g., good- Based on the poor state, a first optimization path may be identified. However, the second opposite During recovery, if the given patient vectors are the same (i.e., the same optimization profile is identified) Even if ( ), different optimization paths are based on changes in state (e.g., poor-good state). It may be specified. In 914, the optimal specified by the technology disclosed herein The pathway may be provided to a designated patient and / or healthcare provider 7.

[0107] Steps 502-517 in Figure 5A, 512-530 in Figure 5B, and 902-91 in Figure 9 While 4 is depicted in a specific order, the principles of this disclosure are not limited to the order in which they are depicted.

[0108] Figure 10A is a diagram 1002 including chart 1002 of CGM events by time and day. It indicates 00. Such diagrams or other visual outputs are CGM jars for a given period of time. To make it easier to understand, the application (for example, the mHealth application) It may also be provided to healthcare provider 7 or user 8 via (1). Diagram 1000 The Y-axis represents the number of days in the journey, and the X-axis represents time. Viewers should refer to Diagram 1. Upon receiving 000, easily create a pattern spanning a predetermined day, time, and / or number of days or times. You may make a judgment based on this.

[0109] Figure 10B shows Chart 1012 of total carbohydrate intake by time and meal type. Diagram 1010 is shown. Such diagrams or other visual outputs are predetermined. To make it easier to understand eating habits during that period, an application (e.g., mHealth) h may be provided to healthcare providers 7 or users 8 via application 1). 1010 includes the total number of carbohydrates on the Y axis and time on the X axis. Viewers can view the diagram. Gram 1010 received, and the pattern over a specified day, time, and / or number of days or times The user may be able to easily determine the type of food consumed during the day, You can easily see the calories associated with each meal type.

[0110] Figure 11 shows the first patient and Chart 1102, as shown via Chart 1102. As shown via 4, the severity score count for the second patient is shown. Chart 1 Each bar in charts 102 and 1104 represents the first patient and This represents the number of occurrences shown in the CGM data of the second patient. Generally, such distributions are A higher count with a lower severity score may indicate better diabetes management. There may be cases where a medical institution or healthcare provider 7 has treated one or more individuals over one or more periods. Even if you receive distribution charts for multiple patients (e.g., chart 1102 and chart 1104) Often, distribution charts can be used to monitor the progression of the entire patient group. Alternatively or additionally, The distribution map is based on specific patient groups (for example, based on treatment time, based on the medical team). Even if patient attributes, patient vectors, etc. are generated, and the trends of a specific patient group are analyzed, Often, it is helpful to compare trends across multiple patient groups.

[0111] Figure 12 shows a CGM-based approach to providing coaching to patients based on CGM data. Diagram 1200 shows the implementation form of C. As shown, one or more attributes are C This can be provided to the GM message generator 1212. The attributes are glucose value 1202, gluco -Trend 1204 (e.g., CGM trends, CGM event data, etc.), carbohydrate information Report 1206, Activity Information 1208, Insulin Information 1210, etc., or combinations thereof. This may include, but is not limited to, the following. The message level may be determined based on the attributes. For example, the message level is action level 1222, which indicates that urgent action is required (for example) Therefore, users should take insulin or consume carbohydrates to avoid harm. (For example, user) (Glucose should be carefully monitored), or an advisory message will be provided. Slevel 1226 (for example, no user action is required) may be possible.

[0112] The CGM message generator 1212 generates patient vectors (e.g., attributes 1202-1210) ) may be used to provide an optimized path based on Figure 5A, 518 or Figure This may be applied in 9 of 914. For example, the optimization path is attribute 1202-1210 The determination may be made at least partially based on the CGM message generator 1212 It may be provided to patients who require it.

[0113] Figure 13A shows an exemplary message provided using the CGM message generator 1212. This indicates message 1302. Message 1302 is provided via user 8's electronic device 19. It is possible. In the embodiment provided in Figure 13A, the message is a message of behavior level 1222. This is a message and may include the necessary actions. As shown, exemplary message 130 2 is, "Action required: Hey Charlie, your glucose is high and rising rapidly." To return to the range, go to the insulin computer and administer insulin. Go and get it." Message 1302 is sent with high importance. It can be provided to user 8 via mobile phone 1300. The high importance lies in electronic device 1 9 provides, in addition to message 1302, audible warnings, tactile warnings, visual warnings, etc. It may bring about.

[0114] Figure 13B shows the use of a CGM message generator 1212 via a mobile phone 1300. Another example message 1314 is provided. Message 1314 is a warning level 1 This is a 224 message and may not include urgent action. As shown, it is illustrative. Message 1314 says, "No action required: Charlie, your glucose is within the target range." It's rising slightly. There's no need to take any action at this time. Furthermore, Additional information such as blood glucose level 1312 is also provided via mobile phone 1300, and related messages. It may be available near page 1314.

[0115] Figure 13C shows the use of a CGM message generator 1212 via a mobile phone 1300. Another exemplary message 1322 is provided. Message 1322 is an advice This is a message from Bell 1226, including general advice for patients. (Figure 13C) Message 1322 also includes carbohydrate information, blood glucose levels, insulin dosage, etc. This may include other patient vectors.

[0116] Therefore, as shown through the embodiments in Figures 13A-13C, machine learning-driven self Dynamic user coaching or CGM feedback may be provided to user 8. For example, The stem and method require extremely important actions, such as in cases of hypoglycemia or extreme hyperglycemia. It can be used to provide a warning in cases of less critical glucose measurements. Knowledge messages may also be provided. Insulin administration support is provided via an optimization pathway. Therefore, as one of the patient vector corrections, corrective insulin is recommended based on the glucose trend. It can be provided. According to one implementation, the insulin dosage is based on the glucose trend (e.g., C The patient vector may be adjusted based on the GM event. The current blood glucose level is also It may also be a factor when determining the amount of insulin to adjust. For example, bolus insulin It may take a period of time (e.g., 30 minutes) to deliver the intended result, and accordingly Therefore, predicting the trend at the end of that period may be more useful than relying solely on current blood glucose levels. Trends can be an important component.

[0117] According to one implementation of the subject matter to be disclosed, an insulin computer may be provided. The insulin computer includes, for example, the amount of insulin to be administered at a predetermined time. To provide power, a context computer receives one or more factors as input. It may also be a data type. The insulin computer may be part of the CGM monitor. Alternatively, it may be outside the CGM monitor (for example, one or more electronic devices 19) (Partial use is also acceptable). External insulin computers are connected to electronic networks such as 32. 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. Alternatively, it may be an application. For example, an insulin computer is mHeal It may be part of application 1. The insulin computer may be one or more. It may receive a complex number of inputs and provide an action output. The action output is insulin. Whether it is needed, how much insulin is needed, and whether glucose is needed Whether or not this is the case, how much glucose is needed, and what food intake is required. Whether or not, how much food intake is needed, whether or not exercise is needed, This includes, but is not limited to, the amount of exercise required, one or more lines. The function of the insulin computer may be an indication or numerical value having a dynamic output category. The behavioral output may change based on the user's state. For example, the behavioral output may change based on the user's blood glucose level. It may be modified within a range to keep it safe and effective. In this embodiment, CGM training The input can be used to determine the optimal blood glucose level.

[0119] An insulin computer may receive CGM trends as input. As disclosed in detail, CGM trends include CGM traces, CGM events, etc. This may also be based on CGM traces, CGM events, etc. This may be based on changes in two or more glucose measurements over a period of time. CGM trend This may be based on glucose measurements provided by a CGM device. The trend is corrected or updated by additional glucose measurements. It may change over time as obtained. Past CGM trends can also be used as input. It's okay.

[0120] The insulin computer may receive meal information as input. The input may contain the content of the food before consumption (e.g., images, videos, etc.) or examples. Symbolic food examples (for example, an image of pizza found online to represent the food eaten) Provide insulin to the insulin computer by any applicable method, such as by entering it. The content may be entered using an electronic device 19, or - May be received from resources such as applications that track food intake. Information is stored at a certain point in time (for example, when a computer is used to determine the action output). Regarding user 8 in ), the insulin-carbohydrate ratio may be included, or insulin A computer may calculate the insulin-carbohydrate ratio. An insulin computer is a user The behavioral output based on dietary information for -za8 is generated when another user with the same dietary information on a given day The effects of food intake on user 8 may be individualized, as they may vary depending on the individual. The insulin computer uses dietary information and / or insulin-carbohydrate ratios to determine one or Multiple behavioral outputs may be adjusted. Past meal information may also be used as input. stomach.

[0121] Even if the insulin computer receives exercise information (e.g., any activity) as input, Good. Exercise information is generated by user input (e.g., past or planned exercises), (e.g.) For example, by an exercise or health tracker (from an electronic device 19), one of the CGM monitors This can be done by any applicable method, such as by multiple components, one or more sensors, etc. This information may be provided to the insulin computer. Exercise information includes calorie information and heart rate information. This may include the duration of exercise, exercise intensity, and physical strain. The action output based on user 8's exercise information is determined by another user who has the same exercise information on a predetermined day. The effects of the movement on user 8 may be individualized, as they may vary depending on the user. The Slin computer may adjust one or more behavioral outputs based on motor information. Past exercise information may also be used as input.

[0122] The insulin computer receives information about previous insulin doses as input. This may also be discussed further herein with reference to Figure 13D, the insulin computer The data is one or more factors associated with user 8 (e.g., diet, exercise, individual body). Considering factors such as individual characteristics, CGM trends, and past data, the previous dosage and previous administration The action output can be determined based on the given time. For example, if user 8 administers additional insulin... When determining whether or not to administer insulin, the insulin co The computer calculates how much insulin is still present in user 8's body from the previous dose. You may also determine whether or not it is the case.

[0123] The insulin computer may receive information about the current blood glucose level as input. Furthermore, the insulin computer takes CGM trends (for example, the body of user 8) as input. Information regarding the rate of change in glucose within the body may be received. Current blood glucose and / or CG levels. M Trend is the trend of blood glucose levels in the body of the insulin computer (for example, increasing It may be possible to determine the rate of change (such as addition, reduction, and stabilization). Based on this, the insulin computer may adjust one or more behavioral outputs. The blood glucose level may also be used as input.

[0124] The insulin computer receives the user's sensitivity to insulin as input. This is also acceptable. Sensitivity to insulin may be based on a predetermined value, or Even if it is based on past data received by a computer, CGM monitor, etc. i. According to one implementation, sensitivity is adjusted over time based on the user's insulin use. It may be done. Therefore, the insulin computer periodically checks the sensitivity to insulin. It may be updated periodically, or each time the user's action output is calculated.

[0125] The insulin computer may receive the user 8's hypoglycemia history as input. When the lance computer provides behavioral output, between the calculation of the hypoglycemic event and the behavioral output The duration may also be considered. Insulin computers also provide behavioral output when low The severity of the blood glucose event may also be considered. As an example, the history of user 8 is used in the calculation of the behavioral output. If a hypoglycemic event occurs within the past two days, or if user 8 has a blood glucose level exceeding 4% for three consecutive days... If you experience hypoglycemia, the recommendation from the insulin computer is to use insulin if there was no hypoglycemic event. It may be more conservative than the case described.

[0126] Figure 13D shows an exemplary insulin computer 1 in one implementation form of the disclosed subject. 330 is provided. As shown, of the four time zones provided in Figure 13D One time zone is the behavioral output provided by the insulin computer 1330. Insulators can be modified based on the time zone when providing action output. It can be provided as input to computer 1330. Each time zone can be provided by user 8. This can be determined based on the time of the last time a bolus injection was received. Zone 1334 is the time of mealtime bolus administration of insulin from time 1332 It may be. The second zone 1336 is within 2 hours of mealtime bolus administration 1332. It is possible. Zone 3 (1338) is 2 to 4 hours after a mealtime bolus (1332). This is possible. The fourth zone 1340 is more than 4 hours after a mealtime bolus administration 1332. This can happen. Each of the four zones is insulin-based, as shown in Figure 13D. This includes the Insulin-Carbohydrate Bloc (IOB), Correction Factor (CF), and Insulin-Carbohydrate Ratio (ICR), It may have associated attributes. If it is within the first zone 1334, IOB, CGM Based on the trend, adjusted CF and ICR administration may all be considered. Zone 1 of the second zone If within 336, IOB and CF may not be considered, but ICR administration is considered. This is possible. If it is within the third time zone 1338, CF without IOB and CGM adjustments. , and ICR administration may be considered. If within the fourth zone 1340, IOB, CGM Trend-based adjusted CF and ICR administration can all be considered.

[0127] Therefore, based on the factors discussed herein, the insulin computer is Whether insulin is needed, how much insulin is needed, glucose Whether it is needed, how much glucose is needed, and what food intake is needed Whether or not it will happen, how much food intake is needed, and whether or not exercise is needed. , how much exercise is needed, or a combination of these, It may provide behavioral outputs, not limited to those. The insulin computer is As a result of one or more factors such as the different histories of each patient, the first user having input The user can be individualized so that they can receive different behavioral outputs from a second user with similar inputs. The contextualized behavior output may be provided.

[0128] According to one implementation configuration, one or more behavioral outputs are part of the insulin computer. or determined using machine learning models associated with insulin computers. It is also fine. The machine learning model is based on known good outputs and / or the machine learning model Past behavioral output provided by the user, and past CGM data after providing the past behavioral output. A supervised model trained to provide behavioral output based on corresponding changes in trends. It may also be one or It may be configured to provide actionable outputs based on multiple inputs. (Machine learning mode) It may receive updated CGM trends after providing action output. Dell analyzes CGM trends and updates its models based on CGM trends (for example, heavy (Update neural networks, layers, etc.) and the results provided by the machine learning model The output of future actions may be improved. The machine learning model may (for example, the actions provided to the user) Based on the output and subsequent user CGM trends, this can be done for individuals, or for one or more people. Based on feedback from a number of users (i.e., CGM trends), multiple users You may update the model in that regard.

[0129] Figure 14 shows, for example, a host server and a decision server configured to function as a decision server for healthcare providers. This is a simplified functional block diagram of a computer that can be configured as follows. Figure 14 represents a network or host computer platform 1400. Those skilled in the art will know that They are familiar with the structure, programming, and general operation of such computer equipment. Therefore, the drawings are considered self-evident.

[0130] For example, a platform such as Server 1400 is a data server for packet data communication. It may include a communication interface 1460. The platform also includes program instructions. A central processing unit (CPU) 1420 in the form of one or more processors for execution It may include. The platform typically has an internal communication bus 1410, a program storage The ROM1430 and RAM1440, and other components are processed by the platform. and / or includes data storage for various data files to be communicated. Hardware elements, operating systems, and programming languages ​​of devices such as This is actually a conventional method, and it is presumed that those skilled in the art are well familiar with it. Server 1400 also supports keyboard, mouse, touchscreen, monitor, and display. Input / output port 1450 and communication port 1450 for connecting to input / output devices such as I. It may include 60. Of course, various server functions distribute the processing load, and many of the same It may be implemented in a distributed manner across various platforms. Alternatively, the server may be 1 Implemented by appropriate programming of one computer hardware platform It's okay.

[0131] As described herein, this disclosure covers software, hardware, firmware, etc. A, and / or the entity shown in the figure may be implemented in many different embodiments. This will be obvious to those skilled in the art in the relevant technical field. Hardware for implementing the embodiment Any actual software code involving specialized control of A is not limited to detailed explanation. Therefore, the examples, considering the level of detail presented herein, have been modified and The explanation is given with the understanding that deformation is possible. The described aspects of the subject are machine-readable. A form of executable code and / or related data that is executed or embodied in a certain medium. It can be considered as a "product" or "manufactured item" in terms of its state. A "storage" type medium is a con Any or all of the tangible memory such as computers and processors, or various semiconductors This includes memory, tape drives, disk drives, and their related modules. It can provide non-temporary storage at any time for software programming. Some software is sometimes used on the internet or various other telecommunications networks. Communication may be conducted through a mobile communication network. For example, such communication may be conducted through 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 be possible to load the software onto another computer or processor. However, Another type of medium that may involve software elements is a physical medium between local devices. Through interfaces, via wired and optical fixed-line networks, and various air links Including light waves, radio waves and electromagnetic waves, such as those used via wired connections. Alternatively, physical elements such as wireless links and optical links can also be considered as media accompanied by software. This may be considered. It is not limited to non-temporary, tangible “storage” media as used herein. Unless otherwise specified, terms such as "computer" or "machine" are used in relation to the process for execution. Refers to any medium involved in providing instructions to Sass.

[0132] Both the general explanation above and the detailed explanation below are illustrative and descriptive only. Please understand that the requested examples are not limited to those disclosed.

[0133] Other embodiments of the invention disclosed herein can be found in the description and examples disclosed herein. This should be clear to the public. The specification and examples are considered to be illustrative only, and this publication The true scope and essence of the invention are intended to be shown by the following claims.

[0134] As is clear from the figures, text, and examples presented above, various embodiments are conceivable. However, it is not limited to those.

[0135] 1. A computer implementation method for managing a user's glucose state, The user's blood glucose levels are received using a continuous glucose monitoring (CGM) device. That thing, The determination of the time in which the user's blood glucose level is within range (TIR) ​​is, the TIR value This is based on the amount of time the user's blood glucose level is within the threshold range over a reference period. The above determination, The TIR state is determined based on the aforementioned TIR value, This involves receiving at least the blood glucose fluctuation (GV) value based on the user's blood glucose level, The GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is the base Considering the standard deviation of the blood glucose levels over the specified period, the fluctuations in the user's blood glucose levels are shown. To receive the information, The GV state is determined based on the aforementioned GV value, The start state is determined based on the TIR state and the GV state, Determining that the aforementioned starting state corresponds to a non-ideal state, To reach an ideal state based on one or more user vectors and the starting state The process involves generating an optimized path, wherein the optimized path is used by one or more users The generation includes adjusting one or more vectors, Providing the aforementioned optimization path to the user, Methods that include...

[0136] 2. The threshold range is approximately between 70 mg / dL and 180 mg / dL, according to the embodiment. The method described in 1.

[0137] 3. The method according to Embodiment 1, wherein the reference period is 24 hours.

[0138] 4. The CV value is calculated by taking the standard deviation of the blood glucose level over the reference period. The method according to Embodiment 1, which is determined by dividing by the mean.

[0139] 5. The TIR state is selected from either a good TIR state or a bad TIR state. The method according to Embodiment 1, which is a binary state.

[0140] 6. The above-mentioned good TIR condition corresponds to a TIR value greater than the TIR cutoff, The method described in Method 5.

[0141] 7. The GV state is selected from one of either a good GV state or a bad GV state. The method according to Embodiment 1, which is in a binary state.

[0142] 8. The above-mentioned good GV state corresponds to a GV value greater than the GV cutoff in the embodiment. The method described in 7.

[0143] 9. The user vector may be related to drugs, food intake, exercise values, psychosocial parameters, or social factors. The method according to Embodiment 1, comprising one or more of the determinant parameters.

[0144] 10. Classifying one or more CGM events based on the user's blood glucose level. The classification described above is based on the severity associated with each of the one or more CGM events. The classification, which is based at least on the score, and one or more of the classified C Embodiment 1 further includes generating the optimization path based on GM events. Methods used.

[0145] 11. The optimization path is further based on user attributes, and these user attributes are social One of the following: physical attributes, medical attributes, user preferences, metabolic attributes, or user demographics. A method according to Embodiment 1, selected from multiple options.

[0146] 12. The optimization path involves increasing one or more state improvement habits, and / or one or The method according to Embodiment 1, including the reduction of multiple habits that worsen the condition.

[0147] 13. A computer implementation method for managing a user's glucose state, To reach an ideal state from a non-ideal state, multiple optimization profiles are received. Therefore, the aforementioned ideal state is a good time-in-time (TIR) ​​state and good blood glucose fluctuations. Corresponding to the (GV) state, the non-ideal state is a poor TIR state or a poor GV state. Receiving, which includes at least one of the following, The current TIR state is determined based on the TIR value of the user's blood glucose level, The aforementioned TIR value is based on the amount of time the user's blood glucose level is within the threshold range over a reference period. Therefore, the current TIR state is one of two conditions: a good TIR state or a bad TIR state. There is one, the determination mentioned above, The current GV status is determined based on the GV value associated with the user's blood glucose level. Therefore, the GV value indicates the standard deviation (SD) or coefficient of variation (CV) of blood glucose levels. The aforementioned CV is calculated by considering the standard deviation of the blood glucose levels over the reference period, and the user's blood glucose level The determination that indicates the change in value, Receiving one or more user vectors from the aforementioned user, Based on the one or more user vectors and one or more user attributes, the most Identifying one of the optimization profiles, Based on the identified optimization profile, the TIR state, and the GV state, Identifying an optimization path, wherein the optimization path is one or more user vectors The aforementioned identification includes one or more adjustments to the rule, Providing the aforementioned optimization path to the user, Methods that include...

[0148] 14. Each of the above-mentioned optimization profiles is a user vector and a user The method according to Embodiment 13, including different combinations of attributes.

[0149] 15. Each of the above-mentioned optimization profiles is associated with a plurality of optimization paths. Each of the above-mentioned multiple optimization paths is one of the potential TIR states or potential GV states. The method according to Embodiment 14, as identified based on one or more of the following.

[0150] 16. The machine learning model takes the optimization profile, the TIR state, and as inputs, The method according to Embodiment 13, which receives the GV state and outputs the optimized path.

[0151] 17. Receiving one or more user attributes and the one or more user attributes Furthermore, based on this, further including identifying one of the aforementioned optimization profiles, The method according to Embodiment 13.

[0152] 18. The CV value is calculated by taking the standard deviation of the blood glucose levels over the reference period. The method according to Embodiment 13, which is determined by dividing by the average value.

[0153] 19. A system for managing a user's blood glucose level, Memory for storing processor-readable instructions, A programmer configured to access the memory and execute the processor-readable instructions Rossessa and, Includes, When the processor-readable instruction is executed by the processor, To obtain glucose values ​​using a component that penetrates the user's skin. Using a continuous glucose monitoring (CGM) device configured as described above, the user's blood Receiving sugar levels electronically, The determination of the time in which the user's blood glucose level is within range (TIR), The value is based on the amount of time the user's blood glucose level is within the threshold range over a reference period. The threshold range is approximately between 70 mg / dL and 180 mg / dL, and the reference period The determination is that it is 24 hours, The TIR state is determined based on the TIR value, wherein the TIR state is good. The determination is made to select between a good TIR state or a bad TIR state. At least receiving a blood glucose fluctuation (GV) value based on the user's blood glucose level, The GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is the Considering the standard deviation of the blood glucose levels over the reference period, the fluctuations in the user's blood glucose levels are shown. Receiving the above, The GV state is determined based on the GV value, wherein the GV state is good G The determination that it is either a V state or a poor GV state, The start state is determined based on the TIR state and the GV state, Determining that the aforementioned starting state corresponds to a non-ideal state, The detection of a CGM event based on the user's blood glucose level, Multi-parameter CGM classification or characterization of severity and CGM event trace shape Characterizing the CGM event based on one or more of the following, the multip The Lameter CGM classification is based on the blood glucose level at the start of the CGM event, its severity, and the CGM event itself. The characterization includes the blood glucose level at the end of the process, Based on one or more account vectors and the characterization of the CGM event The process involves generating an optimized path in order to reach an ideal state, and the said optimized path The path includes the adjustment of one or more of the one or more account vectors that generate to, Providing the aforementioned optimization path to the user, A system that configures the processor to perform a method including the following.

[0154] 20. Providing the optimization path to the user means that, based on the optimization path, The system according to Embodiment 19 includes providing the user with text-based instructions. Tem.

[0155] Additional embodiments include the following:

[0156] 1. A system for providing glucose trend-based action output, Multiple glucose measurements are output based on the analysis of body fluids over a certain period of time. A continuous glucose monitoring (CGM) device is configured as follows: A memory configured to store the plurality of glucose measurement values, It is a processor, The multiple output by the CGM device and / or stored in the memory Based on changes in the number of glucose measurements, the CGM trend is determined. Based on the aforementioned CGM trend and at least one additional factor, at least one action can be taken. Determine the force, The process is configured to provide the user with at least one of the aforementioned action outputs. Ssa, A system that includes this.

[0157] 2. The CGM device, after a certain period of time, uses the body fluid to determine subsequent glucose levels. The processor is further configured to output a measured value, and the processor is configured to output the subsequent glucose - It is further configured to determine updated CGM trends based on the measured values. The system described in Embodiment 1.

[0158] 3. The at least one behavioral output indicates whether insulin is needed and how much. Whether insulin is needed, whether glucose is needed, and how much glucose Whether a course is required, whether food intake is required, and how much food intake is required. Is it required, is exercise necessary, or how much exercise is needed? The system according to Embodiment 1 corresponds to at least one behavior category selected from among the following. Tem.

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

[0160] 5. The system according to Embodiment 1, wherein the at least one additional factor includes dietary information. .

[0161] 6. The system according to Embodiment 5, wherein the dietary information includes the insulin-carbohydrate ratio.

[0162] 7. The system according to Embodiment 1, wherein the at least one additional factor includes motion information. .

[0163] 8. The exercise information includes calorie information, heart rate information, duration of exercise, exercise intensity, or body The system according to Embodiment 7, which may include at least one of the physical burdens.

[0164] 9. The at least one additional factor includes information regarding previous insulin doses. The system described in Embodiment 1.

[0165] 10. The system according to Embodiment 1, wherein the at least one additional factor includes blood glucose levels. .

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

[0167] 12. The occurrence of hypoglycemia within the threshold time triggers the behavioral output of the insulin recommendation behavior category, as described above. The behavioral activity of the insulin recommendation behavior category that does not involve the occurrence of hypoglycemia within the threshold time. The system according to embodiment 11, which is more conservative compared to force.

[0168] 13. The processor has one or more past behavior outputs and past CGM trends Configured to output the at least one action output based on the corresponding change. The system according to Embodiment 1, including a machine learning model.

[0169] 14. The at least one action output is the CGM monitor, electronic device, or app. Embodiment 1 provides to the user using at least one of the applications. The system described.

[0170] 15. A computer implementation method for providing glucose trend-based action output. There is, The CG analyzes body fluids over a certain period of time using a continuous glucose monitor (CGM) device. Receiving multiple glucose measurements based on the M device, Based on the changes in the plurality of glucose measurements output by the CGM device , determining the CGM trend, Based on the aforementioned CGM trend, determine at least one behavioral output, To provide the user with at least one of the aforementioned action outputs, Methods that include...

[0171] 16. The at least one behavioral output indicates whether insulin is needed and to what extent Whether insulin or glucose is needed for leprosy, and how much... Whether glucose is needed, whether food intake is needed, and how much food intake Whether it is needed, whether exercise is needed, or how much exercise is needed Embodiment 15 describes an embodiment corresponding to at least one behavioral category selected from among the following. The method.

[0172] 17. The CGM device, after a certain period of time, based on the body fluid, subsequently generates glucose It is further configured to output glucose measurements, based on the subsequent glucose measurements. The embodiment described in Embodiment 15 further includes determining the updated CGM trend. Law.

[0173] 18. The processor receives the updated CGM trend, and the updated Based on the observed CGM trend, determine at least one updated behavioral output, Implementation further includes providing the user with at least one updated behavioral output. The method described in morphology 17.

[0174] 19. A system for providing glucose trend-based action output, Multiple glucose measurements are output based on the analysis of body fluids over a certain period of time. A continuous glucose monitoring (CGM) device configured as follows, wherein the CGM device The vice accesses the bodily fluids through the user's skin, and the CGM device is 5 The continuous glucose monitor is configured to acquire glucose measurements in units of less than a minute. CGM (Continuous Global Mass) devices and A memory configured to store the plurality of glucose measurement values, It is a processor, The multiple output by the CGM device and / or stored in the memory Determining the CGM trend based on changes in the number of glucose measurements, CGM trends are CGMs that map glucose measurements over a certain period of time. Determined using the race, the CGM trend is among the CGM events or severity scores. The determination described above is further based on at least one of the following: Receiving at least one additional factor, wherein the at least one additional factor is , dietary information, exercise information, carbohydrate-insulin ratio, information on previous insulin dosage, Receiving the information, including one or more of the following: blood glucose levels and information regarding a history of hypoglycemia. and, Based on the aforementioned CGM trend and at least one additional factor, insulin is required. Whether or not, how much insulin is needed, how much glucose is needed. How much glucose is needed, and how much food intake is required? How much food intake is needed, whether exercise is needed, how much energy Identify at least one behavioral category selected from whether movement is required, Based on the aforementioned CGM trends and the aforementioned at least one additional factor, at least one Determining the behavioral output, wherein the at least one behavioral output is the at least one From one identified behavioral category, the at least one behavioral output is one or more Based on the number of past behavioral outputs and the corresponding changes in past CGM trends, the less The determination is made using a machine learning model configured to output a single action. The above determination, Based on the aforementioned at least one identified behavior category, the graphical user interface Generating a face (GUI) and The user is provided with the at least one action output via the generated GUI. That thing, After providing the user with at least one of the aforementioned action outputs, the updated CGM trend Receiving a signal, the updated CGM trend is the at least one row The receiving is based on glucose measurements after the dynamic output has been provided to the user. and, Updating the machine learning model based on the updated CGM trends, A system that includes this.

[0175] 20. Provide the updated CGM trend as input to the insulin computer. To provide and, by the insulin computer, the updated CGM trend Based on this, determine at least one updated action output, and the at least one The system according to Embodiment 19 further includes providing the user with updated behavioral output. Tem.

[0176] Additional embodiments include the following:

[0177] 1. A system for managing a user's glucose status, Multiple glucose measurements are output based on the analysis of body fluids over a certain period of time. A continuous glucose monitoring (CGM) device is configured as follows: A memory configured to store the plurality of glucose measurement values, It is a processor, A CGM trace is generated based on the plurality of glucose measurements over the aforementioned period. to, Identifying the severity score of the CGM trace, wherein the severity score is The height of the CGM trace and the duration of time the CGM trace remains above the target value. The aforementioned identification is based on time, Identifying the starting state based on the severity score, wherein the starting state is the The above identification indicates the user's glucose health status, To reach an ideal state based on one or more user vectors and the starting state, To generate an optimized path, wherein the optimized path is one or more user vectors The process of generating, which includes adjusting one or more of the torques, Providing the aforementioned optimization path to the user, A system that includes this.

[0178] 2. Identifying a starting parameter, wherein the starting parameter is compared with a target range. The scaled value is determined based on the starting point of the CGM trace, The above identification and the generation of the optimization path based on the above start parameters The system according to Embodiment 1, further comprising the above.

[0179] 3. The aforementioned starting parameters include very high parameters, high parameters, and parameters within a range. Embodiments selected from one of the following: low parameters and very low parameters. The system described in section 2.

[0180] 4. Identifying the termination parameters, wherein the termination parameters are compared with the target range. The scaled value is determined based on the endpoint of the CGM trace, The above identification and the generation of the optimization path based on the above termination parameters The system according to Embodiment 1, further comprising the above.

[0181] 5. The severity score is determined by the height of the CGM trace, and the CGM trace is the same as the target The system according to Embodiment 1, which is determined by multiplying by the duration that exceeds the standard value.

[0182] 6. The height of the CGM trace is given a first weight, and the CGM trace is The duration exceeding the target value is given a second weight different from the first weight. The system described in Embodiment 5.

[0183] 7. When a lower severity score is compared to a higher severity score, the ideal state The system according to Embodiment 1, which corresponds to a starting state that is closer to the state.

[0184] 8. The user vector may be related to drugs, food intake, exercise values, psychosocial parameters, or social factors. The system according to Embodiment 1, comprising one or more of the determinant parameters.

[0185] 9. The optimization path is selected from the optimization profile, and the optimization profile is Embodiment 1 is identified based on the severity score and one or more user characteristics. The system described above.

[0186] 10. Determining the time in range (TIR) ​​value of the CGM trace, The IR value is calculated based on the amount of time the CGM trace is within the threshold range over a reference period. There is, to make the above determination, The TIR state is determined based on the aforementioned TIR value, At least receiving blood glucose fluctuation (GV) values ​​based on the CGM trace, The GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is the Taking into account the standard deviation of the glucose measurements over the reference period, the glucose measurements Receiving the above, which indicates fluctuations, The GV state is determined based on the aforementioned GV value, The start state is determined based on the TIR state and the GV state. The system described in Embodiment 1, including the above.

[0187] 11. A computer implementation method for managing a user's glucose state, From the continuous glucose monitoring (CGM) device, over a certain period of time, the user's Receiving a meter reading, To generate a CGM trace based on the received glucose measurement values, Identifying the severity score of the CGM trace, wherein the severity score is The height of the CGM trace and the duration of time the CGM trace remains above the target value. Based on time, the aforementioned identification, The CGM trace shape of the CGM trace is identified, and the CGM tray The shape is based on at least one of the height or width of the CGM trace, the particular To determine, The starting state is identified based on the severity score and the CGM trace shape. Therefore, the aforementioned starting state indicates the glucose health status of the user, and the aforementioned identification, To reach an ideal state based on one or more user vectors and the starting state The process involves generating an optimized path, wherein the optimized path is used by one or more users The generation includes adjusting one or more vectors, Providing the aforementioned optimization path to the user, Methods that include...

[0188] 12. The CGM trace shape is a wide shape, a narrow shape, a short shape, and a tall shape. One of these is the method described in Embodiment 11.

[0189] 13. The CGM trace shape is determined based on the CGM trace. As identified by a machine learning model configured to output, as described in Embodiment 12 Method of loading.

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

[0191] 15. A system for managing a user's glucose status, Multiple glucose measurements are output based on the analysis of body fluids over a certain period of time. A continuous glucose monitoring (CGM) device configured as follows, wherein the CGM device The vice accesses the bodily fluids through the user's skin, and the CGM device operates for 5 minutes or less. The continuous glucose monitoring system is configured to acquire glucose measurements in units of less than 100 units. (CGM) device and, A memory configured to store the plurality of glucose measurement values, It is a processor, Generate a CGM trace that maps the glucose measurement values ​​over a certain period of time. That thing, Identifying the severity score of the CGM trace, wherein the severity score is The height of the CGM trace and the duration of time the CGM trace remains above the target value. The aforementioned identification is based on time, To identify the CGM trace shape of the CGM trace using a machine learning model. The CGM trace shape is at least the height or width of the CGM trace. Based on one, the aforementioned identification, The starting state is identified based on the severity score and the CGM trace shape. Therefore, the aforementioned starting state indicates the glucose health status of the user, and the aforementioned identification, To reach an ideal state based on one or more user vectors and the starting state The process involves generating an optimized path, wherein the optimized path is used by one or more users The generation includes adjusting one or more vectors, Based on the optimization path described above, a graphical user interface (GUI) is generated. That thing, The user is provided with at least one optimization path via the generated GUI. Toto, After providing the user with the optimized path, the updated CGM trace is received. And so the updated CGM trace provided the user with the optimized path. Based on subsequent glucose measurements, the aforementioned receiving Updating the machine learning model based on the updated CGM trace, A system that includes this.

[0192] 16. Identifying a starting parameter, wherein the starting parameter is proportional to the target range. In comparison, this is a scaled value determined based on the starting point of the CGM trace. Based on the above identification and the above start parameters, the optimized path is generated. The system according to Embodiment 15, further comprising the following:

[0193] 17. The aforementioned starting parameters are very high parameters, high parameters, and parameters within a range. The implementation is selected from one of the following: low parameter, very low parameter, and very low parameter. The system described in form 16.

[0194] 18. Identifying the termination parameter, wherein the termination parameter is relative to the target range. In comparison, this is a scaled value determined based on the endpoint of the CGM trace. Based on the above identification and the above termination parameters, the optimized path is generated. The system according to embodiment 15, further comprising the following:

[0195] 19. The CGM trace shape is a wide shape, a narrow shape, a short shape, and a tall shape. One example is the system described in Embodiment 15.

[0196] 20. The user vector may be related to drugs, food intake, exercise values, psychosocial parameters, or social factors. The system according to Embodiment 15, comprising one or more of the determinant parameters.

Claims

1. A computer implementation method for managing a user's glucose state, The user's blood glucose levels are received using a continuous glucose monitoring (CGM) device. That thing, The determination of the time in which the user's blood glucose level is within range (TIR) This is based on the amount of time the user's blood glucose level is within the threshold range over a reference period. The above determination, The TIR state is determined based on the aforementioned TIR value, This involves receiving at least the blood glucose fluctuation (GV) value based on the user's blood glucose level, The GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is the base Considering the standard deviation of the blood glucose levels over the specified period, the fluctuations in the user's blood glucose levels are shown. To receive the information, The GV state is determined based on the aforementioned GV value, The start state is determined based on the TIR state and the GV state, Determining that the aforementioned starting state corresponds to a non-ideal state, To reach an ideal state based on one or more user vectors and the starting state The process involves generating an optimized path, wherein the optimized path is used by one or more users The generation includes adjusting one or more vectors, Providing the aforementioned optimization path to the user, Methods that include...

2. The threshold range is approximately between 70 mg / dL and 180 mg / dL, as described in claim 1. Method of loading.

3. The method according to claim 1, wherein the reference period is 24 hours.

4. The CV value is the average of the blood glucose levels over the reference period, calculated by taking the standard deviation of the blood glucose levels. The method according to claim 1, which is determined by dividing by a value.

5. The TIR state is selected from one of a good TIR state or a bad TIR state. The method according to claim 1, wherein the state is a binary state.

6. The aforementioned good TIR state corresponds to a TIR value greater than the TIR cutoff, claim The method described in 5.

7. The GV state is a binary value selected from one of two options: a good GV state or a bad GV state. The method according to claim 1, which is in the state described above.

8. The aforementioned good GV state corresponds to a GV value greater than the GV cutoff, as described in claim 7. Method of loading.

9. The aforementioned user vectors include drug, food intake, exercise values, psychosocial parameters, or social decisions. The method according to claim 1, comprising one or more of the determinant parameters.

10. The process involves classifying one or more CGM events based on the user's blood glucose level, The classification is based on the severity score associated with each of the one or more CGM events. The above classification is based at least on, Based on the one or more CGM events classified above, the optimization path is generated. The method according to claim 1, further comprising the following:

11. The optimization path is further based on user attributes, and these user attributes are social attributes. , medical attributes, user preferences, metabolic attributes, or user demographics, one or more of these The method according to claim 1, selected from the above.

12. The optimization path involves increasing one or more state improvement habits, and / or one or more The method according to claim 1, including a reduction in habits that worsen the condition.

13. A computer implementation method for managing a user's glucose state, To reach an ideal state from a non-ideal state, multiple optimization profiles are received. Therefore, the aforementioned ideal state is a good time-in-time (TIR) ​​state and good blood glucose fluctuations. Corresponding to the (GV) state, the non-ideal state is a poor TIR state or a poor GV state. Receiving, which includes at least one of the following: The current TIR state is determined based on the 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 the threshold range over a reference period. Therefore, the current TIR state is one of either a good TIR state or a bad TIR state. There is one, the determination mentioned above, The current GV status is determined based on the GV value associated with the user's blood glucose level. Therefore, the GV value indicates the standard deviation (SD) or coefficient of variation (CV) of blood glucose levels. The aforementioned CV is calculated by taking into account the standard deviation of the blood glucose levels over the reference period, and the user's blood glucose level The determination that indicates the change in value, Receiving one or more user vectors from the aforementioned user, Based on the one or more user vectors and one or more user attributes, the most Identifying one of the optimization profiles, Based on the identified optimization profile, the TIR state, and the GV state, Identifying an optimization path, wherein the optimization path is one or more user vectors The aforementioned specification includes one or more adjustments to the rule, Providing the aforementioned optimization path to the user, Methods that include...

14. Each of the above multiple optimization profiles is a user vector and a user attribute and The method according to claim 13, including different combinations of the.

15. Each of the aforementioned optimization profiles is associated with a plurality of optimization paths, Each of the multiple optimization paths is one of the following: a potential TIR state or a potential GV state. The method according to claim 14, wherein is specified based on a plurality.

16. The machine learning model takes the optimization profile, the TIR state, and the The method according to claim 13, wherein the GV status is received and the optimized path is output.

17. Receiving one or more user attributes, and further adding to the said one or more user attributes A claim further comprising identifying one of the optimization profiles based on the above. Method 13.

18. The CV value is the average of the blood glucose levels over the reference period, calculated by taking the standard deviation of the blood glucose levels. The method according to claim 13, which is determined by dividing by a value.

19. A system for managing a user's blood glucose levels, Memory for storing processor-readable instructions, A programmer configured to access the memory and execute the processor-readable instructions Rossessa and, Includes, When the processor-readable instruction is executed by the processor, To obtain glucose values ​​using a component that penetrates the user's skin. Using a continuous glucose monitoring (CGM) device configured as described above, the user's blood Receiving sugar levels electronically, The determination of the time in which the user's blood glucose level is within range (TIR) The value is based on the amount of time the user's blood glucose level is within the threshold range over a reference period. The threshold range is approximately between 70 mg / dL and 180 mg / dL, and the reference period The determination is that it is 24 hours, The TIR state is determined based on the TIR value, and the TIR state is good. The determination is made to select between a favorable TIR state or a poor TIR state. At least receiving a blood glucose fluctuation (GV) value based on the user's blood glucose level, The GV value is one of the standard deviation or the coefficient of variation (CV), and the CV is the Considering the standard deviation of the blood glucose levels over the reference period, the fluctuations in the user's blood glucose levels are shown. Receiving the above, The GV state is determined based on the GV value, wherein the GV state is good G The determination that it is either a V state or a poor GV state, The start state is determined based on the TIR state and the GV state, Determining that the aforementioned starting state corresponds to a non-ideal state, The CGM event is detected based on the user's blood glucose level, Multi-parameter CGM classification or characterization of severity and CGM event trace shape Characterizing the CGM event based on one or more of the following, the multip The Lameter CGM classification is based on the blood glucose level at the start of the CGM event, its severity, and the CGM event itself. The characterization includes the blood glucose level at the end of the process, Based on one or more account vectors and the characterization of the CGM event The process involves generating an optimized path in order to reach an ideal state, and the said optimized path The path includes the adjustment of one or more of the one or more account vectors that generate to, Providing the aforementioned optimization path to the user, A system that configures the processor to perform a method including the following.

20. Providing the user with the optimization path means that, based on the optimization path, the context The system according to claim 19, further comprising providing the user with instructions for the base.

Images