Method and apparatus for calculating the risk of diabetes-related diseases based on multiple data

Abstract

One embodiment of the present invention discloses a multi-data-based method for calculating the risk of diabetes-related diseases, which can easily and accurately determine diabetes and diabetes-related complications.

Description

Technology Field

[0001] The present invention relates to a method for calculating the risk of developing diabetes and diabetes-related complications based on the results of collecting and analyzing data. Background Technology

[0002] Existing disease prediction models typically adopt an independent structure that calculates the risk of onset limited to a specific single disease. For example, while they provide individual prediction figures for specific conditions such as hypertension or diabetes, they have limitations in that they fail to comprehensively reflect the user's overall health status. Because this fragmented analysis method cannot comprehensively consider the various disease factors organically interconnected within the body, users face difficulties in intuitively and holistically understanding their health condition.

[0003] Although chronic diseases in modern people often stem from complex metabolic syndromes rather than a single cause, existing technologies lack mechanisms to predict multiple metabolic diseases simultaneously and comprehensively. Despite the high degree of interrelationship among metabolic diseases, treating them only as individual variables reduces prediction efficiency. In other words, there is currently no structural design in place to derive the probability of developing multiple diseases through a single analysis process and provide users with an integrated report.

[0004] The biggest limitation of existing disease prediction technologies is their inability to deeply analyze the causal or sequential relationships between diseases, such as diabetes and its associated complications. Lacking the functionality to simultaneously calculate and provide the probability of a specific disease transitioning to a subsequent one, these technologies merely present a one-off "current risk" and fail to offer proactive information regarding "future disease spread pathways." This poses a significant obstacle to users comprehensively understanding the progression of diseases and preparing for them.

[0005] In particular, conventional digital health systems have limitations in that they cannot provide information specialized to users because they only monitor / predict blood glucose values ​​based on CGM, provide only static risk analysis and disease prediction based on health checkup data, or do not incorporate diet, exercise, sleep, and stress data into their analysis and prediction. Prior art literature

[0006] Republic of Korea Published Patent Application No. 10-2025-0061315 (Published May 8, 2025) The problem to be solved

[0007] The technical problem that the present invention aims to solve is to provide a method and apparatus for calculating the risk of diabetes-related diseases based on multiple data. means of solving the problem

[0008] A method according to an embodiment of the present invention for solving the above technical problem comprises: receiving and preprocessing first data measured by a blood glucose meter from a user terminal, second data of non-blood glucose measured by a wearable device, third data which is the user's body data input to the user terminal, and fourth data which is the user's health checkup data; calculating a score set including a plurality of scores for the preprocessed first data, second data, third data, and fourth data; calculating a comprehensive score based on individual scores included in the calculated score set; and calculating data on the risk of developing diabetes and diabetes-related complications for the user based on the calculated comprehensive score.

[0009] In the above method, the step of calculating the score set may calculate a score set including a health checkup score, a blood sugar score, a meal score, an exercise score, a sleep score, and a stress score based on the first data, the second data, the third data, and the fourth data.

[0010] In the above method, the step of calculating the comprehensive score may calculate the comprehensive score by applying a unique weight set for each of the health checkup score, blood sugar score, diet score, exercise score, sleep score, and stress score.

[0011] In the above method, the disease risk data can be calculated by ensembling a first model that processes the fourth data to produce static data and a second model that processes the first data, the second data, and the third data to produce dynamic data.

[0012] In the above method, the first model is a Cox model that calculates general risk using the static data, and the second model may be a correction function model that calculates a correction value using the calculated comprehensive score as an individual correction factor.

[0013] In the above method, the correction value may be a value with an upper limit set to 25%.

[0014] In the above method, the step of calculating the disease risk data may calculate the disease risk data by considering at least one of the calculated comprehensive score and the sensitivity coefficient, which is the rate of change of the probability of disease onset for changes in the user's lifestyle habits.

[0015] In the above method, the step of calculating the disease risk data may further use at least one of the XGboost survival model and the random survival forest model to calculate the disease risk data.

[0016] In the above method, the method may further include the step of converting the calculated disease risk data into display data that can be displayed on the user terminal and transmitting it to the user terminal.

[0017] An apparatus according to another embodiment of the present invention for solving the above technical problem comprises: a memory in which at least one program is stored; and a processor that performs calculations by executing the at least one program, wherein the processor receives and preprocesses first data measured by a blood glucose meter from a user terminal, second data of non-blood glucose measured by a wearable device, third data which is the user's body data input to the user terminal, and fourth data which is the user's health checkup data, calculates a score set including a plurality of scores for the preprocessed first data, second data, third data, and fourth data, calculates a comprehensive score based on individual scores included in the calculated score set, and calculates data on the risk of developing diabetes and diabetes-related complications for the user based on the calculated comprehensive score.

[0018] In the above device, the processor can calculate a score set including a health checkup score, a blood sugar score, a meal score, an exercise score, a sleep score, and a stress score based on the first data, the second data, the third data, and the fourth data.

[0019] In the above device, the processor can calculate the overall score by applying a unique weight set for each of the health checkup score, blood sugar score, meal score, exercise score, sleep score, and stress score.

[0020] In the above device, the disease risk data can be calculated by ensembling a first model that processes the fourth data to produce static data and a second model that processes the first data, the second data, and the third data to produce dynamic data.

[0021] In the above device, the first model is a Cox model that calculates a general risk level using the static data, and the second model may be a correction function model that calculates a correction value using the calculated comprehensive score as an individual correction factor.

[0022] One embodiment of the present invention may provide a computer-readable recording medium storing a program for executing the method. Effects of the invention

[0023] According to the present invention, the long-term risk of developing diabetes and diabetes-related complications can be accurately and easily provided to the user by utilizing multiple data such as the user's health checkup data, the user's biosignal data, and the user's blood glucose data. Brief explanation of the drawing

[0024] FIG. 1 is a schematic diagram illustrating an example of a diabetes-related disease risk calculation system implementing the method according to the present invention. FIG. 2 is a block diagram showing an example of a risk calculation device according to the present invention. Figure 3 is a diagram illustrating a submodule included in the processor of Figure 2. FIG. 4 is a block diagram illustrating an example of a user terminal according to the present invention. FIG. 5 is a diagram for explaining a submodule included in the terminal processor of FIG. 4. FIG. 6 is a flowchart illustrating an example of a method according to the present invention. Specific details for implementing the invention

[0025] The present invention is capable of various modifications and may have various embodiments; specific embodiments are illustrated in the drawings and described in detail in the detailed description. The effects and features of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described below in detail together with the drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various forms.

[0026] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. When describing with reference to the drawings, identical or corresponding components are given the same reference numerals, and redundant descriptions thereof will be omitted.

[0027] In the following embodiments, terms such as first, second, etc. are used not in a limiting sense, but for the purpose of distinguishing one component from another component.

[0028] In the following embodiments, singular expressions include plural expressions unless the context clearly indicates otherwise.

[0029] In the following embodiments, terms such as "include" or "have" mean that the features or components described in the specification are present, and do not preclude the possibility that one or more other features or components may be added.

[0030] Where an embodiment can be implemented differently, a specific process sequence may be performed differently from the order described. For example, two processes described consecutively may be performed substantially simultaneously or proceed in the reverse order of the description.

[0031] FIG. 1 is a schematic diagram illustrating an example of a diabetes-related disease risk calculation system implementing the method according to the present invention.

[0032] Referring to FIG. 1, it can be seen that the diabetes-related disease risk calculation system (1) includes an external device (100), a risk calculation device (200), and a user terminal (300).

[0033] In FIG. 1, the external device (100) refers to a device that primarily provides raw data necessary for the risk calculation device (200) to calculate the risk of developing diabetes and diabetes-related complications of a user to a user terminal (300), and may include a blood glucose meter (101), a wearable device (102), and a medical database (103).

[0034] In FIG. 1, the blood glucose meter (101) refers to a continuous glucose monitoring device that measures the user's blood glucose level by using it periodically when the user is a Type 1 diabetes patient, and can provide the measured blood glucose level to the user terminal (300) based on a short-range communication method such as Bluetooth or Wi-Fi. The data measured by the blood glucose meter (101) may be referred to as blood glucose data to distinguish it from the non-blood glucose data described later. As another example, the blood glucose meter (101) may be a device that estimates blood glucose through a non-invasive method, in addition to a device that measures blood glucose by necessarily collecting the user's blood, which is a conventional traditional method. In the past, blood from the patient was required to determine the patient's blood sugar, but recently, a method has been introduced to estimate the patient's blood sugar through a non-invasive method using Raman spectroscopy. Specifically, by focusing light of a specific wavelength onto a specific part of the body where capillaries are concentrated, and then analyzing the Raman spectrum that has been wavelength-transformed by glucose molecules, the blood sugar concentration can be estimated without collecting the patient's blood. The blood glucose meter (101) of FIG. 1 is considered to be a broad concept that includes a device for estimating (measuring) blood sugar without collecting the user's blood as described above.

[0035] In FIG. 1, the wearable device (102) is mounted on a specific part of the user's body to collect the user's biosignals and convert the collected biosignals into digital data to transmit to a user terminal (300) via a communication network (20). As an example, the wearable device (102) may be a smart watch that can be mounted on a patient's wrist. In FIG. 1, the wearable device (102) is depicted in the form of a smart watch that can be mounted on a patient's wrist; however, since the present invention does not limit the type of wearable device (102) to a specific type, a device mounted on a part other than the patient's wrist may also be included in the wearable device (102). For example, a blood pressure monitor mounted on the patient's upper arm and a band-type measuring device mounted on the patient's neck or ankle may also be wearable devices (102) according to the present invention. Additionally, there may be multiple wearable devices (102) mounted on the user. The biosignal collected by the wearable device (102) and transmitted to the user terminal (300) may be referred to as non-blood glucose data to distinguish it from the blood glucose data measured by the aforementioned blood glucose meter (101).

[0036] In FIG. 1, the medical database (103) refers to a database that stores medical data operated by a medical institution. Here, the medical institution may be a tertiary general hospital corresponding to a tertiary medical institution, or a government agency such as the National Health Insurance Service that provides health checkup data through a prescribed procedure, or a quasi-government agency entrusted with execution. The user terminal (300) of FIG. 1 uses the medical institution's unique API (Application Programming Interface) to enable the user to access data stored in the medical database (103), download it in a usable format, or view the data. As an example, the data that can be provided from the medical database (103) to the user terminal (300) may include the user's health checkup data.

[0037] Health checkup data collected from a medical database (103) may include anthropometric and basic information, lifestyle indicators, biosignal information, and blood and biochemical indicators as sub-parameters. First, anthropometric and basic information may include the subject's age, gender, waist circumference, weight, and height. Lifestyle indicators may include the subject's smoking status, frequency of drinking, and number of days (number of times) of weekday exercise. Biosignal information may include the subject's systolic blood pressure (SBP) and diastolic blood pressure (DBP). Blood and biochemical indicators may include values ​​for fasting blood glucose, total cholesterol, triglycerides (TG), HDL cholesterol, serum creatinine, serum AST, serum GPT, and gamma-GPT. That is, the health checkup data must include a total of 18 sub-parameters.

[0038] As an optional embodiment, if the medical database (103) is a database of a hospital where the user's outpatient records remain, the data that the medical database (103) can provide to the user terminal (300) may further include the user's estimated glomerular filtration rate (eGFR). The estimated glomerular filtration rate can be extracted from the user's hospital medical records (EMR data) and, unlike the 18 sub-parameters mentioned above, can be used as an auxiliary parameter to calculate the risk of developing kidney disease, which is one of the user's diabetes-related complications.

[0039] The risk calculation device (200) is a core device for implementing the disease prediction method according to the present invention, and can perform the function of calculating the risk of diabetes and diabetes-related complications for a user possessing a user terminal (300) by training and enhancing a prediction model based on data collected through external devices (100) and a user terminal (300). The risk calculation device (200) is electrically and logically connected to the user terminal (300) through a communication network (20), and although it is shown in the form of a server in FIG. 1, it can be implemented in various forms.

[0040] The user terminal (300) refers to a terminal held by a user and may be a multi-functional composite terminal including an input unit (not shown) capable of receiving various data from the user, and a display device or speaker capable of visually or audibly outputting processed data to the user. Additionally, although the user terminal (300) is shown in the form of a smartphone in FIG. 1, it may be implemented as a tablet or a PC depending on the embodiment.

[0041] The communication network (20) performs the function of connecting an external device (100), a risk calculation device (200), and a user terminal (300), which are components of the diabetes-related disease risk calculation system (1), and may include various wired and wireless communication networks such as a data network, a mobile communication network, and the Internet. In particular, in the present invention, the communication network (20) includes not only the mobile communication network currently in use but also the old generation mobile communication network that has already been used and discarded, and the next generation mobile communication network for which infrastructure is to be built and used in the future. Therefore, it may be one of the following: GSM (Global System for Mobile communications), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), CDMA 2000, LTE (Long Term Evolution), LTE-A (Long Term Evolution Advanced), 5G (5-Generation), and 6G mobile communication networks scheduled to be serviced in 2030. In addition, the communication network (20) may include a network implemented through satellite communication, such as STARLINK.

[0042] In one embodiment, the diabetes-related disease risk calculation system (1) according to FIG. 1 can effectively predict disease based on data collected through an external device (100) and data input into a user terminal (300). Specifically, the diabetes-related disease risk calculation system (1) may include an external device (100) that collects data based on user input or collects biosignals mounted on a specific part of the user's body, a risk calculation device (200) that receives and processes data from the external device (100) and the user terminal (300), and a user terminal (300) that generates data based on user input, transmits data received from the external device (100) to the risk calculation device (200), and visually displays the disease risk received from the risk calculation device (200).

[0043] In particular, the risk calculation device (200) can provide intuitive and accurate disease prediction information to the user by converting the calculated diabetes-related disease risk data into display data that can be displayed in levels or percentage units on the user terminal (300) and transmitting it to the user terminal (300).

[0044] FIG. 2 is a block diagram showing an example of a risk calculation device according to the present invention.

[0045] Referring to FIG. 2, it can be seen that the risk calculation device (200) includes a communication unit (210), a processor (230), and a memory (250).

[0046] The communication unit (210) may include one or more components that enable wired / wireless communication with an external device. For example, the communication unit (210) may include at least one piece of hardware necessary to implement short-range communication such as Wi-Fi or Bluetooth in a network provided by the communication network (20), or to implement various communications including the Internet when a LAN cable is connected.

[0047] Memory (250) is hardware that stores various data processed within the risk calculation device (200) and can store programs for processing and controlling the processor (230). Memory (250) may include RAM (random access memory) such as DRAM (dynamic random access memory) and SRAM (static random access memory), ROM (read-only memory), EEPROM (electrically erasable programmable read-only memory), CD-ROM, Blu-ray or other optical disc storage, HDD (hard disk drive), SSD (solid state drive), or flash memory.

[0048] The processor (230) can control the overall operation of the risk calculation device (200). For example, the processor (230) can control the operation of the input unit (not shown), display (not shown), communication unit (210), memory (250), etc. included in the risk calculation device (200) by executing programs stored in memory (250).

[0049] As an example, the processor (230) receives and preprocesses first data measured by a blood glucose meter from the user terminal (300), second non-blood glucose data measured by a wearable device (102), third data which is the user's body data input into the user terminal (300), and fourth data which is the user's health checkup data, calculates a score set including multiple scores for the preprocessed first data, second data, third data, and fourth data, calculates a comprehensive score based on the individual scores included in the score set, and calculates data on the risk of developing diabetes and diabetes-related complications for the user using the user terminal (300) based on the comprehensive score. The specific process processed by the processor (230) will be described later through FIG. 3.

[0050] When the risk calculation device (200) is implemented as a physical device, the processor (230) may be implemented using at least one of ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), controllers, micro-controllers, microprocessors, and other electrical units for performing functions.

[0051] In addition, when the risk level calculation device (200) in the present invention is implemented in the form of an application (program) that runs on an integrated data processing device, the processor (230) and memory (250) included in the risk level calculation device (200) may be implemented in the form of a virtual machine that implements hardware such as DSPs, microcontrollers, RAM, ROM, HDD, etc., as software (command script).

[0052] Hereinafter, the process processed by the processor (230) of the risk level calculation device (200) according to the present invention will be described by example.

[0053] In one embodiment, the processor (230) receives and preprocesses first data measured by a blood glucose meter (101) from a user terminal (300), second data of non-blood glucose measured by a wearable device (102), third data which is the user's body data input into the user terminal (300), and fourth data which is the user's health checkup data, calculates a score set including multiple scores for the preprocessed first data, second data, third data, and fourth data, calculates a comprehensive score based on the individual scores included in the score set, and calculates data on the risk of developing diabetes and diabetes-related complications for a user using the user terminal (300) based on the comprehensive score. According to the present embodiment, the data transmitted to the risk calculation device (200) may be three types (first data, second data, and fourth data) from an external device (100) and one type (third data) from the user terminal (300). According to an embodiment, when a user manually inputs all items listed in the user's health checkup data through the input unit of the user terminal (300), the fourth data may be included in the third data.

[0054] In one embodiment, the processor (230) of the risk calculation device (200) can calculate a score set including a health checkup score, a blood sugar score, a meal score, an exercise score, a sleep score, and a stress score based on the first data, the second data, the third data, and the fourth data. A method for calculating the health checkup score, the blood sugar score, the meal score, the exercise score, the sleep score, and the stress score will be described later in FIG. 3.

[0055] In one embodiment, the processor (230) of the risk calculation device (200) can calculate a comprehensive score by applying a unique weight set for each of the health check score, blood sugar score, meal score, exercise score, sleep score, and stress score. The weights set for each of the health check score, blood sugar score, meal score, exercise score, sleep score, and stress score will be described later through Table 1.

[0056] In one embodiment, the disease risk data calculated by the processor (230) of the risk calculation device (200) may be data calculated by ensembling a first model that calculates static data by processing the fourth data and a second model that calculates dynamic data by processing the first to third data. In this case, the first model may be a Cox model that calculates general risk as static data, and the second model may be a correction function model that calculates a correction value using the calculated comprehensive score as an individual correction factor, and the correction value may have an upper limit set to 25%. The ensemble model of the first model and the second model and the correction value will be described later in FIG. 3.

[0057] In one embodiment, the processor (230) of the risk calculation device (200) can calculate the risk of onset data by considering at least one of a comprehensive score and a sensitivity coefficient, which is the rate of change of the probability of onset for changes in the user's lifestyle habits. At this time, the processor (230) of the risk calculation device (200) can calculate the risk of onset data by further using at least one of an XGboost survival model and a random survival forest model.

[0058] In one embodiment, the processor (230) of the risk calculation device (200) can convert the disease risk data into display data that can be displayed on the user terminal (300) and transmit it to the user terminal (300). An example of the display data displayed on the user terminal (300) will be described later.

[0059] Figure 3 is a diagram illustrating a submodule included in the processor of Figure 2.

[0060] Referring to FIG. 3, it can be seen that the processor (230) includes a preprocessing module (231), a score calculation module (233), and a risk calculation module (235). Since the preprocessing module (231), score calculation module (233), and risk calculation module (235) shown in FIG. 3 are modules that are logically and conceptually separated to explain the process performed by the processor (230) in the process of implementing the method according to the present invention, FIG. 3 shows three sub-modules, but depending on the embodiment, the processor (230) may include fewer than three or more than three sub-modules.

[0061] In addition, since the preprocessing module (231), score calculation module (233), and risk calculation module (235) of FIG. 3 are sub-modules of the processor (230), they can be implemented using at least one of ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), controllers, micro-controllers, microprocessors, and other electrical units for performing functions, just like the processor (230).

[0062] First, the preprocessing module (231) can perform preprocessing on the first to fourth data when the risk calculation device (200) receives from the user terminal (300) the first data measured by the blood glucose meter (101), the second data of non-blood glucose measured by the wearable device (102), the third data which is the user's body data input into the user terminal (300), and the fourth data which is the user's health checkup data. Since the first data, second data, third data, and fourth data after preprocessing are further highlighted with the unique characteristics of each data, the preprocessing module (231) can be seen as operating as a feature extraction layer in the risk calculation device (200).

[0063] For example, the first data is merely a measurement value (blood glucose value) received from a blood glucose meter (101), but the first data after preprocessing is data representing the user's blood glucose variability, and at least one of MAGE, CV, and SD can be produced as feature data. The second data is merely a biosignal collected by a wearable device (102), but the second data after preprocessing is merely a feature data, and at least one of the user's exercise intensity (MET, heart rate by zone, etc.), sleep rhythm (consistency, deep sleep, REM sleep, etc.), and stress (HRV: Heart Rate Variability, trend of change in heart rate variability) can be produced as feature data. Additionally, the third data is merely answers regarding meal history, sleep history, smoking status, drinking frequency, and number of weekday exercise days (number of times) entered by the user terminal (300), but the third data after preprocessing can be produced as a feature vector of the user's lifestyle pattern entered by the user. In addition, although the fourth data is merely health checkup data, the preprocessed fourth data may be the result of calculating an obesity index, such as BMI, by combining the user's height, gender, age, weight, and stature included in the health checkup data.

[0064] Next, the score calculation module (233) can calculate a score set including multiple scores for the preprocessed first data, second data, third data, and fourth data, and calculate a comprehensive score based on the individual scores included in the score set. The score calculation module (233) can be seen as operating as a score engine layer in the risk calculation device (200).

[0065] Hereinafter, the score set, individual scores included in the score set, and the total score are sequentially described as processing information calculated by the score calculation module (233). First, the score set refers to a unit that includes multiple scores as components, and more specifically, the score set includes scores for the preprocessed first data, second data, third data, and fourth data as components. For example, if the score set includes scores for the first to fourth data, it can be seen that the score set includes a total of four scores.

[0066] In another embodiment, the score set may include a total of six types of individual scores based on preprocessed first data, second data, third data, and fourth data. Here, the six types of individual scores refer to a health checkup score, a blood sugar score, a meal score, an exercise score, a sleep score, and a stress score, and can be calculated based on the preprocessed first data, second data, third data, and fourth data.

[0067]

[0068] Mathematical Formula 1 represents the formula used by the score calculation module (233) to calculate the health examination score. The score calculation module (233) can calculate the health examination score by subtracting from 100 the result of a weighted sum of the degree of deviation from the clinical normal range of the total 18 parameters (examination variables) constituting the health examination data described in FIG. 1. In Mathematical Formula 1, S check is the health checkup score, W k is a weight for each of the 18 parameters (by screening variable), which can be a parameter of the Cox model, and Value k is the actual measured value of 18 parameters, and Median k represents the mean value of the control group for the corresponding age / gender, and SDk represents the standard deviation of the control group for the corresponding age / gender.

[0069]

[0070] Mathematical formula 2 represents the formula used by the score calculation module (233) to calculate the blood glucose score. The score calculation module (233) can calculate the blood glucose score by applying respective weights to the TIR, CV, and MAGE obtained through the blood glucose meter (101) of FIG. 1 and summing them. In mathematical formula 2, S glucose represents the blood glucose score, TIR represents the Time In Range, CV represents the Coefficient of Variation, and MAGE represents the Mean Amplitude of Glycemic Excursions, respectively. Referring to Equation 2, it can be seen that a normalized value of MAGE is used to obtain a reliable range of blood glucose scores.

[0071]

[0072] Mathematical Formula 3 represents the formula used by the score calculation module (233) to calculate the meal score. The score calculation module (233) can calculate the meal score through the result of preprocessing life log data input through the user terminal (300) of FIG. 1. Here, the meal score is a value obtained by combining the food name, food quantity, calories, carbohydrate / fat / protein composition, meal time, etc. The food name, food quantity, calories, and carbohydrate / fat / protein composition consumed by the user can be obtained by the user directly inputting numerical values ​​into the user terminal (300), or by identifying the type and quantity of food through the result of an AI model analyzing the uploaded photo if the user uploads a photo of the food. In Mathematical Formula 3, S meal is the dietary score, GI k CA is the glycemic index of consumed food, an indicator of the speed at which food raises the user's blood sugar. k represents the total amount of net carbohydrates consumed by the user in grams (g), RDA is the user's recommended dietary allowance, and SF is a scale factor that is dynamically adjusted according to the individual's basal metabolic rate (BMR) and activity level.

[0073]

[0074] Mathematical formula 4 represents the formula used by the score calculation module (233) to calculate an exercise score. The score calculation module (233) can calculate an exercise score through the result of preprocessing the biosignal obtained through the wearable device (102) of FIG. 1. In mathematical formula 4, S exercise is the exercise score, W zonek is a range weight, a weight for blood glucose consumption efficiency according to heart rate, MET k is the metabolic rate, and the corresponding exercise intensity coefficient, T krepresents the duration of exercise in each exercise intensity range (in minutes), and TAI represents the daily recommended energy expenditure index set by considering the user's age, weight, and basal metabolic rate as the target activity index.

[0075]

[0076] Mathematical formula 5 represents the formula used by the score calculation module (233) to calculate a sleep score. The score calculation module (233) can calculate a sleep score through the result of preprocessing the biosignal obtained through the wearable device (102) of FIG. 1. In mathematical formula 5, S sleep Efficiency is the value obtained by dividing the actual sleep time (in minutes) measured by the wearable device (102) by the time the user stayed in bed (in minutes) and then multiplying by 100, and Consistency is the standard score value for the time of sleep over the past 7 days.

[0077]

[0078] Mathematical Equation 6 represents the formula used by the score calculation module (233) to calculate the stress score. The score calculation module (233) can calculate the stress score through the result of preprocessing the biosignal obtained through the wearable device (102) of FIG. 1. In Mathematical Equation 6, S stress RMSSD (Root Mean Squarer of Successive Differences) is a value used to indicate the activity of the parasympathetic nervous system, and SDNN (Standard Deviation of NN Intervals) represents the standard deviation of heart rate intervals during the entire measurement period as the autonomic nervous system's regulatory ability and resilience.

[0079]

[0080] Mathematical formula 7 represents a formula used by the score calculation module (233) to calculate a comprehensive score using the health checkup score, blood sugar score, meal score, exercise score, sleep score, and stress score as individual scores. In mathematical formula 7, S total Eun is the total score, W k is the weight for each individual score, S k represents the health checkup score, blood sugar score, diet score, exercise score, sleep score, and stress score, respectively. In addition, Table 1 explains the weights for each individual score.

[0081]

[0082] As shown in Equation 7 and Table 1, the composite score can be calculated by differentially assigning weights to each item affecting blood glucose based on its contribution to the prevention of diabetes and complications. The risk calculated based on the composite score is not merely a list of information for the user, but rather information intended to provide intuitive management indicators by generating a unified indicator through Weighted Integration based on clinical significance. In Table 1, the weights for each individual score may be changed according to pre-set logic.

[0083] Next, the risk calculation module (235) can calculate data on the risk of developing diabetes and diabetes-related complications for the user based on the comprehensive score calculated by the score calculation module (233). Specifically, the comprehensive score is applied to the probability of developing diabetes and complications by year by the risk calculation module (235) and used as a dynamic variable, and the user can be provided with simulation results in which the probability of developing diabetes and diabetes-related complications decreases in real time whenever the user takes action to manage blood sugar to an appropriate level.

[0084]

[0085] Mathematical formula 8 represents the formula used by the risk calculation module (235) to calculate the incidence risk data. In mathematical formula 8, Risk f (t) is the disease risk data, Risk cox (t) is the probability of developing diabetes / diabetes-related complications within t years calculated based on health checkup data (18 parameters) using a Cox model, f(S total ) is the total score S in the correction function f(x). total As a result of applying [it], each represents an exponential function correction coefficient that increases or decreases the risk level according to the overall score.

[0086]

[0087] Equation 9 represents the exponential correction coefficient based on the composite score explained in Equation 8. The correction function is applied to reflect changes in the risk of developing diabetes and related complications according to changes in the composite score variable. In Equation 9, K represents the reference point, and 50 points may be set as the neutral point for correction. Additionally, in Equation 9, a represents the sensitivity coefficient, which is a constant that controls the influence of changes in the user's lifestyle pattern (lifestyle habits) on the actual risk of developing diabetes / complications. Depending on clinical significance, a value selected by a predetermined criterion from among values ​​between 0.2 and 0.5 may be applied for the sensitivity coefficient.

[0088] The correction function is a model that reflects the clinical impact of real-time lifelogs through a non-linear correction model by exponentially attenuating the risk when the comprehensive score exceeds the reference value of 50 points, and exponentially amplifying the risk when it falls below 50 points. The user can directly check the risk corrected by lifestyle patterns through the user terminal (300) and experience the impact of the current lifestyle pattern on the probability of future disease at a percentage level. However, lifestyle habits alone do not correspond to the Risk corresponding to the base risk coxBecause there is a limit to the correction effect, the risk calculation module (235) in the present invention limits the maximum correction range of the correction function to 25% to ensure the stability of the model. In particular, when the meal score or blood sugar score drops sharply, a risk-avoiding prevention logic can be implemented by applying asymmetric weights that are set to increase the risk more sensitively than the increase in other scores.

[0089]

[0090] Table 2 is an exemplary table showing the case where K is 50 and a is 0.4 in the correction function according to Equation 9. Referring to Table 2, when the user's overall score is between 0 and 29, the value of the correction function ranges from 1.09 to 1.22; accordingly, the Risk corresponding to the final disease risk data f It can be seen that the incidence rate increases by 9% to 22%.

[0091] FIG. 4 is a block diagram illustrating an example of a user terminal according to the present invention.

[0092] Referring to FIG. 4, it can be seen that the user terminal (300) includes a terminal communication unit (310), a terminal processor (330), and a terminal memory (350).

[0093] The terminal communication unit (310) may include one or more components that enable wired / wireless communication with an external device. For example, the terminal communication unit (310) may include at least one piece of hardware necessary to implement short-range communication such as Wi-Fi or Bluetooth in a network provided by the communication network (20), or to implement various communications including the Internet when a LAN cable is connected.

[0094] The terminal memory (350) is hardware that stores various data processed within the user terminal (300) and can store a program for processing and controlling the terminal processor (330). The terminal memory (350) may include RAM (random access memory) such as DRAM (dynamic random access memory) and SRAM (static random access memory), ROM (read-only memory), EEPROM (electrically erasable programmable read-only memory), CD-ROM, Blu-ray or other optical disc storage, HDD (hard disk drive), SSD (solid state drive), or flash memory.

[0095] The terminal processor (330) can control the overall operation of the user terminal (300). For example, the terminal processor (330) can control the operation of the terminal input unit (not shown), terminal display (not shown), terminal communication unit (310), terminal memory (350), etc. included in the user terminal (300) by executing programs stored in the terminal memory (350).

[0096] As an example, the terminal processor (330) can generate third data based on user input, transmit it to the risk level calculation device (200) together with the first data, second data, and fourth data collected from external devices (100), and receive display data from the risk level calculation device (200) and process the data to display it through the terminal display (not shown) of the user terminal (300).

[0097] When the user terminal (300) is implemented as a physical device, the terminal processor (330) may be implemented using at least one of ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), controllers, micro-controllers, microprocessors, and other electrical units for performing functions.

[0098] In addition, when the user terminal (300) in the present invention is implemented in the form of an application (program), the terminal processor (330) and terminal memory (350) included in the user terminal (300) may be implemented in the form of a virtual machine that implements hardware such as DSPs, microcontrollers, RAM, ROM, HDD, etc., as software (command script).

[0099] FIG. 5 is a diagram for explaining a submodule included in the terminal processor of FIG. 4.

[0100] Referring to FIG. 5, it can be seen that the terminal processor (330) includes a data input processing module (331) and a data display processing module (333). Since the data input processing module (331) and the data display processing module (333) illustrated in FIG. 5 are modules that are logically and conceptually separated to explain the process performed by the terminal processor (330) in the process of implementing the method according to the present invention, FIG. 5 shows two sub-modules, but depending on the embodiment, the terminal processor (330) may include fewer than two or more than two sub-modules. In addition, since the data input processing module (331) and data display processing module (333) of FIG. 5 are sub-modules of the processor (230), they can be implemented using at least one of ASICs, DSPs, DSPDs, PLDs, FPGAs, controllers, microcontrollers, microprocessors, and other electrical units for performing functions, just like the terminal processor (330).

[0101] The data input processing module (331) can detect the second health checkup data input through the input section of the user terminal and process it into a state that can be transmitted to the risk calculation device (200). According to an embodiment, the data input processing module (331) may also perform the function of providing guidelines so that the user does not omit items or input incorrect figures in accordance with a predetermined format.

[0102] When the risk of occurrence data processed by the risk calculation device (200) is converted into display data and received, the data display processing module (333) reads the received display data and processes and displays the probability of occurrence of diabetes and diabetes-related complications in a data form that the user can recognize.

[0103] FIG. 6 is a flowchart illustrating an example of a method according to the present invention.

[0104] Since the method according to FIG. 6 can be implemented by the risk calculation device (200) and the sub-module included in the risk calculation device (200) described in FIG. 1 to 3, it will be explained with reference to FIG. 1 to 3, and explanations that overlap with already explained content will be omitted.

[0105] The risk calculation device (200) can receive and preprocess first data measured by a blood glucose meter (101) from a user terminal (300), second data of non-blood glucose measured by a wearable device (102), third data which is the user's body data input to the user terminal (300), and fourth data which is the user's health checkup data (S610).

[0106] The risk calculation device (200) can calculate a score set including multiple scores for preprocessed first data, second data, third data and fourth data (S630).

[0107] The risk calculation device (200) can calculate a total score based on individual scores included in the calculated score set (S650).

[0108] The risk calculation device (200) can calculate the risk of developing diabetes and diabetes-related complications for the user based on the calculated total score (S670).

[0109] According to an embodiment, the risk calculation device (200) can convert the disease risk data calculated in step S670 into display data that can be displayed in levels and % units on the user terminal (300) and transmit it to the user terminal (300), and the user can visually recognize the information regarding diabetes and diabetes-related complications displayed on the user terminal (300) and check the likelihood of the user developing diabetes and diabetes-related complications.

[0110] According to the present invention, the long-term risk of developing diabetes and diabetes-related complications can be calculated using various data including a user's health checkup data and provided to the user.

[0111] The embodiments according to the present invention described above may be implemented in the form of a computer program that can be executed through various components on a computer, and such a computer program may be recorded on a computer-readable medium. In this case, the medium may include a magnetic medium such as a hard disk, a floppy disk, and a magnetic tape, an optical recording medium such as a CD-ROM and a DVD, a magneto-optical medium such as a floptical disk, and a hardware device specifically configured to store and execute program instructions, such as a ROM, RAM, or flash memory.

[0112] Meanwhile, the above-mentioned computer program may be one specifically designed and configured for the present invention, or one known and available to those skilled in the art of computer software. Examples of computer programs may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0113] The specific embodiments described in this invention are examples and do not limit the scope of the invention in any way. For the sake of brevity of the specification, descriptions of prior electronic configurations, control systems, software, and other functional aspects of said systems may be omitted. Additionally, the connections of lines or connecting members between components shown in the drawings are illustrative of functional connections and / or physical or circuit connections, and may be replaced or additionally represented as various functional connections, physical connections, or circuit connections in actual devices. Furthermore, unless specifically stated as “essential,” “importantly,” etc., a component may not be strictly necessary for the application of the invention.

[0114] In the specification of the present invention (particularly in the claims), the use of the term “the above” and similar descriptive terms may be in both singular and plural. Furthermore, where a range is described in the present invention, it is to include the invention to which individual values ​​belonging to said range are applied (unless otherwise stated), and is equivalent to describing each individual value constituting said range in the detailed description of the invention. Finally, regarding the steps constituting the method according to the present invention, unless explicitly stated or otherwise stated, said steps may be performed in a suitable order. The present invention is not necessarily limited by the order in which said steps are described. The use of all examples or exemplary terms (e.g., etc.) in the present invention is merely for the purpose of describing the present invention in detail, and the scope of the present invention is not limited by said examples or exemplary terms unless limited by the claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and changes may be made according to design conditions and factors within the scope of the claims or equivalents to which they are added. Explanation of the symbols

[0115] 1: Diabetes-Related Disease Risk Calculation System 20: Communication network 100: External device 200: Risk calculation device 300: User terminal

Claims

Claim 1 A step of receiving and preprocessing first data measured by a blood glucose meter from a user terminal, second non-blood glucose data measured by a wearable device, third data which is the user's body data input into the user terminal, and fourth data which is the user's health checkup data; a step of calculating a score set including a plurality of scores for the preprocessed first data, second data, third data, and fourth data; and a step of calculating a comprehensive score based on individual scores included in the calculated score set.The method includes a step of calculating data on the risk of developing diabetes and diabetes-related complications for the user based on the comprehensive score calculated above, wherein the step of calculating the score set calculates a score set including a health checkup score, blood sugar score, meal score, exercise score, sleep score, and stress score based on the first data, the second data, the third data, and the fourth data, and the step of calculating the comprehensive score calculates the comprehensive score by applying a unique weight set for each of the health checkup score, blood sugar score, meal score, exercise score, sleep score, and stress score, wherein the comprehensive score is calculated using an asymmetric weight setting method in which the weight applied to the meal score and the blood sugar score is set larger than the weight applied to the remaining scores so that when the meal score or the blood sugar score drops below a preset standard, the comprehensive score is calculated higher than the increase of the remaining scores, and the risk of developing data comprises a first model that processes the fourth data to produce static data and a dynamic data model that processes the first data, the second data, and the third data. A multi-data-based method for calculating the risk of diabetes-related diseases, wherein a second model is calculated by ensembling the first model, the first model is a Cox model that calculates general risk using static data, and the second model is a correction function model that calculates a correction value using the calculated comprehensive score as an individual correction factor, and includes a reference value which is a neutral point of correction and a sensitivity coefficient that quantifies the influence of changes in the user's lifestyle habits on the risk of developing diabetes and diabetes-related complications, wherein the reference value is set to 50, the sensitivity coefficient is set to a value selected by a predetermined standard among values ​​between 0.2 and 0.5, and the correction value has an upper limit of 25%. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete Claim 7 delete Claim 8 In claim 1, the step of calculating the incidence risk data further calculates the incidence risk data using at least one of an XGboost survival model and a random survival forest model, a multi-data-based method for calculating the risk of diabetes-related diseases. Claim 9 A method for calculating risk of diabetes-related diseases based on multiple data, wherein, in claim 1, the method further comprises the step of converting the calculated disease risk data into display data that can be displayed on the user terminal and transmitting it to the user terminal. Claim 10 A computer-readable recording medium storing a program for executing the method according to paragraph 1. Claim 11 Memory where at least one program is stored; The apparatus includes a processor that performs operations by executing at least one program, wherein the processor receives and preprocesses first data measured by a blood glucose meter from a user terminal, second non-blood glucose data measured by a wearable device, third data which is the user's body data input to the user terminal, and fourth data which is the user's health checkup data, and calculates a score set including a plurality of scores for the preprocessed first data, second data, third data, and fourth data, calculates a comprehensive score based on individual scores included in the calculated score set, and calculates data on the risk of developing diabetes and diabetes-related complications for the user based on the calculated comprehensive score, and the processor calculates a score set including a health checkup score, blood glucose score, meal score, exercise score, sleep score, and stress score based on the first data, second data, third data, and fourth data, and applies a unique weight set for each of the health checkup score, blood glucose score, meal score, exercise score, sleep score, and stress score. Calculate a comprehensive score using an asymmetric weighting method in which the weights applied to the meal score and the blood sugar score are set larger than the weights applied to the remaining scores, so that when the meal score or the blood sugar score decreases by exceeding a preset standard, the comprehensive score is calculated to be higher than the increase of the remaining scores; calculate the disease risk data by ensembling a first model that processes the fourth data to produce static data and a second model that processes the first data, the second data, and the third data to produce dynamic data; wherein the first model is a Cox model that calculates general risk using the static data, and the second model is a correction function model that calculates a correction value using the calculated comprehensive score as an individual correction factor.A multi-data-based diabetes-related disease risk calculation device comprising a reference value which is a neutral point for correction and a sensitivity coefficient that quantifies the influence of changes in the user's lifestyle habits on the risk of developing diabetes and diabetes-related complications, wherein the reference value is set to 50, the sensitivity coefficient is set to a value selected by a predetermined standard among values ​​between 0.2 and 0.5, and the correction value has an upper limit of 25%. Claim 12 delete Claim 13 delete Claim 14 delete Claim 15 delete

Images