WO2024144337A1

WO2024144337A1 - Diabetes management service method and device

Info

Publication number: WO2024144337A1
Application number: PCT/KR2023/021958
Authority: WO
Inventors: 김대훈; 서인근; 박성호; 메데이로스 루카피멘타
Original assignee: 주식회사 누비랩
Priority date: 2022-12-30
Filing date: 2023-12-29
Publication date: 2024-07-04

Abstract

The present invention relates to a diabetes management service method and device. The diabetes management service method according to one embodiment of the present invention comprises the steps of: receiving meal data and blood glucose measurement data of a user; identifying a meal event by using the received meal data and/or blood glucose measurement data; identifying at least one food from the received meal data and integrating same into the identified meal event; mapping, to blood glucose measurement data for each meal event, the blood glucose measurement data corresponding to the integrated meal event; using the mapped blood glucose measurement data for each meal event and a blood glucose prediction model trained in advance, so as to calculate the blood glucose level of the user and predict a blood glucose change; and outputting the calculated blood glucose level and the predicted blood glucose change.

Description

Diabetes management service method and device

The present invention relates to diabetes management technology, and more specifically, to diabetes management service technology for understanding and predicting the interaction between food and blood sugar, beyond simply recording meal data.

Diabetes refers to a condition in which the level of glucose in the blood is higher than normal. It is a disease named because glucose, which should be used as energy in our body, is excreted through urine. Diabetes occurs when sugar floats around in the blood without being broken down. When you eat food, it is digested in the gastrointestinal tract, converted into glucose, and absorbed into the blood. Glucose that enters the blood is called blood sugar, and blood sugar is moved to cells and used as energy needed for our activities. In order for blood sugar to enter cells and be used as energy, a hormone called insulin, produced by the pancreas, is required. Diabetes is when insulin is not secreted normally or does not function properly even when secreted. As a result, glucose that cannot enter the cells continues to accumulate in the blood, causing hyperglycemia.

Diabetes is caused by genetics, viruses, obesity, aging, incorrect eating habits, stress, and drugs, and aging can lead to a decline in cellular function in the body. In countries with higher per capita income, diabetes is more likely to occur, and continued stress can interfere with cortisol secretion and insulin action. When people with genetic factors for diabetes take medications such as diuretics, oral contraceptives, and thyroid hormones, there is a high risk of developing or worsening diabetes. Diabetes is known to cause temporary elevations in blood sugar levels to awaken genes related to cancer, and is known to increase oxidative stress, thereby increasing the risk of developing heart disease. Therefore, to manage diabetes, managing postprandial blood sugar (Postprandial glucose) is becoming important.

The goal of controlling and maintaining blood sugar is important to manage the nutrient called carbohydrates. However, carbohydrates are essential nutrients. If you consume extremely limited carbohydrates while eating a high-protein and high-fat diet, your blood sugar level may improve, but normal metabolic processes cannot occur, resulting in ketoacidosis or adverse health effects. Therefore, carbohydrate intake must account for 40-60% of total calories consumed.

Accordingly, diabetes can be appropriately managed if an appropriate amount of carbohydrate intake is suggested. For example, if a user simply eats an appropriate amount of rice to manage diabetes, that is, by controlling the amount of rice properly, the postprandial blood sugar level can be controlled. Rice is easy to identify, and rice has a uniform composition in most cases. To maximize the effectiveness of diabetes management services, rice is the main source of carbohydrates in the Korean diet. Energy intake was significantly lower for women than men at 2,125 kcal for men and 1,591 kcal for women, and the macronutrient intake ratio (%) for men was 68.7:16.4:14.9 for carbohydrates: fat: protein, and 71.0:15.1 for women. It was 13.8. Therefore, there is an urgent need to suggest an appropriate amount of carbohydrate intake for diabetes management services.

Embodiments of the present invention provide a diabetes management guide that displays recommended intake based on user data such as meal scan data and blood sugar measurement data and blood sugar predicted through a blood sugar prediction model, thereby helping the user with diabetes by consuming an appropriate amount of carbohydrates. We aim to provide diabetes management service methods and devices that can easily manage diabetes.

Embodiments of the present invention use meal data to accurately classify food, integrate it into meal events, and connect integrated meal events with blood sugar changes, so that they go beyond simply recording meal data and understand the interaction between food and blood sugar. We aim to provide predictable diabetes management service methods and devices.

However, the problem to be solved by the present invention is not limited to this, and may be expanded in various environments without departing from the spirit and scope of the present invention.

According to an embodiment of the present invention, a diabetes management service method provided by a diabetes management service device, comprising: receiving meal data and blood sugar measurement data of a user; Identifying a meal event using at least one of the received meal data and blood sugar measurement data; identifying at least one food from the received meal data and incorporating it into the identified meal event; mapping blood sugar measurement data corresponding to the integrated meal event into blood sugar measurement data for each meal event; calculating the user's blood sugar level and predicting changes in blood sugar using the mapped blood sugar measurement data for each meal event and a pre-learned blood sugar prediction model; and outputting the calculated blood sugar level and the predicted blood sugar change. A diabetes management service method may be provided.

The meal data may include at least one of images, text, and metadata related to the user's meal.

In the step of integrating into the identified meal event, at least one key value classified by food may be extracted from the received meal data, and at least one food may be identified using the extracted at least one key value.

The step of identifying the meal event includes at least one of the following: when meal data including an image is received, when meal data including text is received, and when the user's blood sugar level exceeds a preset blood sugar level value. can be set as the start point of the meal event.

The method includes calculating an evaluation index for evaluating a blood sugar response to the identified meal event using an average blood sugar value of the meal event, and evaluating the blood sugar response for each meal event through the calculated evaluation index. It may further include.

The method further includes normalizing the mapped blood sugar measurement data for each meal event by considering the blood sugar level for each user, and comparing the blood sugar response of the user group in a standardized manner through the normalized blood sugar measurement data for each meal event. It can be included.

The method may further include recommending at least one of meals, meal order, and meal speed personalized for each theme to the user using the at least one extracted key value.

In the step of recommending a meal, elements that are not suitable for meal recommendation may be processed as exceptions using the extracted at least one key value, and the meal may be recommended by excluding the exceptions.

The method uses external data containing at least one of personal information, health checkup data, and public institution data to create a cohort of other users with a preset similarity or more to the user, and collects blood sugar response data of the created cohort. A step of analyzing and providing customized blood sugar management recommendations to the user may be further included.

Meanwhile, according to another embodiment of the present invention, a communication module; memory that stores one or more programs; and a processor executing the one or more stored programs, wherein the processor receives the user's meal data and blood sugar measurement data, and identifies a meal event using at least one of the received meal data and blood sugar measurement data. , Identifying at least one food from the received meal data and integrating it into the identified meal event, mapping blood sugar measurement data corresponding to the integrated meal event to blood sugar measurement data for each meal event, and mapping the mapped meal event to blood sugar measurement data for each meal event. A diabetes management service device may be provided that calculates the user's blood sugar level and predicts blood sugar changes using star blood sugar measurement data and a pre-learned blood sugar prediction model, and outputs the calculated blood sugar level and the predicted blood sugar change. there is.

The processor may extract at least one key value classified by food from the received meal data, and cook at least one meal using the extracted at least one key value.

The processor starts a meal event in at least one of the following cases: when meal data including an image is received, when meal data including text is received, and when the user's blood sugar level exceeds a preset blood sugar level value. It can be set to any point in time.

The processor may calculate an evaluation index for evaluating a blood sugar response for the identified meal event using the average blood sugar value of the meal event, and evaluate the blood sugar response for each meal event through the calculated evaluation index. .

The processor may normalize the mapped blood sugar measurement data for each meal event by considering the blood sugar level for each user, and compare the blood sugar responses of user groups in a standardized manner through the normalized blood sugar measurement data for each meal event.

The processor may recommend at least one of a meal, meal order, and meal speed personalized for each theme to the user using the at least one extracted key value.

The processor may use the extracted at least one key value to process elements that are not suitable for meal recommendation as exceptions and recommend a meal by excluding the exceptions.

The processor creates a cohort of other users with a preset similarity or higher to the user using external data containing at least one of personal information, health examination data, and public institution data, and provides blood sugar response data of the created cohort. A step of analyzing and providing customized blood sugar management recommendations to the user may be further included.

The disclosed technology can have the following effects. However, since it does not mean that a specific embodiment must include all of the following effects or only the following effects, the scope of rights of the disclosed technology should not be understood as being limited thereby.

Embodiments of the present invention provide a diabetes management guide that displays recommended intake based on user data such as meal scan data and blood sugar measurement data and blood sugar predicted through a blood sugar prediction model, thereby helping the user with diabetes by consuming an appropriate amount of carbohydrates. can be easily managed.

Embodiments of the present invention allow users to easily manage diabetes by consuming an appropriate amount of carbohydrates, thereby reducing the intake of an appropriate amount of carbohydrates (e.g., rice, porridge, noodles, bread, etc.) from a health management perspective, and diabetic patients. The pre-stage population can be converted into a normal blood sugar population. Through this, embodiments of the present invention can help create healthy lifestyle habits.

Embodiments of the present invention use meal data to accurately classify food, integrate it into meal events, and connect integrated meal events with blood sugar changes, so that they go beyond simply recording meal data and understand the interaction between food and blood sugar. It is predictable.

Embodiments of the present invention can more accurately record and analyze blood sugar changes related to a meal by setting the start point of the meal event.

Embodiments of the present invention can cut blood sugar measurement data for each meal event and map it to each meal event.

Embodiments of the present invention can establish the connection between blood sugar data and meal events more accurately and flexibly.

Embodiments of the present invention can ensure the accuracy of meal data and, as a result, provide more accurate information in analyzing the relationship between blood sugar measurement data and food.

In embodiments of the present invention, when a user takes a photo of a meal or inputs food-related text, the photo is analyzed and a blood sugar prediction result is provided to show a predicted value of how the meal will affect blood sugar.

Embodiments of the present invention can provide information about blood sugar response to the user in real time, and the user can make healthy meal choices based on this.

Embodiments of the present invention provide users with immediate feedback on blood sugar management and can help form healthy eating habits in the long term.

Embodiments of the present invention can show the overall blood sugar response and the differences between different groups through blood sugar conversion rates.

Embodiments of the present invention provide personalized meal recommendations, and can identify and exclude unhealthy elements from the recommendations.

Embodiments of the present invention can provide personalized blood sugar management recommendations by utilizing external information such as health checkup data.

Figure 1 is a diagram showing the configuration of a diabetes management service system according to an embodiment of the present invention.

Figure 2 is a diagram showing the operation of a diabetes management service according to an embodiment of the present invention.

Figure 3 is a flowchart showing a diabetes management service method according to an embodiment of the present invention.

Figures 4 to 6 are diagrams showing examples of diabetes management service operations according to an embodiment of the present invention.

Figure 7 is a configuration diagram of a diabetes management service device according to an embodiment of the present invention.

Figure 8 is a diagram showing the operation of a diabetes management service device according to another embodiment of the present invention.

Figure 9 is a flowchart of a diabetes management service method according to another embodiment of the present invention.

Figure 10 is a diagram showing an example of a food data classification operation by a diabetes management service device according to another embodiment of the present invention.

Figures 11 and 12 are diagrams showing an example of a blood sugar index and an example of data normalization by a diabetes management service device according to another embodiment of the present invention.

Since the present invention can be modified in various ways and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and may be understood to include all transformations, equivalents, and substitutes included in the technical idea and scope of the present invention. In describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted.

Terms such as first, second, etc. may be used to describe various components, but the components are not limited by the terms. Terms are used only to distinguish one component from another.

The terms used in the present invention are only used to describe specific embodiments and are not intended to limit the present invention. The terms used in the present invention are general terms that are currently widely used as much as possible while considering the function in the present invention, but this may vary depending on the intention of a person skilled in the art, precedents, or the emergence of new technology. In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the relevant invention. Therefore, the terms used in the present invention should be defined based on the meaning of the term and the overall content of the present invention, rather than simply the name of the term.

Singular expressions include plural expressions unless the context clearly dictates otherwise. In the present invention, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that it does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description with reference to the accompanying drawings, identical or corresponding components will be assigned the same drawing numbers and overlapping descriptions thereof will be omitted. do.

As shown in Figure 1, the diabetes management service system 10 according to an embodiment of the present invention includes a user terminal 101, a blood sugar meter 102, a health device 103, a diabetes management service device 100, and Includes a medical data server 200. However, not all of the illustrated components are essential components. The diabetes management service system 10 may be implemented with more components than the components shown, or the diabetes management service system 10 may be implemented with fewer components.

Hereinafter, the specific configuration and operation of each component of the diabetes management service device 100 of FIG. 1 will be described.

The blood sugar meter 102 is attached to the user's body and measures the user's blood sugar using a blood collection method or a non-invasive method. In other words, a blood glucose meter uses the principle of directly measuring glucose contained in blood, while a non-invasive blood glucose meter measures glucose contained in the user's sweat rather than blood, but the glucose level in sweat is proportional to blood sugar. The principle is to measure blood sugar indirectly using glucose in sweat.

If the blood glucose meter 102 is non-invasive, it may be provided in the form of a wearable device such as a patch-type sensor attached to the body, a wristwatch-type sensor, or a clothing-type sensor. When a predetermined measurement start command is input or the operation of the imaging module is performed, the blood glucose meter 102 may be provided in the form of a wearable device. When this is detected, blood sugar measurement begins. The blood sugar meter 102 can measure blood sugar periodically and continuously for a predetermined period of time from the time of food intake, either manually or automatically. If the blood glucose meter 102 is a manual method, it may provide means such as a special alarm sound or vibration for the start time, measurement cycle, and end time to encourage the user to operate the buttons. The blood sugar meter 102 is connected to the user terminal 101 through short-distance wireless communication such as Bluetooth, Wi-Fi, or Zigbee.

The health device 103 may monitor user health information, such as exercise information, sitting time, and number of steps, and transmit it to the user terminal 101. The health device 103 measures physical activity or exercise data (e.g., activity time, duration according to activity type such as walking, running, swimming, activity intensity such as low, medium, high, etc.) by measuring the user terminal (101). ) can be passed on. The health device 103 may obtain cardiovascular data, respiration data, body temperature data (eg, skin temperature data), sleep data, stress level data, etc. and transmit them to the user terminal 101 .

In the above, the blood sugar meter 102 and the health device 103 were described separately from a functional perspective, but the blood sugar meter 102 may be built into the health device 103, or the blood sugar meter and the health device 103 may be implemented as one product. It may be possible.

The user terminal 101 photographs the food the user wants to eat and uses the meal scan data including the photographed food image, the blood sugar measurement data received from the blood glucose meter 102, and the user status data received from the health device 103 for diabetes. It can be transmitted to the management service device 100 and provide a blood sugar change trend when consuming the food based on the blood sugar received from the diabetes management service device 100. And when a dangerous situation is expected based on the provided blood sugar change trend, the user terminal 101 sends the diabetes management guide received from the diabetes management service device 100 through the user terminal 101 to guide the user on appropriate action tips. It can be provided to the user.

The medical data server 200 collects and manages the user's medical my data. The medical data server 200 may transmit the managed user's medical my data to the diabetes management service device 100 according to the medical data request. Medical MyData includes medical history or health data (e.g. weight, age, sleep patterns, health conditions, cholesterol levels, type of diabetes, family history, patient health history, diagnoses, blood pressure, etc.), user personal data (e.g. For example, name, gender, demographics, social network information, etc.), medication data (e.g., timing, dosages of drugs such as insulin), etc.

The diabetes management service device 100 receives meal scan data, blood sugar measurement data, and user state data from the user terminal 101, and provides blood sugar image index information using the received meal scan data, blood sugar measurement data, and user state data. Calculate The diabetes management service device 100 receives user medical data from the medical data server 200 and generates user health management information using the received user medical data. The diabetes management service device 100 predicts blood sugar by inputting it into a pre-learned blood sugar prediction model using the calculated blood sugar image index information and the generated user health management information, and displays the recommended intake amount based on the predicted blood sugar. Provides diabetes management guide.

To this end, the diabetes management service device 100 may learn the trend of blood sugar changes for the user each time the user eats food in conjunction with the user terminal 101. The diabetes management service device 100 photographs the food the user wants to eat through the user terminal 101 and specifies the name of the food from the photographed meal scan data. And the diabetes management service device 100 receives and records blood sugar measurement data after the user consumes food. The diabetes management service device 100 may receive blood sugar measurement data for a predetermined period of time from the time of food intake and record changes in the user's blood sugar according to food intake. The diabetes management service device 100 may receive blood sugar measurement data immediately before food intake. This is to predict the reference point and amount of change in the user's blood sugar change when predicting the user's blood sugar change trend. Additionally, the diabetes management service device 100 may learn the food intake amount derived from a secondary photograph of the end-of-meal image after a meal, in addition to the food name. That is, the diabetes management service device 100 can learn in more detail the blood sugar change trend according to the final intake of the food, in addition to the blood sugar change trend according to the intake of a specific food. The diabetes management service device 100 may learn a blood sugar prediction model using this blood sugar change trend. Here, the blood sugar prediction model may be any one of a supervised learning model using the Boltzmann machine algorithm, a linear regression analysis model, a plurality of sets of machine learning models, and a metabolic model.

Thereafter, the diabetes management service device 100 may assume that the user consumes the photographed food and predict the user's blood sugar change trend according to the intake through a pre-learned blood sugar prediction model. The diabetes management service device 100 may provide a diabetes management guide that displays a recommended intake amount to the user when the predicted change in blood sugar exceeds the user's preset threshold range. Diabetes management guides may include guidelines such as limiting the intake of certain ingredients in the food, reducing the overall intake of the food, consuming supplementary foods to accompany the food, or suggesting exercise after the meal. .

According to embodiments, the diabetes management service device 100 may provide a push notification for diabetes management through the user terminal 101. For example, the user terminal 101 may output a notification of the blood sugar measurement time (eg, 30 minutes, 1 hour, 2 hours) through the blood sugar meter 102. Additionally, the user terminal 101 may provide a push alarm on/off menu for each function. The user terminal 101 may provide a push alarm for a suggestion to start exercise after a meal in conjunction with the health device 103.

According to embodiments, the diabetes management service device 100 may provide a meal navigator function in conjunction with the user terminal 101. For example, the diabetes management service device 100 may induce meal records by linking with the user terminal 101, such as “What are you eating for lunch today?” or “Don’t forget to record it with a photo.” The diabetes management service device 100 can measure meal start and end times in conjunction with the health device 103 and the user terminal 101. Additionally, the diabetes management service device 100 may measure the type of meal the user eats through meal scan data. The diabetes management service device 100 may measure carbohydrate, protein, and fat nutrients per meal from meal scan data. The diabetes management service device 100 may use meal scan data, blood sugar measurement data, and user status data to provide an estimated blood sugar level increase after a certain period of time (eg, 2 hours) from the start of a meal. The diabetes management service device 100 may use additional meal scan data including leftover food at the end of a meal to provide a corrected expected blood sugar increase after a certain period of time (eg, 2 hours). When the user takes a picture of food at the start of a meal, the user terminal 101 calculates a recommended meal duration and provides “time remaining to become healthy” as a timer. When a photo of a meal is taken at the start of a meal, the user terminal 101 may display a space or a camera button to encourage taking a photo after the meal is over. When the camera is activated, the user terminal 101 may display at least one food area representing the area where the food is located on the capture screen in real time and guide the user to take a photo including the food area.

According to embodiments, the diabetes management service device 100 may store the user's blood sugar measurement record using blood sugar measurement data measured through the blood sugar meter 102. Additionally, the diabetes management service device 100 may calculate blood sugar measurement data and provide an analysis report to the user terminal 101. Additionally, the diabetes management service device 100 may predict the user's glycated hemoglobin based on blood sugar measurement data and a blood sugar prediction model.

According to embodiments, even if the user is measuring blood sugar using a continuous blood sugar meter, blood sugar level at 30 minutes, 1 hour, or 2 hours can be measured and recorded using the blood sugar meter 102.

According to embodiments, the diabetes management service device 100 may link with the user terminal 101 to deliver specialized information about the diabetes management service to the user.

According to embodiments, the diabetes management service device 100 may provide a diabetes management service by linking the user terminal 101 with a community for the diabetes management service. For example, the diabetes management service device 100 connects other users' diet introduction blogs, etc., such as "xxx, who controls blood sugar well, prepared breakfast like this," or analyzes other users' data similar to blood sugar measurement data. Results can be linked. The diabetes management service device 100 introduces other users' exercise plans, such as "xxx, who overcame blood sugar through exercise, exercise like this," or provides data-based analysis results of other users similar to the user's blood sugar measurement data. It can be connected.

According to embodiments, the diabetes management service device 100 may provide a regular delivery service including standard food and measurement records that may be helpful in diabetes management in connection with an external regular delivery service.

According to embodiments, the diabetes management service device 100 may provide diabetes management services in conjunction with a health device (eg, smart watch, etc.). For example, if the user sits down after taking a photo at the beginning of the meal, but detects that the user got up without taking a photo at the end of the meal, the diabetes management service device 100 takes the photo through the user terminal 101 or the health device 103. Provide push notifications for “Are you finished eating?” Or, you can encourage people to take pictures after eating, such as “Please take a picture of your meal after the meal.” Here, the way to detect the end of a user's meal is when the user takes a picture of the food at the beginning of the meal and sits for a while, but the user gets up for more than a certain period of time, sits but no longer hears conversation sounds, moves, exercises, etc. This may be a pattern for ending a meal. The diabetes management service device 100 detects this meal completion pattern in conjunction with the user terminal 101 or the health device 103, and when the meal completion is detected, the blood sugar measurement data at that time is reflected in the blood sugar prediction model to determine the blood sugar level. The forecast can be modified.

According to embodiments, the diabetes management service device 100 may obtain the user's medical my data from the medical data server 200 and predict blood sugar by reflecting personal characteristics in a blood sugar prediction model.

According to embodiments, the diabetes management service device 100 may reflect the estimated blood sugar level as 1/N of the intake when the user eats with N other users in the blood sugar prediction process.

According to embodiments, the user terminal 101 may collect blood sugar measurement data from a continuous glucose monitor (CGM) or a blood sugar meter 102 using wireless communication. The user terminal 101 can collect blood sugar measurement data through NFC communication or BLE communication with a continuous blood glucose meter (CGM), and can collect blood sugar measurement data through LTE communication or BLE communication with the blood sugar meter 102. .

According to embodiments, the diabetes management service device 100 may recommend a diet containing a recommended intake suitable for diabetes management services. For example, the diabetes management service device 100 may say, "This morning was too heavy. How about taking off the weight this time?" Alternatively, a diabetes management guide that displays the recommended amount of carbohydrates that can be consumed based on the predicted blood sugar level, such as "The carbohydrates consumed for a week is too low. How about adding weight?" can be provided through the user terminal 101.

The user terminal 101 sends meal scan data, which scans a meal scene containing food, blood sugar measurement data received from the blood sugar meter 102, and user status data received from the health device 103, to the diabetes management service device 100. send to Food type, food amount, meal time identification information, etc. can be recognized from meal scan data, blood sugar measurement data, and user status data. Blood sugar image index information may be calculated from meal scan data, blood sugar measurement data, and user status data and transmitted to the diabetes management service device 100.

The medical data server 200 may transmit user medical data, that is, user health management information, including the user's general data and user health data to the diabetes management service device 100. User medical data may include information that can identify an individual user, medical my data that can confirm the user's health, and health device information.

The diabetes management service device 100 predicts blood sugar by inputting it into a pre-learned blood sugar prediction model using blood sugar image index information and user health management information, and displays a diabetes management guide that displays the recommended intake amount based on the predicted blood sugar. Can be provided to the user terminal 101.

The user terminal 101 can provide blood sugar prediction data and diabetes management guide to the user through the diabetes management service app.

In step S101, the diabetes management service device 100 according to an embodiment of the present invention receives meal scan data, blood sugar measurement data, and user status data from the user terminal 101.

In step S102, the diabetes management service device 100 calculates blood sugar image index information using the received meal scan data, blood sugar measurement data, and user status data.

In step S103, the diabetes management service device 100 receives user medical data from the medical data server 200.

In step S104, the diabetes management service device 100 generates user health care information using the received user medical data.

In step S105, the diabetes management service device 100 predicts blood sugar by inputting the calculated blood sugar image index information and the generated user health management information into a pre-learned blood sugar prediction model.

In step S106, the diabetes management service device 100 provides a diabetes management guide that displays the recommended intake amount based on the predicted blood sugar level.

As shown in FIGS. 4 to 6, the diabetes management service device 100 may provide diabetes management services in conjunction with the user terminal 101. The user's diet can be managed healthily through the meal scan data of the user terminal 101. The diabetes management service device 100 predicts blood sugar through a pre-learned blood sugar prediction model and guides the user to consume the recommended amount of intake through the predicted blood sugar, thereby providing a diabetes management guide to prevent the user from developing diabetes or manage diabetes. You can. Additionally, the diabetes management service device 100 can analyze personalized nutrient information to more accurately predict postprandial blood sugar levels and provide an appropriate recommended amount of food.

As shown in FIG. 4, the user terminal 101 performs a meal scan operation for diabetes management service. When the user runs the camera through the user terminal 101, the user terminal 101 detects food on the capture screen and displays the detected food area on the capture screen in real time. And when a user takes a picture of food using a camera, the user terminal 101 cuts out the food area in the taken picture. When the user selects a food to consume from the cut food area, the user terminal 101 may save the selected food as a diet record and transmit it to the diabetes management service device 100.

As shown in FIG. 5, the diabetes management service device 100 can predict blood sugar through a blood sugar prediction model if there is food to be consumed in the photograph taken from meal scan data. The diabetes management service device 100 may provide, through the user terminal 101, a blood sugar prediction graph in which the user's blood sugar level may increase when the user eats a meal. At this time, the diabetes management service device 100 may provide a comparison between the user's current blood sugar prediction graph and the previously learned normal user's blood sugar prediction graph. For example, the diabetes management service device 100 may provide the user with a blood sugar prediction graph showing that the blood sugar level may be higher by an average of 40 mg/dl than that of a normal user. At this time, the user terminal 101 may take a photo again for the user or provide a diet recommendation for a diabetes management service.

As shown in FIG. 6, the diabetes management service device 100 provides a diabetes management guide that displays the recommended intake amount based on predicted blood sugar level. For example, the diabetes management service device 100 displays the recommended intake amount (e.g., 187 g of pasta and 257 kcal calories) in the photographed food photo and displays the remaining food excluding the recommended intake amount as the intake limit through the user terminal. It can be provided at (101). The user terminal 101 may display the diabetes management guide received from the diabetes management service device 100 on a capture screen or a photo of food.

In relation to the diabetes management guide, the diabetes management service device 100 may provide the recommended carbohydrate intake for the user to consume from meal scan data. For example, the diabetes management service device 100 recognizes how much rice is in the rice bowl from meal scan data and guides the user on how much rice to eat excluding the amount appropriate for the recommended intake. A management guide can be provided through the user terminal 101. At this time, the diabetes management service device 100 can infer the amount of rice by using the volume in the rice air as a reference point. Additionally, the diabetes management service device 100 may provide a diabetes management guide by adjusting the amount of rice according to the individual's intake history.

According to embodiments, the type of rice, etc. is specified by targeting places where recipes have already been determined, such as schools and hospitals. In situations where menus can be specified in this way, such as schools and hospitals, recipes are already set and food types are imaged. Accordingly, the diabetes management service device 100 may provide a blood sugar prediction value and a diabetes management guide by considering the user's overall data on whether to take a little rice from the plate.

According to embodiments, the cafeteria collects information that a typical middle school child eats this much rice for lunch. The diabetes management service device 100 may link the cafeteria information collected from these cafeterias with the user terminal 101 to provide a diabetes management guide for consuming the same amount of rice for lunch in the summer for people of similar age and gender.

According to embodiments, the diabetes management service device 100 determines the type of rice bowl based on the similarity of images and determines the volume of the rice bowl using triangulation using a reference object or a plurality of photos to infer the amount of rice. can do. The diabetes management service device 100 may determine the user's intake amount or visually provide the amount of rice and how much rice should be consumed through rice images taken from food photos. Additionally, in order to record the intake of not only rice but also specific foods related to carbohydrate nutrients, the user can specify specific foods (e.g., rice, noodles, noodles, chocolate, cola, etc.).

According to embodiments, the above-described determination of the volume of the rice bowl does not occur once, and the volume of the rice bowl may change when the user goes to a different restaurant or changes the atmosphere. The user terminal 101 may reset the rice volume again, or inquire from the user how much it has grown or decreased compared to the amount of rice consumed previously. For example, the user terminal 101 can understand that the user inputs that the amount of rice consumed has increased by 10% or decreased by 10% compared to the previous amount, or that although it is not accurate, the user has operated it by taking that into account, so the accuracy is low. If you want to measure accurately, you can say, “Perform initialization again.”

According to embodiments, the diabetes management service device 100 searches for information on users of similar categories (e.g., age group, gender, season, diet pattern, etc.) based on cafeteria information collected from the cafeteria (doppelganger search). , it is possible to predict appropriate blood sugar levels for users in similar categories and provide blood sugar predictions, or provide diabetes management guides tailored to users in similar categories. The diabetes management service device 100 can effectively utilize food intake-related data of a specific user and a doppelganger user (e.g., another user with similar food tastes or similar diabetes symptoms, etc.) for health management or diabetes management of the specific user. Blood sugar prediction factors can be analyzed, and at least one analyzed blood sugar prediction factor can be reflected in a blood sugar prediction model to provide a diabetes management guide for health management or diabetes management for a specific user.

According to embodiments, cafeterias serve small amounts of food at the beginning of the meal out of concern for food shortage, and serve more at the end of the meal to reduce leftovers. However, assuming that users within the distribution consume food uniformly, there may be a lot of leftover food in the later stages of the distribution, resulting in a lot of food waste. If the weight of the food waste bin is continuously acquired, statistical data on this is accumulated, and if this is acquired over a certain period of time (for example, a month), the distribution pattern can be extracted. The diabetes management service device 100 may match the food distribution pattern and food type according to the distribution time and provide management standards (for example, in the form of an image or weight) for the appropriate amount of food according to the distribution time at the cafeteria.

As shown in FIG. 7, the diabetes management service device 100 according to an embodiment of the present invention includes a communication module 110, a database 120, a memory 130, and a processor 140. However, not all of the illustrated components are essential components. The diabetes management service device 100 may be implemented with more components than the illustrated components, or the diabetes management service device 100 may be implemented with fewer components.

Hereinafter, the specific configuration and operation of each component of the diabetes management service device 100 of FIG. 7 will be described.

The communication module 110 communicates with the user terminal 101 and the medical data server 200.

The database 120 stores data related to diabetes management services or a blood sugar prediction model for diabetes management services.

Memory 130 stores one or more programs related to diabetes management services.

Processor 140 executes one or more programs related to diabetes management services stored in memory 130. The processor 140 receives meal scan data, blood sugar measurement data, and user status data from the user terminal 101 through the communication module 110. The processor 140 calculates blood sugar image index information using the received meal scan data, blood sugar measurement data, and user state data. The processor 140 receives user medical data from the medical data server 200 through the communication module 110. The processor 140 generates user health care information using the received user medical data. The processor 140 predicts blood sugar by inputting it into a pre-learned blood sugar prediction model using the calculated blood sugar image index information and generated user health management information, and displays the recommended intake amount based on the predicted blood sugar for diabetes management. A guide is provided to the user terminal 101.

According to embodiments, the diabetes management service device 100 may use meal scan data to correct and predict the expected blood sugar rise after a specific time (eg, 2 hours) using the amount of food before the meal and the amount of leftover food after the meal.

According to embodiments, the diabetes management service device 100 captures an image of food before a meal, then detects the end of the meal (for example, getting up from the seat after a meal, etc.) through linkage with the health device 103 and notifies the image of the food after the meal. Alternatively, the exercise state can be detected and reflected in the blood sugar model to modify the blood sugar prediction value.

According to embodiments, the diabetes management service device 100 learns by reflecting specific data most relevant to postprandial blood sugar prediction among the data input to the blood sugar prediction model in order of weight, and learns the blood sugar prediction model to provide more accurate postprandial blood sugar prediction. Blood sugar levels can be predicted.

According to embodiments, the diabetes management service device 100 uses a general-purpose blood sugar model suitable for general users to create a personalized blood sugar model that matches meal log data such as meals of a specific individual, and uses the personalized blood sugar model. This allows you to predict postprandial blood sugar levels. Additionally, the diabetes management service device 100 may collect feedback from postprandial predictions of specific individuals to calibrate and update the general blood sugar model.

According to embodiments, the diabetes management service device 100 not only predicts postprandial blood sugar levels, but also determines whether or not to enter the pre-diabetes stage, the period until entry, the period from entry to entry, or whether or not to enter the previously learned blood sugar level. The model can be used to predict postprandial blood sugar levels. Alternatively, the diabetes management service device 100 may predict pre-diabetes by learning a new pre-diabetes model that is different from the previously learned blood sugar model.

The diabetes management service device 100 includes a communication module 110, a database 120, a memory 130 that stores one or more programs, and a processor 140 that executes one or more programs stored in the memory.

The processor 140 receives the user's meal data and blood sugar measurement data, identifies a meal event using at least one of the received meal data and blood sugar measurement data, and identifies at least one food from the received meal data. Integrate the blood sugar measurement data corresponding to the integrated meal event into blood sugar measurement data for each meal event, and use the mapped blood sugar measurement data for each meal event and a pre-learned blood sugar prediction model to level the user's blood sugar level. Calculates and predicts blood sugar changes, and outputs the calculated blood sugar level and predicted blood sugar changes.

According to embodiments, meal data may include at least one of an image, text, and metadata related to the user's meal.

According to embodiments, the processor 140 may extract at least one key value classified by food from the received meal data and identify at least one food using the at least one extracted key value.

According to embodiments, the processor 140 may perform at least one of the following: when meal data including an image is received, when meal data including text is received, and when the user's blood sugar level exceeds a preset blood sugar level value. The case can be set as the start point of the meal event.

According to embodiments, the processor 140 calculates an evaluation index for evaluating the glycemic response for the identified meal event using the average blood sugar value of the meal event, and calculates the blood sugar response for each meal event through the calculated evaluation index. Responses can be evaluated.

According to embodiments, the processor 140 normalizes the mapped blood sugar measurement data for each meal event by considering the blood sugar level for each user, and normalizes the blood sugar response of the user group in a standardized manner through the normalized blood sugar measurement data for each meal event. You can compare.

According to embodiments, the processor 140 may recommend at least one of a personalized meal, meal order, and meal speed for each theme to the user using at least one extracted key value.

According to embodiments, the processor 140 may use at least one extracted key value to process elements that are not suitable for meal recommendation as exceptions and recommend a meal by excluding the exceptions.

According to embodiments, the processor 140 uses external data including at least one of personal information, health checkup data, and public institution data to create a cohort of other users with a preset similarity or more to the user, and generates the generated cohort of other users. By analyzing the blood sugar response data of the cohort, customized blood sugar management recommendations can be provided to users.

Meanwhile, the diabetes management service device 100 according to another embodiment of the present invention can classify various foods consumed by the user and map blood sugar data corresponding to each meal. This mapping process is important in understanding how individual foods combine to affect blood sugar. Various foods consumed by the user affect blood sugar levels through the digestion process. Therefore, the diabetes management service device 100 uses the image, text, or other metadata of each meal to accurately classify foods, integrate the classified foods into meal events, and associate the integrated meal events with blood sugar changes. You can. The diabetes management service device 100 goes beyond simply recording meal data and can contribute to understanding and predicting the interaction between food and blood sugar.

Here, the diabetes management service device 100 may be implemented in a user terminal or as a diabetes management server that communicates with the user terminal.

Looking at the user interface, users upload photos of their meals through a mobile app. This can be achieved through the app's camera function. Additionally, the user can enter additional text or just text.

The diabetes management service device 100 can identify food from uploaded meal images or text through AI image analysis and perform classification based on the characteristics (e.g., color, shape) of each food.

In order to process meal data, the diabetes management service device 100 may link food information classified by AI with a meal database and connect nutritional information for each meal event with possible blood sugar response.

In order to integrate the CGM data, the diabetes management service device 100 can receive the user's continuous glucose monitor (CGM) data in real time to the app and track the user's blood sugar level and changes.

The diabetes management service device 100 can combine CGM data and meal data classified by AI for analysis and prediction, and predict the user's current blood sugar level and expected blood sugar change due to the meal.

In order to provide all information to the user, the diabetes management service device 100 may visually provide all of the above information so that the user can check the expected blood sugar change after a meal through the app. This can help users make immediate decisions about managing their blood sugar.

The diabetes management service device 100 records all data through a mobile app for recording and tracking, tracks the user's meal history and blood sugar change trends, and can be used for long-term health management and analysis.

The diabetes management service device 100 may provide a personalized meal recommendation service based on accumulated individual meal data, blood sugar response, and other characteristics.

The diabetes management service device 100 may support a user to leave a personal note along with the blood sugar response to each meal for personal note and metadata recording. These notes may include how you felt about the meal, your taste rating, and additional information about the eating situation. This memo data can be used to record the user's personal experience.

The diabetes management service device 100 allows users to share their meal records and related metadata with the community within the app for community sharing. This increases interaction between users and provides motivation and support for forming healthy eating habits.

The diabetes management service device 100 may include a reward system to encourage the user's health care behavior for behavior-based rewards. For example, the diabetes management service device 100 may continuously upload CGM data and provide incentives to users who regularly leave meal records. This can motivate users to continuously manage their health data.

In step S201, the diabetes management service device receives the user's meal data and blood sugar measurement data.

In step S202, the diabetes management service device identifies a meal event using at least one of the received meal data and blood sugar measurement data.

In step S203, the diabetes management service device identifies at least one food from the received meal data and integrates it into the identified meal event.

In step S204, the diabetes management service device maps blood sugar measurement data corresponding to the integrated meal event to blood sugar measurement data for each meal event.

In step S205, the diabetes management service device calculates the user's blood sugar level and predicts blood sugar changes using blood sugar measurement data for each mapped meal event and a pre-learned blood sugar prediction model.

In step S206, the diabetes management service device outputs the calculated blood sugar level and the predicted blood sugar change.

To classify and map food data, the diabetes management service device 100 may identify food through image analysis and user input, integrate it into a meal event, and map it to blood sugar response data.

For example, when an image is uploaded, the diabetes management service device 100 can extract various key values using a large language model (LLM). This is not only for food classification, but also for understanding the overall characteristics of the image. This may include whether harmful elements are present, whether captured, whether indoors or outdoors, whether shared or not, or the type of meal (e.g. fast food, alcohol, drugs). Additionally, the diabetes management service device 100 may analyze the amount of meal, the state of the meal (e.g., eating all or only part of the meal), etc. Additionally, the diabetes management service device 100 can distinguish foods using key values including various characteristics such as shape, taste, and color. The diabetes management service device 100 can identify all foods included in the meal based on the characteristics of each food thus classified.

Thereafter, the diabetes management service device 100 may identify a meal event using various conditions and integrate images or user text inputs into a meal event for event identification and integration.

The diabetes management service device 100 performs a blood sugar data mapping operation for each meal, which may be based on the concept of a 'meal event'. When a user inputs a meal through image or text input, etc., this 'meal' may be identified as a specific 'meal event'. A meal event can start from the moment a user uploads an image or enters text, and can last for a set amount of time (e.g., 2 hours or more). Blood sugar measurement data during this period can be linked to this meal event and analyzed. Rather than simply using the entire daily blood sugar data, the diabetes management service device 100 may cut the blood sugar measurement data by meal event and map it to each meal event.

A meal event can be triggered when a user inputs an image of food. This may coincide with the point in time when the user actually starts eating. The start of a meal event is linked to image upload, allowing blood sugar changes related to meals to be more accurately recorded and analyzed.

Starting a meal event can be done in a variety of ways. The most common case is when a user uploads an image of food. However, meal events can also be triggered under other conditions. For example, the user can manually enter the time when they started eating, or it can be started automatically based on certain biosignals (such as changes in blood sugar levels). In this way, the diabetes management service device 100 determines the average blood sugar level when 1) a user uploads a food image, 2) a user enters meal information in text, or 3) a change in blood sugar level is detected. If at least one of the cases is higher or exceeds a set threshold, a meal event may be automatically started through time series data analysis.

The diabetes management service device 100 may utilize information based on various key values extracted from images or text to determine whether to use the meal event starting point. For example, the diabetes management service device 100 may set a condition such that if there is no food in the image, the image is not considered to start a meal event. The diabetes management service device 100 may contribute to improving the validity of integrated outcome data of meal events.

The diabetes management service device 100 can also flexibly set the event end. Meal events can end after a certain amount of time (e.g. 2 hours) by default, but this can be adjusted depending on the user's needs. Additionally, the diabetes management service device 100 may detect that the blood sugar level falls back to the average level and end the event. In this way, the diabetes management service device 100 can more accurately and flexibly establish the connection between blood sugar data and meal events.

After identifying the meal event, the diabetes management service device 100 can now map foods related to the meal event. These foods can be identified through image recognition or user input (text). One meal event may include multiple images, and in this case, the diabetes management service device 100 may analyze the images and integrate them into one meal event.

The process of integrating multiple foods into a meal event may involve the use of unique values. This can be important to remove duplicates if the same food is photographed multiple times during a meal. The diabetes management service device 100 can identify a unique value in each food image or text data and regard it as one food when the same food appears repeatedly. Through this, the diabetes management service device 100 can ensure the accuracy of meal data and, as a result, provide more accurate information in analyzing the relationship between blood sugar measurement data and food.

Meanwhile, the diabetes management service device 100 may evaluate a meal event. The diabetes management service device 100 may identify characteristics of blood sugar response for each meal event and evaluate blood sugar impact using a meal event-based glucose management indicator (E-GMI). For example, it can be calculated as E-GMI = 3.31 + (0.02392 * average (blood sugar levels measured within the event). Here, a blood sugar management index was selected to evaluate the glycemic response to a meal, and the blood sugar management index is The diabetes management service device 100 may calculate GMI using the average blood sugar value of a meal event, which may be referred to as 'E-GMI' (Event-based GMI). Because E-GMI is used for events that raise blood sugar levels, such as meals, it tends to be higher than normal glycated hemoglobin levels, making direct comparison with glycated hemoglobin used to diagnose diabetes difficult. In addition, the diabetes management service device 100 identifies the starting point, peak point, blood sugar level after 1 hour, or blood sugar level after 2 hours within the meal event and uses it as a blood sugar management indicator. You can.

Meanwhile, the diabetes management service device 100 can classify the food consumed by the user through a blood sugar prediction model and predict changes in blood sugar by mapping it with blood sugar measurement data.

The diabetes management service device 100 has data such as nutrient information, serving size, and glycemic index (GI) and glycemic load index (GL) of each food based on key values classified by food. , this can be estimated using the large language model. This information includes carbohydrates, proteins, fats, etc.

When a user takes an image of a meal, the diabetes management service device 100 can search food information in a database based on the integrated food key value. The diabetes management service device 100 may group previously accumulated data using a combination of food key values and average them to provide a blood sugar response to the food. Additionally, the diabetes management service device 100 may provide the user's predicted blood sugar response based on data of other users similar to the user. Accordingly, when the user takes a photo of a meal or inputs food-related text, the diabetes management service device 100 analyzes the photo and provides a blood sugar prediction result to show a predicted value of how the meal will affect blood sugar. .

The diabetes management service device 100 may also be able to approach in the opposite direction during the data analysis process. For example, the diabetes management service device 100 may analyze the blood sugar responses of various people to black bean noodles and identify events showing significant differences among them. Then, the diabetes management service device 100 can check the specific meal event in more detail and analyze what foods the meal consists of.

Through this, the diabetes management service device 100 can discover that one of the people who ate Jjajangmyeon also consumed soju, so the change in blood sugar level was not significant. Through this analysis, the diabetes management service device 100 can further understand the relationship between a specific meal and blood sugar response and provide insightful information such as 'This is why these people reacted this way.'

Through this process, the diabetes management service device 100 can provide information about blood sugar response to the user in real time, and the user can make healthy meal choices based on this. Through these functions, the diabetes management service device 100 provides users with immediate feedback on blood sugar management and can help form healthy eating habits in the long term.

As shown in FIG. 11, the diabetes management service device 100 displays the maximum value (MAX) 310 of the glucose index according to the main menu names AA to JJ and the average value of the glucose index (Average). )(320) can be obtained.

For data normalization, the diabetes management service device 100 may normalize blood sugar measurement data by considering individual blood sugar levels and perform standardized comparison and analysis.

Here, for meal-specific recommendations and data sharing, the normalization process of data may be important. This is because it takes into account different blood sugar levels for each individual. For example, when eating black bean noodles, the blood sugar response of healthy people and diabetic patients may be different. If this data is not normalized, the blood glucose data after 2 hours will be dominated by the high response of diabetic patients. This can distort data from healthy people. Therefore, a process of normalizing the data may be necessary to obtain an undistorted average.

For data normalization, the diabetes management service device 100 may use blood sugar measurement data for a specific period. For example, after attaching a continuous blood glucose monitor (CGM), the data from the first day can be averaged and used as a reference point. Another method is to calculate the average value based on the glycated hemoglobin value entered by the user.

Through data normalization, the diabetes management service device 100 can compare the blood sugar responses of various user groups in a standardized manner. For example, the diabetes management service device 100 may calculate blood sugar change rates 410 to 450 of menus AA to EE according to the event relative time. The diabetes management service device 100 can analyze the blood sugar response of various people who eat food BB. At this time, the diabetes management service device 100 may set the starting point to the reference value '1' and calculate the blood sugar change rate of each user as a percentage. In this way, the diabetes management service device 100 can clearly compare the difference in blood sugar response between a healthy person and a diabetic patient.

The diabetes management service device 100 can perform cohort analysis based on this data. The diabetes management service device 100 may use other categories, such as health checkup data, to analyze the blood sugar response of certain types of people when they eat the food BB. The diabetes management service device 100 can show the overall response and differences between different groups through the blood sugar conversion rate.

Meanwhile, the diabetes management service device 100 provides personalized meal recommendations for meal recommendation services and exception processing, and can identify elements unsuitable for health and exclude them from the recommendations.

The diabetes management service device 100 can classify food by various key values such as color and shape. Through this, the diabetes management service device 100 may recommend at least one of themed meals, meal order, and meal speed to the user. For example, the diabetes management service device 100 may provide meal recommendations by theme based on data such as blood sugar response to brown food, indoor/outdoor eating patterns, consumption of specific types of food, etc. These themed meal recommendations add a fun element to the user experience and can contribute to increasing users' interest and dwell time in the app. This process can serve as an important factor in promoting user participation as well as personalized meal recommendations.

Additionally, the diabetes management service device 100 may be able to recommend meals tailored to various health goals by analyzing the user's health data, past meal patterns, or similar user data.

Conversely, the diabetes management service device 100 may include an exception handling function in the recommendation algorithm. For example, the diabetes management service device 100 may analyze data of a person whose blood sugar level rose high after eating black bean noodles and a person whose blood sugar level rose low after eating soju together. At this time, the diabetes management service device 100 does not provide unrealistic recommendations such as 'Drink soju to lower blood sugar level'. Instead, the diabetes management service device 100 may identify whether the meal includes alcohol through image analysis and process this as an exception in the recommendation algorithm.

In this way, the diabetes management service device 100 may classify meals containing alcohol or other specific elements as 'meals not suitable for recommendation'. Additionally, the diabetes management service device 100 is not limited to such exception processing simply to alcohol, but can extend it to identify elements unsuitable for recommendation through other key values. The diabetes management service device 100 can play an important role in suggesting realistic and healthy meal choices to the user through this exception handling function.

According to embodiments, the diabetes management service device 100 receives GPS information (eg, latitude and longitude) included in the metadata of the image. Through this, the diabetes management service device 100 can record the location information of the image and recommend a restaurant based on location, similar to meal recommendation. The diabetes management service device 100 may map the blood sugar response to a meal event as well as map the blood sugar response to a location to collect the blood sugar response for each location.

And similar to utilizing a medical data server to perform cohort analysis on users, the diabetes management service device 100 determines the location of the restaurant based on restaurant information in an external database (e.g., utilizing the API of a map platform). It can be converted to . If it is difficult to specify a restaurant based on location alone, the diabetes management service device 100 may increase the accuracy of restaurant identification by providing a UI to the user to distinguish between restaurants. Additionally, the diabetes management service device 100 can identify menus more accurately by collecting restaurant menu names and restaurant preferences (eg, star ratings) using an external database related to restaurant information. Accordingly, the diabetes management service device 100 collects the blood sugar response by location and maps the blood sugar response according to the menu of each restaurant corresponding to the location, so that the user can select a restaurant or restaurant menu where blood sugar level rises less than the preset blood sugar level. It can be recommended to Additionally, because each restaurant has different cooking methods, even if the food is the same, users' blood sugar response tendencies may vary depending on the restaurant (location). The diabetes management service device 100 may provide the user with the blood sugar tendency of other user groups for each restaurant.

Meanwhile, the diabetes management service device 100 may provide personalized blood sugar management recommendations by utilizing external information such as health checkup data. The diabetes management service device 100 may be connected to an external data source to provide personalized services. External information may include personal information entered by the user, health checkup data, public institution data, etc. The diabetes management service device 100 may form a 'cohort', a group of other people with similar characteristics to the user, based on outsourced information. By analyzing the blood sugar response data of people in this cohort, personalized blood sugar management recommendations can be provided to these users. This approach can be important in providing users with a more personalized experience, beyond just general recommendations.

Meanwhile, according to an embodiment of the present invention, the various embodiments described above are implemented as software including instructions stored in a machine-readable storage media (e.g., a computer). It can be. The device is a device capable of calling instructions stored from a storage medium and operating according to the called instructions, and may include an electronic device (eg, electronic device A) according to the disclosed embodiments. When an instruction is executed by a processor, the processor may perform the function corresponding to the instruction directly or using other components under the control of the processor. Instructions may contain code generated or executed by a compiler or interpreter. A storage medium that can be read by a device may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' only means that the storage medium does not contain signals and is tangible, and does not distinguish whether the data is stored semi-permanently or temporarily in the storage medium.

Additionally, according to an embodiment of the present invention, the method according to the various embodiments described above may be provided and included in a computer program product. Computer program products are commodities and can be traded between sellers and buyers. The computer program product may be distributed on a machine-readable storage medium (e.g. compact disc read only memory (CD-ROM)) or online through an application store (e.g. Play Store™). In the case of online distribution, at least a portion of the computer program product may be at least temporarily stored or created temporarily in a storage medium such as the memory of a manufacturer's server, an application store's server, or a relay server.

In addition, according to an embodiment of the present invention, the various embodiments described above are stored in a recording medium that can be read by a computer or similar device using software, hardware, or a combination thereof. It can be implemented in . In some cases, embodiments described herein may be implemented in a processor itself. According to software implementation, embodiments such as procedures and functions described in this specification may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein.

Meanwhile, computer instructions for performing processing operations of devices according to the various embodiments described above may be stored in a non-transitory computer-readable medium. Computer instructions stored in such a non-transitory computer-readable medium, when executed by a processor of a specific device, cause the specific device to perform processing operations in the device according to the various embodiments described above. A non-transitory computer-readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as registers, caches, and memories. Specific examples of non-transitory computer-readable media may include CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, etc.

In addition, each component (e.g., module or program) according to the various embodiments described above may be composed of a single or multiple entities, and some of the sub-components described above may be omitted, or other sub-components may be omitted. Sub-components may be further included in various embodiments. Alternatively or additionally, some components (e.g., modules or programs) may be integrated into a single entity and perform the same or similar functions performed by each corresponding component prior to integration. According to various embodiments, operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or at least some operations may be executed in a different order, omitted, or other operations may be added. It can be.

In the above, preferred embodiments of the present invention have been shown and described, but the present invention is not limited to the specific embodiments described above, and may be used in the technical field pertinent to the disclosure without departing from the gist of the present invention as claimed in the claims. Of course, various modifications can be made by those skilled in the art, and these modifications should not be understood individually from the technical idea or perspective of the present invention.

[Explanation of symbols]

10: Diabetes management service system

100: Diabetes management service device

101: User terminal

102: Blood sugar meter

103: Health device

200: Medical data server

110: communication module

120: database

130: memory

140: processor

Claims

In a diabetes management service method by a diabetes management service device,

Receiving the user's meal data and blood sugar measurement data;

Identifying a meal event using at least one of the received meal data and blood sugar measurement data;

identifying at least one food from the received meal data and incorporating it into the identified meal event;

mapping blood sugar measurement data corresponding to the integrated meal event into blood sugar measurement data for each meal event;

calculating the user's blood sugar level and predicting changes in blood sugar using the mapped blood sugar measurement data for each meal event and a pre-learned blood sugar prediction model; and

Diabetes management service method comprising outputting the calculated blood sugar level and the predicted blood sugar change.
According to paragraph 1,

The meal data is,

A diabetes management service method, comprising at least one of images, text, and metadata related to the user's meals.
According to paragraph 1,

The step of integrating into the identified meal event is,

A diabetes management service method that extracts at least one key value classified by food from the received meal data and identifies at least one food using the extracted at least one key value.
According to paragraph 1,

The step of identifying the meal event is,

Setting at least one of the following cases, when meal data including an image is received, when meal data including text is received, and when the user's blood sugar level exceeds a preset blood sugar level value, is set as the start point of the meal event. , how to get diabetes management services.
According to paragraph 1,

For the identified meal events, calculating an evaluation index for evaluating the glycemic response using the average blood sugar value of the meal event, and further comprising evaluating the blood sugar response for each meal event through the calculated evaluation index. , how to get diabetes management services.
According to paragraph 1,

Diabetes, further comprising normalizing the mapped blood sugar measurement data for each meal event in consideration of the blood sugar level for each user, and comparing blood sugar responses of user groups in a standardized manner through the normalized blood sugar measurement data for each meal event. Management service method.
According to clause 3,

Diabetes management service method further comprising recommending at least one of personalized meals, meal order, and meal speed for each theme to the user using the extracted at least one key value.
In clause 7,

The steps for recommending the above meal are:

A diabetes management service method that uses the extracted at least one key value to treat elements that are not suitable for meal recommendation as exceptions and recommend meals by excluding the exceptions.
According to paragraph 1,

Using external data that includes at least one of personal information, health checkup data, and public institution data, a cohort of other users with a preset similarity level or higher is created, and the blood sugar response data of the created cohort is analyzed to inform the user. A method of providing diabetes management services, further comprising providing personalized blood sugar management recommendations.
communication module;

memory that stores one or more programs; and

a processor that executes the one or more stored programs,

The processor,

Receive the user's meal data and blood sugar measurement data,

Identifying a meal event using at least one of the received meal data and blood sugar measurement data,

Identify at least one food from the received meal data and integrate it into the identified meal event,

Map the blood sugar measurement data corresponding to the integrated meal event to blood sugar measurement data for each meal event,

Calculate the user's blood sugar level and predict changes in blood sugar using the mapped blood sugar measurement data for each meal event and the previously learned blood sugar prediction model,

A diabetes management service device that outputs the calculated blood sugar level and the predicted blood sugar change.
According to clause 10,

The meal data is,

A diabetes management service device comprising at least one of images, text, and metadata related to the user's meals.
According to clause 10,

The processor,

A diabetes management service device that extracts at least one key value classified by food from the received meal data and identifies at least one food using the extracted at least one key value.
According to clause 10,

The processor,

When meal data including an image is received, when meal data including text is received, and when the user's blood sugar level exceeds a preset blood sugar level value, at least one case is set as the start point of the meal event. , Diabetes Management Services Unit.
According to clause 10,

The processor,

A diabetes management service device that calculates an evaluation index for evaluating blood sugar response for the identified meal event using the average blood sugar value of the meal event, and evaluates the blood sugar response for each meal event through the calculated evaluation index. .
According to clause 10,

The processor,

A diabetes management service device that normalizes the mapped blood sugar measurement data for each meal event by considering the blood sugar level for each user, and compares blood sugar responses of user groups in a standardized manner through the normalized blood sugar measurement data for each meal event.
According to clause 12,

The processor,

A diabetes management service device that recommends at least one of personalized meals, meal order, and meal speed for each theme to the user using the extracted at least one key value.
According to clause 16,

The processor,

A diabetes management service device that processes elements that are not suitable for meal recommendation as exceptions using the extracted at least one key value, and recommends a meal by excluding the exceptions.
According to clause 10,

The processor,

Using external data that includes at least one of personal information, health checkup data, and public institution data, a cohort of other users with a preset similarity level or higher is created, and the blood sugar response data of the created cohort is analyzed to inform the user. A diabetes management service device, further comprising providing personalized blood sugar management recommendations.