Information processing system, information processing program, and information processing method

Abstract

Provided is an information processing system that provides users more useful information for improving blood sugar levels. The information processing system acquires meal information about meals consumed by a user, and outputs recommendation information on the basis of the meal information, user information, blood glucose level information, and behavioral information. Additionally, the information processing system outputs predicted blood glucose levels on the basis of meal information, user information, blood glucose level information, and behavioral information, the predicted blood glucose levels being for the case in which the user consumes a meal corresponding to the meal information. The information processing system also outputs blood glucose level prediction information on the basis of the recommended information. Moreover, the information processing system provides the user with blood glucose level prediction information based on meal information, recommendation information, and blood glucose level prediction information based on the recommendation information.

Description

Information Processing System, Information Processing Program, and Information Processing Method

[0001] One example of the present invention relates to an information processing system, an information processing program, and an information processing method for improving a user's blood glucose level.

[0002] Conventionally, an information providing device that provides information for improving a user's blood glucose level has been proposed (for example, Patent Document 1).

[0003] International Publication No. 2023 / 127285

[0004] However, in the above conventional technology, there is room for improvement in providing more beneficial information to the user.

[0005] Therefore, one object of the present invention is to provide an information processing system that provides more beneficial information to a user for improving blood glucose levels.

[0006] In order to solve the above problems, the present invention adopts the following configuration.

[0007] (First Configuration) The first configuration of the present invention is an information processing system that provides information for improving a user's blood glucose level. The information processing system includes an information acquisition means, a recommended information output means, a blood glucose prediction information output means, and a presentation means. The information acquisition means acquires user information regarding the user's body, information regarding the user's blood glucose level, dietary information regarding the user's diet, and behavior information regarding the user's behavior. The recommended information output means outputs recommended information for improving the user's blood glucose level based on the user information, the information regarding the user's blood glucose level, the dietary information, and the behavior information. The blood glucose prediction information output means outputs blood glucose prediction information regarding the user's blood glucose level when the user consumes a diet according to the dietary information based on the user information, the dietary information, and the behavior information. The presentation means presents the recommended information and the blood glucose prediction information. The blood glucose prediction information output means outputs the blood glucose prediction information corresponding to the recommended information based on the recommended information.

[0008] According to the above, users can be provided with recommended information for improving blood glucose levels and predicted blood glucose levels corresponding to those recommendations. This allows users to know the predicted blood glucose levels if they follow the recommendations, and for example, they can gain motivation to follow the recommendations.

[0009] (Second Configuration) The second configuration may be the same as the first configuration, wherein the recommendation information output means is a learner that takes user information, blood glucose prediction information or information relating to blood glucose levels, meal information and behavior information as inputs and outputs the recommendation information, and the blood glucose prediction information output means is a learner that takes user information, meal information and behavior information as inputs and outputs the blood glucose prediction information. The information processing system may further include: a second information acquisition means for acquiring recommendation information implementation information showing the user's implementation status of the recommendation information; a third information acquisition means for acquiring blood glucose evaluation information relating to the user's evaluation of blood glucose levels in relation to the blood glucose prediction information; a first retraining means for causing the recommendation information output means to retrain using the recommendation information and the recommendation information implementation information as training data; and a second retraining means for causing the blood glucose prediction information output means to retrain using the blood glucose prediction information and the blood glucose evaluation information as training data.

[0010] Based on the above, the recommendation information output means and the blood glucose level prediction information output means can be retrained based on the recommendation information implementation information and blood glucose level evaluation information from the user, thereby improving the accuracy of the recommendation information and blood glucose level prediction information.

[0011] (Third Configuration) In the third configuration, the recommendation information output means may include a meal suggestion means that suggests improved meal information for an improved meal when the user consumes a meal corresponding to the meal information and the blood glucose level prediction information satisfies specific conditions. The blood glucose level prediction information output means may output blood glucose level prediction information when the user consumes an improved meal corresponding to the improved meal information.

[0012] According to the above, if the blood glucose level prediction information for a meal consumed by a user meets specific conditions, an improved meal can be suggested. For example, if an increase in blood glucose level is predicted, an improved meal that suppresses the increase in blood glucose level can be suggested.

[0013] (Fourth configuration) In the third configuration, the presenting means may present the blood glucose level prediction information when the user consumes a meal corresponding to the meal information, and the blood glucose level prediction information when the user consumes an improved meal corresponding to the improved meal information.

[0014] According to the above, users can compare and select between blood glucose prediction information for meals before improvement and blood glucose prediction information for meals after improvement.

[0015] (Fifth configuration) In the fifth configuration, in the third or fourth configuration, the meal suggestion means may suggest, as improved meal information, first improved meal information relating to a first improved meal obtained by improving the meal using the first method, and second improved meal information relating to a second improved meal obtained by improving the meal using the second method. The blood glucose level prediction information output means may output first blood glucose level prediction information when the user consumes the first improved meal, and second blood glucose level prediction information when the user consumes the second improved meal.

[0016] According to the above, it is possible to propose a first and second improved diet to the user and present blood glucose level prediction information for each.

[0017] (Sixth configuration) The sixth configuration may include a comparison means for comparing the first blood glucose prediction information and the second blood glucose prediction information in the fifth configuration. The presentation means may present information according to the results of the comparison.

[0018] According to the above, users can compare and choose between the first and second improved meals.

[0019] (Seventh configuration) In the seventh configuration, satisfying the above-mentioned specific condition in any of the third to sixth configurations may mean that a blood glucose spike occurs when the user consumes a meal corresponding to the meal information.

[0020] Based on the above, it is possible to suggest improved diets when blood sugar spikes occur.

[0021] (Eighth configuration) The eighth configuration may further include a second information acquisition means for acquiring recommended information implementation information that shows the user's implementation status of the recommended information, in any of the first to seventh configurations. It may also include a recommended information evaluation means for evaluating the user's implementation status of the recommended information based on the recommended information and the recommended information implementation information.

[0022] According to the above, users can input the status of their implementation of recommended information and receive an evaluation of their implementation status.

[0023] (Ninth configuration) The ninth configuration is that, in the eighth configuration, the recommended information implementation information may include information on alternative meals or alternative behaviors that replace the meals or behaviors recommended by the recommended information. The recommended information evaluation means may evaluate the user's implementation status of the recommended information based on the alternative meals or alternative behaviors.

[0024] Based on the above, the implementation status of alternative behaviors to the recommended behaviors can also be evaluated.

[0025] (Tenth Configuration) The tenth configuration is an information processing system that provides information for improving the user's blood glucose level, and comprises an information acquisition means, a blood glucose level prediction information output means, and a meal suggestion means. The information acquisition means acquires user information relating to the user's body and meal information relating to the meals consumed by the user. The blood glucose level prediction information output means outputs blood glucose level prediction information relating to whether or not a blood glucose spike will occur when the user consumes a meal corresponding to the meal information, based on the user information and the meal information. If a blood glucose spike is predicted to occur, the meal suggestion means outputs improved meal information relating to an improved meal that is an improved version of the meal.

[0026] Based on the above, it is possible to predict whether or not a blood glucose spike will occur based on dietary information regarding the meals consumed by the user, and if a blood glucose spike is predicted to occur, an improved diet can be suggested.

[0027] The other invention may be an information processing program or an information processing method.

[0028] According to one example of the present invention, users can be provided with recommended information for improving blood glucose levels and blood glucose level prediction information corresponding to said recommended information.

[0029] Diagram of the schematic configuration of the information processing system 1 of this embodiment Diagram showing an example of the configuration of the user terminal 100 Diagram showing an example of the configuration of the server Diagram showing an example of the functional configuration of the server 500 Diagram showing an example of the operation overview of the information processing system 1 when outputting the first recommended information Diagram showing an example of the operation overview of the information processing system 1 when outputting the second recommended information Diagram showing the processing flow of the information processing system 1, which shows the processing flow of the user terminal and server over one day Diagram showing an example of the screen displayed on the user terminal 100 when the second recommended information is transmitted in step S1 Diagram showing the input screen for meal information in step S2, and in steps S3 and S4, "blood glucose level prediction information" corresponding to the meal information and A diagram showing an example of the screen displayed on the user terminal 100 when the "first recommended information" is output. A diagram showing an example of the input screen for blood glucose evaluation information in step S6. A diagram showing an example of the input screen for recommended information implementation information in step S7 and the screen displayed on the user terminal 100 when evaluation information is generated in step S8. A diagram showing an overview of the retraining of the recommended information output unit 522 and the blood glucose prediction information output unit 523. A flowchart showing an example of terminal processing performed on the user terminal 100 of the information processing system 1 of this embodiment. A flowchart showing an example of server processing performed on the server 500 of the information processing system 1 of this embodiment. A diagram showing an overview of the information processing system 2 of the second embodiment.

[0030] <1. Overview of the Information Processing System> The information processing system of this embodiment will be described below with reference to the drawings. The information processing system according to this embodiment is a system that provides information to improve a user's blood glucose level, for example, to prevent diabetes. Figure 1 is a schematic diagram of the information processing system 1 of this embodiment.

[0031] As shown in Figure 1, the information processing system 1 comprises a user terminal 100 for at least one user and a server 500, which are connected by a network 700 (e.g., the Internet, LAN, etc.).

[0032] The user terminal 100 is a device used by the user, and may be, for example, a smartphone, tablet, personal computer, or dedicated terminal. An example of the configuration of the user terminal 100 will be described later.

[0033] A measuring device 101 is connected to the user terminal 100 wirelessly or via a wired connection. The measuring device 101 is a device for measuring the user's blood glucose level. The measuring device 101 continuously measures blood glucose levels and is composed of, for example, a known CGM (Continuous Glucose Monitoring) or FGM (Flash Glucose Monitoring). For example, the measuring device 101 measures the glucose concentration in the subcutaneous tissue interstitial fluid and calculates the blood glucose level from the measurement result. However, the measuring device 101 is not limited to these, and is not particularly limited as long as it can measure blood glucose levels continuously or periodically (for example, every 5 minutes). In other words, there are methods for measuring blood glucose levels, such as methods of acquiring and analyzing blood, methods of measuring with a device that is constantly inserted with a needle, methods of measuring by contacting the retina, and methods of measuring by reflecting light from a light source onto blood vessels, but the method is not limited. The blood glucose level measured by the measuring device 101 is transmitted to the user terminal 100 periodically or continuously, and stored in the user terminal 100 as information regarding the blood glucose level.

[0034] The user terminal 100 stores user information, information about the user's blood glucose levels, information about the user's meals, and information about the user's behavior. This information is transmitted to the server 500 via the network 700. Details of this information will be described later.

[0035] Server 500 generates recommendation information and blood glucose prediction information based on user information, blood glucose information, meal information, behavioral information, etc., and transmits them to the user terminal 100. Details of the recommendation information and blood glucose prediction information will be described later.

[0036] <2. User Terminal> Next, the configuration of the user terminal 100 will be described. Figure 2 shows an example of the configuration of the user terminal 100.

[0037] As shown in Figure 2, the user terminal 100 comprises a control unit 11, a storage unit 12, a communication unit 13, a display 14, and an input unit 15.

[0038] The control unit 11 includes, for example, a CPU (Central Processing Unit) for information processing, RAM (Random Access Memory), and ROM (Read Only Memory). The control unit 11 may also include a GPU (Graphics Processing Unit).

[0039] The storage unit 12 is composed of, for example, a hard disk drive, a solid-state drive, an optical disk, a magnetic disk, flash memory, a memory card, etc. The storage unit 12 stores various data such as terminal programs 121, user information 122, blood glucose level information 123, meal information 124, behavioral information 125, bacterial flora information 126 related to the user's gut microbiota, and recommendation information 127 and blood glucose level prediction information 140 transmitted from the server 500. The storage unit 12 also stores gene information 130, recommendation information implementation information 150, and blood glucose level evaluation information 160.

[0040] The terminal program 121 is a program for performing terminal processing, which will be described later, and is a program for performing processing related to the acquisition, processing, management, and output of various types of data.

[0041] User information 122 includes the user's physical information, such as the user's age, gender, height, weight, chest circumference, waist circumference, BMI, body fat percentage, medical history, blood pressure, pulse rate, and other information related to health checkup results. User information 122 may also include information from the user's responses to questionnaires. This information may be entered directly into the user terminal 100 by the user or obtained from an external database.

[0042] The blood glucose information 123 is information about the user's measured blood glucose level, and may include, for example, information about at least one blood glucose level measured by the user within the past year. For example, the blood glucose information 123 may include the user's fasting blood glucose level and HbA1c measured within the past year. The blood glucose information 123 may also include information about the user's average blood glucose level, maximum blood glucose level, minimum blood glucose level, blood glucose rise curve, area under the blood glucose rise curve (AUC), etc., over a predetermined period within the past year. Note that the blood glucose information 123 is not limited to information within the past year, but may also include this information from the past several years. Furthermore, the blood glucose information 123 may also be at least one blood glucose level measured by the measuring device 101 over the past few hours, the past day, or the past few days. Furthermore, the information regarding blood glucose levels 123 may also be information regarding the user's blood glucose levels at predetermined intervals (for example, every hour) over the past few hours, the past day, or the past several days (for example, information regarding the average blood glucose level, maximum blood glucose level, minimum blood glucose level, blood glucose elevation curve, area under the blood glucose elevation curve, etc.).

[0043] Furthermore, the information regarding blood glucose levels 123 may include not only information on measured blood glucose levels but also information on estimated blood glucose levels. For example, the information regarding blood glucose levels 123 may include information on the user's subjective evaluation related to blood glucose levels (for example, the results of a questionnaire to the user regarding symptoms related to blood glucose levels, such as postprandial drowsiness or fatigue). The information regarding blood glucose levels 123 may also include information on blood glucose levels estimated based on user information 122. The information regarding blood glucose levels 123 may also include blood glucose prediction information 140 output from the blood glucose prediction information output unit 523, which will be described later.

[0044] The dietary information 124 is information regarding the user's diet (e.g., breakfast, lunch, dinner, and snacks. Including beverages), and includes information on the diet the user has taken and information on the diet the user is going to take in the future. For example, the dietary information 124 may be information regarding the content and amount of the diet the user has taken, the time when the diet was taken, etc. For example, the dietary information 124 may include information on the immediately previous diet, the diet taken in the past one day, past several days, or past several months. Also, the dietary information 124 may be information regarding the content and amount of the diet the user is going to take, the time when the diet will be taken, etc. The dietary information 124 is not particularly limited, but for example, it may be information regarding menu information for the diet (e.g., staple food: rice, main dish: tonkatsu, side dish: salad, etc.), estimated raw materials, estimated energy, nutrients and their amounts, etc.

[0045] The dietary information 124 is input, for example, by the user using the user terminal 100. For example, the user may input the dietary information 124 by selecting the diet taken or to be taken from a pre-prepared list, or may directly input the dietary information 124 by inputting text. Note that the input method of the dietary information 124 is not limited to these. For example, the dietary information 124 may be input when the user uses a camera to image the diet, and the user terminal 100 or the server 500 determines the diet content based on the imaged image.

[0046] Behavioral information 125 includes information about the user's actions (e.g., exercise) over the past few hours, past day, past few days, and past months. Behavioral information 125 may also include information such as the type of exercise, duration of exercise, and time of day when exercise was performed. Behavioral information 125 is not particularly limited and may include, for example, estimated energy expenditure, estimated load, estimated heart rate / pulse, and oxygen saturation (SpO2). This information may be obtained by the user directly entering text, or it may be obtained automatically based on measurement data from the measuring device 101 or other equipment. Furthermore, behavioral information 125 may also include information about the user's drinking history, smoking history, and daily activity level. This information may also be obtained based on questionnaires from the user. For example, daily activity level as behavioral information 125 may be information that represents the user's habits, such as "sitting all day," "standing all day," "exercising 2-3 days a week," or "hardly exercising." Furthermore, drinking history may include, for example, the type and amount of alcohol consumed by the user per day, the type and amount of alcohol consumed per week, the number of days per week on which alcohol is consumed and the amount consumed. Similarly, smoking history may include the number of cigarettes smoked per day, the number of cigarettes smoked per week, the number of years smoking, and the smoking index.

[0047] The gut microbiota information 126 is information about the user's gut microbiota. Specifically, the gut microbiota information 126 includes information such as the diversity of bacteria in the intestines and the proportion of major bacteria, and is measured from the user's stool by a specialized institution. The gut microbiota information 126 may be, for example, information obtained at least once within the past year. The gut microbiota information 126 may be obtained from a specialized institution. The gut microbiota information 126 may not be stored in the user terminal 100 but may be stored in the server 500. Furthermore, the gut microbiota information 126 may be stored in a device different from the user terminal 100 and the server 500. In addition, the gut microbiota information 126 is not limited to information actually measured, but may also be information about the gut microbiota estimated from user information 122, dietary information 124, behavioral information 125, information about the user's bowel movements, etc.

[0048] The recommended information 127 is information transmitted from the server 500, which is information recommended to improve the user's blood glucose level. For example, it is information regarding meals that lower blood glucose levels, supplements (including medicines), and actions (such as exercise). The recommended information 127 includes first recommended information and second recommended information. The first recommended information is information recommended to the user in a relatively short period. For example, it includes information regarding meals to be consumed hereafter. Also, the first recommended information may include information regarding supplements recommended to be taken before or after meals in relation to the meals to be consumed hereafter, and information regarding actions recommended after meals. The second recommended information is information recommended to the user in a period longer than the first recommended information. For example, it may be information regarding meals, supplements, and actions in the next one day or the next few days. Specific examples of the first recommended information and the second recommended information will be described later.

[0049] Here, examples of meals that lower blood glucose levels include foods with a low GI value, such as leafy vegetables, yogurt, nuts, and in the case of carbohydrates, brown rice, etc. Also, examples of supplements (including medicines) that lower blood glucose levels include those added with indigestible dextrin, those containing salacinol derived from Salacia, biguanide drugs that suppress sugar synthesis in the liver, and thiazolidine drugs that improve insulin resistance in skeletal muscle and the liver. Also, examples of actions that lower blood glucose levels include exercises such as walking, jogging, and strength training. By performing these exercises within one hour after a meal, the post-meal increase in blood glucose level can be suppressed. The recommended information 127 may include this information that lowers blood glucose levels.

[0050] The gene information 130 is information regarding the user's genes. For example, it may be gene information that can determine a high risk of diabetes. The gene information 130 is information obtained by the user's genetic test and may be obtained from a specialized institution that conducts genetic tests. Note that the gene information 130 may not be stored in the user terminal 100 and may be stored in the server 500. Also, the gene information 130 may be stored in a device different from the user terminal 100 and the server 500.

[0051] The blood glucose level prediction information 140 is information transmitted from the server 500 and is information regarding the user's predicted blood glucose level, which is output from the blood glucose level prediction information output unit 523, which will be described later.

[0052] Recommended Information Implementation Information 150 is information regarding the user's implementation status of the recommended information. Recommended Information Implementation Information 150 includes information on the extent to which the user implemented the recommended information. Specific examples of Recommended Information Implementation Information 150 will be described later.

[0053] Blood glucose level evaluation information 160 is information regarding the evaluation of the user's blood glucose level. Specific examples of blood glucose level evaluation information 160 will be described later.

[0054] The communication unit 13 is, for example, a wired LAN (Local Area Network) module or a wireless LAN module, and communicates with other devices (specifically, the server 500) via the network.

[0055] The display 14 is a display device that displays characters and images, and may be a known type such as a liquid crystal display or an organic EL display. The input unit 15 is an input device that receives user input, and may include, for example, a touch panel, a keyboard, a pointing device, etc.

[0056] Note that the hardware configuration of user terminal 100 is merely an example, and components can be omitted, replaced, or added.

[0057] <3. Server> Next, the configuration of the server 500 will be described. Figure 3 is a diagram showing an example of the server configuration. As shown in Figure 3, the server 500 comprises a control unit 51, a storage unit 52, and a communication unit 53. Note that the server 500 may be composed of one or more devices.

[0058] The control unit 51 includes a CPU, RAM, ROM, GPU, etc. The CPU performs, for example, server processing as described later. The GPU is used to train the first and second learners with data.

[0059] The storage unit 52 is composed of, for example, a hard disk drive, a solid-state drive, an optical disk, a magnetic disk, flash memory, a memory card, etc. Various types of information are stored in the storage unit 52, such as the information processing program 520, the first learner 522, the second learner 523, training data 525 and 529, the user database 526, and the recommendation information database 527.

[0060] The information processing program 520 is a program for performing server processing, which will be described later. For example, the information processing program 520 is a program for acquiring various information from the user terminal 100 and for outputting recommendation information 127 and blood glucose level prediction information 140 based on the acquired information.

[0061] The first learning unit 522 is an example of the recommendation information output unit 522 described later, and is a learning unit that has been trained to output recommendation information 127 as output information when user information 122, blood glucose level information 123, meal information 124, and behavior information 125 are input information. In addition to this information, gut microbiota information 126 may also be input to the first learning unit 522. The first learning unit 522 is pre-trained based on training data obtained from a large number of subjects (for example, a dataset in which the subject's physical information, measured blood glucose level information, information on the meals consumed, information on the actions taken by the subject, and the subject's gut microbiota information are input information, and the meals, supplements, and actions taken when blood glucose levels decreased are output information). Specifically, the first learning unit 522 is, for example, a multi-layered neural network used in so-called deep learning, and comprises an input layer, an intermediate layer (hidden layer), and an output layer. The first learning unit 522 may also be configured using other known machine learning algorithms.

[0062] The second learning unit 523 is an example of the blood glucose level prediction information output unit 523 described later, and is a learning unit that has been trained to output blood glucose level prediction information 140 as output information when user information 122, blood glucose level information 123, meal information 124, and behavior information 125 are input information. In addition to this information, gut microbiota information 126 and genetic information 130 may also be input information to the second learning unit 523. The second learning unit 523 is pre-trained based on training data obtained from a large number of subjects (for example, a dataset that takes the subject's physical information, information on the meals consumed, information on the actions taken by the subject, the subject's gut microbiota information, and the subject's genetic information as input information, and the measured blood glucose level information as output information). Specifically, the second learning unit 523 is, for example, a multi-layered neural network used in so-called deep learning, and comprises an input layer, an intermediate layer (hidden layer), and an output layer. The second learning unit 523 may also be configured using other known machine learning algorithms.

[0063] The training data 525 and 529 are data used for retraining the first learner 522 and the second learner 523. Details of the training data 525 and 529 will be described later.

[0064] The user database 526 is a database that stores information about users, including user information 122 transmitted from the user terminal 100 to the server 500, blood glucose level information 123, dietary information 124, behavioral information 125, gut microbiota information 126, genetic information 130, and recommendation information 127 transmitted to the user terminal. Note that gut microbiota information 126 may not be transmitted from the user terminal 100 but may be transmitted from a specialized institution. In addition to this information, various other user information related to this information processing system, such as the user's usage history, may also be stored in the user database 526.

[0065] The recommendation information database 527 is a database that stores information on recommended foods, supplements, and behaviors. For example, the recommendation information database may store information associated with a particular food (e.g., pasta) that is less likely to raise blood glucose levels than that food (e.g., low-carbohydrate pasta). Similarly, the recommendation information database may store information associated with a particular behavior (e.g., walking) that is more likely to lower blood glucose levels than or equal to that behavior (e.g., jogging).

[0066] The communication unit 53 is, for example, a wired LAN (Local Area Network) module or a wireless LAN module, and communicates with other devices (specifically, the server 500) via the network.

[0067] Figure 4 shows an example of the functional configuration of server 500. As shown in Figure 4, server 500 includes an information acquisition unit 521, a recommendation information output unit 522, a blood glucose level prediction information output unit 523, and an evaluation unit 524. Each of these units is realized when the control unit 51 (e.g., CPU) of server 500 executes an information processing program 520 stored in the storage unit 52.

[0068] The information acquisition unit 521 acquires user information 122, blood glucose level information 123, diet information 124, behavioral information 125, gut microbiota information 126, and genetic information 130, etc. For example, the information acquisition unit 521 may acquire this information by communicating with the user terminal 100, or by accessing the user database 526. For example, of the user information 122, blood glucose level information 123, diet information 124, behavioral information 125, gut microbiota information 126, and genetic information 130, only updated information may be acquired from the user terminal 100, and unupdated information may be acquired from the user database 526. The acquired information is input to the recommendation information output unit 522 (first learner) and the blood glucose level prediction information output unit 523 (second learner).

[0069] The recommendation information output unit 522 outputs recommendation information 127 based on user information 122, blood glucose level information 123, meal information 124, and behavioral information 125. Specifically, the recommendation information output unit 522 is the first learning device that has been trained to take user information 122, blood glucose level information 123, meal information 124, behavioral information 125, and gut microbiota information 126 as inputs and output recommendation information 127. The recommendation information output unit 522 is not limited to a trained learning device as long as it is configured to output recommendation information 127 based on these inputs. For example, the recommendation information output unit 522 may be configured to output recommendation information 127 based on a rule-based system. The recommendation information output unit 522 does not necessarily need to receive gut microbiota information 126 as input.

[0070] The blood glucose level prediction information output unit 523 outputs blood glucose level prediction information 140 based on user information 122, blood glucose level information 123, meal information 124, and behavioral information 125. Specifically, the blood glucose level prediction information output unit 523 is the second learning device that has been trained to take user information 122, blood glucose level information 123, meal information 124, behavioral information 125, gut microbiota information 126, and genetic information 130 as input and output blood glucose level prediction information 140. The blood glucose level prediction information output unit 523 may be input with information on measured blood glucose levels as "blood glucose level information 123," or instead (or in addition to) information on the user's subjective evaluation related to blood glucose levels, or information on blood glucose levels estimated from user information 122, etc., may be input. Furthermore, the blood glucose level prediction information output unit 523 is not necessarily required to be input with information on blood glucose levels 123. In other words, the blood glucose level prediction information output unit 523 may be configured to output blood glucose level prediction information 140 based on at least user information 122 and meal information 124. The blood glucose level prediction information output unit 523 is not limited to a pre-trained learner; it is configured to output blood glucose level prediction information 140 based on this input information. For example, the blood glucose level prediction information output unit 523 may be configured to output blood glucose level prediction information 140 based on a rule-based system. Furthermore, the blood glucose level prediction information output unit 523 does not necessarily need to receive at least one of the behavioral information 125, gut microbiota information 126, and genetic information 130 as input.

[0071] The evaluation unit 524 evaluates the implementation status of the recommended information 127 based on the recommended information 127 from the recommended information output unit 522 and the recommended information implementation information 150 input by the user.

[0072] <Overview of Information Processing System Operation> Next, the overview of the operation of the information processing system 1 in this embodiment will be described. Figure 5 is a diagram showing an example of the overview of the operation of the information processing system 1 when outputting the first recommended information.

[0073] As shown in Figure 5, for example, before user A eats a meal, user A inputs meal information 124 about the meal to be eaten into the user terminal 100. The input meal information 124 is transmitted to the server 500. The recommendation information output unit 522 outputs first recommendation information 127 based on the input meal information 124, user information 122 of user A stored in the user DB 526, information on blood glucose levels 123, behavioral information 125, and gut microbiota information 126. The first recommendation information 127 output from the recommendation information output unit 522 is transmitted to the user terminal 100 via the network 700 and presented to the user.

[0074] The blood glucose information 123 input to the recommendation information output unit 522 may be, for example, the blood glucose levels of user A measured immediately before, in the past day, in the past few days, in the past few weeks, or in the past year. For example, the blood glucose information 123 input to the recommendation information output unit 522 may be fasting blood glucose levels, HbA1c, etc., measured within the past year. In addition, the behavioral information 125 input to the recommendation information output unit 522 may be information about user A's actions (e.g., exercise) performed immediately before, in the past day, in the past few days, or in the past few weeks. Furthermore, in addition to meal information 124 about the meal to be consumed, the recommendation information output unit 522 may also receive meal information and supplement information about meals consumed by the user within the most recent predetermined time. For example, if the meal to be consumed is dinner, the meal information for meals consumed within the most recent predetermined time may include meal information for "lunch," "snack after lunch," and "breakfast" consumed by the user.

[0075] When meal information 124 regarding a meal to be consumed is input, the recommendation information output unit 522 outputs, for example, improved meal information indicating an improved meal as first recommendation information 127. The improved meal is a meal that has been improved so that user A's blood glucose level is lower than when user A consumed the meal corresponding to the meal information 124. For example, the improved meal may include a first improved meal obtained by improving the meal corresponding to the meal information 124 using a first method, and a second improved meal obtained by improving the meal corresponding to the meal information 124 using a second method. The first method may be, for example, reducing the amount of a specific nutrient (for example, carbohydrates). The second method may be, for example, substituting a certain ingredient with a similar low-carbohydrate ingredient. For example, the second method may be substituting regular pasta with low-carbohydrate pasta.

[0076] The recommendation information output unit 522 may output recommended behavior information indicating recommended actions to lower blood glucose levels as the first recommendation information 127, in addition to (or instead of) the improved diet information. In addition to (or instead of) the improved diet information and recommended behavior information, the recommendation information output unit 522 may output recommended supplement information indicating supplements recommended to lower blood glucose levels as the first recommendation information 127.

[0077] Furthermore, information on undesirable foods, supplements, and behaviors for each user may be stored in the server 500, and information on foods, supplements, and behaviors excluding these may be presented to the user as the first recommended information 127.

[0078] Meanwhile, the blood glucose level prediction information output unit 523 outputs blood glucose level prediction information 140 based on the input meal information 124, user information 122 of user A stored in the user DB 526, blood glucose level information 123, behavioral information 125, gut microbiota information 126, and genetic information 130, for when user A consumes a meal corresponding to the meal information 124. The blood glucose level prediction information 140 based on the input meal information 124 is presented to user A.

[0079] For example, the blood glucose prediction information 140 may be the blood glucose level predicted after a predetermined time has elapsed since the meal (for example, several tens of minutes to several hours). Alternatively, the blood glucose prediction information 140 may also be information regarding the average, maximum, minimum blood glucose levels, blood glucose elevation curve, area under the blood glucose elevation curve (AUC), etc., predicted for a predetermined period after the meal (for example, the period from the meal to two hours later).

[0080] Furthermore, the blood glucose level prediction information output unit 523 outputs blood glucose level prediction information 140' based on the first recommendation information 127 output from the recommendation information output unit 522. For example, if improved meal information is output from the recommendation information output unit 522 as the first recommendation information 127, the improved meal information is input to the blood glucose level prediction information output unit 523. Based on the improved meal information and various information stored in the user DB 526, the blood glucose level prediction information 140' outputs blood glucose level prediction information 140' for when user A consumes the improved meal corresponding to the improved meal information.

[0081] Then, the first recommendation information 127 output from the recommendation information output unit 522 and the blood glucose level prediction information 140' based on the first recommendation information 127 are presented to user A.

[0082] As described above, when the information processing system 1 of this embodiment receives meal information 124 about a meal that the user is about to eat, it outputs blood glucose level prediction information 140 for when the user eats a meal corresponding to the meal information 124. This allows the user to, for example, check how much their blood glucose level will rise after eating based on the meal they are about to eat, and to change the content of the meal they eat if it is predicted that their blood glucose level will rise. In addition, the information processing system 1 outputs first recommendation information 127 for improving the user's blood glucose level based on the input meal information 124, and also outputs blood glucose level prediction information 140' based on the first recommendation information 127. This allows the user to know how much their blood glucose level will rise (how much the rise in blood glucose level can be suppressed) if they eat or act according to the recommended diet or behavior. For example, the user can find out about improved diets that can suppress the rise in blood glucose level, and how much the rise in blood glucose level can be suppressed when they eat improved diets.

[0083] Next, we will explain how to output the second recommendation information using the information processing system 1. Figure 6 shows an example of the operation overview of the information processing system 1 when outputting the second recommendation information.

[0084] As shown in Figure 6, the recommendation information output unit 522 outputs second recommendation information 127 based on, for example, user information 122 of user A stored in the user DB 526, information on blood glucose levels 123, meal information 124, behavioral information 125, and gut microbiota information 126. The recommendation information output unit 522 may, for example, automatically output the second recommendation information 127 at a predetermined time in the morning, or it may output the second recommendation information 127 based on instructions from user A. The second recommendation information 127 is transmitted to the user terminal 100 and presented to user A.

[0085] The blood glucose information 123 input to the recommendation information output unit 522 may, for example, be the blood glucose levels of user A measured within the past day, past few days, past weeks, or past year. For example, the blood glucose information 123 input to the recommendation information output unit 522 may be fasting blood glucose levels, HbA1c, etc., measured within the past year. The meal information 124 input to the recommendation information output unit 522 may also be meal information relating to meals consumed by user A within the past day, past few days, or past weeks. The behavioral information 125 input to the recommendation information output unit 522 may also be behavioral information relating to user A's actions (e.g., exercise) performed within the past day, past few days, or past weeks.

[0086] The second recommendation information 127 may, for example, be information about meals, supplements, and activities recommended for user A today. For example, the second recommendation information 127 may include information about breakfast, lunch, and dinner recommended for user A. Furthermore, the second recommendation information 127 may include information about exercise recommended for user A.

[0087] Furthermore, information on undesirable foods, supplements, and behaviors for each user may be stored in the server 500, and information on foods, supplements, and behaviors excluding these may be presented to the user as second recommendation information 127.

[0088] As shown by the dashed line in Figure 6, the second recommendation information 127 output from the recommendation information output unit 522 is input to the blood glucose level prediction information output unit 523, and blood glucose level prediction information 140' based on the second recommendation information 127 may be presented to user A together with the second recommendation information 127. The blood glucose level prediction information 140' may be, for example, the predicted blood glucose level after a predetermined time has elapsed if user A consumes each meal according to the second recommendation information 127.

[0089] Furthermore, after the second recommendation information 127 is presented, user A inputs recommendation information implementation information 150 into the user terminal 100, which indicates the status of implementation of the second recommendation information 127. For example, user A inputs information as recommendation information implementation information 150 indicating the extent to which each meal or activity recommended in the second recommendation information 127 has been implemented. This recommendation information implementation information 150 is stored in the user DB 526.

[0090] The evaluation unit 524 evaluates the input recommended information implementation information 150 and presents the evaluation results to user A.

[0091] <Processing Flow and Screen Examples of Information Processing System 1> Next, we will explain the processing flow in Information Processing System 1, as well as an example of a screen displayed on the user terminal 100.

[0092] Figure 7 is a diagram showing the processing flow of the information processing system 1, and illustrates the processing flow at the user terminal and server over a one-day period.

[0093] As shown in Figure 7, the server 500 transmits second recommendation information to the user terminal 100 at a predetermined time in the morning (step S1). The user terminal 100 receives the second recommendation information and displays it on its screen. The second recommendation information includes, for example, information on the content and quantity of meals recommended to the user for breakfast, lunch, and dinner, and information on exercise recommended to the user for the day.

[0094] Figure 8 shows an example of a screen displayed on the user terminal 100 when the second recommendation information is transmitted in step S1. As shown in Figure 8, on the user terminal 100, for example, advice on yesterday's results is displayed, and "Today's Recommendation" is displayed as the second recommendation information. "Today's Recommendation" is information about the meals the user should eat today, and includes information about recommended meals for breakfast, lunch, and dinner. "Today's Recommendation" also includes information about recommended actions for today. For example, "Walk 10,000 steps" is displayed as an example of today's recommended action. Note that multiple recommended meals and recommended actions may be displayed for each meal.

[0095] Furthermore, supplements may be presented as a second recommendation on the "Today's Recommendation" screen in Figure 8. Users may also be able to order the presented supplements on that screen.

[0096] Returning to Figure 7, the user inputs meal information about the meal they are about to eat into the user terminal 100 before consuming the meal (step S2). The meal information may be entered in any format. For example, the user may input menu information or information about each food item of the meal they will be consuming. Alternatively, the user may input meal information by taking an image of the meal with a camera. For example, the user terminal 100 may determine the contents of the meal and generate meal information based on the image it has taken of the meal. The entered meal information is transmitted to the server 500.

[0097] The server 500 generates blood glucose level prediction information based on the meal information from the user terminal 100 (step S3). The server 500 also generates first recommendation information based on the meal information from the user terminal 100 (step S4). Specifically, the blood glucose level prediction information output unit 523 generates blood glucose level prediction information based on the meal information about the input meal, assuming the user consumes that meal. Furthermore, if, for example, the blood glucose level prediction information assuming the user consumes that meal meets certain conditions, the recommendation information output unit 522 generates first recommendation information regarding an improved meal that is an improvement on the input meal. For example, if the predicted blood glucose level when the user consumes the input meal exceeds a predetermined value, the recommendation information output unit 522 generates first recommendation information regarding an improved meal that is an improvement on the input meal. Furthermore, the blood glucose level prediction information output unit 523 generates blood glucose level prediction information based on the first recommendation information. The blood glucose level prediction information based on the input meal information, the first recommendation information, and the blood glucose level prediction information based on the first recommendation information are then transmitted to the user terminal 100.

[0098] Figure 9 shows an example of the input screen for meal information in step S2, and the screen displayed on the user terminal 100 when "blood glucose level prediction information" and "first recommendation information" corresponding to the meal information are output in steps S3 and S4.

[0099] As shown in Figure 9(a), on the meal information input screen, the user inputs meal information about the meal they will be eating, such as the type and amount of staple food, the type and amount of main dish, the type and amount of side dishes, and the type and amount of other foods. The method of inputting meal information may be by the user directly typing text, or by the user selecting from pre-prepared options. Alternatively, meal information may be input by the user taking a picture of the meal with a camera.

[0100] In addition, in the meal information input screen shown in Figure 9(a), the meal recommended by "Today's Recommendation" is initially entered, and the user may input meal information by modifying this.

[0101] When meal information is entered, blood glucose level prediction information based on the entered meal information is displayed, for example, as shown in the Recommended Information Output Screen Example 1 in Figure 9 (b-1). For example, the graph shown by the solid line in Figure 9 (b-1) is displayed as blood glucose level prediction information based on the entered meal information. This solid line graph shows the time change of the predicted blood glucose level within a few hours after eating. In addition to the graph showing the predicted blood glucose level, a graph shown by a dashed line is also displayed. This dashed line graph shows the time change of the normal blood glucose level within a few hours after eating. Here, the normal blood glucose level may be the average of the user's post-meal blood glucose levels over a predetermined period in the past (for example, the past few days, past weeks, past months, etc.). If the blood glucose level prediction information based on the entered meal information meets certain conditions (for example, if the blood glucose level predicted based on the entered meal information is higher than the normal blood glucose level), a warning such as "Your blood glucose level may rise higher than usual" is displayed. In this case, as the first recommendation information, improved meal information about an improved meal that improves upon the entered meal is displayed. For example, "Improved Meal 1" is displayed as improved meal information. For example, if "250g of rice" is entered as meal information, "200g of rice" will be displayed as "Improved Meal 1." Furthermore, a graph (dashed line) predicting the blood glucose level when the user consumes this Improved Meal 1 will be displayed. Since the dashed line graph is below the solid line graph, it indicates that consuming Improved Meal 1 will suppress the rise in blood glucose levels more than consuming the entered meal. This encourages the user to consume Improved Meal 1. In addition, since the user's normal post-meal blood glucose level trend, the predicted blood glucose level information based on the entered meal, and the predicted blood glucose level information based on Improved Meal 1 are displayed, the user can easily compare these and be motivated to choose Improved Meal 1.

[0102] Furthermore, if meal information is entered, as shown in example 2 of the recommended information output screen in (b-2) of Figure 9, symptoms that appear due to elevated blood glucose levels (e.g., drowsiness, fatigue, decreased concentration, etc.) may be displayed as blood glucose level prediction information based on the entered meal information. For example, if the blood glucose level prediction information based on the entered meal information meets certain conditions (for example, if the blood glucose level predicted based on the entered meal information is higher than the normal blood glucose level), a warning such as "Your blood glucose level may be higher than usual" will be displayed, similar to the above. In this case, the first recommendation information will also be displayed. For example, improved meal 1 will be displayed as the first recommendation information, along with recommended actions. For example, if an increase in blood glucose levels is predicted around 2 p.m., the recommended action may be "Set an alarm for around 2 p.m."

[0103] Furthermore, in the recommendation information output screen shown in Figure 9 (b-1) or (b-2), supplements may be presented as the first recommendation. Users may also be able to order the presented supplements on that screen.

[0104] Users consume meals by referring to blood glucose level prediction information based on the entered meal information, first recommendation information, and blood glucose level prediction information based on the first recommendation information. For example, users may eat the recommended improved diet or eat their preferred diet which differs from the improved diet.

[0105] Furthermore, if the blood glucose level prediction information based on the entered meal information does not meet certain conditions (for example, if the predicted blood glucose level based on the entered meal information is lower than the normal blood glucose level), the first recommendation information does not need to be displayed. For example, if a user enters a meal that matches the second recommendation information as the meal they will eat, the rise in blood glucose level predicted after the meal will be suppressed. In this case, the first recommendation information does not need to be displayed.

[0106] Returning to Figure 7, after the user has consumed a meal, they input meal information about the meal into the user terminal 100 (step S5). For example, an input screen similar to that in Figure 9(a) is displayed, and the user inputs meal information about the meal they consumed on this input screen. The entered meal information is transmitted to the server 500 and stored in the user DB 526.

[0107] Furthermore, the user inputs blood glucose level evaluation information, for example, after consuming a meal (for example, two hours after eating) (step S6).

[0108] Figure 10 shows an example of the input screen for blood glucose level evaluation information in step S6. For example, on the input screen for blood glucose level evaluation information, as shown in Figure 10(a), the blood glucose level evaluation information may include the measurement result of the blood glucose level from the measuring device 101 and the time the blood glucose level was measured. The input blood glucose level evaluation information is transmitted to the server 500 and stored in the user DB 526. The server 500 performs an evaluation of the input blood glucose level evaluation information and transmits the evaluation result to the user terminal 100. The evaluation result is displayed on the user terminal 100. For example, the input blood glucose level may be evaluated and displayed on a scale of A to D. Also, as shown in Figure 10(b), the blood glucose level evaluation information may include subjective evaluations by the user, such as drowsiness or fatigue felt by the user. For example, if the user did not feel drowsy after a meal, "A" is entered, and if they felt very drowsy, "D" is entered. Furthermore, subjective information from the user, as shown in Figure 10(c), may also be entered as blood glucose level evaluation information.

[0109] Returning to Figure 7, steps S2 to S6 described above are performed with each meal. Steps S5 and S6 may be performed simultaneously. For example, steps S5 and S6 may be performed before going to bed.

[0110] Next, the user enters information on the implementation of recommended information, for example, after dinner or before going to bed (step S7). The information on the implementation of recommended information includes information indicating the implementation status of the "second recommended information" in step S1. The information on the implementation of recommended information may also include information indicating the implementation status of the "first recommended information" in step S4. The information on the implementation of recommended information is transmitted to the server 500.

[0111] Next, the server 500 generates evaluation information for the recommended information implementation information (step S8) and transmits it to the user terminal 100.

[0112] Figure 11 shows an example of the input screen for recommended information implementation in step S7 and the screen displayed on the user terminal 100 when evaluation information is generated in step S8.

[0113] As shown in Figure 11(a), the Recommended Information Implementation Information Input Screen allows users to input their implementation status of the second recommended information (today's recommendations) output in step S1. For example, the Recommended Information Implementation Information may include information indicating the extent to which the recommended meals were followed at each meal. It may also include information indicating the content and quantity of meals consumed by the user. Furthermore, it may include information indicating the extent to which recommended behaviors were followed. It may also include information indicating the content and quantity (time, distance, and number of times) of actions (exercises) performed by the user. Additionally, information regarding alternative behaviors to recommended behaviors may be entered. The Recommended Information Implementation Information may be entered by the user directly typing text, or by other methods. For example, Recommended Information Implementation Information for recommended meals may be entered by the user taking a picture of the meal immediately before consumption with a camera. Alternatively, the exercise performed by the user may be calculated based on data from a device that detects the user's movements, and this calculated exercise may be entered as Recommended Information Implementation Information for recommended behaviors.

[0114] When recommended information implementation information is entered, the recommended information implementation information is sent to the server 500. The server 500 evaluates the recommended information implementation information and sends the evaluation result to the user terminal 100. As a result, an evaluation result screen, such as the one shown in Figure 11(b), is displayed. For example, the evaluation result screen may display information such as whether the user achieved the recommended meal and the degree of achievement. It may also display information such as whether the user achieved the recommended action and the degree of achievement. Furthermore, if the user performs an alternative action different from the recommended action, the evaluation result including that alternative action will be displayed. For example, if the recommended action is "walk 10,000 steps," and the user walked 5,000 steps and jogged for 10 minutes today, the 10 minutes of jogging will be evaluated as an alternative action. The actual actions performed by the user (5,000 steps walking and 10 minutes of jogging) are compared with the recommended action (10,000 steps walking), and the evaluation result for the recommended action is displayed. For example, if the sum of the estimated calories burned based on the number of steps taken and the estimated calories burned based on the alternative activity is greater than or equal to the estimated calories burned by performing the recommended activity "walk 10,000 steps," an evaluation result indicating that today's recommended activity has been achieved may be displayed.

[0115] Furthermore, the recommended information implementation information may include the user's implementation status regarding the first recommended information. For example, if the first recommended information is information about an improved diet that the user is about to consume, the user may enter the implementation status of that improved diet on the recommended information implementation information input screen. Also, if the user consumes a substitute diet for the improved diet, information about that substitute diet may be entered as recommended information implementation information.

[0116] Furthermore, the user terminal 100 can display various information, not limited to the information described above. For example, the user terminal 100 may access the server 500 at appropriate times and display various reports. For instance, the user terminal 100 may output reports showing the changes in user information 122 (such as weight and BMI) and blood glucose levels over a predetermined period in the past (for example, the past day, several days, several weeks, or one month).

[0117] <Retraining of Recommendation Information Output Unit 522 and Blood Glucose Prediction Information Output Unit 523> Next, the retraining of the Recommendation Information Output Unit 522 and the Blood Glucose Prediction Information Output Unit 523 will be explained. As described above, the Recommendation Information Output Unit 522 is trained to output recommendation information 127 (first recommendation information and second recommendation information) to improve the user's blood glucose level based on pre-prepared training data from a large number of subjects. The Blood Glucose Prediction Information Output Unit 523 is also trained to output blood glucose prediction information 140 based on pre-prepared training data from a large number of subjects. In order to improve the accuracy of the recommendation information 127 and the blood glucose prediction information 140, the information processing system 1 of this embodiment obtains feedback from the user and retrains the Recommendation Information Output Unit 522 and the Blood Glucose Prediction Information Output Unit 523.

[0118] Figure 12 shows an overview of the retraining of the recommendation information output unit 522 and the blood glucose level prediction information output unit 523.

[0119] As described above, after the recommendation information 127 (first recommendation information, second recommendation information) is presented to the user, the user inputs recommendation information implementation information 150, which indicates the status of implementation of the recommendation information 127. The recommendation information implementation information 150 includes, for example, information indicating whether the user has implemented the meals or actions recommended by the recommendation information 127 (first recommendation information and second recommendation information), and to what extent. For example, the user terminal 100 prompts the user to input the recommendation information implementation information 150 at predetermined times (for example, after each meal or before going to bed), and the user inputs the recommendation information implementation information 150 using the user terminal 100.

[0120] Furthermore, as described above, the user inputs blood glucose evaluation information 160. For example, blood glucose levels may be measured regularly and continuously by the measuring device 101, and the measured blood glucose levels may be automatically input as blood glucose evaluation information 160. Alternatively, blood glucose levels measured using the measuring device 101 a predetermined time after the user has consumed a meal may be input as blood glucose evaluation information 160. In addition, the user may input information regarding their subjective evaluation, such as the degree and type of postprandial drowsiness caused by the rise in blood glucose levels, as blood glucose evaluation information 160. Furthermore, objective indicators other than blood glucose levels, such as the user's postprandial heart rate, may also be input as blood glucose evaluation information 160.

[0121] The recommendation information 127 output from the recommendation information output unit 522, the recommendation information implementation information 150 corresponding to the recommendation information 127, the blood glucose level evaluation information 160, and the various input information used to output the recommendation information 127 (user information 122, blood glucose level information 123, meal information 124, behavioral information 125, and gut microbiota information 126) are stored in the server 500 as training data 525. Then, at an appropriate timing, the recommendation information output unit 522 is retrained using the training data 525.

[0122] Furthermore, the blood glucose prediction information 140 output from the blood glucose prediction information output unit 523, the blood glucose evaluation information 160 corresponding to the blood glucose prediction information 140, and the various input information used to output the blood glucose prediction information 140 (user information 122, blood glucose information 123, diet information 124, behavioral information 125, gut microbiota information 126, and genetic information 130) are stored in the server 500 as training data 529. Then, at an appropriate timing, the blood glucose prediction information output unit 523 is retrained using the training data 529.

[0123] By using such training data 525 and 529, the recommendation information output unit 522 and the blood glucose level prediction information output unit 523 are retrained, resulting in the output of more accurate blood glucose level prediction information 140, and more favorable recommendation information 127 (first recommendation information, second recommendation information) for the user.

[0124] <Processing Details> Next, we will describe the processing details in the information processing system 1 of this embodiment. First, we will describe the terminal processing performed at the user terminal 100, and then we will describe the server processing performed at the server 500.

[0125] <Terminal Processing> Figure 13 is a flowchart showing an example of terminal processing performed at the user terminal 100 of the information processing system 1 of this embodiment.

[0126] In the following description, it will be explained that the control unit 11 of the user terminal 100 executes the terminal program 121 to perform the processes shown in Figure 13. Note that at least a portion of the processes shown in Figure 13 may be executed not only by the control unit 11, but also by other processors or other devices. Furthermore, at least a portion of the processes shown in Figure 13 may be executed by the server 500. Also, the processes shown in Figure 13 are merely examples, and the order and content of the processes may be changed.

[0127] As shown in Figure 13, the user terminal 100 first performs information registration and update processing (step S100). In step S100, registration processing for users not registered in the information processing system 1 and update processing for information of registered users are performed. In the registration process, the user inputs user information 122, which includes their own physical information, information on blood glucose levels 123, gut microbiota information 126, and genetic information 130 into the user terminal 100. The input information is sent to the server 500 and stored in the user DB 526. In the update process, this registered information is updated based on the user's instructions. For example, if there is a change in the user's physical information, the user uses the user terminal 100 to activate the user information update screen and update the user information 122.

[0128] Next, the user terminal 100 performs the process of acquiring and displaying the second recommendation information (step S101). Specifically, the user terminal 100 determines whether or not it has received the second recommendation information indicating "Today's Recommendation" from the server 500, and if it has received it, it displays "Today's Recommendation". As a result, for example, the screen shown in Figure 8 is displayed. For example, the user terminal 100 may receive the second recommendation information indicating "Today's Recommendation" from the server 500 at a predetermined time in the morning and present it to the user.

[0129] Next, the user terminal 100 performs pre-meal input processing (step S102). Here, the user terminal 100 displays, for example, the meal information input screen shown in Figure 9(a), based on the user's instructions. The user uses the meal information input screen to input meal information about the meal to be consumed. The entered meal information is transmitted to the server 500. The user terminal 100 may also determine whether it is mealtime or not, and if it is mealtime, it may prompt the user to input meal information about the meal to be consumed using the meal information input screen.

[0130] Next, the user terminal 100 performs the first recommendation information acquisition and display processing (step S103). Here, the user terminal 100 determines whether it has received response information from the server 500 corresponding to the meal information entered in step S102, and if it has received it, it displays information according to the response information. The response information includes blood glucose level prediction information based on the entered meal information, first recommendation information based on the meal information, and blood glucose level prediction information based on the first recommendation information. The first recommendation information may include, for example, improved meal information regarding an improved meal that is an improvement on the meal information, and recommended action information regarding recommended actions according to the meal information. In step S103, for example, the recommendation information output screen shown in (b-1) and (b-2) of Figure 9 is displayed.

[0131] Next, the user terminal 100 performs post-meal input processing (step S104). Here, the user inputs information about the meal they have consumed. In step S104, for example, a meal information input screen similar to that in step S102 is displayed, and the meal information of the consumed meal is entered. For example, the user terminal 100 may display the meal information input screen a predetermined time after the meal and prompt the user to input information about the meal they have consumed.

[0132] Next, the user terminal 100 performs a blood glucose level evaluation information input process (step S105). Here, the user terminal 100 displays, for example, the blood glucose level evaluation information input screen shown in Figure 10, based on the user's instructions. For example, the user may input the blood glucose level measured by the measuring device 101 on the blood glucose level evaluation information input screen (Figure 10(a)), or input information indicating a subjective evaluation related to the blood glucose level (Figures 10(b), (c)). For example, the user terminal 100 may display the blood glucose level evaluation information input screen after a predetermined time has passed since the meal and prompt the user to input the blood glucose level evaluation information. Note that, for example, steps S104 and S105 may be performed simultaneously. For example, in the post-meal input process, meal information of the meal consumed by the user and blood glucose level evaluation information may be input.

[0133] Next, the user terminal 100 performs the recommended information implementation information input process (step S106). Here, the user inputs recommended information implementation information indicating the status of implementation of the second recommended information. For example, the user terminal 100 may determine whether it is a predetermined time (for example, a predetermined time at night), and if it is the predetermined time, it may display the recommended information implementation information input screen shown in Figure 11(a) to prompt the user to input recommended information implementation information for the second recommended information. Also, in step S106, the user inputs recommended information implementation information indicating the status of implementation of the first recommended information. For example, the user terminal 100 may display a recommended information implementation information input screen at a predetermined time after a meal to prompt the user to input the status of implementation of the first recommended information. The input recommended information implementation information is transmitted to the server 500.

[0134] Next, the user terminal 100 performs evaluation information acquisition and display processing (step S107). Here, when recommended information implementation information is entered, the user terminal 100 determines whether or not it has received evaluation information from the server 500, and if it determines that it has received it, it displays the evaluation result screen. As a result, for example, the evaluation result screen shown in Figure 11(b) is displayed.

[0135] If the user terminal 100 has performed the process in step S107, it will perform the process in step S100 again.

[0136] <Server Processing> Figure 14 is a flowchart showing an example of server processing performed in the server 500 of the information processing system 1 of this embodiment.

[0137] In the following description, the control unit 51 of the server 500 will execute the information processing program 520 to perform the processes shown in Figure 14. Note that the processes shown in Figure 14 may be executed by multiple servers. Furthermore, at least a portion of the processes shown in Figure 14 may be executed by other processors, not just the control unit 51. Also, at least a portion of the processes shown in Figure 14 may be executed by the user terminal 100. Additionally, the processes shown in Figure 14 are merely examples, and the order and content of the processes may be changed as appropriate.

[0138] As shown in Figure 14, the server 500 performs information registration and update processing (step S500). In this step, in accordance with the processing in step S100, the server 500 performs registration processing for users not yet registered and update processing for information about registered users. Specifically, when the server 500 receives a request for registration or update processing from the user terminal 100, it registers a new user in the user DB 526 or updates the information of registered users stored in the user DB 526 based on the information received from the user terminal 100.

[0139] Next, the server 500 performs a second recommendation information transmission process (step S501). Here, the server 500 generates second recommendation information indicating "Today's Recommendation" for each user and transmits it to each user terminal 100. For example, the server 500 generates the second recommendation information at a predetermined time in the morning and transmits it to each user's user terminal 100. The server 500 may generate the second recommendation information at different times for each user and transmit it to the user terminal 100. Specifically, the server 500 inputs user information 122, blood glucose level information 123, meal information 124, behavior information 125, and gut microbiota information 126 stored in the user DB 526 to the recommendation information output unit 522, and obtains the output from the recommendation information output unit 522 as the second recommendation information. The meal information 124 and behavior information 125 input to the recommendation information output unit 522 may be meal information 124 and behavior information 125 stored in the past day or past few days.

[0140] Next, the server 500 performs a first recommendation information transmission process (step S502). Here, when the server 500 receives meal information entered in the pre-meal input process in step S102 from the user terminal 100, it sends response information including the first recommendation information to the user terminal 100. Specifically, the server 500 inputs the received meal information into the blood glucose level prediction information output unit 523 and obtains blood glucose level prediction information based on the meal information from the blood glucose level prediction information output unit 523. The server 500 also inputs the received meal information into the recommendation information output unit 522 and obtains first recommendation information based on the meal information from the recommendation information output unit 522. The first recommendation information may include, for example, improved meal information regarding an improved meal that is an improvement on the meal information, and recommended behavior information regarding recommended behaviors in accordance with the meal information. The first recommendation information may also include information on supplements recommended to lower blood glucose levels. Furthermore, the server 500 inputs the first recommendation information to the blood glucose level prediction information output unit 523 and obtains blood glucose level prediction information based on the first recommendation information from the blood glucose level prediction information output unit 523. The server 500 then transmits response information to the user terminal 100, which includes the blood glucose level prediction information based on the meal information, the first recommendation information, and the blood glucose level prediction information based on the first recommendation information. The meal information input to the recommendation information output unit 522 and the blood glucose level prediction information output unit 523 may include meal information related to the meal the user has most recently consumed. Also, the behavior information 125 input to the recommendation information output unit 522 and the blood glucose level prediction information output unit 523 may be behavior information 125 stored within the most recent, past day, past few days, or past few weeks.

[0141] Next, the server 500 performs post-meal information reception processing (step S503). Here, the server 500 receives the meal information entered in the post-meal input processing in step S104 from the user terminal 100. The received meal information, along with the date and time, is stored in the user DB 526.

[0142] Next, the server 500 performs a blood glucose level evaluation information reception process (step S504). Here, the server 500 receives the blood glucose level evaluation information entered in the blood glucose level evaluation information input process in step S105 from the user terminal 100. The received blood glucose level evaluation information, along with the date and time, is stored in the user DB 526. This blood glucose level evaluation information is used for relearning the blood glucose level prediction information output unit 523.

[0143] Next, the server 500 performs a recommended information implementation information reception process (step S505). Here, the server 500 receives the recommended information implementation information input in the recommended information implementation information input process in step S106 from the user terminal 100. The received recommended information implementation information is stored in the user DB 526. This recommended information implementation information is used for retraining of the recommended information output unit 522.

[0144] Next, the server 500 performs evaluation information output processing (step S506). Here, the evaluation unit 524 of the server 500, upon receiving the recommended information implementation information input in step S106 from the user terminal 100, performs an evaluation of the recommended information implementation information. The server 500 transmits the evaluation result of the recommended information implementation information to the user terminal 100 as evaluation information.

[0145] Next, the server 500 performs a retraining process (step S507). Here, the server 500 retrains the blood glucose level prediction information output unit 523 based on the blood glucose level evaluation information acquired in step S504. The server 500 also retrains the recommendation information output unit 522 based on the recommendation information implementation information acquired in step S505.

[0146] If the server 500 has performed the process in step S507, it will execute the process in step S500 again.

[0147] As described above, the information processing system 1 of this embodiment, when inputting meal information related to a meal consumed by the user, presents the user with blood glucose level prediction information based on the meal consumed by the user. The information processing system 1 also presents the user with first recommendation information, including improved meals and behaviors, and blood glucose level prediction information based on the first recommendation information. This allows the user to know, for example, whether or not their blood glucose level will rise after consuming a meal, and how their blood glucose level will improve if they take the improved meals or behaviors in accordance with the first recommendation information.

[0148] Furthermore, the information processing system 1 can output second recommendation information based on user information, information on blood glucose levels, meal information regarding meals consumed by the user, and behavioral information regarding actions taken by the user. The second recommendation information may, for example, be the content of meals and actions recommended for the user for each meal today. This allows the user to plan meals and actions to suppress the rise in blood glucose levels.

[0149] Furthermore, the information processing system 1 acquires blood glucose level evaluation information and recommended information implementation information from the user, and retrains the blood glucose level prediction information output unit 523 and the recommended information output unit 522 based on the acquired blood glucose level evaluation information and recommended information implementation information. This improves the accuracy of the blood glucose level prediction information and recommended information.

[0150] <Information Processing System of the Second Embodiment> Next, the information processing system 2 of the second embodiment will be described. When the blood glucose level prediction information output unit 523 of the information processing system 1 above receives meal information 124 about a meal that the user is about to eat, it outputs a prediction of the blood glucose level when the user eats the meal. In the second embodiment, when meal information about a meal that the user is about to eat is received, information is output indicating whether or not a blood glucose spike will occur when the user eats the meal.

[0151] Figure 15 is a diagram showing an overview of the information processing system 2 of the second embodiment. The information processing system 2 of the second embodiment includes a blood glucose level prediction information output unit 530 and a meal suggestion unit 531. The information processing system 2 of the second embodiment also includes the user terminal 100 and server 500 as a physical configuration. The control units of the user terminal 100 and server 500 execute predetermined programs to configure each part of Figure 15.

[0152] When meal information is input, the blood glucose level prediction information output unit 530 outputs blood glucose spike information 141 as blood glucose level prediction information, indicating whether or not a blood glucose spike will occur. Here, the blood glucose level prediction information output unit that predicts whether or not a blood glucose spike will occur is referred to as the "blood glucose spike prediction model 530". The information processing system 2 of the second embodiment is a system that proposes an improved diet using the blood glucose spike prediction model 530.

[0153] Specifically, as shown in Figure 15, the user inputs meal information 133 about the meal to be consumed into the user terminal 100. Meal information 133 is, for example, menu information about the meal menu. Menu information includes information about the type and amount of staple food, information about the type and amount of main dish, and information about the type and amount of side dishes. For example, the information about the type and amount of staple food is "rice 250g", and the information about the type and amount of main dish is, for example, "pork cutlet 200g". The input meal information 133 is transmitted to the server 500 via the network. Alternatively, instead of meal information 133, meal information similar to the meal information 124 in Figure 5 may be input into the blood glucose spike prediction model 530.

[0154] The server 500 calculates the amount of nutrients by food group based on the standard table of food composition for the received meal information 133.

[0155] The blood glucose spike prediction model 530 outputs blood glucose spike information 141 indicating whether or not a blood glucose spike will occur when a user consumes a meal according to the meal information 133, based on the amount of nutrients by food group calculated based on the meal information 133, user information 122 stored in the user DB 526, blood glucose information 131, and behavioral information 132. Here, a blood glucose spike is a condition in which blood glucose levels rise rapidly after a meal and then rapidly fall back to their original level or below, for example, a condition in which blood glucose levels exceed 140 (mg / dl) within two hours after a meal.

[0156] Specifically, the blood glucose spike prediction model 530 is a learning device that takes user information 122, blood glucose information 131, nutrient amounts by food group, and behavioral information 132 as inputs and outputs blood glucose spike information 141 indicating whether or not a blood glucose spike will occur. For example, the blood glucose spike prediction model 530 may be constructed based on a random forest. The machine learning algorithm used in the blood glucose spike prediction model 530 is not limited to this and may be constructed using any algorithm. For example, the blood glucose spike prediction model 530 may be constructed using a multi-layered neural network.

[0157] User information 122 is the same information as user information 122 in Figure 5 described above. However, user information 122 as input information for the blood glucose spike prediction model 530 may include information different from user information 122 in Figure 5.

[0158] The blood glucose information 131 is information on blood glucose levels measured in the past, and may include, for example, fasting blood glucose levels and HbA1c measured within the past year. The blood glucose information 131 may also be the same information as the blood glucose information 123 in Figure 5 described above. Furthermore, the input blood glucose information 131 is not limited to the user's measured blood glucose levels, but may also be information based on the user's subjective evaluation. Additionally, the input blood glucose information 131 may be information on the user's estimated blood glucose levels based on user information 122 and behavioral information 132. Finally, the input blood glucose information 131 may also be the blood glucose prediction information 140 output by the information processing system 1.

[0159] Behavioral information 132 may include, for example, information indicating the user's drinking history, smoking history, and daily activity level. Drinking history may include, for example, the type and amount of alcohol consumed by the user per day, the type and amount of alcohol consumed per week, and the number of days and amount of alcohol consumed per week. Smoking history may include the number of cigarettes smoked per day, the number of cigarettes smoked per week, the number of years smoking, and the smoking index. Information indicating daily activity level may include, for example, information representing the user's habits based on questionnaires from the user. Information indicating daily activity level may also be objective information representing the user's daily activity level obtained from a device equipped with a pedometer or inertial sensor. Note that behavioral information 132 may be the same information as behavioral information 125 in Figure 5.

[0160] If the blood glucose spike prediction model 530 predicts that no blood glucose spike will occur based on the meal information 133 entered by the user, it outputs blood glucose spike information 141 indicating that no blood glucose spike will occur. This blood glucose spike information 141 is transmitted to the user terminal 100 and presented to the user.

[0161] On the other hand, if a blood glucose spike is predicted to occur based on the meal information 133, the blood glucose spike prediction model 530 outputs blood glucose spike information 141 indicating that a blood glucose spike will occur. In this case, the meal suggestion unit 531 of the server 500 generates improved meal information 170 regarding an improved meal that is an improved version of the meal indicated in the meal information 133. An improved meal is a meal that has been improved so as not to cause a blood glucose spike. For example, an improved meal may include a first improved meal, which is an improved version of the meal indicated in the meal information 133 using a first method, and a second improved meal, which is an improved version using a second method. The first method may be, for example, reducing the amount of a specific nutrient (e.g., carbohydrates). The second method may be, for example, substituting a certain ingredient with a similar low-carbohydrate ingredient. For example, the second method may be substituting regular pasta with low-carbohydrate pasta.

[0162] For example, the meal suggestion unit 531 generates improved meal information 170 on a rule-based basis. For example, if it is predicted that a blood glucose spike will occur when a user consumes a meal corresponding to the input meal information 133, the meal suggestion unit 531 generates improved meal information 170 with a reduced amount of carbohydrates in the meal based on a predetermined algorithm. The meal suggestion unit 531 may determine the amount of carbohydrate reduction based on, for example, user information 122 (e.g., height, weight, age, gender, etc.). Alternatively, the meal suggestion unit 531 may refer to a database that stores a correspondence between a certain food and a low-carbohydrate substitute food, and generate improved meal information 170 in which a specific food in the input meal is replaced with a low-carbohydrate substitute food.

[0163] Based on the improved meal information 170 output from the meal suggestion unit 531, the amount of nutrients by food group is calculated and input into the blood glucose spike prediction model 530. As a result, blood glucose spike information 141 is output indicating whether or not a blood glucose spike will occur if the user consumes the improved meal indicated in the improved meal information 170. If information on multiple improved meals, including a first improved meal and a second improved meal, is output as improved meal information 170, blood glucose spike information 141 indicating whether or not a blood glucose spike will occur is output based on each improved meal.

[0164] The information processing system 2 then presents the user with the meal information 133 and the improved meal information 170 entered by the user. If multiple improved meals, including the first improved meal and the second improved meal, are output by the information processing system 2, it visualizes and presents to the user the results of comparing each improved meal with the meal entered by the user. For example, the information processing system 2 visualizes and presents to the user the results of comparing the meal information 133 with the first improved meal. It also visualizes and presents to the user the results of comparing the meal information 133 with the second improved meal. As a result, the user can learn about the effect of suppressing the rise in blood glucose levels for each of the first and second improved meals, and can select their preferred improved meal from the first and second improved meals.

[0165] Furthermore, in addition to user information 122, blood glucose information 131, behavioral information 132, and dietary information 133 (nutrient amounts by food group calculated based on dietary information 133), the blood glucose spike prediction model 530 may also be input with the aforementioned gut microbiota information 126 and / or genetic information 130. By learning from this information as well, the accuracy of predicting the occurrence of blood glucose spikes can be improved.

[0166] Furthermore, some of the user information 122, blood glucose information 131, behavioral information 132, dietary information 133, gut microbiota information 126, and genetic information 130 may be input into the blood glucose spike prediction model 530. For example, the blood glucose spike prediction model 530 may be input with user information 122, blood glucose information 131, and dietary information 133, but behavioral information 132 may not be input. Alternatively, the blood glucose spike prediction model 530 may be input with user information 122 and dietary information 133, but blood glucose information 131 and behavioral information 132 may not be input.

[0167] In the information processing system 2, menu information is input as meal information 133, and the amount of nutrients by food group calculated based on the menu information is input to the blood glucose spike prediction model 530. In other embodiments, menu information may be input to the blood glucose spike prediction model 530 as meal information 133, and improved meal information 170 may be output.

[0168] Furthermore, the meal suggestion unit 531 may be configured as a trained model generated by machine learning. For example, the meal suggestion unit 531 may be a trained model that takes user information 122, blood glucose level information 131, and meal information 133 as inputs and outputs improved meal information 170 regarding an improved meal that has been modified to prevent blood glucose spikes.

[0169] Furthermore, the meal suggestion unit 531 may be included as part of the blood glucose spike prediction model 530 described above.

[0170] As described above, the information processing system 2 of the second embodiment predicts whether or not a blood glucose spike will occur when meal information regarding the meals consumed by the user is input, and if a blood glucose spike is predicted to occur, it proposes an improved meal. This makes it possible to suppress a rapid rise in blood glucose levels.

[0171] (Modifications) The above embodiment has been described, but the above embodiment is merely an example, and modifications such as the following may be made.

[0172] For example, in the above embodiment, when a user inputs meal information about a meal they are about to eat, if the blood glucose level prediction information for when the user eats that meal meets certain conditions, the first recommendation information is output; if the blood glucose level prediction information does not meet the certain conditions, the first recommendation information is not output. In other embodiments, when a user inputs meal information about a meal they are about to eat, the first recommendation information may be output regardless of whether or not the blood glucose level prediction information for when the user eats that meal meets certain conditions.

[0173] In the above embodiment, the recommendation information output unit 522 output improved diet information and recommended behavior information as first recommendation information. In other embodiments, the recommendation information output unit 522 may output at least one of the following as first recommendation information: improved diet information, recommended behavior information, and information on recommended supplements.

[0174] Furthermore, in the above embodiment, the information processing system 1 displayed blood glucose level prediction information based on meal information related to the meals consumed by the user, first recommendation information, and blood glucose level prediction information based on the first recommendation information on a single screen. The method of presenting this information to the user is not limited to this. For example, depending on the user's operation, this information may be displayed on separate screens, or this information may be displayed in a scrolling manner. The same applies to the second recommendation information. Similarly, in the information processing system 2, the information presented to the user may be displayed in any manner.

[0175] In the above embodiment, the user inputs an alternative action to replace the recommended action based on the recommended information, the evaluation unit evaluates the alternative action, and presents the evaluation result to the user. In another embodiment, the user inputs an alternative meal to the improved meal based on the recommended information, the evaluation unit evaluates the alternative meal, and presents the evaluation result to the user.

[0176] Furthermore, in the above embodiment, the user inputs meal information regarding the meal to be consumed, and the recommendation information output unit 522 outputs first recommendation information, which includes improved meal information regarding an improved meal that is an improvement on the meal in question, based on the meal information. In another embodiment, the user inputs meal information regarding the meal consumed after eating, and the recommendation information output unit 522 outputs first recommendation information, which includes recommended behavior information indicating recommended post-meal behaviors, based on the meal information. The first recommendation information may include recommendation information regarding the following meals.

[0177] Furthermore, in the information processing system 1 of Figure 5, the blood glucose level prediction information output unit 523 is used to predict the user's post-meal blood glucose level, and in the information processing system 2 of Figure 15, the blood glucose spike prediction model 530 is used to predict whether or not a blood glucose spike will occur after a meal. In other embodiments, the blood glucose spike prediction model 530 of Figure 15 may be used instead of the blood glucose level prediction information output unit 523 of Figure 5. In this case, based on the meal information 124 input by the user, the blood glucose spike prediction model 530 outputs blood glucose level prediction information indicating whether or not a blood glucose spike will occur and presents it to the user. If a blood glucose spike is predicted to occur, the recommendation information output unit 522 outputs improved meal information indicating an improved meal that is an improvement on the meal according to the input meal information 124 as first recommendation information 127. Furthermore, the blood glucose spike prediction model 530 outputs blood glucose level prediction information indicating whether or not a blood glucose spike will occur based on the improved meal information, and the improved meal information and the blood glucose level prediction information based on the improved meal information are presented to the user.

[0178] Furthermore, in the above embodiment, "Today's Recommendation" is output as the second recommendation information. However, in addition to this (or instead), information regarding recommended meals, supplements, and behaviors for the user over a predetermined period in the future (e.g., several days, several weeks, one month, etc.) may also be output as the second recommendation information.

[0179] Furthermore, although the above-described information processing systems 1 and 2 are composed of a user terminal 100 and a server 500, this is merely an example, and they may be composed of any hardware. For example, information processing systems 1 and 2 may be composed of a single information processing device.

[0180] Furthermore, the configurations of the above embodiments and their modified forms can be combined in any way, as long as they do not contradict each other. Also, the above is merely an example of the present invention, and various other improvements and modifications may be made.

[0181] 1, 2 Information Processing System 521 Information Acquisition Unit 522 Recommended Information Output Unit 523 Blood Glucose Level Prediction Information Output Unit 524 Evaluation Unit 530 Blood Glucose Level Spike Prediction Model

Claims

1. An information processing system that provides information for improving a user's blood glucose level, comprising: information acquisition means for acquiring user information relating to the user's physical body, information relating to the user's blood glucose level, meal information relating to the user's meals, and behavioral information relating to the user's actions; recommendation information output means for outputting recommendation information for improving the user's blood glucose level based on the user information, information relating to the user's blood glucose level, meal information, and behavioral information; blood glucose level prediction information output means for outputting blood glucose level prediction information relating to the user's blood glucose level when the user consumes a meal corresponding to the meal information, based on the user information, meal information, and behavioral information; and presentation means for presenting the recommendation information and the blood glucose level prediction information, wherein the blood glucose level prediction information output means outputs the blood glucose level prediction information corresponding to the recommendation information based on the recommendation information.

2. The information processing system according to claim 1, further comprising: a recommendation information output means which is a learner trained to take user information, blood glucose level prediction information or information relating to blood glucose levels, meal information and behavior information as inputs and output the recommendation information; a blood glucose level prediction information output means which is a learner trained to take user information, meal information and behavior information as inputs and output the blood glucose level prediction information; a second information acquisition means for acquiring recommendation information implementation information showing the user's implementation status of the recommendation information; a third information acquisition means for acquiring blood glucose level evaluation information relating to the user's evaluation of blood glucose levels in relation to the blood glucose level prediction information; a first retraining means for causing the recommendation information output means to retrain using the recommendation information and the recommendation information implementation information as training data; and a second retraining means for causing the blood glucose level prediction information output means to retrain using the blood glucose level prediction information and the blood glucose level evaluation information as training data.

3. The information processing system according to claim 1, wherein the recommended information output means includes a meal suggestion means that suggests improved meal information relating to an improved meal when the user consumes a meal corresponding to the meal information and the blood glucose level prediction information satisfies specific conditions, and the blood glucose level prediction information output means outputs blood glucose level prediction information when the user consumes an improved meal corresponding to the improved meal information.

4. The information processing system according to claim 3, wherein the presentation means presents the blood glucose level prediction information when the user consumes a meal corresponding to the meal information, and the blood glucose level prediction information when the user consumes an improved meal corresponding to the improved meal information.

5. The information processing system according to claim 3, wherein the meal suggestion means suggests, as improved meal information, first improved meal information relating to a first improved meal obtained by improving the meal by a first method, and second improved meal information relating to a second improved meal obtained by improving the meal by a second method, and the blood glucose level prediction information output means outputs first blood glucose level prediction information when the user consumes the first improved meal and second blood glucose level prediction information when the user consumes the second improved meal.

6. The information processing system according to claim 5, comprising a comparison means for comparing the first blood glucose level prediction information with the second blood glucose level prediction information, wherein the presentation means presents information according to the result of the comparison.

7. The information processing system according to claim 3, wherein satisfying the above-mentioned specific conditions means that a blood glucose spike occurs when the user consumes a meal corresponding to the meal information.

8. The information processing system according to claim 1, further comprising a second information acquisition means for acquiring recommended information implementation information indicating the user's implementation status of the recommended information, and a recommended information evaluation means for evaluating the user's implementation status of the recommended information based on the recommended information and the recommended information implementation information.

9. The information processing system according to claim 8, wherein the recommended information implementation information includes information on alternative meals or alternative behaviors that replace the meals or behaviors recommended by the recommended information, and the recommended information evaluation means evaluates the user's implementation status of the recommended information based on the alternative meals or alternative behaviors.

10. An information processing system that provides information for improving a user's blood glucose level, comprising: an information acquisition means for acquiring user information relating to the user's physical condition and meal information relating to the meals consumed by the user; a blood glucose level prediction information output means for outputting blood glucose level prediction information relating to whether or not a blood glucose spike will occur when the user consumes a meal corresponding to the meal information, based on the user information and the meal information; and a meal suggestion means for outputting improved meal information relating to an improved meal if a blood glucose spike is predicted to occur.

11. An information processing program executed by a computer of an information processing system that provides information for improving a user's blood glucose level, wherein the computer functions as: an information acquisition means for acquiring user information relating to the user's body, information relating to the user's blood glucose level, meal information relating to the user's meals, and behavioral information relating to the user's actions; a recommendation information output means for outputting recommendation information for improving the user's blood glucose level based on the user information, the information relating to the user's blood glucose level, the meal information, and the behavioral information; a blood glucose level prediction information output means for outputting blood glucose level prediction information relating to the user's blood glucose level when the user consumes a meal corresponding to the meal information, based on the user information, the meal information, and the behavioral information; and a presentation means for presenting the recommendation information and the blood glucose level prediction information, wherein the blood glucose level prediction information output means outputs the blood glucose level prediction information corresponding to the recommendation information based on the recommendation information.

12. An information processing program executed by a computer of an information processing system that provides information for improving a user's blood glucose level, wherein the computer functions as: an information acquisition means for acquiring user information relating to the user's body and meal information relating to meals consumed by the user; a blood glucose prediction information output means for outputting blood glucose prediction information relating to whether or not a blood glucose spike will occur when the user consumes a meal corresponding to the meal information, based on the user information and the meal information; and a meal suggestion means for outputting improved meal information relating to an improved meal if a blood glucose spike is predicted to occur.

13. An information processing method for providing information for improving a user's blood glucose level, comprising: an information acquisition step of acquiring user information relating to the user's physical body, information relating to the user's blood glucose level, meal information relating to the user's meals, and behavioral information relating to the user's actions; a recommendation information output step of outputting recommendation information for improving the user's blood glucose level based on the user information, information relating to the user's blood glucose level, meal information, and behavioral information; a blood glucose level prediction information output step of outputting blood glucose level prediction information relating to the user's blood glucose level when the user consumes a meal corresponding to the meal information, based on the user information, meal information, and behavioral information; and a presentation step of presenting the recommendation information and the blood glucose level prediction information, wherein in the blood glucose level prediction information output step, the blood glucose level prediction information corresponding to the recommendation information is output based on the recommendation information.

14. An information processing method for providing information for improving a user's blood glucose level, comprising: an information acquisition step of acquiring user information relating to the user's physical body and dietary information relating to meals consumed by the user; a blood glucose level prediction information output step of outputting blood glucose level prediction information relating to whether or not a blood glucose spike will occur when the user consumes a meal corresponding to the dietary information, based on the user information and the dietary information; and a meal suggestion step of outputting improved meal information relating to an improved meal if a blood glucose spike is predicted to occur.

Images