Prediction method, prediction program, and prediction device

Abstract

To provide a prediction method, a prediction program, and a prediction apparatus for predicting a state of an intestinal flora by simple means other than the examination.SOLUTION: In a prediction method, a computer acquires the result of response to a questionnaire with a plurality of question items including the question item concerning the living habit of a living body to be predicted, and based on the correlation between the response result and the examination data indicating a state of the intestinal flora pre-associated with the questionnaire, the computer is caused to execute processing for predicting the state of the intestinal flora of the living body to be predicted using the acquired response result as an input.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The disclosed technology relates to a prediction method, a prediction program, and a prediction device. [Background technology]

[0002] Conventionally, there is a technique for predicting the state of intestinal flora, which is the state of the intestinal flora in the body, by testing the intestinal flora.

[0003] For example, there is a technology that calculates an intestinal flora score using multiple indices related to intestinal bacteria (see Patent Document 1). This technology describes that information related to a user's human intestinal bacteria is obtained from a testing institution via a network. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-078273 Summary of the Invention [Problem to be solved by the invention]

[0005] However, taking a gut microbiota test requires a relatively high cost. Therefore, users willing to take the test were limited to those who were health conscious and financially well off. Therefore, even if there were users who were health conscious and wanted to understand the state of their gut flora, they did not end up taking a gut microbiota test. Another problem was that such users would not take steps to improve their lifestyle habits because they would not be able to understand the state of their gut flora unless they took a gut microbiota test.

[0006] Given the above background, there is a need for a method that can predict the state of intestinal flora using simple means other than testing.

[0007] The disclosed technology has been made in consideration of the above points, and aims to provide a prediction method, a prediction program, and a prediction device that can predict the state of intestinal flora using simple means other than testing. [Means for solving the problem]

[0008] A first aspect of the present disclosure is a prediction method in which a computer is caused to acquire response results to a questionnaire with multiple questions including questions about lifestyle habits of a living organism to be predicted, and to execute a process using the acquired response results as input to predict the state of the intestinal flora of the living organism to be predicted based on a correlation between the response results and test data indicating the state of the intestinal flora that is pre-associated with the questionnaire. [Effects of the Invention]

[0009] According to the disclosed technology, the state of intestinal flora can be predicted by simple means other than testing. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 2 is a block diagram showing the hardware configuration of a learning device and a prediction device according to the embodiment. [Figure 2] FIG. 2 is a block diagram illustrating an example of a functional configuration of a learning device according to an embodiment. [Figure 3] (A) A graph showing the distribution of intestinal flora states from actual test data, and (B) a graph showing the distribution of intestinal flora states predicted using a prediction model based on questionnaire response information. [Figure 4] 10 is a graph showing the results of a residual analysis that verified the predictive performance of the trained predictive model. [Figure 5] 1 is a block diagram showing a configuration of a prediction system including a prediction device according to an embodiment. [Figure 6] 10 is an example of a questionnaire response screen. [Figure 7] 10 is an example of a display screen showing the predicted results of the intestinal flora state and suggestions for improvement. [Figure 8] 10 is a flowchart illustrating a flow of a learning process performed by the learning device according to the embodiment. [Figure 9] 10 is a flowchart illustrating a flow of a prediction process performed by the prediction device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] An example of an embodiment of the disclosed technology will be described below with reference to the drawings. Note that the same reference numerals are used to designate identical or equivalent components and parts in each drawing. Also, the dimensional proportions in the drawings are exaggerated for the sake of explanation and may differ from the actual proportions.

[0012] An overview of an embodiment of the present disclosure will be described. As described above, there are problems with testing intestinal microbiota. Therefore, in this embodiment, the state of the intestinal flora, which is the state of the intestinal microbiota in the body, is predicted using the results of responses to a questionnaire regarding lifestyle habits. The intestinal microbiota is an example of the intestinal microbiota (normal microbiota). The intestinal microbiota here refers to the microbiota (normal microbiota) that is normally present in a part of the body. The normal intestinal microbiota can include the intestinal microbiota in the oral cavity, orbit, etc.

[0013] To predict the state of the intestinal flora, a prediction model that learns the correlation between the questionnaire responses and the test data associated with the questionnaire is used. The test data is used as numerical data (intestinal flora score) that indicates the state of the intestinal flora.

[0014] In the following description, the prediction target is the state of the intestinal flora of a human user, but the present invention is not limited to this. For example, the method of this embodiment can be similarly applied to any living organism, such as a household pet or a domestic animal, whose owner can answer a questionnaire.

[0015] The embodiments of the present disclosure will be described separately for a learning device and a prediction device.

[0016] 1 is a block diagram showing the hardware configuration of the learning device 100 and the prediction device 200. The learning device 100 and the prediction device 200 can have the same hardware configuration.

[0017] 1, learning device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each component is connected to each other via a bus 19 so that they can communicate with each other.

[0018] The CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 controls each of the above components and performs various arithmetic processing in accordance with the program stored in the ROM 12 or the storage 14. In this embodiment, a prediction program is stored in the ROM 12 or the storage 14.

[0019] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a working area. The storage 14 is configured with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including the operating system and various data.

[0020] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to perform various inputs.

[0021] The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may function as the input unit 15 by adopting a touch panel system.

[0022] The communication interface 17 is an interface for communicating with other devices such as terminals, etc. For this communication, for example, a wired communication standard such as Ethernet (registered trademark) or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark) is used.

[0023] Similarly, the prediction device 200 has a CPU 21, a ROM 22, a RAM 23, a storage 24, an input unit 25, a display unit 26, and a communication I / F 27. Each component is connected to each other via a bus 29 so that they can communicate with each other. A prediction program is stored in the ROM 22 or the storage 24. The description of each component of the hardware configuration is omitted because it is the same as that of the learning device 100. (Learning device) The following describes each functional configuration of the learning device 100. Fig. 2 is a block diagram showing the configuration of the learning device 100 of this embodiment. Each functional configuration is realized by the CPU 11 reading out a learning program stored in the ROM 12 or storage 14, expanding it in the RAM 13, and executing it.

[0024] As shown in FIG. 2, the learning device 100 functionally includes a learning data storage unit 102, a model storage unit 104, an acquisition unit 110, and a learning unit 112.

[0025] The acquisition unit 110 acquires training data and stores it in the training data storage unit 102. The training data storage unit 102 stores the training data acquired by the acquisition unit 110. The training data includes test data that quantifies the state of the intestinal flora based on multiple bacterial composition data of the intestinal flora, and responses to a questionnaire with multiple questions corresponding to the test data. The acquisition unit 110 converts the training data into features and stores them in the training data storage unit 102 to use them for training the prediction model. The source of the training data can be set as appropriate, such as an existing database constructed from test data indicating the state of the intestinal flora, an external database, or other necessary source. Intestinal flora tests are performed by testing institutions. During the test, stool samples and a questionnaire regarding lifestyle habits are administered, and test data is obtained by analyzing these. Therefore, the database from which the test data and the responses to the questionnaire at the time of the test are stored and used as training data.

[0026] The quantified test data uses the intestinal flora score, which relates to the state of the intestinal flora, described in the aforementioned Patent Document 1. The intestinal flora score described in Patent Document 1 is indicated as a score of E for less than 50 points, D for 50 to 60 points, C for 60 to 70 points, B for 65 to 70 points, and A for 70 points or more. A higher intestinal flora score indicates a better state of the intestinal flora, while a lower score indicates a worse state of the intestinal flora. A score of D or E encourages lifestyle improvement. The intestinal flora score is calculated by adding up the scores for each index indicated by the bacterial composition data of the intestinal flora (human intestinal bacterial data), which is the test result. The indices indicated by the bacterial composition data include a diversity-related index, a short-chain fatty acid production index, an intestinal immunity-related index, an oral bacteria-related index, and a diarrhea / constipation-related index, and a judgment score is calculated for each index. The test data also determines the type of intestinal flora (intestinal flora type). In this embodiment, the intestinal flora score of the test data is used to predict the state of the intestinal flora. Note that the state of the intestinal flora to be predicted is not limited to the intestinal flora score, and any test data that numerically represents the state of the intestinal flora in the body can be applied. Furthermore, in the intestinal flora test described above, the collected sample must be processed at a testing institution to obtain the above-mentioned indicators. Therefore, the test is relatively expensive due to the costs involved in the testing process, etc.

[0027] The questionnaire with multiple questions consists of multiple questions, such as questions about attribute information and questions about lifestyle habits, and the response results include answers to each question. The questionnaire will be described in detail in the explanation of the prediction device 200.

[0028] The learning unit 112 uses the training data stored in the training data storage unit 102 to train a prediction model for predicting the state of intestinal flora (intestinal flora score), and stores the trained model in the model storage unit 104. Any learning method capable of predicting the state of intestinal flora as a regression problem can be used for the prediction model. For example, a method can be used in which a posterior distribution is generated based on a hierarchical Bayesian model using test data as a prior distribution and answer results as covariates. The learning unit 112 uses any learning method to derive regression coefficients for each category of multiple question items in the answer results to the training data, using the intestinal flora scores of the test data as training data, and trains a prediction model for predicting the state of intestinal flora in the body using the regression coefficients. Table 1 shows an example of regression coefficients for each category of multiple question items obtained from the training data. [Table 1]

[0029] The items in Table 1 correspond to the categories of questions in the questionnaire, including age (age_rank), gender (gender_M), drinking habits (drink_score), exercise habits (exercise_score), eating habits (food_eat_score), sleeping habits (sleep_score), and smoking habits (smoke_score). The regression coefficients for each item are expressed as negative values, and the larger the absolute value, the stronger the influence of the answers on the intestinal flora score. The prediction model trained by the learning unit 112 is configured to output a prediction of the state of the intestinal flora through a simulation based on the regression coefficients. By training the prediction model using the regression coefficients, the correlation between the questionnaire response results and the test data is learned. By simulating the state of the user's intestinal flora using the trained prediction model, a pseudo-synthesized intestinal flora score can be output. Note that the prediction model may not only be trained in advance to predict the state of the intestinal flora as a score as described above, but may also be a model that stores regression coefficients indicating the correlation between the questionnaire response results and the test data so that predictions can be derived.

[0030] [Verification of the prediction model] The prediction performance of the prediction model learned by the learning unit 112 was verified. In the verification, a comparison with actual test data and a residual analysis were performed.

[0031] Figure 3 shows (A) a graph showing the distribution of intestinal flora state from actual test data, and (B) a graph showing the distribution of intestinal flora state predicted using a prediction model from questionnaire response information. (A) and (B) are graphs that tally men on the right and women on the left, respectively, with the horizontal axis of the graph representing flora score and the vertical axis representing the number of samples. The predicted intestinal flora state in (B) is able to output a distribution that is close to the intestinal flora state of the actual test data in (A).

[0032] Figure 4 is a graph showing the results of a residual analysis that verified the predictive performance of the trained predictive model. Graph (A) shows the frequency of the number of samples against the residual between the observed value and the predicted value, with the residuals concentrated around zero. Graph (B) shows the residual against the predicted value. The closer the residual is to zero, the better the predictive performance, and verification has confirmed that significant predictive performance can be obtained in both graphs.

[0033] In this way, the learning device 100 can generate a prediction model for predicting the state of intestinal flora using as input the response results to a questionnaire that includes questions about lifestyle habits.

[0034] (Prediction device) The functional components of the prediction device 200 will be described. FIG. 5 is a block diagram showing the configuration of a prediction system 2 including the prediction device 200 of this embodiment. In the prediction system 2, the prediction device 200 and a terminal 250 are connected via a network N. The terminal 250 is a smartphone, tablet, personal computer, or the like that can be operated by a user, and displays a questionnaire response screen necessary for predicting the state of the user's intestinal flora and accepts responses from the user. The functional components of the prediction device 200 are realized by the CPU 11 reading out a prediction program stored in the ROM 12 or storage 14, expanding it into the RAM 13, and executing it. The prediction device 200 and the terminal 250 may be configured as an integrated unit.

[0035] As shown in FIG. 5, the prediction device 200 functionally includes a model storage unit 202, a questionnaire storage unit 204, an acquisition unit 210, a prediction unit 212, and a presentation unit 214.

[0036] The model storage unit 202 stores a prediction model that has been trained in advance in the learning device 100. The questionnaire storage unit 204 stores the questionnaire response results acquired by the acquisition unit 210. Note that the prediction model may not only be trained in advance as described above, but may also be a model that stores regression coefficients that indicate the correlation between the questionnaire response results and the test data, allowing predictions to be derived.

[0037] The acquisition unit 210 acquires the response results of the questionnaire to be predicted from the terminal 250 and stores them in the questionnaire storage unit 204. The prediction target in this embodiment is a user. The acquisition unit 210 converts the response results into features and stores them in the questionnaire storage unit 204 to use them as input for a prediction model.

[0038] The questionnaire consists of multiple questions, including questions about attribute information and lifestyle habits, and the response results include answers to each question. The attribute information questions are questions about information indicating the user's attributes, such as "gender," "age," and "weight." The attribute information questions may also include questions about family structure, type of work, and physical constitution. The lifestyle habits questions are further divided into categories of questions that are further subdivided according to the nature of the lifestyle. The lifestyle habits category includes a first question about eating habits, a second question about exercise habits, a third question about bowel movements, a fourth question about drinking habits, a fifth question about smoking habits, a sixth question about sleeping habits, and a seventh question about medical history. Furthermore, as shown in the example of the response screen below, multiple more detailed questions are prepared within each category. Of the lifestyle habits questions, questions 4 to 7 may be set arbitrarily. The attribute information questions may also be set arbitrarily.

[0039] FIG. 6 shows an example of a questionnaire response screen. The questionnaire response screen (6A) in FIG. 6 is displayed on, for example, the terminal 250, and is a screen for accepting responses from the user. The response screen (6A) displays multiple options for each question. The response screen (6A) in FIG. 6 shows an example in which questions about attribute information such as "gender," "age group," and "weight" are displayed, and questions about lifestyle habits such as "dietary habits," "exercise habits," and "bowel movements" are displayed. Note that the questions on the response screen (6A) are a screen that displays some of the attribute information questions and the first to seventh questions about lifestyle habits, and may also include other questions. The user can answer the questionnaire by checking the options for each question on the response screen (6A). In this way, the user can answer the questionnaire using a simple selection method.

[0040] For "Gender," participants were asked to select {Male} or {Female}. For "Age Group," participants were asked to select a response from the age range options of {20s}, {30s}, {40s}, etc. For "Weight," participants were asked to select a value from the options indicating kilograms. For "Dietary Habits," participants were asked to select the frequency of intake for each type of food. For "Dietary Habits," participants were asked to select the frequency of intake for each type of food: "Polished Rice / Bread," "Brown Rice / Multigrain Rice / Unrefined Rice," "Light-Colored Vegetables (Lettuce, Cabbage, Cucumber, etc.)," ​​"Green and Yellow Vegetables (Peppers, Pumpkin, Spinach, etc.)," ​​"Fruits," "Natto," and "Mushrooms." For intake frequency, participants were asked to select from five options: {Did Not Eat}, {1-3 Times a Week}, {4-6 Times a Week}, {Once a Day}, and {Twice a Day or More}. For "Exercise Habits," participants were asked to select {Yes} or {No} for the question, "Have you been exercising / playing sports regularly for the past three months or more?" Regarding "bowel movement status," there are two questions: (A) "On average, how often have you had bowel movements over the past three months?" and (B) "What was the most common type of bowel movement over the past three months?" For (A), the respondent can select from the following options: {3 or more times / day}, {1-2 times / day}, {4-6 times / week}, {2-3 times / week}, or {1 time or less / week}. For illustration purposes, the response screen (6A) omits the options for question (B) regarding "bowel movement status." However, the respondent can select from the following options: {Hard, round, rabbit droppings-like stool (difficult to pass)}, {Sausage-shaped but uneven (lumpy) stool}, {Sausage-shaped stool with cracks on the surface}, and so on. Note that the options and number of options listed above are merely examples and can be set appropriately depending on the design of the prediction model.

[0041] In addition, the answer screen (6A) may include questions such as "drinking habits," "smoking habits," "sleeping habits," "skin type / constitution," and "medical history." For the "medical history" question, specific medications may be included as a question item, and whether or not a drug has been used may be an option. The answer screen (6A) may also have a free-form entry field regarding diarrhea and constipation, and the entered content may be analyzed or manually determined.

[0042] As described above, specific questions are set for the questionnaire. For example, in the case of "dietary habits," questions are set about types of food that are particularly related to the state of intestinal flora, and five options for the frequency of intake per week are set for each type of food. By setting specific questions that have a strong correlation with the state of intestinal flora and narrowing down the questions to specific content and obtaining answers from users, the state of intestinal flora can be predicted. These questions have been verified to have a significant correlation by comparing actual test data with the answers to a questionnaire administered at the time of the test corresponding to the test data.

[0043] The acquisition unit 210 converts the questionnaire response results into features based on the answer options listed above and stores the results in the questionnaire storage unit 204. To convert the response results into features, it is sufficient to assign features based on the options. For example, for the question item "Dietary Habits," one of five options for intake frequency is obtained as the response result for each meal type. Therefore, each dimension of the feature is assigned to each meal type based on the options, and the feature is converted into a feature. The five meal type options for "polished rice / bread" are assigned feature dimensions 1-5, the five meal type options for "brown rice / multigrain rice / unrefined rice" are assigned feature dimensions 6-10, and the five meal type options for "light-colored vegetables" are assigned feature dimensions 11-15. Similarly, features are converted into features based on the options for other question items. Note that known dimensionality reduction techniques may be used to compress the dimensions and convert the features into features. The acquisition unit 110 of the learning device 100 also converts the features into features in a similar manner.

[0044] The prediction unit 212 acquires the prediction model from the model storage unit 202 and the response results to the questionnaire to be predicted from the questionnaire storage unit 204. The prediction unit 212 inputs the acquired response results to the questionnaire to be predicted into the prediction model, and predicts the state of the intestinal flora to be predicted. The prediction model takes the response results to the above-mentioned multiple questions as input, and outputs the state of the intestinal flora in the body of the user to be predicted. The state of the intestinal flora is predicted as a score that is a pseudo-synthesized intestinal flora score of the user through a simulation using the prediction model. Note that the state of the intestinal flora is not limited to being predicted as an intestinal flora score, and may also be predicted as a composition such as the occupancy rate or number of bacterial flora in the intestine.

[0045] The presentation unit 214 presents the judgment result of the intestinal flora state A to E based on the predicted intestinal flora score and the improvement proposal according to the judgment result. The presentation unit 214 presents the following to the user by displaying a presentation screen on the terminal 250.

[0046] FIG. 7 is an example of a display screen showing the results of the predicted intestinal flora state and suggestions for improvement. The display screen (7A) in FIG. 7 displays the title "Your intestinal flora score predicted from your questionnaire responses" and the following explanation: "Based on your responses to the lifestyle questionnaire, we have predicted the quality of your intestinal environment on a five-point scale, from A to E. Those who received an A or B rating have a good intestinal environment, those who received a C rating are average, and those who received a D or E rating are likely to have an unhealthy intestinal environment." The display screen (7A) also displays the user's position on the scale of A to E as the predicted intestinal flora state assessment result (7x) (in this example, D). The display screen (7A) also displays suggestions for improving the assessment result (7y). The improvement suggestion may read, "Based on your questionnaire responses, your intestinal flora was predicted and you received a D rating. A D rating indicates a high possibility of an unhealthy intestinal environment. We recommend that you improve your lifestyle habits based on your questionnaire responses below and take an intestinal flora test." The "Your Questionnaire Response Results" section also displays the response results for each question. A score ranging from A to E may also be displayed. Furthermore, based on the responses to lifestyle-related questions, the results of A to E for each category of questions, such as "diet," "exercise," "sleep time," "drinking frequency," and "tobacco," may be displayed as a lifestyle assessment, and improvement suggestions may be presented according to the assessment results for each category of questions. For example, if the assessment result for a category related to dietary habits is poor, improvement suggestions for dietary habits may be presented. In this way, the presentation unit 214 may output a lifestyle assessment calculated from the prediction of the intestinal flora state for each lifestyle-related question. Improvement suggestions for supplements, nutrients, foods, medicines, exercise methods, etc. may also be output.

[0047] (Processing flow) Next, the operations of the learning device 100 and the prediction device 200 will be described.

[0048] 8 is a flowchart showing the flow of the learning process by the learning device 100. The learning process is performed by the CPU 11 reading out a learning program from the ROM 12 or storage 14, expanding it into the RAM 13, and executing it. The CPU 11 functions as each part of the learning device 100 to execute the following processes.

[0049] In step S100, the CPU 11 acquires training data and stores it in the training data storage unit 102. The training data includes test data that quantifies the state of the intestinal flora based on multiple bacterial composition data of the intestinal bacterial flora, and responses to a questionnaire with multiple questions corresponding to the test data. Note that the training data is converted into features and stored in the training data storage unit 102 in order to be used for training a prediction model.

[0050] In step S102, the CPU 11 uses the learning data stored in the learning data storage unit 102 to learn a prediction model for predicting the state of the intestinal flora (intestinal flora score), and stores the model in the model storage unit 104.

[0051] As described above, the learning device 100 of this embodiment can learn a prediction model for predicting the state of intestinal flora using test data and questionnaire responses corresponding to the test data as learning data.

[0052] 9 is a flowchart showing the flow of prediction processing by the prediction device 200. The prediction processing is performed by the CPU 21 reading out a prediction program from the ROM 22 or the storage 24, expanding it in the RAM 23, and executing it. The CPU 11 functions as each part of the prediction device 200 to execute the following processing.

[0053] In step S200, the CPU 21 acquires the response results of the questionnaire to be predicted from the terminal 250 and stores them in the questionnaire storage unit 204. The response results are converted into feature quantities and stored in the questionnaire storage unit 204 to be used as input for the prediction model.

[0054] In step S202, the CPU 21 acquires the prediction model from the model storage unit 202 and the response results of the questionnaire to be predicted from the questionnaire storage unit 204 for prediction.

[0055] In step S204, the CPU 21 inputs the acquired questionnaire response results of the prediction target into the prediction model, and predicts the state of the intestinal flora of the prediction target.

[0056] In step S206, the CPU 21 presents to the terminal 250 the predicted determination results A to E of the state of the intestinal flora and improvement suggestions according to the determination results.

[0057] As described above, the prediction device 200 of this embodiment can predict the state of intestinal flora by simple means other than testing.

[0058] In the above embodiments, the learning process or prediction process executed by the CPU after reading the software (program) may be executed by various processors other than the CPU. Examples of such processors include programmable logic devices (PLDs) (such as field-programmable gate arrays (FPGAs)) whose circuit configuration can be changed after manufacture, graphics processing units (GPUs), and application-specific integrated circuits (ASICs) that are dedicated electrical circuits that are processors with circuit configurations specifically designed to execute specific processes. The learning process or prediction process may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (e.g., multiple FPGAs, multiple GPUs, or a combination of a CPU and an FPGA). The hardware structure of these various processors is, more specifically, an electrical circuit that combines circuit elements such as semiconductor devices.

[0059] In the above embodiment, the learning program or the prediction program is pre-stored (installed) in storage, but the present invention is not limited to this. The program may be provided in a form stored in a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The program may also be downloaded from an external device via a network. [Explanation of symbols]

[0060] 100 Learning Device 102 Learning data storage unit 104 Model Memory Unit 110, 210 Acquisition Department 112 Learning Department 2. Prediction System 200 Prediction Device 202 Model Memory Unit 204 Questionnaire memory section 212 Prediction Department 214 Presentation section 250 devices

Claims

1. On the computer, Acquire answers to a questionnaire with multiple questions including questions about lifestyle habits of the living body to be predicted; The question items include at least a first question item regarding predetermined dietary habits, a second question item regarding exercise habits, and a third question item regarding bowel movements, as question items regarding the lifestyle habits related to the state of intestinal flora, The first question item regarding the dietary habits, in relation to the state of the intestinal flora, distinguishes between polished rice, bread, brown rice, mixed grain rice, and unrefined rice with respect to the type of rice, and distinguishes between light-colored vegetables, green and yellow vegetables, fruits, natto, and mushrooms with respect to the type of vegetables, and the answers obtained are based on the questions. Using the acquired answer results as input, predict the state of the intestinal flora of the living body to be predicted based on a correlation between the answer results and test data indicating the state of the intestinal flora previously associated with the questionnaire. The prediction method that causes the process to run.

2. The first question regarding frequency of intake is answered with options of {did not eat}, {1-3 times a week}, {4-6 times a week}, {once a day}, and {2 or more times a day}, and the third question regarding frequency is answered with options of {3 or more times / day}, {1-2 times / day}, {4-6 times / week}, {2-3 times / week}, and {less than once a week}, The prediction method according to claim 1, wherein the improvement proposal for the state of the intestinal flora is presented as an improvement proposal for the lifestyle habits, the improvement proposal including improvement content for at least one of the type of rice, the type of vegetables, and the frequency of intake.

3. 2. The prediction method according to claim 1, wherein the plurality of questions in the questionnaire include a question about attribute information including at least one of age and gender, and at least one of a fourth question about drinking habits, a fifth question about smoking habits, a sixth question about sleeping habits, and a seventh question about medical history as the question about lifestyle habits.

4. The prediction method according to any one of claims 1 to 3, wherein the prediction outputs an evaluation for each category of the question items, which is obtained from a prediction of the state of the intestinal flora for each of the plurality of question items related to lifestyle habits.

5. The prediction method according to any one of claims 1 to 4, further comprising presenting the predicted state of the intestinal flora and a predetermined improvement proposal for the state of the intestinal flora.

6. The prediction method according to claim 5 , wherein the improvement suggestions further include improvement suggestions determined in advance for each category of the plurality of question items regarding the lifestyle habits.

7. the correlation is configured as a predictive model; The prediction method according to any one of claims 1 to 6, wherein the prediction model is configured to use, as learning data, the test data in which the state of the intestinal flora is quantified based on multiple bacterial composition data of the indigenous bacterial flora in the body, and the answers to the multiple questions corresponding to the test data, to calculate predetermined regression coefficients for categories of the multiple questions corresponding to the test data, and to output a prediction of the state of the intestinal flora by simulation based on the regression coefficients.

8. On the computer, Acquire answers to a questionnaire with multiple questions including questions about lifestyle habits of the living body to be predicted; The question items include at least a first question item regarding predetermined dietary habits, a second question item regarding exercise habits, and a third question item regarding bowel movements, as question items regarding the lifestyle habits related to the state of intestinal flora, The first question item regarding the dietary habits, in relation to the state of the intestinal flora, distinguishes between polished rice, bread, brown rice, mixed grain rice, and unrefined rice with respect to the type of rice, and distinguishes between light-colored vegetables, green and yellow vegetables, fruits, natto, and mushrooms with respect to the type of vegetables, and the answers obtained are based on the questions. Using the acquired answer results as input, predict the state of the intestinal flora of the living body to be predicted based on a correlation between the answer results and test data indicating the state of the intestinal flora previously associated with the questionnaire. A prediction program that executes the process.

9. The method includes: an acquisition unit that acquires answer results to a questionnaire with a plurality of questions including questions about lifestyle habits of a living body that is a prediction target; and a prediction unit, The question items include at least a first question item regarding predetermined dietary habits, a second question item regarding exercise habits, and a third question item regarding bowel movements, as question items regarding the lifestyle habits related to the state of intestinal flora, The first question item regarding the dietary habits, in relation to the state of the intestinal flora, distinguishes between polished rice, bread, brown rice, mixed grain rice, and unrefined rice with respect to the type of rice, and distinguishes between light-colored vegetables, green and yellow vegetables, fruits, natto, and mushrooms with respect to the type of vegetables, and the answers obtained are based on the questions. The prediction unit uses the acquired answer results as input and predicts the state of the intestinal flora of the living body being predicted based on the correlation between the answer results and test data indicating the state of the intestinal flora pre-associated with the questionnaire.

Images