Disease onset-risk reduction support system

Abstract

To provide a disease onset-risk reduction support system excellent in convenience and accuracy.SOLUTION: A disease onset-risk reduction support system according to one form of the present invention is provided with a body fat amount measurement unit, an onset-risk risk evaluation unit, an advice generation unit, an onset-risk risk prediction unit, and a presentation information generation unit. The body fat amount measurement unit acquires a body type image and measures the amount of body fat of a user on the basis of the body type image. The onset-risk risk evaluation unit inputs the amount of body fat measured by the body fat amount measurement unit to a prediction model to evaluate the risk that the user is affected by a disease. The advice generation unit generates an advice for reducing the risk according to a result of evaluation made by the onset-risk risk evaluation unit. The onset-risk risk prediction unit inputs the amount of body fat of the user achieved by following the advice to the prediction model to predict the risk that the user is affected by the disease. The presentation information generation unit generates presentation information including the result of evaluation made by the onset-risk risk evaluation unit, a result of prediction made by the onset-risk risk prediction unit, and the advice.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a disease prevention support system that utilizes body fat mass. [Background technology]

[0002] Body fat mass is a cause of various diseases such as metabolic syndrome and lifestyle-related diseases, and understanding body fat mass is important from the perspective of preventing or treating these diseases. However, measuring body fat mass is not as easy as other biomarkers such as weight or blood pressure. For example, body fat mass measurements using computed tomography (CT) or impedance spectroscopy require the use of medical equipment and therefore require the supervision of a medical professional. While there are methods for estimating body fat mass from abdominal circumference measurements using a tape measure or similar, these methods are prone to error. In response to this issue, various methods for estimating body fat mass have been investigated in recent years. For example, Patent Document 1 discloses a method for estimating body fat mass from measurements of the width and thickness of the torso.

[0003] Various studies have also been conducted on disease prevention and treatment methods that utilize estimated body fat mass, and for example, a technology has been developed that uses various biomarkers of a user to determine which diseases the user is suffering from or may suffer from in the future (see Patent Document 2), and a technology that suggests improvements (see Patent Document 3).Technology has also been developed for diagnosing the state of metabolic syndrome and lifestyle-related diseases using biomarkers such as body fat mass (see Patent Document 4). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2018-198800 A [Patent Document 2] Japanese Patent Publication No. 2023-113955 [Patent Document 3] Japanese Patent Publication No. 2020-60803 [Patent Document 4] Japanese Patent Application Publication No. 2023-143891 Summary of the Invention [Problem to be solved by the invention]

[0005] As described above, measuring body fat mass is not easy, which poses a problem when using body fat mass to prevent and treat diseases. The present inventors have investigated a simple and accurate method for measuring body fat mass and a disease prevention measure using the same, and have found an effective method.

[0006] The present invention relates to a disease morbidity prevention support system that is highly convenient and accurate. [Means for solving the problem]

[0007] A disease morbidity prevention support system according to one embodiment of the present invention is a disease morbidity prevention support system that supports a user in taking measures to prevent disease morbidity, and includes a body fat mass measurement unit, a morbidity risk assessment unit, an advice generation unit, a morbidity risk prediction unit, and a presentation information generation unit. The body fat mass measuring unit acquires a body shape image by capturing an image of the user's body, and measures the body fat mass of the user based on the body shape image. The morbidity risk assessment unit inputs the body fat mass measured by the body fat mass measurement unit into a prediction model that defines the relationship between body fat mass and the incidence rate of a specific disease, and assesses the risk of the user contracting the disease. The advice generation unit generates advice for reducing the risk in accordance with the evaluation result by the morbidity risk evaluation unit. The morbidity risk prediction unit inputs the body fat mass of the user that is predicted to be achieved by following the advice into the prediction model, and predicts the risk of the user contracting the disease. The presentation information generation unit generates presentation information including the evaluation result by the morbidity risk evaluation unit, the prediction result by the morbidity risk prediction unit, and the advice.

[0008] A disease morbidity prevention support program according to one embodiment of the present invention is a disease morbidity prevention support program executed by an information processing device, which supports a user's disease morbidity prevention measures, the program including: acquiring a body shape image by capturing an image of the user's body, and measuring the body fat mass of the user based on the body shape image; a step of inputting the body fat mass measured in the step of measuring the body fat mass into a prediction model that defines a relationship between body fat mass and the incidence of a specific disease, and assessing the risk of the user contracting the disease; generating advice for reducing the risk according to an assessment result of the risk assessment step; A step of inputting the body fat mass of the user that is predicted to be achieved by following the advice into the prediction model to predict the risk of the user suffering from the disease; generating presentation information including an evaluation result from the step of evaluating the risk, a prediction result from the step of predicting the risk, and the advice; Execute the following.

[0009] A disease morbidity prevention support method according to one embodiment of the present invention is a disease morbidity prevention support method for supporting a user's disease morbidity prevention measures, acquiring a body shape image by capturing an image of the user's body, and measuring the body fat mass of the user based on the body shape image; inputting the body fat mass measured in the step of measuring the body fat mass into a prediction model that defines a relationship between body fat mass and the incidence of a specific disease, and assessing the risk of the user developing the disease; generating advice for reducing the risk according to an evaluation result obtained by the step of evaluating the risk; inputting the user's body fat mass that is predicted to be achieved by following the advice into the prediction model to predict the user's risk of developing the disease; Presentation information is generated that includes the evaluation result from the step of evaluating the risk, the prediction result from the step of predicting the risk, and the advice. [Effects of the Invention]

[0010] According to the present invention, it is possible to realize a disease morbidity prevention support system that is highly convenient and accurate. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing the configuration of a disease morbidity prevention support system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram showing the configurations of an eating habit classification generation unit and an advice generation unit included in the disease morbidity prevention support system. [Figure 3] 10 is an example of a screen displaying questions presented to a user by the disease morbidity prevention support system. [Figure 4] 10 is an example of a presentation screen of the eating habit classification results presented to the user by the disease morbidity prevention support system. [Figure 5] 10 shows an example of advice presented to a user by the disease morbidity prevention support system. [Figure 6] 10 is an example of information presented to a user from the disease morbidity prevention support system. [Figure 7] 4 is a flowchart showing the operation of the disease morbidity prevention support system. [Figure 8] 4 is a flowchart showing the operation of the disease morbidity prevention support system. DETAILED DESCRIPTION OF THE INVENTION

[0012] A disease morbidity prevention support system according to an embodiment of the present invention will be described below with reference to the drawings and tables. The present invention is not limited to the following embodiments, and various modifications may be made without departing from the spirit and scope of the present invention. The disease morbidity prevention support system according to an embodiment of the present invention also functions as a metabolic syndrome risk assessment system, a metabolic syndrome morbidity prevention support system, a body fat reduction support system, a visceral fat reduction support system, and a diet support system. In this specification, the term "system" includes one or more information processing devices. For example, a system can be constituted by a single information processing device, or by multiple information processing devices working together to perform functions such as a web server. Furthermore, the term "system" may include an information processing device functioning as a web server, one or more terminal devices, and one or more testing devices.

[0013] [Overall structure of the disease prevention support system] The configuration of the disease incidence countermeasure support system 1 according to this embodiment will be described. The disease incidence countermeasure support system 1 is a system that provides support for countermeasures against diseases related to body fat mass to a specific user. Diseases related to body fat mass include metabolic syndrome, diabetes, dyslipidemia, and hypertension. Hereinafter, one or more diseases for which the disease incidence countermeasure support system 1 supports countermeasures will be referred to as "target diseases," and a user for whom the disease incidence countermeasure support system 1 supports countermeasures will be referred to as "target user."

[0014] As shown in Figure 1, the disease morbidity countermeasure support system 1 according to this embodiment includes a body fat mass measurement unit 11, a prediction model generation unit 12, a morbidity risk assessment unit 13, a dietary habit classification generation unit 14, an advice generation unit 15, a body fat mass prediction unit 16, a morbidity risk prediction unit 17, and a presentation information generation unit 18. The disease morbidity countermeasure support system 1 may be realized by one information processing device or by multiple information processing devices. The configuration of each unit will be described briefly below, and details will be provided later.

[0015] The body fat mass measurement unit 11 acquires a body shape image of the target user's body and measures the target user's body fat mass based on the body shape image. Hereinafter, the body fat mass measured by the body fat mass measurement unit 11 will be referred to as the "current body fat mass." Note that body fat mass includes various amounts indicating the amount of body fat, as well as related alternative expressions such as symbols indicating relative superiority or inferiority. Typically, it is the visceral fat area (VFA). Abdominal information related to the size or shape of the abdomen (typically, abdominal circumference or abdominal thickness) may be used instead of the visceral fat area. Visceral fat area is preferred because it can better utilize the effects of the present invention in terms of measurement difficulty. The body fat mass measurement unit 11 supplies the current body fat mass to the morbidity risk assessment unit 13 and the advice generation unit 15.

[0016] The prediction model generation unit 12 generates a "prediction model," which is a machine learning model that defines the relationship between body fat mass and the incidence rate of a target disease. Typically, the prediction model defines the incidence rate of a target disease over a predetermined prediction period, which is, for example, three years. The prediction model generation unit supplies the generated prediction model to the morbidity risk assessment unit 13 and the morbidity risk prediction unit 17.

[0017] The morbidity risk assessment unit 13 inputs the current body fat mass supplied from the body fat mass measurement unit 11 into the prediction model supplied from the prediction model generation unit 12, and assesses the risk of the target user contracting a target disease (hereinafter referred to as "morbidity risk"). Typically, the morbidity risk is assessed in the form of "Whether or not the target disease will be contracted within three years: Yes / No" or "Probability of contracting the target disease within three years: 0%". The morbidity risk assessment unit 13 supplies the assessment result of the morbidity risk to the advice generation unit 15 and the presentation information generation unit 18.

[0018] The eating habit classification generation unit 14 generates "eating habit classifications" that classify the eating habits of a large number of analysis subjects. The eating habit classifications are typically classified into categories such as "low health consciousness," "snacking," "late-night snacking," "fast eating," "lack of exercise," "late-night eating," "low basal metabolic rate or metabolism," and "other (does not fall into any category)." The eating habit classification generation unit 14 supplies the generated eating habit classifications to the advice generation unit 15.

[0019] The advice generating unit 15 generates advice for reducing the morbidity risk according to the evaluation result by the morbidity risk evaluating unit 13. Typically, the advice generating unit 15 sets a target value for the target user's body fat mass (hereinafter referred to as "target body fat mass") required to reduce the morbidity risk. The target body fat mass may be predetermined according to the gender, age, etc. of the target user, or may be specified by the target user, a doctor, etc.

[0020] The advice generation unit 15 determines at least one of exercise content (exercise intensity, exercise time, etc.) and dietary content (calorie intake, nutrients, etc.) so that the target user's current body fat mass will be equal to or less than the target body fat mass within a predetermined period (hereinafter referred to as the "advice implementation period"), and generates advice proposing these. The advice generation unit 15 can further classify the target user's eating habits using the eating habit classification supplied from the eating habit classification generation unit 14, and generate advice according to the classification result. The advice generation unit 15 supplies the generated advice to the presentation information generation unit 18.

[0021] The body fat mass prediction unit 16 predicts the target user's body fat mass that will be achieved by the target user following the advice generated by the advice generation unit 15. Hereinafter, the body fat mass of the target user predicted by the body fat mass prediction unit 16 will be referred to as the "future body fat mass." Typically, the body fat mass prediction unit 16 calculates the amount of body fat reduction that the target user will achieve at the end of the advice implementation period, assuming that the target user performs the exercises included in the advice or eats the diet included in the advice, and subtracts this from the current body fat mass to predict the future body fat mass. The body fat mass prediction unit 16 may predict the future body fat mass only before the start of the advice implementation period, or may predict the future body fat mass before the start of the advice implementation period and at regular intervals (e.g., every week) during the advice implementation period. The body fat mass prediction unit 16 supplies the future body fat mass to the morbidity risk prediction unit 17.

[0022] The morbidity risk prediction unit 17 inputs the future body fat mass supplied from the body fat mass prediction unit 16 into the prediction model supplied from the prediction model generation unit 12, and predicts the morbidity risk at the end of the advice implementation period. The prediction model is the same as that described above. The morbidity risk prediction unit 17 may predict the morbidity risk only before the start of the advice implementation period, or may predict the morbidity risk at the timing when the future body fat mass is supplied from the body fat mass prediction unit 16 before the start and during the advice implementation period. The morbidity risk prediction unit 17 supplies the morbidity risk prediction result to the presentation information generation unit 18.

[0023] The presentation information generation unit 18 generates presentation information to be presented to the target user. The presentation information includes the morbidity risk assessment result by the morbidity risk assessment unit 13, the morbidity risk prediction result by the morbidity risk prediction unit 17, and the advice generated by the advice generation unit 15. The presentation information may further include the current body fat mass measured by the body fat mass measurement unit 11 and the future body fat mass predicted by the body fat mass prediction unit 16. The presentation information generation unit 18 generates images, videos, and audio including the presentation information and presents them to the target user.

[0024] In this way, the disease morbidity prevention support system 1 presents the target user with the evaluation result of the morbidity risk of the target disease, advice for reducing the morbidity risk, and the predicted morbidity risk after the advice is implemented.

[0025] Although the morbidity risk assessment unit 13 has been described as inputting the current body fat mass, which is the body fat mass supplied from the body fat mass measurement unit 11, into the prediction model to assess the morbidity risk, this is not limited to this. The target user's body fat mass may be measured by other means, and for example, the morbidity risk assessment unit 13 may obtain the target user's body fat mass by direct input by the target user, and input the body fat mass into the prediction model to assess the morbidity risk.

[0026] Furthermore, the prediction model used by the morbidity risk assessment unit 13 and the morbidity risk prediction unit 17 is not limited to one that defines the body fat mass and the target disease morbidity rate. This prediction model may define the relationship between the target disease morbidity rate and at least one of body fat mass, BMI (Body Mass Index), number of cigarettes smoked, diastolic blood pressure, systolic blood pressure, sex, and age.

[0027] The disease incidence prevention support system 1 is a functional configuration realized by the cooperation of hardware and software of one or more information processing devices each consisting of a CPU (Central Processing Unit), RAM (Random Access Memory), etc. The software is a program executable by the information processing device, and may be a program recorded on a recording medium readable by the information processing device.

[0028] [Body fat mass measurement by the body fat mass measurement department] As described above, the body fat mass measuring unit 11 can acquire a body type image by capturing an image of the target user's body, and measure the body fat mass of the target user based on the body type image.

[0029] A typical method for measuring body fat mass using the body fat mass measurement unit 11 is the method described in Japanese Patent Application Laid-Open No. 2018-198800. Specifically, this method involves obtaining a first captured image of the target user's body from the front, and measuring a first length, which is the width of the front of the target user's torso, in this first captured image. Also, obtaining a second captured image of the target user's body from the side, and measuring a second length, which is the width of the front of the target user's torso, in this second captured image.

[0030] Furthermore, at least one of the measured first length and second length of the target user is compared with "relationship information." The relationship information is information prepared in advance that indicates the relationship between the body fat mass and at least one of the first length, second length, and parameters based on the first length or second length. Therefore, by comparing at least one of the first length and second length of the target user with the relationship information, the body fat mass of the target user's torso can be measured.

[0031] As a specific method for measuring body fat mass by the body fat mass measurement unit 11, the method described in International Publication No. 2020 / 206049 can also be used. Specifically, this method acquires a body shape image of the target user's body, and uses an annotation deep learning module to identify annotation key points, which are body components under the target user's clothing, from the body shape image. Furthermore, geometric features of the target user are calculated based on the annotation key points, and a weight machine learning module is used based on the geometric features and parameters of the target user to measure the target user's body fat mass.

[0032] In addition to the above methods, the body fat mass measurement unit 11 can measure the body fat mass of the target user using any method that can measure the body fat mass of the target user based on a body shape image taken of the target user's body.

[0033] [Prediction model generation by the prediction model generation unit] The prediction model generation unit 12 generates a prediction model, which is a machine learning model that defines the relationship between body fat mass and disease incidence as described above. The prediction model generation unit 12 can create a prediction model based on health checkup data covering multiple years for a large number of subjects. The multiple years may be, for example, six years. The disease incidence countermeasure support system 1 according to this embodiment further includes a prediction model generation unit 12 that supplies this prediction model to the incidence risk assessment unit 13 and the incidence risk prediction unit 17. Note that the disease incidence rate not only indicates the probability of disease incidence, but also includes related paraphrases such as symbols indicating relative superiority or inferiority.

[0034] Typically, the prediction model generation unit 12 selects subjects who do not suffer from the target disease. Table 1 below shows diagnostic criteria for metabolic syndrome as an example. Furthermore, the prediction model generation unit 12 assigns a label to subjects who do not suffer from the target disease indicating whether or not they will develop the target disease within a specified prediction period. The prediction model generation unit 12 randomly divides the health checkup data of subjects to which this label has been assigned into validation data and test data. For example, the validation data may be 80% and the test data may be 20%.

[0035] Next, the prediction model generation unit 12 constructs a prediction model using the label of the target disease within the prediction period as an output variable and biomarkers related to the target disease as input variables. The biomarkers related to the target disease include body fat mass, such as VFA (Visceral Fat Area). In addition to body fat mass, biomarkers related to the target disease may include at least one of gender, age, BMI (Body Mass Index), smoking information, alcohol intake information, systolic blood pressure, and diastolic blood pressure. Preferred are: VFA alone; VFA, systolic blood pressure, BMI, diastolic blood pressure, gender, age, and smoking information; and VFA, systolic blood pressure, BMI, diastolic blood pressure, gender, age, smoking information, and alcohol intake information. From the perspective of improving user convenience and information processing speed by reducing the input variables of the biomarker, using VFA alone is more preferable. From the perspective of improving the accuracy of metabolic syndrome onset prediction, using VFA, systolic blood pressure, BMI, diastolic blood pressure, gender, age, and smoking information is more preferable. Note that from the perspective of improving the accuracy of metabolic syndrome onset prediction, it is preferable not to use alcohol intake information as a biomarker. The prediction model generation unit 12 can be constructed using five-fold cross-validation, and the model with the highest AUC (Area Under the Curve) when five cross-validations are performed can be used as the prediction model.

[0036] The prediction model is not limited to the one described above, and any machine learning model that defines the relationship between body fat mass and the incidence rate of a target disease may be used.

[0037] [Advice generation by the advice generation unit] As described above, the advice generating unit 15 generates advice for reducing the morbidity risk in accordance with the evaluation result by the morbidity risk evaluating unit 13.

[0038] Specifically, the advice generating unit 15 provides the target user with advice for reducing body fat mass or preventing lifestyle-related diseases. "Reducing body fat mass" includes reducing visceral fat, reducing abdominal circumference, reducing abdominal thickness, reducing weight, reducing body fat percentage, and eliminating metabolic syndrome. "Preventing lifestyle-related diseases" includes improving high blood pressure, improving hyperglycemia, improving dyslipidemia, and eliminating obesity.

[0039] The advice generating unit 15 generates advice using bioindicators for advice generation. The bioindicators for advice generation are indices that represent the state of a living body, such as abdominal circumference, VFA (Visceral Fat Area), FBS (Fasting blood sugar), TG (Triglyceride), HDL-C (High Density Lipoprotein Cholesterol), SBP (Systolic Blood Pressure), and DBP (Diastolic Blood Pressure).

[0040] The advice generation unit 15 uses a specific biometric indicator depending on the symptoms of the person to whom advice is to be provided. Hereinafter, the biometric indicator used by the advice generation unit 15 will be referred to as a "specific biometric indicator." Table 1 below shows examples of symptoms and specific biometric indicators. Note that the symptoms and specific biometric indicators are not limited to those shown here. For example, the specific biometric indicator for body fat mass may be waist circumference or weight.

[0041] [Table 1]

[0042] The advice generation unit 15 obtains the "eating habit classification" from the eating habit classification generation unit 14 as shown in Fig. 1 and uses it to generate advice. The eating habit classification generation unit 14 analyzes multiple people whose specific biomarkers are equal to or greater than the reference values ​​shown in Table 1 as analysis subjects, and generates an eating habit classification. There is no particular limit to the number of analysis subjects, but 100 or more people is preferable.

[0043] 2, the eating habit classification generation unit 14 includes a data acquisition unit 21, an eating habit analysis unit 22, and a ranking determination unit 23. The data acquisition unit 21 acquires "eating habit information," which is information about the eating habits (including dietary habits, eating behavior, lifestyle behavior, exercise habits, and personality) of each analysis subject, and data on specific biomarkers for multiple analysis subjects.

[0044] The dietary information consisted of responses to 35 questions about dietary habits, including but not limited to the following: Q01 Even if I'm not hungry, I end up eating when something smells delicious. Q02 If someone around me is eating something, I eat it too Q03 I overeat when I'm feeling depressed Q04 I consciously limit my food intake to avoid gaining weight Q05 I feel hungry all day Q06 I feel guilty about eating too much, so I eat less. Q07 I try to buy low-calorie foods Q08 I think I eat faster than other people Q09 I actively eat green and yellow vegetables Q10 I like hamburgers, donuts, and potato chips Q11 I actively choose foods that are high in dietary fiber. Q12 I prefer fish to meat Q13 I often eat a late-night snack after dinner Q14 I try to limit my intake of animal fats and instead eat vegetable fats and fish fats. Q15 I think I am more prone to gaining weight than other people. Q16 I feel like I won't feel energized if I don't eat. Q17 Meal times are random Q18 I have a sweet tooth Q19 Eating a late-night snack Q20 When I see fruit or sweets, I can't help but reach for them. Q21 When I receive food, I eat it because I feel it would be a waste Q22 I feel uneasy unless I buy more food than I need. Q23 I often have late dinners because of work. Q24 If there is an elevator or escalator, I will use it. Q25 I finish dinner more than two hours before going to bed Q26 I don't like walking or cycling. Q27 I don't exercise much Q28 I hardly ever chew Q29 I think it's more that I'm not exercising enough than that I'm eating too much. Q30 I'm a night owl and have trouble in the mornings Q31 I'm lazy Q32 Insightful Q33 Cooperative Q34 Proactive Q35 I get nervous easily

[0045] The subject answers each question by selecting one of the following: "Does not apply = 1," "Does not really apply = 2," "Can't say either way = 3," "Somewhat applies = 4," or "Applies = 5," and eating habit information is generated. Note that the above questions are just an example, and other questions may be used. The number of questions is not limited to 35. The data acquisition unit 21 supplies the acquired eating habit information and biometric data to the eating habit analysis unit 22.

[0046] The eating habit analysis unit 22 performs factor analysis on the data supplied from the data acquisition unit 21, and generates multiple eating habit classifications according to factors for the biomarkers of the eating habit information. The eating habit classifications are classifications of the eating habits of the person being analyzed, and are typically classifications such as "low health consciousness," "snacking," "late-night snacking," "fast eating," "lack of exercise," "late-night eating," "low basal metabolic rate or metabolism," and "other (does not fall into any category)."

[0047] "Low health consciousness" is a category of eating habits in which people are indifferent to diet and exercise and eat without any consideration for health, and has the following characteristics, for example: [Diet] Not buying low-calorie foods Not actively eating green and yellow vegetables - Not actively choosing foods high in dietary fiber - Not limiting animal fats and consuming vegetable fats and fish fats I like hamburgers, donuts, and potato chips. I don't like fish more than meat - Not consciously limiting the amount of food you eat to avoid gaining weight Feeling guilty about overeating and not reducing the amount you eat Meal times are random Eat a late-night snack [motion] Use elevators and escalators if available Low step count -Doesn't exercise much I think it's more a lack of exercise than eating too much. [Personality] ·Being lazy Lack of insight Not proactive

[0048] "Snacking" is a category of eating habits in which people have a sweet tooth and can't help but eat sweets, and has the following characteristics: [Diet] I like hamburgers, donuts, and potato chips. - Has a sweet tooth When fruits and sweets are available, I can't help but reach for them. If someone around me is eating something, I eat it too Even if I'm not hungry, I can't help but eat something that smells delicious. Overeating when feeling depressed · If I receive food, I eat it because I don't want to waste it. [Meal Contents] High intake of fat, vegetable fat, saturated fatty acids, monounsaturated fatty acids, and sucrose [motion] Low muscle mass [Personality] I think I'm more prone to gaining weight than other people I feel like I won't have energy if I don't eat I feel uncomfortable unless I buy more groceries than I need. Short -Low basal metabolic rate

[0049] "Late-night snacking" is a classification of eating habits in which a person is a night owl and eats a late-night snack after dinner, and has the following characteristics, for example: [Diet] Eat a late-night snack after dinner Eat a late-night snack Dinner is often late due to work Meal times are random If someone around me is eating something, I eat it too Even if I'm not hungry, I can't help but eat something that smells delicious. - Not consciously limiting the amount of food you eat to avoid gaining weight Overeating when feeling depressed Feeling hungry all day - Has a sweet tooth When fruits and sweets are available, I can't help but reach for them. · If I receive food, I eat it because I don't want to waste it. [Meal Contents] High energy, carbohydrate, and sucrose intake Low intake of animal protein, vitamin D, niacin, vitamin B6, vitamin B12, and alcohol [motion] -Doesn't exercise much [Personality] I think I'm more prone to gaining weight than other people I feel uncomfortable unless I buy more groceries than I need. I'm a night owl and have a weak morning temperament. · Lazy Easily nervous Younger ·Emotionally unstable

[0050] "Speed ​​eaters" is a classification of eating habits of people with a large build who eat food without chewing it much, and has the following characteristics: [Diet] I think I eat faster than other people Hardly chews · If I receive food, I eat it because I don't want to waste it. - Not having dinner more than two hours before going to bed [Personality] Heavy weight High muscle mass High basal metabolic rate

[0051] "Lack of exercise" is a classification of eating habits such as disliking exercise and not eating enough meat or vegetables, and has the following characteristics: [Diet] Not actively eating green and yellow vegetables - Not actively choosing foods high in dietary fiber I don't like fish more than meat - Not limiting animal fats and consuming vegetable fats and fish fats Feeling guilty about overeating and not reducing the amount you eat [Meal Contents] Low intake of beta-carotene [motion] I don't like walking or cycling Low step count -Doesn't exercise much I think it's more a lack of exercise than eating too much. [Personality] · Lazy High body fat percentage

[0052] "Late-night eating" is a category of eating habits such as eating dinner while drinking alcohol late at night, and has the following characteristics, for example: [Diet] Dinner is often late due to work - Not having dinner more than two hours before going to bed Meal times are random - Has a sweet tooth [Meal Contents] Low intake of lipids, K, Ca, Fe, Zn, Cu, vitamin B1, vitamin C, saturated fatty acids, soluble dietary fiber, insoluble dietary fiber, total dietary fiber, and sucrose High alcohol intake

[0053] "Decreased basal metabolic rate or metabolism" is a classification of eating habits that result in weight gain even if you eat the same amount of food as before, and has the following characteristics, for example: [Meal content, meal amount, meal time, exercise] No problem [Personality] Older

[0054] The dietary habit analysis unit 22 performs factor analysis on the dietary habit information and biometric data of multiple subjects. While not limited to specific methods, a preferred method is data obtained by statistically investigating the dietary habit information and biometric data. For example, various models can be employed, including multiple regression analysis using statistical data, logistic regression models, multilayer perceptrons, neural networks such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), support vector machines using arbitrary kernel functions such as Gaussian kernels, random forests modeled as regression trees, models using hidden Markov models, statistical models, probabilistic models, factor analysis, and correspondence tables. A model that combines various models to perform a comprehensive assessment can also be employed. Among these, factor analysis using a principal factor method with varimax rotation is particularly preferred.

[0055] Tables 2 and 3 below show the "factor loadings" of each question item, "eigenvalues," "contribution rates," and "cumulative contribution rates" for each factor calculated using the "principal factor method with varimax rotation" for visceral fat area (VFA), a biomarker of visceral fat accumulation.

[0056] [Table 2]

[0057] [Table 3]

[0058] The eating habit analysis unit 22 sorts each question item into factors according to the factor loading, as shown in Tables 2 and 3. For example, Q04, Q07, Q06, Q09, and Q11 are sorted into "Factor 1" because their factor loadings (inside the black squares) for "Factor 1" exceed the reference value. The reference value is, for example, 0.4. Similarly, Q02, Q01, and Q20 are sorted into "Factor 2," Q19 and Q13 into "Factor 3," and Q08 and Q28 into "Factor 4." Furthermore, Q27 and Q29 are sorted into "Factor 5," and Q25 and Q23 into "Factor 6." The eating habit analysis unit 22 generates an eating habit classification, such as "low health consciousness," for each factor according to the content of the question items sorted into each factor.

[0059] Similarly, for biomarkers other than VFA, the dietary habit analysis unit 22 sorts each question item by factor according to the factor loading. Table 4 below shows the dietary habit classification by factor for each symptom. As shown in Table 4, the dietary habit classification by factor for each symptom and the corresponding question items differ. The dietary habit analysis unit 22 supplies the generated dietary habit classification and "contribution rate" to the ranking determination unit 23.

[0060] [Table 4]

[0061] The ranking determination unit 23 determines the priority order among the eating habit categories in descending order of contribution rate. In the examples of Tables 2 and 3, the contribution rate (0.12) of "low health consciousness" is the highest, followed by the contribution rate (0.11) of "snacking." The contribution rates decrease in the following order: "late-night snack," "quick eating," "lack of exercise," and "late-night eating." For this reason, the ranking determination unit 23 gives the highest priority to "low health consciousness," and then prioritizes "snacking," "late-night snack," "quick eating," "lack of exercise," and "late-night eating" in that order.

[0062] Table 4 shows the priority order of eating habit classifications based on the contribution rate for each symptom. As shown in Table 4, the ranking unit 23 assigns the highest priority to "late-night snacking" for "dyslipidemia," followed by "overeating," "lack of exercise," and "low health consciousness" in that order. For "high blood pressure," the ranking unit 23 assigns the highest priority to "late-night snacking," followed by "snacking," "lack of exercise," "low health consciousness," and "fast eating" in that order. For "hyperglycemia," the ranking unit 23 assigns the highest priority to "snacking," followed by "late-night snacking," "lack of exercise," "irregular meal times," and "food portion control" in that order. The ranking unit 23 provides the eating habit classifications and priorities to the advice generation unit 15. Note that the values ​​obtained for the eating habit classification names, contribution rates, and priorities vary depending on the data used. The results disclosed herein are merely examples. The priorities may be changed depending on the improvement goals desired by the user.

[0063] The advice generation unit 15 generates advice for reducing body fat or preventing lifestyle-related diseases, i.e., advice for reducing the risk of morbidity, in accordance with the evaluation results by the morbidity risk evaluation unit 13. The advice generation unit 15 includes an eating habit information acquisition unit 31, an eating habit classification unit 32, a record information acquisition unit 33, an evaluation unit 34, and an advice creation unit 35, as shown in FIG.

[0064] The eating habit information acquisition unit 31 acquires the eating habit information of the target user. The eating habit information is the answers to the 35 questions about eating habits described above. The eating habit information acquisition unit 31 can generate a screen for presenting questions as shown in FIG. 3, and by displaying one question per screen, it is possible to make it easier for the user to answer. The eating habit information acquisition unit 31 may also acquire personal data that affect the target user's eating habits, such as the target user's gender, age, place of residence, place of origin, occupation, income, etc.

[0065] The eating habit classification unit 32 classifies the eating habits of the target user based on the eating habit information into one of the eating habit categories generated by the eating habit analysis unit 22. As shown in FIG. 2 , the eating habit classification unit 32 includes an eating habit category determination unit 36 ​​and an eating habit category determination unit 37.

[0066] The eating habit classification determination unit 36 ​​determines whether the eating habits of each target user correspond to each eating habit classification based on the eating habit information of each target user. The eating habit classification determination unit 36 ​​can determine whether the target user's eating habits correspond to each eating habit classification based on the answer scores to the questions selected for each eating habit classification. The rotation scores are the above-mentioned scores of "does not apply = 1," "does not really apply = 2," "cannot say either way = 3," "somewhat applies = 4," and "applies = 5."

[0067] For example, with regard to visceral fat accumulation, if the answer scores for Q04, Q07, Q06, Q09, and Q11, which are questions selected as "low health consciousness" (see Table 4), are all "not applicable" (1) or "somewhat not applicable" (2), the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall under "low health consciousness." Furthermore, if the answer scores for Q02, Q01, and Q20, which are questions selected as "snacking," are all "applicable" (5) or "somewhat applicable" (4), the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall under "snacking." Similarly, the eating habit classification determination unit 36 ​​determines whether the target user's eating habits fall under "late-night snacking," "quick eating," and "lack of exercise" based on the answer scores for the questions selected as "snacking." Furthermore, if the answer score for Q25, a question item selected as "eating late at night," is "not applicable = 1" or "somewhat not applicable = 2," and the answer score for Q23 is "applicable = 5" or "somewhat applicable = 4," the eating habit classification determination unit 36 ​​determines that the target user's eating habit falls under "eating late at night."

[0068] Here, if the answer scores meet the conditions in multiple eating habit categories, the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall into multiple eating habit categories. Also, if the answer scores do not meet the conditions in any of the eating habit categories, the eating habit classification determination unit 36 ​​determines that the target user's eating habits do not fall into any eating habit category.

[0069] As another method, the eating habit classification determination unit 36 ​​determines that the target user's eating habits correspond to "low health consciousness" if the total of the answer scores for Q04, Q07, Q06, Q09, and Q11, which are question items selected as "low health consciousness" (see Table 4), is equal to or greater than a threshold. Furthermore, the eating habit classification determination unit 36 ​​determines that the target user's eating habits correspond to "snacking" if the total of the answer scores for Q02, Q01, and Q20, which are question items selected as "snacking," is equal to or greater than a threshold. Similarly, the eating habit classification determination unit 36 ​​determines whether the target user's eating habits correspond to "late-night snacking," "fast eating," "lack of exercise," and "late-night eating" based on the answer scores for the question items classified as each of these eating habits.

[0070] In this case, the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall into multiple eating habit classifications if the answer scores in multiple eating habit classifications are equal to or greater than the thresholds. On the other hand, the eating habit classification determination unit 36 ​​determines that the target user's eating habits do not fall into any eating habit classification if the answer scores in any eating habit classifications are less than the thresholds.

[0071] The eating habit classification determination unit 37 determines the eating habit classification of the target user in response to the determination result by the eating habit classification determination unit 36. When the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall into one eating habit classification, the eating habit classification determination unit 37 determines that the target user's eating habits fall into that eating habit classification.

[0072] Furthermore, when the eating habit classification determination unit 36 ​​determines that the target user's eating habits fall into multiple eating habit classifications, the eating habit classification determination unit 37 determines that the target user's eating habits should be classified into the eating habit classification with the highest priority (see Table 4) determined by the ranking determination unit 23. For example, when the eating habit classification determination unit 36 ​​determines that the eating habits of a specific target user with regard to visceral fat accumulation fall into "snacking" and "lack of exercise," the eating habit classification determination unit 37 determines that the target user's eating habits should be classified into the eating habit classification of "snacking," which has the higher priority.

[0073] Furthermore, if the eating habit classification determination unit 36 ​​determines that the target user's eating habits do not fall into any of the eating habit classifications, the eating habit classification determination unit 37 can classify the target user's eating habits into a classification other than the eating habit classifications generated for each factor. For example, if the target user's eating habits are determined to fall into any of the eating habit classifications for visceral fat accumulation, the eating habit classification determination unit 37 classifies the target user's eating habits into "decreased basal metabolic rate or metabolism." This is because, even though there is nothing wrong with the target user's eating habits, the amount of visceral fat accumulation exceeds a specified value, and therefore it can be determined that this is due to a decreased basal metabolic rate.

[0074] The eating habit classification unit 32 classifies the user's eating habits into one of multiple eating habit classifications based on body fat accumulation factors or lifestyle-related disease susceptibility factors, and classifies the target user's eating habits into one of these classifications in the above manner. The eating habit classification unit 32 can generate an eating habit classification presentation screen such as that shown in FIG. 4, and visually notify the target user of the classification results. The radar chart shows the answer scores for each eating habit classification. The eating habit classification unit 32 may also accept a selection of an eating habit classification from the target user, and the eating habit classification selected by the target user may be used as the user's eating habit classification. This is effective when the target user is aware of their own eating habit classification that was previously notified to them. The eating habit classification unit 32 supplies the target user's eating habit classification to the record information acquisition unit 33.

[0075] The record information acquisition unit 33 acquires record information, which is information about the meal contents and activity contents recorded by the target user. For each target user, the record information acquisition unit 33 generates a record item group including record items according to the target user's eating habit classification. Tables 5 and 6 below are examples of record item groups generated by the record information acquisition unit.

[0076] [Table 5]

[0077] [Table 6]

[0078] As an option, the items recorded may also include the intake of functional ingredients such as chlorogenic acid and ALA-DAG (alpha-linolenic acid diacylglycerol), which contribute to suppressing the accumulation of body fat and preventing lifestyle-related diseases.

[0079] As shown in Tables 5 and 6, the record item group generated by the record information acquisition unit includes "eating habit category items." "Eating habit category items" are record items for each eating habit category classified by the eating habit classification unit 32, and Tables 5 and 6 show examples of record items for "low health consciousness." Table 7 below shows record items for other eating habit categories. As shown in these tables, the record items for each eating habit category have different content for each eating habit category. Furthermore, the record items for each eating habit category may also include calories burned other than from walking.

[0080] [Table 7]

[0081] Furthermore, the group of recording items may include "nutrient intake," as shown in Table 5. "Nutrient intake" is a recording item for the intake of one or more nutrients selected from protein, omega-3 fatty acids, and dietary fiber. Furthermore, the group of recording items may include recording items such as "lifestyle points," "daily calorie intake," "number of steps," and "body measurements," as shown in Table 6. Of the group of recording items, items other than "items by dietary habit classification" are common to each target user. Hereinafter, "items by dietary habit classification," "protein," "dietary fiber," "omega-3 fatty acids," and "lifestyle points" will each be referred to as a "major item."

[0082] The record information acquisition unit 33 presents the generated group of record items to the target user. The target user inputs the record content, and the record information acquisition unit 33 acquires the record information. The target user inputs the record content on a daily or weekly basis. The record information acquisition unit 33 supplies the acquired record information to the evaluation unit 34.

[0083] The evaluation unit 34 evaluates the cause of body fat accumulation or the cause of lifestyle-related disease for each target user based on the recorded information. Tables 8 and 9 below show the evaluation method used by the evaluation unit 34.

[0084] [Table 8]

[0085] [Table 9]

[0086] The evaluation unit 34 can compile the recorded information for each target user for a certain evaluation period, for example, one week, and score each recorded item by calculating the average number of affirmative responses for that recorded item. For example, for the recorded item "I ate meals with consideration for nutritional balance" under "low health consciousness," the evaluation unit 34 calculates the score for that recorded item by calculating the average number of "o"s over one week. Similarly, for other recorded items such as "I did not eat until I was full at any meal," the score for that recorded item is calculated by calculating the average number of "o"s over one week.

[0087] The evaluation unit 34 can extract the record items with the highest scores (hereinafter referred to as the record items) and the record items with the lowest scores (hereinafter referred to as the lower record items) from among the record items. The higher record items are, for example, the record items with the highest scores from 1st to 3rd, and the lower record items are, for example, the record items with the lowest scores from 1st to 3rd. The evaluation unit 34 can evaluate the lower record items as items that require improvement in order to eliminate the causes of body fat accumulation or the causes of lifestyle-related disease.

[0088] Furthermore, the evaluation unit 34 can evaluate the intake amount of nutrients for each target user based on the recorded information. The evaluation unit 34 can estimate the intake ratio and intake status of each nutrient from the scores for each recorded item of "protein," "omega-3 fatty acids," and "dietary fiber." The evaluation unit 34 can also estimate the "protein / fat ratio," "omega-3 fatty acids / fat ratio," and "dietary fiber / sugar or carbohydrate ratio" from the scores for each recorded item of "protein," "omega-3 fatty acids," and "dietary fiber."

[0089] In addition, the evaluation unit 34 can evaluate the "dietary behavior" or "personal information" for each target user based on the recorded information. "Dietary behavior" refers to one or more of the target user's meal quality, quantity, time, eating style, and activity level, and "personal information" refers to at least one or more of weight, waist circumference, etc. The priority of the evaluation by the evaluation unit 34 is not limited, but as an example, it is preferable to give the highest priority to "meal amount," followed by "activity level," "meal time," "meal quality," "eating style," and "weight," in that order. The evaluation unit 34 supplies the evaluation results to the advice creation unit 35.

[0090] The advice creation unit 35 creates advice for reducing body fat or preventing lifestyle-related diseases, i.e., advice for reducing the risk of developing a lifestyle-related disease. Specifically, the advice creation unit 35 calculates the difference between the target body fat mass of the target user required to reduce the risk of developing a lifestyle-related disease and the current body fat mass supplied from the body fat mass measurement unit 11. The advice creation unit 35 determines at least one of exercise content (exercise intensity, exercise time, etc.) and diet content (calorie intake, nutrients, etc.) so that the target user's body fat mass will be equal to or less than the target body fat mass during the advice implementation period, and creates advice suggesting these.

[0091] Furthermore, the advice creation unit 35 can create advice based on the evaluation results of the evaluation unit 34 on the causes of body fat accumulation or lifestyle-related disease, i.e., the scores of each record item. Specifically, the advice creation unit 35 can create advice that praises the higher-ranking record items and suggests improvements for the lower-ranking record items. The advice creation unit 35 can also select record items with lower scores that are likely to be improved and create advice that suggests improvements for the record items with higher scores. The advice creation unit 35 can select record items with higher scores that are likely to be improved from the lower-ranking record items by excluding items with low scores in each record. Furthermore, the advice creation unit 35 can create advice that encourages the target user's eating habits from the next time onwards.

[0092] For example, as shown in FIG. 5, the advice creation unit 35 can create advice including advice S1 praising higher-level recorded items during the evaluation period, advice S2 suggesting improvements to lower-level recorded items, and advice S3 encouraging the next evaluation period.

[0093] The advice creation unit 35 can also create advice using standard phrases prepared for each record item. Tables 10 and 11 below are examples of standard phrases for each record item. The advice creation unit 35 can create advice that includes at least one of the standard phrases for higher-level record items and the standard phrases for lower-level record items. The standard phrases may include specific menu suggestions, and it is preferable that the menus are easy to eat. The standard phrases for each record item may have different content depending on the season, and for example, they may suggest seasonal ingredients and cooking methods that are easy to find and purchase depending on the season. The advice creation unit 35 can create advice according to the season at the time of creation by using standard phrases prepared for each season.

[0094] [Table 10]

[0095] [Table 11]

[0096] The advice creating unit 35 can also create advice according to the evaluation results regarding the target user's nutrient intake. The advice creating unit 35 can create advice that praises the intake of nutrients with a high intake amount and encourages the intake of nutrients with a low intake amount regarding the intake amounts of "protein," "omega-3 fatty acids," and "dietary fiber" estimated by the evaluation unit 34. Furthermore, the advice creating unit 35 can create advice that praises points with a good evaluation and encourages points with a poor evaluation according to the evaluation results of the target user's "dietary behavior" or "personal information."

[0097] Furthermore, the advice creating unit 35 can create advice that praises the intake of nutrients with a high "protein / fat ratio," "omega-3 fatty acid / fat ratio," and "dietary fiber / sugar or carbohydrate ratio" estimated by the evaluation unit 34, and encourages the intake of nutrients with a low ratio. In this case, the advice creating unit 35 may create advice that suggests the intake of functional foods, foods for specified health uses, and foods containing functional ingredients. Note that the advice creating unit 35 does not necessarily have to create advice regarding nutrient intake.

[0098] The method of creating advice by the advice creating unit 35 is not particularly limited, but in addition to using the above-mentioned fixed phrases, the advice creating unit 35 can accumulate obtained data, perform machine learning, and create advice using the learning results. Furthermore, the advice creating unit 35 can create advice based on nutritional guidance by an expert such as a registered dietitian.

[0099] Alternatively, the advice creating unit 35 may create advice for each target user according to the eating habit classification. The advice creating unit 35 is not limited to creating advice according to the evaluation result of the recorded information by the evaluation unit 34, and may create advice according to the eating habit classification once the eating habit classification of the target user is determined by the eating habit classification unit 32.

[0100] [Disease Prevention Support System Operation] The operation of the disease incidence prevention support system 1 will be explained. The disease incidence prevention support system 1 provides a disease risk reduction program to target users. The disease incidence prevention support system 1 operates "before the program is implemented" and "during the program is implemented" as shown in Figures 7 and 8. Table 12 below shows the contents of "input," "judgment," and "output" before and during the program is implemented. Note that the implementation period of the disease risk reduction program is synonymous with the advice implementation period described above.

[0101] [Table 12]

[0102] Before the program is implemented, as shown in Figures 7 and 8, the prediction model generation unit 12 generates a prediction model based on the subject's health checkup data over several years (St1). The eating habit classification generation unit 14 generates an eating habit classification based on the subject's eating habit information and data related to biomarkers (St2). The body fat mass measurement unit 11 then measures the subject's current body fat mass based on a body image of the subject (St3).

[0103] Next, the morbidity risk assessment unit 13 inputs the current body fat mass into the prediction model and assesses the morbidity risk of the target user (St4). Next, the eating habit information acquisition unit 31 acquires the eating habit information of the target user (St5), and the eating habit classification unit 32 determines the eating habit classification of the target user (St6). The eating habit information acquisition unit 31 accepts input of eating habit information from the target user as shown in Table 12, and the eating habit classification unit 32 determines the eating habit classification of the target user.

[0104] Next, the advice generating unit 15 generates advice for reducing the morbidity risk in accordance with the evaluation result by the morbidity risk evaluating unit 13 (St7). The advice generating unit 13 calculates the difference between the target body fat mass of the target user and the current body fat mass supplied from the body fat mass measuring unit 11, and determines the exercise and diet contents so that the target user's body fat mass will be equal to or less than the target body fat mass during the implementation period of the program, and can generate advice suggesting these.

[0105] Next, the body fat mass prediction unit 16 predicts the future body fat mass of the target user that will be achieved by following the advice (St8). The body fat mass prediction unit 16 can predict the future body fat mass at the end of the program implementation period, assuming that the target user follows the advice generated by the advice generation unit 15.

[0106] Next, the morbidity risk prediction unit 17 inputs the future body fat mass into a prediction model and predicts the morbidity risk at the end of the advice implementation period (St9). Next, the presentation information generation unit 18 generates presentation information including the evaluation result by the morbidity risk evaluation unit 13, the prediction result by the morbidity risk prediction unit 17, and the advice generated by the advice generation unit 15, and presents it to the target user (St10). Note that the order of the steps (St1 to St10) before implementing the program may be reversed.

[0107] During the execution of the program, as shown in FIG. 8, the record information acquisition unit 33 acquires record information (see Table 5) (St11), and the evaluation unit 34 evaluates the record information (St12). The record information acquisition unit 33 accepts input of record information from the target user as shown in Table 12. The evaluation unit 34 calculates the score for each record item and each major item, and extracts the record items and sub-record items. The evaluation unit 34 may also evaluate the amount of nutrient intake.

[0108] Next, the advice creation unit 35 generates advice for eliminating the causes of body fat accumulation or lifestyle-related disease according to the dietary habit classification (St13), and the presentation information generation unit 18 presents presentation information including the advice to the target user (St14). As shown in Fig. 6, the presentation information generation unit 18 generates presentation information including advice, major item scores and time trends, upper record items and lower record items, etc., and outputs it to each target user. Note that the disease incidence countermeasure support system 1 may perform the steps from the current body fat mass measurement step (St3) to the presentation information generation step (St9) even during the execution of the program.

[0109] After performing the operations "before program implementation," the disease morbidity prevention support system 1 performs the operations "during program implementation" for each evaluation period. The evaluation period is, for example, one week. The program implementation period (advice implementation period) is, for example, one month, and the disease morbidity prevention support system 1 performs the operations "during program implementation" one or more times during the program implementation period.

[0110] In this way, the disease morbidity prevention support system 1 provides advice to reduce the risk of disease to each target user. The operation of the disease morbidity prevention support system 1 described above is an example, and the disease morbidity prevention support system 1 may provide advice to the target user to reduce the risk of disease at least either before or during the program. In addition to providing advice, the target user's eating habits classification can be used for product recommendations, service provision and collaboration, future predictions (body type, lifestyle-related diseases, etc.), introductions to friends (community formation), etc.

[0111] [Effects of the disease prevention support system] As described above, the disease morbidity prevention support system 1 measures the target user's body fat mass based on a body image, evaluates the risk of contracting a target disease based on that body fat mass, and generates advice to reduce the risk. Furthermore, the body fat mass that will be achieved by implementing the advice is predicted, and the risk of contracting the disease based on that body fat mass is also predicted. By presenting this information to the target user, the target user can understand their current risk of contracting a disease and the risk that will be reduced by implementing the advice, enabling them to actively work on implementing the advice.

[0112] Furthermore, in the disease prevention support system 1, the advice creation unit 35 creates advice based on the target user's eating habits classified by the eating habit classification unit 32, making it possible to provide effective advice for reducing the risk of disease for each target user.

[0113] In this case, the evaluation unit 34 evaluates the causes of body fat accumulation or lifestyle-related disease based on the record information on dietary content and activity content input by the target user, and the advice creation unit 35 creates advice based on the evaluation results, making it possible to provide advice that is in line with the dietary content and activity content of the target user. Also, the record information acquisition unit 33 presents the target user with a group of record items including record items according to the dietary habit classification results, making it possible to acquire effective record information for each dietary habit classification and use it to create advice.

[0114] Furthermore, the advice creation unit 35 can create advice using standard phrases prepared for each record item, thereby automatically creating advice according to the evaluation results of the record item, and by preparing standard phrases for each season, it is possible to provide advice appropriate for each season, such as seasonal ingredients.

[0115] Furthermore, the evaluation unit 34 evaluates the intake of nutrients from the recorded information, and the advice creation unit 35 creates advice based on the evaluation results, making it possible to provide advice according to the excess or deficiency of nutrients taken by the user. As nutrients, protein, omega-3 fatty acids, and dietary fiber have a large impact on body fat accumulation and the risk of lifestyle-related diseases, advice regarding these nutrients is particularly effective.

[0116] In providing advice, the disease incidence prevention support system 1 may refer to the content of the previous advice and the above-mentioned record information, have the evaluation unit 34 determine whether the target user was able to perform the items pointed out in the previous advice, and have the advice creation unit 35 create advice including an evaluation of the items pointed out in the previous advice according to the determination result. For example, if the target user was able to perform the pointed out items, the advice creation unit 35 may create a standard phrase praising the item, and if the target user was not able to perform the item, it may create a standard phrase encouraging the item.

[0117] Regarding the generation of eating habit classifications, the eating habit analysis unit 22 can generate multiple eating habit classifications for each factor by performing factor analysis on the eating habit information and data related to biomarkers, and can appropriately classify the target user's eating habits by prioritizing them based on the contribution rates obtained from the factor analysis. In this case, the eating habit analysis unit 22 calculates the factor loadings of the questions presented to the target user and selects each question item for each factor according to the factor loadings, thereby generating eating habit classifications based on the contents of the selected questions.

[0118] It is also possible to provide a separate advice selection unit that allows the target user to select whether or not to adopt the advice created by the advice creation unit 35. In this way, it becomes possible to provide advice that is easier for the target user to continue.

[0119] The optimal insurance product may be proposed in combination with the disease morbidity prevention support system 1. Furthermore, as part of the services provided by health guidance institutions, etc., the disease morbidity prevention support system 1 may be provided in combination with health consultations, health checkups, health education, health coaching, health programs for smoking and lack of exercise, etc. Furthermore, as part of the services provided by screening institutions, etc., the disease morbidity prevention support system 1 may be provided in combination with health checkups, cancer screenings, lifestyle-related disease screenings, prenatal checkups, and checkups conducted by companies for employee health management, etc.

[0120] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention. Note that, in the disease morbidity countermeasure support program and disease morbidity countermeasure support method, preferred examples are the same as those described above regarding the disease morbidity countermeasure support system 1. [Explanation of symbols]

[0121] 1. Disease prevention support system 11…Body fat measurement section 12...Prediction model generation unit 13. Morbidity Risk Assessment Department 13...Advice generation unit 14…Eating habits classification generation part 15...Advice generation unit 16...Body fat mass prediction section 17. Morbidity Risk Prediction Department 18…Presentation information generation unit 21...Data acquisition section 22…Eating habits analysis department 23...Ranking Division 31…Eating habits information acquisition department 32…Eating habits classification department 33...Recorded information acquisition unit 34...Evaluation Department 35...Advice Creation Department 36…Eating habits classification determination section 37…Eating habits classification determination unit

Claims

1. A disease morbidity prevention support system that supports a user in taking measures against disease morbidity, a body fat mass measuring unit that acquires a body shape image by capturing an image of the user's body and measures the user's body fat mass based on the body shape image; a disease risk assessment unit that inputs the body fat mass measured by the body fat mass measurement unit into a prediction model that defines the relationship between body fat mass and the incidence rate of a specific disease, and assesses the risk of the user developing the disease; an advice generating unit that generates advice for reducing the risk according to the evaluation result by the morbidity risk evaluation unit; a disease risk prediction unit that inputs the body fat mass of the user that is predicted to be achieved by following the advice into the prediction model and predicts the risk of the user contracting the disease; a presentation information generation unit that generates presentation information including the evaluation result by the morbidity risk evaluation unit, the prediction result by the morbidity risk prediction unit, and the advice; A disease prevention support system equipped with the above.

2. The disease is metabolic syndrome The disease morbidity prevention support system according to claim 1.

3. The body fat mass is VFA (Visceral Fat Area) The disease morbidity prevention support system according to claim 1.

4. The present invention further includes a prediction model generation unit that generates a prediction model, which is a machine learning model that defines the relationship between body fat mass and the morbidity rate, and supplies the prediction model to the morbidity risk assessment unit and the morbidity risk prediction unit. The disease prevention support system according to claim 1.

5. The prediction model generation unit generates the prediction model in which a relationship between the incidence rate and at least one of body fat mass, BMI (Body Mass Index), number of cigarettes smoked, diastolic blood pressure, systolic blood pressure, sex, and age is defined. The disease prevention support system according to claim 4.

6. The advice generation unit sets a target value of the user's body fat mass necessary to reduce the risk, and generates the advice so that the user's body fat mass will be equal to or less than the target value within a predetermined advice implementation period. The disease prevention support system according to claim 1.

7. The advice includes suggestions for at least one of exercise and diet. The disease prevention support system according to claim 6.

8. The body fat mass measuring unit measures a first length, which is a widthwise length of the front of the user's trunk, and a second length, which is a thicknesswise length of the side of the trunk, in the body type image, and measures the amount of body fat contained in the trunk from the first length and the second length. The disease prevention support system according to claim 1.

9. The body fat mass measuring unit measures the first length in a first captured image obtained by capturing the user's body from the front, and measures the second length in a second captured image obtained by capturing the user's body from the side. The disease morbidity prevention support system according to claim 8.

10. The advice generation unit has a dietary habit classification unit that classifies the user's dietary habits into one of a plurality of dietary habit classifications according to body fat accumulation factors or lifestyle-related disease morbidity factors based on dietary habit information that is information about the user's dietary habits, and generates advice to the user for eliminating causes of body fat accumulation or causes of lifestyle-related disease morbidity according to the dietary habit classification result. The disease prevention support system according to claim 1.

11. A disease prevention support program executed by an information processing device to support a user's disease prevention measures, the information processing device comprising: acquiring a body shape image by capturing an image of the user's body, and measuring the body fat mass of the user based on the body shape image; a step of inputting the body fat mass measured in the step of measuring the body fat mass into a prediction model that defines a relationship between body fat mass and the incidence of a specific disease, and assessing the risk of the user contracting the disease; generating advice for reducing the risk according to an assessment result of the risk assessment step; A step of inputting the body fat mass of the user that is predicted to be achieved by following the advice into the prediction model to predict the risk of the user suffering from the disease; generating presentation information including an evaluation result from the step of evaluating the risk, a prediction result from the step of predicting the risk, and the advice; A disease prevention support program that implements the following.

12. A disease morbidity prevention support method for supporting a user in taking measures against disease morbidity, acquiring a body shape image by capturing an image of the user's body, and measuring the body fat mass of the user based on the body shape image; inputting the body fat mass measured in the step of measuring the body fat mass into a prediction model that defines a relationship between body fat mass and the incidence of a specific disease, and assessing the risk of the user developing the disease; generating advice for reducing the risk according to an evaluation result of the step of evaluating the risk; inputting the user's body fat mass that is predicted to be achieved by following the advice into the prediction model to predict the user's risk of developing the disease; Generate presentation information including the evaluation result from the step of evaluating the risk, the prediction result from the step of predicting the risk, and the advice Methods for supporting disease prevention.

Images