JP2008176434A - Individual health guidance support system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an individual health guidance support system for determining guidance items showing high relevancy commonly to a plurality of diseases, and for presenting life style improvement items for reducing any risk to the whole plurality of diseases. <P>SOLUTION: This individual health guidance support system is provided with: an item classification means 111 for generating life style item classification for classifying the life style items of medical checkup information for every life style to be guided; a disease-categorized contribution calculation means 112 for analyzing the relations of the life style items and the disease development of symptoms, and for calculating the rate of the development of symptoms for every value of the life style items, and for calculating the contribution of every development of symptom; the number of items/combination contribution calculation means 113 for combining the life style items seem to be relevant to the development of symptoms for every classification of the life style items, and for calculating the weighted combination contribution as the sum of contributions weighted by the rate of the development of symptoms of the life style whose value is bad for every combination; and a guidance item candidate selection means 114 for selecting the combination of the life style items whose number of items is the minimum, and whose weighted combination contribution is the maximum as guidance items for every life style item classification. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an individual health guidance support system that presents information for individual disease prevention / health promotion from health checkup results.

Analyze the relationship between health checkup information accumulated at health checkup facilities and the onset of diseases such as diabetes, hypertension, and hyperlipidemia, and determine lifestyle related items that are related as guidance items. Next, risk knowledge is created by calculating the risk that a person who has the condition of the instruction item and the related inspection item or a combination of these values will develop each disease. Then, there is an individual health guidance support system that supports the health guidance by obtaining the disease risk corresponding to the health check result of the person being instructed from the risk knowledge. For example, in Patent Document 1, risk knowledge is generated by calculating the risk of occurrence of a test abnormality for each related test item and lifestyle item. A system that provides health guidance by obtaining the risk of laboratory abnormalities that correspond to the results of the health examination of the person being instructed based on the risk knowledge is introduced.
Next, the guidance item determination method used for disease risk calculation is described. As mentioned above, this system first determines the guidance items, but the amount of medical examination information held by a single medical examination facility may not always be sufficient, so all related lifestyle items are used as guidance items. This is difficult to realize because fewer people have conditions that combine these values. Therefore, it is necessary to select effective guidance items from the related lifestyle items of each disease, and as a method, it is commonly high for multiple diseases such as diabetes, hypertension, and hyperlipidemia as follows. A method of selecting a lifestyle item showing a relationship can be considered.
(1) (1) Analyze the relationship between lifestyle items and the onset of each disease, and calculate the onset rate by value of each item. Next, the value of lifestyle items with a high incidence rate is defined as a bad lifestyle value, and the value of lifestyle items with a low incidence rate is defined as a good lifestyle value. The degree of contribution (odds ratio) is calculated. Here, the odds ratio is (P ₁ / 1- P ₁ ) / (P ₀ / 1- P ₀ ), where P _{1 is} the incidence of bad lifestyles and P ₀ is the incidence of good lifestyles. )
(2) (2) Combining related lifestyle items, finding the sum of the number of items combined and the degree of contribution for each combination, and using the combination of items with the smallest number of items and the largest sum of contributions as guidance items select.

JP 2002-183647 A

The odds ratio used in the above method is calculated as (P ₁ / 1- P ₁ ) / (P ₀ / 1-P ₀ ), so for example, “breakfast days / four days or less → five days or more” Change in rate P ₁ : 40% → P ₀ : 32% and change in onset rate of “number of eating out days / week 2 days or more → 1 day or less” P ₁ : 45% → P ₀ : 37% odds ratio is 1.4 There is a problem that it becomes the same value. From the viewpoint of guidance, it is better to determine the guidance items in consideration of the change in the incidence rate, but the above method did not consider this point.

  In order to solve the above-mentioned problems and realize the purpose, the individual health guidance support system of the present invention generates items of lifestyle items including a plurality of lifestyle items classified for each type of lifestyle to be taught. Analyzing the relationship between the classification means and the lifestyle item and the onset of disease, calculating the onset rate indicating the proportion of people who have developed the value according to the value of the lifestyle item, and the onset rate is The value of the lifestyle item having a high value is a bad lifestyle value, the value of the lifestyle item having a low onset rate is a good value for the lifestyle, and the incidence rate of the good lifestyle and the value of the lifestyle is bad. A combination of lifestyle items included in the lifestyle item classification is generated and generated for each lifestyle item classification generated by the disease classification contribution calculating unit for calculating the contribution by disease from the incidence rate and the item classification unit, and generated To the combination The lifestyle item included in the generated combination and the weighted combination contribution that is the sum of the contribution calculated by the disease-specific contribution calculating means for the lifestyle item to be weighted by the onset rate of the lifestyle having a bad value And the number of items calculated by the number of items / combination contribution calculation means for each lifestyle item classification generated by the item classification means, It is characterized by having guidance item candidate selection means for selecting, as a guidance item, a combination of the lifestyle items having the smallest number of items and the largest weighted contribution contribution from the weighted combination contribution.

  Furthermore, the individual health guidance support system of the present invention is calculated by the disease-specific contribution calculating means for each lifestyle item included in the combination for each combination of lifestyle items generated by the number of items / combination contribution calculating means. The combination contribution and the number of items calculated by the item / combination contribution calculation means for each lifestyle item classification generated by the item classification means and the means for calculating the combination contribution that is the sum of the contributions Compare the means for selecting the lifestyle item combination with the smallest number of items and the largest contribution to the combination with the combination of the selected lifestyle item and the lifestyle item combination selected by the guidance item candidate selection means. If they are different, the combination of both is displayed as a guidance item candidate according to the lifestyle item classification generated by the item classification means.

  Furthermore, the individual health guidance support system according to the present invention includes a condition in which the values of the guidance item and the test item are combined, and an onset ratio indicating the ratio of persons who have developed among persons having the combination, and the number of people who have developed the total number of persons. It is characterized by having rule creation means for calculating the degree of support indicating the reliability of the onset rate as a rule for each disease, and creating risk knowledge for each disease.

  Furthermore, the individual health guidance support system of the present invention includes a rule search means for searching for the rule indicating the onset rate of each disease corresponding to the health check result of the target person from the risk knowledge of each disease, The lifestyle item is good one by one for the risk to people with good lifestyle and a bad value for the lifestyle item, showing the relative risk for the rule that all the lifestyle items are changed to good values for the lifestyle item Relative risk calculation means for calculating the relative risk for each item indicating the relative risk for the rule changed to a value for each disease, the combination risk indicating the sum of the relative risk for each item for each guidance item, and the disease for which the relative risk for each item is calculated The disease number / combination risk calculation means for calculating the number of diseases, which is the number of diseases, and the items for which the relative risk for each item exists are extracted as recommended improvement items. It is characterized by having an improvement item extracting means for extracting an item having the maximum number of diseases and the maximum combination risk as the most recommended improvement item from the number of diseases / combination risk.

  Furthermore, the individual health guidance support system according to the present invention is characterized in that the number of diseases / combination risk calculating means weights each guidance item according to the onset ratio of each disease according to the rule corresponding to the health check result of the guidance subject. The weighted combination risk indicating the sum of the relative risks and the number of diseases, which is the number of diseases for which the item-specific relative risk is calculated, are calculated, and the improvement item extracting means recommends items for which the item-specific relative risk exists. And the item having the largest number of diseases and the largest weighted combination risk is extracted as the most recommended improvement item from the disease number / weighted combination risk.

  Furthermore, the individual health guidance support system according to the present invention includes a guidance content creation means for displaying a list of the onset rate of each disease of the guidance target person, risks for people with good lifestyle habits, recommended improvement items, and most recommended improvement items. It is characterized by having.

  Furthermore, the individual health guidance support system according to the present invention displays, as a bar graph, the onset rate of each disease of the target person and the change in the onset rate when the recommended improvement item and the most recommended improvement item are good values. It is characterized by having instruction content creation means.

  Furthermore, the individual health guidance support system of the present invention displays the change rate of the onset rate of each disease of the target person and the change rate of the occurrence rate when the recommended improvement item and the most recommended improvement item are good values in a pie chart. It is characterized by having teaching content creation means.

  In the individual health guidance support system of the present invention, the number of items / combination contribution calculation means calculates the weighted combination contribution and the number of items for each combination of lifestyle items, and the guidance item candidate selection means weights with the minimum number of items. Since the combination of items with the maximum combination contribution is selected as a guidance item, there is an effect that it is possible to determine an important guidance item that shows a high relation in common with respect to multiple diseases and raises the incidence rate more.

  Furthermore, in the individual health guidance support system of the present invention, the item number / combination contribution calculating means calculates the weighted combination contribution, the number of items, and the combination contribution for each combination of lifestyle items, and the guidance item candidate selecting means , A combination of items in which the maximum weighted combination contribution and the maximum combination contribution reverse with the minimum number of items is extracted as a guidance item candidate. And, since the guidance item candidates are displayed according to the classification and the operator decides the final guidance item, the operator can determine the guidance item suitable for the guidance while checking the combination of items to be reversed. .

  Furthermore, in the individual health guidance support system according to the present invention, the rule creation means creates risk knowledge of each disease using the guidance items determined by the guidance item candidate selection means. This has the effect of creating risk knowledge to be presented.

  Furthermore, in the individual health guidance support system of the present invention, the number of diseases / combination risk calculation means calculates the combination risk and the number of diseases for each guidance item, and the improvement item extraction means is the most recommended improvement item as the maximum number of diseases. Since the item with the maximum combination risk is extracted in step 1, there is an effect that it is possible to present lifestyle improvement items that reduce the risk for all of the multiple diseases.

  Further, in the individual health guidance support system of the present invention, the disease number / combination risk calculation means calculates the weighted combination risk and the number of diseases for each guidance item, and the improvement item extraction means sets the number of diseases as the most recommended improvement item. Since the item with the highest weighted combination risk is extracted at the maximum, there is an effect that an improvement item that reduces the disease risk with a high onset ratio of the target person can be presented.

  Furthermore, in the individual health guidance support system of the present invention, the guidance content creation means displays a list of the onset rate of each disease of the guidance target person, risks for people with good lifestyle habits, recommended improvement items, and most recommended improvement items. Therefore, the operator can easily understand the improvement point of the person to be instructed, and has an effect of giving priority to the point.

  Furthermore, the individual health guidance support system according to the present invention provides the guidance content creation means that changes the incidence rate of each disease of the guidance subject and the incidence rate when the recommended improvement item and the most recommended improvement item are good values. Since it is displayed as a bar graph or pie chart, it has the effect of providing guidance that gives a visual impact to the target person.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings. In the following explanation, metabolic syndrome (visceral fat syndrome), diabetes, hypertension, and hyperlipidemia are given as examples of diseases, and the risk of these diseases and the content of improvements that reduce them are presented to the instructor. The explanation will be made assuming that the guidance is supported.

  FIG. 1 is a diagram showing a configuration example of an individual health guidance support system that is an embodiment of the present invention. The individual health guidance support system includes a health guidance support terminal 101 and a database 106. The health guidance support terminal 101 is a computer device, an input unit 102 such as a mouse or a keyboard, an output unit 104 such as a display or a printer, and a lifestyle that is related to each disease from the relationship between medical examination information and disease onset. Risk knowledge creating means 105 for creating risk knowledge having a plurality of rules indicating the condition in which the guidance item is determined and combining the values of the medical examination items and the onset rate of the condition, etc., and the guidance target input by the input unit 102 A guidance content creating means 103 for creating guidance contents such as the onset rate of each disease and improvement items corresponding to the health check result of the person from risk knowledge and displaying the guidance contents on the output unit 104 is provided.

  The risk knowledge creation means 105 uses an item classification means 111 that classifies lifestyle items of medical examination information by genre such as drinking, exercise, tobacco, etc., and the relationship between lifestyle items and each disease occurrence using a logistic regression model or the like. And a disease-specific contribution calculating means 112 for statistically analyzing and calculating a disease-specific contribution. In addition, combinations of lifestyle items that have been found to be related, and for each combination, the sum of contributions (combination contributions), the sum of contributions weighted by the incidence of bad lifestyles (weighted combination contributions), combined The number of items / combination contribution calculation means 113 that calculates the number of lifestyle items and the combination that minimizes the number of items and maximizes the weighted combination contribution and the combination that maximizes the combination contribution are operated as guidance item candidates. Guidance item candidate selection means 114 that prompts the user to make a decision. In addition, analysis by correlation rule mining is performed using the determined guidance items, test items (abdominal circumference, fasting blood glucose level, etc.) used to determine the onset of each disease, basic items (sex, age), etc. A rule creation means 115 for creating a plurality of rules for each disease, indicating the incidence rate (number of affected people / number of affected people) and support (an index indicating the reliability of the incidence rate) of a person having a combination of item values Have.

  The guidance content creation means 103 includes a rule retrieval means 107 that retrieves a rule indicating the incidence of each disease corresponding to the health check result of the guidance subject from the risk knowledge of each disease, and a condition of lifestyle habit guidance items in the retrieval rule Relative risk calculation means 108 for calculating the relative risk for each rule by changing the bad lifestyle value to a good value for each disease, and the number of diseases for which the sum of the relative risks and the number of diseases for which the relative risk has been calculated for each guidance item A combination risk calculation unit 109 and an improvement item extraction unit 110 that extracts lifestyle improvement items recommended for the guidance target person from the number of diseases and combination risk.

  The database 106 includes a medical examination information management means 119 for managing medical examination information, a contribution information management means 120 for managing the contribution of each lifestyle item by disease, the number of items, the contribution of the combination, and the contribution of the weighted combination, Instruction item information management means 121 for managing habit instruction items, risk knowledge management means 122 for managing risk knowledge of each disease, and instruction content management means 121 for managing instruction contents corresponding to the health checkup result of the person being instructed Have.

  FIG. 2 is a diagram showing an example of medical examination information managed by the medical examination information management means 119. As shown in FIG. Examination items such as BMI210, abdominal circumference 211, fasting blood glucose level, such as health checkup ID 201 that specifies health checkup information, personal ID 202 that specifies individuals, checkup date 203, basic items such as gender 204, age 205 at the time of checkup Information such as 212, systolic blood pressure 213, diastolic blood pressure 214, neutral fat 215, HDL cholesterol (HDL-C) 216 is managed. In addition, lifestyle items include weight gain from age 20 220, alcohol consumption / 221, drinking days / week 222, breakfast days / week 223, eating out days / week 224, meal size 225, meal speed 226, meal balance 227, smoking 228, regular exercise 229, sweating exercise 230, etc., as determined by the doctor from the test results, glucose metabolism determination 231 for diabetes, blood pressure determination 232 for hypertension, lipid determination 233 for hyperlipidemia It manages information such as.

FIG. 3 is a diagram showing an example of disease-specific contribution information managed by the contribution information management means 120. An item 301, a classification 302 of the item, and contributions 303 to 306 for each disease are managed. For example, a metabolic syndrome contribution of 4.0 with a weight gain of 10 kg or more from the age of 20 indicates an odds ratio of 10 kg or more to less than 10 kg, and 38% and 13% in parentheses indicate an onset rate P of 10 kg or more, respectively. ₁ , showing an onset rate P _{0 of} less than 10 kg. If there is a contribution value, it indicates that the item has a statistically significant association with the onset of the disease, and the greater the contribution, the higher the association. Moreover, when there is no value of contribution, it has shown that the item is not related to disease onset.

  FIG. 4 is a diagram showing an example of the number of items / combination contribution information managed by the contribution information management means 120. As shown in FIG. Item combination 401, classification 302, contributions 303 to 306 for each disease, number of combined items 402, combination contribution 403 that is the sum of contributions 303 to 306, weighted by the incidence of bad lifestyles A weighted combination contribution 404 that is the sum of contributions 303 to 306 is managed. A combination having a small number of items 402 and a high combination contribution 403 indicates that it is commonly related to the onset of all diseases and is highly related. In addition, the combination with a small number of items 402 and a high weighted combination contribution 404 is an important combination that raises the onset rate in addition to showing a common high association with the onset of all diseases. It is shown that.

  FIG. 5 is a diagram showing an example of a guidance item candidate display screen 501 displayed on the output unit 104 by the guidance item candidate selection unit 114. This screen displays the combination 401 of items classified by category selected as guidance item candidates from the number of items / combination contribution information in FIG. The vertical axis of the left bar graph on the screen 501 is the weighted combination contribution, and the vertical axis of the right bar graph is the combination contribution. Also, 502 to 506 are radio buttons for selecting a classification for displaying instruction item candidates, 510 to 513 are radio buttons for selecting the number of items, 530 to 531 are radio buttons for selecting an instruction item selection method, and 514 is , Buttons for determining final instruction items are shown. In this example, the operator selects the meal radio button 505. From the item number / combination contribution information in FIG. 4, the radio button 511 having the smallest number of meal items is automatically selected. As guidance item candidates for meals, combinations 413 and 414 of items whose weighted combination contribution maximum and combination contribution maximum are reversed are extracted and displayed. In the bar graph of the weighted combination contribution on the left side of the screen 501, 414 is indicated by 520 and 413 is indicated by 521, and in the bar graph of the right combination contribution, 414 is indicated by 522 and 413 is indicated by 523. The instruction item selection radio button automatically selects the combination having the largest weighted combination contribution from the displayed instruction item candidates as the most powerful instruction item candidate. In this example, the item combination 520 is selected. The final decision is made by the operator pressing the decision button 514. When the number of items is to be changed or the instruction item selection method is to be changed, the item number selection radio buttons 510 to 513, the instruction item selection method, respectively. Change the selection radio button 530-531. The instruction items determined on this screen are managed by the instruction item information management means 121 of the database 106 in the format shown in FIG.

  FIG. 6 is a diagram showing an example of instruction item information managed by the instruction item information management means 121. As shown in FIG. The classification 302 and the guidance items 601 to 604 for each finally determined disease are managed.

  FIG. 7 is a diagram showing an example of risk knowledge managed by the risk knowledge management means 122. As shown in FIG. Risk knowledge is created for each disease. In this example, risk knowledge for metabolic syndrome is shown. Risk knowledge is the metabolic syndrome of a person who has a combination of multiple conditions such as rule ID 801 that identifies the rule, basic items such as gender 204, examination items such as waist circumference 211, and guidance items such as weight gain 220 from age 20 Rules 810 to 817 indicating the onset rate 806 (number of affected persons 805 / number of affected persons 804) and support level 807 are managed. Here, the onset rate 806 indicates the number of onset persons 805 in the group (number of applicable persons 804) corresponding to the same examination / interview result condition divided by the number of persons in the group. The support level 807 is obtained by dividing the number 805 of patients in the group corresponding to the same examination / interview result condition by the number of the population, and indicates the reliability of the onset rate.

  Next, the operation will be described in detail using a flowchart and a sequence diagram. First, an example of a procedure for creating risk knowledge from medical examination information will be described with reference to the flowchart of FIG. 8 and the sequence diagram of FIG. 14 showing the exchange between the health guidance support terminal 101 and the database 106.

  When creation of risk knowledge is started (701), an item classification step 702 is performed first. Here, the item classification means 111 acquires the lifestyle items of the medical examination information of FIG. 2 managed by the medical examination information management means 119. Next, the item classification means 111 classifies lifestyle items according to lifestyles to be instructed. For example, classify into weight, alcohol consumption, tobacco, diet, exercise, etc. In this case, weight gain from 20 years old is 220, body weight, alcohol consumption / 221, drinking days / week 222 drinking, smoking 228 cigarettes, breakfast days / week 223, eating out days / week 224, diet 225, fast eating 226, dietary balance 227 is classified as diet, regular exercise 229, and sweating exercise 230. As a result, it is possible to select various instruction items according to classification, thereby enabling various instruction.

Next, a disease-specific contribution calculation step 703 is performed. Here, first, the disease-specific contribution calculation means 112 acquires the lifestyle items of the medical examination information of FIG. 2 managed by the medical examination information management means 118. Next, the disease-specific contribution calculating means 112 statistically analyzes the relationship between lifestyle items and the onset of each disease using a logistic regression model, etc., and among those who have the value for each lifestyle item value. The onset rate indicating the rate of people who have developed symptoms is calculated. The values of lifestyle items with a high incidence rate are those with bad lifestyle habits, the values of lifestyle items with a low incidence rate are those with good lifestyle habits, and the incidence rates with good lifestyle habits and those with bad lifestyle habits The contributions 303 to 306 for each disease onset are calculated from the onset rate. In addition, when the onset rate of the good value is equal to the onset rate of the bad value, it is treated as having no contribution. Specifically, the odds ratio and 95% confidence interval are calculated for each disease and item using this model. Here, the odds ratio is the odds of onset in the group with bad lifestyle habits relative to the odds that occur in groups with good lifestyle habits (ratio of probability p to onset to 1-p probability of not developing disease). It is a ratio, and shows the height of the onset risk of the group with a bad value to the group with a good lifestyle habit. For example, if the odds ratio of the group with smoking to the group without smoking is 1.5, it means that the risk of developing the disease with the group with smoking is 1.5 times higher than that with the group without smoking. Assuming that the incidence rates of the group having a good lifestyle habit and the group having a bad habit are p ₀ and p ₁ , they are calculated by the following equations.

また，信頼区間は，真のオッズ比が存在していると思われる区間とその信頼度を示すものである。例えば，あるオッズ比の95％信頼区間が，1.2〜1.8であった場合，真のオッズ比は，1.2〜1.8の区間にあると考えられ，その信頼度は95％であることを意味する。信頼度は，通常95％がよく用いられる。つまり，ある健診項目のオッズ比の95％信頼区間を求め，その最低値が1より大きい値であれば，その項目はその疾病発症に対して統計的に有意な関連が見られる項目となる。そこで，疾病別，項目別に，オッズ比と95％信頼区間を算出し，95％信頼区間の最低値が1より大きい場合は，算出されたオッズ比がその項目の寄与度となる。1以下の場合は，寄与がない項目となる。これにより，疾病発症に有意な関連が見られる項目を抽出できる。算出された疾病別寄与度情報は，図3の形式でデータベース106に記録される。例えば，喫煙313の糖尿病寄与度304の値1.6は，喫煙なし群に対する喫煙あり群の糖尿病発症リスクが1.6倍高いことを示しており，喫煙313が，糖尿病発症に有意な関連が見られる項目であることを示している。一方，喫煙313のメタボリックシンドローム寄与度303の値が書かれていないところは，喫煙313が，メタボリックシンドロームの発症に有意な関連が見られない項目であることを示している。

The confidence interval indicates an interval where a true odds ratio is considered to exist and its reliability. For example, if the 95% confidence interval of a certain odds ratio is 1.2 to 1.8, the true odds ratio is considered to be in the interval of 1.2 to 1.8, which means that the reliability is 95%. A reliability level of 95% is usually used. In other words, if a 95% confidence interval for the odds ratio of a health check item is obtained and the minimum value is greater than 1, that item is a statistically significant item for the onset of the disease. . Therefore, the odds ratio and 95% confidence interval are calculated for each disease and item. If the minimum value of the 95% confidence interval is greater than 1, the calculated odds ratio is the contribution of the item. If the value is 1 or less, the item has no contribution. This makes it possible to extract items that are significantly related to disease onset. The calculated disease contribution information is recorded in the database 106 in the format of FIG. For example, a value of 1.6 for diabetes contribution 304 for smoking 313 indicates that the risk of developing diabetes in the smoked group is 1.6 times higher than that in the non-smoking group, and smoking 313 is an item that is significantly associated with the onset of diabetes. It shows that there is. On the other hand, where the value of the metabolic syndrome contribution 303 of smoking 313 is not written, smoking 313 indicates that there is no significant association with the onset of metabolic syndrome.

Next, the number of items / combination contribution calculation step 704 is performed. Here, first, the number-of-items / combination contribution calculation unit 113 acquires the disease-specific contribution information of FIG. 3 managed by the contribution information management unit 120. Next, the number-of-items / combination contribution calculation unit 113 combines items having a contribution value for each category, and calculates the number of items 402, the combination contribution 403, and the weighted combination contribution 404 for each combination. The item number 402 is the number of combined items. The combination contribution 403 is the sum of the disease-specific contributions 303 to 306, and the weighted combination contribution 404 is the sum of the disease-specific contributions 303 to 306 weighted by the onset rate P ₁ of the group having a bad lifestyle. , Respectively, are calculated by the following equations.

図3の食事の例では，メタボリックシンドロームの寄与度の値がある項目は，外食日数/週，食事量，食事早さの3個，糖尿病では朝食日数/週，外食日数/週の2個，高血圧では食事量の1個，高脂血症では朝食日数/週，外食日数/週，食事早さ，食事バランスの4個である。したがって，その組合せ数は，3×2×1×4で24となり，その組合せ別に項目数と組合せ寄与度と重み付け組合せ寄与度を算出する。例えば，項目の組合せとして，食事量（メタボリックシンドローム），朝食日数/週（糖尿病），食事量（高血圧），朝食日数/週（高脂血症）を組合せた場合，項目数は，食事量，朝食日数/週の2，組合せ寄与度は，各疾病の寄与度，1.5（メタボリックシンドローム），1.5（糖尿病），1.4（高血圧），1.4（高脂血症）の和である5.8となる。また，重み付け組合せ寄与度は，生活習慣が悪い値を持つ群の発症割合P₁で重み付けした各疾病の寄与度，0.24×1.5（メタボリックシンドローム），0.13×1.5（糖尿病），0.30×1.4（高血圧），0.40×1.4（高脂血症）の和である1.5となる。算出された項目数・組合せ寄与度情報は，図4の形式でデータベース106に記録される。

In the meal example of Figure 3, the items with metabolic syndrome contribution values are the number of eating out days / week, the amount of meals, and the speed of eating, 3 for diabetes, breakfast days / week, eating out days / week, For hypertension, there is 4 meals; for hyperlipidemia, 4 meals are breakfast days / week, eating days / week, meal speed, and meal balance. Accordingly, the number of combinations is 3 × 2 × 1 × 4, which is 24, and the number of items, combination contributions, and weighted combination contributions are calculated for each combination. For example, if the combination of items is meal amount (metabolic syndrome), breakfast days / week (diabetes), meal amount (hypertension), breakfast days / week (hyperlipidemia), the number of items is the amount of meal, The number of breakfast days / week, 2, the combined contribution is 5.8, which is the sum of the contribution of each disease, 1.5 (metabolic syndrome), 1.5 (diabetes), 1.4 (hypertension), and 1.4 (hyperlipidemia). Further, the weighting combination contribution is contribution of each disease weighted by onset rate P ₁ of the group with lifestyle bad value, 0.24 × 1.5 (metabolic syndrome), 0.13 × 1.5 (diabetes), 0.30 × 1.4 (hypertension ), 0.40 × 1.4 (hyperlipidemia), which is 1.5. The calculated number of items / combination contribution information is recorded in the database 106 in the format of FIG.

  In the sequence diagram of FIG. 14, the health guidance support terminal 101 acquires medical examination information 1403 from the database 106, calculates disease-specific contribution information, item count / combination contribution information, and performs registration 1404 thereof.

  Next, a guidance item candidate display step 705 is performed. Here, first, the guidance item candidate selection unit 114 acquires the number of items / combination contribution level information of FIG. 4 managed by the contribution level information management unit 120. Next, the guidance item candidate selection unit 114 determines, from the acquired item number / combination contribution information, as a guidance item candidate, a combination of items in which the maximum weighted combination contribution degree and the maximum combination contribution degree are reversed with the minimum number of items for each classification. In particular, the maximum weighted combination contribution is selected as the most promising instruction item candidate. As a result, it is possible to select an important instruction item that is highly related to all diseases and that has a higher incidence rate. For example, in the case of meals, the minimum number of items and the maximum weighted combination contribution are the combination of the number of eating out days / week and the amount of meal 414, and the maximum combination contribution is the combination of breakfast days / week and the amount of meal 413, which are reversed. The combination of items that are reversed in this way is extracted as a guidance item candidate, and among them, the combination of the number of eating out days / weeks and the amount of meals that has the maximum contribution to the weighted combination is selected as the most likely guidance item candidate. If there is no reversal, the combination of items with the minimum number of items and the maximum weighted combination contribution is the most likely instruction item candidate.

  Next, a guidance item determination step 706 is performed. Here, first, the instruction item candidate selecting means 114 displays the selected instruction item candidate on the output unit 104 as in the instruction item candidate display screen 501 of FIG. Let me decide. The operator first selects a classification for displaying instruction item candidates from the classification radio buttons 502 to 506. Next, instruction items are selected from instruction item selection method selection radio buttons 530 to 531. In the default state, the most promising guidance item candidate with the minimum number of items and the maximum weighted combination contribution is selected. Finally, press the decision button 514 to decide the final instruction item. For example, if the meal radio button 505 is pressed, item combinations 414 (FIG. 5: 520) and 413 (FIG. 5: 523) are displayed as instruction items for meals. Selection method selection radio button 530 is selected. If this item is acceptable, press the decision button 514 to decide the instruction item. If not, the selection of the number-of-items radio buttons 510 to 513 and the instruction item selection method selection radio buttons 530 to 531 is changed. The determined guidance item information is recorded in the database 106 in the format of FIG.

  In the sequence diagram of FIG. 14, the health guidance support terminal 101 acquires contribution information 1405 from the database 106, determines guidance items, and performs registration 1406 thereof.

  Next, risk knowledge creation step 707 is performed. Here, first, the rule creation means 115 acquires the guidance item information of FIG. 6 managed by the guidance item management means 121 and the medical examination information of FIG. 2 managed by the medical examination information management means 119. Next, the rule creation means 115 uses the guidance items, the test items used to determine the onset of each disease (abdominal circumference, fasting blood glucose level, etc.), basic items (gender, age), etc. Analyzing by mining, creating multiple rules indicating the condition part that combines the values of the health checkup items, the incidence rate (number of affected persons / number of applicable persons) and the degree of support for each condition part, and the risk knowledge of each disease create. Here, the onset rate of the disease is obtained by extracting people who are not sick at the first time from the health examination information for a plurality of years, and obtaining the rate of those who subsequently developed the disease. For example, in the metabolic syndrome, the onset of disease is determined using academic criteria (abdominal circumference: 85 cm or more for men, 90 cm or more for women, and abnormalities in two or more of hyperglycemia, lipid abnormalities, and high blood pressure). In addition, as shown in FIG. 7, the created risk knowledge includes a condition part that combines values of a plurality of medical examination items, an onset rate 806 (number of affected persons 805 / number of applicable persons 804), This is data in which the support degree 807 is recorded. For example, rule 810 is male, age 40s, waist circumference 85 cm or more, BMI 25 or more, systolic blood pressure less than 130 and diastolic blood pressure less than 85, triglyceride less than 150 and HDL-C40 or more, weight gain from 20 years old over 10 kg, meal The incidence of metabolic syndrome is 53% (10 patients / 19 patients), and the degree of support indicating the reliability of the incidence is high. It is 0.1%. For risk knowledge, a rule having such a combination of various conditions is prepared. When risk knowledge for all diseases is created, the creation of risk knowledge is terminated (708).

  In the sequence diagram of FIG. 19, the health guidance support terminal 101 acquires guidance items and medical examination information 1407 from the database 106, creates risk knowledge, and performs registration 1408 thereof.

  Next, an example of the flow of processing from health check result input to instruction content display will be described with reference to the flowchart of FIG. 13, the sequence diagram of FIG. 14, the risk knowledge of FIG. 7, and FIGS. 9, 10, 11, and 12. explain. This process is a process in the case where a leader such as a doctor or a public health nurse provides guidance to a guidance target person such as a health check-up person by presenting the danger to each disease and the content of improvement to reduce it.

  FIG. 9 is a diagram showing an example of instruction content corresponding to the health check result of the instructed person managed by the instruction content management means 123. It manages the disease 901, the incidence 910 for the disease, the number of affected people 911, the number of affected people 912, the risk 913 for people with good lifestyle habits, and the relative risks 914 to 919 for each guidance item. In addition, the number of diseases 906 in which there is a relative risk for each instruction item and the combined risk 907 that is the sum of the relative risks are managed. For example, 902 were 20 people (911) with 76% (number of affected people: 911) who were in the same health condition as a certain target person, and the ratio was 28.3% ( Incidence rate 910), people with good lifestyle habits, here is less than 10kg weight gain from age 20 years, alcohol consumption / 3 days or less, people with regular exercise are 1.6 times higher risk of onset It shows that weight gain from 20 years old is 1.4 times higher for people under 10 kg, 1.3 times higher for people under 3 alcohol / day, and 1.2 times higher for people with regular exercise. ing. In addition, guidance items that do not have a relative risk indicate that the health check result of the subject is good, is not a related item of the disease, or is not a major factor in the onset of the disease for the subject. ing.

  10 and 11 are diagrams showing an example of a medical examination result input screen 1001 displayed on the output unit 104 by the guidance content creation means 103, and FIG. 10 shows a state before inputting the medical examination result of the person to be instructed. FIG. 11 is a diagram showing a state after inputting the health check result of the instructed person. This screen is created using the medical examination items used to create the risk knowledge. 1003 to 1015 are medical checkup items, 1020 to 1021 are buttons for selecting sex, 1022 to 1024 are buttons for selecting age, 1025 to 1026 are buttons for selecting abdominal circumference, 1027 to 1028 are buttons for selecting BMI, and 1029 to 1031 is a button for selecting a fasting blood glucose level, 1032 to 1033 are buttons for selecting a systolic blood pressure and a minimum blood pressure, 1034 to 1035 are buttons for selecting a triglyceride / HDL-C, and 1036 to 1037 are weights from the age of 20 Button to select increase, buttons to select alcohol amount / day for 1038 to 1039, buttons to select smoking for 104040 to 1041, buttons to select meal amount for 1042 to 1043, days to eat for 1044 to 1045 , Buttons 1046 to 1047 are buttons for selecting a regular exercise, and 1002 is an execution button. FIG. 12 is a diagram showing an example of a guidance content display screen 1201 displayed on the output unit 104 by the guidance content creation means 103, which corresponds to the health check result of the target person entered on the health check result input screen 1001. It is a figure which shows the state which displayed the guidance content which was performed. On the health examination result input screen 1001, select conditions from the health examination items 1003 to 1015 and press the execution button 1002. On the guidance content display screen 1201, onset rate display fields 1210, 1220, 1230, 1240 The ratio, the number of affected persons display fields 1211, 1221, 1231, and 1241 display the number of patients with each disease, and the number of applicable persons display fields 1212, 1222, 1232, and 1242, respectively. The risk display fields 1213, 1223, 1233, and 1243 are risks to people with good lifestyle habits, the recommended improvement item display fields 1214, 1224, 1234, and 1244 are the guidance items for which the relative risk is calculated, and the most recommended improvement item display field 1250. The guidance item with the maximum number of diseases selected from the guidance content information and the maximum combination risk is displayed.

  When the processing of FIG. 13 is started (1301), first, a medical examination result input step 1302 is performed. In the health check result input step 1302, the rule search means 107 inputs the health check result of the guidance target person. First, the screen of FIG. 10 is displayed on the output unit 104, and the input of the health check result of the instructed person is awaited. Then, when the operator inputs the health check result of the guidance target person for the health check items 1003 to 1015 by operating the input unit 102 and presses the execution button 1002, the rule search unit 107 acquires the input condition. Here, the results of the health checkups for the guidance subjects are male (1020), age 40s (1023), abdominal circumference less than 85 (1025), BMI less than 25 (1027), fasting blood glucose level of 100 to 109 (1030), and systolic blood pressure 130 or more or diastolic blood pressure 85 or more (1033), triglyceride less than 150 and HDL-C40 or more (1034), weight gain from 20 years of age more than 10kg (1037), alcohol amount more than 3 days a day (1039), smoking Yes (1041), large amount of meal (1043), number of eating out days / two days or more (1045), regular exercise (1047).

  Next, a rule search step 1303 is performed. Here, the rule search means 107 searches for a rule corresponding to the medical examination result input in the medical examination result input step 1302 from the risk knowledge. Specifically, in order to raise the prevention awareness of the person to be instructed, a rule that is reliable and has the maximum onset rate is searched from rules that can be combined with the values of the input health checkup results. In this case, a rule 812 that is reliable (here, 0.1% or more) and has the maximum onset rate among the rules that can be combined with the values of the inputted medical examination results is used as the search result. The onset rate, the number of affected persons, and the number of applicable persons as search results are recorded in the database 106 in the format of FIG.

  Next, a relative risk calculation step 1304 is performed. Here, the relative risk calculation means 108 calculates a risk for a person with good lifestyle and a relative risk for each instruction item for each disease from the risk knowledge of each disease. Specifically, first, conditions with bad lifestyle habits are extracted from the rules selected in the rule search step 1303 under the conditions of the guidance items of the guidance target person. In this case, the rule 812 in FIG. 7 extracts a weight increase of 10 kg or more from the age of 20 and an amount of alcohol more than 3 days / day and no regular exercise. Next, select a rule that changes only the extracted conditions to a condition with good lifestyle, and as a risk to a person with good lifestyle, before the change to the rule after the change (good lifestyle) (bad lifestyle) Calculate the odds ratio for the rule. In this case, weight gain from age 20 under the conditions of rule 812, alcohol consumption / day, regular exercise only, weight gain from 20 years old over 10 kg → less than 10 kg, alcohol amount per day over 3 days → over 3 In the following, when the rule 814 that has been changed from “No regular exercise” to “Yes” is selected, and the odds ratio of the rule 812 to the rule 814 is calculated, the risk for a person with good lifestyle is 1.6 times. Subsequently, the relative risk for each instruction item is calculated. The odds ratio is calculated by selecting a rule in which the condition of bad lifestyle is changed to a good condition one by one under the condition of the guidance item of the guidance target person. In this case, first, rule 815 in which only the weight gain from the age of 20 under the rule 812 is changed from 10 kg to less than 10 kg is selected, and an odds ratio of 1.4 is calculated as a relative risk of weight gain from the age of 20. Next, rule 816 in which only the alcohol amount / day is changed from more than 3 to 3 or less under the condition of rule 812 is selected, and the odds ratio 1.3 is calculated as the relative risk of alcohol amount / day. Then, the rule 817 in which only the regular exercise is changed to “None → Yes” under the condition of the rule 812 is selected, and the odds ratio 1.2 is calculated as the relative risk of the regular exercise. The calculated risk for a person with good lifestyle and the relative risk for each instruction item are recorded in the database 106 in the format of FIG.

  Next, a disease number / combination risk calculation step 1305 is performed. Here, the number of diseases / combination risk calculation means 109 calculates the number of diseases and the combination risk from the relative risk for each instruction item in FIG. 9 calculated in the relative risk calculation step 1304. Specifically, a combined risk that is the sum of the number of diseases for which there is a relative risk for each instruction item and the relative risk of each disease is calculated. For example, in the case of weight gain from age 20, there is a relative risk for all diseases, so the number of diseases is 4, and the combined risk is calculated as 6.1. In addition, since there is a relative risk for smoking only in diabetes and hyperlipidemia, the number of diseases is 2, and the combined risk is calculated as 3.6. The calculated number of diseases and combination risk are recorded in the database 106 in the format of FIG.

In the sequence diagram of FIG. 14, the health guidance support terminal 101 makes a rule acquisition request 1409, acquires a rule 1410 from the database 106, and performs instruction content registration 1411.
Next, an improvement item extraction step 1306 is performed. Here, first, the improvement item extraction means 110 extracts instruction items having relative risks from the instruction contents shown in FIG. 9 for each disease, and sets these as recommended improvement items for each disease. Next, the improvement item extraction means 110 extracts the guidance item having the maximum number of diseases and the maximum combination risk from the number of diseases / combination risk, and makes this the most recommended improvement item. Thereby, the improvement item which reduces the onset risk of each disease and the improvement item which reduces the onset risk of all the diseases in common can be extracted. In this case, the recommended improvement items for metabolic syndrome are less than 10 kg of weight gain from 20 years old, alcohol amount / three days or less, and regular exercise, and the recommended improvement items for diabetes are less than 10 kg weight gain from age 20 years. , Alcohol / day 3 or less, no smoking. The recommended improvement items for hypertension are less than 10 kg of weight gain from age 20 years, alcohol consumption is less than 3 days a day, and the amount of food is not much. Recommended improvement items for hyperlipidemia are less than 10 kg weight gain from age 20 years , Alcohol / day 3 or less, no smoking, eating out days / week or less. Furthermore, the most recommended improvement is less than 10 kg of weight gain from the age of 20 with the highest number of diseases and the highest combined risk.

  Next, instruction content display step 1307 is performed. Here, the guidance content creation means 103 displays the guidance content in FIG. 9, the recommended improvement item extracted in the improvement item extraction step 1306 and the most recommended improvement item on the guidance content display screen in FIG. 12. At this time, recommended improvement items for each disease are displayed in descending order of relative risk. Using this screen, the instructing operator presents the instructor with the recommended rate of improvement and the most recommended improvement items to reduce the risk, such as the incidence of each disease and the risk to people with good lifestyle habits. And give guidance. When the instruction is over, the process ends (1308).

  As described above, according to the individual health guidance support system of the present invention, the number of items / combination contribution calculation means calculates the weighted combination contribution and the number of items for each combination of lifestyle items, and the guidance item candidate selection means. However, since the combination of items with the smallest number of items and the largest weighted combination contribution is selected as the guidance item, important guidance items that are commonly associated with multiple diseases and that have a higher incidence are determined. There is an effect that can be done.

  Also, in the individual health guidance support system of the present invention, the number of items / combination contribution calculating means calculates the weighted combination contribution, the number of items, and the combination contribution for each combination of lifestyle items, and the guidance item candidate selecting means , A combination of items in which the maximum weighted combination contribution and the maximum combination contribution reverse with the minimum number of items is extracted as a guidance item candidate. And, since the guidance item candidates are displayed according to the classification and the operator decides the final guidance item, the operator can determine the guidance item suitable for the guidance while checking the combination of items to be reversed. .

  In the individual health guidance support system of the present invention, the rule creation means creates risk knowledge of each disease using the guidance items determined by the guidance item candidate selection means. This has the effect of creating risk knowledge to be presented.

  In addition, the individual health guidance support system of the present invention is such that the disease number / combination risk calculation means calculates the combination risk and the number of diseases for each guidance item, and the improvement item extraction means is the most recommended improvement item as the maximum number of diseases. Since the item with the maximum combination risk is extracted in step 1, there is an effect that it is possible to present lifestyle improvement items that reduce the risk for all of the multiple diseases.

  Also, in the individual health guidance support system of the present invention, the guidance content creation means displays a list of the onset rate of each illness of the target person, risks to people with good lifestyle habits, recommended improvement items, and most recommended improvement items. Therefore, the operator can easily understand the improvement point of the person to be instructed, and has an effect of giving priority to the point.

  In the above embodiment, in the disease number / combination risk calculation step 1305, the disease number / combination risk calculation means 109 calculates a combination risk that is the sum of the number of diseases and the relative risk, and in the improvement item extraction step 1306, an improvement item extraction means. 110 explained the example of extracting the guidance item with the maximum number of diseases and the maximum combination risk as the most recommended improvement item from the number of diseases and the combination risk. However, in the disease number / combination risk calculation step 1305, the disease number / combination risk calculation means 109 calculates the sum of the relative risks weighted by the incidence 910 of each disease, and is most recommended from the disease number / weight combination risk. As an improvement item, a guidance item having the maximum number of diseases and the maximum weighted combination risk may be extracted. This will be specifically described with reference to FIG. For example, for weight gain 914 from age 20, weighted combined risk 908 is relative risk weighted by the incidence of each disease, 0.26 x 1.4 (metabolic syndrome), 0.19 x 1.5 (diabetes), 0.61 x 1.9 (hypertension) , 0.51 x 1.3 (hyperlipidemia), and the value is 2.47. In addition, in the case of the amount of alcohol / day, the sum is 0.26 × 1.3 (metabolic syndrome), 0.19 × 1.3 (diabetes), 0.61 × 2.0 (hypertension), 0.51 × 1.4 (hyperlipidemia), and the value is 2.52. Become. Then, from the calculated number of diseases / weighted combination risk, the most recommended improvement item is the alcohol amount / day 3 or less with the maximum number of diseases 906 and the maximum weighted combination risk 908. Thereby, it is effective in extracting the improvement item which reduces mainly the disease risk with a high onset rate of the guidance subject.

  In the above embodiment, in the guidance content display step 1307, the guidance content creation means 103 sets the guidance content, recommended improvement items, and most recommended improvement items in FIG. 9 as numerical values as shown in the guidance content display screen in FIG. In this example, the operator gives guidance. However, in the guidance content display step 1307, the guidance content creation means 103 displays the onset rate of each disease of the guidance subject as bar graphs 1610 to 1613 and pie graphs 1910 to 1913 as in the examples of FIGS. Also good. In addition, the guidance content creation means 103 buttons the recommended improvement items and the most recommended improvement items for each disease (1630-1632: metabolic syndrome, 1640-1642: diabetes, 1650-1652: hypertension, 1660-1663: high Lipidemia, 1670: The most recommended improvement item), and when the operator presses the button, the incidence rate of good lifestyle habits as in the examples of FIG. 17, FIG. 18, FIG. 20, and FIG. You may display. FIGS. 17 and 20 are examples of screens of bar graphs and pie charts when the buttons of 1630, 1640, 1651, and 1662 for weight gain of less than 10 kg from the age of 20 are pressed, and the onset rate of each disease less than 10 kg 1710-1713 (Bar graph) Shows 2010-2013 (pie graph). Figures 18 and 21 are examples of bar graphs and pie charts when the 1670 button for alcohol content / day 3 or less, which is the most recommended improvement item, is pressed. The onset rate 1810-1813 (bar graph) 2110-2113 (pie graph) of each disease is shown. Thus, the operator can instruct the instructing subject by presenting the change in the onset rate in a graph, thereby instructing the instructor to give a more visual impact.

  Further, in the above embodiment, in the guidance item determination step 706, the guidance item candidate selection means 114 displays the guidance item candidates on the guidance item candidate display screen 501 in FIG. Although the example of determining is described, the instruction item determining step 706 may be omitted, and the combination having the minimum number of items and the maximum weighted combination contribution may be used as the final instruction item. Thereby, an operator's effort can be reduced.

  In the above embodiment, metabolic syndrome, diabetes, hypertension, and hyperlipidemia have been described as examples of diseases, but other diseases may be used. It can be used for medical examination items and lifestyle-related diseases such as hyperuricemia and LDL cholesterolemia. By doing in this way, there is an effect that can support guidance for various diseases.

  In the above embodiment, the method for calculating the odds ratio using a logistic regression model has been described as a method for calculating the contribution by disease, but other methods may be used. For example, other statistical models such as the Cox proportional hazard model can be used. The contribution may not be an odds ratio. For example, the ratio or difference between the onset rate of bad and good lifestyle habits may be used, or the operator who guides may set his own contribution. By doing in this way, there exists an effect which can show the guidance content which reflected the operator's intention more.

  In the above embodiment, the case where the onset rate is used as the risk of illness has been described as an example, but other indicators may be used. Indicators calculated by other methods such as health and risk, statistical indicators, and all indicators related to health and illness can be used. Moreover, although the said Example demonstrated the method of using an association rule mining as a method of creating risk knowledge, you may create risk knowledge using another mining method.

  In the above-described embodiment, the method of inputting the medical examination result with the button or the like has been described. However, other methods may be used. For example, a text input field may be provided and input from a keyboard or the like, or a slide bar type input I / F may be provided to set a numerical value. Various user interfaces can be used.

  Further, in the above embodiment, a method has been described in which the operator inputs the health check result using the input unit 104. However, the rule search unit 107 stores the health check result of the person to be instructed and the health check information in FIG. It may be obtained from the prepared database 106. As a result, it is possible to reduce the time and effort for the operator to input the medical checkup result.

The figure which shows the example of 1 structure of the individual health guidance assistance system of this invention. The figure which shows an example of the medical examination information which a medical examination information management means manages. The figure which shows an example of the contribution information classified by disease which a contribution information management means manages. The figure which shows an example of the number of items and combination contribution information managed by the contribution information management means. The figure which shows an example of the guidance item candidate display screen which determines a guidance item. The figure which shows an example of the guidance item information which guidance item management information manages. The figure which shows an example of the risk knowledge which a risk knowledge management means manages. The flowchart which shows an example of the flow of the process which creates risk knowledge from medical examination information. The figure which shows an example of the instruction content which the instruction content management means manages It is a figure which shows an example of the medical examination result input screen which inputs the medical examination result of a guidance subject, and is a figure which shows the state before input. It is a figure which shows an example of the medical examination result input screen which inputs the medical examination result of a guidance object person, and is a figure which shows the state after input. The figure which shows an example of the guidance content display screen which displays the guidance content. The flowchart which shows an example of the flow of a process from a medical examination result input to guidance content display. The sequence diagram which shows an example of exchange with a health guidance support terminal and a database. The figure which shows an example of the guidance content at the time of using the weighting combination risk which a guidance content management means manages. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a bar graph, and is a figure which shows the state before pressing an improvement item button. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a bar graph, and is a figure which shows the state after pushing a recommendation improvement item button. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a bar graph, and is a figure which shows the state after pressing the most recommended improvement item button. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a pie chart, and is a figure which shows the state before pressing an improvement item button. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a pie chart, and is a figure which shows the state after pushing a recommendation improvement item button. It is a figure which shows an example of the instruction | indication content display screen which displays an onset rate with a pie chart, and is a figure which shows the state after pressing the most recommended improvement item button.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Health guidance support terminal, 102 ... Input part, 103 ... Instruction content creation means, 104 ... Output part, 105 ... Risk knowledge creation means, 106 ... Database, 107 ... Rule search means, 108 ... Relative risk calculation means, 109 ... Disease number / combination risk calculation means, 110 ... improvement item extraction means, 111 ... item classification means, 112 ... disease-specific contribution calculation means, 113 ... number of items / combination contribution calculation means, 114 ... guidance item candidate selection means, 115 ... rule creation means, 119 ... medical examination information management means, 120 ... contribution information management means, 121 ... guidance item information management means, 122 ... risk knowledge management means, 123 ... guidance content management means, 201 ... health examination ID, 202 ... Personal ID, 203 ... Date of consultation, 204 ... Gender, 205 ... Age, 210 ... BMI, 211 ... Abdominal circumference, 212 ... Fasting blood glucose level, 213 ... Maximum blood pressure, 214 ... Minimum blood pressure, 215 ... Neutral fat, 216 ... HDL-C, 220 ... weight gain from age 20, 221 ... alcohol / week, 222 ... drinking days / week, 223 ... breakfast / Week, 224 ... number of eating out days / week, 225 ... meal amount, 226 ... meal speed, 227 ... meal balance, 228 ... smoking, 229 ... regular exercise, 230 ... sweat exercise, 231 ... determining glucose metabolism, 232 ... Blood pressure judgment, 233 ... Lipid judgment, 301 ... Item, 302 ... Classification, 303-306 ... Contribution rate to each disease and incidence of bad and good habits, 310-320 ... Contribution by disease of each item 401, item combination, 402 ... number of items, 403 ... combination contribution, 404 ... weighting combination contribution, 410 to 424 ... item combination and number of items / combination contribution / weighted combination contribution by combination, 501 ... Instruction item candidate display screen, 502 to 506 ... Classification radio button, 510 to 513 ... Number of items selection radio button, 514 ... Decision button, 520 to 521 ... Bar graph indicating the weighted combination contribution of item combinations, 522 to 523 ... Bar graph indicating combination contribution of item combinations, 530 to 531 ... Instruction item selection method selection radio button 601 to 604 ... guidance items for each disease, 702 ... item classification step, 703 ... disease contribution level calculation step, 704 ... number of items / combination contribution level calculation step, 705 ... guidance item candidate selection step, 706 ... guidance item Decision step, 707 ... Risk knowledge creation step, 801 ... Rule ID, 802 ... Maximum blood pressure / minimum blood pressure, 803 ... Neutral fat / HDL-C, 804 ... Number of applicable people, 805 ... Number of affected people, 806 ... Onset rate, 807 ... Support, 810 to 817 ... Rule, 901 ... Disease, 902 to 905 ... Disease, 906 ... Disease number, 907 ... Combination risk, 908 ... Weighted combination risk, 910 ... Rate of onset, 911 ... Number of onset patients, 912 … Number of applicable people, 913… Risk for people with good lifestyle habits, 914 to 919… Relative risk by disease of each guidance item, 1001… Checkup result input screen, 1002… Execute button, 1003 to 1015… Checkup item, 1020 ~ 1047 ... Condition input button, 1201 ... Instruction content display screen, 1210, 1220, 1230, 1240 ... Occurrence of each disease Proportion, 1211, 1221, 1231, 1241 ... Number of patients with each disease, 1212, 1222, 1232, 1242 ... Number of people with each disease, 1213, 1223, 1233, 1243 ... Risk for people with good lifestyles of each disease , 1214, 1224, 1234, 1244 ... Recommended improvement items for each disease, 1250 ... Most recommended improvement items, 1302 ... Checkup result input step, 1303 ... Rule search step, 1304 ... Relative risk calculation step, 1305 ... Number of diseases Combination risk calculation step, 1306: Instruction item extraction step, 1307 ... Instruction content display step, 1403 ... Medical checkup information, 1404 ... Contribution information registration, 1405 ... Contribution information, 1406 ... Instruction item registration, 1407 ... Instruction item, Health checkup information registration, 1408 ... Risk knowledge registration, 1409 ... Rule acquisition request, 1410 ... Rules, 1411 ... Instruction content registration, 1412 ... Instruction content, 1601 ... Instruction content display screen (bar graph version), 1610-1613 ... for each disease Bar graph showing the incidence before pressing the improvement item button, 162 0-1623… Risk for people with good lifestyles of each disease, 1630-1632, 1640-1642, 1650-1652, 1660-1663… Recommended improvement item buttons for each disease, 1670… Most recommended improvement item buttons, 1710 ~ 1713, 1810-1813 ... bar graph showing the onset rate after pressing the improvement item button of each disease, 1901 ... instruction content display screen (pie graph version), 1910-1913 ... the onset rate before pressing the improvement item button of each disease Pie chart showing, 2010-2013, 2110-2113 ... Pie chart showing onset rate after pressing improvement item button of each disease.

Claims (8)

An individual health guidance support system that presents information for individual prevention and health promotion related to one or more illnesses from medical examination information obtained by accumulating medical examination results including multiple lifestyle items and examination items for multiple persons. ,
Item classification means for generating a lifestyle item classification including a plurality of lifestyle items classified for each lifestyle type to be instructed;
Analyzing the relationship between the lifestyle items and the onset of illness, calculating the onset rate indicating the proportion of people who have developed the value according to the value of the lifestyle item, and calculating the above-mentioned lifestyle habit with a high calculated incidence rate The value of the item is a value with a bad lifestyle, the value of the lifestyle item with a low calculated incidence rate is a value with a good lifestyle, the incidence rate of the value with a good lifestyle and the incidence rate with a value with a bad lifestyle A disease-specific contribution calculation means for calculating a disease-specific contribution from the disease,
For each lifestyle item classification generated by the item classification means, a combination of lifestyle items included in the lifestyle item classification is generated, and calculated by the disease-specific contribution calculating means for the lifestyle item included in the generated combination. The number of items / combination contribution that calculates the weighted combination contribution that is the sum of the contributions weighted by the incidence of bad lifestyles and the number of items that are the number of lifestyle items included in the generated combination A degree calculating means;
For each lifestyle item classification generated by the item classification means, the life with the smallest number of items and the largest weighted combination contribution is calculated from the number of items and the weighted combination contribution calculated by the number of items / combination contribution calculating means. An individual health guidance support system comprising guidance item candidate selection means for selecting a combination of habit items as a guidance item. The individual health guidance support system according to claim 1, wherein each of the combinations of lifestyle items generated by the number of items / combination contribution calculating means is calculated by the disease-specific contribution calculating means for the lifestyle items included in the combination. Means for calculating a combination contribution that is the sum of the contributions
For each lifestyle item classification generated by the item classification means, the lifestyle item having the smallest number of items and the largest combination contribution from the combination contribution and the number of items calculated by the number of items / combination contribution calculation means A means of selecting a combination of
The selected combination of lifestyle items and the guidance item selected by the guidance item candidate selection means are compared, and if they are different, the combination is displayed as a guidance item candidate according to the lifestyle item classification generated by the item classification means. An individual health guidance support system characterized by doing.   The individual health guidance support system according to claim 1 or 2, wherein a condition that combines the values of the guidance item and the test item and a rate of onset that indicates a percentage of people who develop the condition and the number of people who develop the disease An individual health guidance support system characterized by having rule creation means for calculating the degree of support indicating the reliability of the onset rate for each disease as a rule, and creating risk knowledge for each disease. In the individual health guidance support system according to claim 3, a rule retrieval means for retrieving the rule indicating the onset rate of each disease corresponding to the health check result of the guidance subject from the risk knowledge of each disease,
A risk for a person with good lifestyle showing a relative risk with respect to a rule in which all of the lifestyle items in the search rule are changed to a good value for the lifestyle item, and the lifestyle for each person with a good value for the lifestyle item A relative risk calculation means for calculating the relative risk for each item indicating the relative risk with respect to the rule whose item is changed to a good value;
Recommended combination risk that indicates the sum of the relative risk of each item for each guidance item, the number of illnesses / combination risk calculation means for calculating the number of diseases for which the relative risk for each item is calculated, and items for which the relative risk for each item exists Individual health guidance support characterized by having an improvement item extracting means for extracting as an improvement item and extracting from the disease number / combination risk the item having the maximum number of diseases and the maximum combination risk as the most recommended improvement item system.   5. The individual health guidance support system according to claim 4, wherein the number of diseases / combination risk calculating means is weighted by the onset rate of each disease in the rule corresponding to the health check result of the instructed subject for each guidance item. Calculating a weighted combination risk indicating the sum of relative risks and the number of diseases for which the item-specific relative risk is calculated, and the improvement item extracting means extracts an item having the item-specific relative risk as a recommended improvement item; An individual health guidance support system, wherein an item having the maximum number of diseases and the maximum weighted combination risk is extracted as the most recommended improvement item from the number of diseases / weighted combination risk.   6. The individual health guidance support system according to claim 4 or 5, wherein the guidance content is created by displaying a list of the onset rate of each disease of the target person, risks for people with good lifestyle habits, recommended improvement items, and most recommended improvement items. An individual health guidance support system characterized by having means.   6. The individual health guidance support system according to claim 4 or 5, wherein the onset rate of each disease of the target person and the change in the onset rate when the recommended improvement item or the most recommended improvement item is a good value in a bar graph. An individual health guidance support system characterized by having guidance content creation means for displaying.   6. The individual health guidance support system according to claim 4, wherein the onset rate of each disease of the target person and changes in the onset rate when the recommended improvement item and the most recommended improvement item are good values. An individual health guidance support system characterized by having a guidance content creation means for displaying on the screen.

Images