Precise recommendation system and method for health food and non-instantaneous computer readable storage medium

Abstract

A health food accurate recommendation system and method and a non-instantaneous computer readable storage medium, the system including a database, a score window element, a disease risk unit, an arithmetic unit and a risk management recommendation unit, the method can utilize a plurality of data of clinical trials of a plurality of health food for different diseases, calculating a plurality of scores of the health food for different diseases to obtain a plurality of score tables, then obtaining different disease risk values of congenital genes of the user and different disease risk values of the current body of the user, multiplying the different disease risk values by the score tables, and standardizing the values; the risk management recommendation unit obtains benefit values of the health food for innate gene risks of the user and benefit values of current body risks, the benefit values are combined to generate a judgment module for the user to accurately recommend the health food, and the storage medium can perform operation on the disease risk unit, the operation unit and the risk management recommendation unit to generate the judgment module.

Description

Technical Field

[0001] The present invention relates to health food, and more particularly to a health food accurate recommendation system, method and non-transient computer-readable storage medium. Background Art

[0002] Although there are many health foods available for metabolic disease care, they are mostly used to provide patients with supportive health effects after the onset of the disease, and relatively few provide preemptive health care. However, it is often too late to provide health foods to patients when they actually become ill. Secondly, given the increasing maturity of gene identification technology, it has contributed to the development of modern preventive medicine. Gene identification testing can quantitatively analyze the innate constitution of each subject by comparing statistical data. In addition, another cause of disease is acquired lifestyle habits. Therefore, if the innate and acquired information can be combined, the risks can be known in advance before health problems occur, and diseases can be prevented.

[0003] Medical science has known that probiotics can prevent, regulate, and even treat one or more diseases. If everyone knows what diseases they may have in the future through genetic testing, they can adjust their bodies in advance or immediately. However, there are hundreds of probiotics on the market, and many people have tried them again and again but the results are not good. If probiotics are recommended based on the results of genetic testing and the current situation, it is like going to the hospital to see a doctor and prescribing the right medicine for the right condition. It can provide quick and accurate advice, saving users time and money in trying.

[0004] Secondly, although there are currently technologies that combine genetic testing with recommending health foods, such as Taiwan Patent No. I680468 and Publication No. 201928992, No. 202026426, and Chinese Invention Patents CN 114446482A, most of them only use genetic testing to predict possible future lesions and then find suitable health foods to recommend, and there is no technology that combines current physical condition information to recommend health foods. Summary of the invention

[0005] The main purpose of the present invention is to provide a health food accurate recommendation system, method and non-transient computer readable storage medium, which can accurately recommend health food by combining gene test results with current physical condition information, saving users time and money in trying, and is very practical.

[0006] To achieve the aforementioned purpose, the present invention provides a health food precision recommendation system for users who may suffer from predetermined diseases, comprising a database comprising a plurality of data on clinical trials of a plurality of health foods for different diseases; a score form element connected to the database for calculating a plurality of scores of each health food for different diseases, wherein the scoring method of each score is to give different scores at different stages of the clinical trial, and to accumulate the number of international journals that disclose the efficacy of each health food for different diseases; a disease risk unit connected to the score form element, comprising an objective risk value module and a subjective risk value module, wherein the objective risk value module obtains different diseases of the user's congenital genes through gene detection technology. Risk value, the subjective risk value module obtains the different disease risk values ​​of the user's current body through testing or questionnaires; a calculation unit, connecting the score form element and the disease risk unit, can multiply the different disease risk values ​​of the objective risk value module and the subjective risk value module with each score table respectively, and perform numerical standardization to obtain the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk; a risk management recommendation unit, connected to the calculation unit, combines the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk to generate a judgment module, which uses the judgment module to provide users with accurate health food recommendations.

[0007] Preferably, the probiotics in each health food, and the data include probiotics-related data from the U.S. National Institutes of Health clinical trial database and international journals on the benefits of probiotics to the human body.

[0008] Preferably, each of the health foods comprises a composite health food of several nutrients, and each of the data comprises relevant data of each nutrient from the clinical trial database of the National Institutes of Health of the United States and international journals on the benefits of each nutrient to the human body.

[0009] Preferably, the score table is a table corresponding to different types of health foods and diseases, and the scoring method for each score is as follows: clinical trials are divided into four phases, and the scores decrease from the fourth phase to the first phase. The scores of international journals are given the same score for each article, and the scores of multiple international journals reporting the same effect of the predetermined health food are accumulated.

[0010] Preferably, the objective risk value module converts different disease items detected by genetic testing into risk values, and the subjective risk value module converts different disease items of the user's current body into risk values ​​through a questionnaire.

[0011] Preferably, after the risk values ​​of the objective risk value module and the subjective risk value module are multiplied by each score table respectively, the value is standardized in a way that the multiplication operation values ​​of different health foods for different diseases are summed up and then divided by the risk values ​​of different diseases, and multiplied by 100%, so as to obtain the benefit values ​​of different health foods for users.

[0012] Preferably, the judgment module is a risk management matrix.

[0013] In addition, the present invention further provides a method for accurately recommending health foods using an accurate health food recommendation system, the steps of which at least include: confirming the disease: using genetic testing technology to learn about the diseases that the user may develop in the future, and obtaining the different diseases of the user's current body through testing or questionnaires; extracting data from a database: collecting data from clinical trials of various health foods for different diseases, and international journals on the benefits of various health foods to the human body from the database; establishing a score table: creating a number of score tables using the data collected from the database, the score table containing multiple scores of each health food for different diseases, and the scoring method of each score is based on different scores given at different stages of the clinical trial. The same score, and the accumulation of the number of international journals that reveal the efficacy of each health food for different diseases; obtain disease risk value: obtain the different disease risk values ​​of the user's congenital genes and current body through genetic testing technology and testing or questionnaires; obtain benefit value: multiply the different disease risk values ​​of the user's congenital genes and current body with each score table respectively, and standardize the values ​​to obtain the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk; accurate recommendation: combine the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk to accurately recommend health foods with better benefit values.

[0014] Preferably, in the step of obtaining the benefit value, the risk values ​​of different diseases are multiplied by the scores of the health food fields of the corresponding diseases on the score table, and then the values ​​are standardized.

[0015] In addition, the present invention also provides a non-transient computer-readable storage medium that stores multiple instructions, which are executed by a processing unit of a computer device to operate on the disease risk unit, operation unit and risk management recommendation unit of the health food precision recommendation system to generate precise recommendations for health foods with better benefit values. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a system diagram of a preferred embodiment of the present invention.

[0017] Figure 2 It is a disease risk value map obtained by calculation of the objective risk value module in a preferred embodiment of the system of the present invention.

[0018] Figure 3 It is the benefit value of various probiotic strains obtained by calculation by the calculation unit in the system of a preferred embodiment of the present invention.

[0019] Figure 4 It is a judgment module diagram obtained by calculation of the risk management recommendation unit in a preferred embodiment system of the present invention.

[0020] Figure 5 It is a flow chart of a method of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0021] Hereinafter, several preferred embodiments of the present invention are given, and further detailed description is given with reference to the accompanying drawings as follows:

[0022] First, see Figure 1 As shown, a health food accurate recommendation system 10 according to a preferred embodiment of the present invention comprises a database 12, a score form unit 14, a disease risk unit 16, a calculation unit 18 and a risk management recommendation unit 19.

[0023] The database 12 includes a plurality of data on clinical trials of a plurality of health foods for different diseases. Each of the health foods at least includes probiotics. The health foods in this embodiment take probiotics as an example. Each of the data includes probiotic-related data from the U.S. National Institutes of Health Clinical Trials database and international journals on the benefits of probiotics to the human body (such as various medical-related international journals), but is not limited to the aforementioned U.S. National Institutes of Health Clinical Trials database and medical international journals. Each of the data can be downloaded in advance to a storage unit of a computer device (not shown in the figure).

[0024] The score form element 14 is built in the computer device and connected to the database 12. It takes the data about the probiotics on different diseases in the database 12, integrates and calculates the scores to form a score table. The scoring method of each score is to give different scores according to different stages of clinical trials, and to accumulate the number of international journals that reveal that each health food has the effect on different diseases. In detail, clinical trials in medicine are divided into four phases, and the scores decrease from the fourth phase to the first phase. For example, the score of the fourth phase is 0.8 points, the score of the third phase is 0.6 points, the score of the second phase is 0.4 points, and the score of the first phase is 0.2 points. The scores of international journals are given the same score for each article, for example, 0.1 points for one article. The scores of multiple international journals reporting the same effect of the predetermined health food are accumulated, and the maximum score is 1 point. The scores of each data are integrated and calculated to form a score table, as shown in Table 1, and stored in the computer device.

[0025] Table 1:

[0026] Bacteria name / disease type Cardiovascular skin liver Stomach Intestinal Bacillus coagulans 0 0.5 0 0.4 0.3 Bifidobacterium longum 0 0 0 0.7 0.2 Lactobacillus acidophilus 0 0 0 0.8 0 Lactobacillus brevis 0.2 0 0.1 0 0 Lactobacillus bulgaricus 0 0 0 0 0.8 Lactobacillus casei 0 0.6 0.2 0.2 0.1 Lactobacillus helveticus 0.7 0 1 0.5 0.5

[0027] In addition, most probiotics on the market are in powder form, usually composed of two or more probiotics and compound ingredients. In this case, we can first analyze the probiotic types, content and amount of live bacteria in various probiotic powders, and then integrate them with Table 1 to generate a new score table for different probiotic powders for different diseases (each probiotic is based on 1 billion active probiotics, which refers to the definition of daily effective probiotic intake by the US NIH and many international journals. If it exceeds 1 billion CFU, it is calculated as 1 unit, and if it is below 1 billion CFU, it is calculated as a multiple. Prebiotics and compounds are based on the efficacy reported in international journals + 0.1 points).

[0028] Of course, if multiple health foods are used, multiple score tables can be prepared.

[0029] The disease risk unit 16 is connected to the score form unit 14, and includes an objective risk value module 22 and a subjective risk value module 24. The objective risk value module 22 obtains different disease items that the user's congenital genes have the possibility of suffering from through conventional gene detection technology, and then converts the different disease items into risk values. The subjective risk value module 24 obtains different disease items of the user's current body through detection or questionnaire, and then converts the different disease items of the current body into risk values.

[0030] In detail, the gene detection technology used by the objective risk value module 22 is mainly predictive gene detection, which uses gene chips as detection. The gene detection risk determination method and calculation method are shown in Taiwan Invention No. I764817. Generally, the PRS value of each gene detection item of the user is obtained, and the standard deviation and average value of the item (PRS calculated using Taiwanese genetic data, Taiwanese genetic data refer to the Chinese cell line and gene database in Taiwan) are obtained. These values ​​are used using an inverse function (normal distribution back-testing) to obtain the risk value of each disease, with the highest being 100%. The closer the value is to 100%, the more serious it is, such as Figure 2 As shown. The detection method of the subjective risk value module 24 can be a wearable device or a physical examination, but it is not limited to the questionnaire method. The questionnaire method uses the conventional hierarchical analysis method (Analytic Hierarchy Process, AHP). The items of the questionnaire (disease items) are the same as the genetic detection technology items used by the objective risk value module 22. The details of the questionnaire method (for example, the items are divided into four levels A, B, C, D for comparison of importance, the questionnaire results are converted into a matrix table, the matrix calculates the risk value, etc.) are not repeated here, so that the risk value of each disease can be obtained.

[0031] The calculation unit 18, which connects the score form unit 14 and the disease risk unit 16, can multiply the different disease risk values ​​of the objective risk value module 22 and the subjective risk value module 24 with the score table respectively, and perform numerical standardization, sum up the multiplication operation values ​​of various probiotics for different diseases respectively, divide them by the risk values ​​of different diseases, and multiply them by 100%, and then obtain the benefit values ​​of various probiotics for the user's innate genetic risk and the benefit values ​​of the current physical risk. In detail, the objective risk value module 22 part: for example, the genetic test risk value of a certain user is shown in Table 2, taking the cardiovascular project as an example, the cardiovascular risk value is multiplied by the value of each cardiovascular strain field in Table 1, and after multiplication, a multiplied value table will be obtained, such as Table 3, and numerical standardization is performed to obtain the comprehensive benefit of each probiotic strain (that is, the value of the same strain is added to become the sum of the benefit of the strain), and finally divided by the total, and multiplied by 100%, and then the benefit values ​​of various probiotics for the user's innate genetic cardiovascular risk can be obtained.

[0032] Table 2:

[0033] Cardiovascular skin liver Stomach Intestinal 90％ 15％ 20％ 6％ 70％

[0034] Table 3:

[0035]

[0036]

[0037] For example, the benefit of Bacillus coagulans for this user is the sum divided by the total disease risk value: 30.9÷(90+15+20+6+70)X100%=15.37%. This is the benefit value of this probiotic for this user. Figure 3 .

[0038] In addition, when probiotics are in powder form (the same applies to comprehensive health foods), the calculation is the same.

[0039] Secondly, the subjective risk value module 24 is the same as above. After obtaining the risk value of each disease, multiply it with the score table respectively, and standardize the value. The multiplication values ​​of various probiotics for different diseases are summed up and divided by the risk values ​​of different diseases, and multiplied by 100%. The benefit value of various probiotics for the user's current physical risk can be obtained, which can produce an approximate Figure 3 Graph (not shown).

[0040] The risk management recommendation unit 19 is connected to the operation unit 18, and is a judgment module that combines the benefit value of various probiotics for the user's innate genetic risk and the benefit value of the current physical risk. As shown in the fourth figure, the judgment module is a risk management matrix, that is, a judgment module that integrates the possible future (innate gene) risks and the current diseases that have occurred, so that users can simply and clearly choose accurate probiotics and obtain the effect of accurately recommending probiotics. Of course, each health food is not limited to probiotics.

[0041] Based on this, modern medicine has confirmed that probiotics can prevent, regulate and even treat one to multiple diseases. The present invention uses the gene detection of the objective risk value module 22 and the detection and questionnaire of the subjective risk value module 24, as well as the many-to-many calculation logic and recommendation of the operation unit 18, so that the risk management recommendation unit 19 can combine subjective benefits and objective benefits to provide users with accurate probiotic solutions, just like the concept of going to the hospital to see a doctor and prescribing the right medicine for the disease. It can quickly and accurately recommend probiotics, saving users time and money in trying, and is very practical.

[0042] In addition, Figure 5 As shown, the present invention also provides a method 100 for accurately recommending health foods, wherein the first step is to confirm the disease 110: the genetic testing technology of the objective risk value module 22 is used to learn about the diseases that the user may develop in the future, and the different diseases of the user's current body are obtained through the testing or questionnaire method of the subjective risk value module 24.

[0043] The second step of the present invention is to extract data from the database 120: collect clinical trial data of various health foods for different diseases and international journals on the benefits of various health foods to the human body from the database 12. As mentioned above, the present invention takes probiotics as an example, but is not limited.

[0044] The third step of the present invention is to establish a score table 130: the data collected from the database 12 is used to create a score table such as Table 1, which includes multiple scores of various probiotics for different diseases. The scoring method of each score is as described above, which is to give different scores at different stages of clinical trials and accumulate the number of international journals that reveal that each health food has the efficacy of different diseases.

[0045] The fourth step of the present invention is to obtain the disease risk value 140: through the gene detection technology of the objective risk value module 22, and the detection or questionnaire method of the subjective risk value module 24, the average value and standard deviation of different disease items of the user's congenital genes and current body can be obtained, and the inverse function calculation of the average value and standard deviation can be used to obtain the different disease risk values ​​of the user's congenital genes and current body, such as Figure 2 shown.

[0046] The fifth step of the present invention is to obtain the benefit value 150: the user's innate genes and the different disease risk values ​​of the current body are multiplied by the health food field scores of the corresponding diseases on the score table, and the values ​​are standardized to obtain the benefit values ​​of various probiotics for the user's innate genetic risks and the benefit values ​​of the current physical risks, as shown in the third figure.

[0047] The final step of the present invention is accurate recommendation 160: combining the benefit values ​​of various probiotics for the user's innate genetic risk and the benefit values ​​of the current physical risk, the judgment module, i.e., the risk management matrix, can be obtained, such as Figure 4 As shown, probiotics with better user benefit values ​​can be accurately recommended.

[0048] Furthermore, as mentioned above, the health food disclosed in the present invention is not limited to probiotics, but can also be a composite health food composed of several nutrients (such as tablet-shaped health food), and the composite health food is more accurately recommended:

[0049] For example, the health food precision recommendation system 10 of the present invention can obtain data on the relationship between nutrients such as magnesium and vitamins and diseases from a database 12 (such as the clinical trial database of the National Institutes of Health in the United States). The score form element 14 can integrate the data on the relationship between nutrients such as magnesium and vitamins and diseases to calculate the score (many-to-many operation). The scoring method of each score is as described above. For example, clinical trials in medicine are divided into four phases, the score of the fourth phase is 4 points, the score of the third phase is 3 points, the score of the second phase is 2 points, and the score of the first phase is 1 point. Each international journal reporting the efficacy of nutrients such as magnesium and vitamins is 0.1 points. If multiple journals report the same effect, the score is accumulated, with a maximum of 1 point, and a score table as shown in Table 4 can be obtained.

[0050] Table 4:

[0051] Category / Health Products Thromboembolism High blood pressure Type 2 diabetes Hyperthyroidism obesity ... Vitamin A 2 5 3 7 2 ... Vitamin B6 2 1 0 2 3 ... Vitamin C 0 5 1 3 6 ... Vitamin B12 2 2 5 4 0 ... Vitamin K 2 3 5 0.5 2 ... Omega-3 (EPA) 3 2 2 2 0 ... Omega-3 (DPA) 5 2 1 0 4 ... Omega-3 (DHA) 4 3 2 0 3 ... Zinc 1 3 7 4 5 ... Iron 5 4 3 2 4 ... Calcium 1 5 4 1 2.5 ... Magnesium 0 5 2 1 4 ... ... ... ... ... ... ... ...

[0052] Based on this, the calculation unit 18 of the present invention can multiply the above-mentioned score table with the different disease risk values ​​of the objective risk value module 22 and the subjective risk value module 24 respectively, to obtain the benefit values ​​of various nutrients such as magnesium and vitamins for the user's congenital genetic risk and the benefit values ​​of the current physical risk. The risk management recommendation unit 19 can also generate a judgment module (not shown in the figure) through each benefit value, so that the user can simply and clearly choose nutrients such as magnesium and vitamins, thereby achieving the effect of accurately recommending compound health foods.

[0053] In addition, the present invention further provides a non-transitory computer readable storage medium 20, such as Figure 1As shown, multiple instructions are stored, and the multiple instructions are for execution by a processing unit of the aforementioned computer device to operate the disease risk unit 16, the operation unit 18 and the risk management recommendation unit 19. The executed instructions (such as program codes, as shown in Appendix 1) can generate the judgment module to obtain accurate recommendations for health foods with better benefit values.

[0054] The above descriptions are merely some preferred embodiments of the present invention and are not intended to limit the present invention. Any person having ordinary knowledge in the technical field may make some modifications, changes and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the appended claims.

Claims

1. A health food accurate recommendation system, characterized in that: Provided to users who may suffer from predetermined diseases, including: a database comprising a plurality of data on clinical trials of a plurality of health foods for different diseases; a score form element, connected to the database, for calculating the multiple scores of each health food for different diseases and establishing a plurality of score tables, wherein the scoring method of each score is to give different scores at different stages of clinical trials and to accumulate the number of international journals that disclose the efficacy of each health food for different diseases; a disease risk unit, connected to the score form element, comprising an objective risk value module and a subjective risk value module, wherein the objective risk value module obtains different disease risk values ​​of the user's congenital genes through gene detection technology, and the subjective risk value module obtains different disease risk values ​​of the user's current body through detection or questionnaire; a calculation unit, connected to the score form unit and the disease risk unit, for multiplying the different disease risk values ​​of the objective risk value module and the subjective risk value module with each score table respectively, and performing numerical standardization to obtain the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk; and A risk management recommendation unit is connected to the operation unit, and generates a judgment module by combining the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk, so as to provide the user with accurate health food recommendations using the judgment module.

2. The health food accurate recommendation system according to claim 1, characterized in that: The probiotics in various health foods include probiotics-related data from the U.S. National Institutes of Health Clinical Trials Database (ClinicalTrials) and international journals on the benefits of probiotics to the human body.

3. The health food accurate recommendation system according to claim 1, characterized in that: Each of the health foods is a composite health food containing several nutrients, and each of the data includes relevant data of each nutrient from the clinical trial database of the National Institutes of Health of the United States and international journals on the benefits of each nutrient to the human body.

4. The health food accurate recommendation system according to claim 1, characterized in that: The score table is a table that corresponds to different types of health foods and diseases. The scoring method for each score is as follows: clinical trials are divided into four phases, and the scores decrease from the fourth phase to the first phase. The score of international journals is given the same score for each article. If multiple international journals report the same effect of the predetermined health food, the scores are accumulated.

5. The health food accurate recommendation system according to claim 4, characterized in that: The objective risk value module converts different disease items of genetic testing into risk values ​​(up to 100%), and the subjective risk value module converts different disease items of the user's current body into risk values ​​(up to 100%) through a questionnaire.

6. The health food accurate recommendation system according to claim 5, characterized in that: After the risk values ​​of the objective risk value module and the subjective risk value module are multiplied by each score table respectively, the value is standardized in a way that the multiplication values ​​of different health foods for different diseases are summed up and then divided by the risk values ​​of different diseases, and multiplied by 100%, so as to obtain the benefit values ​​of different health foods for users.

7. The health food accurate recommendation system according to claim 5, characterized in that: The judgment module is the risk management matrix.

8. A method for accurately recommending health foods using the accurate health food recommendation system according to any one of claims 1 to 7, characterized in that: The steps include at least: Confirming diseases: Using genetic testing technology to learn about the diseases that the user may have in the future, and using tests or questionnaires to learn about the different diseases that the user currently has; Data extraction from database: Data on clinical trials of various health foods for different diseases and international journals on the benefits of various health foods to the human body are collected from the database; Establishing score tables: The data collected from the database are used to create a number of score tables, which contain multiple scores of each health food for different diseases. The scoring method for each score is to give different scores at different stages of clinical trials and to accumulate the number of international journals that reveal that each health food has the efficacy for different diseases; Obtain disease risk values: Obtain the different disease risk values ​​of the user's innate genes and current body through genetic testing technology and testing or questionnaires; Obtaining benefit value: multiplying the user's innate gene and current disease risk values ​​of the body with each score table, and normalizing the values ​​to obtain the benefit value of each health food for the user's innate gene risk and current physical risk; Accurate recommendations: Combine the benefit value of each health food for the user's congenital genetic risk and the benefit value of the current physical risk to accurately recommend health foods with better benefit values.

9. The method for accurately recommending health food according to claim 7, characterized in that: In the step of obtaining the benefit value, the risk values ​​of different diseases are multiplied by the scores of the health food fields of the corresponding diseases on the score table, and then the values ​​are standardized.

10. A non-transitory computer-readable storage medium, characterized in that: A plurality of instructions are stored, and the plurality of instructions are executed by a processing unit of a computer device to operate the disease risk unit, the operation unit and the risk management recommendation unit of the health food precision recommendation system of claim 1 to generate a health food precision recommendation with a better benefit value.

Images