TWI893445B - Intelligent nutrient assessment system and method - Google Patents

Abstract

This invention provides an intelligent nutrient evaluation system and method. Based on academic literature, this invention deduces the nutrient weighting values between negative factors and specific nutrients, and establishes negative factor weighting values through a negative factor grading table. Therefore, the invention can provide customized nutrient supplement recommendations based on nutrient weighting values and negative factor weighting values, according to the subject's data (such as a health questionnaire).

Description

Intelligent nutrient assessment system and method

The present invention relates to the technical field of functional medicine, and more particularly to a method of combining non-invasive examinations with computer algorithms to provide objective nutritional assessment information to subjects.

Due to the high stress of modern life, more and more people are suffering from various illnesses. Even with the advancement of medicine today, it remains difficult to escape the burden of illness. Furthermore, within the current medical system, doctors often prescribe a specific condition (such as upper respiratory tract inflammation, asthma, or allergic rhinitis) and then prescribe appropriate medication. However, some medications merely alleviate symptoms, while most rely on the body's own immune system for recovery. Patients with these conditions often lack the ability to improve their overall health, beyond simply following their doctor's instructions and maintaining a healthy lifestyle.

Therefore, various health foods are marketed, or nutritionists provide nutritional information for various foods, hoping that people can supplement their nutrition and improve their health. However, this leads to people consuming large amounts of nutrients without any basis, which in turn leads to additional side effects (for example, consuming large amounts of health foods may increase the burden on liver and kidney metabolism).

Furthermore, many people have long-term bad habits (such as staying up late, eating a high-fat diet) or mild symptoms (such as night cramps, migraines, and frequent colds). Even if they go to the doctor, they may only get some medication to relieve the symptoms.

In view of this, and based on the above-mentioned issues, there is currently no non-invasive, objective, customized, and convenient nutritional assessment method, making it an unresolved problem in the relevant technical field.

The present invention provides a smart nutrient assessment system, signal-connected to at least one literature database. The smart nutrient assessment system includes a literature retrieval module, a literature weighting module, a negative factor weighting module, and a processing module. The literature retrieval module is signal-connected to the at least one literature database. The literature retrieval module uses a search command composed of keywords containing multiple negative factors and multiple nutrients to obtain from the at least one literature database multiple literature reports that simultaneously record at least one of the multiple negative factors and at least one of the multiple nutrients. The multiple negative factors include diseases and unhealthy lifestyle habits. The literature weighting module is signal-connected to the literature retrieval module. The literature weighting module classifies the plurality of literature reports using a literature rating table and, based on the literature ratings corresponding to the plurality of literature reports, sets nutrient weighting values corresponding to the literature ratings for the nutrients recorded in the plurality of literature reports, wherein the nutrient weighting value corresponds to one of the plurality of negative factors. The processing module is signal-connected to the literature weighting module. The processing module selects and sums the nutrient weighted values of each nutrient associated with each of the multiple negative factors. The nutrient weighted values are then arithmetic averaged based on the number of corresponding literature reports to generate an average nutrient weighted value corresponding to each of the multiple negative factors.

The processing module is signal-connected to the execution module, which is signal-connected to the user device for receiving subject information corresponding to the subject provided by the user device. The subject information records at least one of a plurality of negative factors. The processing module extracts the average weighted value of the nutrient corresponding to each negative factor recorded in the subject information as the first customized nutrient supplement recommendation value.

The literature weighting module uses natural language processing (NLP) to determine the nutritional benefits of each nutrient reported in multiple literature reports for negative factors. It then assigns a corresponding literature-level weighted sum of the positive nutrient weights, the negative nutrient weights, and the ineffective nutrient weights for each nutrient reported in the multiple literature reports. The module then arithmetic averages these sums based on the number of corresponding literature reports to generate the average positive nutrient weights, the average negative nutrient weights, and the average ineffective nutrient weights for each negative factor.

The processing module is signal-connected to the execution module, which is signal-connected to the user device for receiving subject information corresponding to the subject provided by the user device. The subject information records at least one of a plurality of negative factors. The processing module extracts the average weighted value of positive nutrients, the average weighted value of negative nutrients, and the average weighted value of ineffective nutrients corresponding to each negative factor recorded in the subject information as a second customized nutrient supplement recommendation value.

The processing module also divides the average weighted value of negative nutrients and the average weighted value of ineffective nutrients by the average weighted value of positive nutrients to generate a nutrient negative benefit ratio and a nutrient ineffective benefit ratio, respectively, as the third customized nutrient supplement recommendation value.

The processing module is further connected to a negative factor weighting module. The negative factor weighting module provides a corresponding negative factor weighting value for each of the multiple negative factors based on a negative factor grading table, so that the processing module multiplies the average nutrient weighting value corresponding to each of the multiple negative factors by the negative factor weighting value to generate a nutrient composite weighting value corresponding to each of the multiple negative factors.

The processing module is further signal-connected to the execution module, which is signal-connected to the user device for receiving subject information corresponding to the subject provided by the user device. The subject information records at least one of a plurality of negative factors. The processing module extracts the nutrient complex weight value corresponding to each negative factor recorded in the subject information as a fourth customized nutrient supplement recommendation value.

The processing module is further connected to the negative factor weighting module. The negative factor weighting module provides corresponding negative factor weighting values for each of the multiple negative factors based on the negative factor grading table, so that the processing module multiplies the average weighted value of positive nutrients, the average weighted value of negative nutrients, and the average weighted value of ineffective nutrients corresponding to each of the multiple negative factors by the negative factor weighting values, respectively, to generate the positive nutrient composite weighting value, the negative nutrient composite weighting value, and the ineffective nutrient composite weighting value corresponding to each of the multiple negative factors.

The processing module is further signal-connected to the execution module, which is signal-connected to the user device for receiving subject information corresponding to the subject provided by the user device. The subject information records at least one of a plurality of negative factors. The processing module extracts the positive nutrient composite weight, negative nutrient composite weight, and ineffective nutrient composite weight corresponding to each negative factor recorded in the subject information as the fifth customized nutrient supplement recommendation value.

The processing module also divides the negative nutrient complex weights and the ineffective nutrient complex weights by the positive nutrient complex weights to generate a complex nutrient negative benefit ratio and a complex nutrient ineffectiveness ratio, respectively, as the sixth customized nutrient supplement recommendation value.

The present invention further provides a smart nutrient evaluation method, comprising the following steps: obtaining a plurality of literature reports that simultaneously record at least one of the plurality of negative factors and at least one of the plurality of nutrients through a search instruction consisting of a plurality of negative factors and a plurality of nutrients; grading the plurality of literature reports through a literature grading table; and grading the plurality of literature reports according to the literature grades corresponding to the plurality of literature reports. Nutrient weighting values corresponding to the literature level were set for each nutrient reported in multiple literature reports, where the nutrient weighting value corresponded to one of the multiple negative factors. The nutrient weighting values of each nutrient associated with each multiple negative factors were screened and summed up, and then arithmetic averaged based on the number of corresponding nutrient weighting values in the multiple literature reports was performed to generate the average nutrient weighting value corresponding to each multiple negative factors.

When obtaining the subject information, it also includes: extracting the average weighted value of nutrients corresponding to each negative factor recorded in the subject information.

The step of setting nutrient weighting values corresponding to the literature levels for the nutrients recorded in the plurality of literature reports includes: determining the nutritional benefits of at least one of the plurality of nutrients recorded in the plurality of literature reports relative to the corresponding negative factors, and splitting the nutrient weighting values into positive nutrient weighting values, negative nutrient weighting values, and ineffective nutrient weighting values.

The smart nutrient assessment method further includes the following steps: multiplying the average weighted value of the nutrients corresponding to each of the multiple negative factors by the weighted value of the negative factors to obtain the composite weighted value of the nutrients corresponding to each of the multiple negative factors.

The step of multiplying the average weighted value of nutrients corresponding to each of the plurality of negative factors by the weighted value of the negative factors to obtain the composite weighted value of nutrients corresponding to each of the plurality of negative factors further includes: splitting the average weighted value of nutrients into an average weighted value of positive nutrients, an average weighted value of negative nutrients, and an average weighted value of ineffective nutrients, wherein the average weighted value of positive nutrients, the average weighted value of negative nutrients, and the average weighted value of ineffective nutrients corresponding to each of the plurality of negative factors are respectively multiplied by the weighted value of the negative factors to obtain the composite weighted value of positive nutrients, the composite weighted value of negative nutrients, and the composite weighted value of ineffective nutrients corresponding to each of the plurality of negative factors.

The smart nutrient assessment method further includes the following steps: dividing the negative nutrient composite weights and the ineffective nutrient composite weights by the positive nutrient composite weights, respectively, to generate a nutrient negative benefit ratio and a nutrient ineffective benefit ratio, respectively.

Compared with the prior art, the present invention can claim the following effects.

(1) Providing intelligent nutritional assessment: The present invention can analyze various negative factors (such as diseases or bad living habits) to understand their relationship with nutrients.

(2) Ensuring the reference value of literature reports: This invention, in the spirit of evidence-based medicine (EBM), uses a credible (evidence-based) literature classification method to assign a corresponding level to each literature report. Correspondingly, the same nutrient weighting value is given to each nutrient reported in the literature report based on the level of each literature report.

(3) Providing objective professional advice: The risk levels of various negative factors on the human body vary, but the public is often unable to distinguish the severity of negative factors. Based on this, and in order to eliminate the different opinions of individual experts (e.g., doctors, nutritionists) on the risk level of negative factors, the present invention can use the Delphi method to conduct multiple rounds of questionnaire assessments to make the experts' opinions on the risk level of negative factors on the human body more unified, thereby quantifying it as the negative factor weighted value of each negative factor.

(4) Customized nutrient supplementation recommendations: Based on the nutrient weights and negative factor weights, the nutrients corresponding to the negative factors reported in each literature report can be weighted (i.e., the nutrient weights multiplied by the negative factor weights) to obtain the various nutrient supplementation recommendations corresponding to the various negative factors. Therefore, by evaluating the negative factors recorded in the subject information through the present invention, the corresponding customized nutrient supplementation recommendations can be generated through the present system or method.

The following text provides a detailed description of the present invention through specific embodiments and accompanying drawings, which will make it easier to understand the purpose, technical content, features and effects achieved by the present invention.

The embodiments of the present invention are further explained below with reference to the accompanying drawings. Whenever possible, identical reference numerals in the drawings and the specification represent identical or similar components. It should be understood that components not specifically shown in the drawings or described in the specification are of a type known to those skilled in the art. Those skilled in the art may make various changes and modifications based on the teachings of the present invention.

First embodiment

As shown in Figure 1, Figure 1 is a flow chart of the smart nutrient assessment method according to an embodiment of the present invention, including steps S1 to S7, which are described in detail below.

In step S1, multiple literature reports are obtained that simultaneously record negative factors and nutrients. According to some embodiments, the negative factors and nutrients may be, for example, those listed in Table 1 below. This embodiment can retrieve a large number of literature reports from at least one academic literature database using keywords for various negative factors and nutrients. Therefore, the specific selection of negative factors and nutrients is not limited in this embodiment and can be configured accordingly.

Table 1: Examples of Negative Factors and Nutrients Negative factors disease 1. Cardiovascular and metabolic systems: hyperglycemia, stroke, etc. 2. Hormonal system: polycystic ovary syndrome, menopause, etc. 3. Immune system: atopic dermatitis, lupus erythematosus, etc. 4. Nervous and psychiatric systems: insomnia, anxiety, etc. 5. Musculoskeletal system: fibromyalgia, osteoporosis, etc. 6. Intestinal system: gastric ulcers, irritable bowel syndrome, etc. 7. Other chronic diseases: chronic kidney disease, macular degeneration, etc. Bad living habits Frequently staying up late, drinking, smoking, eating out, low-fiber diet, high-fat diet, high-sugar diet, high-stress lifestyle, frequent use of consumer electronics, and use of more than five chronic medications. nutrients vitamins Vitamin A, Vitamin B1 (Thiamine), Vitamin B2 (Riboflavin), Vitamin B3 (Niacin), Vitamin B5 (Pantothenic Acid), Vitamin B6 (Pyridoxine), Vitamin B7 (Biotin), Vitamin B9 (Folic Acid), Vitamin B12 (Cholesterol), Vitamin C, Vitamin D, Vitamin E, Vitamin K Minerals Calcium, magnesium, potassium, phosphorus, iron, zinc, copper, selenium, chromium, manganese, cobalt, sodium, potassium, sulfur Other nutrients Fish oil (DHA, EPA), Q10, probiotics, dietary fiber, glutathione, digestive enzymes, lutein, crayfish, serotonin, dopamine, gamma-aminobutyric acid (GABA), etc.

In step S2, the literature reports are graded using the literature grading table, and nutrient weights corresponding to the literature grade are assigned to each nutrient reported in the literature report. The nutrient weights of each nutrient associated with each of the multiple negative factors are screened and summed. The nutrient weights are then arithmetic averaged based on the number of corresponding literature reports to generate the average nutrient weight for each of the multiple negative factors.

According to some implementations, the "screening and summing up the nutrient weighted values of each nutrient associated with each nutrient, and then performing arithmetic averaging based on the number of corresponding literature reports" may include the following: (1) summing up the nutrient weighted values of the same nutrient based on the nutrient weighted values corresponding to each literature level, as shown in Table 3 below; (2) summing up the nutrient weighted values of the same nutrient based on the nutrient weighted values corresponding to at least one literature level (e.g., EBM Level 1, the highest literature level).

According to some embodiments, the reference basis of the literature rating table can be, for example, evidence-based medicine (EBM), impact factor, or any combination thereof.

Specifically, the literature level can be determined based on the level of evidence in literature reports, as determined by evidence-based medicine. Since a lower EBM Level score indicates greater reliability of the evidence, the literature level can be, for example, [e.g., 6 - EBM Level]. This embodiment assigns a literature level to each literature report, as shown in Table 2.

Table 2: Literature Rating Table Publication Type EBM Level Literature Level Meta-Analysis 1 5 Randomized Controlled Trial 1 5 Systematic Review 1 5 Comparative Study 3 3 Controlled Clinical Trial 3 3 Adaptive Clinical Trial 4 2 Case Reports 4 2 Clinical Study 4 2 Clinical Trial 4 2 Clinical Trial, Phase I 4 2 Clinical Trial, Phase II 4 2 Clinical Trial, Phase III 4 2 Clinical Trial, Phase IV 4 2 Observational Study 4 2 Classical Article 5 1 Comment 5 1 Guideline 5 1 Practice Guideline 5 1

Furthermore, the literature levels and nutrient weighting values of all literature reports can be shown in Table 3 below.

Table 3: Example of the relationship between literature level and nutrient weighting Literature Report Literature Level Negative factors Nutrient weighted value calcium Magnesium Vitamin D Lutein AAA 1 osteoporosis 1 BBB 3 osteoporosis 3 3 CCC 4 Insomnia 4 4 DDD 1 Insomnia 1 1 EEE 3 Macular degeneration 3 FFF 4 Macular degeneration 4 GGG 5 Macular degeneration 5

For example, if the negative factor is osteoporosis and the nutrient is calcium, the average nutrient weighted value is (1 + 3) / 2 = 2.5. Therefore, using the average nutrient weighted value balances all nutrient weights, preventing a particular nutrient from increasing in weight as more literature discusses it. For ease of explanation, some values in subsequent steps will still be expressed using nutrient weighted values.

According to some implementations, since the evidence of literature reports with low literature levels may not be sufficient, the literature level range for calculating the average nutrient weighted value can be preset, which may include the following: (1) Only the nutrient weighted values corresponding to the highest literature level (such as EBM Level 1) are used to calculate the average nutrient weighted value. For example: when the negative factor is macular degeneration and the nutrient is lutein, the average nutrient weighted value = (5)/1 = 5. (2) Only the nutrient weights corresponding to the highest and second-highest literature levels (e.g., EBM Level 1, EBM Level 2) are used to calculate the average nutrient weight. For example, when the negative factor is macular degeneration and the nutrient is lutein, the average nutrient weight = (4 + 5) / 2 = 4.5. This is merely an illustration of some specific calculation methods for the average nutrient weight. Any changes or substitutions that are equal or equivalent to the above-mentioned embodiments should be interpreted as being within the spirit or scope of the present invention.

Therefore, this method provides a flexible way to calculate average nutrient weights, allowing adjustments based on the current state of research on negative factors and nutrients. For example, if numerous studies have shown a relationship between osteoporosis and calcium, using nutrient weights corresponding to high-quality literature will help enhance the weight of evidence and credibility of the average nutrient weight. It also prevents the use of low-quality nutrient weights in subsequent arithmetic averaging (due to a larger denominator as the number of literature increases), which could lower the average nutrient weight. Conversely, for emerging research topics where there has been limited literature on the impact of a particular nutrient on a particular negative factor, using nutrient weightings corresponding to lower-level literature reports can diversify the "average nutrient weightings" and provide professionals with more reference options to assist participants. For example, as shown in Table 3, when a subject has the negative factor of insomnia, if only nutrient weights from literature level 4 or above are applied, the "average nutrient weights" are calcium = 4/1 = 4, and magnesium = 4/1 = 4. However, if nutrient weights from all literature levels are applied, the "average nutrient weights" are calcium = 4 + 1/2 = 2.5, magnesium = 4/1 = 4, and vitamin D = 1/1 = 1. This increases the diversity of the "average nutrient weights," allowing professionals to consider recommending a small amount of vitamin D supplementation for the subject.

In step S3: based on the negative factor ranking table, negative factor weighted values corresponding to the multiple negative factors are obtained.

According to some embodiments, the negative factor rating scale can be based on, for example, the Delphi method, the top ten causes of death in China (for diseases), or any combination thereof.

Specifically, the Delphi method can involve multiple rounds of questionnaires designed in advance, allowing experts to assess the health risks of various negative factors based on their respective knowledge and experience (e.g., clinical experience and academic knowledge). For example, the first round of questionnaires asks experts to directly assign specific risk values to specific negative factors (e.g., 0 for no effect; 1 for mild, with persistent, long-term health consequences; 2 for moderate, with immediate and severe health consequences; 3 for severe, with immediate and severe health consequences). The second round of questionnaires will provide the same content as the first round, but will also provide the average risk values of other anonymous experts. The third and fourth rounds of questionnaires can be used to further investigate areas of disagreement among experts (this can be objectively determined through statistical analysis of dispersion) to converge their opinions. Finally, the final risk values from all experts across multiple rounds of questionnaires are averaged to obtain the weighted negative factor values. The correlation between the weighted negative factor values and each negative factor can be shown in Table 4 below.

Table 4: Examples of the relationship between negative factor weightings and each negative factor Negative factors Negative factor weighting osteoporosis 2.1 Insomnia 2.3 Macular degeneration 1.7 Lupus erythematosus 2.7 Migraine 1.8 Anxiety disorder 2.5

In step S4: multiply the nutrient weighted value corresponding to each negative factor and the negative factor weighted value to obtain the nutrient composite weighted value.

According to some embodiments, corresponding to the same negative factor, the nutrient weighted value in Table 3 is multiplied by the negative factor weighted value in Table 4, and the final nutrient composite weighted value can be obtained, for example, as shown in Table 5 below.

Table 5: Examples of nutrient complex weights Literature Report Literature Level Negative factors Nutrient complex weight calcium Magnesium Vitamin D Lutein AAA 1 osteoporosis 1*2.1 BBB 3 osteoporosis 3*2.1 3*2.1 CCC 4 Insomnia 4*2.3 4*2.3 DDD 1 Insomnia 1*2.3 1*2.3 EEE 3 Macular degeneration 3*1.7 FFF 4 Macular degeneration 4*1.7 GGG 5 Macular degeneration 5*1.7

Specifically, the formula for the example in Table 5 is: Nutrient weighted value = Literature level.

Furthermore, when calculating the average weighted nutrient value, the formula is as follows: Average weighted nutrient value = (Reference Level 1 + Reference Level 2 + … + Reference Level N) / N; Combined nutrient weight = (Reference Level 1 + Reference Level 2 + … + Reference Level N) / N * Negative factor weight.

In step S5: obtain subject information.

In step S6: extract the corresponding nutrient complex weights based on the negative factors in the subject's information.

In step S7: nutritional supplementation recommendations are provided to the subject based on the nutrient complex weights.

According to steps S6-S7, according to some implementations, if the negative factors recorded in the subject's information are osteoporosis and insomnia, the subject's nutritional supplement recommendations are: calcium = (1*2.1) + (3*2.1) + (4*2.3) + (1*2.3)/4 = 4; magnesium = (4*2.3)/1 = 4.6; and vitamin D = (3*2.1) + (1*2.3)/2 = 4.3. Therefore, after reviewing the subject's nutritional supplement recommendations, the dietitian should recommend the following order of nutritional supplementation: magnesium, vitamin D, then calcium. This embodiment only lists some of the recommended nutrient supplement values for the subjects. In fact, the subject information also contains other factors that can serve as the basis for analysis in this embodiment.

Furthermore, when the subject's information indicates that the subject falls within the currently medically recognized elderly group, is too thin, or is overweight, the default values of this implementation (for example, age > 65 is considered a negative factor indicating a significant aging trend) are considered as negative factors and used as the basis for analyzing recommended nutrient supplement values.

According to other implementations, step S2 may further determine the nutritional benefit of at least one of the plurality of nutrients recorded in each of the plurality of literature reports with respect to the corresponding negative factor through natural language processing, so as to set a positive nutrient weighting value (positive) corresponding to the literature level (nutrient helps improve the negative factor), a negative nutrient weighting value (negative) corresponding to the negative benefit (nutrient will worsen the negative factor), and a null nutrient weighting value (null) corresponding to the literature level for each nutrient recorded in the plurality of literature reports. Specifically, through natural language processing, the benefits of negative factors reported in the literature can be determined. The recognition characters of natural language processing can be, for example, as shown in Table 6.

Table 6: Recognition characters for natural language processing to determine nutritional benefits benefit Character recognition Positive positive outcome, beneficial, successful, favorable, significant improvement, positive response, therapeutic success, efficacy, positive impact, enhances, boosts Negative detrimental, harmful, adverse effects, failure, negative outcome Null (no benefit) no significant difference, equivalent, similar outcomes, no advantage, equivalent efficacy, similar effectiveness

Furthermore, corresponding identification details can be further configured based on the textual descriptions used in existing medical journals, papers, or clinical trial reports. For example, the above identification characters must appear in the same paragraph or sentence as at least one nutrient in order to serve as evidence of the nutrient's effect on a negative factor.

Therefore, based on the application of natural language processing, the example in Table 3 above can be further modified as shown in Table 7 below.

Table 7: Examples of the relationship between literature level, nutrient benefits, and nutrient weightings Literature Report Literature Level Negative factors effect Nutrient weighted value calcium Magnesium Vitamin D Lutein AAA 1 osteoporosis Positive 1 BBB 3 osteoporosis Positive 3 3 CCC 4 Insomnia Positive 4 CCC 4 Insomnia Negative 4 CCC 4 Insomnia Null 4 DDD 1 Insomnia Positive 1 1 EEE 3 Macular degeneration Positive 3 FFF 4 Macular degeneration Positive 4

As shown in Table 7, natural language processing allows for a more precise analysis of the negative effects of various nutrients reported in various literature reports. Thus, a simple search might identify calcium, magnesium, and vitamin D as nutrients that are beneficial for improving insomnia (as shown in Table 3). However, further natural language processing reveals that only calcium is a nutrient that is beneficial for improving insomnia (as shown in Table 7). Therefore, using the weighted nutrient values obtained through natural language processing as a basis for generating nutrient supplement recommendations will be more valuable. It should be noted that the negative effects and nutrient benefits listed in Tables 3 and 7 are merely examples for illustrative purposes in this manual.

According to some embodiments, after applying natural language processing, positive nutrient weights, negative nutrient weights, and ineffective nutrient weights can be obtained. Therefore, steps S3-S6 of FIG. 1 can be further performed to generate corresponding positive nutrient composite weights, negative nutrient composite weights, and ineffective nutrient composite weights, respectively. Furthermore, by dividing the negative nutrient composite weights and ineffective nutrient composite weights by the positive nutrient composite weights, respectively, a composite nutrient negative benefit ratio and a composite nutrient ineffective benefit ratio can be generated, respectively. Alternatively, the negative nutrient complex weights and the ineffective nutrient complex weights are added together and then divided by the positive nutrient complex weights to produce a complex nutrient benefit ratio.

According to other implementations, a big data database can be established through the intelligent nutrient assessment method. For example, information on relatively healthy subjects or those with slightly negative factors can be collected and standard values can be established. After analyzing the actual subject information, the average weighted value of nutrients, the average weighted value of positive nutrients, the average weighted value of negative nutrients, the average weighted value of ineffective nutrients, the nutrient benefit ratio, the nutrient negative factors, and the average weighted value of the nutrients can be obtained. If at least one of the benefit ratio, nutrient ineffectiveness ratio, nutrient complexity weight, positive nutrient complexity weight, negative nutrient complexity weight, ineffective nutrient complexity weight, complex nutrient benefit ratio, complex nutrient negative benefit ratio, and complex nutrient ineffectiveness ratio exceeds the standard value, customized nutrient supplementation recommendations will be provided, allowing professionals to recommend specific nutrients to the subject.

According to some implementations, when a (multi-)nutrient negative benefit ratio or a (multi-)nutrient no-benefit ratio is greater than 0.3, the corresponding nutrient will not be recommended as a priority intake. For example, based on a subject's information, if the no-benefit ratio of vitamin D is analyzed to be greater than 0.3, vitamin D will not be recommended to the subject as a priority intake.

According to other embodiments, in step S7, the subject can be provided with food intake recommendations based on a plurality of preset food nutrient information records, including the nutrient content of each specific food (e.g., carrots) per a certain weight (e.g., 100 grams), and customized nutrient supplementation recommendations. Furthermore, if the subject's information indicates a desire to implement a special nutritional plan, the recommended intake information is adjusted. Special nutrition plans include: diabetic diets, weight loss diets, lactose intolerance diets, high-protein diets, high-fiber diets, allergen elimination diets, gluten-free diets, vegetarian diets, ketogenic diets, low-sodium diets, and low-fat diets.

The intelligent nutrient assessment method based on the first embodiment can be adjusted accordingly as needed to be set in hardware, devices, or systems. However, this does not mean that the hardware, devices, or systems need to apply all of the first embodiment and its implementation aspects, as detailed in the following embodiments.

Second embodiment: generating a first customized nutrient supplement recommendation value

As shown in Figure 2, a first block diagram of an intelligent nutrient assessment system according to an embodiment of the present invention is provided. This embodiment provides an intelligent nutrient assessment system 100, which is signal-connected to at least one literature database 200. The intelligent nutrient assessment system 100 includes a literature retrieval module 110, a literature weighting module 120, and a processing module 130. The literature search module 110 is signal-connected to at least one literature database 200. The literature search module 110 uses a search command composed of keywords including a plurality of negative factors and a plurality of nutrients to obtain from the at least one literature database 200 a plurality of literature reports that simultaneously record at least one of the plurality of negative factors and at least one of the plurality of nutrients. The plurality of negative factors include diseases and unhealthy lifestyle habits (as shown in Table 1). The literature weighting module 120 is signal-connected to the literature retrieval module 110. The literature weighting module 120 ranks the plurality of literature reports using a literature rating table (as shown in Table 2). Based on the literature ratings corresponding to the plurality of literature reports, the module sets nutrient weighting values corresponding to the literature ratings for the nutrients recorded in the plurality of literature reports (as shown in Table 3). The nutrient weighting value corresponds to one of the plurality of negative factors. Processing module 130 is signal-connected to literature weighting module 120. Processing module 130 selects and sums the nutrient weights of each nutrient associated with each of the multiple negative factors. It then arithmetic averages these sums based on the number of corresponding literature reports to generate an average nutrient weight for each of the multiple negative factors.

As shown in Figure 3, a second block diagram of an intelligent nutrient assessment system according to an embodiment of the present invention is provided. According to some embodiments, the processing module 130 is further signally connected to the execution module 140, which is signally connected to the user device 400. The execution module 140 receives subject information corresponding to a subject provided by the user device 400. The subject information records at least one of a plurality of negative factors. The processing module 130 extracts the average weighted nutrient value corresponding to each negative factor recorded in the subject information as a first customized nutrient supplement recommendation.

Third embodiment: generating a second customized nutrient supplement recommendation value

As shown in FIG2 , based on the second embodiment and in some implementations, the literature weighting module 120 further determines the nutritional benefits of each of the plurality of nutrients recorded in the plurality of literature reports for the negative factors based on natural language processing, and sets the sum of the positive nutrient weights, the sum of the negative nutrient weights, and the sum of the ineffective nutrient weights corresponding to the literature level for the nutrients recorded in each of the plurality of literature reports. The module then performs an arithmetic average based on the number of corresponding literature reports to generate an average positive nutrient weight, an average negative nutrient weight, and an average ineffective nutrient weight corresponding to each of the plurality of negative factors.

As shown in Figure 3, the processing module 130 is further signal-connected to the execution module 140, and the execution module 140 is signal-connected to the user device 400 for receiving subject information corresponding to the subject provided by the user device 400. The subject information records at least one of a plurality of negative factors. The processing module 130 extracts the average weighted value of positive nutrients, the average weighted value of negative nutrients, and the average weighted value of ineffective nutrients corresponding to each negative factor recorded in the subject information as the second customized nutrient supplement recommendation value.

Fourth embodiment: generating a third customized nutrient supplement recommendation value

As shown in FIG3 , continuing with the third embodiment, the processing module 130 further divides the average weighted value of the negative nutrients and the average weighted value of the ineffective nutrients by the average weighted value of the positive nutrients to generate a nutrient negative benefit ratio and a nutrient ineffective benefit ratio, respectively, as the third customized nutrient supplement recommendation. Alternatively, the average weighted value of the negative nutrients and the average weighted value of the ineffective nutrients are first added together and then divided by the composite weighted value of the positive nutrients to generate a nutrient benefit ratio as the third customized nutrient supplement recommendation.

Fifth embodiment: generating a fourth customized nutrient supplement recommendation value

As shown in Figure 4, a third block diagram of the smart nutrient assessment system according to an embodiment of the present invention is provided. Based on the second embodiment, and in some embodiments, processing module 140 is further signal-connected to negative factor weighting module 150. Negative factor weighting module 150 provides a corresponding negative factor weighting value (as shown in Table 4) for each of the multiple negative factors based on a negative factor grading table. Processing module 140 then multiplies the average nutrient weighting value corresponding to each of the multiple negative factors by the negative factor weighting value to generate a composite nutrient weighting value corresponding to each of the multiple negative factors (as shown in Table 5).

As shown in Figure 3, the processing module is further signal-connected to the execution module 140, which is signal-connected to the user device 400 for receiving subject information corresponding to the subject provided by the user device 400. The subject information records at least one of a plurality of negative factors. The processing module 130 extracts the nutrient complex weight corresponding to each negative factor recorded in the subject information as the fourth customized nutrient supplement recommendation value.

Sixth embodiment: generating the fifth customized nutrient supplementation recommendation value

As shown in FIG4 , based on the second embodiment, and according to some implementations, the processing module 140 is further connected to the negative factor weighting module 150. The negative factor weighting module 150 provides a corresponding negative factor weighting value (as shown in Table 4) for each of the plurality of negative factors based on the negative factor grading table, so that the processing module 140 multiplies the average weighted value of positive nutrients, the average weighted value of negative nutrients, and the average weighted value of ineffective nutrients corresponding to each of the plurality of negative factors by the negative factor weighting value, respectively, to generate the composite weighted value of positive nutrients, the composite weighted value of negative nutrients, and the composite weighted value of ineffective nutrients corresponding to each of the plurality of negative factors. Alternatively, the average weight of the positive nutrients is multiplied by the weight of the negative factors to produce a composite weight of the positive nutrients for each multiple negative factors.

As shown in Figure 3, the processing module is further signally connected to the execution module 130, which is signally connected to the user device 400. The execution module 130 is configured to receive subject information corresponding to a subject provided by the user device 400. The subject information records at least one of a plurality of negative factors. The processing module 130 extracts the positive nutrient composite weight, negative nutrient composite weight, and/or ineffective nutrient composite weight corresponding to each negative factor recorded in the subject information as the fifth customized nutrient supplementation recommendation. Alternatively, the processing module 130 only extracts the positive nutrient composite weight corresponding to each negative factor recorded in the subject information as the fifth customized nutrient supplementation recommendation.

Seventh embodiment: generating the sixth customized nutrient supplementation recommendation value

As shown in FIG3 , continuing with the sixth embodiment, the processing module 130 further divides the negative nutrient composite weights and the ineffective nutrient composite weights by the positive nutrient composite weights to generate a composite nutrient negative benefit ratio and a composite nutrient ineffective benefit ratio, respectively, as the sixth customized nutrient supplementation recommendation. Alternatively, the negative nutrient composite weights and the ineffective nutrient composite weights are first added together and then divided by the positive nutrient composite weights to generate a composite nutrient benefit ratio as the sixth customized nutrient supplementation recommendation.

Based on the second to seventh embodiments, the following embodiments may be further included.

According to some implementations, the user device 400 may be a device with network and telecommunication capabilities, such as a desktop computer or a portable device. The execution module 140 may be an application programming interface (API) for receiving subject information.

According to other embodiments, the intelligent nutrient assessment system 100 can be implemented on a computer device equipped with computational processing capabilities. The literature database 200 can be, for example, websites of major scientific journals, international clinical trial databases, or academic paper databases. Furthermore, it can be, for example, PubMed (a database used to search MEDLINE).

As shown in FIG2 or FIG4 , according to other embodiments, the negative factor weighting module 130 is further signal-connected to a plurality of expert devices 300 to receive negative factor weighting values corresponding to each of the plurality of negative factors provided by a plurality of experts. The expert devices 300 may be, for example, devices with network and telecommunication capabilities, such as desktop computers or portable devices.

According to some other embodiments, the literature search module 110 executes the search command through an application program interface of any literature search platform, such as the PubMed API.

According to other embodiments, the processing module 130 further includes a plurality of food nutrient information, which records the various nutrient contents of foods of the same weight and provides the subject with food intake recommendation information for selecting food based on customized nutrient supplementation recommendations.

Furthermore, the processing module 130 also includes a plurality of special nutritional plans. When the subject information records one of the plurality of special nutritional plans, the processing module 130 adjusts the recommended intake information according to the special nutritional plan to be implemented in the subject information.

In summary, the first to sixth customized nutrient supplementation recommendation values can be generated based on the subject's information (recording at least one negative factor) using the smart nutrient assessment method or the smart nutrient assessment system 100 to provide professionals (e.g., doctors or nutritionists) with quantitative recommended values for each nutrient. The following are some bases and examples to assist in determining whether the subject needs to supplement a specific nutrient: (1) First customized nutrient supplementation recommendation value: Based on the subject's negative factors, the "average weighted nutrient value" of each nutrient is calculated as the "first customized nutrient supplementation recommendation value." For example, if a subject suffers from osteoporosis, arrhythmia, tingling or spasms in the hands and face, and night blindness, based on the literature reports that indicate that these diseases can be supplemented with nutrients, the "average weighted values of nutrients" for calcium and vitamin A may be provided as the first customized nutrient supplement recommendation. (2) Second customized nutrient supplement recommendation: Based on the subject's negative factors and natural language recognition, the "average weighted value of positive nutrients", "average weighted value of negative nutrients", and "average weighted value of ineffective nutrients" for each nutrient are calculated as the "second customized nutrient supplement recommendation". For example, if the subject also suffers from osteoporosis, based on literature reports documenting the effects of certain nutrients on osteoporosis, the "positive nutrient average weighted value" for calcium and/or the "negative nutrient average weighted value" for sodium may be provided (nutrient effects are indicative only and do not represent actual effects). (3) Third customized nutrient supplementation recommendation value: Based on the negative factors of the subject and natural language recognition, the "average weighted value of negative nutrients" and "average weighted value of ineffective nutrients" of each nutrient are divided by the "weighted value of positive nutrients" to generate the "nutrient negative benefit ratio" and "nutrient ineffectiveness ratio" respectively, which serve as the "third customized nutrient supplementation recommendation value". For example: Based on the subject information of a subject, the nutrient ineffectiveness ratio of vitamin D is analyzed to be greater than 0.3, so vitamin D will not be recommended to the subject for priority intake. (4) The fourth customized nutrient supplement recommendation value: Based on the negative factors and the negative factor weighted values of the subjects, the "average nutrient weighted value" corresponding to each negative factor is multiplied by the "negative factor weighted value" to generate the "nutrient composite weighted value" corresponding to each negative factor as the "fourth customized nutrient supplement recommendation value". For example, if a subject suffers from osteoporosis, arrhythmia, tingling or cramps in the hands and face, and night blindness, since arrhythmia is a more serious negative factor, it has a higher negative factor weighting value and can be improved by calcium supplementation. Therefore, after adding the "negative factor weighting value", calcium will produce a higher "nutrient complex weighting value" as the "fourth customized nutrient supplement recommendation value." (5) The fifth customized nutritional supplement recommendation value: Based on the negative factors of the subjects, natural language recognition and the weighted value of the negative factors, the "positive nutritional complex weight value", "negative nutritional complex weight value" and "ineffective nutritional complex weight value" are used as the "fifth customized nutritional supplement recommendation value". (6) Sixth customized nutrient supplementation recommendation value: Based on the negative factors of the subjects, natural language recognition and the weighted values of the negative factors, the "negative nutrient composite weight" and "ineffective nutrient composite weight" of each nutrient are divided by the "positive nutrient composite weight" to generate the "complex nutrient negative benefit ratio" and "complex nutrient ineffectiveness ratio" respectively, which serve as the "sixth customized nutrient supplementation recommendation value".

The above description is merely an example of a preferred embodiment of the present invention and is not intended to limit the scope of implementation. Any simple substitution or equivalent variation based on the scope of the present invention and the contents of the patent specification shall fall within the scope of the present invention.

100: Smart Nutrition Assessment System 110: Literature Search Module 120: Literature Weighting Module 130: Processing Module 140: Execution Module 150: Negative Factor Weighting Module 200: Literature Database 300: Expert Device 400: User Device S1-S7: Steps

Figure 1 is a flow chart of the smart nutrient assessment method according to an embodiment of the present invention. Figure 2 is a schematic diagram of the first block of the smart nutrient assessment system according to an embodiment of the present invention. Figure 3 is a schematic diagram of the second block of the smart nutrient assessment system according to an embodiment of the present invention. Figure 4 is a schematic diagram of the third block of the smart nutrient assessment system according to an embodiment of the present invention.

S1~S7: Steps

Claims (12)

A smart nutrient evaluation system is signal-connected to at least one literature database. The smart nutrient evaluation system includes: a literature retrieval module, signal-connected to the at least one literature database, the literature retrieval module forms a retrieval instruction with a keyword having a plurality of negative factors and a plurality of nutrients, and obtains a plurality of literature reports from the at least one literature database that simultaneously record at least one of the negative factors and at least one of the nutrients, wherein the negative factors include diseases and unhealthy living habits; a literature weighting module, signal-connected to the literature retrieval module The document weighting module classifies the document reports using a document rating table, and sets a nutrient weighting value corresponding to the document level for each nutrient recorded in the document reports according to the document level corresponding to each of the document reports, wherein the nutrient weighting value corresponds to one of the negative factors, and the document weighting module determines a nutrient benefit of each of the nutrients recorded in the document reports for the negative factor based on a natural language processing; a processing module, signal-connected to the document weighting module, the processing module screens each of the nutrients recorded in the document reports. The nutrient weighted values of each nutrient associated with the negative factors are summed up, and then arithmetic average is performed based on the number of corresponding literature reports to generate the average nutrient weighted value corresponding to each negative factor. According to the nutrient benefits, the sum of the positive nutrient weighted values, the sum of the negative nutrient weighted values, and the sum of the ineffective nutrient weighted values corresponding to the literature level of the nutrient settings recorded in each of the literature reports are calculated, and arithmetic average is performed based on the number of corresponding literature reports to generate an average nutrient weighted value corresponding to each negative factor. an average weighted value of positive nutrients, an average weighted value of negative nutrients, and an average weighted value of ineffective nutrients; and a negative factor weighting module, signal-connected to the processing module. The negative factor weighting module provides a corresponding negative factor weighting value for each of the negative factors based on a negative factor grading table, so that the processing module multiplies the average weighted value of the positive nutrients, the average weighted value of the negative nutrients, and the average weighted value of the ineffective nutrients corresponding to each of the negative factors by the negative factor weighting value to generate a nutrient composite weighted value corresponding to each of the negative factors. As described in claim 1, the intelligent nutrient assessment system, wherein the processing module is further signal-connected to an execution module, and the execution module is signal-connected to a user device for receiving subject information corresponding to a subject provided by the user device, wherein the subject information records at least one of the negative factors, and the processing module extracts the average weighted value of the nutrient corresponding to each negative factor recorded in the subject information as a first customized nutrient supplement recommendation value. A smart nutrient assessment system as described in claim 1, wherein the processing module is further signal-connected to an execution module, which is signal-connected to a user device and is used to receive subject information corresponding to a subject provided by the user device, wherein the subject information records at least one of the negative factors, and the processing module extracts the average weighted value of the positive nutrients, the average weighted value of the negative nutrients, and the average weighted value of the ineffective nutrients corresponding to each negative factor recorded in the subject information as a second customized nutrient supplement recommendation value. The intelligent nutrient assessment system as described in claim 3, wherein the processing module further divides the average weighted value of the negative nutrients and the average weighted value of the ineffective nutrients by the average weighted value of the positive nutrients, respectively, to generate a nutrient negative benefit ratio and a nutrient ineffectiveness ratio, respectively, as a third customized nutrient supplement recommendation value. As described in claim 1, the intelligent nutrient assessment system, wherein the processing module is further signal-connected to an execution module, and the execution module is signal-connected to a user device for receiving subject information corresponding to a subject provided by the user device, wherein the subject information records at least one of the negative factors, and the processing module extracts the nutrient complexity weight corresponding to each negative factor recorded in the subject information as a fourth customized nutrient supplement recommendation value. The intelligent nutrient assessment system as described in claim 1, wherein the negative factor weighting module further provides a corresponding negative factor weighting value for each of the negative factors based on the negative factor grading table, so that the processing module multiplies the average weighted value of the positive nutrients, the average weighted value of the negative nutrients, and the average weighted value of the ineffective nutrients corresponding to each of the negative factors by the negative factor weighting value, respectively, to generate a positive nutrient composite weighting value, a negative nutrient composite weighting value, and an ineffective nutrient composite weighting value corresponding to each of the negative factors. A smart nutrient assessment system as described in claim 6, wherein the processing module is further signal-connected to an execution module, which is signal-connected to a user device and is used to receive subject information corresponding to a subject provided by the user device, wherein the subject information records at least one of the negative factors, and the processing module extracts the positive nutrient composite weight value, the negative nutrient composite weight value, and the ineffective nutrient composite weight value corresponding to each negative factor recorded in the subject information as a fifth customized nutrient supplement recommendation value. The intelligent nutrient assessment system as described in claim 7, wherein the processing module further divides the negative nutrient composite weight and the ineffective nutrient composite weight by the positive nutrient composite weight, respectively, to generate a composite nutrient negative benefit ratio and a composite nutrient ineffectiveness ratio, respectively, as a sixth customized nutrient supplement recommendation value. A smart nutrient evaluation method includes: obtaining a plurality of literature reports that simultaneously record at least one of the negative factors and at least one of the nutrients through a search instruction consisting of a plurality of negative factors and a plurality of nutrients; grading the literature reports through a literature grading table; and grading the literature reports according to the literature grades corresponding to the literature reports. The nutrient recorded in the literature reports is set with a nutrient weighted value corresponding to the literature level, wherein the nutrient weighted value corresponds to one of the negative factors; the nutrient benefit of at least one of the nutrients recorded in the literature reports corresponding to the negative factor is determined, and the nutrient weighted value is split into a positive nutrient weighted value, a negative nutrient weighted value and a negative nutrient weighted value according to the nutrient benefit. and an ineffective nutrient weighted value; screen a positive nutrient weighted value, a negative nutrient weighted value, and an ineffective nutrient weighted value of each nutrient associated with each of the negative factors, add them up, and then perform arithmetic averaging based on the number of corresponding literature reports to generate an average positive nutrient weighted value, an average negative nutrient weighted value, and an average negative nutrient weighted value corresponding to each of the negative factors. an average weighted value of ineffective nutrients; providing a corresponding negative factor weighted value for each of the negative factors according to a negative factor grading table, and multiplying the average weighted value of the positive nutrients corresponding to each of the negative factors, the average weighted value of the negative nutrients, and the average weighted value of the ineffective nutrients by the weighted value of the negative factors to obtain a composite weighted value of the nutrients corresponding to each of the negative factors. The smart nutrient assessment method as described in claim 9, when obtaining subject information, further includes: extracting the average weighted value of the nutrient corresponding to each negative factor recorded in the subject information. The smart nutrient assessment method as described in claim 9, wherein the average weighted value of the positive nutrients, the average weighted value of the negative nutrients, and the average weighted value of the ineffective nutrients corresponding to each of the negative factors are respectively multiplied by the weighted value of the negative factor to obtain a positive nutrient composite weighted value, a negative nutrient composite weighted value, and an ineffective nutrient composite weighted value corresponding to each of the negative factors. The smart nutrient assessment method as described in claim 11 further includes: generating a complex nutrient negative benefit ratio and a complex nutrient ineffective benefit ratio by dividing the negative nutrient complex weight and the ineffective nutrient complex weight by the positive nutrient complex weight, respectively.