Home nurse diet therapy index intervention of chronic disease composition intelligent decision method and system

Abstract

This invention provides a method and system for intelligent decision-making regarding dietary intervention for chronic diseases using a home nurse dietary index. It relates to the fields of dietary intervention for chronic diseases and intelligent decision-making for nutritional health. The method includes: constructing a foundational database and a non-nutrient chronic disease effect anchoring map based on the home nurse dietary index, generating corresponding mapping relationships; performing dynamic weight calibration through a coupling model to obtain the optimal weight vector; collecting and standardizing multi-dimensional patient data; generating personalized weights based on the weight vector and individual data, calculating the dietary index score, and completing risk grading and target anchoring; screening foods and verifying synergistic effects using relevant data to generate a candidate food pool; optimizing the proportions using an optimization algorithm to obtain an initial plan; constructing a validation model to complete effect and safety assessments, generating a final plan; and ultimately forming a personalized chronic disease intervention strategy throughout the entire process.

Description

Technical Field

[0001] This invention relates to the field of dietary intervention for chronic diseases and intelligent decision-making for nutritional health, and in particular to a method and system for intelligent decision-making of combinations for family nurses to intervene in chronic diseases using dietary therapy index. Background Technology

[0002] Dietary intervention is a core component of the whole-cycle prevention and control of chronic diseases. The non-nutrient active components in plant-based foods can play a role in chronic disease intervention through anti-inflammatory, antioxidant stress and metabolic improvement pathways, which is a key research direction in the fields of nutritional science and chronic disease prevention and control.

[0003] The existing dietary evaluation system has formed a relatively mature technical framework, mainly including three types of evaluation indices: nutrient-oriented, food-type-oriented, and composite index-oriented. This provides a standardized reference for assessing the quality of residents' diets and has broad application value in dietary guidance and chronic disease prevention. Related dietary recommendations are based on general dietary guidelines, providing basic dietary intervention references for patients with chronic diseases and promoting the standardized development of nutritional intervention techniques.

[0004] However, existing dietary indices mostly adopt a fixed weight model, do not include the dimension of non-nutrient chronic disease intervention effects, lack targeted screening and synergistic effect verification for composition recommendations, and have not formed a closed-loop decision-making mechanism from weight calibration and ratio optimization to effect verification. Summary of the Invention

[0005] To address the technical issues that existing dietary indices mostly adopt a fixed weighting model, do not incorporate the dimension of non-nutrient chronic disease intervention effects, lack targeted screening and synergistic effect verification in composition recommendations, and do not form a closed-loop decision-making mechanism from weight calibration and ratio optimization to effect verification, this invention provides a method and system for intelligent decision-making of compositions for family nurses to intervene in chronic diseases using dietary therapy indices.

[0006] The technical solutions provided by the embodiments of the present invention are as follows: The first aspect of this invention provides a smart decision-making method for family nurses' dietary therapy index intervention in chronic diseases, comprising: S1: Based on the family nurse dietary therapy index, construct the underlying support database and the non-nutrient-chronic disease effect anchoring map to generate basic data and non-nutrient-chronic disease intervention effect mapping relationship; S2: Through the coupling model, complete the dynamic weight calibration of chronic disease-specific family nurse dietary therapy index and generate the optimal weight vector of non-nutrient indicators. S3: Collect multi-dimensional individual data of patients with chronic diseases and perform standardized processing to generate standardized individual datasets; S4: Based on the optimal weight vector of non-nutrient indicators and the standardized individual dataset, generate a personalized non-nutrient indicator weight vector, calculate the baseline dietary family nurse dietary therapy index score, complete health risk classification and intervention target anchoring, and generate an intervention target list. S5: Combining basic data, the mapping relationship between non-nutrients and chronic disease intervention effects, standardized individual datasets, personalized non-nutrient index weight vectors, and a list of intervention targets, we completed the screening of plant-derived foods and the verification of non-nutrient synergistic effects, generating a pool of candidate food combinations. S6: Based on the personalized non-nutrient index weight vector, the list of intervention targets, and the candidate food pool for the composition, the composition ratio is optimized through an optimization algorithm to generate the initial composition plan. S7: Based on the mapping relationship between non-nutrients and chronic disease intervention effects, construct a validation model, input the initial composition scheme into the validation model, complete the validation of intervention effects and safety assessment, and generate the final composition scheme; S8: Generate personalized chronic disease intervention strategies based on standardized individual datasets and final composite schemes.

[0007] The second aspect of this invention provides a smart decision-making system for home nurse dietary therapy index intervention in chronic diseases, comprising: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the intelligent decision-making method for family nurse dietary therapy index intervention of chronic diseases as described in the first aspect.

[0008] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: This invention addresses the shortcomings of existing dietary indices, such as fixed weights, lack of inclusion of non-nutrient intervention effects, lack of targeted screening and synergistic validation of compositions, and absence of a closed-loop decision-making mechanism. It constructs a non-nutrient chronic disease effect anchoring map to incorporate non-nutrient chronic disease intervention effects. A coupled model is used for dynamic weight calibration, replacing the fixed-weight approach. Multi-dimensional data is combined to screen plant-based foods and validate non-nutrient synergistic effects, achieving targeted screening and synergistic enhancement. Based on optimized algorithm formulation, validation model evaluation, and a comprehensive strategy, a closed-loop decision-making mechanism is constructed, encompassing weight calibration, formulation optimization, and effect validation, comprehensively improving the accuracy and practicality of dietary composition interventions for chronic diseases. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the intelligent decision-making method for family nurses' dietary therapy index intervention in chronic diseases, as provided in an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of the intelligent decision-making system for family nurse dietary therapy index intervention of chronic diseases provided in an embodiment of the present invention. Detailed Implementation

[0012] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0013] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0014] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0015] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0016] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0017] Reference manual attached Figure 1 The diagram illustrates a flowchart of the intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases provided by an embodiment of the present invention.

[0018] This invention provides a method for intelligent decision-making regarding the composition of a home nurse's dietary therapy index intervention for chronic diseases. This method can be implemented using a home nurse's dietary therapy index intelligent decision-making device for chronic disease intervention, which can be a terminal or a server. The processing flow of the home nurse's dietary therapy index intelligent decision-making method for chronic disease intervention may include the following steps: S1: Based on the family nurse dietary therapy index, construct the underlying support database and the non-nutrient-chronic disease effect anchoring map to generate basic data and non-nutrient-chronic disease intervention effect mapping relationship.

[0019] Among them, the Family Nurse Dietary Therapy Index is a dynamic scoring system used to quantitatively evaluate the value of non-nutrient active ingredients in the diet for chronic disease intervention. The underlying supporting database includes a food-non-nutrient content database and a non-nutrient-health effect quantitative database. The non-nutrient-chronic disease effect anchoring map is a knowledge graph that establishes the mapping relationship between non-nutrients and effect dimensions, chronic disease pathological targets, effect intensity, and evidence-based level. The basic data are the non-nutrient content and health effect quantitative values ​​stored in the underlying supporting database. The non-nutrient-chronic disease intervention effect mapping relationship is an intervention knowledge set that includes the synergistic and antagonistic effects of non-nutrients.

[0020] In one possible implementation, S1 specifically includes sub-steps S101 to S103: S101: Based on the family nurse dietary therapy index, construct an underlying support database, which includes a food-non-nutrient content database and a non-nutrient-health effect quantification database.

[0021] Among them, the food-nonnutrient content database records the content data of each nonnutrient in 100 grams of edible plant-based food, and the nonnutrient-health effect quantitative database records the standardized effect scores of each nonnutrient in the dimensions of anti-inflammatory effect, antioxidant stress effect, and metabolic regulation effect.

[0022] The food-nonnutrient content database records the content of each nonnutrient in 100 grams of edible plant-based foods. For example, it collects nonnutrient content data for 100g of edible portions of common plant-based foods in my country, covering all food categories including grains, tubers, legumes, vegetables, fruits, soybeans, nuts, and algae.

[0023] Specifically, the Non-Nutrient-Health Effects Quantification Database records the standardized effect scores of each non-nutrient in the dimensions of anti-inflammatory effects, antioxidant stress effects, and metabolic regulation effects. These scores are calculated based on a weighted scoring system of evidence-based literature: 10 points for clinical intervention studies, 8 points for prospective cohort studies, 7 points for case-control studies, 6 points for cross-sectional studies, 5 points for animal experiments, 3 points for cell experiments, and 1 point for molecular experiments. Then, the raw effect scores for each dimension are calculated using the following formula:

[0024] in, N represents the raw effect score of the q-th non-nutrient in the m-th effect dimension. q,m The score represents the total number of valid studies included in the corresponding effect dimension for this non-nutrient. i This represents the evidence-based weighted score of the i-th document under this dimension.

[0025] Furthermore, the scores are mapped to the [0,1] interval through min-max normalization:

[0026] Among them, S q,m This represents the standardized effect score of the q-th non-nutrient in the m-th effect dimension. This represents the maximum original effect score for all core non-nutrients included in the evaluation in this effect dimension. This represents the minimum original effect score for all core non-nutrients included in the evaluation in this effect dimension.

[0027] S102: Based on the food-nonnutrient content database and the nonnutrient-health effect quantification database, generate basic data and establish a multi-dimensional mapping relationship between nonnutrients and chronic disease effects to form a nonnutrient-chronic disease effect anchoring map.

[0028] Among them, the multi-dimensional mapping relationship includes the correspondence between non-nutrients and effect dimensions, pathological targets of chronic diseases, effect intensity, and evidence-based level.

[0029] Specifically, for target chronic diseases, the core pathological features and intervention targets should be clearly identified first: The core pathology of type 2 diabetes is insulin resistance, glucose metabolism disorder, chronic low-grade inflammation, and oxidative stress damage. The core pathology of essential hypertension is vascular endothelial damage, oxidative stress, chronic low-grade inflammation, and water and sodium metabolism disorder. The core pathology of dyslipidemia is lipid metabolism disorder, vascular endothelial inflammation, and oxidative stress. The core pathology of obesity is fat accumulation, chronic low-grade inflammation, and insulin resistance. The core pathology of chronic low-grade inflammation-related chronic diseases is persistently elevated inflammatory factors and oxidative stress damage.

[0030] S103: Based on the non-nutrient-chronic disease effect anchoring map, the synergistic and antagonistic effects between non-nutrients are marked, and the non-nutrient-chronic disease intervention effect mapping relationship is generated.

[0031] Among them, synergistic effect refers to the relationship that the intervention effect after the combination of two or more non-nutrients is greater than the sum of their individual effects, while antagonistic effect refers to the relationship that the intervention effect after the combination is less than the sum of their individual effects.

[0032] For example, anthocyanins and proanthocyanidins have a synergistic antioxidant effect, and soy isoflavones and quercetin have a synergistic anti-inflammatory effect. Antagonistic attenuation refers to a relationship where the combined intervention effect is less than the sum of the individual effects.

[0033] In this embodiment of the invention, a food-nonnutrient content database and a nonnutrient-health effect quantification database are constructed. Based on evidence-based literature weighted scoring and min-max standardization, a five-dimensional anchoring map of nonnutrient-chronic disease effects is established. At the same time, the synergistic and antagonistic effects between nonnutrients are marked, realizing the precise matching of active ingredients with chronic disease pathological targets. This provides a standardized and quantifiable data foundation for subsequent dynamic weight calibration and food screening.

[0034] S2: Through the coupling model, complete the dynamic weight calibration of chronic disease-specific weights for the family nurse dietary therapy index, and generate the optimal weight vector for non-nutrient indicators.

[0035] Among them, the coupling model is a weight optimization model that combines fuzzy hierarchical analysis with the maximum entropy principle. Chronic disease-specific dynamic weight calibration refers to the process of adjusting the weights of non-nutrient indicators for different chronic disease types and individual patient pathological states. The optimal weight vector of non-nutrient indicators is a set of weights that satisfies the maximum weight distribution entropy and conforms to the constraints of the fuzzy judgment matrix.

[0036] In one possible implementation, S2 specifically includes sub-steps S201 to S203: S201: Based on triangular fuzzy numbers, construct a fuzzy judgment matrix between non-nutrient indicators, and perform a consistency check on the fuzzy judgment matrix.

[0037] Among them, the triangular fuzzy number is a fuzzy mathematical expression of the relative importance of two indicators using three parameters: lower bound, median, and upper bound. The fuzzy judgment matrix is ​​a square matrix composed of the comparison values ​​of the importance of the triangular fuzzy numbers between each indicator. The consistency test is used to judge the logical self-consistency of the matrix.

[0038] Among them, the triangular fuzzy number is a fuzzy mathematical expression that uses three parameters—lower bound, median, and upper bound—to represent the relative importance of two indicators, denoted as […]. ,in The lower realm The median value. For the upper boundary, satisfy .

[0039] Specifically, we first define the triangular fuzzy number, let the triangular fuzzy number be... ,in The lower realm The median value. For the upper boundary, satisfy .

[0040] The operation rules for triangular fuzzy numbers are as follows: ,in This represents triangular fuzzy number addition. This represents triangular fuzzy number multiplication. and These are two triangular fuzzy numbers.

[0041] Then construct the fuzzy judgment matrix. ,in The number of primary core indicators, here , As an indicator relative to indicators The importance triangular fuzzy number, the triangular fuzzy number corresponding to the importance level is: equal importance is Slightly more important Obviously important , strongly important Extremely important And satisfy .

[0042] Next, the fuzzy weight vector is calculated. First, the geometric mean of the triangular fuzzy numbers in each row is calculated. : ,in Indicates from arrive Triangular fuzzy number multiplication, Representing triangular fuzzy numbers The root is then used to calculate the fuzzy weights. :

[0043] in, Indicates from arrive Triangular fuzzy numbers are added together. It represents the reciprocal of the triangular fuzzy number.

[0044] For example, five experts each from the fields of nutrition, chronic disease medicine, and food science, totaling 15 experts, were invited to construct fuzzy judgment matrices for type 2 diabetes, essential hypertension, dyslipidemia, obesity, and chronic low-grade inflammation-related chronic diseases. The 15 fuzzy judgment matrices for each chronic disease were then aggregated to obtain the aggregated fuzzy judgment matrix. The aggregation method is to take the average of the medians of the triangular fuzzy numbers of each expert as the median of the aggregated triangular fuzzy number, take the average of the lower bounds as the lower bound of the aggregated triangular fuzzy number, and take the average of the upper bounds as the upper bound of the aggregated triangular fuzzy number.

[0045] Then, a consistency test is performed on the aggregated fuzzy judgment matrix. The consistency test is performed by calculating the fuzzy consistency ratio. To achieve this, first calculate the fuzzy consistency index. :

[0046] in, The largest eigenvalue of the aggregated fuzzy judgment matrix is ​​obtained by calculating the eigenvalue of the triangular fuzzy number.

[0047] S202: Using a coupled model of fuzzy hierarchical analysis and the principle of maximum entropy, with the goal of maximizing the entropy of the weight distribution, and in combination with constraints, the optimal weights of non-nutrient indicators are solved.

[0048] Among them, the weight distribution entropy is used to measure the uniformity of weight distribution, and the constraints include fuzzy weight boundary constraints and weight normalization constraints.

[0049] Specifically, the entropy of the weights Defined as:

[0050] in, As an indicator Clear weights, It is the natural logarithm function.

[0051] The constraints include constraints on fuzzy weights, i.e., clear weights. It needs to fall under the fuzzy weight Within the range: , Simultaneously satisfying the weight normalization constraint: And nonnegativity constraints: , .

[0052] The objective function of the coupled model is to maximize entropy. And simultaneously satisfy all the above constraints, that is:

[0053] in, This indicates a constraint.

[0054] Calculate the fuzzy consistency ratio again :

[0055] in, As a random consistency index, for , .when If the median is less than 0.1, the aggregated fuzzy judgment matrix is ​​considered to have passed the consistency test; otherwise, the fuzzy judgment matrix needs to be adjusted.

[0056] In the second stage, the Lagrange multiplier method combined with gradient descent algorithm is used to solve the objective function of the coupled model, and the optimal clear weight vector is obtained. , ..., First, construct the Lagrange function. :

[0057] in, For normalized constrained Lagrange multipliers, For Lagrange multipliers with lower bound constraints, Let be an upper bounded Lagrange multiplier that satisfies the following conditions: , .

[0058] Then, regarding the Lagrange function... Find the partial derivatives and set them to 0:

[0059]

[0060] Taking into account the constraints, the gradient descent algorithm is used for iterative solution, and the iterative formula is as follows:

[0061]

[0062]

[0063]

[0064] in, For the number of iterations, The learning rate is set here. The iteration termination condition is At this point, we get That is, the optimal clarity weight .

[0065] For different chronic diseases, corresponding optimal clear weight vectors are trained and obtained, denoted as follows: Used for type 2 diabetes Used for essential hypertension. Used for dyslipidemia Used for obesity It is used for chronic low-grade inflammation-related chronic diseases.

[0066] S203: Based on the type of chronic disease, generate an optimal weight vector of non-nutrient indicators that is suitable for the type of chronic disease according to the optimal weight.

[0067] Among them, chronic diseases include type 2 diabetes, essential hypertension, dyslipidemia, obesity, and chronic low-grade inflammation-related chronic diseases.

[0068] Specifically, for target chronic disease patients, the first step is to determine the type of chronic disease and select the corresponding optimal clear weight vector. Then, based on the patient's complications and pathological state, the weights are recalibrated. This recalibration is achieved by adjusting the constraint range of the fuzzy judgment matrix. The adjusted constraint range is then input into the coupled model to re-solve for the patient's personalized weight vector. , ..., .

[0069] In this embodiment of the invention, a coupled model is constructed by integrating fuzzy hierarchical analysis and the maximum entropy principle. A judgment matrix is ​​constructed using triangular fuzzy numbers and a consistency test is completed. The optimal weight vector is solved with the goal of maximizing the weight distribution entropy. This approach takes into account both expert experience and data objectivity, breaks through the limitations of the fixed weight model, realizes dynamic weight calibration specific to chronic diseases, and can perform secondary personalized calibration by adjusting the coefficient α based on the patient's pathological state.

[0070] S3: Collect multi-dimensional individual data of patients with chronic diseases and perform standardized processing to generate standardized individual datasets.

[0071] The multi-dimensional individual data includes chronic disease clinical diagnosis and treatment data, dietary intake and eating habits data, lifestyle and medication-related data, intervention needs and adherence-related data, and the standardization process includes data cleaning, dimensional standardization and structured decomposition.

[0072] In one possible implementation, S3 specifically includes sub-steps S301 to S305: S301: Collect multi-dimensional individual data of patients with chronic diseases.

[0073] S302: Perform data cleaning operations on multi-dimensional individual data to generate cleaned data.

[0074] Data cleaning operations include removing outliers and invalid values, as well as filling in missing values.

[0075] The data cleaning process includes removing outliers and invalid values, as well as filling in missing values. Specifically, missing values ​​are filled using the mean value matched with characteristics of the same population group or by reasonable logical inference, and abnormal data that exceeds a reasonable range undergoes secondary verification and correction.

[0076] S303: Perform dimensional standardization on the cleaned data to generate standardized data.

[0077] Among them, dimensional standardization operations include min-max standardization or z-score standardization.

[0078] Specifically, the min-max standardization formula is:

[0079] in The original data, The minimum value in the dataset. The maximum value in the dataset. This is the standardized data.

[0080] The z-score standardization formula is:

[0081] in, The mean of the dataset. denoted as the standard deviation of the dataset.

[0082] It should be noted that the data structuring process follows a pre-defined structured template, organizing all data into standardized individual datasets, which are then divided into five subsets: patient basic information subset, clinical indicator subset, dietary data subset, contraindications and preferences subset, and intervention needs subset. This step ultimately outputs a standardized, multi-dimensional individual dataset of the target chronic disease patients, containing these five structured subsets.

[0083] S304: Perform a structured splitting operation on standardized data to generate a subset of structured data.

[0084] The structured data subset includes a subset of basic patient information, a subset of clinical indicators, a subset of dietary data, a subset of contraindications and preferences, and a subset of intervention needs.

[0085] S305: Integrate a subset of structured data into a standardized individual dataset.

[0086] In this embodiment of the invention, by collecting multi-dimensional individual data of patients with chronic diseases, performing data cleaning, dimensional standardization (min-max or z-score), and structured splitting, a standardized individual dataset containing five subsets including basic patient information, clinical indicators, dietary data, contraindications and preferences, and intervention needs is generated, ensuring the accuracy, consistency, and usability of the input data.

[0087] S4: Based on the optimal weight vector of non-nutrient indicators and the standardized individual dataset, generate a personalized non-nutrient indicator weight vector, calculate the baseline dietary family nurse dietary therapy index score, complete the health risk classification and intervention target anchoring, and generate an intervention target list.

[0088] Among them, the personalized non-nutrient index weight vector is a set of weights after secondary calibration of the optimal weight vector based on the patient's chronic disease type, complications and pathological status; the baseline dietary family nurse dietary therapy index score is the patient's comprehensive evaluation value of daily diet; the health risk classification is divided into excellent, good, insufficient and deficient levels according to the family nurse dietary therapy index score; and the intervention target anchoring is to determine the core pathological targets that need intervention by combining clinical abnormal indicators and dietary deficiencies.

[0089] In one possible implementation, S4 specifically includes sub-steps S401 to S404: S401: Generate personalized non-nutrient index weight vectors for patients with chronic diseases based on the optimal weight vector of non-nutrient indicators and standardized individual datasets.

[0090] It should be noted that this personalized weight vector is obtained by secondary calibration based on the optimal weight vector, taking into account the patient's complications and pathological state. Specifically, this is achieved by adjusting the constraint range (adjustment coefficient) of the fuzzy judgment matrix. The coupling model was then re-solved to obtain the result.

[0091] S402: Calculate the baseline dietary home nurse dietary therapy index score based on the personalized non-nutrient index weight vector and combined with the average daily food intake.

[0092] The average daily food intake was derived from a subset of dietary data in a standardized individual dataset.

[0093] Specifically, first calculate the dynamic calibration score of the personalized home nurse dietary therapy index for a single food item:

[0094] in, For the first The personalized home nurse diet therapy index for this type of food is dynamically calibrated. For the first Personalized weights for each primary core indicator; For 100g The first of the food The content values ​​of non-nutrient indicators are 5, and the number of primary core indicators is 5.

[0095] In some embodiments, the formula for calculating the dynamic calibration score of the food-grade home nurse dietary index is as follows:

[0096] in, For the first The first of the food Standardized scores of the three-dimensional comprehensive health effects of each non-nutrient indicator.

[0097] It should be noted that in the general dietary assessment scenario for the entire population, the equal-weighted arithmetic mean method is used to synthesize the standardized scores of the three effect dimensions, generating the standardized score S of the three-dimensional comprehensive health effect of non-nutrients. q,k The calculation formula is:

[0098] in, , , These represent the standardized effect scores of the q-th non-nutrient in the dimensions of anti-inflammatory effect, antioxidant stress effect, and metabolic regulation effect. The standardized score of the overall health effect of the same non-nutrient q is fixed and is independent of the food type k, determined solely by the biological effects of the non-nutrient itself and evidence-based data.

[0099] In some embodiments, for precision intervention scenarios targeting specific chronic diseases, based on the core pathological characteristics of the target chronic disease, the weights of the three major effect dimensions are calibrated using fuzzy hierarchical analysis, and a standardized score for the three-dimensional comprehensive health effect of non-nutrients is synthesized using a dynamic weighting method. The calculation formula is:

[0100] in, , , The calibration weights are respectively for the anti-inflammatory effect, the anti-oxidative stress effect, and the metabolic regulation effect, satisfying... The weights can be adjusted to a secondary personalized calibration based on the target chronic disease type, the patient's individual pathological state, and complications.

[0101] The formula for calculating the dynamic score of the dietary therapy index for home nurses is as follows:

[0102] in, The dynamic score of the patient's daily diet combined with the family nurse's dietary therapy index is calculated. The number of different types of food consumed by the patient. For the first The average daily intake of this type of food.

[0103] S403: Based on the baseline dietary home nurse dietary therapy index score, complete the health risk classification according to the home nurse dietary therapy index classification standard corresponding to chronic diseases.

[0104] The grading criteria are as follows: Excellent (score > 43.3), Good (36.6~43.3), Insufficient (30.0~36.6), and Deficient (≤ 30.0).

[0105] S404: Based on the health risk classification results, and combined with clinical abnormal indicators and dietary deficiencies in standardized individual datasets, anchor personalized core intervention targets and generate an intervention target list.

[0106] Among them, abnormal clinical indicators include indicators related to glucose metabolism, lipid metabolism, inflammation, and oxidative stress. Dietary deficiencies refer to the gap in the intake of core non-nutrients in the patient's current diet and the problem of excessively high proportion of foods with low family nurse dietary therapy index scores.

[0107] Clinically abnormal indicators include glucose metabolism indicators (fasting blood glucose, glycated hemoglobin, fasting insulin, insulin resistance index), lipid metabolism indicators (triglycerides, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol), inflammatory indicators (interleukin-6, high-sensitivity C-reactive protein, tumor necrosis factor-alpha), and oxidative stress indicators (malondialdehyde, superoxide dismutase). Dietary deficiencies refer to a gap in the intake of core non-nutrients in the patient's current diet and an excessively high proportion of foods with low home nurse dietary therapy index scores.

[0108] In this embodiment of the invention, a personalized weight vector exclusive to the patient is generated based on the optimal weight vector and individual data. The family nurse's dietary therapy index score is calculated by combining the average daily food intake with the average daily food intake. The health risk is classified according to the grading standard (excellent >43.3, good 36.6~43.3, insufficient 30.0~36.6, deficient ≤30.0). At the same time, personalized core intervention targets are anchored by combining clinical abnormal indicators and dietary deficiencies, realizing accurate assessment and target identification from the group to the individual.

[0109] S5: Combining basic data, the mapping relationship between non-nutrients and chronic disease intervention effects, standardized individual datasets, personalized non-nutrient index weight vectors, and a list of intervention targets, we complete the screening of plant-derived foods and the verification of synergistic effects of non-nutrients, generating a pool of candidate food combinations.

[0110] Among them, plant-derived food refers to edible materials derived from plants, and the candidate food pool of the composition is a collection of foods that can be used for subsequent ratio optimization after targeted screening and synergistic effect verification.

[0111] In one possible implementation, S5 specifically includes sub-steps S501 to S504: S501: Set multi-dimensional constraints based on the standardized individual dataset and the list of intervention targets.

[0112] The multi-dimensional constraints include target matching constraints, contraindication constraints, safety constraints, compliance constraints, and family nurse dietary therapy index score constraints.

[0113] The multi-dimensional constraints include target matching constraints (foods must be rich in the highest priority core non-nutrients in the intervention target list, with a corresponding effect score ≥0.8), contraindications constraints (excluding allergies, dietary restrictions, and prohibited foods), safety constraints (excluding foods that have adverse effects on comorbidities), compliance constraints (prioritizing foods within the patient's dietary preferences), and family nurse dietary therapy index score constraints (if a single food score reaches a preset qualified threshold, priority should be given to foods that rank highly in the same category).

[0114] S502: Based on multi-dimensional constraints, basic data, personalized non-nutrient index weight vectors, and the mapping relationship between non-nutrients and chronic disease intervention effects, plant-derived foods are subjected to preliminary screening, fine screening, and targeted screening in sequence to obtain food combinations after preliminary screening.

[0115] Among them, the initial screening is used to eliminate foods that are contraindicated or do not meet safety standards, the fine screening is used to screen foods that are rich in core non-nutrients and meet the family nurse dietary therapy index score, and the targeted screening is used to match foods with the highest effect strength ranking corresponding to the intervention target.

[0116] The screening process involves several steps: initial screening to eliminate contraindicated and unsafe foods; fine screening to select foods rich in core non-nutrients and meeting the family nurse dietary index score (retaining the top 10 foods in each category); and targeted screening to match the top 5 foods with the strongest effects corresponding to the intervention target. For example, anti-inflammatory targets correspond to turmeric, onions, and broccoli; blood sugar-lowering targets correspond to blueberries, mulberry leaves, and bitter melon; and antioxidant targets correspond to grapes, purple sweet potatoes, and goji berries.

[0117] S503: Based on the mapping relationship between non-nutrients and chronic disease intervention effects, determine whether there are antagonistic or diminishing effects between non-nutrients in the food combinations after initial screening. If so, eliminate food combinations with antagonistic or diminishing effects. Otherwise, retain the food combinations as candidate combinations.

[0118] Specifically, combinations that reduce the bioavailability of non-nutrients and cancel out each other's intervention effects should be completely eliminated.

[0119] S504: Based on the mapping relationship between non-nutrients and chronic disease intervention effects, screen food combinations with synergistic effects from candidate combinations, and generate a pool of candidate food combinations based on the screened food combinations.

[0120] Furthermore, calculate the synergistic effect magnitude:

[0121] in, To maximize synergistic effects, The overall home nurse dietary therapy index score for food combinations. This represents the number of food types in the food combination. This represents the average of the individual home nurse dietary therapy index scores for each food item in the food combination.

[0122] In this embodiment of the invention, plant-derived foods are subjected to initial screening, fine screening, and targeted screening based on multi-dimensional constraints (target matching, contraindications, safety, compliance, and family nurse dietary therapy index score constraints). Antagonistic combinations are eliminated and synergistic combinations are screened based on the non-nutrient-chronic disease effect mapping relationship. The synergistic effect is quantitatively verified through the formula of synergistic effect magnitude, and a candidate food pool with complete information is generated to ensure the targeting, safety, and synergistic intervention effect of the food.

[0123] S6: Based on the personalized non-nutrient index weight vector, the list of intervention targets, and the candidate food pool for the composition, the composition ratio is optimized through an optimization algorithm to generate the initial composition plan.

[0124] The optimization algorithm is an adaptive multi-objective chaotic particle swarm optimization algorithm, and the composition ratio optimization refers to determining the weight ratio of each food in the candidate food pool in a unit composition.

[0125] In one possible implementation, S6 specifically includes sub-steps S601 to S604: S601: Based on the candidate food pool for the composition, generate an initial set of formulation schemes through chaotic mapping.

[0126] Among them, the chaotic mapping is the Logistic chaotic mapping, which is used to improve the diversity of the initial population.

[0127] Wherein, the chaotic mapping is the Logistic chaotic mapping, and the formula is:

[0128] in For the first Chaotic variables in the next iteration. This is a control parameter, set here. At this point, the Logistic mapping is in a completely chaotic state. .

[0129] Initial position generated based on chaotic variables :

[0130] in, For the first The first particle The initial chaotic variable of dimension is generated by randomly generating initial values. Then, the chaotic variables in the remaining dimensions are obtained by iterating the Logistic mapping.

[0131] S602: Based on the personalized non-nutrient index weight vector and the list of intervention targets, set optimization objectives, and perform multi-objective iterative optimization on the initial set of formulation schemes through adaptive weights and learning factors to generate a set of candidate formulation schemes.

[0132] The optimization objectives include maximizing the dietary therapy index score per unit weight of the composition for home nurses and maximizing the comprehensive effect score of the composition on the core intervention target. The adaptive weight is the inertial weight that is dynamically adjusted with the number of iterations, and the learning factor is the individual experience weight and the group experience weight that are dynamically adjusted with the number of iterations.

[0133] Furthermore, we define two objective functions, the first objective function... Personalized Home Nurse Dietary Therapy Index Score per Unit Weight of Composition:

[0134] in, The first in the candidate food pool The personalized home nurse dietary therapy index for a variety of foods is dynamically calibrated, and the calculation of this score depends on the personalized weights of the S2 output. ,Right now:

[0135] Second objective function The overall effect score of the composition on the patient's core intervention target:

[0136] in, The first in the candidate food pool The comprehensive effect score of a food on the patient's core intervention target is determined by the five-dimensional anchoring map of non-nutrient-chronic disease effects output by S1.

[0137] Next, we define the particle velocity update formula and position update formula, and introduce adaptive inertia weight and adaptive learning factor. Inertia weight... Adaptive adjustment based on iteration count:

[0138] in, The maximum inertia weight is set here. , To minimize inertia weight, it is set here. , This represents the current iteration number. The maximum number of iterations is set here. .

[0139] Learning factor and The algorithm adapts to the number of iterations, focusing on individual experience in the early stages and group experience in the later stages.

[0140]

[0141] in, , , , .

[0142] The speed update formula is:

[0143] in, and A random number between 0 and 1 For the first The first particle The optimal position of an individual in dimension. For the first The globally optimal position of the dimension.

[0144] The position update formula is:

[0145] After the update, the location is normalized to ensure that it meets the requirements. and :

[0146] Finally, the Pareto dominance relation is used to determine the individual optimal position and the global optimal position for two particles. and ,like and If at least one of the inequalities is strictly true, then it is called... Pareto Domination All particles not dominated by other particles are grouped into a Pareto optimal solution set. Three representative particles are selected from the Pareto optimal solution set as the final composite scheme, namely, the strong intervention type, the equilibrium compliance type, and the economical type.

[0147] S603: Select feasible ratio schemes from the candidate ratio scheme set through a penalty function.

[0148] The penalty function is used to penalize and reduce the objective function value of schemes that violate the constraints.

[0149] The process involves defining constraints, including intake constraints, nutritional balance constraints, contraindications and safety constraints, compliance constraints, and synergistic effect constraints. All constraints are derived from the outputs of previous steps. A penalty function method is used to process the constraints, transforming them into penalty terms that are then added to the objective function to form a penalized objective function. and :

[0150]

[0151] in, The penalty coefficient is set here. , The number of constraints, For the first The degree of violation of each constraint condition, if the constraint is satisfied... ,otherwise .

[0152] S604: Based on feasible formulation schemes, generate multiple sets of differentiated initial composition schemes.

[0153] Among them, the multiple differentiated initial combination protocols include the high-potency intervention type, the balanced compliance type, and the cost-effective type.

[0154] Among them, the initial treatment options for the multiple differentiated compositions include a high-potency intervention type (maximum first objective function value) and a balanced compliance type (selected using the ideal point method, with the distance formula as follows:

[0155] The particle with the smallest distance is selected as the equilibrium compliance type, and the particle with the lowest food cost is selected as the economical type. The total cost of each particle is calculated based on the food cost information in the candidate food pool, and the particle with the lowest total cost is selected.

[0156] In this embodiment of the invention, an adaptive multi-objective chaotic particle swarm optimization algorithm is adopted. The population is initialized through Logistic chaotic mapping, and adaptive inertia weights and learning factors are set. Multi-objective iterative optimization is carried out with two objective functions (maximizing the score of the family nurse's dietary therapy index per unit weight and maximizing the comprehensive effect score of the core intervention target). The penalty function is used to handle the constraints. Based on the Pareto dominance relationship, three differentiated initial schemes are selected: a powerful intervention type, a balanced compliance type (ideal point method), and an economical type. This achieves precise optimization of the composition ratio and diverse output.

[0157] S7: Based on the mapping relationship between non-nutrients and chronic disease intervention effects, construct a validation model, input the initial composition scheme into the validation model, complete the validation of intervention effects and safety assessment, and generate the final composition scheme.

[0158] The validation model is a coupled model of Bayesian network and Monte Carlo simulation. The intervention effect validation is to predict the improvement effect of the composition on the core clinical indicators of chronic diseases. The safety assessment includes the risk of non-nutrient excess, the safety of comorbidities, and the risk of drug-food interactions.

[0159] In one possible implementation, S7 specifically includes sub-steps S701 to S704: S701: Based on the mapping relationship between non-nutrients and chronic disease intervention effects, construct a validation model that includes non-nutrient nodes, effect dimension nodes, and clinical indicator nodes.

[0160] Among them, the non-nutrient nodes correspond to core indicators such as anthocyanins, proanthocyanidins, resveratrol, soy isoflavones, and quercetin, while the effect dimension nodes include anti-inflammatory effects, antioxidant stress effects, and metabolic regulation effects. The clinical indicator nodes are set according to the type of chronic disease.

[0161] Specifically, the Bayesian network is a directed acyclic graph, where non-nutrient nodes point to effect dimension nodes, and effect dimension nodes point to clinical indicator nodes.

[0162] Specifically, the structure learning of Bayesian networks uses the K2 algorithm, where the connections between nodes minimize the Bayesian information criterion. The formula for the Bayesian information criterion is:

[0163] in, The log-likelihood function value of the model. The number of parameters in the model. This represents the sample size of the training data.

[0164] Furthermore, after structure learning is completed, parameter learning is performed. The maximum likelihood estimation method is used to learn the parameters of the conditional probability table, providing a probabilistic basis for subsequent Monte Carlo simulations.

[0165] S702: Input the non-nutrient intake in the initial formulation of the composition as evidence into the validation model.

[0166] S703: Obtain the expected distribution of changes in clinical indicators through simulated sampling.

[0167] The simulation sampling is a Monte Carlo simulation, and the expected variation distribution includes the mean, standard deviation, and confidence interval.

[0168] The simulation sampling was a Monte Carlo simulation. The marginal probability distribution of non-nutrient nodes adopted a Gaussian distribution.

[0169] in, For the first Daily intake of non-nutrients This represents the average daily intake of this non-nutrient from the database. Let Variance be the variance.

[0170] For the effect dimension nodes, their conditional probability tables are determined based on the non-nutrient-health effect quantification database output by S1, and fitted using a logistic regression model. The conditional probabilities are:

[0171] in, For the first Each effect dimension node The effect is sufficient. For the intercept term, For the first The first non-nutrient to the first The regression coefficients for each effect dimension were obtained by fitting data from the literature.

[0172] For clinical indicator nodes, their conditional probability tables are determined based on dose-response data from publicly available clinical intervention studies, and fitted using a linear regression model. The expected changes in the clinical indicators are:

[0173] in, For the first The expected change in each clinical indicator; positive values ​​indicate improvement, and negative values ​​indicate deterioration. For the intercept term, For the first The effect on the first The regression coefficients of each clinical indicator The error term follows a mean of 0 and a variance of . The Gaussian distribution.

[0174] S704: Based on the expected change distribution, complete the virtual verification of the intervention effect and safety assessment, and generate the final composition scheme.

[0175] It should be noted that this step also includes a reverse iterative optimization process based on the verification results. Specifically, the verification results are fed back to S2 and S6, the weighting system and constraints are adjusted, and the composition scheme is regenerated and verified until the intervention effect reaches the target. Interactive formula:

[0176] in, To adjust the coefficient, determined from the verification results, if a certain non-nutrient... If the actual effect is higher than expected, then ,otherwise The adjustment range is This step ultimately outputs the optimal, clear weight vectors for different chronic diseases and personalized weight vectors for patients, while also completing two-way interaction.

[0177] Specifically, the interactive formula represents the degree of constraint violation after adjustment:

[0178] in, The constraint relaxation factor is determined by the verification results of S7. Simultaneously, the weights of the objective function are adjusted to form a weighted objective function. :

[0179] in, The weights for the first objective function are determined by the verification results. In Pareto dominance, particles with larger weighted objective function values ​​are given priority.

[0180] Furthermore, the Bayesian network model was validated by comparing the predicted changes in clinical indicators with the actual changes in historical studies, calculating the prediction error, and verifying the model's accuracy. The prediction error was calculated using the mean squared error.

[0181] in, To verify the sample size of the data, For the first The predicted change value for each sample, For the first The actual change value of each sample. If the mean squared error is greater than the preset threshold, then the structure learning and parameter learning are repeated, and the structure and parameters of the Bayesian network are adjusted until the mean squared error is less than the preset threshold.

[0182] Based on the simulation results, the effectiveness of each treatment group in achieving the patient's core intervention goals was evaluated. The intervention effect of each treatment group was quantitatively scored using the following formula:

[0183] in, To quantify the intervention effect, The number of core clinical indicators, For the first The weights of each clinical indicator are determined by the list of intervention targets output by S4. For the first The mean of the expected changes in each clinical indicator For the first Target changes in each clinical indicator. Simultaneously, a comprehensive safety assessment is conducted, including assessments of non-nutrient excess risk, comorbidity safety, drug-food interactions, and allergy and intolerance risks. Based on the assessment results, composition regimens are screened, and those with safety risks are eliminated.

[0184] The formula for calculating the expected dietary home nurse's dietary therapy index score is as follows:

[0185] in, The expected dietary home nurse's dietary therapy index score. The patient's baseline diet and the family nurse's dietary therapy index score. This refers to the patient's average total daily food intake. The comprehensive home nurse dietary therapy index score for the composition. This is the recommended total daily weight of the composition.

[0186] In this embodiment of the invention, a Bayesian network coupled with Monte Carlo simulation is constructed, including non-nutrient nodes (anthocyanins, proanthocyanidins, resveratrol, soy isoflavones, quercetin), effect dimension nodes (anti-inflammatory, antioxidant, metabolic regulation), and clinical indicator nodes. Gaussian distribution, logistic regression, and linear regression are used to fit the conditional probabilities. The expected change distribution of clinical indicators is obtained through multiple simulations (e.g., 10,000 times). The intervention effect is quantified using a scoring formula. At the same time, a two-way interaction mechanism with S2 and S6 is established (adjustment coefficient α∈[-0.2,0.2], relaxation coefficient β∈[0,0.1], weight γ∈[0.3,0.7]) to achieve reverse optimization of weights and constraints based on the validation results until the intervention effect reaches the target (e.g., Score_eff≥80), which significantly improves the safety and clinical efficacy prediction ability of the composition regimen.

[0187] S8: Generate personalized chronic disease intervention strategies based on standardized individual datasets and final composite schemes.

[0188] The personalized chronic disease intervention strategy includes a combination diet plan, guidance on optimizing daily diet, a phased intervention implementation plan, monitoring and follow-up recommendations, and recommendations for coordinated lifestyle interventions.

[0189] In one possible implementation, S8 specifically includes sub-steps S801 to S803: S801: Develop a phased implementation plan based on the standardized individual dataset and the final composite scheme.

[0190] The phased implementation plan includes the frequency of consumption of the composition, the time of consumption, the cooking method, and the phased intervention goals.

[0191] The phased implementation plan includes the frequency of consumption of the combination, the timing of consumption, the cooking method, and the phased intervention goals. For example, the goal for the first month is to increase the dietary therapy index score of the home nurse to above 36.6 and reduce glycated hemoglobin by 0.5%.

[0192] S802: Based on the phased implementation plan, generate dietary optimization guidance, monitoring programs, and lifestyle-related intervention measures.

[0193] The dietary optimization guidance includes recommendations for foods with high dietary value index for home nurses and suggestions for controlling the intake of low-value foods. The monitoring plan includes the time points and frequencies for home monitoring indicators and hospital follow-up indicators.

[0194] The dietary optimization guidance includes recommendations for high-value foods for home nurses (such as blueberries, broccoli, and turmeric) and suggestions for controlling the intake of low-value foods. The monitoring plan includes the time points and frequencies for home monitoring indicators (collecting dietary intake data and home clinical indicators weekly) and hospital follow-up indicators (reviewing glucose and lipid metabolism, inflammation, oxidative stress, and liver and kidney function indicators every 3 months).

[0195] S803: Based on the phased implementation plan, dietary optimization guidance, monitoring program, and lifestyle-related intervention measures, a personalized full-process strategy for chronic disease intervention is formed.

[0196] The personalized chronic disease intervention strategy includes a combination diet plan, guidance on optimizing daily diet, a phased intervention implementation plan, monitoring and follow-up recommendations, and recommendations for coordinated lifestyle interventions.

[0197] It should be noted that, to verify the predictive value of the family nurse dietary therapy index for the intervention effect of chronic diseases, the correlation between changes in patients' family nurse dietary therapy index scores and improvements in clinical indicators can be analyzed during the intervention follow-up phase. The Pearson correlation coefficient can be used to calculate the correlation.

[0198] in, The Pearson correlation coefficient is used. To track the sample size of the data, For the first The dietary therapy index score of the family nurses who were followed up. To track the mean score, For the first Clinical indicator values ​​from the follow-up period. This represents the mean value of the clinical indicators.

[0199] Furthermore, based on the attribution analysis results, the home nurse dietary therapy index model was iteratively calibrated. The fuzzy judgment matrix of the fuzzy hierarchical analysis-maximum entropy principle coupled model was optimized, and the weights of non-nutrients and effect dimensions were adjusted for different chronic diseases and pathological states. The calculation formulas for food-grade and dietary-grade home nurse dietary therapy indices were optimized, and model parameters were improved based on real-world dose-response relationship data. The basic database and the five-dimensional anchoring map of non-nutrient-chronic disease effects were updated simultaneously, supplementing the model with new evidence-based medicine, new data on the non-nutrient content of foods, and new synergistic effect research results.

[0200] It should be noted that the intelligent decision-making optimization rules for the composition were updated based on the attribution analysis results. The food selection rules in S5 were optimized by adjusting the food selection thresholds corresponding to different intervention targets, adding foods with significant actual intervention effects, and removing foods whose actual effects did not meet expectations. The adaptive multi-objective chaotic particle swarm optimization algorithm in S6 was optimized by adjusting the weight allocation of the objective function, adding a compliance-related optimization dimension, and updating the boundary parameters of the constraints. The Bayesian network-Monte Carlo simulation coupling model in S7 was optimized by updating the structure and parameters of the Bayesian network based on real clinical data, and supplementing drug-food interaction data and comorbidity safety constraint data.

[0201] For example, for patients currently undergoing intervention, the effectiveness and goal achievement are evaluated based on full-cycle tracking data. The combination therapy regimen, dietary optimization suggestions, and phased intervention goals are dynamically adjusted based on changes in the family nurse's dietary therapy index score, improvement in clinical indicators, and adherence. This generates a personalized intervention strategy for the next stage, achieving dynamic chronic disease intervention management throughout the patient's lifecycle. The final output includes an intervention effectiveness attribution analysis report, an iteratively calibrated family nurse dietary therapy index model and dynamic weight calibration system, updated intelligent decision-making rules for the entire combination therapy process, and a dynamically optimized intervention strategy for the next stage for the current patient. Simultaneously, the iteratively updated model, rules, and database are fed back to the entire process (S1, S2, S5, S6, S7) for intelligent decision-making for subsequent new patients, forming a complete and continuously optimized methodological closed loop.

[0202] In this embodiment of the invention, a phased implementation plan is formulated based on a standardized individual dataset and a final composite scheme (e.g., the first month's goal: to increase the family nurse's dietary therapy index score to above 36.6 and reduce glycated hemoglobin by 0.5%). This generates dietary optimization guidance (recommendations of high-value foods such as blueberries, broccoli, and turmeric), a monitoring plan (weekly home data collection and hospital follow-up every 3 months), and lifestyle-related intervention measures, forming a comprehensive strategy. Simultaneously, a full-cycle data tracking system is established, using Pearson correlation coefficient analysis to analyze correlations, and attribution analysis to iteratively calibrate the model and optimization rules, thereby achieving continuous dynamic optimization and closed-loop management of personalized interventions throughout the entire process.

[0203] Reference manual attached Figure 2 The diagram shows a schematic of the intelligent decision-making system for family nurse dietary therapy index intervention of chronic diseases provided by the present invention.

[0204] The present invention also provides a composition intelligent decision-making system 20 for home nurse dietary therapy index intervention in chronic diseases, applied to the above-mentioned composition intelligent decision-making method for home nurse dietary therapy index intervention in chronic diseases, including: Processor 201.

[0205] The memory 202 stores computer-readable instructions, which, when executed by the processor 201, implement the intelligent decision-making method for family nurse dietary therapy index intervention of chronic diseases as described in the method embodiment.

[0206] The intelligent decision-making system 20 for family nurse dietary therapy index intervention of chronic diseases provided by the present invention can execute the above-mentioned intelligent decision-making method for family nurse dietary therapy index intervention of chronic diseases and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0207] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0208] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0209] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0210] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0211] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0212] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0213] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0214] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0215] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0216] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0217] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0218] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0219] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the intelligent decision-making method for family nurse dietary therapy index intervention of chronic diseases as described in the method embodiments.

[0220] The present invention provides a computer-readable storage medium that can realize the steps and effects of the intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases in the above-described method embodiments. To avoid repetition, the present invention will not repeat the details.

[0221] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0222] The following points need to be explained: (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0223] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0224] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0225] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A smart decision-making method for family nurses' dietary therapy index intervention in chronic diseases, characterized in that, include: S1: Based on the family nurse dietary therapy index, construct the underlying support database and the non-nutrient-chronic disease effect anchoring map to generate basic data and non-nutrient-chronic disease intervention effect mapping relationship; S2: Through the coupling model, complete the dynamic weight calibration of chronic disease-specific family nurse dietary therapy index and generate the optimal weight vector of non-nutrient indicators. S3: Collect multi-dimensional individual data of patients with chronic diseases and perform standardized processing to generate standardized individual datasets; S4: Based on the optimal weight vector of the non-nutrient indicators and the standardized individual dataset, generate a personalized non-nutrient indicator weight vector, calculate the baseline dietary family nurse dietary therapy index score, complete the health risk classification and intervention target anchoring, and generate an intervention target list. S5: Combining the basic data, the non-nutrient-chronic disease intervention effect mapping relationship, the standardized individual dataset, the personalized non-nutrient index weight vector, and the intervention target list, complete the screening of plant-derived foods and the verification of non-nutrient synergistic effects, and generate a pool of candidate food compositions. S6: Based on the personalized non-nutrient index weight vector, the list of intervention targets, and the candidate food pool of the composition, the composition ratio is optimized through an optimization algorithm to generate an initial composition plan. S7: Based on the non-nutrient-chronic disease intervention effect mapping relationship, construct a verification model, input the initial composition scheme into the verification model, complete the intervention effect verification and safety assessment, and generate the final composition scheme; S8: Based on the standardized individual dataset and the final composite scheme, generate a personalized chronic disease intervention strategy for the entire process.

2. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S1 specifically includes: S101: Based on the family nurse dietary therapy index, construct a basic support database, which includes a food-non-nutrient content database and a non-nutrient-health effect quantification database; S102: Based on the food-nonnutrient content database and the nonnutrient-health effect quantification database, generate the basic data and establish a multi-dimensional mapping relationship between nonnutrients and chronic disease effects to form the nonnutrient-chronic disease effect anchoring map. S103: Based on the non-nutrient-chronic disease effect anchoring map, mark the synergistic and antagonistic effects between non-nutrients, and generate the non-nutrient-chronic disease intervention effect mapping relationship.

3. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S2 specifically includes: S201: Based on the triangular fuzzy number, construct a fuzzy judgment matrix between non-nutrient indicators, and perform a consistency check on the fuzzy judgment matrix. S202: Using a coupled model of fuzzy hierarchical analysis and the principle of maximum entropy, with the goal of maximizing the entropy of the weight distribution, and in combination with constraints, the optimal weights of the non-nutrient indicators are solved. S203: Based on the chronic disease type, generate an optimal weight vector for the non-nutrient index that is adapted to the chronic disease type according to the optimal weight.

4. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S3 specifically includes: S301: Collect the multi-dimensional individual data of the patients with chronic diseases; S302: Perform data cleaning operations on the multi-dimensional individual data to generate cleaned data; S303: Perform a dimensional standardization operation on the cleaned data to generate standardized data; S304: Perform a structured splitting operation on the standardized data to generate a structured data subset; S305: Integrate the structured data subset into the standardized individual dataset.

5. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S4 specifically includes: S401: Based on the optimal weight vector of the non-nutrient indicators and the standardized individual dataset, generate the personalized non-nutrient indicator weight vector for the chronic disease patient. S402: Calculate the baseline dietary family nurse diet therapy index score based on the personalized non-nutrient index weight vector and the average daily food intake. S403: Based on the baseline dietary home nurse dietary therapy index score, complete the health risk classification according to the home nurse dietary therapy index classification standard corresponding to chronic diseases; S404: Based on the health risk classification results, and combined with the clinical abnormal indicators and dietary deficiencies in the standardized individual dataset, anchor personalized core intervention targets and generate the intervention target list.

6. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S5 specifically includes: S501: Based on the standardized individual dataset and the list of intervention targets, set multi-dimensional constraints; S502: According to the multi-dimensional constraints, and based on the basic data, the personalized non-nutrient index weight vector and the non-nutrient-chronic disease intervention effect mapping relationship, the plant-derived foods are subjected to preliminary screening, fine screening and targeted screening in sequence to obtain the food combination after preliminary screening. S503: Based on the non-nutrient-chronic disease intervention effect mapping relationship, determine whether there is an antagonistic or diminishing effect relationship between non-nutrients in the food combinations after initial screening; if so, remove the food combinations with the antagonistic or diminishing effect relationship; otherwise, retain the food combinations as candidate combinations. S504: Based on the non-nutrient-chronic disease intervention effect mapping relationship, screen food combinations with synergistic effects from the candidate combinations, and generate the candidate food pool of the composition based on the screened food combinations.

7. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S6 specifically includes: S601: Based on the candidate food pool of the composition, an initial set of formulation schemes is generated through chaotic mapping; S602: Based on the personalized non-nutrient index weight vector and the list of intervention targets, set optimization objectives, and perform multi-objective iterative optimization on the initial formula scheme set through adaptive weights and learning factors to generate a candidate formula scheme set; S603: Select feasible proportion schemes from the set of candidate proportion schemes using a penalty function; S604: Based on the feasible formulation scheme, generate multiple sets of differentiated initial formulations of the composition.

8. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, Specifically, S7 includes: S701: Based on the non-nutrient-chronic disease intervention effect mapping relationship, construct a validation model that includes non-nutrient nodes, effect dimension nodes, and clinical indicator nodes; S702: Input the non-nutrient intake in the initial formulation of the composition as evidence into the validation model; S703: Obtain the expected distribution of changes in clinical indicators through simulated sampling; S704: Based on the expected change distribution, complete the virtual verification of the intervention effect and safety assessment, and generate the final version of the composition scheme.

9. The intelligent decision-making method for family nurse dietary therapy index intervention in chronic diseases according to claim 1, characterized in that, S8 specifically includes: S801: Based on the standardized individual dataset and the final composite scheme, formulate a phased implementation plan; S802: Based on the phased implementation plan, generate dietary optimization guidance, monitoring programs, and lifestyle intervention measures; S803: Based on the phased implementation plan, the dietary optimization guidance, the monitoring plan, and the lifestyle synergistic intervention measures, a personalized chronic disease intervention strategy is formed.

10. A smart decision-making system for family nurses' dietary therapy index intervention in chronic diseases, characterized in that: include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the intelligent decision-making method for family nurse dietary therapy index intervention of chronic diseases as described in any one of claims 1 to 9.

Images