Precise intervention treatment system and method based on personalized nutrition metabolism evaluation

Abstract

The invention belongs to the technical field of medical health, and provides a precise intervention treatment system and method based on personalized nutrition metabolism assessment, and the system comprises a data collection module which is used for collecting the physiological data, nutrition intake data, gene information, microbiome data and lifestyle data of an individual; the metabolism evaluation module is used for inputting the data acquired by the data acquisition module into a metabolism evaluation model based on deep learning; the intervention scheme generation module is used for evaluating a result and an individual health target based on the metabolic condition; through joint cooperation of the data acquisition module, the metabolism evaluation module and the intervention scheme generation module, in combination with deep learning and a multi-objective optimization algorithm, the system can provide a more personalized and scientific health management scheme according to the specific condition of an individual, the health management effect and the success rate of disease intervention are improved, and the system is suitable for popularization and application. And powerful support is provided for individual health management and disease prevention.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical health, and specifically relates to a precise intervention treatment system and method based on personalized nutritional metabolism assessment. Background Art

[0002] With the improvement of people's living standards and the enhancement of health awareness, personalized nutrition management has gradually become a research hotspot in the field of health management. However, there are many challenges in the process of formulating existing personalized nutrition programs. First of all, data collection is not comprehensive enough, often relying only on individual physiological data or simple nutritional intake records, ignoring the important influence of genetic information, microbiome data, and lifestyle data on individual nutritional metabolism. Secondly, existing intervention plans often lack a dynamic adjustment mechanism and cannot be adjusted in a timely manner according to the real-time feedback of individuals, thus affecting the health management effect and the success rate of disease intervention.

[0003] In view of the above problems, existing personalized nutrition management systems and methods have certain limitations. On the one hand, traditional data collection methods are single, making it difficult to obtain comprehensive individual health data, resulting in inaccurate evaluation results; on the other hand, traditional intervention plan generation methods often only consider a single goal, lack the ability of multi-objective optimization, and lack a dynamic adjustment mechanism, unable to meet the health needs of individuals at different stages.

[0004] Therefore, those skilled in the art have proposed a precise intervention treatment system and method based on personalized nutritional metabolism assessment to solve the problems raised in the background art. Summary of the Invention

[0005] Based on this, in view of the above technical problems, the purpose of the present invention is to provide a precise intervention treatment system and method based on personalized nutritional metabolism assessment, so as to achieve precise assessment and dynamic adjustment of individual nutritional needs, and improve the health management effect and the success rate of disease intervention.

[0006] In the first aspect, the present application provides a precise intervention treatment system based on personalized nutritional metabolism assessment, including:

[0007] A data collection module for collecting individual physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data;

[0008] A metabolic assessment module for inputting the data collected by the data collection module into a metabolic assessment model based on deep learning to obtain an assessment result of an individual's metabolic status;

[0009] An intervention plan generation module, configured to construct a multi-objective optimization model based on the metabolic status assessment result and the individual's health goals, and solve the multi-objective optimization model through a genetic algorithm to obtain an optimal intervention plan, where the optimal intervention plan includes nutritional intake suggestions, lifestyle adjustment suggestions, and medication treatment suggestions.

[0010] In one embodiment, the data acquisition module includes a data preprocessing subunit, configured to:

[0011] Clean and standardize the collected physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data to obtain normalized data;

[0012] Extract features from the normalized data to obtain a comprehensive feature vector including physiological features, nutritional intake features, genetic features, microbiome features, and lifestyle features;

[0013] Perform dimensionality reduction on the comprehensive feature vector to obtain a refined feature vector, which is used to reduce data redundancy and improve the efficiency of subsequent processing.

[0014] In one embodiment, the metabolic assessment module includes a metabolic assessment analysis subunit, configured to:

[0015] Input the refined feature vector into a deep learning-based metabolic assessment model, and extract a high-level abstract representation of the features through a convolutional neural network to obtain deep features;

[0016] Perform a non-linear transformation on the deep features through a fully connected layer to obtain a metabolic status assessment result, which is used to characterize the individual's metabolic health status.

[0017] In one embodiment, the intervention plan generation module includes a multi-objective optimization model construction subunit, configured to:

[0018] Determine an objective function based on meeting the individual's health goals and the metabolic status assessment result, where the objective function includes a nutritional intake optimization goal, a lifestyle adjustment goal, and a medication treatment optimization goal;

[0019] Construct constraint conditions according to nutritional intake constraints, lifestyle adjustment constraints, and medication treatment constraints, where the constraint conditions are used to ensure the feasibility of the intervention plan;

[0020] Obtain a multi-objective optimization model through the objective function and the constraint conditions.

[0021] In one embodiment, the constraint conditions are:

[0022] min(N n )≤N n (t)≤max(N n), where N n (t) is the intake of the nth nutrient at time t, n is the nutrient number, and its value range is from 1 to M; t is time, and its value range is from 1 to T; min(N n ) is the minimum intake of the nth nutrient, and max(N n ) is the maximum intake of the nth nutrient;

[0023] 0 ≤ L(t) ≤ 1, where L(t) is the lifestyle adjustment variable at time t, 0 means no adjustment, and 1 means adjustment;

[0024] min(D) ≤ D(t) ≤ max(D), where D(t) is the drug dose at time t, min(D) is the minimum drug dose, and max(D) is the maximum drug dose.

[0025] In one embodiment, the intervention plan generation module further includes a model solving sub-unit, which is used for:

[0026] Initializing the genetic algorithm parameters and randomly generating an initial population based on the multi-objective optimization model, where each individual in the initial population is used to represent an intervention plan;

[0027] Decoding the intervention plan according to each individual in the initial population, calculating the health improvement effect and side effects of the intervention plan, and combining the constraint penalty term for constraint to obtain the fitness value of each individual, and the fitness value is used to evaluate the intervention plan corresponding to the individual;

[0028] Updating the next-generation individuals in the population according to the fitness values of each individual, and dynamically adjusting the crossover probability and mutation probability to generate a new generation of population;

[0029] Performing an iterative optimization process based on the new generation of population to obtain the optimal population, where the excellent individuals of the previous generation of population are used for the population update of each generation;

[0030] Constructing the optimal intervention plan for each period through the analysis of the optimal individual in the optimal population;

[0031] Generating real-time guidance suggestions according to the optimal intervention plan to obtain the final precise intervention treatment plan.

[0032] In one embodiment, the calculation formula of the fitness value is:

[0033] where f1(x) is the nutritional intake optimization objective function, f2(x) is the lifestyle adjustment objective function, f3(x) is the drug treatment optimization objective function, α, β, and γ are the weights of each objective function respectively, λ is the weight of the constraint penalty term, and penalty(t) is the constraint penalty value at time t.

[0034] In a second aspect, the present application also provides a precise intervention treatment method based on personalized nutritional metabolism assessment, and this method includes:

[0035] S1. Collect physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data of an individual;

[0036] S2. Input the collected data into a metabolic assessment model based on deep learning to obtain an assessment result of the individual's metabolic status;

[0037] S3. Based on the metabolic status assessment result and the individual's health goals, construct a multi-objective optimization model, and solve the multi-objective optimization model through a genetic algorithm to obtain an optimal intervention plan, where the optimal intervention plan includes nutritional intake suggestions, lifestyle adjustment suggestions, and drug treatment suggestions.

[0038] In a third aspect, the present application also provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes, it implements the steps in the first aspect.

[0039] In a fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps in the first aspect.

[0040] The above-mentioned precise intervention treatment system and method based on personalized nutritional metabolism assessment, computer device, and storage medium, through the joint cooperation of a data collection module, a metabolic assessment module, and an intervention plan generation module, combined with deep learning and multi-objective optimization algorithms, effectively solve the problems existing in the existing personalized nutrition programs; specifically, the data collection module can ensure the comprehensiveness and accuracy of the data by comprehensively collecting multi-dimensional data of an individual; the metabolic assessment module can accurately assess the individual's metabolic status through a metabolic assessment model based on deep learning, thereby providing a basis for the subsequent generation of an intervention plan; the intervention plan generation module can generate an optimal intervention plan by constructing a multi-objective optimization model and using a genetic algorithm for solution, thereby achieving precise intervention and dynamic adjustment of an individual; compared with traditional methods, this system can provide a more personalized and scientific health management plan according to the specific situation of an individual, improve the effect of health management and the success rate of disease intervention, and provide strong support for realizing individual health management and disease prevention.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The present invention comprehensively collects an individual's physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data through a data collection module, and uses a data preprocessing subunit to clean, standardize, and extract features from the data, effectively reducing data redundancy and improving the efficiency and accuracy of subsequent processing.

[0043] 2. The present invention adopts a metabolic assessment model based on deep learning. It extracts high-level abstract representations of features through a convolutional neural network and uses a fully connected layer for non-linear transformation to obtain an assessment result of an individual's metabolic status. This model can fully mine the deep features in the data and improve the accuracy of metabolic assessment.

[0044] 3. The present invention constructs a multi-objective optimization model based on the assessment result of an individual's metabolic status and health goals, and solves the model through a genetic algorithm to obtain an optimal intervention plan. This plan includes nutritional intake suggestions, lifestyle adjustment suggestions, and drug treatment suggestions, which can meet the health needs of individuals at different stages. At the same time, the present invention also provides a dynamic adjustment mechanism that can timely adjust the intervention plan according to the individual's real-time feedback data to ensure the effectiveness and adaptability of the intervention plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic structural diagram of a precise intervention treatment system based on personalized nutritional metabolism assessment provided by an exemplary embodiment of the present invention;

[0046] Figure 2 It is a flowchart of a precise intervention treatment method based on personalized nutritional metabolism assessment provided by an exemplary embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] To more clearly illustrate the embodiments of the present application, the following will be combined with the attached Figure 1 and the attached Figure 2 The specific implementation manners of the precise intervention treatment system and method based on personalized nutritional metabolism assessment will be described in detail. The present invention relates to the fields of medical health and computer technology. In particular, by comprehensively collecting multi-dimensional data of individuals, combining deep learning and multi-objective optimization algorithms, it realizes the precise assessment and dynamic adjustment of individual nutritional needs, and improves the effect of health management and the success rate of disease intervention.

[0048] The system structure of the present invention is as Figure 1As shown, it includes a data acquisition module, a metabolic assessment module, and an intervention plan generation module. The data acquisition module is used to collect an individual's physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data; through a variety of sensors and devices, such as body fat scales, blood glucose meters, blood pressure monitors, genetic detectors, gut microbiota detectors, etc., various health data of the individual can be comprehensively and accurately collected; in addition, lifestyle data of the individual, such as exercise volume, sleep quality, eating habits, etc., are collected through questionnaires or smart wearable devices to ensure the comprehensiveness and accuracy of the data.

[0049] The data acquisition module also includes a data preprocessing subunit for preprocessing the collected data; the preprocessing process mainly includes data cleaning and standardization processing; data cleaning refers to removing invalid or abnormal data points, such as missing values, outliers, etc., to ensure the integrity and reliability of the data; standardization processing refers to uniformly processing data from different sources so that they are on the same scale and within the same range, facilitating subsequent model training and evaluation; the specific steps are as follows:

[0050] S1.1. Data cleaning: Preprocess the physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data, and identify and remove invalid or abnormal data points; for example, for physiological data, such as blood glucose values, blood pressure values, etc., by setting reasonable thresholds, remove data points outside the threshold range; for genetic information, such as SNP data, through quality control standards, remove low-quality SNP sites.

[0051] S1.2. Data standardization: Standardize the cleaned data so that they are on the same scale and within the same range; specific methods can use Z-score standardization or Min-Max standardization; for example, for blood glucose values, Z-score standardization can be used, and the formula is: where x is the blood glucose value, μ is the average value of the blood glucose value, and σ is the standard deviation of the blood glucose value; for diet frequency data, Min-Max standardization can be used, and the formula is: where x is the original data, x′ is the standardized data, and min(x) and max(x) are the minimum and maximum values of the data respectively.

[0052] S1.3. Feature extraction: Extract features from the standardized data to obtain a comprehensive feature vector including physiological features, nutritional intake features, genetic features, microbiome features, and lifestyle features. For example, for physiological data, physiological features can be extracted by calculating indicators such as daily average blood glucose value and blood pressure value; for genetic data, genetic features can be extracted by calculating indicators such as gene frequency and gene mutation probability; for microbiome data, microbiome features can be extracted by calculating indicators such as intestinal flora diversity and flora relative abundance; for lifestyle data, lifestyle features can be extracted by calculating indicators such as daily exercise amount and sleep time.

[0053] S1.4. Dimensionality reduction processing: Perform dimensionality reduction processing on the comprehensive feature vector to obtain a concise feature vector, so as to reduce data redundancy and improve the efficiency of subsequent processing. The specific method can use principal component analysis (PCA) or independent component analysis (ICA). For example, for the comprehensive feature vector, use PCA for dimensionality reduction, and the formula is: Y = XW, where X is the original feature vector, W is the principal component analysis matrix, and Y is the feature vector after dimensionality reduction.

[0054] The metabolic assessment module is used to input the concise feature vector after dimensionality reduction processing into a metabolic assessment model based on deep learning to obtain an assessment result of an individual's metabolic status. The metabolic assessment model mainly includes two parts: a convolutional neural network (CNN) and a fully connected layer (FC). The specific steps are as follows:

[0055] S2.1 Model construction: Construct a metabolic assessment model based on deep learning, which mainly includes a convolutional neural network and a fully connected layer. The convolutional neural network is used to extract high-level abstract representations of the concise feature vector, and the fully connected layer is used to perform non-linear transformation on the high-level features to obtain an assessment result of the metabolic status.

[0056] S2.2. Data input: Input the concise feature vector after dimensionality reduction processing into the metabolic assessment model. For example, input the concise feature vector y including features such as blood glucose value, blood pressure value, gene frequency, and intestinal flora diversity into the model.

[0057] S2.3 Feature extraction: Extract high-level abstract representations of the concise feature vector through a convolutional neural network to obtain deep features. The formula of the convolutional neural network is: z = ReLU(W * y + b), where W is the convolutional kernel, * is the convolutional operation, b is the bias term, and ReLU is the activation function.

[0058] S2.4. Nonlinear transformation: Perform a nonlinear transformation on the deep features through a fully connected layer to obtain the metabolic status assessment result. The formula for the fully connected layer is: p = soft max(W1z + b1), where W1 is the weight matrix, b1 is the bias term, and softmax is the activation function. The obtained assessment result p is used to characterize the metabolic health status of an individual.

[0059] The intervention plan generation module is used to construct a multi-objective optimization model based on the metabolic status assessment result and the individual's health goals, and solve the multi-objective optimization model through a genetic algorithm to obtain the optimal intervention plan. The specific steps are as follows:

[0060] S3.1. Data input: Input the metabolic status assessment result p obtained by the metabolic assessment module and the individual's health goals into the intervention plan generation module. The individual's health goals can include specific goals such as reducing blood sugar, controlling weight, and lowering blood pressure.

[0061] S3.2. Objective function construction: Determine the objective function of the multi-objective optimization model according to the individual's health goals and the metabolic status assessment result. The objective function includes the nutrition intake optimization goal, the lifestyle adjustment goal, and the drug treatment optimization goal. For example, for the goal of reducing blood sugar, the objective function can be expressed as: where G(t) is the blood sugar value at time t, and G target (t) is the target blood sugar value at time t.

[0062] S3.3. Constraint condition construction: Construct constraint conditions according to the nutrition intake constraints, lifestyle adjustment constraints, and drug treatment constraints. The constraint conditions ensure the feasibility of the intervention plan. For example, the nutrition intake constraint can be expressed as: min(N n ) ≤ N n (t) ≤ max(N n ), where N n (t) is the intake of the nth nutrient at time t, and min(N n ) and max(N n ) are the minimum and maximum intakes of the nth nutrient respectively. The lifestyle adjustment constraint can be expressed as: 0 ≤ L(t) ≤ 1, where L(t) is the lifestyle adjustment variable at time t, 0 means no adjustment, and 1 means adjustment. The drug treatment constraint can be expressed as: min(D) ≤ D(t) ≤ max(D), where D(t) is the drug dose at time t, and min(D) and max(D) are the minimum and maximum doses of the drug respectively.

[0063] S3.4. Multi-objective optimization model construction: Construct a multi-objective optimization model through the objective function and the constraint conditions. For example, the multi-objective optimization model can be expressed as:

[0064] minimize(f1(x), f2(x), f3(x));

[0065] subject to (min(N n ) ≤ N n (t) ≤ max(N n ))), 0 ≤ L(t) ≤ 1, min(D) ≤ D(t) ≤ max(D));

[0066] where f1(x) is the objective function for reducing blood sugar, f2(x) is the objective function for controlling body weight, and f3(x) is the objective function for reducing blood pressure;

[0067] S3.5. Genetic algorithm initialization: Initialize the genetic algorithm parameters and randomly generate the initial population based on the multi-objective optimization model; Each individual in the initial population represents an intervention plan, such as different combinations of nutritional intake, lifestyle adjustments, and drug treatment plans; The specific steps are as follows:

[0068] S3.51 Initialize the population: Randomly generate the initial population, and each individual represents an intervention plan; For example, generate 100 individuals, and each individual contains the nutrient intake N n (t), the lifestyle adjustment variable L(t), and the drug dose D(t);

[0069] S3.52 Parameter setting: Set the parameters of the genetic algorithm, such as population size, crossover probability, mutation probability, number of iterations, etc.; For example, set the population size to 100, the crossover probability to 0.8, the mutation probability to 0.1, and the number of iterations to 1000 times;

[0070] S3.6 Model solution: Solve the multi-objective optimization model through the genetic algorithm to obtain the optimal intervention plan;

[0071] S3.7 Fitness value calculation: According to each individual in the initial population, decode to obtain the intervention plan, and calculate the health improvement effect and side effects of the intervention plan, and combine the constraint penalty term for constraint to obtain the fitness value of each individual; The fitness value is used to evaluate the intervention plan corresponding to the individual; For example, calculate the fitness value of each intervention plan as: where α, β, and γ are the weights of each objective function respectively, and λ is the weight of the constraint penalty term;

[0072] S3.8 Population update: Update the next-generation individuals in the population according to the fitness values of each individual, and dynamically adjust the crossover probability and mutation probability to generate a new generation of population; For example, select individuals with higher fitness values as parents, perform crossover and mutation operations to generate new-generation individuals;

[0073] S3.9 Iterative Optimization: Based on the new generation of population, an iterative optimization process is carried out. The excellent individuals of the previous generation are used for the population update in each generation. For example, through multiple generations of iteration, the individuals in the population are gradually optimized, and finally the optimal population is obtained.

[0074] S3.10 Optimal Solution Generation: By analyzing the optimal individuals in the optimal population, the optimal intervention solutions for each period are constructed. For example, select the individual with the highest fitness value and decode to obtain the optimal nutrient intake, lifestyle adjustment variables, and drug dosage at time t.

[0075] S3.11 Real-time Guidance Suggestions: According to the optimal intervention solutions, real-time guidance suggestions are generated to obtain the final precise intervention treatment plan. For example, generate daily nutrition intake suggestions, exercise suggestions, and drug treatment suggestions, and push them to the individual in real time through a mobile application or smart wearable device.

[0076] In a specific embodiment, it is assumed that individual A has health goals of reducing blood sugar, controlling weight, and lowering blood pressure. The data acquisition module comprehensively acquires the physiological data (such as blood sugar value, blood pressure value, body fat percentage, etc.), nutritional intake data (such as the amount of various nutrients ingested daily), genetic information (such as SNP data), microbiome data (such as gut microbiota diversity), and lifestyle data (such as daily exercise amount, sleep time, etc.) of individual A through devices such as a body fat scale, blood glucose meter, sphygmomanometer, genetic detector, and gut microbiota detector.

[0077] The data preprocessing subunit cleans and standardizes the acquired data to obtain normalized data. For example, remove the outliers in the blood sugar value and perform Z-score standardization to obtain the standardized blood sugar value; perform quality control on the gene data to remove low-quality SNP sites; calculate the gut microbiota diversity of the microbiome data to obtain the standardized diversity value; perform Min-Max standardization on the lifestyle data to obtain the standardized exercise amount and sleep time.

[0078] The feature extraction subunit extracts features from the normalized data to obtain a comprehensive feature vector. For example, calculate physiological features such as the daily average blood sugar value, blood pressure value, and body fat percentage; calculate the amount of various nutrients ingested daily to obtain nutritional intake features; calculate gene features such as gene mutation probability and SNP frequency; calculate microbiome features such as gut microbiota diversity and relative abundance of the flora; calculate lifestyle features such as daily exercise amount and sleep time.

[0079] The dimensionality reduction processing subunit performs dimensionality reduction on the comprehensive feature vector to obtain a refined feature vector. For example, use the PCA dimensionality reduction method to reduce the comprehensive feature vector from 100 dimensions to 50 dimensions.

[0080] The metabolic assessment module inputs the reduced feature vector after dimensionality reduction into a deep learning-based metabolic assessment model to obtain the metabolic status assessment result of Individual A; the convolutional neural network extracts the high-level abstract representation of the reduced feature vector to obtain deep features; the fully connected layer performs a non-linear transformation on the deep features to obtain the metabolic status assessment result; for example, the assessment result is that Individual A has a high blood sugar level, normal blood pressure and weight, but low gut microbiota diversity.

[0081] Based on the metabolic status assessment result and the health goals of Individual A, the intervention plan generation module constructs a multi-objective optimization model and solves the multi-objective optimization model through a genetic algorithm to obtain the optimal intervention plan; the specific steps are as follows:

[0082] S3.01. Construction of the objective function: According to the health goals and metabolic status assessment result of Individual A, determine the objective function of the multi-objective optimization model; for example, the objective function for reducing blood sugar is: The objective function for controlling weight is: The objective function for reducing blood pressure is: where G(t), W(t), and B(t) are the blood sugar value, weight, and blood pressure value at time t, respectively, and G target (t), W target (t), and B target (t) are the target blood sugar value, target weight, and target blood pressure value at time t, respectively;

[0083] S3.02. Construction of the constraint conditions: Construct the constraint conditions according to the nutritional intake constraint, lifestyle adjustment constraint, and drug treatment constraint; for example, the nutritional intake constraint is: min(N n ) ≤ N n (t) ≤ max(N n ), the lifestyle adjustment constraint is: 0 ≤ L(t) ≤ 1, and the drug treatment constraint is: min(D) ≤ D(t) ≤ max(D);

[0084] S3.03. Construction of the multi-objective optimization model: Construct the multi-objective optimization model through the objective function and the constraint conditions; for example, the model can be expressed as:

[0085] minimize(f1(x), f2(x), f3(x));

[0086] subject to (min(N n ) ≤ N n (t) ≤ max(N n ), 0 ≤ L(t) ≤ 1, min(D) ≤ D(t) ≤ max(D));

[0087] S3.04. Genetic Algorithm Initialization: Initialize the parameters of the genetic algorithm, set the population size to 100, the crossover probability to 0.8, the mutation probability to 0.1, and the number of iterations to 1000; randomly generate the initial population, where each individual represents an intervention plan, including the nutrient intake, lifestyle adjustment variables, and drug dosage at time t;

[0088] S3.05. Model Solving: Solve the multi-objective optimization model through the genetic algorithm to obtain the optimal intervention plan;

[0089] S3.06. Fitness Value Calculation: Calculate the fitness value of each intervention plan. For example, the fitness value is:

[0090] S3.07. Population Update: Update the individuals in the population according to the fitness value, select the individuals with higher fitness values as the parents, and perform crossover and mutation operations to generate a new generation of population;

[0091] S3.08. Iterative Optimization: Gradually optimize the individuals in the population through multiple generations of iteration to finally obtain the optimal population;

[0092] S3.09. Optimal Plan Generation: Select the individual with the highest fitness value and decode it to obtain the optimal nutrient intake, lifestyle adjustment variables, and drug dosage at time t;

[0093] S3.101. Real-time Guidance Suggestions: Generate real-time guidance suggestions according to the optimal intervention plan. For example, generate daily nutrition intake suggestions (reduce carbohydrate intake and increase dietary fiber intake), exercise suggestions (perform at least 30 minutes of moderate-intensity aerobic exercise per day and at least 5 times a week), and drug suggestions (take 1000 IU of vitamin D and 10 g of unsaturated fatty acids per day). According to the severity of the metabolic risk and the individual's lifestyle preferences, rank the diet suggestions and drug suggestions first and the exercise suggestions second; during the implementation of the intervention plan, the individual uploads feedback data such as physiological index changes, diet records, and exercise records in real time through a mobile application. When specifically implementing, update the metabolic assessment results according to the real-time feedback data. If the individual's blood glucose level and blood lipid level decrease significantly, the individual can adjust the intervention plan, reduce the carbohydrate intake suggestion, increase the protein and fat intake suggestions, and at the same time reduce the exercise frequency and time to keep the individual's metabolic level within the normal range.

[0094] As Figure 2 shown, the present application also provides a precise intervention treatment method based on personalized nutrition metabolism assessment, and this method includes:

[0095] S1. Collect the physiological data, nutrition intake data, gene information, microbiome data, and lifestyle data of the individual;

[0096] S2. Input the collected data into a deep learning-based metabolic assessment model to obtain the assessment result of an individual's metabolic status;

[0097] S3. Based on the metabolic status assessment result and the individual's health goals, construct a multi-objective optimization model, and solve the multi-objective optimization model through a genetic algorithm to obtain an optimal intervention plan, where the optimal intervention plan includes nutritional intake suggestions, lifestyle adjustment suggestions, and drug treatment suggestions.

[0098] The present invention also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes it, the steps in the above-mentioned precise intervention treatment system based on personalized nutritional metabolism assessment are implemented. Specifically, the computer device can be a personal computer, a smart phone, a server, etc. The memory can be a hard disk, a solid-state drive, a memory, etc., and the processor can be a central processing unit (CPU), a graphics processing unit (GPU), etc. The computer program stored in the memory includes subroutines such as a data processing module, a model training module, an intervention plan generation module, and a dynamic adjustment module. When the processor runs, it calls these subroutines to implement functions such as data processing, model training, intervention plan generation, and dynamic adjustment. For example, the data processing module is responsible for collecting and preprocessing multi-dimensional data, the model training module is responsible for training and updating the metabolic assessment model, the intervention plan generation module is responsible for generating personalized nutritional intervention plans, and the dynamic adjustment module is responsible for adjusting the intervention plan according to real-time feedback data.

[0099] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned precise intervention treatment system based on personalized nutritional metabolism assessment are implemented. Specifically, the computer-readable storage medium can be an optical disc, a disk, a USB flash drive, a hard disk, etc. The computer program stored in the storage medium includes subroutines such as a data processing module, a model training module, an intervention plan generation module, and a dynamic adjustment module. When the processor runs, it calls these subroutines to implement functions such as data processing, model training, intervention plan generation, and dynamic adjustment. For example, the data processing module is responsible for collecting and preprocessing multi-dimensional data, the model training module is responsible for training and updating the metabolic assessment model, the intervention plan generation module is responsible for generating personalized nutritional intervention plans, and the dynamic adjustment module is responsible for adjusting the intervention plan according to real-time feedback data.

[0100] In summary, the present invention provides a precise intervention treatment system and method based on personalized nutritional metabolism assessment. Through the collaborative work of a data collection module, a metabolism assessment module, an intervention plan generation module, and a dynamic adjustment module, combined with deep learning and a dynamic adjustment mechanism, it effectively solves the problems of incomplete data collection, inaccurate assessment methods, and lack of dynamic adjustment in traditional personalized nutrition programs. This system and method can comprehensively analyze, precisely assess, and dynamically adjust individual characteristics in different application scenarios, thereby improving the effectiveness of health management and the success rate of disease intervention. For example, in diabetes management, this system can generate personalized diet, exercise, and medication recommendations through comprehensive data collection and precise assessment, and dynamically adjust according to the individual's real-time feedback data to ensure the effectiveness and adaptability of the intervention plan, thereby significantly reducing the individual's blood sugar level and diabetes risk. In obesity management, this system can generate personalized diet, exercise, and medication recommendations in a similar manner, and dynamically adjust according to the individual's real-time feedback data to ensure the effectiveness and adaptability of the intervention plan, thereby significantly reducing the individual's weight and obesity risk. In hypertension management, this system can generate personalized diet, exercise, and medication recommendations through comprehensive data collection and precise assessment, and dynamically adjust according to the individual's real-time feedback data to ensure the effectiveness and adaptability of the intervention plan, thereby significantly reducing the individual's blood pressure and hypertension risk.

[0101] The embodiments of the present invention are not limited to the above specific descriptions and can also be extended and applied in other similar application scenarios. For example, in chronic kidney disease management, this system can generate personalized diet, exercise, and medication recommendations by collecting an individual's genetic data, microbiome data, physiological index data, eating habit data, and lifestyle data, and dynamically adjust according to the individual's real-time feedback data, thereby significantly improving the individual's renal function and reducing the incidence of chronic kidney disease.

[0102] The embodiments of the present invention are given for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A precise intervention treatment system based on personalized nutritional metabolic assessment, characterized by: The system comprises: A data collection module is used to collect individual physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data; A metabolic assessment module, which is used to input the data collected by the data acquisition module into a metabolic assessment model based on deep learning to obtain an individual's metabolic status assessment result; The intervention plan generation module is used to build a multi-objective optimization model based on the metabolic status assessment results and the individual's health goals, and solve the multi-objective optimization model through a genetic algorithm to obtain the optimal intervention plan. The optimal intervention plan includes nutritional intake recommendations, lifestyle adjustment recommendations, and drug treatment recommendations.

2. A precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in claim 1, characterized in that: The data acquisition module includes a data preprocessing subunit, which is used to: Clean and standardize the collected physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data to obtain standardized data; Perform feature extraction on the normalized data to obtain a comprehensive feature vector including physiological features, nutritional intake features, gene features, microbiome features, and lifestyle features; The comprehensive feature vector is subjected to dimensionality reduction processing to obtain a simplified feature vector.

3. A precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in claim 1, characterized in that: The metabolic assessment module includes a metabolic assessment analysis subunit, which is used to: The streamlined feature vector is input into the metabolic assessment model based on deep learning, and the high-level abstract representation of the feature is extracted through the convolutional neural network to obtain the deep-level feature; The deep features are transformed nonlinearly through the fully connected layer to obtain the metabolic status assessment results.

4. A precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in claim 1, characterized in that: The intervention plan generation module includes a multi-objective optimization model building subunit, which is used to: Determine the objective function based on meeting the individual's health goals and metabolic status assessment results, the objective function includes nutrition intake optimization goals, lifestyle adjustment goals and drug treatment optimization goals; Construct constraints based on nutritional intake constraints, lifestyle modification constraints, and medication constraints; Through the objective function and constraints, a multi-objective optimization model is obtained.

5. A precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in claim 4, characterized in that: The constraints are: min(N n )≤N n (t)≤max(N n ), where N n (t) is the intake of the nth nutrient at time t, n is the nutrient number, ranging from 1 to M; t is time, ranging from 1 to T; min(N n ) is the minimum intake of the nth nutrient, max(N n ) is the maximum intake of the nth nutrient; 0≤L(t)≤1, where L(t) is the lifestyle adjustment variable at time t, 0 means no adjustment, and 1 means adjustment; min(D)≤D(t)≤max(D), where D(t) is the drug dose at time t, min(D) is the minimum dose of the drug, and max(D) is the maximum dose of the drug.

6. A precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in claim 1, characterized in that: The intervention plan generation module further includes a model solving subunit, which is used to: Initialize genetic algorithm parameters and randomly generate initial populations based on a multi-objective optimization model, where each individual in the initial population is used to represent an intervention plan; The intervention plan is obtained by decoding each individual in the initial population, and the health improvement effect and side effects of the intervention plan are calculated. The constraint penalty term is combined to constrain and obtain the fitness value of each individual; According to the fitness value of each individual, the next generation of individuals in the population is updated, and the crossover probability and mutation probability are dynamically adjusted to generate a new generation of population; An iterative optimization process is performed based on the new generation population to obtain the optimal population, in which each generation of population is updated using the excellent individuals of the previous generation population; By analyzing the optimal individuals in the optimal population, the optimal intervention plan for each period is constructed; Based on the optimal intervention plan, real-time guidance suggestions are generated to obtain the final precise intervention treatment plan.

7. A precise intervention treatment system based on personalized nutritional metabolic assessment according to claim 6, characterized in that: The calculation formula of the fitness value is: Where f1(x) is the objective function for optimizing nutritional intake, f2(x) is the objective function for lifestyle adjustment, f3(x) is the objective function for optimizing drug treatment, α, β, and γ are the weights of each objective function, λ is the weight of the constraint penalty term, and penalty(t) is the constraint penalty value at the tth moment.

8. A precise intervention treatment method based on personalized nutritional metabolic assessment, characterized in that: Applicable to a precise intervention treatment system based on personalized nutritional metabolic assessment as claimed in any one of claims 1 to 7, the method comprising: S1. Collect individual physiological data, nutritional intake data, genetic information, microbiome data, and lifestyle data; S2. Input the collected data into a metabolic assessment model based on deep learning to obtain an individual's metabolic status assessment result; S3. Based on the metabolic status assessment results and individual health goals, a multi-objective optimization model is constructed, and the multi-objective optimization model is solved by genetic algorithm to obtain the optimal intervention plan. The optimal intervention plan includes nutritional intake recommendations, lifestyle adjustment recommendations, and drug treatment recommendations.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the system according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the system according to any one of claims 1 to 7 is implemented.

Images