Method and system for constructing health database and nutrition requirement database

Abstract

The application belongs to the field of health management and nutrition, and discloses a method for constructing a health database and a nutrient requirement database of different populations, which comprises the following steps: constructing a health index and nutrient requirement database of different healthy populations and sub-healthy populations; constructing a health index and nutrient requirement correlation database; formulating a health status evaluation scheme for different populations; and formulating a precise personalized nutrient intervention scheme for specific populations. Through the construction of the health database and the nutrient requirement database of different populations, the application realizes personalized health evaluation and nutrient intervention based on data analysis, provides precise health management and nutrient intervention suggestions for different populations, and thus improves the overall health level of different populations.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of health management and nutrition, and in particular relates to a method and system for constructing a health database and a nutrition requirement database for different populations. BACKGROUND

[0002] Existing health management and nutrient intake recommendations are mainly targeted at a broad population, without fully considering the special needs of different populations (such as pregnant women, infants, the elderly, etc.). Different populations with different ages, genders, health levels, and lifestyles have different health indicators and nutrient requirements. In order to provide more accurate health management and nutrient intake guidance, it is necessary to construct a health database and a nutrient requirement database for different populations, and to develop corresponding modeling and algorithms to achieve personalized health assessment and intervention measures.

[0003] The technical problems existing in the prior art in the field of health management and nutrient intake recommendation mainly include the following points:

[0004] 1. Lack of personalized recommendations: Current systems mainly make recommendations based on broad population data, without fully considering individual differences. Each person's physical condition, nutritional needs, and health status are unique, so a more personalized recommendation system is needed.

[0005] 2. Incomplete database: Existing health databases and nutrient requirement databases may not be comprehensive, especially for specific populations (such as pregnant women, infants, the elderly, etc.). There is insufficient data collection and analysis for these populations. This leads to recommendations for these populations that may not be accurate enough.

[0006] 3. Limitations of modeling and algorithms: Existing modeling and algorithms may not be able to fully capture all variables that affect health and nutrient requirements, or may not be able to accurately predict and recommend based on these variables. In addition, these algorithms may also not be able to handle the sparsity and imbalance of data well.

[0007] 4. Lack of real-time feedback and adjustment: People's physical condition and nutritional needs change over time. Existing systems may not be able to provide real-time feedback and adjustment suggestions to meet the changing needs of users.

[0008] 5. Unfriendly user interface: Some existing health management and nutrient intake recommendation systems may have an unintuitive and user-unfriendly user interface, making it difficult for users to use the system, thereby reducing the practicality and acceptance of the system.

[0009] To address these issues, future technological development directions may include building more comprehensive and detailed databases, developing more advanced modeling and algorithms to provide more accurate personalized recommendations, implementing real-time feedback and adjustment functions for the system, and optimizing the user interface, etc. SUMMARY

[0010] In view of the problems in the prior art, the present application provides a method for constructing a health database and a nutrition demand database for different populations.

[0011] The present application is implemented by a method for constructing a health database and a nutrition demand database for different populations, which comprises:

[0012] S1: constructing a health index and nutrient demand database for different healthy populations and sub-healthy populations;

[0013] S2: constructing a health index and nutrient demand correlation database;

[0014] S3: developing a health status assessment scheme for different populations;

[0015] S4: developing a precise and personalized nutrient intervention scheme for a specific population.

[0016] Further, the S1 comprises data collection and database construction;

[0017] The data collection specifically comprises:

[0018] Data sources:

[0019] (1) medical institutions, health management platforms, questionnaire surveys, personal health records, etc.;

[0020] (2) literature materials from domestic and foreign journals and databases such as CNKI, Wanfang, VIP, ScienceDirect, PubMed, Web of Science, and WILEY, etc.;

[0021] (3) relevant information on official information platforms such as the National Health Commission of China, the Chinese Nutrition Society, etc.;

[0022] Data types: including health indicators such as body weight, blood pressure, blood glucose, physical activity, and sleep, and nutrient demand data;

[0023] Target population: healthy population: women in the preparatory period, pregnant women (early / middle / late), lactating women, infants, children, adolescents, middle-aged people, and the elderly.

[0024] Sub-healthy population: obese population, hypertensive population, diabetic population, tumor disease population, and cardiovascular disease population, etc.

[0025] Further, S2 specifically includes:

[0026] (1) Data collection and organization: In different health levels of the population, statistics and analysis of the types and intake of nutrients affecting health status;

[0027] (2) Data analysis and processing: Using regression analysis, correlation analysis and machine learning algorithm to evaluate the correlation between nutrients and health level;

[0028] (3) Database building: Establish a database containing the correlation data between health indicators and nutrient requirements.

[0029] Further, the database building specifically includes:

[0030] (1) Demand analysis

[0031] Determine the data types and structures that the database needs to store, such as health indicators, nutrient types and intake, collection time, data sources, etc.;

[0032] (2) Database design: table structure design; field definition;

[0033] (3) Database implementation: Select a suitable database management system (such as MySQL, MongoDB), create a database and table;

[0034] (4) Data import: Import the cleaned and processed data into the database.

[0035] Further, S3 specifically includes:

[0036] (1) Correlation analysis of health indicators: Using statistical analysis and machine learning methods, explore the correlation between different health indicator levels and health status, aiming to understand the internal relationship between indicators and provide basis for subsequent personalized assessment model construction;

[0037] (2) Personalized assessment model, this model mainly realizes: based on the analysis results, build a health status assessment model suitable for different populations; the assessment model inputs different health indicator data, and outputs the health assessment results; The personalized assessment model aims to build a customized health status assessment model for different populations based on the analysis results of health indicators. Through inputting different health indicator data, this model can output health assessment results, providing scientific basis for personalized health management.

[0038] Further, the implementation process of the correlation analysis of health indicators specifically includes:

[0039] (1) Feature selection

[0040] LASSO Regression: Select important features by introducing L1 regularization;

[0041] Random Forest: Evaluate the importance of each indicator using feature importance;

[0042] (2) Modeling and Validation

[0043] Model Selection: Choose appropriate models (e.g., linear regression, decision tree, support vector machine);

[0044] Cross-Validation: Split the dataset for model training and validation to ensure the generalization ability of the model;

[0045] (3) Hyperparameter Tuning

[0046] Optimize model parameters using grid search or random search;

[0047] (4) Result Analysis and Reporting

[0048] Correlation Report: Comprehensive description of the correlation between health indicators;

[0049] Feature Importance: List key health indicators and their importance ranking through methods such as LASSO and Random Forest;

[0050] Model Performance: Evaluate the performance of different models in predicting health indicators.

[0051] Further, the implementation process of the personalized evaluation model is as follows:

[0052] (1) Data Clustering

[0053] According to the characteristics of the population (such as age, gender, health status, etc.), the data is clustered to ensure the pertinence of the model;

[0054] (2) Model Selection

[0055] Model Type: Linear regression, decision tree, random forest, support vector machine, neural network, etc.;

[0056] Model selection basis: According to the data characteristics of health indicators and health level, select appropriate model type;

[0057] (3) Model Training

[0058] Divide the training set and test set; train the model and evaluate its performance;

[0059] (4) Model Evaluation

[0060] Evaluate model performance using evaluation metrics (such as accuracy, AUC, F1-score, etc.);

[0061] Cross-validation is used to ensure the generalization ability of the model.

[0062] (5) Hyperparameter Tuning

[0063] Hyperparameters of the model are tuned using grid search and random search to improve model performance.

[0064] Furthermore, the S4 specifically includes:

[0065] Data Integration: Different sources and types of health data and nutrient requirement data are integrated to establish a complete and unified dataset, providing a foundation for subsequent algorithm model development;

[0066] Algorithm Model: Develop algorithms based on machine learning and statistical models to predict the impact of different nutrient intakes on health levels and generate personalized nutrient intervention recommendations.

[0067] Further, the algorithm model algorithm steps are as follows:

[0068] (1) Feature Selection

[0069] Select key health indicators and nutrient data: Select important features through correlation analysis;

[0070] Feature Engineering: Create new features to improve the predictive ability of the model;

[0071] (2) Train Model

[0072] Select machine learning algorithms: such as linear regression, decision tree, random forest, support vector machine, neural network, etc.

[0073] Data segmentation: divide into training set and test set;

[0074] Model training: train the selected machine learning model;

[0075] (3) Model Evaluation

[0076] Performance evaluation: use indicators such as mean square error (MSE), R 2 , etc. to evaluate the performance of the model;

[0077] Cross-validation: evaluate the stability and generalization ability of the model through K-fold cross-validation;

[0078] (4) Generate Personalized Nutrient Intervention Recommendations

[0079] Rule Engine: Establish a rule engine to provide personalized nutrient intervention recommendations based on model prediction results, such as increasing the intake of a certain nutrient;

[0080] Nutrient intervention optimization algorithm: using linear programming or other optimization algorithms to recommend the optimal nutrient intervention plan for users.

[0081] Another object of the present application is to provide a health database and nutrition requirement database for different populations and a modeling system for implementing the construction method of the health database and nutrition requirement database for different populations, which comprises:

[0082] Database building module, obtain health data and nutrient requirement data from hospitals, health management platforms, etc.; remove duplicate and abnormal data to ensure data quality; use MySQL database storage structure to store structured data to realize health indicators and nutrition requirement database for different populations;

[0083] Data collection and import module, collect data through questionnaires, medical records, health management platforms, etc.; clean and organize the collected data and import them into the database;

[0084] Data analysis and modeling module, based on the collected health data, use statistical and machine learning methods to establish health assessment models for different populations; analyze the relationship between nutrients and health level, and establish personalized nutrient requirement models;

[0085] System implementation module, develop front-end interface to display health evaluation results and nutrient requirement suggestions for different populations; build back-end services to provide API interfaces for health assessment and nutrient requirement prediction; continuously optimize algorithms to improve the accuracy of health assessment and nutrient requirement prediction.

[0086] In combination with the above technical solutions and solved technical problems, the technical solutions to be protected by the present application have the following advantages and positive effects:

[0087] First, use literature search, web crawling, information platform search, etc. to obtain raw data and establish food nutrition and human health related databases; design dynamic and process-based database construction algorithm model.

[0088] The advantages and innovations of the dynamic and process-based database construction method are as follows:

[0089] (1) Dynamic

[0090] Real-time update: dynamic database can reflect the latest research results and data changes in real time, ensuring the timeliness and accuracy of the data. This is particularly important in the fields of nutrition and health science, as these fields are rapidly advancing and new discoveries and data are constantly emerging.

[0091] Automatic expansion: As new data is continuously added, the database can automatically expand its storage capacity and structure without human intervention. This greatly improves the flexibility and scalability of the database, allowing it to adapt to growing data demands.

[0092] (2) Processed

[0093] Standardized operations: The process of database construction algorithm model standardizes data acquisition, processing, storage, updating, etc. to ensure that each step follows established rules and processes. This helps improve the efficiency and consistency of database construction.

[0094] Reducing human error: Through the process of operation, the possibility of human intervention and error is reduced, improving the accuracy and reliability of the data. This is particularly important for scientific research and data analysis, as incorrect data can lead to incorrect conclusions and decisions.

[0095] (3) Innovation

[0096] Technology integration: The invention integrates literature search, web crawling, information platform search and other data acquisition methods to achieve comprehensive and diverse data. This technological integration is an innovative attempt in the field of database construction, which can more comprehensively cover relevant nutrition and health information.

[0097] Algorithm optimization: The dynamic and process-oriented database construction algorithm model optimizes data processing and storage processes, improving the efficiency and stability of the database. This algorithm optimization is an innovation in the field of database technology, which helps to promote the development of related technologies.

[0098] User experience: Dynamic and process-oriented database construction not only improves the accuracy and reliability of data, but also enhances user experience. Users can more conveniently access the latest data and information, improving the practicality of the database and user satisfaction.

[0099] Second, by segmenting different populations, personalized health assessment and personalized nutrient intervention strategies are carried out, so as to provide more accurate health management and dietary recommendations.

[0100] The advantages and innovations of segmenting different populations, personalized health assessment and personalized nutrition intervention are as follows:

[0101] (1) Segmenting different populations

[0102] By segmenting different populations, such as age, gender, weight, height, lifestyle, medical history, etc., the invention can more accurately identify individual health needs and risks. This segmentation makes health assessment and nutrition recommendations more personalized, avoiding "one-size-fits-all" universal recommendations, thereby improving the relevance and effectiveness of health management.

[0103] (2) Personalized health assessment

[0104] Real-time monitoring: Utilizing modern technology such as wearable devices and mobile applications, the invention can monitor users' health conditions such as heart rate, blood pressure, and activity level in real-time. This dynamic assessment enables more timely and accurate health management, allowing for timely detection of health problems and the implementation of appropriate interventions.

[0105] Scientific analysis: Based on a large amount of health data and nutritional data, statistical and machine learning methods are used for scientific analysis and prediction. This data-driven evaluation method can provide more objective and accurate health assessment results, ensuring the scientificity and effectiveness of nutritional recommendations.

[0106] (3) Personalized nutrient intervention

[0107] Customized nutrition recommendations: Considering the differences in nutritional needs and health conditions among individuals, the invention can provide customized nutrition recommendations based on individual circumstances. This personalized nutrition recommendation is more in line with individual actual needs, helping to achieve better health management results.

[0108] Third, individualization: By segmenting different populations, individualized health evaluation criteria and nutrient requirements are developed, providing more accurate health management and dietary recommendations.

[0109] Data-driven: Based on a large amount of real data for analysis and modeling, the scientificity and effectiveness of health assessment and nutrient intervention recommendations are ensured.

[0110] Scalability: The database and algorithm model design has good scalability, which can be continuously optimized and upgraded according to user needs and the development of new technologies.

[0111] The purpose of the invention is:

[0112] (1) Provide health indicators and nutrient requirement databases for different populations (including healthy and sub-healthy populations), providing data support for personalized health management and nutrient intake guidance.

[0113] (2) Develop health status assessment programs and nutrient intervention measures for different populations through statistical analysis and modeling.

[0114] The invention achieves personalized health assessment and nutrient intervention based on data analysis by constructing health databases and nutrient requirement databases for different populations, providing precise health management and nutrient intervention recommendations for different populations, thereby improving the overall health level of different populations.

[0115] Fourth, as the creative evidence of the invention's claims, it is also reflected in the following important aspects:

[0116] (1) The expected income and commercial value of the technical solution of the invention after transformation are:

[0117] The expected income and commercial value of the "dynamic and process-based database construction method"

[0118] Scientific research support:

[0119] The dynamic and process-based database provides rich data resources for the field of nutrition and health science, which helps to promote scientific research and technological innovation in related fields. Researchers can more conveniently obtain the required data and accelerate the research process.

[0120] Health assessment and disease prevention:

[0121] The database can be used for health assessment and disease prevention, providing scientific data support. By analyzing the relationship between food nutrition and human health, it can provide decision-making basis for health managers and policymakers, and improve public health level.

[0122] Economic benefits:

[0123] By optimizing the database construction process, it reduces manual intervention and errors, and reduces operating costs. At the same time, the high efficiency and reliability of the database also provide economic value for related enterprises and institutions, and improve their competitiveness.

[0124] Sustainable development:

[0125] The dynamic and process-based database construction model has the characteristics of sustainable development. With the passage of time and the progress of technology, the database can be continuously updated and expanded to adapt to changing needs and challenges, maintaining its long-term value and application potential.

[0126] The expected income and commercial value of "dividing different groups of people, and conducting personalized health assessment and personalized nutrient intervention strategies"

[0127] Improve the effect of health management:

[0128] Through personalized health assessment and nutrient intervention, it can more accurately identify and manage health risks, and improve the overall effect of health management. This helps to prevent diseases, promote health, and improve the quality of life.

[0129] Promote nutritional balance:

[0130] Personalized nutrition recommendations help users achieve nutritional balance and avoid excessive or insufficient nutrition. This is of great significance to maintaining physical health and preventing chronic diseases.

[0131] Improve user satisfaction:

[0132] Personalized health assessment and nutrition recommendations can better meet the needs of users, improve user satisfaction and trust. Users are more willing to accept and follow these recommendations, thereby improving the compliance of health management.

[0133] Support scientific research:

[0134] The implementation and optimization of personalized health assessment and nutrient intervention models provide rich data and cases for scientific research. This helps to promote research in the fields of health science and nutrition, and promotes the development and application of related theories.

[0135] Economic benefits:

[0136] By improving the effectiveness of health management and user satisfaction, personalized health assessment and nutrient intervention models can bring significant economic benefits. This not only includes reducing medical costs, improving work efficiency, but also improving the overall quality of life of users.

[0137] (2) The technical solution of the present application solves the technical problems that people have been eager to solve but have failed to succeed:

[0138] The present application provides a solution for building and modeling a health database and nutrient demand database for different populations with real-time, personalized and systematic characteristics.

[0139] The present application realizes the construction of a health index-nutrient demand correlation database from dynamic and process-based, to the provision of customized nutrition recommendations through personalized health assessment models and nutrient intervention models. Users can monitor their health status in real time, such as heart rate, blood pressure, activity level, etc. The present application can more accurately identify and manage health risks based on these health data of users, and provide personalized nutrient recommendations.

[0140] The above solution applies the latest research results of nutrition and health to health assessment and disease prevention, while personalized health assessment and nutrition recommendations can better meet the needs of users, improve user satisfaction and trust. Users are more willing to accept and follow these recommendations, thereby improving the compliance of health management. BRIEF DESCRIPTION OF DRAWINGS

[0141] Figure 1 is the overall process logic diagram provided by the embodiment of the present application;

[0142] Figure 2 is the database logical structure diagram provided by the embodiment of the present application;

[0143] Figure 3is a flow logic diagram of a health assessment model and a nutrient intervention recommendation model provided by an embodiment of the present application;

[0144] Figure 4 is a health data (part of the data) collection schematic diagram of a pre-pregnancy population provided by an embodiment of the present application;

[0145] Figure 5 is a schematic diagram of establishing a health index and nutrient requirement database provided by an embodiment of the present application;

[0146] Figure 6 is a weight health data (part of the data) schematic diagram of a pre-pregnancy population provided by an embodiment of the present application;

[0147] Figure 7 is a web page display schematic diagram provided by an embodiment of the present application;

[0148] Figure 8 is a disease population classification algorithm model implementation schematic diagram provided by an embodiment of the present application. DETAILED DESCRIPTION

[0149] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0150] As shown in Figure 1 , an embodiment of the present application provides a method for constructing a health database and a nutrient requirement database of different populations, which comprises:

[0151] S1: constructing a health index and nutrient requirement database of different healthy populations and sub-healthy populations;

[0152] S2: constructing a health index and nutrient requirement correlation database;

[0153] S3: developing a health status assessment scheme for different populations;

[0154] S4: developing a precise personalized nutrient intervention scheme for a specific population.

[0155] As shown in Figure 2 , the S1 comprises data collection and database construction;

[0156] The data collection specifically comprises:

[0157] Data sources:

[0158] (1) medical institutions, health management platforms, questionnaire surveys, personal health records, etc.;

[0159] (2) CNKI, Wanfang, VIP, ScienceDirect, PubMed, Web of Science, and WILEY, etc. domestic and foreign journals and database literature;

[0160] (3) Related information on official information platforms such as the National Health Commission of China, the Chinese Nutrition Society, etc.

[0161] Data types: including health indicators such as weight, blood pressure, blood sugar, physical activity, sleep, and nutrient requirement data;

[0162] Target population: healthy population: women preparing for pregnancy, pregnant women (early / mid / late), lactating women, infants, children, adolescents, middle-aged people, and the elderly.

[0163] Sub-health population: obese population, hypertensive population, diabetic population, tumor disease population, cardiovascular disease population, etc.

[0164] Further, the S2 specifically includes:

[0165] (1) Data collection and sorting: in different health levels of population, statistics and analysis of the types and intake of nutrients affecting health status;

[0166] (2) Data analysis and processing: using regression analysis, correlation analysis and machine learning algorithm to evaluate the correlation between nutrients and health level;

[0167] (3) Database building: building a database containing health indicators and nutrient requirement correlation data.

[0168] Further, the database building specifically includes:

[0169] (1) Demand analysis

[0170] Determine the data types and structures that the database needs to store, such as health indicators, nutrient types and intake, collection time, data sources, etc.

[0171] (2) Database design: table structure design; field definition;

[0172] (3) Database implementation: select a suitable database management system (such as MySQL, MongoDB), create a database and a table;

[0173] (4) Data import: import the cleaned and processed data into the database.

[0174] Further, the S3 specifically includes:

[0175] (1) Correlation analysis of health indicators: Using statistical analysis and machine learning methods, explore the correlation between different health indicators and health status, aiming to understand the internal relationship between indicators, and provide basis for subsequent personalized assessment model construction;

[0176] (2) Personalized assessment model, which mainly realizes: based on the analysis results, construct health status assessment model suitable for different population; input different health indicators data into the assessment model, output health assessment results; personalized assessment model aims to construct customized health status assessment model for different population based on the analysis results of health indicators, through inputting different health indicators data, the model can output health assessment results, and provide scientific basis for personalized health management.

[0177] Further, the implementation process of the correlation analysis of health indicators specifically includes:

[0178] (1) Feature selection

[0179] LASSO regression: by introducing L1 regularization, screen important features;

[0180] Random forest: use feature importance to evaluate the importance of each indicator;

[0181] (2) Modeling and verification

[0182] Model selection: select appropriate model (such as linear regression, decision tree, support vector machine);

[0183] Cross validation: split the data set for model training and verification, to ensure the generalization ability of the model;

[0184] (3) Hyperparameter tuning

[0185] Use grid search or random search to optimize model parameters;

[0186] (4) Result analysis and report

[0187] Correlation report: comprehensive description of the correlation between health indicators;

[0188] Feature importance: list the key health indicators and their importance ranking through methods such as LASSO and random forest;

[0189] Model performance: evaluate the performance of different models in predicting health status indicators.

[0190] Further, the implementation process of the personalized assessment model is as follows:

[0191] (1) Data clustering

[0192] Data grouping based on population characteristics (e.g. age, gender, health status, etc.) to ensure model targeting;

[0193] (2) Model selection

[0194] Model type: linear regression, decision tree, random forest, support vector machine, neural network, etc.

[0195] Model selection basis: select appropriate model type based on data characteristics of health indicators and health level;

[0196] (3) Model training

[0197] Divide training set and test set; train model and evaluate its performance;

[0198] (4) Model evaluation

[0199] Evaluate model performance using evaluation indicators (e.g. accuracy, AUC, F1-score, etc.);

[0200] Ensure model generalization ability through cross-validation;

[0201] (5) Hyperparameter tuning

[0202] As shown in Figure 3 , use grid search and random search to optimize model hyperparameters to improve model performance.

[0203] Further, the S4 specifically includes:

[0204] Data integration: integrate different sources and types of health data and nutrient requirement data to establish a complete and unified data set, providing a foundation for subsequent algorithm model development;

[0205] Algorithm model: develop algorithms based on machine learning and statistical models to predict the impact of different nutrient intakes on health level and generate personalized nutrient intervention recommendations.

[0206] Further, the algorithm model algorithm steps are as follows:

[0207] (1) Feature selection

[0208] Select key health indicators and nutrient data: select important features through correlation analysis;

[0209] Feature engineering: create new features to improve model prediction ability;

[0210] (2) Train model

[0211] Select machine learning algorithm: such as linear regression, decision tree, random forest, support vector machine, neural network, etc.

[0212] Data splitting: divided into training set and test set;

[0213] Model training: training the selected machine learning model;

[0214] (3) Model evaluation

[0215] Performance evaluation: evaluate the model performance using metrics such as mean squared error (MSE), R 2 square, etc.

[0216] Cross-validation: evaluate the stability and generalization ability of the model through K-fold cross-validation;

[0217] (4) Generate personalized nutrient intervention recommendations

[0218] Rule engine: establish a rule engine to provide personalized nutrient intervention recommendations based on model prediction results, such as increasing the intake of a certain nutrient;

[0219] Nutrient intervention optimization algorithm: use linear programming or other optimization algorithms to recommend the optimal nutrient intervention plan for users.

[0220] The embodiment of the present application provides a health database and nutrition demand database of different populations and a modeling system for implementing the construction method of the health database and nutrition demand database of different populations, which comprises:

[0221] Database building module, obtain health data and nutrient demand data from hospitals, health management platforms, etc.; remove duplicate and abnormal data to ensure data quality; use MySQL database storage structure to store structured data, realize health indicators and nutrition demand database of different groups;

[0222] Data collection and import module, collect data through questionnaires, medical records, health management platforms, etc.; clean and organize the collected data and import them into the database;

[0223] Data analysis and modeling module, based on the collected health data, use statistical and machine learning methods to establish health assessment models for different populations; analyze the relationship between nutrients and health level, and establish personalized nutrient demand model;

[0224] System implementation module, develop front-end interface to display health evaluation results and nutrient demand recommendations for different populations; build back-end services to provide API interfaces for health assessment and nutrient demand prediction; continuously optimize algorithms to improve the accuracy of health assessment and nutrient demand prediction.

[0225] Example 1: Health assessment based on database

[0226] Suppose we have collected a large amount of health examination data of people of different ages, including blood pressure, blood sugar, blood lipid, etc. Through data cleaning and analysis, health evaluation standards for different age groups are established. When a user uploads his health examination data to the system, the system automatically evaluates the health status according to his age group and provides personalized health advice.

[0227] Example 2: Personalized nutrient intervention recommendations

[0228] By analyzing the diet records and health status of a certain health level population, it is found that certain nutrients (such as vitamin D, unsaturated fatty acids) have a significant impact on their health status. Based on this, a nutrient intervention recommendation algorithm is established. When the user provides his daily diet record and health goal, the system can provide personalized nutrient intervention recommendations based on the analysis results.

[0229] First, through different domestic and foreign journals, databases and various health data platforms, etc. Data sources to obtain health indicators and nutrient demand related data materials, Figure 4 is the health data (part) of the pre-pregnancy population.

[0230] Then, through the process of data mining algorithm, data cleaning, data structure design, data storage, dynamically and procedurally build the health indicators and nutrient demand database of different healthy and sub-healthy populations. Figure 5 is the database interface of health indicators and nutrient demand, Figure 6 is the weight health data (part of the data) of the pre-pregnancy population in the database.

[0231] Figure 7 The data display module includes a web display interface. Users input personal data through the health indicator collector and disease condition collector, and the data processing module generates personalized nutrient intake plan according to the input data, and displays the intake plan to the user through the web display interface.

[0232] Figure 8 is the decision tree and logistic regression used for multi-label classification to classify disease populations: according to the collected diabetes data set for training, input some physical examination indicators of patients such as triglyceride, etc. Output whether the patient has diabetes and high blood pressure.

[0233] It should be noted that embodiments of the present application can be realized by hardware, software, or a combination of software and hardware. The hardware portion can be realized by a special logic; the software portion can be stored in a memory and executed by a proper instruction execution system, such as a microprocessor or a specially designed hardware. A person of ordinary skill in the art can understand that the above-mentioned apparatus and method can be realized by computer executable instructions and / or included in processor control codes, such as a carrier medium, such as a magnetic disk, CD or DVD-ROM, a programmable memory, such as a read-only memory (firmware), or a data carrier, such as an optical or electronic signal carrier. The apparatus of the present application and its modules can be realized by a hardware circuit, such as a very large scale integrated circuit or a gate array, a semiconductor, such as a logic chip, a transistor, or a programmable hardware device, such as a field programmable gate array, a programmable logic device, or the like, by software executed by various types of processors, or by a combination of the above-mentioned hardware circuit and software, such as firmware.

[0234] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any modification, equivalent replacement, and improvement within the technical range disclosed by the present application, and within the spirit and principle of the present application, should be covered within the protection scope of the present application.

Claims

1. A method for constructing a health database and a nutritional requirement database for different population groups, characterized in that, The method comprises: S1: constructing a health index and nutrient requirement database of different healthy and sub-healthy populations; S2: constructing a health index and nutrient requirement correlation database; S3: developing a health status assessment scheme for different populations; S4: developing a precise and personalized nutrient intervention scheme for specific populations; The S1 comprises data collection and database construction; The data collection specifically comprises: Data sources: (1) medical institutions, health management platforms, questionnaire surveys, and personal health records; (2) literature materials from domestic and foreign journals and databases of CNKI, Wanfang, VIP, ScienceDirect, PubMed, Web of Science, and WILEY; (3) relevant information on official information platforms of the National Health Commission of China and the Chinese Nutrition Society; Data types: including body weight, blood pressure, blood glucose, physical activity, sleep health indicators, and nutrient requirement data; Target population: healthy population: women preparing for pregnancy, pregnant women, lactating women, infants, children, adolescents, middle-aged people, and the elderly; Sub-healthy population: obese population, hypertensive population, diabetic population, tumor disease population, and cardiovascular disease population; The S2 specifically comprises: (1) Data collection and sorting: in different health levels of population, the types and intake of nutrients affecting health status are counted and analyzed; (2) Data analysis and processing: regression analysis, correlation analysis, and machine learning algorithm are used to evaluate the correlation between nutrients and health level; (3) Database construction: a database containing health index and nutrient requirement correlation data is established; The database construction specifically comprises: (1) Demand analysis Determine the data types and structures that the database needs to store, including health indicators, nutrient types and intake, collection time, and data sources; (2) Database design: table structure design; field definition; (3) Database implementation: select a suitable database management system, create a database and tables; (4) Data import: import the cleaned and processed data into the database; The S3 specifically comprises: (1) Correlation analysis of health indicators: statistical analysis and machine learning methods are used to explore the correlation between different health indicator levels and health degree, aiming to understand the internal relationship between indicators and provide basis for subsequent personalized assessment model construction; (2) Personalized assessment model, which mainly realizes: based on the analysis results, a health status assessment model suitable for different populations is constructed; the assessment model inputs different health indicator data and outputs health assessment results; the personalized assessment model aims to construct a customized health status assessment model for different populations based on the analysis results of health indicators, and through inputting different health indicator data, the model can output health assessment results, providing a scientific basis for personalized health management.

2. The method of claim 1, wherein the database of health and nutritional needs of different groups of people is constructed by, The implementation process of the correlation analysis of health indicators specifically comprises: (1) Feature selection LASSO regression: important features are selected by introducing L1 regularization; Random forest: the importance of each indicator is evaluated using feature importance; (2) Modeling and verification Model selection: Choose appropriate models; Cross-validation: Split the dataset for model training and validation to ensure the generalization ability of the model; (3) Hyperparameter tuning Optimize model parameters using grid search or random search; (4) Result analysis and reporting Correlation report: Comprehensive description of the correlation between health indicators; Feature importance: List the key health indicators and their importance ranking through LASSO and random forest methods; Model performance: Evaluate the performance of different models in predicting health indicators.

3. The method of claim 1, wherein the database of health and nutritional needs of different groups of people is constructed by collecting and analyzing the health and nutritional needs of different groups of people. The implementation process of the personalized evaluation model is as follows: (1) Data clustering According to the characteristics of the population, the data is clustered to ensure the pertinence of the model; (2) Model selection Model type: Linear regression, decision tree, random forest, support vector machine, neural network; Model selection basis: According to the data characteristics of health indicators and health level, select the appropriate model type; (3) Model training Divide the training set and test set; train the model and evaluate its performance; (4) Model evaluation Evaluate the performance of the model using evaluation indicators; Ensure the generalization ability of the model through cross-validation; (5) Hyperparameter tuning Optimize the hyperparameters of the model using grid search and random search to improve the performance of the model.

4. The method for constructing health databases and nutritional needs databases for different population groups as described in claim 1, characterized in that, The S4 specifically includes: Data integration: Integrate different sources and types of health data and nutrient demand data to establish a complete and unified dataset, providing a foundation for subsequent algorithm model development; Algorithm model: Develop algorithms based on machine learning and statistical models to predict the impact of different nutrient intakes on health levels and generate personalized nutrient intervention recommendations.

5. The method of claim 4, wherein the database of health and nutritional needs of different groups of people is constructed by: The algorithm model algorithm steps are as follows: (1) Feature selection Select key health indicators and nutrient data: Select important features through correlation analysis; Feature engineering: Create new features to improve the predictive ability of the model; (2) Train the model Select machine learning algorithms: such as linear regression, decision tree, random forest, support vector machine, neural network; Data segmentation: Divide into training set and test set; Model training: Train the selected machine learning model; (3) Model evaluation Performance evaluation: Use metrics such as mean squared error, R 2 to evaluate the model performance; Cross-validation: Evaluate the stability and generalization ability of the model through K-fold cross-validation; (4) Generate personalized nutrient intervention recommendations Rule engine: Establish a rule engine to provide personalized nutrient intervention recommendations based on model prediction results, such as increasing the intake of a certain nutrient; Nutrient intervention optimization algorithm: Use linear programming or other optimization algorithms to recommend the optimal nutrient intervention scheme for users.

6. A database of health and nutritional needs of different groups of people and a modeling system for the database, which implement the method for constructing the database of health and nutritional needs of different groups of people according to any one of claims 1 to 5, characterized in that, The system includes: Database building module, obtain health data and nutrient demand data from hospitals and health management platforms; remove duplicate and abnormal data to ensure data quality; use MySQL database to store structured data to realize the health indicators and nutrient demand database of different groups; Data collection and import module, collect data through questionnaires, medical records, and health management platform channels; clean and organize the collected data and import it into the database; Data analysis and modeling module: Based on the collected health data, statistical and machine learning methods are used to establish health assessment models for different populations. The relationship between nutrients and health levels is analyzed to establish personalized nutrient demand models. System implementation module: Develop front-end interfaces to display health evaluation results and nutrient demand recommendations for different populations. Build back-end services to provide API interfaces for health assessment and nutrient demand prediction. Continuously optimize algorithms to improve the accuracy of health assessment and nutrient demand prediction.

Images