Long-term healthy diet recommendation method, device and system and medium

Abstract

The invention relates to the technical field of artificial intelligence, in particular to a long-term healthy diet recommendation method, device and system and a medium, and the method comprises the steps: obtaining the current original data of a user, carrying out the preprocessing of the original data, and generating a health data vector of the user; dividing the life cycle into different life stages, and determining a standard nutritional requirement vector of each life stage; determining a life stage of the user, and generating a personalized nutritional demand vector of the user based on the standard nutritional demand vector of the life stage and the health data vector of the user; acquiring historical data of the user, generating a comprehensive feature vector based on the historical data, and generating a prediction result of the current life stage of the user based on the personalized nutritional demand vector and the comprehensive feature vector; generating a current diet planning scheme of the user based on the personalized nutritional demand vector of the user in the current life stage; according to the invention, continuous optimization and evolution of long-term healthy diet planning can be realized.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, system and medium for recommending a long-term healthy diet. Background Art

[0002] Most of the current mainstream diet planning methods on the market are based on static rules or short-term health assessments, which have obvious limitations. Traditional solutions are unable to cope with the changing nutritional needs of users throughout their life cycle, mainly reflected in the following aspects: First, it's difficult to balance individual differences with specific needs. Adolescents, middle-aged adults, and the elderly differ significantly in their physiological characteristics, metabolic rates, and nutrient absorption capacity. Existing methods fail to fully account for these stage-specific characteristics, making it difficult to provide precise dietary recommendations.

[0003] Second, the ability to predict health risks is insufficient. Existing methods rely primarily on immediate data and lack forward-looking risk predictions based on long-term health data and historical statistics. This makes it difficult to promptly detect and intervene in potential health risks, often requiring responses only after problems emerge.

[0004] Third, dynamic planning has limitations. Because users’ physiology and lifestyle habits are constantly changing, static diet plans cannot be continuously optimized and adjusted, making it difficult to adapt to the dynamic needs of users at different stages of their lives.

[0005] In view of this, there is an urgent need for an innovative intelligent algorithm that can combine the characteristics of users' life stages, use knowledge graphs to predict health risks, and continuously optimize diet plans through large-scale artificial intelligence models to meet users' healthy diet needs throughout their life cycle from young to old. Summary of the Invention

[0006] In view of this, the purpose of the embodiments of the present invention is to provide a long-term healthy diet recommendation method, device, system and medium to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0007] In one aspect, an embodiment of the present invention provides a method for recommending a long-term healthy diet, the method comprising the following steps: Obtain the user's current raw data, pre-process the raw data, and generate the user's health data vector; Divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; Determining the user's life stage, and generating a personalized nutritional requirement vector for the user based on a standard nutritional requirement vector for the life stage and the user's health data vector; Obtain the user's historical data, generate a comprehensive feature vector based on the historical data, and generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; Generate the user's current diet plan based on the user's personalized nutritional needs vector at the current stage of life.

[0008] Optionally, obtaining the user's original data, preprocessing the original data, and generating the user's health data vector includes: Obtaining the user's original data in the current sampling period, the original data including basic data and dietary health data; the dietary health data including dietary records, daily exercise volume, work and rest time, and physical examination report; The dietary health data is formed into a time series, and the time series is preprocessed through feature engineering to generate the user's health data vector.

[0009] Optionally, dividing the life cycle into different life stages and determining the standard nutritional requirement vector for each life stage includes: Divide the life cycle into different stages based on age, gender, physiological characteristics and health status; For each life stage, a set of standard nutritional requirement parameters are pre-set to form a standard nutritional requirement vector.

[0010] Optionally, the calculation formula of the user nutritional requirement vector is: ; in: represents the personalized nutritional requirement vector, represents the standard nutritional requirement vector, represents the health data vector, represents the life stage of the i-th user; represents life stage specific standard nutrient weights, Represents the adjustment weight based on individual health status, ensuring that the model can reflect the user's unique physiological and lifestyle differences.

[0011] Optionally, acquiring the user's historical data, generating a comprehensive feature vector based on the historical data, and generating a prediction result of the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector include: Obtain historical data of users in past sampling periods and extract time series features from the historical data; A knowledge graph is constructed by combining historical user data. Based on the association information between nodes in the knowledge graph, the user's nutritional intake, lifestyle habits and potential disease risks are associated to form a comprehensive feature vector. The nodes of the knowledge graph include nutritional ingredients, disease risks and lifestyle habits; the edges represent the association information between nodes. Generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector: the prediction result includes a risk prediction value, and nutrients and / or living habits that contribute significantly to the risk prediction value.

[0012] Optionally, the risk prediction value is calculated as follows:

[0013] in, Indicates the risk prediction value of the user's current life stage: represents the risk prediction model, Indicates the user's recent time period The comprehensive feature vector within A personalized nutritional requirement vector for the user's current life stage.

[0014] Optionally, the method further includes: Collecting user monitoring data during the current sampling period; the monitoring data includes physiological indicators, exercise records, and user feedback on diet satisfaction; Standardize and vectorize the monitoring data to form a feedback vector; The pre-trained large model is used to process the current diet plan, risk prediction value and feedback vector to generate a new diet plan.

[0015] In another aspect, an embodiment of the present invention provides a long-term healthy diet recommendation device, comprising: The first module is used to obtain the user's current raw data, pre-process the raw data, and generate the user's health data vector; The second module is used to divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; The third module is used to determine the life stage of the user and generate a personalized nutritional requirement vector for the user based on the standard nutritional requirement vector of the life stage and the health data vector of the user; The fourth module is used to obtain the user's historical data, generate a comprehensive feature vector based on the historical data, and generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; The fifth module is used to generate the user's current diet plan based on the user's personalized nutritional requirement vector at the current stage of life.

[0016] In another aspect, an embodiment of the present invention provides a long-term healthy diet recommendation system, comprising: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0017] On the other hand, an embodiment of the present invention provides a computer-readable storage medium storing a program executable by a processor. When the program is executed by the processor, it is used to perform the above method.

[0018] The embodiments of the present invention include the following beneficial effects: This invention uses historical health data and statistical models to construct a knowledge graph, enabling accurate predictions of potential future health risks. Furthermore, based on user feedback on physiological changes and lifestyle habits at different life stages, a large-scale artificial intelligence model is used to automatically generate and adjust long-term dietary plans. This not only meets the personalized nutritional needs of users at different life stages, but also possesses forward-looking risk intervention capabilities, dynamically adapting to physiological changes in users. This allows for dietary evolutionary planning throughout the entire life cycle, from youth to old age, providing comprehensive and precise guidance for users' long-term healthy diets. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 This is a schematic flow chart of the steps of a method for recommending a long-term healthy diet provided by an embodiment of the present invention; Figure 2 This is a structural block diagram of a long-term healthy diet recommendation device provided by an embodiment of the present invention; Figure 3 This is a structural block diagram of a long-term healthy diet recommendation system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0022] It should be noted that although the device schematics illustrate a modular division and the flowcharts illustrate a logical sequence, in some cases, the steps shown or described may be performed in a different order than the modular division in the device or the sequence in the flowcharts. The terms "first," "second," and so on in the specification, claims, and drawings are used to distinguish similar items and are not necessarily used to describe a specific order or precedence.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0024] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0025] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0026] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0027] This invention combines nutritional modeling, knowledge graph construction, health risk prediction and dynamic optimization technology, and is suitable for application scenarios such as intelligent health management systems, personalized diet recommendation platforms, and long-term chronic disease prevention.

[0028] To address the deficiencies in the prior art, the present invention provides a method, device, system, and medium for recommending a long-term healthy diet. The main technical solutions include the following three aspects: Stage modeling: Based on the physiological characteristics and nutritional needs of users at different life stages (such as adolescence, middle age, and old age), a segmented modeling method is used to establish a corresponding nutritional demand model, thereby providing a personalized basic diet plan for each stage.

[0029] Knowledge graph prediction: Utilize the user's historical health data, diet records, and physical examination data to build a health knowledge graph, and use statistical models and machine learning algorithms to predict possible health risks in the future, providing forward-looking risk warnings and intervention basis for diet planning.

[0030] Dynamic plan generation: Based on large-scale artificial intelligence models, combined with real-time user feedback and the latest health data, the initial diet plan is dynamically adjusted and optimized to ensure that the plan can adapt to the user's physiological changes and the continuous evolution of lifestyle habits.

[0031] Overall, the present invention organically combines the personalized nutritional needs of users at different life stages, health risk prediction based on knowledge graphs, and a dynamic feedback mechanism to achieve continuous optimization and evolution of long-term healthy diet planning.

[0032] like Figure 1 As shown, Figure 1 A method for recommending a long-term healthy diet is provided in an embodiment of the present invention, and the method comprises the following steps: S100, obtaining the user's current raw data, preprocessing the raw data, and generating the user's health data vector; S200, divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; S300, determining the life stage of a user, and generating a personalized nutritional requirement vector for the user based on a standard nutritional requirement vector for the life stage and a health data vector of the user; S400, obtaining historical data of the user, generating a comprehensive feature vector based on the historical data, and generating a prediction result of the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; S500, generating a current diet plan for the user based on the personalized nutritional requirement vector of the user at the current life stage.

[0033] This invention utilizes historical health data and statistical models to construct a knowledge graph, enabling accurate predictions of potential future health risks. Furthermore, based on user feedback on physiological changes and lifestyle habits at different life stages, a large-scale artificial intelligence model is used to automatically generate and adjust long-term dietary plans. This algorithm not only meets the personalized nutritional needs of users at different life stages but also possesses proactive risk intervention capabilities, dynamically adapting to physiological changes and enabling dietary evolutionary planning throughout the life cycle, from youth to old age. This provides comprehensive and precise guidance for users' long-term healthy diets.

[0034] In some improved embodiments, obtaining the user's original data, preprocessing the original data, and generating the user's health data vector includes: S110, obtaining the user's original data in the current sampling period, wherein the original data includes basic data and dietary health data; the dietary health data includes dietary records, daily exercise amount, work and rest time, and physical examination report; S120 , forming the dietary health data into a time series, preprocessing the time series through feature engineering, and generating a user's health data vector.

[0035] The present invention conducts segmented modeling for the nutritional needs of different life stages, such as adolescence, middle age and old age, and combines a health risk prediction module based on knowledge graph and a dynamic planning generation mechanism to achieve continuous optimization of user diet planning.

[0036] It should be noted that this step mainly implements system architecture and data collection, specifically including the following points: 1.1, data collection layer; First, a multi-dimensional data collection module is built to comprehensively collect various types of health information of users.

[0037] (1) Collection of basic user information: Collect users' basic data such as age, gender, height, weight, body fat percentage, blood pressure, blood sugar, etc. through mobile phone apps, wearable devices and smart physical examination terminals.

[0038] For example, the basic data for a 15-year-old user named "Xiao Ming" might be: age 15, height 165cm, weight 55kg, body fat percentage 18%, and resting heart rate 72 beats / minute. For a middle-aged user named "Ms. Li," her basic data might be: age 42, height 160cm, weight 60kg, body fat percentage 28%, and blood pressure 135 / 85 mmHg.

[0039] (2) Health records and living habits: It also collects the user's diet records, daily exercise, sleep schedule, and physical examination reports, automatically integrating data from the past six months or even a year to form a time series.

[0040] For example, record Xiao Ming's daily nutritional data such as calories, protein, fat, carbohydrates, etc., and make a comprehensive assessment based on the number of exercise steps and sleep duration.

[0041] The collected data is integrated from multiple sources, cleaned and standardized to form a unified health record.

[0042] (3) Preprocessing and health record construction: Clean, standardize and normalize the collected raw data to form a high-dimensional health data vector (For example, d includes indicators such as weight, BMI, daily energy intake, and exercise volume.) Through feature engineering, the collected data is denoised, normalized, and missing values ​​are filled to ensure that each user's health record data is highly consistent and comparable.

[0043] For example, if blood sugar levels in physical examination data are abnormal or missing, they are interpolated and corrected using historical data to ensure data integrity and accuracy. This processing makes subsequent data-based model calculations and risk predictions more reliable.

[0044] In some improved embodiments, dividing the life cycle into different life stages and determining the standard nutritional requirement vector for each life stage includes: S210, divide the life cycle into different stages based on age, gender, physiological characteristics and health status; S220 , for each life stage, pre-set a set of standard nutritional requirement parameters to form a standard nutritional requirement vector.

[0045] It should be noted that this step mainly implements phased modeling, which specifically includes the following points: 2.1, Life stage division; The user's life cycle is divided into different stages based on the user's age, gender, physiological characteristics and health status. Each stage represents different physiological and metabolic characteristics. Common divisions include but are not limited to: Adolescence (12 to 18 years old): The body is in a stage of rapid growth and development, and the nutritional needs are mainly high energy, high protein and sufficient trace elements.

[0046] Middle age (35-55 years old): The body's metabolism tends to be stable, but the risk of chronic diseases gradually emerges. Nutritional planning needs to take into account both energy control and disease prevention.

[0047] Elderly people (over 55 years old): Physiological functions gradually decline, and absorption and metabolic abilities decrease. The focus is on supplementing necessary nutrients, enhancing immunity, and preventing osteoporosis and other problems.

[0048] In addition to simple age classification, a secondary judgment can be made based on physical examination reports, genetic factors and lifestyle habits.

[0049] For example, for people who exercise a lot or have special occupations, the classification criteria can be adjusted appropriately to ensure that the classification results are more in line with their actual physiological state.

[0050] The standards for physiological maturity may vary in different cultural contexts, so custom parameters are supported, allowing experts to fine-tune the boundaries of each stage according to actual needs.

[0051] 2.2, stage-specific nutritional requirements model; For each life stage, a set of standard nutritional requirement parameters are pre-set to form a standard nutritional requirement vector.

[0052] (1) Adolescence: ; During adolescence, emphasis is placed on the supply of energy and protein to support the body's rapid growth and muscle development, while ensuring adequate supply of trace elements to support the needs of the endocrine and immune systems.

[0053] (2) Middle age: ; In middle age, the focus is on moderate energy control and a reasonable proportion of fat intake, aiming to prevent weight gain and the risk of chronic diseases, while appropriately increasing fiber and antioxidants to support cardiovascular and metabolic health.

[0054] (3) Elderly: In the elderly stage, the standard nutritional requirement model emphasizes a diet low in calories, fat, fiber, and trace elements. Supplementing calcium, vitamin D, and protein can help prevent osteoporosis and muscle loss while taking into account easy digestibility and high absorption rate.

[0055] 2.3, Personalized adjustments specifically include the following points: (1) Establish a weighted model; Based on the standard nutritional requirement vector, the user's health data vector is used The user's nutritional needs vector is calculated using a weighted formula (including weight, BMI, amount of exercise, medical history, physical examination indicators, etc.).

[0056] In some embodiments, the calculation formula of the user's nutritional requirement vector is: ; in: represents the personalized nutritional requirement vector, represents the standard nutritional requirement vector, represents the health data vector, represents the life stage of the i-th user; Expresses life stage specific standard nutrient weights reflecting the general requirements of that stage, Represents the adjustment weight based on individual health status, ensuring that the model can reflect the user's unique physiological and lifestyle differences.

[0057] (2) Data-driven adjustment mechanism; Dynamic update: The user's basic data will be updated regularly to capture changes in indicators such as weight, blood pressure, and blood sugar in real time, thereby adjusting the corresponding nutritional needs parameters.

[0058] Risk intervention: If the user has specific health risks (such as hypertension, diabetes), special optimization will be carried out based on the standard model, such as reducing sodium intake, increasing dietary fiber or supplementing specific trace elements.

[0059] In some improved embodiments, acquiring the user's historical data, generating a comprehensive feature vector based on the historical data, and generating a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector include: Obtain historical data of users in past sampling periods and extract time series features from the historical data; A knowledge graph is constructed by combining historical user data. Based on the association information between nodes in the knowledge graph, the user's nutritional intake, lifestyle habits and potential disease risks are associated to form a comprehensive feature vector. The nodes of the knowledge graph include nutritional ingredients, disease risks and lifestyle habits; the edges represent the association information between nodes. Generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector: the prediction result includes a risk prediction value, and nutrients and / or living habits that contribute significantly to the risk prediction value.

[0060] It should be noted that this step mainly realizes the construction of knowledge graph and health risk prediction; specifically includes the following points: 3.1, knowledge graph construction; Leverage health literature, clinical data, and user historical data to build a knowledge graph covering nutrients, disease risks, lifestyle habits, and their interrelationships.

[0061] The nodes of the graph include: nutrients (such as fat, protein, carbohydrates), diseases (such as cardiovascular disease, diabetes) and lifestyle habits (such as exercise volume, smoking, etc.), and the edges represent the association or causal relationship between these factors.

[0062] 3.2, health risk prediction module; (1) Data feature extraction and model construction; Feature extraction: Extract time series features from user historical data, such as weight changes, blood lipid levels, blood pressure fluctuations, and dietary intake over the past 6 months.

[0063] Combining the association information between nodes in the knowledge graph, the user's nutritional intake, living habits and potential disease risks are linked to form a comprehensive feature vector.

[0064] Model selection: Based on the amount of data and feature dimensions, various models such as logistic regression, random forest or neural network are selected for comparison.

[0065] In practical applications, we use simple statistical models (such as logistic regression) for preliminary risk assessment, and then use deep neural networks for refined prediction to ensure the stability and accuracy of the prediction results.

[0066] (2) Risk prediction function and dynamic evaluation: In some improved embodiments, the risk prediction function is defined as: ; in, Indicates the risk prediction value of the user's current life stage: represents the risk prediction model, Indicates the user's recent time period The comprehensive feature vector of the past 6 months (e.g., A personalized nutritional requirement vector for the user's current life stage.

[0067] By training the risk prediction model , enabling it to capture the relationship between historical data and current nutritional status, and output the risk probability of developing specific diseases (such as cardiovascular disease) in the future.

[0068] Dynamic evaluation mechanism: By regularly updating user health data, calculating risk prediction values ​​in real time, and combining the weight relationship between nutrition and disease reflected in the knowledge graph, the risk is dynamically corrected.

[0069] The prediction results not only output risk prediction values, but also provide corresponding explanatory information, such as which nutritional intake or lifestyle factors contribute more to the risk, to facilitate targeted adjustments to subsequent dietary plans.

[0070] Case description: A middle-aged user had three episodes of dyslipidemia in the past year. By collecting detailed blood lipid, blood pressure, and weight data from the user over the past six months and combining it with the positive correlation between "high-fat intake and cardiovascular disease" in the knowledge graph, the system predicted a 0.75 risk of cardiovascular disease in the next 12 months (the risk prediction value ranges from 0 to 1).

[0071] Based on this prediction, the system recommends reducing the intake of high-fat foods and increasing the proportion of foods rich in dietary fiber and unsaturated fatty acids in the next cycle, while strengthening exercise intervention to effectively reduce potential risks.

[0072] After the prediction results were verified in practice, the users' subsequent physical examination data showed that their blood lipid levels gradually returned to normal after adjusting their diet and lifestyle habits, proving that risk prediction and intervention measures are highly effective.

[0073] Next, dynamic planning generation and feedback optimization are performed; 4.1 Initial planning generation: (1) Personalized basic plan generation: First, an initial diet plan is generated based on the user's personalized nutritional needs vector at the current stage of life (such as the standard nutritional needs corresponding to adolescence, middle age, and old age, and the results after personalized adjustment).

[0074] For example, for an adolescent user named "Xiao Ming", after receiving his basic health data (height, weight, amount of exercise, daily energy consumption, etc.), an initial plan is generated, recommending a daily intake of 2500kcal, 70g protein, 80g fat, and 300g carbohydrates. At the same time, specific dishes and recipes are recommended based on seasonal ingredients and regional eating habits.

[0075] (2) Data support and rule verification: The initial planning is based not only on static data, but also on historical health records and authoritative nutrition guidelines to ensure that the developed plan is both scientifically based and practically feasible.

[0076] For example, the system will automatically verify whether the energy distribution in "Xiao Ming"'s plan meets the high metabolic needs of adolescence, and whether the ratio of protein and carbohydrates is appropriate, to ensure that the initial plan is reasonable and safe.

[0077] In some improved embodiments, the method further comprises: Collecting user monitoring data during the current sampling period; the monitoring data includes physiological indicators, exercise records, and user feedback on diet satisfaction; Standardize and vectorize the monitoring data to form a feedback vector; The pre-trained large model is used to process the current diet plan, risk prediction value and feedback vector to generate a new diet plan.

[0078] In this embodiment, the current diet plan is dynamically adjusted based on the user's monitoring data in the current sampling period to generate a new diet plan.

[0079] 4.2, Real-time feedback integration: (1) Multi-dimensional data collection: Users upload physiological indicators such as weight, blood pressure, blood sugar, body fat percentage, etc. every day through APP or wearable devices. At the same time, the system will also collect users' diet satisfaction (subjective feedback such as taste, fullness, diet diversity, etc.) and exercise records.

[0080] For example, after "Xiao Ming" uploaded data for a week, the system found that his weight had increased by 1.2 kg compared with the previous period. At the same time, it recorded that his daily exercise steps were less than before and his carbohydrate intake was high.

[0081] (2) Feedback data forms a vector: The collected data is standardized and vectorized to form a feedback vector This vector not only includes quantitative data (such as weight changes and blood sugar levels), but also incorporates qualitative evaluations (diet satisfaction scores), providing all-round information support for subsequent dynamic adjustments.

[0082] Case description: For example, the feedback recorded by "Xiao Ming" in one week shows: Daily weight gain trend; Consuming more calories than expected; Low dietary satisfaction scores; The system automatically integrates this data and determines that the current plan has deviations in energy distribution or recipe diversity and needs to be optimized in the next cycle.

[0083] 4.3 Dynamic planning adjustment: (1) Model-driven solution iteration: Use a pre-trained large model (based on reinforcement learning) to receive the current diet plan , health risk prediction and feedback data Then, a new plan is generated based on the built-in dynamic adjustment algorithm. .

[0084] By comparing the deviation between the current plan and the health goal, the model automatically calculates the indicators that need to be adjusted, such as reducing the proportion of carbohydrates, appropriately increasing the intake of high-fiber foods, and adjusting the total energy intake.

[0085] (2) Customized adjustment example: In the case of Ms. Li, if the system continuously monitors an upward trend in her blood pressure and blood lipid indicators, the model may automatically adjust the plan: Reduce the daily carbohydrate intake from the original 55% to 45%, increase whole grains, vegetables and foods rich in dietary fiber, and recommend increasing light exercise and limiting salt and fat intake.

[0086] This adjustment ensures that the diet plan can more accurately respond to changes in the user's health status, thereby playing a preventive and intervention role.

[0087] (4) Feedback closed-loop mechanism: After dynamic planning adjustments, user data is continuously tracked in the next cycle, forming a closed-loop feedback loop. New data feedback is used to correct model parameters, ensuring that planning adjustments continue to move towards the optimal state.

[0088] 4.4, optimization algorithm core: (1) Loss function construction: In order to minimize the deviation between the actual diet plan and the health goal, the system designs a loss function .

[0089] This function comprehensively considers the user's actual data, risk prediction results, and ideal health status, and continuously reduces the loss value through methods such as gradient descent, so that the diet plan gradually approaches the optimal solution.

[0090] (2) Algorithm iteration and convergence: The model parameters are updated through multiple iterations and the diet plan is fine-tuned in each cycle to ensure that the model can be continuously optimized under continuous feedback.

[0091] For example, in each update, the system calculates the gradient between the current plan and the target, adjusts the planning parameters, and controls the update step size with a learning rate, thereby achieving a smooth and efficient optimization process.

[0092] (3) Reinforcement learning strategy: In practical applications, reinforcement learning methods can also be used to construct a reward mechanism to allow the model to "learn" the best dietary adjustment strategy under different health conditions.

[0093] The model continuously tries different adjustment plans and gives positive rewards based on the improvement of the user's health status, making future decisions more accurate.

[0094] Through the organic combination of initial plan generation, real-time feedback integration, dynamic planning adjustment and optimization algorithm core, the system can respond to changes in the user's health status in real time and continuously optimize the diet plan based on continuous feedback.

[0095] This dynamic programming generation and feedback optimization mechanism not only improves the scientific nature and personalization of dietary plans, but also enables effective intervention in health risks in practice, ensuring that users always have the most appropriate and scientific dietary advice throughout their life cycles.

[0096] An example of life-cycle health management is provided below; 1. Adolescence For example, the diet plan for Xiao Ming during adolescence focuses on meeting the needs of rapid growth, recommending a high-protein, high-energy diet and ensuring nutritional balance based on exercise data. The system recommends a daily intake of approximately 2500kcal and guides healthy eating habits through personalized menus.

[0097] 2. Middle-aged stage: For middle-aged users like Ms. Li, the system identified a risk of dyslipidemia based on their medical examination reports and historical dietary data. Initial recommendations included a daily intake of 2200kcal, a reduced fat-to-carbohydrate ratio, and an increase in fiber-rich foods. If, after a period of follow-up feedback, blood lipid levels fall below standard, the system would further adjust the plan, such as reducing fat intake to around 60g and increasing omega-3 fatty acids.

[0098] 3. Elderly stage: As users age, the system prioritizes preventing risks like osteoporosis and chronic diseases, recommending a low-fat, high-fiber, low-sodium diet while also supplementing with vitamin D and calcium. In this case, a 65-year-old user's initial diet plan for their senior years included a daily intake of 2,000 kcal, with specific instructions for supplementing trace elements and minerals to ensure immune function and bone health.

[0099] 4. Dynamic health intervention: The system identifies the user's physiological changes in real time through continuous monitoring and feedback mechanisms.

[0100] For example, if a user's weight and blood sugar fluctuate abnormally for two consecutive weeks, the system will immediately initiate dynamic planning adjustments, reduce the proportion of high-GI foods, increase low-GI foods, and recommend an appropriate increase in exercise, striving to restore indicator balance in the shortest period.

[0101] From "Xiao Ming" to "Ms. Li," the system tracks users' health records, dynamically assesses health risks, and adjusts dietary plans based on real-time data, truly enabling personalized health management throughout the entire life cycle, from adolescence to old age. Each feedback adjustment is based on specific data and statistical models, providing users with safe and scientific dietary plans, significantly improving the accuracy and real-time nature of health management.

[0102] See Figure 2, an embodiment of the present invention provides a long-term healthy diet recommendation device, comprising: The first module is used to obtain the user's current raw data, pre-process the raw data, and generate the user's health data vector; The second module is used to divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; The third module is used to determine the life stage of the user and generate a personalized nutritional requirement vector for the user based on the standard nutritional requirement vector of the life stage and the health data vector of the user; The fourth module is used to obtain the user's historical data, generate a comprehensive feature vector based on the historical data, and generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; The fifth module is used to generate the user's current diet plan based on the user's personalized nutritional requirement vector at the current stage of life.

[0103] It can be seen that the contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0104] See Figure 3 , an embodiment of the present invention provides a long-term healthy diet recommendation system, comprising: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0105] It can be seen that the contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0106] In addition, the embodiments of the present application further disclose a computer program product or computer program, which is stored in a computer-readable storage medium. The processor of a computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the computer device performs the above-mentioned method. Similarly, the contents of the above-mentioned method embodiment are all applicable to the present storage medium embodiment, and the functions specifically implemented by the present storage medium embodiment are the same as those of the above-mentioned method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned method embodiment.

[0107] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0108] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0109] The terms "first," "second," "third," "fourth," and the like (if any) in the specification of the present application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in orders other than those illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus.

[0110] It should be understood that in this application, "at least one (item)" means one or more, and "more" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or plural.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0112] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0113] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0114] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store programs, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0115] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A long-term healthy diet recommendation method, characterized in that: The method comprises the following steps: Obtain the user's current raw data, pre-process the raw data, and generate the user's health data vector; Divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; Determining the user's life stage, and generating a personalized nutritional requirement vector for the user based on a standard nutritional requirement vector for the life stage and the user's health data vector; Obtain the user's historical data, generate a comprehensive feature vector based on the historical data, and generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; Generate the user's current diet plan based on the user's personalized nutritional needs vector at the current stage of life.

2. The method according to claim 1, characterized in that The method of obtaining the user's original data, preprocessing the original data, and generating the user's health data vector includes: Obtaining the user's original data in the current sampling period, the original data including basic data and dietary health data; the dietary health data including dietary records, daily exercise volume, work and rest time, and physical examination report; The dietary health data is formed into a time series, and the time series is preprocessed through feature engineering to generate the user's health data vector.

3. The method according to claim 1, characterized in that The life cycle is divided into different life stages, and the standard nutritional requirement vector for each life stage is determined, including: Divide the life cycle into different stages based on age, gender, physiological characteristics and health status; For each life stage, a set of standard nutritional requirement parameters are pre-set to form a standard nutritional requirement vector.

4. The method according to claim 3, characterized in that The calculation formula of the user's nutritional demand vector is: ； in: represents the personalized nutritional requirement vector, represents the standard nutritional requirement vector, represents the health data vector, represents the life stage of the i-th user; represents life stage specific standard nutrient weights, Represents the adjustment weight based on individual health status, ensuring that the model can reflect the user's unique physiological and lifestyle differences.

5. The method according to claim 1, wherein The method of obtaining the user's historical data, generating a comprehensive feature vector based on the historical data, and generating a prediction result of the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector includes: Obtain historical data of users in past sampling periods and extract time series features from the historical data; A knowledge graph is constructed by combining historical user data. Based on the association information between nodes in the knowledge graph, the user's nutritional intake, lifestyle habits and potential disease risks are associated to form a comprehensive feature vector. The nodes of the knowledge graph include nutritional ingredients, disease risks and lifestyle habits; the edges represent the association information between nodes. Generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector: the prediction result includes a risk prediction value, and nutrients and / or living habits that contribute significantly to the risk prediction value.

6. The method according to claim 5, characterized in that The calculation formula of the risk prediction value is: ； in, Indicates the risk prediction value of the user's current life stage: represents the risk prediction model, Indicates the user's recent time period The comprehensive feature vector within A personalized nutritional requirement vector for the user's current life stage.

7. The method according to claim 1, characterized in that The method further comprises: Collecting user monitoring data during the current sampling period; the monitoring data includes physiological indicators, exercise records, and user feedback on diet satisfaction; Standardize and vectorize the monitoring data to form a feedback vector; The pre-trained large model is used to process the current diet plan, risk prediction value and feedback vector to generate a new diet plan.

8. A long-term healthy diet recommendation device, characterized in that: include: The first module is used to obtain the user's current raw data, pre-process the raw data, and generate the user's health data vector; The second module is used to divide the life cycle into different life stages and determine the standard nutritional requirement vector for each life stage; The third module is used to determine the life stage of the user and generate a personalized nutritional requirement vector for the user based on the standard nutritional requirement vector of the life stage and the health data vector of the user; The fourth module is used to obtain the user's historical data, generate a comprehensive feature vector based on the historical data, and generate a prediction result for the user's current life stage based on the personalized nutritional requirement vector and the comprehensive feature vector; The fifth module is used to generate the user's current diet plan based on the user's personalized nutritional requirement vector at the current stage of life.

9. A long-term healthy diet recommendation system, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is configured to perform the method according to any one of claims 1 to 7 when executed by the processor.

Images