A multi-index driven disease risk early warning and personalized dietary recommendation system

Abstract

The application relates to the technical field of disease early warning and dietary recommendation, and discloses a multi-index-driven disease risk early warning and individualized dietary recommendation system, which comprises the following steps: constructing a disease risk prediction model based on a health state, obtaining a baseline disease risk index in a preset time period, generating an individualized dietary scheme according to the baseline disease risk index, setting a stage deviation tolerance mechanism, triggering a deviation state-independent risk correction, calculating an execution degree and a risk response direction according to the baseline disease risk index and the deviation state, decoupling and evaluating the dietary scheme based on the calculation result, judging whether the dietary scheme is effective when the execution degree exceeds a threshold value, refining a risk direction based on the baseline disease risk index and early warning an individual, calculating individual credibility according to a historical dietary execution deviation state, dynamically adjusting a dietary scheme recommendation, and improving model adaptation, the accuracy and operability of disease risk prediction and dietary intervention by cooperating with an execution-response decoupling evaluation module.

Description

Technical Field

[0001] This invention relates to the field of disease early warning and dietary recommendation technology, specifically a multi-indicator driven disease risk early warning and personalized dietary recommendation system. Background Technology

[0002] With the rising incidence of chronic and metabolic diseases, personalized dietary recommendations based on individual health status and disease risk can help to proactively intervene before the onset of diseases, improve metabolic levels, and reduce potential health risks by rationally adjusting dietary structure.

[0003] Existing personalized dietary recommendation technologies based on disease risk do not treat dietary adherence deviations as an independent, modelable key object, resulting in unclear time-phase characteristics and attribution of risk changes. First, in terms of time, existing schemes typically treat dietary adherence deviations as static or discrete events, assessing them only at a single point in time or within a single period. This lack of differentiation between short-term, occasional deviations and long-term, persistent deviations makes it difficult to capture the evolutionary trend of deviation behavior over time. Consequently, risk assessments are susceptible to interference from momentary behavior or delayed responses to long-term poor adherence, failing to support phased risk adjustments. Second, regarding risk attribution, existing technologies generally directly superimpose dietary adherence deviations onto overall disease risk indicators. When risk changes, it is difficult to determine whether the cause is an unreasonable dietary plan itself or individual adherence deviations, leading to a lack of clear basis for evaluating the effectiveness of the dietary plan and hindering the precise formulation of intervention strategies. Due to the lack of methods that simultaneously consider both time phases and attribution mechanisms, existing technologies cannot achieve refined modeling of dietary adherence deviations, limiting the stability, interpretability, and effectiveness of personalized dietary recommendations in disease risk early warning and long-term health management.

[0004] This proposal suggests a multi-indicator-driven disease risk early warning and personalized dietary recommendation system. Summary of the Invention

[0005] This invention provides a multi-indicator-driven disease risk early warning and personalized dietary recommendation system to help solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a multi-indicator driven disease risk early warning and personalized dietary recommendation system, comprising: The health status and risk prediction module is used to collect and integrate multiple indicator features to obtain health status, build a disease risk prediction model based on health status, and obtain the baseline disease risk index for a preset time period. The personalized diet execution deviation module is used to generate personalized diet plans based on the baseline disease risk index, establish a diet execution deviation status based on actual intake behavior, set a phased deviation tolerance mechanism, and trigger independent risk correction of the deviation status. The execution-response decoupling assessment module is used to calculate the execution degree and risk response direction based on the baseline disease risk index and deviation status, and to decouple the effectiveness and risk of dietary program execution based on the calculation results. The protocol effectiveness and correction module is used to determine whether a dietary protocol is effective when the implementation rate exceeds a threshold, and to refine the risk direction and issue warnings to individuals when the implementation rate is below the threshold, based on the baseline disease risk index. The credibility-driven diet recommendation module calculates individual credibility based on historical dietary execution deviations and dynamically adjusts diet plan recommendations. The collaborative execution-response decoupling evaluation module improves model adaptation.

[0007] Optionally, the health status and risk prediction module is used to collect and fuse multiple indicator features to obtain a health status, construct a disease risk prediction model based on the health status, and obtain a baseline disease risk index for a preset time period, including: For any individual: After obtaining multiple indicator features, performing nonlinear mapping, and concatenating them, the health status is obtained. ; The multi-indicator characteristics include bioinformatics characteristics, lifestyle characteristics, and physical constitution characteristics; Mapping the preset time period to a high-dimensional space yields... The preset time period is 3 months; splicing and get Construct a disease risk prediction model: Hidden layer: , , It is a non-linear activation function. This is the weight matrix. For bias; Output layer: , Baseline disease risk index; Aggregate baseline disease risk indices over a preset time period: .

[0008] Optionally, the personalized diet execution deviation module is used to generate a personalized diet plan based on the baseline disease risk index, establish a diet execution deviation status based on actual intake behavior, and set a phased deviation tolerance mechanism, including: Health status and baseline disease risk index Input a diet generation function to obtain a diet plan. ; Real-time recording of individual intake behavior ; Calculate dietary performance deviation at each time point ; Define the structured deviation vector: , For deviation from strength, The deviation duration is the duration of the deviation state within a continuous time period; To deviate from the direction, The actual intake was lower than the dietary plan. To ensure that the actual intake conforms to the dietary plan, Actual intake was higher than the dietary plan; Calculate the consistency coefficient of the deviation direction within a preset time period. , Used to prevent division by zero. The closer it is to 1, the more consistent the deviation direction is; Calculate the degree of deviation , Weights at each time point; Design deviation from threshold and ; like This is a short-term phase, during which lifestyle characteristics are not updated; like This is the intermediate stage, where a phased deviation tolerance mechanism is set up; like This is the long-term stage, characterized by updated lifestyle features.

[0009] Optionally, the setting of the phased deviation tolerance mechanism further includes: The baseline disease risk index corresponds to multiple risk sub-items affected by the current deviation status; Update any risk sub-item The value: , They are weights, It is a bivariate function. The set deviation intensity trigger threshold, Baseline risk; When the deviation intensity exceeds the set threshold, If so, then the risk sub-item is modified.

[0010] Optionally, the execution-response decoupling assessment module is used to calculate the execution degree and risk response direction based on the baseline disease risk index and deviation status, and to conduct a decoupling assessment of the effectiveness and risk of the dietary plan execution based on the calculation results, including: The execution level and risk response direction are independent of each other; Calculate the individual's performance :when At that time, ; when At that time, , This is a sensitivity coefficient used to control the rate at which the execution degree decreases as the degree of deviation increases; Calculate the direction of risk response Specifically: The health status corresponding to multiple indicators of an individual at the end of a preset time period under complete adherence to the dietary plan is set. ,based on Calculate the baseline disease risk index ; Obtain the health status at the end of a preset time period based on an individual's actual intake behavior. ,based on Calculate the baseline disease risk index ; Set threshold for changes in disease risk ,calculate ; like Then set ; like Then set ; like Then set .

[0011] Optionally, the scheme effectiveness and correction module is used to determine whether the dietary plan is effective when the implementation rate exceeds a threshold, and to refine the risk direction and issue an early warning to the individual based on the baseline disease risk index when the implementation rate is below the threshold, including: Set the threshold for execution ; like and If so, the dietary plan is effective; like and If so, the dietary plan is invalid; like ,calculate ,judge and Size relationship; like If so, it is recommended that the individual follow the dietary plan; if If implementing the dietary plan increases the risk, it will trigger an alert in the individual.

[0012] Optionally, the credibility-driven diet recommendation module is used to calculate individual credibility based on historical dietary execution deviations, dynamically adjust diet plan recommendations, and improve model adaptation through a collaborative execution-response decoupling evaluation module, including: The individual credibility is calculated as follows: For each individual, the execution rate over multiple historical preset time periods is obtained, and a consistency coefficient is used. Correct execution degree Calculate the mean of execution degree and variance ; Calculate credibility , This is a credibility adjustment coefficient. When the execution rate fluctuates greatly, the credibility decreases, and when the execution rate fluctuates little, the credibility increases. Dietary recommendations are dynamically adjusted based on individual credibility, specifically as follows: Get the meal generation function Set function ; calculate Recommended dynamically adjusted dietary plans To select dietary items by element; when , When the reliability of execution is low, the number of recommended items is reduced; when , When an individual's credibility is high, the number of recommended items increases.

[0013] Optionally, the collaborative execution-response decoupling evaluation module improves model adaptation by further including: Set weights The complexity of the dietary plan is adjusted based on individual credibility. , Weights The maximum and minimum values ​​of the weights, as individual credibility increases, the weights approach... When an individual's credibility decreases, the weight approaches... ; Update calculation , To weight the elements, when an individual's credibility is low, the key nutrients in the dietary items are reduced, and when an individual's credibility is high, the key nutrients in the dietary items are increased. Set acceptability constraints The historical frequency of execution for each dietary item is statistically analyzed. When the execution frequency falls below a set threshold, the corresponding dietary item is identified as rejected. ; When the execution frequency exceeds the set threshold, the corresponding dietary item will be determined as received. ; Update calculation Then, dietary items that individuals reject are removed.

[0014] The present invention has the following beneficial effects: 1. This multi-indicator-driven disease risk early warning and personalized dietary recommendation system quantifies an individual's baseline disease risk based on multidimensional health indicators and time-period information, generates personalized dietary plans, and monitors dietary implementation deviations in real time, including the direction, intensity, and duration of deviations, enabling phased management of short-term occasional deviations and long-term persistent deviations. Through a phased deviation tolerance mechanism and independent correction of risk sub-items, the system can clearly define the specific impact of deviations on health risks, improving the scientific rigor and interpretability of interventions. The implementation-response decoupling assessment module independently evaluates dietary implementation behavior and risk changes, distinguishing between plan effectiveness and individual implementation deficiencies, ensuring accurate and reliable intervention decisions. The credibility-driven module dynamically adjusts the complexity and key nutritional items of the dietary plan based on the individual's historical implementation stability, while eliminating exclusionary items, achieving long-term personalized optimization. The overall solution not only improves the accuracy and operability of disease risk prediction and dietary intervention but also enhances the adaptability of personalized interventions and the effectiveness of long-term health management, providing a comprehensive, efficient, and interpretable technical means for precision nutritional intervention, disease risk control, and individual health optimization.

[0015] 2. This multi-indicator-driven disease risk early warning and personalized dietary recommendation system collects multi-dimensional indicators such as bioinformatics, lifestyle, and physical constitution, performs nonlinear mapping and high-dimensional splicing on them to form a comprehensive and accurate representation of individual health status. Simultaneously, it constructs a disease risk prediction model by combining information from a preset time period, outputting a baseline disease risk index. It can simultaneously capture the multi-dimensional characteristics and temporal evolution trends of individual health, enabling disease risk prediction to not only reflect the current state but also quantify future risk changes, providing a scientific basis for personalized dietary intervention. The design of the nonlinear activation function in the hidden layer enhances the model's ability to identify complex nonlinear health patterns, and the risk index calculated by the output layer can serve as a benchmark indicator for dietary intervention and risk monitoring at various stages. This significantly improves the accuracy and interpretability of risk prediction, allowing dietary plans to be quantitatively optimized for different risk levels and time stages. Furthermore, it supports dynamic updates to individual health status, enabling the system to continuously adapt to individual physiological and behavioral changes during long-term intervention, thereby achieving personalized and dynamic disease risk early warning and dietary plan optimization, improving long-term health management effectiveness and clinical operability.

[0016] 3. This multi-indicator-driven disease risk warning and personalized dietary recommendation system generates personalized dietary plans based on a baseline disease risk index and records individual intake behavior in real time. By calculating deviation vectors, including the direction, intensity, and duration of deviation, it achieves a quantitative representation of dietary adherence deviations. Its advantages lie in its ability to distinguish the different risk implications corresponding to actual intake being higher or lower than the dietary plan, and to categorize deviation behavior into short-term, medium-term, and long-term stages, enabling phased management and intervention. Deviation thresholds and phased tolerance mechanisms ensure that short-term, occasional deviations do not excessively affect risk assessment, while independent risk correction is performed on long-term, persistent deviations, ensuring the scientific rigor and stability of the intervention. By updating risk sub-items affected by deviations, the source of risk changes becomes clearly traceable, providing a basis for dietary plan adjustments and individual risk warnings. Furthermore, the real-time deviation monitoring and phased management mechanism effectively reduces the interference of behavioral fluctuations on risk assessment, enabling refined management of individual long-term dietary behavior and improving the continuous adherence and effectiveness of dietary interventions. It provides a precise execution assessment tool for personalized health management, enhancing the operability, scientific rigor, and interpretability of dietary interventions in practical applications.

[0017] 4. This multi-indicator-driven disease risk early warning and personalized dietary recommendation system achieves decoupled evaluation of dietary plan implementation effectiveness and disease risk changes by independently calculating dietary adherence and risk response direction. It clearly distinguishes the independent roles of individual implementation behavior and plan design in risk changes, making the evaluation of dietary plan effectiveness more accurate. With high adherence, the effectiveness of the dietary plan can be directly determined; with low adherence, the system further refines the risk direction based on risk change attribution and triggers individual early warnings, avoiding the misleading effect of solely using overall risk changes as the basis for intervention. By setting thresholds and difference judgments, it can scientifically differentiate risks arising from insufficient dietary adherence, flawed plan design, or long-term deviations, improving the targeting and accuracy of intervention decisions. It supports the analysis of risk change trends within a preset time period, enabling the system to dynamically adjust for different time stages and individual behavioral performance, ensuring the scientific nature, feasibility, and long-term effectiveness of dietary interventions, and providing reliable and interpretable decision-making basis for disease risk management.

[0018] 5. This multi-indicator-driven disease risk early warning and personalized dietary recommendation system analyzes individual historical execution deviation data to calculate the mean and volatility of execution, generating a reliability index for dynamically adjusting dietary plan recommendations and achieving long-term adaptability. The reliability index reflects the long-term stability of individual execution. When execution fluctuates significantly, the system automatically reduces the complexity of the dietary plan and key nutritional items to reduce execution resistance and improve operability. When execution is stable, the system increases nutritional items and plan complexity to enhance intervention effectiveness and risk control capabilities. Simultaneously, the introduction of acceptability constraints eliminates dietary items rejected by individuals, enhancing the personalization and acceptability of the plan. Through the dynamic combination of historical execution data and real-time feedback, precise adjustments and long-term optimization of the dietary plan are achieved, improving the compliance rate and scientific rigor of personalized interventions. This enhances the system's adaptability to individual behavioral differences, ensuring that dietary recommendations are both feasible and ensure health risk control, significantly improving long-term health management effectiveness, intervention precision, and the practical value of the personalized dietary recommendation system. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, refer to Figure 1 A multi-indicator-driven disease risk early warning and personalized dietary recommendation system, comprising: All health data entered by individuals will be recorded and encrypted in the health management platform's database. The platform supports updating or modifying personal data, and through data processing and model analysis, assesses users' health status and generates personalized health assessment reports. Based on individual health status, physical characteristics, and health needs, the platform automatically recommends suitable food-medicine homology ingredients. Individuals can input relevant keywords or natural language descriptions (such as "kidney tonification," "blood sugar reduction," etc.) to extract matching ingredients from the database and recommend them. Ingredient recommendations not only consider their nutritional components but also reference traditional food-medicine homology theories and their therapeutic effects on specific health issues. Based on the user's selection of food-medicine homology ingredients, a personalized dietary plan will be automatically generated, including: Food pairing suggestions: Recommend which food combinations can maximize health benefits.

[0022] Cooking methods: Provide suitable cooking methods for each ingredient to ensure that its nutritional value is maximized.

[0023] Meal schedule and timing: Based on individual work and rest schedules, the timing of food intake is arranged reasonably to help users better control their health. Deviations from individual dietary plans are inevitable, and the system will dynamically adjust accordingly. Individuals provide monthly or weekly feedback on their dietary adherence, and the system dynamically adjusts recommendations based on their adherence and degree of deviation, providing personalized health plans to help customers achieve better results in weight control, improved fitness, and disease prevention.

[0024] The health status and risk prediction module is used to collect and fuse multiple indicator features to obtain health status, construct a disease risk prediction model based on health status, and obtain a baseline disease risk index for a preset time period, including: In this embodiment, the individual was 45 years old, weighed 68kg, was 162cm tall, and had mild hyperglycemia; Objective: To lower blood sugar and reduce cardiovascular risk; Multiple indicators: blood glucose 110–130 mg / dL, blood pressure 130 / 85 mmHg, weight 68 kg, exercise volume 4500 steps / day, dietary structure score 4 / 10; The indicators were non-linearly mapped to [0,1]: blood glucose → 0.6, blood pressure → 0.55, weight → 0.5, exercise volume → 0.45, dietary score → 0.4, and health status. ; The multi-indicator characteristics include bioinformatics characteristics, lifestyle characteristics, and physical constitution characteristics; Mapping the preset time period to a high-dimensional space yields... A total of 90 dimensions; splicing and get Construct a disease risk prediction model: Hidden layer: , , It is a non-linear activation function. This is the weight matrix. For bias; Output layer: , Baseline disease risk index; Aggregate baseline disease risk index: The higher the value, the higher the risk. The baseline disease risk index provides a benchmark for a system's health status by comprehensively considering multiple individual characteristics. This baseline value can help design personalized dietary plans.

[0025] The personalized diet implementation deviation module is used to generate personalized diet plans based on baseline disease risk indices, establish diet implementation deviation status based on actual intake behavior, and set up a phased deviation tolerance mechanism, including: Health status and baseline disease risk index Input an existing dietary generation function to obtain a dietary plan. The total daily calorie intake for three meals is 1800 kcal, with carbohydrates accounting for 45%, protein for 25%, fat for 30%, fruit for 200g / day, and vegetables for 300g / day. Real-time recording of individual intake behavior Calories 2000 kcal (higher than the plan), carbohydrates 50%, protein 20%, fat 30%, fruit 100g / day, vegetables 250g / day; The discrepancy between dietary plans and actual intake behavior stems from multiple factors, including individual perception, limitations in food access, eating habits, and social influences. To reduce this discrepancy, the system needs to better adapt to individual circumstances, provide more flexible and personalized dietary recommendations, and strengthen the monitoring and feedback of individual behavioral deviations.

[0026] Calculate dietary performance deviation at each time point ; Define the structured deviation vector: , For deviation from strength, The duration of the deviation is the length of time the deviated state is maintained within a continuous period of time; Calculate the consistency coefficient of the deviation direction within a preset time period. , Used to prevent division by zero. The closer it is to 1, the more consistent the deviation direction is; Calculate the degree of deviation , Weights at each time point; Design deviation from threshold and ; In this embodiment, the degree of deviation of the individual This is a short-term phase, and lifestyle characteristics will not be updated. like This is the intermediate stage, where a phased deviation tolerance mechanism is set up; like This is the long-term stage, characterized by updated lifestyle features.

[0027] 1. Short-term phase: The short-term phase refers to occasional dietary deviations, which are usually one-off events and will not have long-term effects on health. Short-term deviations do not require overly strict intervention because their impact on health is limited. For example, occasionally eating some high-sugar foods at a birthday party may deviate from the dietary plan, but there is no need to immediately adjust the plan because it is only a short-term deviation.

[0028] 2. Mid-stage: The mid-stage refers to a period where the deviation has persisted for some time and has begun to have some impact on health, but it is not yet very serious. If a person does not follow the dietary plan for a week, although it is not a long time, it has already begun to have some impact on health, and the system will remind them to pay attention to adjusting their diet to avoid long-term deviation.

[0029] 3. Long-term stage: The long-term stage refers to dietary deviations that have persisted for an extended period and have already posed significant health risks. Long-term deviations have a greater impact on health and require stronger intervention and adjustments. If a person consistently consumes high-salt, high-fat foods, leading to weight gain and elevated blood sugar, this long-term dietary deviation has already caused health problems. It is recommended to change dietary habits to reduce these risks.

[0030] Without differentiation of phases, the system may overreact to minor deviations in the short term, leading to decreased individual adherence and trust, thus rendering the intervention ineffective. Alternatively, the system may react slowly to prolonged deviations, missing the optimal time for intervention.

[0031] The baseline disease risk index corresponds to multiple risk sub-items affected by the current deviation status; In this embodiment, if the deviation from the cumulative total over 3 consecutive days Then update any risk sub-item. The value: , They are weights, It is a bivariate function. The set deviation intensity trigger threshold, Baseline risk; When the deviation intensity exceeds the set threshold, If so, then the risk sub-item is modified.

[0032] In this embodiment, the individual's diabetes risk is 0.6 and cardiovascular risk is 0.3. It is assumed that during the implementation of the dietary plan, the individual consumes excessive sugar for a period of time for some reason, which will directly affect their diabetes risk but have a small or even negligible impact on their cardiovascular risk. In this case, the cardiovascular risk does not need to be updated. Updating the overall disease risk index every time there is a dietary deviation could lead to an overreaction to other health problems, thereby triggering unnecessary interventions.

[0033] The execution-response decoupling assessment module is used to calculate the execution degree and risk response direction based on the baseline disease risk index and deviation status. Based on the calculation results, it performs a decoupling assessment of the effectiveness and risk of dietary program execution, including: The execution level and risk response direction are independent of each other; Calculate the individual's performance In this embodiment At that time, ; when At that time, , This is a sensitivity coefficient used to control the rate at which the execution degree decreases as the degree of deviation increases; Execution rate is an indicator used to measure an individual's actual adherence to a dietary plan, primarily reflecting whether the individual follows the dietary plan. Its value depends on whether the individual consumes the recommended foods and the precision of their execution. When dietary execution aligns with the plan, the execution rate is close to 1, indicating excellent execution. When dietary execution deviates significantly, the execution rate decreases, approaching 0, indicating poor execution.

[0034] Calculate the direction of risk response Specifically: Risk response direction measures the direction of change in health status after deviation from a dietary plan. It compares health status under conditions of complete adherence to the dietary plan with actual deviations to assess changes in health risk.

[0035] If health status improves after the implementation of the dietary plan, and the risk response is negative, it indicates that the risk has decreased and the dietary plan is effective.

[0036] If the health status does not improve and the risk response direction is zero, it indicates that the dietary implementation has not been effective.

[0037] If health deteriorates, the risk response direction is positive, indicating that the dietary plan may lead to increased risk and needs to be adjusted.

[0038] The health status corresponding to multiple indicators of an individual at the end of a preset time period under complete adherence to the dietary plan is set. ,based on Calculate the baseline disease risk index ; Obtain the health status at the end of a preset time period based on an individual's actual intake behavior. ,based on Calculate the baseline disease risk index ; Set threshold for changes in disease risk ,calculate ;like Then set ; In this embodiment Then set ; like Then set .

[0039] The protocol effectiveness and correction module is used to determine the effectiveness of a dietary protocol when the adherence rate exceeds a threshold, and to refine the risk direction and issue individual warnings based on the baseline disease risk index when the adherence rate falls below the threshold; determining the effectiveness of a dietary protocol: If the implementation rate is low (i.e., the individual does not effectively implement the dietary plan) and the risk response direction indicates a deterioration in health status, the system will consider the dietary plan invalid and make corrections.

[0040] When the implementation rate is high and the risk response is negative, the dietary plan is considered effective and should continue to be implemented.

[0041] When the implementation rate is high but the risk response direction is positive, the dietary plan is considered ineffective and needs to be adjusted.

[0042] When the execution rate is low, even if the risk response direction is negative, it is necessary to assess whether the plan needs to be modified.

[0043] Set the threshold for execution ; In this embodiment and If so, the dietary plan is effective; like and If so, the dietary plan is invalid; like ,calculate ,judge and Size relationship; like If the individual is in a critical condition, it is recommended that they follow a dietary plan. In comparison, following a dietary plan will reduce disease risk indicators, so it is recommended that the individual follow a dietary plan. like If implementing a dietary plan increases the risk, it indicates that implementing the dietary plan actually increases the disease risk indicators, thus triggering an early warning for the individual to change the dietary plan. The credibility-driven diet recommendation module calculates individual credibility based on historical dietary adherence deviations, dynamically adjusts diet plan recommendations, and the collaborative execution-response decoupling evaluation module improves model adaptation, including: Calculating individual credibility is crucial. Credibility is a vital component of personalized dietary recommendation systems. By reflecting the stability of an individual's dietary adherence, it helps the system adjust the content and complexity of recommendations, ensuring that individuals can consistently maintain their health goals. Furthermore, credibility enhances an individual's confidence and willingness to follow through, further promoting the sustainability of healthy behaviors.

[0044] For an individual, the execution scores for multiple preset historical time periods were 0.9, 0.85, and 0.8, respectively, using a consistency coefficient. Correct execution degree Calculate the mean of execution degree and variance ; Calculate credibility , This is a credibility adjustment coefficient. When the execution rate fluctuates greatly, the credibility decreases, and when the execution rate fluctuates little, the credibility increases. Dietary recommendations are dynamically adjusted based on individual credibility, specifically as follows: Get the meal generation function Set function ; calculate Recommended dynamically adjusted dietary plans To select dietary items by element; when , When the reliability of execution is low, the number of recommended items is reduced; when , When an individual's credibility is high, the number of recommended items increases.

[0045] Set weights The complexity of the dietary plan is adjusted based on individual credibility. , Weights The maximum and minimum values ​​of the weights, as individual credibility increases, the weights approach... When an individual's credibility decreases, the weight approaches... ; Update calculation , To weight the elements, when an individual's credibility is low, the key nutrients in the dietary items are reduced, and when an individual's credibility is high, the key nutrients in the dietary items are increased. Set acceptability constraints The historical frequency of execution for each dietary item is statistically analyzed. When the execution frequency falls below a set threshold, the corresponding dietary item is identified as rejected. ; When the execution frequency exceeds the set threshold, the corresponding dietary item will be determined as received. ; Update calculation Then, dietary items that individuals reject are removed.

[0046] This program utilizes a personalized dietary recommendation system to monitor and adjust dietary plans in real time based on individual health data, thereby effectively reducing disease risk and improving health status. Specific results are as follows: Significantly Reduced Disease Risk: By calculating a baseline disease risk index, the system can reduce the risk of diabetes and cardiovascular disease by approximately 15%. Controlled Dietary Deviation: The system monitors dietary adherence deviations in real time, reducing the deviation rate from 5% initially to 3%, significantly reducing dietary errors. Improved Health Adherence: Individual dietary adherence improved from 0.8 to 0.85, enhancing individual health management execution. Achievement of Health Goals: Individuals achieved an average weight loss of 5%, effective control of blood sugar and blood pressure, and a 20%-30% increase in health goal achievement rate. Reduced Overall Health Risk: Through continuous monitoring, the system helps individuals reduce their overall disease risk by 10%-20%.

[0047] Conclusion: This program can effectively improve individual health, reduce disease risk, and help individuals achieve their health management goals, demonstrating significant practical value.

[0048] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0049] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multi-indicator-driven disease risk early warning and personalized dietary recommendation system, characterized in that: include: The health status and risk prediction module is used to collect and integrate multiple indicator features to obtain health status, build a disease risk prediction model based on health status, and obtain the baseline disease risk index for a preset time period. The personalized diet execution deviation module is used to generate personalized diet plans based on the baseline disease risk index, establish a diet execution deviation status based on actual intake behavior, set a phased deviation tolerance mechanism, and trigger independent risk correction of the deviation status. The execution-response decoupling assessment module is used to calculate the execution degree and risk response direction based on the baseline disease risk index and deviation status, and to decouple the effectiveness and risk of dietary program execution based on the calculation results. The protocol effectiveness and correction module is used to determine whether a dietary protocol is effective when the implementation rate exceeds a threshold, and to refine the risk direction and issue warnings to individuals when the implementation rate is below the threshold, based on the baseline disease risk index. The credibility-driven diet recommendation module calculates individual credibility based on historical dietary execution deviations and dynamically adjusts diet plan recommendations. The collaborative execution-response decoupling evaluation module improves model adaptation.

2. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 1, characterized in that: The health status and risk prediction module is used to collect and fuse multiple indicator features to obtain health status, construct a disease risk prediction model based on the health status, and obtain a baseline disease risk index for a preset time period, including: For any individual: After obtaining multiple indicator features, performing nonlinear mapping, and concatenating them, the health status is obtained. ; The multi-indicator characteristics include bioinformatics characteristics, lifestyle characteristics, and physical constitution characteristics; Mapping the preset time period to a high-dimensional space yields... The preset time period is 3 months; splicing and get Construct a disease risk prediction model: Hidden layer: , , It is a non-linear activation function. This is the weight matrix. For bias; Output layer: , Baseline disease risk index; Aggregate baseline disease risk indices over a preset time period: .

3. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 2, characterized in that: The personalized diet execution deviation module is used to generate personalized diet plans based on baseline disease risk indices, establish diet execution deviation status based on actual intake behavior, and set up a phased deviation tolerance mechanism, including: Health status and baseline disease risk index Input a diet generation function to obtain a diet plan ; Real-time recording of individual intake behavior ; Calculate dietary performance deviation at each time point ; Define the structured deviation vector: , For deviation from strength, The deviation duration is the duration of the deviation state within a continuous time period; To deviate from the direction, The actual intake was lower than the dietary plan. To ensure that the actual intake conforms to the dietary plan, Actual intake was higher than the dietary plan; Calculate the consistency coefficient of the deviation direction within a preset time period. , Used to prevent division by zero. The closer it is to 1, the more consistent the deviation direction is; Calculate the degree of deviation , Weights at each time point; Design deviation from threshold and ; like This is a short-term phase, during which lifestyle characteristics are not updated; like This is the intermediate stage, where a phased deviation tolerance mechanism is set up; like This is the long-term stage, characterized by updated lifestyle features.

4. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 3, characterized in that: The aforementioned phased deviation tolerance mechanism also includes: The baseline disease risk index corresponds to multiple risk sub-items affected by the current deviation status; Update any risk sub-item The value: , They are weights, It is a bivariate function. The set deviation intensity trigger threshold, Baseline risk; When the deviation intensity exceeds the set threshold, If so, then the risk sub-item is modified.

5. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 4, characterized in that: The execution-response decoupling assessment module is used to calculate the execution degree and risk response direction based on the baseline disease risk index and deviation status, and to conduct a decoupling assessment of the effectiveness and risk of dietary program execution based on the calculation results, including: The execution level and risk response direction are independent of each other; Calculate the individual's performance :when At that time, ; when At that time, , This is a sensitivity coefficient used to control the rate at which the execution degree decreases as the degree of deviation increases; Calculate the direction of risk response Specifically: The health status corresponding to multiple indicators of an individual at the end of a preset time period under complete adherence to the dietary plan is set. ,based on Calculate the baseline disease risk index ; Obtain the health status at the end of a preset time period based on an individual's actual intake behavior. ,based on Calculate the baseline disease risk index ; Set threshold for changes in disease risk ,calculate ; like Then set ; like Then set ; like Then set .

6. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 5, characterized in that: The effectiveness and correction module is used to determine the effectiveness of the dietary plan when the implementation rate exceeds a threshold, and to refine the risk direction and issue warnings to individuals based on the baseline disease risk index when the implementation rate is below the threshold, including: Set the threshold for execution ; like and If so, the dietary plan is effective; like and If so, the dietary plan is invalid; like ,calculate ,judge and Size relationship; like If so, it is recommended that the individual follow the dietary plan; like If implementing the dietary plan increases the risk, it will trigger an alert in the individual.

7. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 6, characterized in that: The credibility-driven diet recommendation module is used to calculate individual credibility based on historical dietary execution deviations, dynamically adjust diet plan recommendations, and improve model adaptation through a collaborative execution-response decoupling evaluation module, including: The individual credibility is calculated as follows: For each individual, the execution rate over multiple historical preset time periods is obtained, and a consistency coefficient is used. Correct execution degree Calculate the mean of execution degree and variance ; Calculate credibility , This is a credibility adjustment coefficient. When the execution rate fluctuates greatly, the credibility decreases, and when the execution rate fluctuates little, the credibility increases. Dietary recommendations are dynamically adjusted based on individual credibility, specifically as follows: Get the meal generation function Set function ; calculate Recommended dynamically adjusted dietary plans To select dietary items by element; when , When the reliability of execution is low, the number of recommended items is reduced; when , When an individual's credibility is high, the number of recommended items increases. These are the minimum and maximum values ​​for the dietary items, respectively.

8. The multi-indicator driven disease risk early warning and personalized dietary recommendation system according to claim 7, characterized in that: The collaborative execution-response decoupling evaluation module improves model adaptation and also includes: Set weights The complexity of the dietary plan is adjusted based on individual credibility. , Weights The maximum and minimum values ​​of the weights, as individual credibility increases, the weights approach... When an individual's credibility decreases, the weight approaches... ; Update calculation , To weight the elements, when an individual's credibility is low, the key nutrients in the dietary items are reduced, and when an individual's credibility is high, the key nutrients in the dietary items are increased. Set acceptability constraints The historical frequency of execution for each dietary item is statistically analyzed. When the execution frequency falls below a set threshold, the corresponding dietary item is identified as rejected. ; When the execution frequency exceeds the set threshold, the corresponding dietary item will be determined as received. ; Update calculation Then, dietary items that individuals reject are removed.

Images