Intestinal inflammation patient whole life cycle health management method, medium and device

Abstract

The application discloses a kind of whole life cycle health management methods, medium and equipment of enteritis patient, based on the whole life cycle health data obtained, key leverage factor is extracted and lever point intervention priority sequence is generated by lever point identification algorithm;Individualized toughness development path planning is generated in combination with the calculation result of toughness evaluation model, and individualized health management strategy is designed according to this structured self-verification experiment scheme;Through scheme execution result, leverage point identification algorithm and toughness evaluation model are continuously updated using adaptive learning algorithm until the optimal parameter configuration and individualized health management strategy are locked;When the stability of key leverage factor ranking is over-standard, start hierarchical response mechanism to generate accurate intervention scheme, and finally output whole life cycle health management comprehensive scheme.The application realizes the change from static matching to dynamic self-adaptation of management strategy by constructing dynamic optimization closed loop, and solves the deficiencies of existing methods in intervention target identification, adaptability quantification and strategy continuous optimization.

Description

Technical Field

[0001] This invention relates to the field of health management technology, specifically to a method, medium, and device for full life-cycle health management of patients with enteritis. Background Technology

[0002] Health management of digestive system diseases such as chronic enteritis has become an important topic in the field of modern health management due to its long course, high recurrence rate, and strong correlation with psychology, environment, and behavioral patterns. Ideal health management not only involves controlling acute symptoms but also achieving long-term, dynamic maintenance and improvement of the patient's health status throughout their entire lifespan. This requires comprehensive consideration of multiple dimensions of information, including individual physiological indicators, psychological state, lifestyle habits, and external environment.

[0003] Currently, technological development in this field mainly follows two paths. One path focuses on personalized solution recommendations based on massive amounts of data and predefined rules. For example, by building a database containing various intervention measures and matching or combining corresponding health management plans based on users' health data tags, this approach can provide a certain degree of personalization. However, its solution generation logic largely relies on static matching rules or decision trees, and adjustments to the plan are often made only after the user executes the plan and provides feedback, requiring a new round of database retrieval and matching. The other path attempts to introduce dynamic models, such as using machine learning algorithms to directly predict and generate the next stage of optimized plans based on the user's health indicators and behavioral data after executing the plan, thereby achieving dynamic updates in management.

[0004] However, when faced with the complex and dynamic management of chronic diseases, existing methods often lack a systematic analysis of the deep-seated key factors influencing the evolution of health status and their interaction mechanisms, resulting in limited precision and efficiency of interventions. Furthermore, existing solutions tend to focus on the application of external measures, lacking effective technological integration on how to quantitatively assess and proactively enhance individuals' intrinsic capacity to cope with change (i.e., their inherent physiological and psychological adaptability). In addition, the system struggles to form a continuous, structured learning and evolutionary closed loop from users' individualized practices, limiting the long-term adaptability and personalized depth of management strategies. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method, medium and device for full life cycle health management of patients with enteritis. By identifying key health influencing factors and constructing a dynamic optimization closed loop that includes self-validation and model iteration, it realizes the transformation from passive intervention to proactive construction of individualized health management strategies, and solves the problems of existing methods in dynamically identifying core intervention targets, quantifying intrinsic adaptability and realizing continuous adaptive optimization of strategies.

[0006] To achieve the above objectives, in a first aspect, this application provides a method for full-life-cycle health management of patients with enteritis, comprising:

[0007] Acquire patients’ full life-cycle health data, which includes dynamic physiological parameter sequences, environmental exposure parameters and behavioral pattern feature vectors. The dynamic physiological parameter sequences include at least intestinal function status indicators and psychophysiological stress indicators.

[0008] Based on full life-cycle health data, a map of factors influencing individual health status is constructed using a leverage point identification algorithm. Time series correlation analysis is used to extract key leverage factors from multi-dimensional data and generate a priority sequence of leverage point interventions.

[0009] Based on the leverage point intervention priority sequence, the physiological system buffer capacity and psychological resilience index are calculated through a resilience assessment model, and an individualized resilience development path plan is generated through multimodal data fusion.

[0010] Based on the priority sequence of leverage interventions and the individualized resilience development path planning, a structured self-validating experimental scheme is generated. The structured self-validating experimental scheme includes a hypothesis formulation module, variable control strategies, and effect evaluation mechanisms.

[0011] Based on the execution results of the structured self-verification experiment scheme, the parameters of the lever point identification algorithm and the connection weights of the resilience assessment model are continuously updated through an adaptive learning algorithm until the execution results of the structured self-verification experiment scheme meet the preset conditions.

[0012] The connection weights between the parameters of the leverage point identification algorithm and the resilience assessment model when the preset conditions are met are determined as the optimal parameter configuration, and the individualized health management strategy for the current stage is locked in.

[0013] Furthermore, when the ranking stability of key leverage factors is detected to exceed the preset stability threshold, a graded response mechanism is activated to generate a precise intervention plan based on the leverage point intervention priority sequence and the current resilience level. The precise intervention plan includes nutritional adjustment strategies, behavioral intervention plans, and psychological support measures.

[0014] Based on individualized health management strategies, precise intervention plans, and individualized resilience development path planning, a comprehensive health management plan for the entire life cycle is generated and output.

[0015] Furthermore, based on full-lifecycle health data, a personalized health status influencing factor map is constructed using a leverage point identification algorithm. Time series association analysis is employed to extract key leverage factors from multi-dimensional data, generating a leverage point intervention priority sequence, including:

[0016] Time alignment processing is performed on dynamic physiological parameter sequences, environmental exposure parameters, and behavioral pattern feature vectors to form a multidimensional data matrix with unified timestamps;

[0017] The multidimensional data matrix is ​​segmented and sampled using a sliding window algorithm. Within each time window, the correlation strength between the dynamic physiological parameter sequence, environmental exposure parameters, behavioral pattern feature vectors, and intestinal function status indicators after a unified timestamp is calculated to generate a dynamic correlation sequence.

[0018] A network graph is constructed based on dynamic correlation sequences. The nodes of the network graph represent dynamic physiological parameter sequences, environmental exposure parameters, and behavioral pattern feature vectors after a unified timestamp, denoted as node parameters. The edge weights represent the maximum mutual information values ​​between node parameters. A community detection algorithm is used to identify high-density connected subgraphs in the network graph.

[0019] The node parameters with the highest centrality are extracted from the high-density connected subgraph as candidate leverage factors, and the causal relationship between the candidate leverage factors and intestinal functional status indicators is verified by a causal inference algorithm.

[0020] Based on the strength of causality and clinical interventionability score, the candidate leverage factors are weighted and ranked to obtain the top-ranked candidate leverage factors, which are recorded as key leverage factors, and the remaining candidate leverage factors are recorded as secondary leverage factors.

[0021] Generate a priority sequence of leverage point interventions that includes key leverage factors and secondary leverage factors.

[0022] Furthermore, based on the leverage point intervention priority sequence, the physiological system buffer capacity and psychological resilience index are calculated through a resilience assessment model. Individualized resilience development path planning is generated through multimodal data fusion, including:

[0023] The resilience assessment model is initialized based on the clinical impact weight of each key leverage factor in the leverage point intervention priority sequence. The resilience assessment model includes a multi-level resilience assessment index system, which includes three assessment dimensions: nutritional metabolic reserves, neuroendocrine regulatory capacity, and psychological adaptability.

[0024] Based on heart rate variability index and cortisol diurnal rhythm data in dynamic physiological parameter sequences, the neuroendocrine regulation capacity score is calculated by autonomic nervous function analysis algorithm, and the nutritional metabolic reserve level is calculated based on the stability characteristics of time series data of nutritional biomarkers.

[0025] The buffer capacity of the physiological system is obtained based on the neuroendocrine regulation capacity score and the nutritional metabolic reserve level.

[0026] Based on the stress coping pattern record in the behavioral pattern feature vector and the social support index in the environmental exposure parameter, the psychological resilience index is calculated by the psychological resilience assessment scale, and the psychological resilience index is generated by combining the stress recovery speed parameter in historical health data.

[0027] The buffer capacity of the physiological system and the psychological resilience index are input into a multimodal feature fusion network, and a comprehensive resilience assessment feature vector is generated through an attention weighting mechanism.

[0028] Based on the comprehensive resilience assessment feature vector, the optimal development path from the current state to the target state is constructed in the resilience development space of the resilience assessment model through the path planning algorithm. The optimal development path includes a sequence of phased training objectives and adaptive challenge tasks.

[0029] Based on the optimal development path, an individualized resilience development path plan is generated. The individualized resilience development path plan includes a training intensity gradient increase scheme, key node evaluation criteria, and dynamic adjustment trigger conditions.

[0030] Furthermore, based on the priority sequence of leverage point interventions and the individualized resilience development path planning, a structured self-validating experimental protocol is generated, including:

[0031] Based on the key leverage factors in the leverage point intervention priority sequence, candidate intervention hypotheses are constructed through a hypothesis generation algorithm. Based on the current development stage goals in the individualized resilience development path planning, candidate intervention hypotheses that meet the clinical safety boundary are selected and denoted as hypotheses to be verified.

[0032] Based on the hypothesis to be verified, a variable control strategy is generated, which includes the setting of the baseline observation period, the implementation guidelines for intervention measures, and the confounding factor control scheme.

[0033] An effectiveness evaluation mechanism was established, which included a multi-dimensional effect indicator collection scheme, statistical testing methods, and clinical significance judgment criteria.

[0034] Based on the hypotheses to be verified, variable control strategies, and effect evaluation mechanisms, a structured self-verification experimental scheme is generated, which includes experimental cycle planning, data acquisition procedures, and safety monitoring measures.

[0035] Furthermore, based on the execution results of the structured self-verification experiment, the parameters of the leverage point identification algorithm and the connection weights of the resilience assessment model are continuously updated using an adaptive learning algorithm until the execution results of the structured self-verification experiment meet preset conditions, including:

[0036] Based on the results of the structured self-validating experimental protocol, experimental validation data were extracted, which included leverage factor intervention effect indicators and resilience development assessment parameters.

[0037] The experimental verification data is input into the parameter optimization engine, and the parameter correction amount of the lever point identification algorithm is calculated through the gradient backpropagation algorithm. At the same time, the connection weights of the resilience assessment model are adjusted and updated using the weight update rule.

[0038] Based on the updated leverage point identification algorithm and resilience assessment model, the leverage point intervention priority sequence and individualized resilience development path planning are recalculated.

[0039] When the ranking stability of key leverage factors in the recalculated leverage point intervention priority sequence reaches a preset stability threshold, and the expected improvement in the individualized resilience development path planning meets the clinical goals, the execution results of the structured self-validating experimental protocol are deemed to meet the preset conditions.

[0040] Furthermore, the connection weights between the parameters of the leverage point identification algorithm and the resilience assessment model corresponding to the preset conditions are determined as the optimal parameter configuration, and the individualized health management strategy for the current stage is locked in, including:

[0041] Based on optimal parameter configuration, a personalized health management strategy is constructed through the strategy generation module. The strategy generation module generates a basic strategy framework that includes core intervention objectives, execution cycle and effect evaluation criteria based on the current leverage point intervention priority sequence and personalized resilience development path planning.

[0042] The basic strategy framework is input into the strategy optimizer, which generates individualized health management strategies that include specific implementation plans and risk control measures based on patients' historical compliance data and clinical guideline constraints through a multi-objective optimization algorithm.

[0043] The individualized health management strategy was piloted within a preset validation period, and data on changes in health status and patient feedback were collected during the strategy implementation period.

[0044] Based on health status change data and patient feedback information, the strategy effectiveness and applicability indicators are calculated through a strategy evaluation model. When the strategy effectiveness and applicability indicators reach the strategy locking threshold, a strategy locking command is triggered.

[0045] In response to the policy lock command, the current personalized health management policy is marked as verified and set as the main execution policy for the current stage. At the same time, a policy execution monitoring plan and an exception handling process are generated.

[0046] Furthermore, when the ranking stability of key leverage factors exceeds a preset stability threshold, a tiered response mechanism is activated. Based on the leverage point intervention priority sequence and the current resilience level, a precise intervention plan is generated. This precise intervention plan includes nutritional adjustment strategies, behavioral intervention plans, and psychological support measures, including:

[0047] A response level is generated based on the volatility of key leverage factors and the current resilience level. The response levels include early warning level, intervention level and emergency control level.

[0048] When an early warning level response is triggered, a primary intervention plan is generated based on the leverage point intervention priority sequence. The primary intervention plan includes dietary adjustments and basic stress management measures.

[0049] When an intervention-level response is triggered, an intermediate-level intervention plan is generated based on the current resilience level. The intermediate-level intervention plan includes targeted nutritional supplementation strategies, structured behavior adjustment plans, and professional psychological support measures.

[0050] When an emergency control-level response is triggered, the probability of health status deterioration is calculated through a risk assessment matrix based on dynamic physiological parameter sequences, environmental exposure parameters, and behavioral pattern feature vectors, and an advanced intervention plan is generated. The advanced intervention plan includes a medical nutrition support plan, a behavioral emergency intervention process, and a crisis psychological intervention mechanism.

[0051] Based on the response level, select one of the corresponding primary, intermediate, or advanced intervention programs to generate a precise intervention plan that includes a response level identifier, specific implementation steps, and expected outcome evaluation criteria.

[0052] Furthermore, based on precise intervention plans and individualized resilience development path planning, a comprehensive full-lifecycle health management plan is generated and output, including:

[0053] The nutritional adjustment strategies, behavioral intervention programs and psychological support measures in the precision intervention program are decomposed into tasks to generate a set of specific operation instructions that include execution time nodes and dosage parameters.

[0054] Based on the phased training objectives and adaptive challenge task sequence in the individualized resilience development path planning, a long-term capability development plan is formulated, which includes progressive training intensity arrangements and periodic assessment nodes.

[0055] The specific operational instruction set is aligned with the long-term capability development plan in time and space to generate a comprehensive solution framework that includes daily execution procedures and special situation response strategies.

[0056] Based on patients' historical compliance characteristics and preference data, the task arrangement logic and prompting mechanism in the comprehensive solution framework are optimized through personalized adaptation algorithms to form an individualized comprehensive health management solution.

[0057] The individualized health management solution is encoded into a standardized data exchange format and synchronously output to the patient's mobile terminal and medical monitoring platform through a secure transmission protocol;

[0058] A visual execution interface is presented on the patient's mobile terminal, and an administrator view containing progress tracking charts and alerts is generated on the medical monitoring platform.

[0059] In a second aspect, the present invention also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the method described in the first aspect.

[0060] In a third aspect, the present invention also provides an electronic device including a memory and a processor, the memory being used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method described in the first aspect.

[0061] Unlike existing technologies, the above-mentioned technical solution provides a method, medium, and device for full-life-cycle health management of enteritis patients. Based on the acquired full-life-cycle health data, it extracts key leverage factors and generates a priority sequence of leverage point interventions through a leverage point identification algorithm. Combined with the calculation results of a resilience assessment model, it generates an individualized resilience development path plan and designs a structured self-validating experimental scheme accordingly. Through the results of the scheme execution, an adaptive learning algorithm continuously updates the leverage point identification algorithm and the resilience assessment model until the optimal parameter configuration and individualized health management strategy are locked. When the stability of the key leverage factor ranking exceeds the standard, a tiered response mechanism is activated to generate a precise intervention plan, ultimately outputting a comprehensive full-life-cycle health management plan. This invention, by constructing a dynamic optimization closed loop, realizes the transformation of management strategies from static matching to dynamic adaptation, solving the shortcomings of existing methods in intervention target identification, adaptive capacity quantification, and continuous strategy optimization.

[0062] The above description of the invention is merely an overview of the technical solution of this application. In order to enable those skilled in the art to better understand the technical solution of this application and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of this application easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of this application. Attached Figure Description

[0063] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of the present invention and other related contents, and should not be considered as limitations on this application.

[0064] In the accompanying drawings of the instruction manual:

[0065] Figure 1 This is a schematic diagram illustrating steps S101 to S107 of the health management method described in the specific implementation embodiment;

[0066] Figure 2This is a schematic diagram illustrating steps S201 to S206 of the health management method described in the specific implementation embodiment;

[0067] Figure 3 This is a schematic diagram illustrating steps S301 to S307 of the health management method described in the specific implementation embodiment;

[0068] Figure 4 This is a schematic diagram illustrating steps S401 to S404 of the health management method described in the specific implementation embodiment;

[0069] Figure 5 This is a schematic diagram of the structure of the electronic device described in a specific embodiment.

[0070] The reference numerals used in the above figures are explained as follows:

[0071] 1. Electronic equipment;

[0072] 11. Memory;

[0073] 12. Processor. Detailed Implementation

[0074] To illustrate the possible application scenarios, technical principles, implementable specific solutions, and achievable objectives and effects of this application in detail, the following description, in conjunction with the listed specific embodiments and accompanying drawings, provides a detailed explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to limit the scope of protection of this application.

[0075] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The term "embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment, nor does it specifically limit its independence or connection with other embodiments. In principle, in this application, as long as there are no technical contradictions or conflicts, the technical features mentioned in each embodiment can be combined in any way to form corresponding implementable technical solutions.

[0076] Unless otherwise defined, the technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the use of related terms herein is merely for the purpose of describing particular embodiments and is not intended to limit this application.

[0077] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A exists, B exists, and A and B exist simultaneously. Additionally, the character " / " in this document generally indicates that the preceding and following objects have an "or" logical relationship.

[0078] In this application, 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 actual quantity, hierarchy or order relationship between these entities or operations.

[0079] Without further limitations, the use of terms such as “comprising,” “including,” “having,” or other similar open-ended expressions in this application is intended to cover non-exclusive inclusion, which does not exclude the presence of additional elements in a process, method, or product that includes the stated elements, such that a process, method, or product that includes a list of elements may include not only those defined elements but also other elements not expressly listed, or elements inherent to such a process, method, or product.

[0080] As understood in the Examination Guidelines, in this application, expressions such as "greater than," "less than," and "exceeding" are understood to exclude the stated number; expressions such as "above," "below," and "within" are understood to include the stated number. Furthermore, in the description of the embodiments in this application, "multiple" means two or more (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups" and "multiple times," unless otherwise explicitly specified.

[0081] In the description of the embodiments of this application, the space-related expressions used, such as "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "vertical," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," indicate the orientation or positional relationship based on the orientation or positional relationship shown in the specific embodiments or drawings. They are only for the purpose of describing the specific embodiments of this application or for the reader's understanding, and do not indicate or imply that the device or component referred to must have a specific position, a specific orientation, or be constructed or operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0082] Please see Figure 1 In a first aspect, this embodiment provides a method for full-life-cycle health management of patients with enteritis, including:

[0083] S101. Obtain the patient’s full life cycle health data, which includes dynamic physiological parameter sequences, environmental exposure parameters and behavioral pattern feature vectors. The dynamic physiological parameter sequences include at least intestinal function status indicators and psychophysiological stress indicators.

[0084] S102. Based on full life cycle health data, construct an individualized health status influencing factor map through leverage point identification algorithm, and use time series correlation analysis to extract key leverage factors from multi-dimensional data to generate a leverage point intervention priority sequence.

[0085] S103. Based on the leverage point intervention priority sequence, the physiological system buffer capacity and psychological resilience index are calculated through a resilience assessment model, and an individualized resilience development path plan is generated through multimodal data fusion.

[0086] S104. Based on the priority sequence of leverage point interventions and the individualized resilience development path planning, generate a structured self-validating experimental scheme. The structured self-validating experimental scheme includes a hypothesis formulation module, variable control strategy and effect evaluation mechanism.

[0087] S105. Based on the execution results of the structured self-verification experimental scheme, continuously update the parameters of the lever point identification algorithm and the connection weights of the resilience assessment model through an adaptive learning algorithm until the execution results of the structured self-verification experimental scheme meet the preset conditions.

[0088] S106. Determine the optimal parameter configuration by connecting the parameters of the leverage point identification algorithm and the resilience assessment model when the preset conditions are met, and lock in the individualized health management strategy for the current stage.

[0089] Furthermore, when the ranking stability of key leverage factors is detected to exceed the preset stability threshold, a graded response mechanism is activated to generate a precise intervention plan based on the leverage point intervention priority sequence and the current resilience level. The precise intervention plan includes nutritional adjustment strategies, behavioral intervention plans, and psychological support measures.

[0090] S107. Generate and output a comprehensive health management plan for the entire life cycle based on individualized health management strategies, precise intervention programs, and individualized resilience development path planning.

[0091] In step S101, it is necessary to acquire a dataset covering the patient's entire health journey, i.e., full-life-cycle health data. The dynamic physiological parameter sequence refers to physiological indicators continuously recorded over time, obtained through wearable devices or periodic monitoring. These indicators include at least intestinal function indicators directly reflecting gut health, and psychophysiological stress indicators characterizing physiological responses to psychological stress. Environmental exposure parameters are used to quantify the impact of the patient's external environment, such as air and water quality data. Behavioral pattern feature vectors are structured descriptions of the patient's lifestyle habits, such as diet, sleep, and exercise. This step, by integrating physiological, environmental, and behavioral data with a time dimension, provides a comprehensive data foundation for constructing a personalized health analysis model.

[0092] In step S102, the construction of a personalized health status influencing factor map is achieved through a leverage point identification algorithm. This algorithm identifies key influencing factors from complex multi-dimensional data that can significantly improve overall health status with relatively low intervention resources; these are the key leverage factors. Specifically, time series association analysis is used to analyze the temporal correlation patterns between dynamic physiological parameters, environmental exposure parameters, and behavioral pattern feature vectors, thereby identifying factors closely related to gut function and with high clinical operability. Finally, based on the identified correlation strength and intervention feasibility, a sequence listing the priority of intervention targets is generated, namely the leverage point intervention priority sequence. This step, through computational analysis, transforms general health influencing factors into a specific, orderly, and efficient list of personalized intervention targets.

[0093] It should be noted that the generation of the aforementioned leverage point intervention priority sequence is not a one-time process. Preferably, in actual management, the system will periodically re-execute the aforementioned data alignment, correlation analysis, network construction, and factor extraction steps based on a preset data collection cycle or the amount of newly accumulated health data, thereby dynamically updating the individual's leverage point intervention priority sequence. This periodic update mechanism provides a data foundation for subsequent monitoring of the dynamic changes in key leverage factors.

[0094] In step S103, the assessment and planning of the patient's intrinsic health capacity is achieved through a resilience assessment model. This model calculates the physiological system buffer capacity, which characterizes the physiological system's ability to absorb daily fluctuations, and the psychological resilience index, which reflects psychological adaptation and recovery capabilities. The calculation process comprehensively utilizes multi-dimensional data from step S101, such as using physiological data to assess neuroendocrine regulatory capacity and using behavioral and environmental data to assess psychological adaptation levels. Subsequently, multimodal data fusion technology integrates assessment results from different sources to generate an individualized resilience development path plan. This plan designs phased, progressive training goals and tasks for the patient, systematically enhancing their intrinsic ability to cope with health challenges. This step shifts the focus of health management from passively responding to external symptoms to proactively building and strengthening the individual's fundamental ability to maintain homeostasis and adapt to change.

[0095] In step S104, the structured self-validating experimental protocol is a personalized experimental framework for validating the intervention's effectiveness. The hypothesis-making module transforms intervention proposals targeting key leverage factors into testable, concrete propositions; the variable control strategy plans how to control other factors that might interfere with the results during the experiment; and the effectiveness evaluation mechanism clarifies the specific indicators and standards for judging the effectiveness of the intervention. This step introduces the rigor of scientific experimentation into personal health management, enabling patients to explore and validate the effectiveness of different interventions in their own practice in a relatively reliable manner, thereby providing empirical evidence for subsequent model optimization.

[0096] In step S105, the core model in the system is continuously optimized using the results of the structured self-verification experimental protocol. Specifically, through an adaptive learning algorithm, the parameters within the leverage point identification algorithm and the connection weights between various evaluation dimensions in the resilience assessment model are dynamically adjusted based on the effect data collected after the structured self-verification experimental protocol is executed. This process continues until the results of the experimental protocol reach preset stability and effectiveness standards, such as the ranking of key leverage factors no longer fluctuating significantly and the intervention effect meeting expectations. This step constructs a technological closed loop that can continuously learn and evolve from individual practice feedback, enabling the entire health management system to continuously self-adjust and become more precise as the patient's health condition changes.

[0097] In steps S106 and S107, after the iterative learning in step S105 meets the preset conditions, the corresponding model parameter configuration is determined as the current optimal parameter configuration, and a set of individualized health management strategies suitable for the current stage is generated and locked accordingly. This strategy integrates validated key intervention targets and their priorities, as well as a development path adapted to the patient's current ability level, forming a specific and executable health management plan, providing patients with clear and reliable action guidelines.

[0098] When the ranking stability of key leverage factors exceeds a preset threshold, indicating potential fluctuations in health status, a tiered response mechanism will be activated. The monitoring of the ranking stability of key leverage factors is based on the aforementioned periodically updated priority sequence of leverage point interventions. Preferably, ranking stability is quantified by comparing the ranking differences of key leverage factors in the current sequence with historical sequences, for example, by calculating the Spearman rank correlation coefficient or the mean of ranking shifts. When this stability index falls below the preset stability threshold, it indicates a significant change or new disturbance in the core factors affecting an individual's health status, at which point the tiered response mechanism is triggered.

[0099] The tiered response mechanism generates precise intervention plans based on the risk level of fluctuations and the patient's current resilience level. These plans include specific measures such as nutritional adjustments, behavioral interventions, and psychological support. Ultimately, it integrates long-term, stable, individualized health management strategies, precise intervention plans to address fluctuations, and long-term resilience development pathways to generate and output a comprehensive, full-life-cycle health management plan. This step ensures that the management plan possesses both the stability of long-term guidance and the flexibility to respond to short-term changes.

[0100] This embodiment constructs a dynamic and adaptive personalized health management closed loop. Through algorithms, it identifies key targets with significant and manageable impacts on an individual's health status from multi-dimensional health data and generates an intervention priority sequence. Simultaneously, a resilience assessment model quantifies the patient's intrinsic physiological and psychological adaptability and plans their development path. Based on this, a structured personal experimental protocol is designed, allowing patients to validate interventions targeting key targets in practice. Using feedback from these experiments, an adaptive learning algorithm continuously optimizes the core identification and assessment model until the model output is stable and reliable. Finally, validated intervention strategies, dynamic response plans to address health fluctuations, and long-term capability development plans are integrated to output a complete personalized health management plan. This embodiment realizes a transformation from static, universal health advice to a precise health management model based on dynamic analysis of individual data, practical validation, and continuous optimization.

[0101] Please see Figure 2 In some embodiments, a personalized health status influencing factor map is constructed based on full life-cycle health data using a leverage point identification algorithm. Time series association analysis is then used to extract key leverage factors from multi-dimensional data, generating a leverage point intervention priority sequence, including:

[0102] S201. Time alignment processing is performed on dynamic physiological parameter sequences, environmental exposure parameters and behavioral pattern feature vectors to form a multidimensional data matrix with unified timestamps.

[0103] S202. The multidimensional data matrix is ​​segmented and sampled using the sliding window algorithm. Within each time window, the correlation strength between the dynamic physiological parameter sequence after the unified timestamp, environmental exposure parameters, behavioral pattern feature vectors, and intestinal function status indicators is calculated to generate a dynamic correlation sequence.

[0104] S203. Construct an association network graph based on dynamic association degree sequence. The nodes of the association network graph represent dynamic physiological parameter sequences, environmental exposure parameters and behavioral pattern feature vectors after unified timestamp, denoted as node parameters. The edge weights represent the maximum mutual information values ​​between node parameters. The community detection algorithm is used to identify high-density connection subgraphs in the association network graph.

[0105] S204. Extract the node parameters with the highest centrality from the high-density connection subgraph as candidate leverage factors, and verify the strength of the causal relationship between the candidate leverage factors and the intestinal functional status indicators through a causal inference algorithm.

[0106] S205. Based on the causal relationship strength and clinical interventionability score, the candidate leverage factors are weighted and ranked to obtain the top-ranked candidate leverage factors, which are recorded as key leverage factors, and the remaining candidate leverage factors are recorded as secondary leverage factors.

[0107] S206. Generate a priority sequence of leverage point interventions that includes key leverage factors and secondary leverage factors.

[0108] In step S201, time alignment processing refers to unifying health data from different sources and with different collection frequencies to the same time point through interpolation or resampling methods, forming a dataset with a consistent timestamp. For example, environmental parameters collected daily and physiological signals collected every minute are aligned to an hourly time granularity. The resulting multidimensional data matrix with unified timestamps has each row representing a unified time point and each column representing a specific health data indicator, thus providing a well-structured data foundation for subsequent time-series correlation analysis. This step ensures the comparability of data from different dimensions on the time axis, which is a prerequisite for effective time-series analysis.

[0109] In step S202, the sliding window algorithm is used to divide continuous time series data into a series of overlapping or continuous time segments. Within each divided time window, the association strength between all node parameters (i.e., aligned health indicators) and intestinal function status indicators is calculated. Specifically, this can be achieved by calculating correlation coefficients, mutual information, or other statistical dependence measures. The dynamic association degree sequence generated by traversing all time windows reflects how the association between various health indicators and intestinal function status changes over time. This step captures the dynamic and phased nature of the associations between health indicators, avoiding the use of global static associations that ignore strong association patterns in specific periods.

[0110] In step S203, the network graph is a graphical model that intuitively displays the complex relationships between various health indicators. Nodes in the graph represent individual node parameters, and edges connect nodes with significant correlations. The weight of an edge is typically quantified by the maximum mutual information value between node parameters, which can capture both linear and non-linear dependencies. The applied community detection algorithm can be the Louvain algorithm, used to identify clusters of nodes with tight internal connections but relatively sparse external connections—i.e., high-density connection subgraphs. These subgraphs often represent a group of health indicators that change synergistically in function or influence. This step automatically discovers indicator clusters with inherent consistency from massive associations using graph theory methods, providing a structured candidate set for subsequent focused analysis.

[0111] In step S204, from the identified high-density connectivity subgraph, node parameters located at the core of the community are further screened based on node centrality indices (such as degree centrality and eigenvector centrality) and used as candidate leverage factors. Subsequently, causal inference algorithms (such as PC algorithms based on conditional independence tests or Granger causality tests using time-series information) are used to verify whether there is a causal relationship between these candidate factors and intestinal functional state indicators, and to quantify the strength of their causal effects. This step upgrades association analysis to causal inference, which can distinguish between simple statistical correlation and potential causal drivers, thereby identifying potential targets with real intervention value and ensuring that the screened factors are more likely to be causes rather than results affecting intestinal state.

[0112] In step S205, the candidate leverage factors that have passed causal verification are finally ranked. The ranking is based on a weighted composite of the causal strength and the clinical interventionability score. The causal strength is obtained in step S204; the clinical interventionability score is a predefined assessment used to measure the feasibility and safety of modulating the factor through existing clinical or behavioral interventions. A higher score indicates easier and safer intervention. The factors are ranked according to the weighted composite score, and the top-ranked factors are identified as key leverage factors, while the rest are secondary leverage factors. This step, based on scientific validity, further incorporates feasibility considerations for clinical practice, ensuring that the generated intervention priority sequence has both theoretical basis and practical operability.

[0113] In step S206, the leverage point intervention priority sequence is an ordered list that clearly identifies the key leverage factors recommended for priority intervention, as well as secondary leverage factors that can serve as subsequent alternative or synergistic intervention targets. The order of the leverage point intervention priority sequence comprehensively reflects the causal influence, dynamic association stability, and clinical intervention convenience of each factor, providing a clear, quantitative, and individualized intervention action guide for the entire health management process.

[0114] This embodiment illustrates the specific process of the leverage point identification algorithm for generating intervention priority sequences. Through multiple steps such as time alignment, dynamic correlation analysis, network community discovery, causal inference verification, and clinical feasibility weighting, it achieves automated and systematic identification and ranking of the most influential intervention targets on an individual's gut health status from raw, multi-dimensional health data. This not only improves the objectivity and efficiency of target discovery but also ensures the scientific validity and practical value of the identified leverage points by introducing causal analysis and clinical evaluation, laying a solid data-driven foundation for the subsequent development of precise individualized health management strategies.

[0115] Please see Figure 3In some embodiments, based on a leverage point intervention priority sequence, the physiological system buffer capacity and psychological resilience index are calculated using a resilience assessment model. Individualized resilience development path planning is generated through multimodal data fusion, including:

[0116] S301. Initialize the resilience assessment model based on the clinical impact weight of each key lever factor in the leverage point intervention priority sequence. The resilience assessment model includes a multi-level resilience assessment index system, which includes three assessment dimensions: nutritional metabolic reserves, neuroendocrine regulatory capacity, and psychological adaptability.

[0117] S302. Based on the heart rate variability index and cortisol diurnal rhythm data in the dynamic physiological parameter sequence, the neuroendocrine regulation capacity score is calculated by the autonomic nervous function analysis algorithm. At the same time, the nutritional metabolic reserve level is calculated based on the stability characteristics of the time series data of nutritional biomarkers.

[0118] S303. The buffer capacity of the physiological system is obtained based on the neuroendocrine regulation capacity score and the nutritional metabolic reserve level.

[0119] S304. Based on the stress coping pattern record in the behavioral pattern feature vector and the social support index in the environmental exposure parameter, calculate the psychological adaptability index through the psychological resilience assessment scale, and generate the psychological resilience index by combining the stress recovery speed parameter in the historical health data.

[0120] S305. Input the physiological system buffer capacity and psychological resilience index into a multimodal feature fusion network, and generate a comprehensive resilience assessment feature vector through an attention weighting mechanism;

[0121] S306. Based on the comprehensive resilience assessment feature vector, the optimal development path from the current state to the target state is constructed in the resilience development space of the resilience assessment model through the path planning algorithm. The optimal development path includes a sequence of phased training objectives and adaptive challenge tasks.

[0122] S307. Generate an individualized resilience development path plan based on the optimal development path. The individualized resilience development path plan includes a training intensity gradient increase scheme, key node evaluation criteria, and dynamic adjustment trigger conditions.

[0123] In step S301, the resilience assessment model is initialized based on the clinical impact weights of key leverage factors. These weights reflect a quantitative assessment of the degree of influence of different leverage factors on an individual's health status, and can be pre-set, for example, through expert scoring or statistical analysis based on historical data. The multi-level resilience assessment index system is a structured assessment framework containing three assessment dimensions corresponding to different aspects of physiological and psychological resilience: nutritional metabolic reserves assess the body's energy and material reserves; neuroendocrine regulatory capacity assesses the balance between the autonomic nervous system and the stress hormone system; and psychological adaptability assesses psychological traits and strategies for coping with stress. This step sets clear dimensions and an individualized initial focus for subsequent quantitative assessments.

[0124] In step S302, the calculation of the neuroendocrine regulation capacity score depends on specific physiological indicators. The autonomic nervous system function analysis algorithm can quantify the balance and flexibility of the autonomic nervous system through time-domain, frequency-domain, or nonlinear analysis of heart rate variability indicators; cortisol diurnal rhythm data are used to assess the rhythmic stability of the hypothalamus-pituitary-adrenal axis. The calculation of nutritional metabolic reserve levels focuses on the fluctuation characteristics of nutritional biomarkers (such as specific protein and vitamin levels) in time-series data, for example, by calculating their coefficient of variation or trend stability within a certain time window to assess the adequacy and stability of reserves. This step transforms abstract physiological regulatory capacity into computable indicators based on objective biological signals.

[0125] In step S303, the physiological system buffer capacity is a comprehensive indicator, synthesized from the neuroendocrine regulation capacity score calculated in step S302 and the nutritional metabolic reserve level through a weighted or function mapping. The physiological system buffer capacity characterizes the upper limit of an individual's physiological system's potential ability to absorb fluctuations without causing functional disorders when coping with internal and external disturbances.

[0126] In step S304, the psychological resilience index is calculated using a standardized psychological resilience assessment scale, which scores patients based on their responses to records of stress coping patterns (such as coping style questionnaires) and social support indicators (such as the perceived social support scale). Stress recovery speed parameters can be extracted from historical health data, such as the time required for relevant physiological or psychological indicators to return to baseline levels after a stressful event. The psychological resilience index, derived from the psychological resilience index and the stress recovery speed parameter, reflects the level of resilience in the psychological dimension.

[0127] In step S305, a multimodal feature fusion network is used to integrate heterogeneous data from different evaluation dimensions. The attention weighting mechanism can automatically learn and assign contribution weights of different features to the final comprehensive evaluation, thereby generating a comprehensive resilience evaluation feature vector that can fully characterize the overall resilience level of an individual.

[0128] In step S306, a path planning algorithm (such as one based on reinforcement learning or optimization) operates within a predefined resilience development space. This space, with each evaluation dimension as a coordinate axis, defines all possible development trajectories from the current resilience state (represented by a comprehensive feature vector) to the ideal target state. The algorithm searches for the trajectory that maximizes efficiency or benefit while satisfying constraints (such as safety and asymptoticity), i.e., the optimal development path, and decomposes it into specific phased training objectives and adaptive challenge tasks.

[0129] In step S307, individualized resilience development path planning is the operationalization and specification of the optimal development path. The training intensity gradient escalation scheme clarifies how the difficulty and load of the training tasks at each stage gradually increase; the key node evaluation criteria set specific indicators and thresholds for judging whether the stage goals have been achieved; and the dynamic adjustment trigger conditions specify under what circumstances (such as low task completion or sudden changes in health status) the path planning needs to be re-evaluated and adjusted.

[0130] This embodiment extracts quantifiable physiological and psychological indicators from multi-source data and utilizes multimodal fusion and path planning technologies to transform the abstract concept of "resilience" into a personalized development plan with clear assessment dimensions, quantitative indicators, and executable steps. This enables health management to shift from passively responding to symptoms to proactively and systematically enhancing the patient's intrinsic ability to maintain homeostasis.

[0131] Please see Figure 4 In some embodiments, a structured self-validating experimental scheme is generated based on the leverage point intervention priority sequence and individualized resilience development path planning, including:

[0132] S401. Based on the key leverage factors in the leverage point intervention priority sequence, candidate intervention hypotheses are constructed through a hypothesis generation algorithm. Based on the current development stage goals in the individualized resilience development path planning, candidate intervention hypotheses that meet the clinical safety boundary are selected and denoted as hypotheses to be verified.

[0133] S402. Generate variable control strategies based on the hypotheses to be verified. Variable control strategies include the setting of the baseline observation period, the implementation guidelines for intervention measures, and the confounding factor control scheme.

[0134] S403. Construct an effect evaluation mechanism, which includes a multi-dimensional effect indicator collection scheme, statistical testing methods, and clinical significance judgment criteria.

[0135] S404. Based on the hypothesis to be verified, variable control strategies, and effect evaluation mechanisms, generate a structured self-verification experimental scheme that includes experimental cycle planning, data acquisition procedures, and safety monitoring measures.

[0136] In step S401, the hypothesis generation algorithm is used to transform key leverage factors into specific, actionable intervention propositions. For example, if the key leverage factor is "insufficient sleep duration," the algorithm can generate "increase daily sleep time to 7-8 hours" as a candidate intervention hypothesis. The screening process references the current developmental stage goals defined in the individualized resilience development path planning. For instance, if the current stage focuses on improving physiological buffer capacity, hypotheses related to nutrition or physiological regulation will be prioritized. Simultaneously, all candidate hypotheses are compared with a pre-defined clinical safety boundary database, eliminating options with known risks or exceeding the patient's current tolerance. The remaining hypotheses are those to be validated.

[0137] In step S402, the variable control strategy is used to ensure the reliability of the experimental results. The baseline observation period requires a period of stable observation before the intervention to record the initial state of the target indicator. The intervention implementation guidelines specify the specific content, dosage, frequency and execution method of the intervention to ensure the consistency of the intervention. The confounding factor control scheme is used to identify and fix other factors that may affect the results as much as possible. For example, patients are required to maintain relatively stable diet, exercise and other lifestyle habits during the experiment, or these factors are recorded for statistical correction during data analysis.

[0138] In step S403, the effect evaluation mechanism defines how to scientifically assess the intervention's effectiveness. The multidimensional effect indicator collection protocol specifies the indicators to be monitored, which may include not only changes in the target leverage factor itself but also related physiological, psychological, and quality-of-life indicators. Statistical tests are used to analyze the collected data, such as paired t-tests to compare differences before and after the intervention. Clinical significance criteria set thresholds for improvements with actual health value, such as a specific percentage reduction in symptom scores or indicators returning to normal ranges, rather than simply statistically significant differences.

[0139] In step S404, the structured self-validating experimental protocol integrates and concretizes all the aforementioned elements. The experimental cycle plan clarifies the total duration of the experiment and the time allocation for each phase (such as the baseline period, intervention period, and follow-up period). The data acquisition procedure details the measurement tools, measurement time points, and recording methods for each indicator. The safety monitoring measures establish contingency plans and response procedures for monitoring adverse reactions or health risks.

[0140] This embodiment transforms the identified intervention targets into a structured experimental protocol that patients can execute individually through systematic hypothesis generation and safety screening, rigorous variable control design, and a multi-dimensional effect evaluation mechanism. This allows patients to explore and verify the effectiveness of personalized interventions in their daily lives in a relatively rigorous and reliable manner, providing a high-quality empirical data foundation for the subsequent learning and optimization of the entire health management system.

[0141] In some embodiments, based on the execution results of the structured self-verification experiment, the parameters of the leverage point identification algorithm and the connection weights of the resilience assessment model are continuously updated using an adaptive learning algorithm until the execution results of the structured self-verification experiment meet preset conditions, including:

[0142] Based on the results of the structured self-validating experimental protocol, experimental validation data were extracted, which included leverage factor intervention effect indicators and resilience development assessment parameters.

[0143] The experimental verification data is input into the parameter optimization engine, and the parameter correction amount of the lever point identification algorithm is calculated through the gradient backpropagation algorithm. At the same time, the connection weights of the resilience assessment model are adjusted and updated using the weight update rule.

[0144] Based on the updated leverage point identification algorithm and resilience assessment model, the leverage point intervention priority sequence and individualized resilience development path planning are recalculated.

[0145] When the ranking stability of key leverage factors in the recalculated leverage point intervention priority sequence reaches a preset stability threshold, and the expected improvement in the individualized resilience development path planning meets the clinical goals, the execution results of the structured self-validating experimental protocol are deemed to meet the preset conditions.

[0146] In this embodiment, the experimental validation data are quantitative results collected during the execution of a structured self-validating experimental protocol. Among them, the leverage factor intervention effect index is used to measure whether the intervention measures targeting specific key leverage factors are effective, such as the change in the value of the factor itself before and after the intervention or the degree of improvement of related symptoms; the resilience development assessment parameter is used to assess whether the patient's physiological system buffer capacity or psychological resilience index has changed in the expected direction during the execution of the experimental protocol.

[0147] The parameter optimization engine is the computational module responsible for model updates. The gradient backpropagation algorithm calculates the error between the model's predictions and the observed experimental data, and propagates this error back along the model's computational graph to determine the direction and magnitude of parameter adjustments needed within the lever point identification algorithm. Simultaneously, for the resilience assessment model, pre-defined weight update rules, such as error-based gradient descent, are used to adjust its connection weights. This process enables targeted fine-tuning of model parameters based on individual feedback data.

[0148] Based on the updated model, the leverage point identification and resilience assessment process is re-executed to generate an updated leverage point intervention priority sequence and individualized resilience development path planning, thereby updating the understanding of individual health based on the latest empirical data.

[0149] The preset conditions include two core judgment criteria: the ranking stability threshold can be measured by calculating the similarity of the ranking order of key lever factors in consecutive iterations, for example, using Kendall's harmony coefficient. When this coefficient exceeds the set threshold, the ranking is considered stable; the expected improvement range meets clinical goals, meaning that the degree of health status improvement predicted by the newly generated individualized resilience development path plan reaches the minimum effective threshold set based on clinical consensus or individualized goals. When both conditions are met simultaneously, the learning process is considered to have converged, and the current model parameters can stably and effectively reflect the individual characteristics of the patient.

[0150] This embodiment extracts the quantitative results of individual experiments as training data, uses optimization algorithms such as gradient backpropagation to update the parameters of the core model, and sets convergence criteria based on ranking stability and clinical improvement goals. This adaptive learning closed-loop mechanism enables the entire health management system to continuously evolve based on each patient's real intervention feedback data, constantly approaching the model state that best reflects the individual's health status and response patterns, thereby achieving continuous personalization and precision in management strategies.

[0151] In some embodiments, the connection weights between the parameters of the leverage point identification algorithm and the resilience assessment model corresponding to the satisfied preset conditions are determined as the optimal parameter configuration, and the individualized health management strategy for the current stage is locked, including:

[0152] Based on optimal parameter configuration, a personalized health management strategy is constructed through the strategy generation module. The strategy generation module generates a basic strategy framework that includes core intervention objectives, execution cycle and effect evaluation criteria based on the current leverage point intervention priority sequence and personalized resilience development path planning.

[0153] The basic strategy framework is input into the strategy optimizer, which generates individualized health management strategies that include specific implementation plans and risk control measures based on patients' historical compliance data and clinical guideline constraints through a multi-objective optimization algorithm.

[0154] The individualized health management strategy was piloted within a preset validation period, and data on changes in health status and patient feedback were collected during the strategy implementation period.

[0155] Based on health status change data and patient feedback information, the strategy effectiveness and applicability indicators are calculated through a strategy evaluation model. When the strategy effectiveness and applicability indicators reach the strategy locking threshold, a strategy locking command is triggered.

[0156] In response to the policy lock command, the current personalized health management policy is marked as verified and set as the main execution policy for the current stage. At the same time, a policy execution monitoring plan and an exception handling process are generated.

[0157] In this embodiment, the strategy generation module is a component that transforms the output of a validated model into an executable plan. It determines the core intervention objectives that need to be prioritized based on the current leverage point intervention priority sequence, such as improving specific key leverage factors; simultaneously, it sets reasonable execution cycles and phased effect evaluation standards with reference to individualized resilience development path planning, thereby forming a basic strategy framework.

[0158] The strategy optimizer is responsible for refining and personalizing the underlying framework. Patient historical adherence data reflects their execution of past health recommendations and can be used to predict and mitigate potential adherence barriers. Clinical guideline constraints ensure the medical safety of the generated strategy. Multi-objective optimization algorithms, such as those based on genetic algorithms or particle swarm optimization, search under these constraints to balance multiple objectives, including intervention effectiveness, feasibility, and patient burden, ultimately outputting a personalized health management strategy that includes specific daily action plans, resource recommendations, and risk warning measures.

[0159] The trial run involves having patients actually implement the strategy within a relatively short, pre-defined validation period. During this time, the system continuously collects data on changes in health status, such as trends in relevant physiological indicators, as well as feedback from patients via questionnaires or interactive interfaces regarding the acceptability and difficulty of the strategy.

[0160] The strategy evaluation model is used to quantify the effectiveness of the trial run. Strategy effectiveness indicators can be calculated based on the degree of improvement in health status data; applicability indicators can be derived by integrating patient feedback, such as satisfaction and perceived burden. When the calculated values ​​of these indicators reach a pre-set strategy locking threshold, the system determines that the strategy is effective and acceptable at the current stage, thereby triggering a strategy locking command.

[0161] Upon responding to the lockout command, the status of the personalized health management strategy is updated to verified, and it is officially activated as the primary execution strategy for the current phase. Simultaneously, the system generates a corresponding strategy execution monitoring plan, clearly defining the indicators and frequencies that need to be monitored regularly, and establishing anomaly handling procedures to address any unexpected situations or unsatisfactory results that may occur during strategy execution.

[0162] This embodiment ensures that the output strategy is not only based on data-driven and model optimization, but also tested and adjusted in real-world execution through a complete process of strategy generation, multi-objective optimization, trial operation verification, and final monitoring. It transforms the learned and optimized model parameters into a stable, executable, and preliminarily verified individualized health management strategy, thereby significantly improving the reliability, operability, and patient acceptance of the strategy, and providing a stable basis for action in long-term health management.

[0163] In some embodiments, when the ranking stability of key leverage factors exceeds a preset stability threshold, a tiered response mechanism is activated to generate a precise intervention plan based on the leverage point intervention priority sequence and the current resilience level. The precise intervention plan includes nutritional adjustment strategies, behavioral intervention plans, and psychological support measures, including:

[0164] A response level is generated based on the volatility of key leverage factors and the current resilience level. The response levels include early warning level, intervention level and emergency control level.

[0165] When an early warning level response is triggered, a primary intervention plan is generated based on the leverage point intervention priority sequence. The primary intervention plan includes dietary adjustments and basic stress management measures.

[0166] When an intervention-level response is triggered, an intermediate-level intervention plan is generated based on the current resilience level. The intermediate-level intervention plan includes targeted nutritional supplementation strategies, structured behavior adjustment plans, and professional psychological support measures.

[0167] When an emergency control-level response is triggered, the probability of health status deterioration is calculated through a risk assessment matrix based on dynamic physiological parameter sequences, environmental exposure parameters, and behavioral pattern feature vectors, and an advanced intervention plan is generated. The advanced intervention plan includes a medical nutrition support plan, a behavioral emergency intervention process, and a crisis psychological intervention mechanism.

[0168] Based on the response level, select one of the corresponding primary, intermediate, or advanced intervention programs to generate a precise intervention plan that includes a response level identifier, specific implementation steps, and expected outcome evaluation criteria.

[0169] In this embodiment, the fluctuation range of key leverage factors can be quantified by calculating the standard deviation or coefficient of variation of their values ​​within a recent time window; the current resilience level is a comprehensive state of the physiological system's buffer capacity and psychological resilience index calculated by the resilience assessment model. Based on preset threshold rules, the risk state is divided into warning level, intervention level, and emergency control level. The preset threshold rules are set based on clinical experience and historical data analysis. Preferably, the fluctuation range (e.g., coefficient of variation) of key leverage factors and the percentile of resilience level can be used as the classification criteria: when the fluctuation range is below a first threshold and the resilience level is above a certain percentile, it is a warning level; when the fluctuation range is between the first and second thresholds, or the resilience level is low, it is an intervention level; when the fluctuation range exceeds the second threshold and is accompanied by a significant deterioration trend in physiological indicators, it is an emergency control level.

[0170] When a warning-level response is triggered, it indicates a slight fluctuation in health status. The initial intervention plan generated at this point allows for early, low-intensity intervention. Its content is directly based on the leverage point intervention priority sequence, such as providing fine-tuning suggestions for top-ranked nutrition or behavioral leverage factors, and recommending basic stress relaxation techniques.

[0171] When an intervention-level response is triggered, it indicates that the fluctuations are already quite significant. The resulting intermediate-level intervention plan needs to be customized based on the patient's current resilience level. For example, for patients with low resilience levels, the behavior modification plan should focus more on feasibility and support; for patients with moderate resilience levels, more challenging structured tasks can be introduced. The plan includes more targeted nutritional supplementation recommendations, structured behavior modification steps that require more effort from the patient, and psychological support measures that may involve professional guidance.

[0172] When an emergency regulatory response is triggered, it indicates a high risk of health deterioration. At this point, a risk assessment matrix is ​​used to comprehensively analyze dynamic physiological parameter sequences, environmental exposure parameters, and behavioral pattern feature vectors to calculate a quantified probability of health deterioration. Advanced intervention programs generated based on this probability include medical nutrition support programs requiring intervention or supervision by healthcare professionals, emergency intervention procedures to immediately block risky behaviors or environments, and intervention mechanisms to address acute psychological crises.

[0173] Finally, based on the determined response level, the system selects the corresponding plan from the three levels mentioned above, and attaches a clear response level label, detailed and actionable specific execution steps, and expected effect evaluation criteria for judging whether the intervention is effective, thus forming a complete precision intervention plan.

[0174] This embodiment describes how to dynamically generate response plans of varying intensity and professionalism based on the severity of the fluctuations and the patient's own capabilities when health status fluctuates. Through a tiered response and precise matching mechanism, it ensures that health management not only possesses long-term planning capabilities but also flexibly, promptly, and appropriately addresses short-term risks. This achieves an organic combination of homeostasis maintenance and emergency response, enhancing the robustness and practicality of the health management system.

[0175] In some embodiments, a comprehensive health management plan for the entire life cycle is generated and output based on a precision intervention plan and an individualized resilience development path plan, including:

[0176] The nutritional adjustment strategies, behavioral intervention programs and psychological support measures in the precision intervention program are decomposed into tasks to generate a set of specific operation instructions that include execution time nodes and dosage parameters.

[0177] Based on the phased training objectives and adaptive challenge task sequence in the individualized resilience development path planning, a long-term capability development plan is formulated, which includes progressive training intensity arrangements and periodic assessment nodes.

[0178] The specific operational instruction set is aligned with the long-term capability development plan in time and space to generate a comprehensive solution framework that includes daily execution procedures and special situation response strategies.

[0179] Based on patients' historical compliance characteristics and preference data, the task arrangement logic and prompting mechanism in the comprehensive solution framework are optimized through personalized adaptation algorithms to form an individualized comprehensive health management solution.

[0180] The individualized health management solution is encoded into a standardized data exchange format and synchronously output to the patient's mobile terminal and medical monitoring platform through a secure transmission protocol;

[0181] A visual execution interface is presented on the patient's mobile terminal, and an administrator view containing progress tracking charts and alerts is generated on the medical monitoring platform.

[0182] In this embodiment, task decomposition can be achieved by parsing the strategy text using natural language processing technology, or by matching and instantiating based on a predefined task template library. For example, a nutrition adjustment strategy can be decomposed into a specific ingredient purchase list, meal recipes, and cooking instructions; a behavioral intervention plan can be decomposed into daily check-in tasks and action guidelines. The specific operation instruction set generated after decomposition will specify the execution date, time, and dosage parameters for each task, such as exercise duration and supplement dosage.

[0183] The development of a long-term capability development plan transforms the abstract, phased goals of individualized resilience development pathway planning into concrete training arrangements with a timeline. Progressive training intensity means that the difficulty or load of training tasks gradually increases over time; for example, exercise intensity increases from low to high, and psychological challenges progress from simple to complex. Periodic assessment nodes are set at key points in the plan to evaluate progress in capability development using pre-designed assessment tools.

[0184] Spatiotemporal alignment is designed to address potential conflicts or overlaps in time and content between short-term intervention tasks and long-term development plans. This process uses a scheduling algorithm to rationally embed daily tasks from the specific operational instruction set into the weekly or monthly framework of the long-term capacity development plan, while considering task compatibility and patient energy allocation. Ultimately, this results in a unified, comprehensive framework that includes daily execution procedures and contingency plans for special circumstances such as holidays and illnesses.

[0185] Personalized adaptation algorithms are used to fine-tune the above framework to improve patient experience and adherence. The algorithm analyzes a patient's historical adherence characteristics, such as which time of day they are more likely to complete tasks or which types of tasks they complete more efficiently; it also considers their explicitly stated preferences, such as dietary restrictions or preferred types of exercise. Based on this information, the algorithm optimizes the task order and timing, and adjusts the timing and method of reminder notifications, thereby creating a more personalized and comprehensive health management plan that better suits the patient's individual habits and preferences.

[0186] The output phase of the solution involves data encoding and synchronization. The personalized health management comprehensive solution is encoded into a standardized data exchange format, such as JSON or XML, to ensure compatibility across different systems. Subsequently, the solution data is synchronously pushed to the patient's mobile terminal and the backend medical monitoring platform via a secure transmission protocol, such as HTTPS.

[0187] At the presentation level, the visual execution interface on the patient's mobile device typically displays daily tasks and long-term plan progress in the form of a calendar, task list, or progress bar. The administrator view of the medical monitoring platform focuses on macro-level monitoring, generating tracking charts that show the patient's overall progress and triggering alerts when declining adherence or abnormal health data are detected.

[0188] This embodiment integrates dynamically generated precise intervention plans with long-term resilience development plans, transforming them into a highly structured, operable, and personalized final health management plan. Through a series of steps including task decomposition, plan formulation, spatiotemporal alignment, personalized adaptation, and multi-device synchronous presentation, it ensures a seamless connection between macro-strategy and micro-implementation, providing clear daily action guidelines and supporting effective remote monitoring by healthcare professionals. This achieves the practical implementation and closed-loop management of the health management plan.

[0189] In a second aspect, this embodiment also provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the method described in the first aspect.

[0190] The computer program involved in this embodiment can be stored in a computer device readable storage medium, which includes, but is not limited to, disks, magnetic tapes, magnetic cards, floppy disks, flash memory, optical disks, optical cards, read-only memory (ROM), random access memory (RAM), erasable programmable ROM (EPROM), and electrically erasable programmable ROM (EEPROM), etc. It also includes other biological, physical, or chemical structures capable of performing similar or equivalent functions to the storage media listed above, such as DNA, RNA, proteins, and other units with information storage capabilities. In specific embodiments, the storage medium involved can be one of the above-mentioned media types or a combination of the above media types. In different embodiments, the computer program involved in the embodiment can be centrally stored in a single medium or distributed across multiple media. The memory containing the computer device readable storage medium can be non-volatile memory or random access memory. These computer device readable storage media can be built into the device or connected to the device involved in the embodiment as an external device or part of an external device. In some embodiments, the memory having a computer device readable storage medium is deployed locally; in other embodiments, the memory may be deployed remotely from the processor, for example, as a network-attached memory accessed via RF circuitry or an external port and a communication network, wherein the communication network may be the Internet, one or more intranets, a local area network (LAN), a wide area network (WLAN), a storage area network (SAN), or a suitable combination thereof, as long as computer device access to the memory is enabled. Furthermore, the computer program involved in the embodiments may be stored in plaintext / ciphertext form, or it may be designed as training data, integrated and recombined through model training and implicitly stored in the parameter states of a deep neural network or other machine learning model.

[0191] Please see Figure 5In a third aspect, this embodiment also provides an electronic device 1, including a memory 11 and a processor 12, wherein the memory 11 is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor 12 to implement the method described in the first aspect.

[0192] The processor described in this embodiment can be implemented by hardware, firmware, software, or a combination thereof. It can be a circuit, one or more of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, or a microprocessor. It also includes other physical, biological, or chemical structures that can implement the same or equivalent functions as the processors listed above, such as biological neurons, quantum computing units, DNA computing units, etc., so that the processor can execute some or all of the steps in the computer program or method involved in the various embodiments of this application, or any combination of the steps mentioned therein.

[0193] By adopting the above technical solutions, this invention differs from existing technologies by constructing a dynamic and adaptive personalized health management closed loop. This solution automatically identifies and prioritizes the most influential interventional targets—key leverage factors—from multi-dimensional health data through a leverage point identification algorithm, thereby transforming general health influencing factors into a specific, orderly, and efficient list of personalized intervention targets. Simultaneously, it quantifies the patient's physiological system buffer capacity and psychological resilience index through a resilience assessment model, generating an individualized resilience development path plan. This shifts the focus of health management from passively responding to external symptoms to proactively building and enhancing the individual's intrinsic capacity to maintain homeostasis and adapt to change.

[0194] Furthermore, a structured self-validating experimental protocol is introduced, enabling patients to explore and verify the effectiveness of interventions in their own practice in a relatively rigorous manner. Using this empirical data, core model parameters are continuously optimized through adaptive learning algorithms, allowing management strategies to continuously adjust and become more precise as the patient's health condition changes. Finally, the system integrates validated individualized health management strategies, tiered response plans to health fluctuations, and long-term capacity development plans, outputting a comprehensive life-cycle health management solution that combines long-term guidance stability with short-term response flexibility. This technical solution represents a fundamental shift from static, universal health advice to a precise health management model based on dynamic analysis of individual data, practical validation, and continuous optimization, effectively improving the targeting, adaptability, and long-term effectiveness of health management.

[0195] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. A method for whole life cycle health management of an enteritis patient, characterized by, The method comprises the following steps: acquiring full-life-cycle health data of a patient, the full-life-cycle health data comprising a dynamic physiological parameter sequence, an environmental exposure parameter, and a behavior pattern feature vector, the dynamic physiological parameter sequence at least containing an intestinal function state indicator and a psychophysiological stress indicator; constructing an individualized health state influence factor graph based on the full-life-cycle health data by a leverage point identification algorithm, extracting key leverage factors from multidimensional data by time series correlation analysis to generate a leverage point intervention priority sequence; calculating physiological system buffer capacity and psychological elasticity index based on the leverage point intervention priority sequence by a resilience evaluation model, and generating an individualized resilience development path plan by multi-modal data fusion; generating a structured self-verification experiment scheme according to the leverage point intervention priority sequence and the individualized resilience development path plan, the structured self-verification experiment scheme comprising a hypothesis proposal module, a variable control strategy, and an effect evaluation mechanism; continuously updating parameters of the leverage point identification algorithm and connection weights of the resilience evaluation model based on execution results of the structured self-verification experiment scheme by an adaptive learning algorithm until the execution results of the structured self-verification experiment scheme meet preset conditions; determining the parameters of the leverage point identification algorithm and the connection weights of the resilience evaluation model corresponding to the time when the preset conditions are met as optimal parameter configurations, and locking the individualized health management strategy of the current stage; when it is monitored that the ordering stability of the key leverage factors exceeds a preset stability threshold, starting a hierarchical response mechanism, and generating a precise intervention scheme according to the leverage point intervention priority sequence and the current resilience level, the precise intervention scheme comprising a nutrition adjustment strategy, a behavior intervention scheme, and a psychological support measure; generating a full-life-cycle health management comprehensive scheme according to the individualized health management strategy, the precise intervention scheme, and the individualized resilience development path plan and outputting the full-life-cycle health management comprehensive scheme; constructing an individualized health state influence factor graph based on the full-life-cycle health data by a leverage point identification algorithm, extracting key leverage factors from multidimensional data by time series correlation analysis to generate a leverage point intervention priority sequence, comprising: performing time alignment processing on the dynamic physiological parameter sequence, the environmental exposure parameter, and the behavior pattern feature vector to form a multidimensional data matrix with unified timestamps; performing segmented sampling on the multidimensional data matrix by a sliding window algorithm, calculating the correlation strength of the dynamic physiological parameter sequence, the environmental exposure parameter, and the behavior pattern feature vector after unified timestamping and the intestinal function state indicator in each time window to generate a dynamic correlation degree sequence; constructing an association network graph based on the dynamic correlation degree sequence, the nodes of the association network graph representing the dynamic physiological parameter sequence, the environmental exposure parameter, and the behavior pattern feature vector after unified timestamping, denoted as node parameters, the edge weights representing the maximum mutual information values between the node parameters, and a community discovery algorithm is used to identify a high-density connected subgraph in the association network graph; extracting the node parameter with the highest centrality from the high-density connected subgraph as a candidate leverage factor, and verifying the causal relationship strength between the candidate leverage factor and the intestinal function state indicator by a causal inference algorithm; According to the causal relationship strength and the clinical intervention score, candidate leverage factors are weighted and ranked to obtain a number of candidate leverage factors ranked at the top, denoted as key leverage factors, and the remaining candidate leverage factors are denoted as secondary leverage factors; A leverage point intervention priority sequence containing the key leverage factors and the secondary leverage factors is generated.

2. The enteritis patient life cycle health management method of claim 1, wherein, Based on the leverage point intervention priority sequence, a physiological system buffer capacity and a psychological resilience index are calculated through a resilience evaluation model, and an individualized resilience development path planning is generated through multi-modal data fusion, including: According to the clinical influence weight of each key leverage factor in the leverage point intervention priority sequence, the resilience evaluation model is initialized, and the resilience evaluation model includes a multi-level resilience evaluation index system, and the multi-level resilience evaluation index system includes three evaluation dimensions of nutritional metabolism reserve, neuroendocrine regulation ability and psychological adaptation ability; Based on the heart rate variability index and the cortisol circadian rhythm data in the dynamic physiological parameter sequence, the neuroendocrine regulation ability score is calculated through an autonomic nervous function analysis algorithm, and the nutritional metabolism reserve level is calculated based on the stability characteristics of the nutritional biomarker time series data; According to the neuroendocrine regulation ability score and the nutritional metabolism reserve level, the physiological system buffer capacity is obtained; Based on the stress coping mode record in the behavior mode feature vector and the social support index in the environmental exposure parameter, the psychological adaptation ability index is calculated through a psychological resilience evaluation scale, and the psychological resilience index is generated in combination with the stress recovery speed parameter in the historical health data; The physiological system buffer capacity and the psychological resilience index are input into a multi-modal feature fusion network to generate a comprehensive resilience evaluation feature vector through an attention weighting mechanism; Based on the comprehensive resilience evaluation feature vector, an optimal development path from the current state to the target state is constructed in the resilience development space of the resilience evaluation model through a path planning algorithm, and the optimal development path includes a phased training target and an adaptive challenge task sequence; According to the optimal development path, an individualized resilience development path planning is generated, and the individualized resilience development path planning includes a training intensity gradient increasing scheme, a key node evaluation standard and a dynamic adjustment trigger condition. 3.The method of claim 1, wherein, According to the leverage point intervention priority sequence and the individualized resilience development path planning, a structured self-verification experiment scheme is generated, including: Based on the key leverage factors in the leverage point intervention priority sequence, candidate intervention hypotheses are constructed through a hypothesis generation algorithm, and candidate intervention hypotheses that meet the clinical safety boundary are selected as verified hypotheses according to the current development stage target in the individualized resilience development path planning. According to the verified hypotheses, a variable control strategy is generated, and the variable control strategy includes a baseline observation period setting, an intervention measure implementation specification and a confounding factor control scheme; An effect evaluation mechanism is constructed, and the effect evaluation mechanism includes a multi-dimensional effect index collection scheme, a statistical test method and a clinical significance judgment standard; Based on the verified hypotheses, the variable control strategy and the effect evaluation mechanism, a structured self-verification experiment scheme containing an experiment cycle planning, a data collection procedure and a safety monitoring measure is generated. 4.The method of claim 1, wherein, Based on the execution result of the structured self-verification experiment scheme, the parameters of the leverage point identification algorithm and the connection weights of the resilience evaluation model are continuously updated by an adaptive learning algorithm until the execution result of the structured self-verification experiment scheme meets the preset condition, including: Based on the execution result of the structured self-verification experiment scheme, extract experimental verification data, which contains leverage factor intervention effect indicators and resilience development evaluation parameters; Input the experimental verification data into the parameter optimization engine, calculate the parameter correction amount of the leverage point identification algorithm by the gradient back propagation algorithm, and adjust the connection weights of the resilience evaluation model and update it at the same time; Based on the updated leverage point identification algorithm and resilience evaluation model, recalculate the leverage point intervention priority sequence and individualized resilience development path planning; When the sorting stability of the key leverage factor in the recalculated leverage point intervention priority sequence reaches the preset stability threshold, and the expected improvement amplitude in the individualized resilience development path planning meets the clinical target, it is determined that the execution result of the structured self-verification experiment scheme meets the preset condition. 5.The method of claim 4, wherein the patient is a patient with enteritis. When the corresponding parameters of the leverage point identification algorithm and the connection weights of the resilience evaluation model meet the preset condition are determined as the optimal parameter configuration, and the individualized health management strategy of the current stage is locked, including: Based on the optimal parameter configuration, build individualized health management strategy through strategy generation module, the strategy generation module generates basic strategy framework containing core intervention target, execution cycle and effect evaluation standard according to current leverage point intervention priority sequence and individualized resilience development path planning; Input the basic strategy framework into the strategy optimizer, generate individualized health management strategy containing specific execution scheme and risk control measures through multi-objective optimization algorithm according to patient historical compliance data and clinical guideline constraint conditions; Run the individualized health management strategy in the preset verification period, collect health status change data and patient feedback information during strategy execution; Based on the health status change data and patient feedback information, calculate the strategy effectiveness and applicability indicators through the strategy evaluation model, and trigger the strategy locking instruction when the strategy effectiveness and applicability indicators reach the strategy locking threshold; In response to the strategy locking instruction, mark the current individualized health management strategy as verified state, and set it as the main execution strategy of the current stage, and generate strategy execution monitoring scheme and abnormal processing flow at the same time. 6.The method of Claim 1, wherein When it is monitored that the sorting stability of the key leverage factor exceeds the preset stability threshold, start the hierarchical response mechanism, generate the precise intervention scheme according to the leverage point intervention priority sequence and the current resilience level, the precise intervention scheme contains nutrition adjustment strategy, behavior intervention scheme and psychological support measures, including: Generate response level based on the fluctuation amplitude of key leverage factor and current resilience level, the response level includes warning level, intervention level and emergency control level; When the warning level response is triggered, generate the primary intervention scheme based on the leverage point intervention priority sequence, the primary intervention scheme contains diet fine-tuning suggestion and basic stress management measures; When the intervention level response is triggered, a medium intervention scheme is generated in combination with the current resilience level, the medium intervention scheme including targeted nutritional supplement strategies, structured behavior adjustment schemes, and specialized psychological support measures; When the emergency regulation level response is triggered, based on the dynamic physiological parameter sequence, environmental exposure parameters, and behavior pattern feature vectors, the health state deterioration probability is calculated through a risk assessment matrix, and a high-level intervention scheme is generated, the high-level intervention scheme including medical nutritional support schemes, behavior emergency intervention processes, and crisis psychological intervention mechanisms; According to the selection of the response level, one of the primary intervention scheme, the medium intervention scheme, and the high-level intervention scheme is selected to generate a precise intervention scheme including response level identification, specific execution steps, and expected effect evaluation standards. 7.The method of Claim 1, wherein According to the individualized health management strategy, the precise intervention scheme, and the individualized resilience development path planning, the whole life cycle health management comprehensive scheme is generated and output, including: The nutritional adjustment strategies, behavior intervention schemes, and psychological support measures in the precise intervention scheme are task decomposed to generate a specific operation instruction set including execution time nodes and dose parameters; Based on the stage training targets and adaptive challenge task sequences in the individualized resilience development path planning, a long-term ability development plan is formulated, including progressive training intensity arrangement and periodic evaluation nodes; The specific operation instruction set and the long-term ability development plan are spatio-temporally aligned to generate a comprehensive scheme framework including daily execution procedures and special situation coping strategies; Based on the patient's historical compliance characteristics and preference data, the task arrangement logic and prompt mechanism in the comprehensive scheme framework are optimized through a personalized adaptation algorithm to form an individualized health management comprehensive scheme; The individualized health management comprehensive scheme is encoded into a standardized data exchange format and output to the patient mobile terminal and the medical monitoring platform through a secure transmission protocol; A visual execution interface is presented on the patient mobile terminal, and an administrator view including progress tracking charts and early warning prompts is generated on the medical monitoring platform.

8. A computer readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions, when executed by the processor, implement the method of any one of claims 1-7.

9. An electronic device comprising a memory and a processor, characterized in that, The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method of any one of claims 1-7.

Images