Food quantification method and system for digestive nursing patient

Abstract

The invention relates to the technical field of data processing and data analysis, in particular to a food quantification method and system for a digestive nursing patient, and the method comprises the following steps: obtaining multi-source heterogeneous time sequence data related to the adjustment of the food of the digestive nursing patient, and carrying out the integration and standardization processing of the obtained multi-source heterogeneous time sequence data, the data after integration and standardization processing are obtained; on the basis of the integrated and normalized data, individual time sequence correlation characteristics between meal adjustment events and subsequent physiological reactions or symptom changes of the digestive nursing patient are mined, and the individual time sequence correlation characteristics are obtained; the problems that real-time or near-real-time individualized adjustment is difficult to realize and an effective feedback optimization mechanism is lacked in the prior art are solved by acquiring multi-source data, mining individualized time sequence correlation features, setting dynamic influence factor weights, generating individualized suggestions and optimizing the weights based on feedback iteration.

Description

Technical Field

[0001] This invention relates to the fields of data processing and data analysis, and in particular to a method and system for quantitatively measuring meals for patients in digestive care. Background Technology

[0002] During the recovery period, patients in digestive care often experience unstable digestive function, requiring strict and dynamically changing dietary needs regarding the type, form, and amount of food consumed per meal. Despite initial individualized meal plans, patients' digestive and absorptive capacity, food tolerance, and appetite are easily affected by various factors and fluctuate rapidly, making it difficult for static plans to match immediate needs and transient digestive capacity. Failure to dynamically adjust the amount of food according to the patient's condition can lead to problems: continuing the planned meals when digestive capacity declines may result in insufficient intake; failing to increase the amount of food when function improves may miss opportunities for nutritional supplementation; and failing to adjust the form or reduce the amount of food when sensitivity increases may induce discomfort such as bloating and diarrhea, affecting comfort and delaying recovery.

[0003] In rehabilitation facilities, multi-dimensional and heterogeneous data are typically collected to manage patients more precisely. This data includes baseline meal plans set by nutritionists, preparation records from the central kitchen, actual food intake and acceptance recorded by nursing staff, subjective symptoms and feelings reported by patients through interactive terminals, and objective physiological parameters (such as heart rate variability, body surface temperature, and activity level) continuously collected through non-invasive devices. Together, these constitute a complex time-series dataset describing the patient's condition.

[0004] However, traditional meal quantity adjustment processes mainly rely on nutritionists periodically reviewing and analyzing these scattered data, and then manually adjusting the plan. This method is slow and difficult to achieve real-time individualization. The core technical challenge in achieving individualized dynamic meal quantity adjustment lies in effectively integrating, cleaning, and synchronizing this scattered and inconsistently formatted data, and extracting reliable temporal correlations between individualized meal adjustment events and subsequent physiological responses or symptom changes. Based on these patterns and the patient's immediate condition data, the system can dynamically generate precise quantitative adjustment suggestions for the next meal or subsequent meals of the day. Furthermore, even if adjustment suggestions can be generated, establishing an effective mechanism to quickly and structurally record and evaluate the patient's immediate physiological responses and symptom changes after adjustment as feedback, and using this feedback information to continuously optimize the influence weights and correlation rules of different data factors in the adjustment logic, forming an efficient "monitoring-analysis-adjustment-feedback-optimization" closed-loop management process, is also a challenge for current technologies. If the evaluation and feedback optimization of adjustment effects still mainly rely on manual experience, the efficiency, accuracy, and adaptability of the entire dynamic quantitative adjustment system will be severely limited.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for quantitatively measuring meals for patients in digestive care.

[0007] In a first aspect, the present invention provides a method for quantitatively measuring meals for patients receiving digestive care, applicable to scenarios where these patients receive nutritional support in rehabilitation institutions. The method includes the following steps: Acquire multi-source heterogeneous time-series data related to the dietary adjustments of the patients receiving digestive care, and integrate and standardize the acquired multi-source heterogeneous time-series data to obtain integrated and standardized data; Based on the integrated and standardized data, individualized temporal correlation features between meal adjustment events and subsequent physiological responses or symptom changes in patients with digestive care are mined to obtain the individualized temporal correlation features. Based on the individualized time-series correlation features and the preset basic rules for meal adjustment, dynamic influence factor weights are set for multiple data factors that affect the generation of quantitative meal adjustment suggestions, and the dynamic influence factor weights are obtained. Based on the real-time status data of patients receiving digestive care, and combined with the weights of the dynamic influencing factors, an individualized meal quantity adjustment suggestion is generated for the patient's next meal or subsequent meals of the day, thus obtaining the individualized meal quantity adjustment suggestion; After the implementation of the individualized meal quantity adjustment recommendations, the physiological responses and symptom changes of patients receiving digestive care are tracked as feedback on the adjustment effect, and the weights of the dynamic influencing factors are iteratively updated based on the adjustment effect feedback.

[0008] Secondly, a system for quantitatively distributing meals to patients requiring digestive care is provided, applicable to scenarios where patients receiving nutritional support in rehabilitation facilities. The system comprises: The data acquisition and processing module is used to acquire multi-source heterogeneous time-series data related to the adjustment of meals for patients receiving digestive care, and to integrate and standardize the acquired multi-source heterogeneous time-series data. The feature mining module is used to mine individualized temporal correlation features between the meal adjustment events and subsequent physiological reactions or symptom changes of the patients receiving digestive care, based on the integrated and normalized data. The weight setting module is used to set the dynamic influence factor weights of multiple data factors that affect the generation of quantitative adjustment suggestions for meals, based on the individualized time-series correlation characteristics and the preset basic rules for meal adjustment. The suggestion generation module is used to generate individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day based on the real-time status data of the digestive care patient and in combination with the dynamic influencing factor weights. The feedback and optimization module is used to track physiological reactions and symptom changes after the implementation of the individualized dietary quantitative adjustment recommendations as feedback on the adjustment effect, and to iteratively update the weights of the dynamic influencing factors based on the adjustment effect feedback.

[0009] Compared with the prior art, the present invention has the following beneficial effects: By acquiring multi-source data, mining individualized temporal correlation features, setting dynamic influencing factor weights, generating individualized suggestions, and iteratively optimizing weights based on feedback, this approach addresses the limitations of existing technologies in matching patients' immediate needs for each meal and the instantaneous carrying capacity of their digestive system in real time. This makes it difficult to achieve truly real-time or near-real-time individualized adjustments and lacks an effective feedback optimization mechanism. It has the advantages of being able to dynamically generate individualized quantitative meal adjustment suggestions based on the patient's immediate state and continuously optimize the adjustment logic based on feedback on the adjustment effect, thereby improving the real-time performance, accuracy, and individualization level of meal adjustments. Attached Figure Description

[0010] Figure 1 Schematic diagram of the method of the present invention.

[0011] Figure 2 This is a schematic diagram of the system structure of the present invention.

[0012] In the diagram: 201, Data Acquisition and Processing Module; 202, Feature Mining Module; 203, Weight Setting Module; 204, Suggestion Generation Module; 205, Feedback and Optimization Module. Detailed Implementation

[0013] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0014] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0015] Traditional methods for adjusting meal quantification are ill-suited to the dynamic needs of patients with rapidly fluctuating physiological states, resulting in untimely and inaccurate adjustments. Furthermore, rehabilitation facilities often collect heterogeneous time-series data from multiple sources, including nutrition plans, kitchen records, nursing records, patient feedback, and physiological monitoring. Integrating and analyzing this data, and establishing a reliable link between this data and precise adjustments to meal quantification, presents a significant technical challenge for achieving individualized dynamic meal quantification.

[0016] For example, suppose in a rehabilitation facility, the meal management of a digestive patient relies on an initial plan set by a nutritionist, feeding records kept by caregivers, symptom information entered by the patient through an interactive terminal, and vital sign data collected by continuous physiological monitoring equipment. These data types are diverse and collected at varying frequencies. The patient's digestive function may change between meals due to medication or stress responses. In this scenario, if meal adjustments are based solely on a fixed plan or periodic manual assessments, it will be unable to respond promptly to changes in the patient's instantaneous condition, potentially leading to a mismatch between nutritional supply and actual needs.

[0017] If the above problems are not addressed, the lag and inaccuracy of meal quantity recommendations will directly affect the efficiency of patients' nutrient intake and their digestive system tolerance. This may lead to malnutrition or digestive discomfort, delaying the recovery process. Without a feedback optimization mechanism based on actual results, adjustment strategies cannot be continuously improved and will struggle to cope with the evolution of patients' physiological states during long-term recovery.

[0018] like Figure 1 The method for quantitatively measuring meals for patients with digestive health conditions, as shown, is applied in scenarios where these patients receive nutritional support in rehabilitation facilities. The method includes the following steps: S101. Acquire multi-source heterogeneous time-series data related to dietary adjustments for patients requiring digestive care, and integrate and standardize the acquired multi-source heterogeneous time-series data to obtain integrated and standardized data. S102. Based on the integrated and standardized data, we will mine the individualized temporal correlation features between meal adjustment events and subsequent physiological reactions or symptom changes in patients with digestive care, and obtain individualized temporal correlation features. S103. Based on individualized time-series correlation characteristics and preset basic rules for meal adjustment, dynamic influence factor weights are set for multiple data factors that affect the generation of quantitative meal adjustment suggestions, and dynamic influence factor weights are obtained. S104. Based on the real-time status data of patients receiving digestive care and combined with the weights of dynamic influencing factors, generate individualized meal quantity adjustment suggestions for the next meal or subsequent meals of the day for the patient, thus obtaining individualized meal quantity adjustment suggestions. S105. After the implementation of individualized meal quantity adjustment recommendations, the physiological reactions and symptom changes of patients receiving digestive care are used as feedback on the adjustment effect, and the dynamic influencing factor weights are iteratively updated based on the adjustment effect feedback.

[0019] Among them, multi-source heterogeneous time-series data refers to data related to dietary adjustments for patients in digestive care, originating from different sources, in different formats, and possessing time-series characteristics. This can include nutritionist planning data, kitchen preparation records, nursing meal records, patient symptom feedback, physiological monitoring data, etc., with the aim of providing a comprehensive data foundation for subsequent individualized analysis. Integration and standardization processing refers to cleaning, format conversion, and time synchronization of the acquired multi-source heterogeneous time-series data. This can be achieved using data warehouse technology, ETL tools, or specialized data processing algorithms, with the aim of eliminating data variability and ensuring data consistency and usability. Individualized time-series correlation characteristics refer to… The aim of this study is to uncover the temporal correlation between specific patient meal adjustment events and subsequent physiological responses or symptom changes over a period of time, revealing unique physiological response patterns and providing a data-driven basis for individualized adjustments. Pre-defined basic rules for meal adjustment refer to general principles or logic for meal adjustment applicable to patients with digestive health conditions, established based on medical knowledge, clinical experience, or nutritional guidelines. Their purpose is to provide fundamental and universally applicable guidance for individualized adjustments. Dynamic influencing factor weights are numerical values ​​assigned to each data factor influencing the generation of quantitative meal adjustment recommendations, reflecting their current importance or influence. These values ​​can be adjusted over time or based on feedback. Adjustments are made to the effectiveness of data, aiming to ensure that different data factors play a role in generating recommendations that aligns with their current relevance. Real-time status data refers to data collected in real-time or near real-time reflecting the patient's current physiological condition, symptoms, and subjective feelings. This data may include body temperature, heart rate, blood pressure, blood oxygen saturation, pain score, nausea level, and appetite. Its purpose is to provide the patient with immediate information as direct input for generating recommendations for adjusting the next meal. Individualized meal quantity adjustment recommendations are generated based on a specific patient's real-time status and historical data analysis, specifying details regarding the type, form, or volume of food for the patient's next meal or subsequent meals of the day. The adjustment plan aims to provide actionable guidance on dietary adjustments tailored to the patient's current needs. Adjustment effect feedback refers to the information used to evaluate the effectiveness of the adjustment after implementing individualized quantitative dietary adjustments, through monitoring and recording data such as the patient's physiological responses, symptom changes, and food intake. This information serves as a basis for subsequent strategy optimization. Iterative updates to the dynamic influencing factor weights refer to the process of periodically or trigger-based correction of these weights based on the adjustment effect feedback. This can be achieved using machine learning algorithms, reinforcement learning methods, or rule-based adaptive adjustment mechanisms. The goal is to continuously learn the patient's response patterns and optimize the accuracy of the adjustment strategy.

[0020] Specifically, the technical solution of this application first acquires and integrates multi-source heterogeneous time-series data, laying a data foundation for subsequent analysis. Next, based on the integrated data, it mines individualized time-series correlation characteristics of patients, revealing their unique physiological response patterns. Then, combining the mined individualized characteristics with preset basic rules, it sets dynamic influencing factor weights, integrating individual experience with general knowledge and assigning different weights to different factors. When suggestions need to be generated, it acquires the patient's real-time status data and uses the previously set dynamic influencing factor weights to calculate and generate individualized quantitative adjustment suggestions for the next meal or subsequent meals of the day. After the suggestions are implemented, the system tracks the patient's physiological responses and symptom changes, using this as feedback on the adjustment effect. Finally, based on the adjustment effect feedback, iteratively updates the dynamic influencing factor weights, correcting the weights through actual effects, making subsequent suggestion generation more accurate and achieving continuous strategy optimization. The entire process forms a closed loop of data collection, analysis, decision-making, implementation, feedback, and optimization.

[0021] As one embodiment of the present invention, the step of iteratively updating the weights of dynamic influence factors based on adjustment effect feedback includes: Obtain records of non-meal-related comorbidities in patients receiving digestive care during the assessment period corresponding to the adjustment effect feedback; Based on the acquired concurrent intervention event records, the potential impact of concurrent intervention events on the feedback of adjustment effects is assessed, and the potential impact assessment results are obtained. Based on the feedback on the adjustment effect and the results of the potential impact assessment, the dynamic impact factor weights are iteratively updated. The iterative update includes adjusting the update magnitude or update direction of the dynamic impact factor weights according to the results of the potential impact assessment.

[0022] Among them, records of non-meal-related concurrent intervention events refer to written or structured data records of medical or nursing interventions, physiological changes, or external events received by the patient during the period of assessing the effectiveness of meal adjustments that are not directly related to meal intake or digestion and absorption. These records can be implemented using medical orders, nursing records, examination reports, patient self-reports, etc., within an electronic medical record system. Their purpose is to identify factors that may confound or interfere with the assessment of the effectiveness of meal adjustments. Potential impact assessment results refer to the quantitative or qualitative assessment output regarding how non-meal-related concurrent intervention events may affect the patient's physiological responses or symptom changes, thereby affecting the judgment of the effectiveness of meal adjustments. This can be achieved using evaluation scores, impact levels, or correction factors based on a preset rule base, expert knowledge system, or statistical model. The purpose is to provide a basis for subsequent weight updates. Adjusting the update magnitude or direction of dynamic impact factor weights refers to amplifying or reducing the magnitude of the change based on the potential impact assessment results when calculating the weight change amount based on the feedback of the adjustment effect, or changing the increasing or decreasing trend of weight adjustment. This can be achieved by multiplying by a correction coefficient determined based on the assessment results, or by directly adjusting the sign of the weight change amount based on the assessment results. The purpose is to reduce the adverse impact of concurrent intervention events on the accuracy of weight updates.

[0023] This application's solution acquires records of non-meal-related concurrent intervention events occurring during the assessment period and assesses their potential impact on the adjustment effect feedback based on these records. This allows the system to combine the adjustment effect feedback and the potential impact assessment results when iteratively updating the weights of dynamic impact factors. Because the potential impact of concurrent intervention events is considered during the weight update process, and the update magnitude or direction is adjusted based on the assessment results, the system can distinguish between responses caused by meal adjustments and responses caused by concurrent events. This avoids incorrectly attributing physiological or symptom changes caused by concurrent events to the meal adjustment effect, effectively reducing the interference of concurrent intervention events on the weight update process. This makes the iterative update of dynamic impact factor weights more accurate, thereby improving the reliability of the subsequently generated quantitative meal adjustment recommendations.

[0024] As one embodiment of the present invention, the step of mining individualized temporal correlation features between meal adjustment events and subsequent physiological responses or symptom changes in patients receiving digestive care, and obtaining individualized temporal correlation features, includes: Based on the integrated and standardized data, the first mining of individualized temporal association features between meal adjustment events and subsequent physiological responses or symptom changes in patients with digestive care was carried out, and the first mining of individualized temporal association features was obtained. Obtain subsequent patient data to validate the individualized temporal association features mined in the initial mining. The subsequent patient data includes the patient’s subsequent dietary adjustment events and the corresponding physiological responses or symptom changes. Under the preset verification trigger conditions, based on the acquired subsequent patient data, the applicability of the initially mined individualized temporal correlation features in the current physiological state of patients receiving digestive care is examined, and the applicability test results are obtained. Based on the applicability test results, the validity representation of the first-mined individualized temporal association features is updated. The validity representation reflects the reliability of the feature under the current physiological state, and the updated validity representation is obtained. If the updated validity representation indicates that the initially mined individualized temporal association features are no longer applicable to the current physiological state, then based on the updated dataset containing subsequent patient data, the mining of individualized temporal association features is re-executed for the original associated meal adjustment events and physiological responses or symptom changes, to obtain individualized temporal association features adapted to the current physiological state, and these individualized temporal association features adapted to the current physiological state are used as individualized temporal association features. If the updated validity representation indicates that the initially mined individualized temporal association features are still applicable to the current physiological state, then the initially mined individualized temporal association features or the individualized temporal association features adjusted based on the validity representation will be used as individualized temporal association features.

[0025] The initial individualized temporal correlation features identified in this study refer to a set of patterns or rules discovered using time-series data mining methods based on historical patient data. These patterns or rules reflect statistically significant temporal correlations between specific dietary adjustments and subsequent changes in specific physiological responses or symptoms. The aim is to establish an individualized initial response model for each patient. Subsequent patient data refers to temporal data collected over time after the initial mining, including records of dietary adjustments, physiological parameters, and symptom feedback. This data aims to provide evidence reflecting the patient's current physiological state and immediate response to dietary adjustments. Pre-set validation trigger conditions are the conditions for initiating feature applicability testing. These conditions can be set based on various factors such as time intervals, accumulation of new data, and fluctuations in key physiological indicators of the patient. The purpose is to determine when the effectiveness of existing features needs to be evaluated. Applicability test results are quantitative or qualitative results evaluating the performance of the initially mined features in subsequent patient data. The purpose is to determine whether existing features are still applicable to the patient's current physiological state. Effectiveness characterization refers to attributes used to quantify or identify the reliability of individualized temporal correlation features in the current physiological state. This can be a numerical score or a status indicator. This provides a quantitative basis for subsequent feature usage or update decisions; updated validity representations no longer applicable to the current physiological state refer to the value or state of the representation reaching a preset "invalidation" standard after updating the validity representation based on the applicability test results. Its purpose is to identify features that can no longer accurately reflect the patient's current condition; updated datasets containing subsequent patient data refer to datasets formed by merging the dataset used in the initial mining with newly acquired subsequent patient data. Its purpose is to provide a more comprehensive and more reflective data foundation for re-mining; personalized temporal correlation features adapted to the current physiological state refer to new features re-mined from the updated dataset containing subsequent patient data, reflecting the temporal correlation between the patient's current dietary adjustments and physiological responses. Its purpose is to provide a basis that can accurately guide current dietary adjustments; personalized temporal correlation features adjusted based on validity representation refer to adjustments made to the feature itself or its application in subsequent processing when the feature is still valid but the validity representation has decreased. For example, adjusting its weight in generating suggestions based on the validity score aims to reduce its impact on decision-making and avoid introducing bias when the feature's validity decreases.

[0026] First, initial data mining is performed based on the patient's historical data to establish an individualized initial response model. As the patient's recovery progresses, their physiological state changes, and the original associated features may no longer be accurate. Therefore, the system continuously acquires subsequent patient data and, under preset trigger conditions, uses this subsequent data to test the applicability of the initially mined features. The test results are used to update the feature validity representation, which quantifies the reliability of the feature in the current state. It is precisely because of the introduction of the validity representation and its dynamic update mechanism that the system can perceive the "health" status of features in real time. When the validity representation indicates that a feature is no longer applicable to the current physiological state, the system will not continue to rely on that invalid feature. Instead, based on the updated dataset containing the latest data, it will re-perform the mining of individualized time-series associated features, targeting the original associated meal adjustment events and physiological responses or symptom changes, thereby obtaining new features adapted to the patient's current physiological state. If the validity representation indicates that the feature is still valid, then the feature will continue to be used, or its influence will be appropriately adjusted according to the validity representation. This closed-loop process of continuous monitoring, evaluation, updating, and re-mining ensures that the individualized temporal correlation features used to guide dietary adjustments accurately reflect the patient's current physiological state, avoiding bias caused by using outdated or ineffective features. Therefore, this approach dynamically maintains the effectiveness and timeliness of individualized temporal correlation features, adapting them to the patient's constantly changing physiological state, thereby improving the accuracy of dietary adjustment recommendations.

[0027] As one embodiment of the present invention, based on individualized time-series correlation features and preset basic rules for meal adjustment, dynamic influence factor weights are set for multiple data factors affecting the generation of quantitative meal adjustment suggestions. The steps to obtain the dynamic influence factor weights include: For each data factor that influences the generation of recommendations for adjusting meal portions: To obtain an effective characterization of the individualized time-series correlation features associated with the data factor; Obtain the applicable priority or confidence level of each preset basic rule for adjusting meals associated with this data factor; Based on the obtained validity characterization, the feature contribution of each individualized time-series correlation feature associated with the data factor to the dynamic influence factor weight of the data factor is determined. Based on the applicable priority or confidence level obtained, determine the rule guidance strength for the dynamic impact factor weight of each preset meal adjustment basic rule associated with the data factor. By integrating the characteristic contribution and rule guidance strength determined for this data factor, the dynamic influence factor weight of this data factor is calculated.

[0028] Among them, data factors refer to various patient-related information categories that affect the generation of meal quantity adjustment suggestions. These can be represented by categories such as physiological parameters, symptoms and signs, eating records, and medication. The purpose is to decompose the patient's complex state into quantifiable influence dimensions. Among them, the effectiveness representation of individualized time-series correlation features refers to the quantitative index that reflects the reliability or applicability of a specific individualized time-series correlation feature in the patient's current physiological state. It can be represented by a value between 0 and 1, a confidence interval, or an applicability level. Its purpose is to assess the current value of individualized features and avoid using outdated or inapplicable features. Among them, the applicable priority or confidence level of the pre-set basic rules for meal adjustment refers to the quantitative indicators that reflect the importance, authority or reliability of a specific pre-set basic rules for meal adjustment in general clinical practice. It can be expressed by a pre-set priority value, an expert score or a statistical confidence level based on historical data. Its purpose is to distinguish the importance of different rules and ensure that key rules are given sufficient attention. Among them, the feature contribution refers to the quantitative impact of a specific individualized time-series correlation feature on the dynamic influence factor weight calculation of a certain data factor. It can be determined by linear mapping value based on validity representation, nonlinear function output or lookup table result. Its purpose is to transform the reliability of individualized features into the impact on weight. Among them, the rule guidance strength refers to the quantitative impact of a specific preset basic rule for adjusting meals on the dynamic influence factor weight calculation of a certain data factor. It can be determined by a proportional factor, weight coefficient or guidance score based on the applicable priority or confidence level. Its purpose is to transform the importance of universal rules into the influence on weights.

[0029] This application's approach involves acquiring the validity characterization of individualized time-series correlation features associated with each data factor influencing the generation of quantitative meal adjustment recommendations, as well as the applicability priority or confidence level of pre-set meal adjustment rules. Because the validity characterization of individualized time-series correlation features reflects the reliability of those features in the patient's current state, and the applicability priority or confidence level of pre-set meal adjustment rules reflects the universal importance of those rules, it is possible to subsequently determine the feature contribution of each associated individualized time-series correlation feature to the dynamic impact factor weight of that data factor based on the acquired validity characterization. Simultaneously, based on the acquired applicability priority or confidence level, the rule guidance strength of each associated pre-set meal adjustment rule to the data factor is determined. Finally, by integrating the feature contribution and rule guidance strength determined for that data factor, the dynamic impact factor weight of that data factor is calculated. This process does not simply superimpose individualized features and universal rules, but rather evaluates their respective value in the current context (effectiveness, priority / confidence level), and transforms this value into a quantitative impact on the final weight calculation (feature contribution, rule guidance strength), thereby obtaining a weight that can dynamically balance individual differences and universal principles.

[0030] As one embodiment of the present invention, after performing the step of generating individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day based on the patient's real-time status data and combined with dynamic influencing factor weights, the method further includes: Identify at least one real-time state data item that plays a major role in the generation of the generated individualized meal quantitative adjustment suggestions, and obtain the real-time state data item that plays a major role in the generation. Obtain the dynamic impact factor weights corresponding to the real-time status data items of the main contributions, and get the corresponding dynamic impact factor weights; By combining the real-time status data items of the main contributions, the corresponding dynamic influencing factor weights, and the preset interpretation logic, an explanation of the adjustment basis for the individualized meal quantity adjustment recommendations is generated and presented to patients receiving digestive care.

[0031] The "real-time status data item" refers to a single data point or indicator reflecting the current physiological, symptom, or behavioral status of a patient receiving digestive care, such as body temperature readings, heart rate values, abdominal distension scores, appetite feedback, and activity records. This data can be obtained through sensor data collection, nursing staff recording, or patient input. The "major contribution role" refers to the data item that has a significant or decisive impact on the final adjustment result during the calculation or decision-making process for generating this individualized meal quantity adjustment recommendation. It can be determined by analyzing the weight allocation within the generation model, sensitivity analysis, or feature importance assessment. The "dynamic influence factor weight" refers to a pre-defined numerical value reflecting the importance of different data factors in generating the meal quantity adjustment recommendation. This weight is dynamically adjusted based on the patient's historical data and adjustment feedback. It can be represented as a numerical value, a coefficient, or an influence level. The "pre-defined interpretation logic" refers to a set of predefined rules, templates, or algorithms used to transform the identified major contribution data items and their influence weights into natural language or structured text that is easy for the patient to understand, explaining the reasons for the adjustment recommendation. This can include an interpretation element library, narrative structure rules, or a rule-based text generation engine. The adjustment justification refers to the text or information snippet ultimately generated and presented to patients receiving digestive care, explaining the specific real-time status data and their impact on which the recommended meal quantity adjustment was based. It can be a sentence, a paragraph, or a structured explanatory interface.

[0032] Specifically, the solution analyzes the recommendation generation process to identify the immediate state data items that have the greatest impact on the current adjustment recommendations. This is akin to tracing the source of the decision to find the most crucial evidence. Subsequently, the solution obtains the dynamic influence factor weights corresponding to these key data items. These weights reflect the magnitude of these data items' influence on the adjustment recommendations in the current context. Because these data items and their weights collectively determine the direction and scope of the adjustment recommendations, the solution can combine this information with a pre-defined explanatory logic. This pre-defined explanatory logic translates these technical data and weights into language that patients can understand, forming an explanation of the basis for the adjustment. In this way, patients not only know "this adjustment is needed," but also "why this adjustment is needed," thus greatly improving the transparency of the recommendations. This transparency allows patients to understand the rationale behind the adjustment recommendations, enhancing their trust in the recommendations and thereby increasing their adherence to the recommendations.

[0033] As one embodiment of the present invention, the steps for generating an explanation of the basis for the individualized meal quantity adjustment recommendations for patients receiving digestive care, by combining the main contributing real-time status data items, the corresponding dynamic influencing factor weights, and the preset interpretation logic, include: Based on the category and current value of the main contributing real-time status data item, and the magnitude of the corresponding dynamic influence factor weight, the selected explanatory elements are selected from the preset explanatory element library for the digestive care scenario, which match the category, current value, and magnitude of the corresponding dynamic influence factor weight of the main contributing real-time status data item. The actual values ​​of the main contributing real-time state data items and the interpretation information of the dynamic influence factor weights are embedded into the selected explanatory elements to form standardized explanatory fragments. Based on the narrative structure rules in the preset interpretation logic, one or more standardized interpretation fragments are combined to generate an adjustment basis description for patients with digestive diseases.

[0034] The pre-built explanatory element library refers to a pre-constructed collection of various explanatory templates or text fragments. These templates or fragments are used to explain the meaning of different types of data items, different ranges of data values, and different influence weights. They can be implemented using structured text templates, parameterized statement frames, or predefined explanatory phrases, and their purpose is to provide basic explanatory units for generating adjustment basis explanations. An explanatory element refers to one or more specific explanatory templates or text fragments selected from the pre-built explanatory element library, used to explain specific major contribution data items and their weights. The magnitude of the dynamic influence factor weight refers to the range or level of the dynamic influence factor weight value, and its purpose is to transform continuous weight values ​​into discrete, easily understood level information. A standardized explanatory fragment refers to a text unit with a unified structure and clear expression formed by filling or embedding the actual values ​​of the real-time status data items of major contributions and the interpretation information of the dynamic influence factor weights into the selected explanatory elements. It can be implemented using string formatting, template filling, or parameter replacement, and its purpose is to transform abstract data information into specific, personalized explanatory statements. The pre-defined explanatory logic refers to a set of rules that guide the organization and presentation of adjustment justification explanations. These rules can be implemented using logical judgment rules, priority ordering rules, or narrative structure rules, aiming to ensure that the generated adjustment justification explanations are logically clear and easy to understand. Narrative structure rules, defined within the pre-defined explanatory logic, define how to combine one or more standardized explanatory segments. For example, they specify the presentation order of explanatory segments, the connection methods between different segments, or the overall framework of the explanation. These rules can be implemented using sequence rules, hierarchical rules, or template rules, aiming to construct highly readable adjustment justification explanations. The adjustment justification explanation refers to the final textual information generated and presented to patients receiving digestive care, explaining the reasons for the proposed dietary quantity adjustments. It can be presented in structured text paragraphs, lists, or a combination of text and graphics, aiming to help patients understand the rationale behind the adjustments.

[0035] This application's solution selects matching explanatory elements from a pre-defined explanatory element library based on the category, current value, and corresponding dynamic influence factor weights of the main contributing real-time status data items, ensuring the relevance of subsequent explanations. Next, the actual values ​​of the main contributing real-time status data items and the interpretation information of the dynamic influence factor weights are embedded into the selected explanatory elements to form standardized explanatory fragments. It is precisely because specific patient data and weight information are integrated into the pre-defined explanatory framework that the explanations are individualized and specific. Finally, according to the narrative structure rules in the pre-defined explanatory logic, one or more standardized explanatory fragments are combined to generate adjustment guidelines for digestive health care patients. It is this structured combination method that allows complex adjustment guidelines to be presented in a logically clear and easily understandable way. This application's solution, combined with the steps of identifying the main contributing real-time status data items and obtaining the corresponding dynamic influence factor weights, enables the system not only to determine the key factors influencing adjustment recommendations and their importance but also to translate this key information into language that patients can understand. This effectively solves the problem of difficulty in understanding simply listing data and weights, improving patients' understanding and acceptance of adjustment recommendations.

[0036] As one embodiment of the present invention, the step of combining one or more standardized interpretation fragments according to the narrative structure rules in the preset interpretation logic includes: Based on the category of the main contributing real-time status data items and the adjustment direction of the individualized meal quantitative adjustment suggestions, a narrative structure template matching the category and adjustment direction is selected from the preset narrative structure template library to obtain the selected narrative structure template. Fill one or more standardized explanatory fragments into the preset positions of the selected narrative structure template to generate an adjustment basis description.

[0037] The pre-defined explanatory logic includes several key elements: Narrative structure rules, which guide the organization of explanatory content and can be implemented using grammatical rules, logical order rules, and emphasis rules. The goal is to ensure the generated explanatory text is readable and logically sound. The categories of the main contributing real-time status data items refer to the data types of the primary reasons for the meal adjustment recommendations. These can be categorized using data such as blood sugar, bloating, appetite, and weight, aiming to identify key factors influencing the recommendations. The adjustment direction of the individualized meal quantity adjustment recommendations specifies whether to increase, decrease, or maintain the current meal quantity, clarifying the specific operational steps. The pre-defined narrative structure template library refers to pre-designed templates with specific narrative frameworks and... The placeholder text template set can be implemented using a template set containing different placeholders and text structures. Each template corresponds to a combination of one or more data item categories and adjustment directions, with the aim of providing a structured explanatory framework. Selecting a narrative structure template that matches the category and adjustment direction means searching and determining the most suitable template for constructing the explanatory text from the template library based on the category of the current main contributing data item and the direction of the adjustment suggestion, with the aim of ensuring the relevance of the explanatory text. Filling one or more standardized explanatory fragments into the preset positions of the selected narrative structure template means embedding explanatory fragments containing specific values ​​and interpretation information into the selected narrative structure template according to the reserved placeholders or specified positions in the template, with the aim of integrating specific information into the structured framework.

[0038] As one embodiment of the present invention, the step of generating individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day includes: A comprehensive adjustment index is obtained by weighting the current value of the real-time status data item with the weight of the dynamic influence factor. Based on the comprehensive adjustment index, individualized meal quantity adjustment recommendations are determined as individualized meal quantity adjustment recommendations for the patient's next meal or subsequent meals of the day.

[0039] The "Real-Time Status Data Item" refers to data reflecting the patient's current physiological status, symptoms, or feelings, and can include objective physiological parameters and subjective feedback. The "Dynamic Influence Factor Weight" refers to the importance or influence coefficient assigned to different data factors affecting the meal quantity adjustment recommendations. Its value can be dynamically adjusted based on the patient's historical data and feedback on the adjustment effect. "Weighted Calculation" is a mathematical operation that multiplies multiple data items by their corresponding weights and then sums or combines them in other forms to obtain a comprehensive value. The "Comprehensive Adjustment Index" is a quantitative indicator obtained by weighting the current value of the real-time status data item with the dynamic influence factor weights. This indicator reflects the direction and magnitude of the meal adjustment needed at the current moment. "Individualized Meal Quantity Adjustment Recommendation" refers to a specific adjustment plan proposed for a particular patient's next meal or subsequent meals of the day, based on the original meal plan. This can involve increases, decreases, or changes in total energy, specific nutrients, food form, or supply volume.

[0040] This application's solution calculates a comprehensive adjustment index by weighting the current value of real-time status data items with the weights of dynamic influencing factors. This weighted calculation method combines the patient's real-time physiological status with the importance of various influencing factors, thus more comprehensively reflecting the patient's actual needs. Real-time status data items reflect the patient's current physiological status, while dynamic influencing factor weights reflect the degree of influence of different factors on the dietary adjustment recommendations. Through weighted calculation, this information can be integrated into a comprehensive index, providing a basis for the subsequent generation of adjustment recommendations. Because of this weighted calculation method, the influence of different real-time status data items on the final adjustment recommendations can be distinguished and quantified according to their corresponding dynamic influencing factor weights, thus avoiding errors that may arise from simple combinations. Based on this, individualized quantitative dietary adjustment recommendations are determined according to the comprehensive adjustment index. The comprehensive adjustment index can be regarded as a quantitative adjustment signal; its magnitude and direction directly reflect the need for dietary adjustments. By establishing a mapping relationship between the comprehensive adjustment index and the quantitative dietary adjustment recommendations, the automatic generation of dietary adjustment recommendations can be achieved. This method not only improves adjustment efficiency but also ensures the rationality and consistency of adjustment recommendations. Within the overall framework of data acquisition, feature mining, dynamic weight setting, suggestion generation, and feedback optimization, this solution effectively integrates real-time status data and dynamic influencing factor weights by introducing weighted calculation and comprehensive adjustment index. This enables the generation of more accurate individualized meal quantitative adjustment suggestions, better meeting the actual needs of patients.

[0041] As one embodiment of the present invention, the step of mining individualized temporal correlation features between meal adjustment events and subsequent physiological responses or symptom changes in patients receiving digestive care, and obtaining individualized temporal correlation features, includes: Based on the integrated and standardized data, the first mining of individualized temporal association features between meal adjustment events and subsequent physiological responses or symptom changes in patients with digestive care was carried out. The first mining included temporal association rule mining under the preset maximum time delay window ΔT_max, minimum support threshold min_Sup, and minimum confidence threshold min_Conf to obtain the individualized temporal association features mined in the first mining. The validity representation of each feature in the individualized temporal association features mined in the first mining was initialized with an initial value of 1.0. Under the preset verification trigger conditions, based on the acquired subsequent patient data, the applicability of the first-mined individualized temporal association features in the current physiological state of digestive care patients is tested. The test includes calculating the actual confidence level Conf_new_k of the first-mined individualized temporal association features in subsequent patient data, and obtaining the applicability test results. Based on the applicability test results, the validity representation of the initially mined individualized temporal association features is updated. The validity representation reflects the reliability of the feature under the current physiological state, and the update is calculated using the following formula: `Validity_Score_k_new = (1 - α) * Validity_Score_k_old + α * (Conf_new_k / Association_Strength_k)` Where `Validity_Score_k_new` is the updated validity representation, `Validity_Score_k_old` is the original validity representation, `α` is the learning rate or decay factor, `Conf_new_k` is the actual confidence level, and `Association_Strength_k` is the association strength of the first mined individualized temporal association features, thus obtaining the updated validity representation. If the updated validity representation indicates that the initially mined individualized temporal association features are no longer applicable to the current physiological state (i.e., Validity_Score_k_new is less than the preset validity score threshold Validity_Threshold), then based on the updated dataset containing subsequent patient data, the mining of individualized temporal association features is re-executed for the meal adjustment events and physiological responses or symptom changes originally associated with these features. The re-mining includes temporal association rule mining under the preset maximum time delay window ΔT_max, the preset minimum support threshold min_Sup_re, and the preset minimum confidence threshold min_Conf_re to obtain individualized temporal association features adapted to the current physiological state, and these individualized temporal association features adapted to the current physiological state are used as individualized temporal association features. If the updated validity representation indicates that the first mined individualized temporal association feature is still applicable to the current physiological state, i.e., Validity_Score_k_new is not less than the preset validity score threshold Validity_Threshold, then the first mined individualized temporal association feature or the individualized temporal association feature adjusted based on the validity representation will be used as the individualized temporal association feature.

[0042] Among them, validity characterization refers to a numerical value reflecting the reliability of individualized temporal correlation features under the current physiological state, which can be implemented using floating-point numbers between 0 and 1; the preset verification trigger condition refers to the condition for initiating the applicability test of the first mined individualized temporal correlation features, which can be implemented by accumulating a certain amount of subsequent patient data, waiting for a preset time interval, or detecting a significant change in the patient's physiological state; the actual confidence score Conf_new_k is the ratio of the actual frequency of the first mined individualized temporal correlation feature in the acquired subsequent patient data to the frequency of the antecedent event of the feature in the subsequent patient data; validity score The Validity_Threshold is a critical value used to determine whether the initially mined individualized temporal association features are still applicable to the current physiological state. It can be implemented using a preset fixed value. The learning rate or decay factor α is a parameter used to control the weight of the influence of subsequent patient data on the updated validity representation. It can be implemented using a floating-point number between 0 and 1. The Association_Strength_k is the association strength of the individualized temporal association feature calculated during the initial mining process. It can be implemented using the confidence, lift, or other association metrics at the time of the initial mining.

[0043] like Figure 2 The system shown is a quantitative meal distribution system for patients receiving digestive care, applied in scenarios where these patients receive nutritional support in rehabilitation facilities. The system includes: The data acquisition and processing module 201 is used to acquire multi-source heterogeneous time-series data related to the adjustment of meals for patients in digestive care, and to integrate and standardize the acquired multi-source heterogeneous time-series data. The feature mining module 202 is used to mine individualized temporal association features between meal adjustment events and subsequent physiological reactions or symptom changes in patients with digestive care based on integrated and normalized data. The weight setting module 203 is used to set the dynamic influence factor weights of multiple data factors that affect the generation of quantitative adjustment suggestions for meals, based on individualized time-series correlation characteristics and preset basic rules for meal adjustment. It is suggested that module 204 be used to generate individualized meal quantity adjustment suggestions for the next meal or subsequent meals of the day for patients with digestive care based on real-time status data and dynamic influencing factor weights. The feedback and optimization module 205 is used to track physiological reactions and symptom changes after the implementation of individualized meal quantitative adjustment recommendations as feedback on the adjustment effect, and to iteratively update the weights of dynamic influencing factors based on the adjustment effect feedback.

[0044] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for quantitatively measuring meals for patients requiring digestive care, applied to scenarios where patients requiring digestive care receive nutritional support in rehabilitation institutions, characterized in that... The method includes the following steps: Acquire multi-source heterogeneous time-series data related to the dietary adjustments of the patients receiving digestive care, and integrate and standardize the acquired multi-source heterogeneous time-series data to obtain integrated and standardized data; Based on the integrated and standardized data, individualized temporal correlation features between meal adjustment events and subsequent physiological responses or symptom changes in patients with digestive care are mined to obtain the individualized temporal correlation features. Based on the individualized time-series correlation features and the preset basic rules for meal adjustment, dynamic influence factor weights are set for multiple data factors that affect the generation of quantitative meal adjustment suggestions, and the dynamic influence factor weights are obtained. Based on the real-time status data of patients receiving digestive care, and combined with the weights of the dynamic influencing factors, an individualized meal quantity adjustment suggestion is generated for the patient's next meal or subsequent meals of the day, thus obtaining the individualized meal quantity adjustment suggestion; After the implementation of the individualized meal quantity adjustment recommendations, the physiological responses and symptom changes of patients receiving digestive care are tracked as feedback on the adjustment effect, and the weights of the dynamic influencing factors are iteratively updated based on the adjustment effect feedback.

2. The method for quantitatively measuring meals for patients with digestive health issues according to claim 1, characterized in that, Based on the feedback of the adjustment effect, the steps for iteratively updating the weights of the dynamic influence factors include: Obtain records of non-meal-related comorbidities that occurred in the patients receiving digestive care during the evaluation period corresponding to the feedback on the adjustment effect; Based on the acquired concurrent intervention event records, the potential impact of the concurrent intervention events on the adjustment effect feedback is evaluated, and the potential impact evaluation results are obtained. Based on the feedback on the adjustment effect and the assessment results of the potential impact, an iterative update of the weights of the dynamic impact factors is performed.

3. The method for quantitatively measuring meals for patients with digestive issues according to claim 2, characterized in that, The iterative update includes adjusting the update magnitude or direction of the dynamic impact factor weights based on the potential impact assessment results.

4. The method for quantitatively measuring meals for patients with digestive issues according to claim 1, characterized in that, The steps for obtaining the individualized temporal association features between meal adjustment events and subsequent physiological responses or symptom changes in patients receiving digestive care include: Based on the integrated and standardized data, the first mining of individualized temporal correlation features between meal adjustment events and subsequent physiological reactions or symptom changes in patients with digestive care was carried out, and the first mining of individualized temporal correlation features was obtained. Obtain subsequent patient data to verify the individualized temporal correlation features mined in the first mining, the subsequent patient data including the patient's subsequent dietary adjustment events and corresponding physiological reactions or symptom changes; Under preset verification trigger conditions, based on the acquired subsequent patient data, the applicability of the initially mined individualized temporal correlation features to the current physiological state of the digestive care patient is tested, and the applicability test results are obtained. Based on the applicability test results, the validity representation of the first mined individualized temporal correlation feature is updated. The validity representation reflects the reliability of the feature under the current physiological state, resulting in the updated validity representation. If the updated validity representation indicates that the initially mined individualized temporal association feature is no longer applicable to the current physiological state, then based on the updated dataset containing the subsequent patient data, the mining of the individualized temporal association feature is re-executed for the original associated meal adjustment events and physiological reactions or symptom changes, to obtain an individualized temporal association feature adapted to the current physiological state, and this individualized temporal association feature adapted to the current physiological state is used as the individualized temporal association feature. If applicable to the current physiological state, the first mined individualized temporal correlation feature or the individualized temporal correlation feature adjusted based on the validity representation shall be used as the individualized temporal correlation feature.

5. A method for quantitatively measuring meals for patients requiring digestive care according to claim 1, characterized in that, The step of setting dynamic influence factor weights for multiple data factors affecting the generation of quantitative meal adjustment suggestions based on the individualized time-series correlation features and preset basic rules for meal adjustment includes: For each data factor that influences the generation of recommendations for adjusting meal portions: To obtain an effective characterization of each of the individualized time-series correlation features associated with the data factor; Obtain the applicable priority or confidence level of each of the preset basic rules for adjusting meals associated with the data factor; Based on the acquired validity characterization, the feature contribution of each of the individualized time-series correlation features associated with the data factor to the dynamic influence factor weight of the data factor is determined. Based on the applicable priority or confidence level obtained, determine the rule guidance strength for the dynamic influence factor weight of each of the preset meal adjustment basic rules associated with the data factor. By integrating the feature contribution degree determined for the data factor with the rule guidance strength, the dynamic influence factor weight of the data factor is calculated.

6. The method for quantitatively measuring meals for patients with digestive health care according to claim 1, characterized in that, After performing the step of generating individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day based on the real-time status data of the digestive care patient and in combination with the dynamic influencing factor weights, the method further includes: Identify at least one of the instantaneous state data items that plays a major role in the generation of the generated individualized meal quantity adjustment suggestions, and obtain the instantaneous state data item that plays a major role in the generation. Obtain the dynamic impact factor weights corresponding to the real-time status data items of the main contributions, and obtain the corresponding dynamic impact factor weights; By combining the real-time status data items of the main contributions, the corresponding dynamic influence factor weights, and the preset interpretation logic, an explanation of the adjustment basis for the individualized meal quantity adjustment suggestions is generated and presented to the digestive care patient.

7. A method for quantitatively measuring meals for patients requiring digestive care according to claim 6, characterized in that, The steps for generating an explanation of the basis for the individualized meal quantity adjustment recommendations, which is presented to the digestive care patient, by combining the real-time status data items of the main contributions, the corresponding dynamic influence factor weights, and the preset interpretation logic, include: Based on the category and current value of the real-time status data item of the main contribution, and the magnitude of the corresponding dynamic influence factor weight, an explanatory element that matches the category, current value and magnitude of the corresponding dynamic influence factor weight of the real-time status data item of the main contribution is selected from the preset explanatory element library for the digestive care scenario, and the selected explanatory element is obtained. The actual values ​​of the real-time state data items of the main contributions and the interpretation information of the dynamic influence factor weights are embedded into the selected explanatory elements to form a standardized explanatory fragment, thus obtaining a standardized explanatory fragment. Based on the narrative structure rules in the preset explanation logic, one or more of the standardized explanation fragments are combined to generate an adjustment basis description for the digestive care patient, thus obtaining the adjustment basis description.

8. A method for quantitatively measuring meals for patients with digestive issues according to claim 7, characterized in that, The step of combining one or more of the standardized explanatory fragments according to the narrative structure rules in the preset explanatory logic includes: Based on the category of the real-time status data item of the main contribution and the adjustment direction of the individualized meal quantitative adjustment suggestion, a narrative structure template matching the category and adjustment direction is selected from the preset narrative structure template library to obtain the selected narrative structure template. The one or more standardized explanatory fragments are filled into the preset positions of the selected narrative structure template to generate the adjustment basis description, thus obtaining the adjustment basis description.

9. A method for quantitatively measuring meals for patients requiring digestive care according to claim 1, characterized in that, The steps for generating individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day include: A comprehensive adjustment index is obtained by weighting the current value of the real-time status data item with the weight of the dynamic influence factor. Based on the comprehensive adjustment index, the individualized meal quantity adjustment recommendations are determined as individualized meal quantity adjustment recommendations for the patient's next meal or subsequent meals of the day.

10. A system for quantitatively distributing meals to patients requiring digestive care, applied in scenarios where patients requiring digestive care receive nutritional support in rehabilitation facilities, characterized in that... The system includes: The data acquisition and processing module is used to acquire multi-source heterogeneous time-series data related to the adjustment of meals for patients receiving digestive care, and to integrate and standardize the acquired multi-source heterogeneous time-series data. The feature mining module is used to mine individualized temporal correlation features between the meal adjustment events and subsequent physiological reactions or symptom changes of the patients receiving digestive care, based on the integrated and normalized data. The weight setting module is used to set the dynamic influence factor weights of multiple data factors that affect the generation of quantitative adjustment suggestions for meals, based on the individualized time-series correlation characteristics and the preset basic rules for meal adjustment. The suggestion generation module is used to generate individualized meal quantity adjustment suggestions for the patient's next meal or subsequent meals of the day based on the real-time status data of the digestive care patient and in combination with the dynamic influencing factor weights. The feedback and optimization module is used to track physiological reactions and symptom changes after the implementation of the individualized dietary quantitative adjustment recommendations as feedback on the adjustment effect, and to iteratively update the weights of the dynamic influencing factors based on the adjustment effect feedback.

Images