Chronic disease management method, system and equipment and storage medium

Abstract

The invention relates to the technical field of large models, in particular to a chronic disease management method, system and device and a storage medium. The method comprises the steps of firstly analyzing user query content, extracting query intention features and generating medical suggestions in combination with personalized parameters; collecting medical equipment data in real time, analyzing health state characteristics in combination with medical suggestion information, and outputting a health monitoring result; according to the health monitoring result and a preset medical scheme, task type features and patient behavior features are fully considered, intervention execution features are determined in a progressive task decomposition mode, and then a standardized intervention instruction is generated; acquiring related historical data from the memory system based on the intervention instruction for matching, and finally integrating a matching result and real-time data to update the health record of the patient; and an execution plan better conforming to the actual condition of the patient is made through a task decomposition mode, so that the pertinence of the medical scheme is improved.

Description

Technical Field

[0001] This application relates to the technical field of large models, and in particular to a method, system, device and storage medium for chronic disease management. Background Technology

[0002] As the situation regarding chronic disease prevention and control becomes increasingly severe, intelligent medical management systems play a crucial role in patients' daily health monitoring and intervention guidance. Effective chronic disease management requires tailoring intervention plans to individual patient characteristics and ensuring that patients consistently adhere to medical recommendations.

[0003] Existing chronic disease management systems typically employ pre-set standardized medical protocols. The system sends task reminders to patients at fixed intervals and monitors their adherence through device data. This approach relies solely on preset rules and thresholds to determine patient performance and generate uniform intervention instructions. This simplistic and mechanical approach leads to ineffective implementation of the medical protocols and warrants further improvement. Summary of the Invention

[0004] To address the issue of ineffective implementation of existing medical protocols, this application provides a method, system, device, and storage medium for chronic disease management, employing the following technical solution: In a first aspect, this application provides a method for managing chronic diseases, comprising the following steps: The query content is parsed to obtain query intent features, and medical suggestion information is generated based on the query intent features and personalized parameters. Based on real-time data from medical devices and the medical advice information, health status characteristics are obtained, and health monitoring results are generated based on these health status characteristics. Based on the health monitoring results and the preset medical plan, intervention execution characteristics are determined through progressive task decomposition based on task type characteristics and patient behavior characteristics. Based on the intervention execution characteristics, standardized intervention instructions are generated. According to the standardized intervention instructions, retrieval features are obtained from the memory system, and historical data matching results are obtained based on the retrieval features; Based on the historical data matching results and real-time data, storage features are calculated, and the patient's health record is updated based on the storage features.

[0005] By adopting the above-mentioned technical solutions, in the field of chronic disease management, the formulation and implementation of medical plans directly affect the patient's recovery process. Due to significant individual differences among patients, their execution abilities and behavioral habits vary. For example, some patients may be unable to take medication on time due to busy work schedules, while others may have difficulty understanding complex dietary requirements due to cognitive impairments. This makes it difficult for uniform medical plans to achieve ideal results. This application first analyzes the user's query content, extracts query intent features, and generates medical suggestions by combining them with personalized parameters. It collects medical device data in real time, analyzes health status characteristics by combining medical suggestion information, and outputs health monitoring results. Based on the health monitoring results and the preset medical plan, it fully considers the characteristics of task types and patient behavior, adopts a progressive task decomposition method to determine the intervention execution characteristics, and then generates standardized intervention instructions. Based on the intervention instructions, it retrieves relevant historical data from the memory system for matching, and finally integrates the matching results and real-time data to update the patient's health record. By developing an execution plan that is more in line with the patient's actual situation through task decomposition, the targeting of the medical plan is improved.

[0006] Optionally, the query content is parsed to obtain query intent features, and medical suggestion information is generated based on the query intent features and personalized parameters, specifically including the following steps: The system parses the user's query to obtain text, voice, and image input. A unified feature representation is obtained based on text, voice, and image input; Based on the unified feature representation, the query intent features are determined; Personalized parameters are obtained based on the user's historical data; Based on the query intent features and the personalized parameters, medical advice information is generated.

[0007] By adopting the above technical solution, this application can simultaneously process user input in multiple forms such as text, voice, and images to obtain a unified feature representation, thereby improving the system's interactive convenience. By analyzing the unified feature representation, the user's query intent can be identified. Combined with personalized parameters extracted from historical data, medical advice information that better meets the user's actual needs can be generated, improving the accuracy and applicability of medical advice.

[0008] Optionally, the real-time data from the medical devices includes glucose concentration data from a continuous glucose monitor, blood pressure data from an ambulatory blood pressure monitor, heart rate data from an electrocardiogram monitor, and weight data from a smart scale. Based on the real-time data from the medical devices and the medical advice information, health status characteristics are obtained, and based on these health status characteristics, health monitoring results are generated, specifically including the following steps: A multi-scale standardized data stream is obtained based on continuous sampling of glucose concentration data, intermittent sampling of blood pressure data, real-time sampling of heart rate data, and periodic sampling of weight data. Based on the glucose concentration fluctuation rate, blood pressure diurnal rhythm, heart rate variability index, and weight change trend in the multi-scale standardized data stream, and combined with the medical advice information, multidimensional physiological indicator characteristics are obtained. Based on the multidimensional physiological indicators and health risk models for diabetes, hypertension, and cardiovascular disease, health monitoring results are generated.

[0009] By adopting the above technical solution, this application can process data from various sampling frequencies from different medical devices, transforming continuous sampling of glucose concentration, intermittent sampling of blood pressure, real-time sampling of heart rate, and periodic sampling of weight into a standardized data stream, which is conducive to unified data management. By analyzing multidimensional physiological indicators such as glucose concentration fluctuation rate, blood pressure diurnal rhythm, heart rate variability index, and weight change trend, combined with medical advice information, a more comprehensive reflection of the patient's health status can be achieved. Based on the characteristics of multidimensional physiological indicators, combined with health risk models of diabetes, hypertension, and cardiovascular diseases, health monitoring results are generated, which helps to identify potential health risks in a timely manner.

[0010] Optionally, the preset medical plan includes a medication reminder plan, a dietary intervention plan, an exercise guidance plan, and a follow-up visit plan. Based on the health monitoring results and the preset medical plan, intervention execution characteristics are determined through progressive task decomposition based on task type characteristics and patient behavior characteristics. Based on the intervention execution characteristics, standardized intervention instructions are generated, specifically including the following steps: Based on the health monitoring results and the medication reminder plan, dietary intervention plan, exercise guidance plan, and follow-up visit plan, task type characteristics are obtained, wherein the task type characteristics include feedback frequency, task difficulty, and cognitive load. Based on the task type characteristics and the patient's previous execution records, the intervention execution characteristics are obtained through progressive task decomposition. Based on the intervention execution characteristics and medical standard requirements, standardized intervention instructions are generated.

[0011] By adopting the above technical solutions, this application analyzes medical plans in multiple aspects, such as medication reminders, dietary interventions, exercise guidance, and follow-up visits. It extracts task type characteristics from dimensions such as feedback frequency, task difficulty, and cognitive load, which helps to comprehensively grasp the execution requirements of medical plans. Combining the patient's previous execution records, a progressive task decomposition approach is adopted to make the intervention execution characteristics more aligned with the patient's actual situation. Based on the intervention execution characteristics, standardized intervention instructions are generated in accordance with medical guidelines, which helps to improve the execution effect of medical plans.

[0012] Optionally, based on the task type characteristics and the patient's previous performance records, intervention performance characteristics are obtained through progressive task decomposition, specifically including the following steps: Based on the temporal regularity, environmental relevance, and emotional state in the patient's past performance records, the patient's behavioral characteristics are obtained. The temporal regularity includes the diurnal fluctuations and periodic changes in task performance, and the environmental relevance includes the differences in task completion rates between indoor and outdoor, work and home scenarios. Based on the patient's behavioral characteristics, and considering task difficulty, execution time, and environmental conditions, the medical task is decomposed into multiple micro-task sequences. Based on the micro-task sequence and task completion probability, intervention execution characteristics are obtained.

[0013] By adopting the above technical solutions, this application can gain a deeper understanding of patients' behavioral patterns by analyzing the temporal regularity and environmental correlation in patients' past execution records, as well as factors such as emotional state; based on task difficulty, execution time and environmental conditions, medical tasks are decomposed into micro-task sequences, making tasks easier to execute; by considering micro-task sequences and task completion probabilities, intervention execution characteristics are generated, which is conducive to improving the execution effect of medical plans.

[0014] Optionally, the method further includes the following steps: An initial interaction analysis is performed based on the patient's health record to obtain interaction state characteristics; Based on the interaction status characteristics and monthly health indicators, trend analysis is performed to obtain health trend characteristics; Based on the health trend characteristics and the medical knowledge base, rule-based reasoning is performed to obtain knowledge application characteristics; Based on the characteristics of knowledge application, an interactive interface is generated, and multimodal input data is obtained based on the interactive interface.

[0015] By adopting the above technical solutions, this application can grasp the patient's interaction status characteristics in real time through initial interactive analysis of the patient's health records; trend analysis based on interaction status characteristics and monthly health indicators is beneficial for understanding the dynamic changes in the patient's health status; rule reasoning combined with health trend characteristics and medical knowledge base yields knowledge application characteristics, realizing hierarchical storage and efficient retrieval of information, which helps the system continuously optimize intervention strategies; in addition, by generating an interactive interface based on knowledge application characteristics, multimodal input interaction methods are provided for users, enabling patients to interact with the system through various forms such as text, voice, and images, improving the system's ease of use and the completeness of information collection.

[0016] Secondly, this application provides a chronic disease management system, comprising: The suggestion information generation module is used to parse the user's query content to obtain query intent features, and generate medical suggestion information based on the query intent features and personalized parameters; The monitoring result generation module is used to obtain health status characteristics based on real-time data from medical devices and the medical advice information, and generate health monitoring results based on the health status characteristics; The intervention instruction generation module is used to determine the intervention execution characteristics based on the health monitoring results and the preset medical plan, through progressive task decomposition based on task type characteristics and patient behavior characteristics, and to generate standardized intervention instructions based on the intervention execution characteristics. The historical data matching module is used to obtain retrieval features from the memory system according to the standardized intervention instructions, and obtain historical data matching results based on the retrieval features; The health record update module is used to calculate storage features based on the historical data matching results and real-time data, and update the patient's health record based on the storage features.

[0017] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described chronic disease management method.

[0018] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described chronic disease management method.

[0019] In summary, this application includes at least one of the following beneficial technical effects: This application first analyzes user queries, extracts query intent features, and generates medical suggestions by combining them with personalized parameters; it then collects medical device data in real time, analyzes health status characteristics based on medical suggestion information, and outputs health monitoring results; based on the health monitoring results and preset medical plans, it fully considers task type characteristics and patient behavior characteristics, adopts a progressive task decomposition approach to determine intervention execution characteristics, and then generates standardized intervention instructions; based on the intervention instructions, it retrieves relevant historical data from the memory system for matching, and finally integrates the matching results and real-time data to update the patient's health record; by using the task decomposition approach to formulate an execution plan that is more in line with the patient's actual situation, the targeting of the medical plan is improved. This application can process data from various sampling frequencies from different medical devices, transforming continuous glucose concentration sampling, intermittent blood pressure sampling, real-time heart rate sampling, and periodic weight sampling into a standardized data stream, which is beneficial for unified data management. By analyzing multidimensional physiological indicators such as glucose concentration variability, blood pressure diurnal rhythm, heart rate variability, and weight change trends, combined with medical advice, it can more comprehensively reflect the patient's health status. Based on the characteristics of multidimensional physiological indicators, combined with health risk models for diabetes, hypertension, and cardiovascular diseases, health monitoring results are generated, which helps to identify potential health risks in a timely manner. This application analyzes the temporal regularity and environmental correlation in patients' past execution records, as well as factors such as emotional state, to gain a deeper understanding of patients' behavioral patterns. Based on task difficulty, execution duration, and environmental conditions, medical tasks are decomposed into a sequence of micro-tasks, making the tasks easier to execute. By considering the micro-task sequence and the probability of task completion, intervention execution characteristics are generated, which helps to improve the execution effect of medical plans. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a chronic disease management method according to an embodiment of this application; Figure 2 This is a flowchart illustrating step S100 in a chronic disease management method according to an embodiment of this application. Figure 3 This is a flowchart illustrating step S200 in a chronic disease management method according to an embodiment of this application; Figure 4 This is a flowchart illustrating step S300 in a chronic disease management method according to an embodiment of this application; Figure 5 This is a flowchart illustrating step S320 in a chronic disease management method according to an embodiment of this application; Figure 6 This is a schematic diagram of the data interaction process in a chronic disease management method according to an embodiment of this application; Figure 7 This is a schematic diagram of a module of a chronic disease management system according to an embodiment of this application; Figure 8 This is an internal structural diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0021] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0022] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0023] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0024] Firstly, this application provides a method for chronic disease management, referring to... Figure 1 It includes the following steps: S100. Analyze the user's query content to obtain query intent features, and generate medical suggestion information based on the query intent features and personalized parameters.

[0025] In this embodiment, the query intent feature refers to the medical consultation purpose implied in the user's input content, including four categories: symptom description, medication consultation, lifestyle advice, and follow-up appointment; personalized parameters include the user's basic disease information, treatment plan, past medical records, and lifestyle records.

[0026] S200: Based on real-time data from medical devices and medical advice, obtain health status characteristics and generate health monitoring results based on these characteristics.

[0027] In this embodiment, the real-time data from the medical devices includes glucose concentration data from a continuous glucose monitor, blood pressure data from an ambulatory blood pressure monitor, heart rate data from an electrocardiogram monitor, and weight data from a smart scale.

[0028] Specifically, the system establishes a health indicator reference interval mapping table, compares real-time data with normal reference values, and determines the degree of abnormality. A sliding window method is used to analyze data fluctuation trends, with the window size being the previous period. Frequency statistics methods are used to identify patterns in lifestyle behaviors and generate a health monitoring and assessment report.

[0029] S300: Based on health monitoring results and pre-set medical plans, determine intervention execution characteristics through progressive task decomposition based on task type characteristics and patient behavior characteristics, and generate standardized intervention instructions based on intervention execution characteristics.

[0030] In this embodiment, task type characteristics include feedback frequency, task difficulty, and cognitive load; patient behavior characteristics include compliance scores, activity patterns, and scenario preferences. Progressive task decomposition refers to breaking down complex medical tasks into multiple simple and executable sub-tasks.

[0031] S400. Based on the standardized intervention instructions, retrieve the retrieval features from the memory system and obtain the historical data matching results based on the retrieval features.

[0032] In this embodiment, the search features include time range, data type, and degree of correlation; the historical data matching results include similar cases, treatment records, and follow-up information.

[0033] S500 calculates storage features based on historical data matching results and real-time data, and updates patient health records based on storage features.

[0034] In this embodiment, the storage features include data timeliness, association attributes, and storage priority; the patient health record includes medical information, monitoring data, and execution feedback.

[0035] In one embodiment, refer to Figure 2 In step S100, the user's query content is parsed to obtain query intent features. Based on the query intent features and personalized parameters, medical suggestion information is generated, specifically including the following steps: S110. Parse the user's query to obtain text, voice, and image input.

[0036] S120. Based on the text, voice and image inputs, a unified feature representation is obtained.

[0037] Specifically, the system pre-establishes a multimodal feature extraction process. For text input, keyword extraction is used to identify the core requirements; for speech input, speech recognition is used to transcribe it into text, and speech feature parameters are extracted; for image input, a pre-trained medical image classification model is used to extract visual features. The features from each modality are then converted into a unified representation using a feature fusion matrix.

[0038] S130. Determine the query intent features based on the unified feature representation.

[0039] Specifically, the system establishes an intent recognition rule base, containing a set of feature-intent mapping rules. A rule matching method is used to match the unified feature representation with the rule conditions to determine the query intent category. For complex intent scenarios, the primary intent is determined by priority ranking.

[0040] S140. Obtain personalized parameters based on user historical data.

[0041] In this embodiment, the personalized parameter acquisition module extracts personalized features from the user's historical data. Personalized parameters include basic information, medical history records, lifestyle habits, and interaction characteristics.

[0042] Specifically, the system constructs a user profile model and extracts personalized parameters from historical data through feature extraction rules. A data update mechanism is employed to periodically refresh user characteristics. A parameter importance scoring system is established to filter key personalized features.

[0043] S150. Generate medical advice information based on query intent characteristics and personalized parameters.

[0044] In one embodiment, refer to Figure 3 The real-time data from medical devices includes glucose concentration data from a continuous glucose monitor, blood pressure data from an ambulatory blood pressure monitor, heart rate data from an electrocardiogram monitor, and weight data from a smart scale. In step S200, based on the real-time data from the medical devices and medical advice information, health status characteristics are obtained, and based on these health status characteristics, health monitoring results are generated, specifically including the following steps: S210. Based on continuous sampling of glucose concentration data, intermittent sampling of blood pressure data, real-time sampling of heart rate data, and periodic sampling of weight data, a multi-scale standardized data stream is obtained.

[0045] S220. Based on the glucose concentration fluctuation rate, blood pressure diurnal rhythm, heart rate variability index and weight change trend in the multi-scale standardized data stream, combined with medical advice information, multidimensional physiological indicator characteristics are obtained.

[0046] In this embodiment, the multidimensional physiological indicators include short-term fluctuation characteristics, diurnal rhythm characteristics, and temporal variation characteristics. Glucose concentration fluctuation rate is obtained by calculating the change amplitude between adjacent time points; blood pressure diurnal rhythm is calculated by the difference between day and night blood pressure; heart rate variability indicators include time domain analysis parameters and frequency domain analysis parameters; weight change trend is obtained by calculating the change rate between adjacent sampling points.

[0047] Specifically, the system pre-establishes a feature extraction rule base and defines the calculation methods for various physiological indicators. It focuses on analyzing relevant indicators in medical advice information and adjusts the feature extraction weights accordingly. A sliding window method is used to calculate time-series features, and multi-dimensional feature vectors are formed through feature combination.

[0048] S230. Generate health monitoring results based on multidimensional physiological indicators and health risk models for diabetes, hypertension, and cardiovascular diseases.

[0049] In this embodiment, the health risk model is built based on medical guidelines and includes risk scoring standards and warning thresholds. Diabetes risk assessment focuses on blood glucose control levels and variability; hypertension risk assessment focuses on blood pressure target achievement rate and diurnal rhythm; cardiovascular disease risk assessment focuses on heart rate variability and body mass index.

[0050] Specifically, the system establishes a tiered early warning mechanism, classifying risk levels into four levels: normal, attention, warning, and intervention. A weighted scoring method is used to calculate a comprehensive risk score based on the degree of abnormality of each indicator. A health monitoring and assessment report is generated through rule-based reasoning, including risk level determination and specific indicator analysis.

[0051] In one embodiment, refer to Figure 4 The pre-set medical plan includes a medication reminder plan, a dietary intervention plan, an exercise guidance plan, and a follow-up visit plan. In step S300, based on the health monitoring results and the pre-set medical plan, the intervention execution characteristics are determined through progressive task decomposition based on task type characteristics and patient behavior characteristics. Based on the intervention execution characteristics, standardized intervention instructions are generated, specifically including the following steps: S310. Based on the health monitoring results and medication reminder plan, dietary intervention plan, exercise guidance plan, and follow-up visit plan, the task type characteristics are obtained.

[0052] Among them, task type characteristics include feedback frequency, task difficulty, and cognitive load.

[0053] In this embodiment, the task analysis module performs feature analysis on different medical plans. Feedback frequency is divided into real-time feedback, daily feedback, weekly feedback, and monthly feedback; task difficulty is divided into basic tasks, advanced tasks, and challenging tasks; cognitive load includes information complexity, number of operation steps, and attention requirements.

[0054] Specifically, the system pre-constructs a task feature evaluation matrix. Medication reminder plans focus on medication timing, dosage, and administration method; dietary intervention plans focus on meal arrangements, ingredient selection, and cooking methods; exercise guidance plans focus on exercise type, duration, and intensity control; and follow-up visit tracking plans focus on follow-up cycles, examination items, and indicator records. Feature values ​​for each task type are calculated using feature extraction rules.

[0055] S320. Based on the characteristics of the task type and the patient's previous execution records, the intervention execution characteristics are obtained through progressive task decomposition.

[0056] In this embodiment, the execution scheduling module decomposes and adjusts tasks based on their characteristics. The patient's past execution records include completion status, execution effect, and degree of adaptation.

[0057] S330. Generate standardized intervention instructions based on the characteristics of intervention implementation and medical standards.

[0058] In this embodiment, the instruction generation module converts the intervention execution features into specific instructions.

[0059] Specifically, the system pre-establishes a command template library, including text commands, graphic commands, and voice commands. Execution features are integrated into the command templates through template filling. Command optimization rules are used to adjust the expression and level of detail of the commands. A command confirmation mechanism is established to ensure that patients understand and accept the command content.

[0060] In one embodiment, refer to Figure 5 In step S320, based on the task type characteristics and the patient's previous execution records, the intervention execution characteristics are obtained through progressive task decomposition, specifically including the following steps: S321. Based on the time regularity, environmental relevance, and emotional state in the patient's past performance records, obtain the patient's behavioral characteristics.

[0061] Among them, temporal regularity includes the diurnal fluctuations and periodic changes in task execution, and environmental relevance includes the differences in task completion rates between indoor and outdoor, work and home scenarios.

[0062] In this embodiment, the behavior pattern recognition unit analyzes the behavioral patterns of patients performing tasks. Temporal regularity reflects the differences in performance ability at different times, including differences in performance in the morning, noon, and night, as well as cyclical changes on weekdays and rest days; environmental relevance reflects the performance effect in different scenarios, including completion status in indoor static scenarios, outdoor activity scenarios, busy work scenarios, and home relaxation scenarios; emotional state includes three types: active cooperation, passive coping, and resistance and rejection.

[0063] Specifically, the behavior pattern recognition unit is divided into four sub-units. The temporal regularity feature extraction sub-unit is used to obtain temporal features based on the patient's past performance records, including diurnal fluctuation features and periodic change features; the environmental correlation feature extraction sub-unit is used to obtain scene features, including completion rates for indoor static scenes, outdoor activity scenes, busy work scenes, and home relaxation scenes; the emotional state feature extraction sub-unit is used to obtain emotional features from feedback content, including the degree of active cooperation, the degree of passive coping, and the degree of resistance and rejection; the feature fusion sub-unit obtains the patient's behavioral features through weighted feature fusion.

[0064] S322. Based on patient behavior characteristics, and considering task difficulty, execution time, and environmental conditions, medical tasks are decomposed into multiple micro-task sequences.

[0065] In this embodiment, the task adaptive adjustment unit achieves fine-grained task decomposition. A micro-task sequence refers to the basic operational units into which a complex medical task is broken down. Task difficulty is categorized based on operational complexity; execution duration is determined by the decomposition granularity based on the duration; environmental conditions include site requirements, equipment needs, and personnel coordination.

[0066] Specifically, the task adaptation adjustment unit is divided into four sub-units. The task difficulty assessment sub-unit calculates the operational complexity score based on the patient's behavioral characteristics, including the number of operation steps, the level of attention required, and the difficulty of skill mastery; the execution duration calculation sub-unit determines the time parameters, including the duration of a single execution, the task interval duration, and the number of repetitions per day; the environmental condition matching sub-unit obtains the environmental requirement parameters, including site space requirements, medical device configuration requirements, and support staff cooperation requirements; and the micro-task generation sub-unit generates a sequence of micro-tasks based on the above parameters using a task decomposition algorithm. Each micro-task includes a description of the execution steps, completion criteria, and monitoring methods.

[0067] S323. Based on the micro-task sequence and task completion probability, the intervention execution characteristics are obtained.

[0068] Specifically, the system establishes an execution plan template library, pre-setting execution schemes for different scenarios. It calculates the probability of completing a sequence of micro-tasks using a probabilistic evaluation model; employs heuristic algorithms to optimize task sorting, improving overall execution efficiency; and establishes a dynamic adjustment mechanism to update the execution plan based on real-time feedback. Combined with time window constraints, it generates a specific execution schedule.

[0069] In one embodiment, refer to Figure 6 The method also includes the following steps: S610. Perform initial interaction analysis based on the patient's health record to obtain interaction state characteristics.

[0070] S620. Perform trend analysis based on interaction status characteristics and monthly health indicators to obtain health trend characteristics.

[0071] S630. Based on health trend characteristics and medical knowledge base, perform rule-based reasoning to obtain knowledge application characteristics.

[0072] S640. Generate an interactive interface based on the characteristics of knowledge application, and obtain multimodal input data based on the interactive interface.

[0073] Specifically, the system establishes an adaptive interface template library, pre-setting interface layout schemes for different scenarios. Adopting responsive design principles, it supports multi-terminal adaptation, including mobile, tablet, and desktop display modes. Interface elements include information display areas, operation control areas, and feedback recording areas, and flexible interface configuration is achieved through modular organization.

[0074] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0075] Secondly, this application provides a chronic disease management system. The chronic disease management system of this application will be described below in conjunction with the above-mentioned chronic disease management methods.

[0076] Reference Figure 7 A chronic disease management approach, comprising: The suggestion information generation module is used to parse the user's query content to obtain query intent features, and generate medical suggestion information based on the query intent features and personalized parameters; The monitoring result generation module is used to obtain health status characteristics based on real-time data from medical devices and medical advice information, and generate health monitoring results based on these health status characteristics. The intervention instruction generation module is used to determine the intervention execution characteristics based on the health monitoring results and the preset medical plan, through progressive task decomposition based on task type characteristics and patient behavior characteristics, and generate standardized intervention instructions based on the intervention execution characteristics. The historical data matching module is used to retrieve retrieval features from the memory system according to standardized intervention instructions, and obtain historical data matching results based on the retrieval features; The health record update module is used to calculate storage characteristics based on historical data matching results and real-time data, and update the patient's health record based on the storage characteristics.

[0077] In one embodiment, this application provides an electronic device, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps of a chronic disease management method.

[0078] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0079] In one embodiment, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0081] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method of chronic disease management, characterized by, Comprising the following steps: According to the user query content analysis, get query intention characteristics, based on the query intention characteristics and personalized parameters, generate medical advice information; According to the medical equipment real-time data and the medical advice information, obtain health status characteristics, based on the health status characteristics, generate health monitoring results; According to the health monitoring results and the preset medical scheme, based on the task type characteristics and the patient behavior characteristics through the progressive task decomposition to determine the intervention execution characteristics, based on the intervention execution characteristics, generate standardized intervention instructions; According to the standardized intervention instructions, retrieve features are obtained from the memory system, based on the retrieval features, the historical data matching results are obtained; According to the historical data matching results and real-time data, calculate storage characteristics, based on the storage characteristics, update the patient health record.

2. The chronic disease management method of claim 1, wherein, According to the user query content analysis, get query intention characteristics, based on the query intention characteristics and personalized parameters, generate medical advice information, specifically including the following steps: According to the user query content analysis, get text, voice and image input; According to the text, voice and image input, get unified feature representation; According to the unified feature representation, determine the query intention characteristics; According to the user historical data, get personalized parameters; According to the query intention characteristics and the personalized parameters, generate medical advice information.

3. The chronic disease management method of claim 2, wherein, The medical equipment real-time data includes glucose concentration data of continuous blood glucose monitor, blood pressure data of ambulatory blood pressure monitor, heart rate data of electrocardiogram monitor and body weight data of intelligent body weight scale; According to the medical equipment real-time data and the medical advice information, obtain health status characteristics, based on the health status characteristics, generate health monitoring results, specifically including the following steps: According to the continuous sampling of the glucose concentration data, the intermittent sampling of the blood pressure data, the real-time sampling of the heart rate data and the periodic sampling of the body weight data, get multi-scale standardized data flow; According to the glucose concentration fluctuation rate, blood pressure circadian rhythm, heart rate variability index and body weight change trend in the multi-scale standardized data flow, combined with the medical advice information, get multi-dimensional physiological index characteristics; According to the multi-dimensional physiological index characteristics and the health risk model of diabetes, hypertension and cardiovascular disease, generate health monitoring results.

4. The chronic disease management method of claim 1, wherein, The preset medical scheme includes medication reminder scheme, diet intervention scheme, exercise guidance scheme and follow-up tracking scheme, according to the health monitoring results and the preset medical scheme, based on the task type characteristics and the patient behavior characteristics through the progressive task decomposition to determine the intervention execution characteristics, based on the intervention execution characteristics, generate standardized intervention instructions, specifically including the following steps: According to the health monitoring results and the medication reminder scheme, diet intervention scheme, exercise guidance scheme and follow-up tracking scheme, get task type characteristics, wherein the task type characteristics include feedback frequency, task difficulty and cognitive load; According to the task type characteristics and the patient's past execution record, through the progressive task decomposition, get the intervention execution characteristics; According to the intervention execution characteristics and medical specification requirements, generate standardized intervention instructions.

5. The chronic disease management method of claim 4, wherein, According to the task type characteristics and the patient's past performance records, intervention execution characteristics are obtained through progressive task decomposition, specifically including the following steps: According to the time regularity, environmental correlation and emotional state in the patient's past performance records, patient behavior characteristics are obtained, wherein the time regularity includes day-to-day fluctuations and periodic changes in task performance, and the environmental correlation includes differences in task completion in indoor and outdoor, work and home scenarios; According to the patient behavior characteristics, medical tasks are decomposed into multiple micro-task sequences based on task difficulty, execution time and environmental conditions; According to the micro-task sequence and task completion probability, intervention execution characteristics are obtained.

6. The chronic disease management method of claim 1, wherein, The method further includes the following steps: According to the patient health records, an initial interaction analysis is performed to obtain interaction state characteristics; According to the interaction state characteristics and monthly health indicators, trend analysis is performed to obtain health trend characteristics; According to the health trend characteristics and the medical knowledge base, rule reasoning is performed to obtain knowledge application characteristics; According to the knowledge application characteristics, an interaction interface is generated, and based on the interaction interface, multi-modal input data is obtained.

7. A chronic disease management system, characterized by, It includes: The suggestion information generation module is used to analyze the user query content to obtain query intention characteristics, and generate medical suggestion information based on the query intention characteristics and personalized parameters; The monitoring result generation module is used to obtain health status characteristics based on real-time data of medical equipment and the medical suggestion information, and generate health monitoring results based on the health status characteristics; The intervention instruction generation module is used to determine intervention execution characteristics based on task type characteristics and patient behavior characteristics through progressive task decomposition based on the health monitoring results and the preset medical scheme, and generate standardized intervention instructions based on the intervention execution characteristics; The historical data matching module is used to obtain retrieval characteristics from the memory system based on the standardized intervention instructions, and obtain historical data matching results based on the retrieval characteristics; The health record updating module is used to calculate storage characteristics based on the historical data matching results and real-time data, and update the patient health records based on the storage characteristics.

8. An electronic device, comprising: The computer program is executed by the processor to realize the steps of the chronic disease management method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the chronic disease management method of any one of claims 1-6.

Images