Method and System for Monitoring Sign Data Based on Medical Internet of Things

Abstract

This application provides a method and system for monitoring physical sign data based on the medical Internet of Things. First, a first graph generation model is used to convert the sparsely collected first medical interaction scenario data to the target medical service feature domain, effectively solving the data sparsity problem and ensuring the consistency of data across different scenarios. Subsequently, based on the converted graph feature representation data, knowledge point prediction is performed through a basic medical service knowledge point prediction network, and the network parameters are continuously optimized through an error feedback mechanism, ultimately generating a high-performance medical service knowledge point prediction network. This medical service knowledge point prediction network can not only accurately predict the medical service knowledge points of the remote session trajectory data originating from the densely collected second medical interaction scenario, but also automatically generate a personalized physical sign data monitoring plan based on the prediction results, providing scientific and efficient decision-making support for doctors, and at the same time promoting the precision and intelligence of patient health management.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology. Specifically, it relates to a method and system for monitoring vital sign data based on the Internet of Medical Things (IoMT). Background Art

[0002] With the rapid development of medical technology and the wide application of Internet of Things technology, the Internet of Medical Things (IoMT) has shown great potential in the fields of health monitoring, disease diagnosis and management. The IoMT integrates various sensors, wearable devices, remote medical systems, etc., to achieve real-time collection, transmission and analysis of patients' vital sign data, providing strong support for personalized medicine and precision health management. However, in practical applications, there are problems such as uneven collection density, diverse data formats, and lack of knowledge annotation in vital sign data under different medical interaction scenarios, which seriously restrict the accuracy and efficiency of vital sign data monitoring.

[0003] Traditional vital sign data monitoring methods often rely on data in a single scenario and are difficult to effectively cope with the challenges brought by cross-scenario data differences. Especially in remote areas or resource-constrained medical interaction scenarios, the collection density of vital sign data is low and the data quality is uneven, which greatly reduces the performance of knowledge discovery and prediction models based on these data. At the same time, the vital sign data under different medical interaction scenarios often follow their own unique feature distributions and service models. Directly migrating models for cross-scenario prediction often has poor results and may even lead to misleading results.

[0004] To overcome the above problems, related technologies have begun to explore the use of advanced technologies such as graph neural networks (GNNs) to model and analyze IoMT data. Graph neural networks can effectively capture feature information in complex networks by constructing graph structures to represent the relationships between data, providing new ideas for cross-scenario data fusion and knowledge prediction. However, most of the existing technologies focus on data modeling in a single scenario and lack effective methods for multi-scenario data fusion and transformation. Especially when facing medical interaction scenarios with significant differences in collection density, how to construct a unified target feature domain and achieve efficient knowledge prediction remains a technical problem to be solved urgently. Summary of the Invention

[0005] In view of the problems mentioned above, in combination with the first aspect of this application, embodiments of this application provide a method for monitoring vital sign data based on the Internet of Medical Things. The method includes:

[0006] Obtain first remote session trajectory training data from a first medical interaction scenario. The first remote session trajectory training data carries first medical service knowledge point annotation data, and the first medical service knowledge point annotation data represents the actual medical service knowledge points of the first remote session trajectory training data;

[0007] Using a first graph generation model to perform graph feature representation on the first remote session trajectory training data, generating first graph feature representation data. The first graph generation model is used to transform remote session trajectory data originating from the first medical interaction scenario into a target medical service feature domain, where the target medical service feature domain is the medical service feature domain corresponding to the graph feature representation data of remote session trajectory data originating from a second medical interaction scenario. The first medical interaction scenario and the second medical interaction scenario are different medical interaction scenarios, and the acquisition density of remote session trajectory data originating from the first medical interaction scenario is less than that of remote session trajectory data originating from the second medical interaction scenario;

[0008] Based on the first graph feature representation data, using a basic medical service knowledge point prediction network to perform knowledge point prediction, generating predicted medical service knowledge points;

[0009] Based on the error between the predicted medical service knowledge points and the first medical service knowledge point annotation data, updating the network parameter information of the basic medical service knowledge point prediction network, generating a medical service knowledge point prediction network, so that the medical service knowledge point prediction network is used to predict the medical service knowledge points of remote session trajectory data originating from the second medical interaction scenario, and generating a corresponding physical sign data monitoring plan based on the medical service knowledge points of the remote session trajectory data originating from the second medical interaction scenario.

[0010] On the other hand, an embodiment of the present application further provides a medical Internet of Things service system, including a processor and a machine-readable storage medium. The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0011] Based on the above aspects, the embodiments of the present application realize data feature conversion and knowledge prediction across different medical interaction scenarios by combining a graph generation model and a deep learning network, significantly improving the accuracy and generalization ability of medical service knowledge point prediction. Specifically, first, the first graph generation model is used to convert the sparsely collected first medical interaction scenario data to the target medical service feature domain, effectively solving the data sparsity problem and ensuring data consistency across different scenarios. Subsequently, based on the converted graph feature representation data, knowledge point prediction is performed through a basic medical service knowledge point prediction network, and the network parameters are continuously optimized through an error feedback mechanism, finally generating a high-performance medical service knowledge point prediction network. This medical service knowledge point prediction network can not only accurately predict the medical service knowledge points of the remote session trajectory data from the second medical interaction scenario with dense collection, but also automatically generate a personalized physical sign data monitoring plan based on the prediction results, providing scientific and efficient decision-making support for doctors, and at the same time promoting the precision and intelligence of patient health management. Thereby, the medical service efficiency and quality are greatly improved. Brief Description of the Drawings

[0012] Figure 1 It is a schematic flowchart of the execution process of the physical sign data monitoring method based on the medical Internet of Things provided by the embodiments of the present application.

[0013] Figure 2 It is a schematic hardware architecture diagram of the medical Internet of Things service system provided by the embodiments of the present application. Detailed Embodiments

[0014] The present application will be specifically described below in conjunction with the accompanying drawings of the specification. Figure 1 It is a schematic flowchart of the physical sign data monitoring method based on the medical Internet of Things provided by an embodiment of the present application. The physical sign data monitoring method based on the medical Internet of Things will be introduced in detail below.

[0015] Step S110, obtain the first remote session trajectory training data from the first medical interaction scenario, where the first remote session trajectory training data carries the first medical service knowledge point annotation data, and the first medical service knowledge point annotation data represents the actual medical service knowledge points of the first remote session trajectory training data.

[0016] In this embodiment, in the medical field, different medical interaction scenarios will generate a large amount of remote session trajectory data, which contains rich medical service information and interaction details between patients and doctors. As the execution entity, the server will first obtain the first remote session trajectory training data from the first medical interaction scenario, and this process usually involves extracting data from a hospital information system (HIS), an electronic medical record system (EMR), or a telemedicine platform.

[0017] For example, assume that the first medical interaction scenario is an online medical consultation platform in a remote area. Due to geographical and resource limitations, the remote session data between users (patients) and doctors on this online medical consultation platform is relatively scarce and sparse. The server regularly pulls remote session records from the database of this online medical consultation platform. These remote session records include the basic information of patients, the main complaints, the doctor's interrogation process, diagnostic suggestions, and follow-up treatment suggestions, etc. Each remote session record is regarded as a first remote session trajectory training data, and each first remote session trajectory training data is attached with first medical service knowledge point annotation data. These first medical service knowledge point annotation data are manually annotated by professional medical staff according to the session content. For example, according to the session content, the key medical knowledge points involved are manually annotated, such as "respiratory infection management", "chronic disease follow-up guidance", etc. The server associates these first medical service knowledge point annotation data with the corresponding first remote session trajectory data to form a training data set.

[0018] Step S120, use the first graph generation model to perform graph feature representation on the first remote session trajectory training data to generate first graph feature representation data. The first graph generation model is used to convert the remote session trajectory data from the first medical interaction scenario to the target medical service feature domain. The target medical service feature domain is the medical service feature domain where the graph feature representation data corresponding to the remote session trajectory data from the second medical interaction scenario is located. The first medical interaction scenario and the second medical interaction scenario are different medical interaction scenarios. The acquisition density of the remote session trajectory data from the first medical interaction scenario is less than the acquisition density of the remote session trajectory data from the second medical interaction scenario.

[0019] In this embodiment, due to the low data acquisition density of the first medical interaction scenario, directly using these first remote session trajectory training data to train the knowledge point prediction model may not yield good results. Therefore, the server uses the first graph generation model to convert these sparse remote session trajectory data into a more general target medical service feature domain, which is usually constructed based on the data of the second medical interaction scenario with a higher acquisition density.

[0020] For example, the server first loads a pre-trained first graph generation model. This first graph generation model is based on a graph neural network (GNN) architecture and can capture the complex relationship structure in the remote session trajectory. The input of the first graph generation model is the first remote session trajectory training data in the first medical interaction scenario. Each piece of the first remote session trajectory training data is represented as a graph structure, where the nodes can be patients, doctors, medical events, etc., and the edges represent the relationships between them (such as consultation, diagnosis, advice, etc.). The first graph generation model extracts high-level features from the graph structure through multi-layer graph convolution operations and generates corresponding first graph feature representation data. These first graph feature representation data are mapped into the target medical service feature domain, which is constructed based on an online medical consultation platform (the second medical interaction scenario) of a large urban hospital, and the data collection density of this online medical consultation platform is much higher than that of the platform in remote areas.

[0021] Step S130: Based on the first graph feature representation data, use the basic medical service knowledge point prediction network to perform knowledge point prediction and generate predicted medical service knowledge points.

[0022] In this embodiment, after obtaining the converted graph feature representation data, the server then uses the basic medical service knowledge point prediction network to perform knowledge point prediction. This basic medical service knowledge point prediction network is usually a deep learning model, such as a combination of a convolutional neural network (CNN) or a recurrent neural network (RNN) and an attention mechanism, which can extract key information from the graph feature representation data and predict the corresponding medical service knowledge points.

[0023] For example, the server inputs the first graph feature representation data into the basic medical service knowledge point prediction network. The basic medical service knowledge point prediction network first further processes the graph feature representation data through several convolutional layers or recurrent layers to extract features related to medical service knowledge points. Then, the attention mechanism is used to focus on the most relevant features to generate predicted medical service knowledge points.

[0024] Step S140: Based on the error between the predicted medical service knowledge points and the first medical service knowledge point annotation data, update the network parameter information of the basic medical service knowledge point prediction network to generate a medical service knowledge point prediction network, so that the medical service knowledge point prediction network is used to predict the medical service knowledge points of the remote session trajectory data from the second medical interaction scenario, and generate a corresponding physical sign data monitoring plan based on the medical service knowledge points of the remote session trajectory data from the second medical interaction scenario.

[0025] In this embodiment, to improve the accuracy of the basic medical service knowledge point prediction network, the server calculates the error between the predicted medical service knowledge points and the first medical service knowledge point annotation data, and updates the network parameter information of the basic medical service knowledge point prediction network according to this error. This process is usually implemented through the backpropagation algorithm, aiming to minimize the prediction error and enable the model to more accurately predict medical service knowledge points.

[0026] For example, the server compares the predicted knowledge points with the annotated knowledge points and calculates the error between them (such as cross-entropy loss). Then, using optimization algorithms such as gradient descent, the weights and biases of the prediction network are updated according to the error backpropagation. After multiple iterative trainings, the performance of the prediction network gradually improves and the error gradually decreases. Finally, the server obtains a trained medical service knowledge point prediction network. This medical service knowledge point prediction network can not only process sparse data in the first medical interaction scenario, but also adapt to dense data in the second medical interaction scenario, providing strong support for the subsequent generation of the physical sign data monitoring scheme.

[0027] Thus, the first remote session trajectory training data from the sparse acquisition scenario is converted into a general graph feature representation, and a high-performance medical service knowledge point prediction network is trained using this knowledge. This medical service knowledge point prediction network can not only improve the efficiency and quality of remote medical consultations, but also provide a scientific basis for the subsequent generation of the physical sign data monitoring scheme, thereby promoting the rational allocation of medical resources and the effective management of patient health.

[0028] For example, to generate a physical sign data monitoring scheme, the server needs to pre-define a set of physical sign monitoring templates. These physical sign monitoring templates are designed according to different medical service knowledge points and contain information such as physical sign monitoring indicators, monitoring frequencies, and monitoring methods for different diseases or health problems. For example, for the knowledge point of "hypertension management", the physical sign monitoring template may include respiratory monitoring indicators, monitoring frequency (twice a day, morning and evening), and monitoring methods.

[0029] The server matches the predicted medical service knowledge points with the physical sign monitoring templates to find the physical sign monitoring templates corresponding to the knowledge points.

[0030] Although the vital sign monitoring template provides a basic framework for vital sign monitoring, the specific conditions of different patients may vary. Therefore, the server also needs to personalize the monitoring plan based on the patient's personal information (such as age, gender, medical history, etc.) and specific information in the session track (such as symptom description, doctor's advice, etc.). For example, for elderly patients or patients with severe conditions, it may be necessary to increase the monitoring frequency or add additional monitoring indicators. After matching and personalization, the server finally generates a vital sign data monitoring plan for each remote session track, and these vital sign data monitoring plans are sent to patients or medical staff in the form of electronic documents or application notifications to guide them in monitoring and recording vital sign data. At the same time, the server can also integrate a vital sign data collection system, allowing patients to upload monitoring data in real time through wearable devices, mobile applications, etc. The server analyzes and processes these data in real time and issues an alarm immediately once an abnormal situation is detected, so that medical staff can intervene and handle it in a timely manner.

[0031] Based on the above steps, the embodiment of the present application realizes data feature conversion and knowledge prediction across different medical interaction scenarios by combining the graph generation model with the deep learning network, significantly improving the accuracy and generalization ability of medical service knowledge point prediction. Specifically, first, the first graph generation model is used to convert the sparsely collected data of the first medical interaction scenario to the target medical service feature domain, effectively solving the data sparsity problem and ensuring the consistency of data across different scenarios. Subsequently, based on the converted graph feature representation data, knowledge point prediction is performed through the basic medical service knowledge point prediction network, and the network parameters are continuously optimized through the error feedback mechanism, and finally a high-performance medical service knowledge point prediction network is generated. This medical service knowledge point prediction network can not only accurately predict the medical service knowledge points of the remote session track data from the second medical interaction scenario with dense collection, but also automatically generate a personalized vital sign data monitoring plan based on the prediction results, providing scientific and efficient decision-making support for doctors, and at the same time promoting the precision and intelligence of patient health management. Thus, the efficiency and quality of medical services are greatly improved.

[0032] In a possible implementation manner, the method further includes:

[0033] Step A110, obtaining the to-be-predicted remote session track data sourced from the second medical interaction scenario.

[0034] In this embodiment, the server regularly pulls new to-be-predicted remote session track data from the database of the second medical interaction scenario (the online medical consultation platform of a large urban hospital) as the prediction object. These to-be-predicted remote session track data have not been knowledge point annotated, but have a similar format to the training data, including the patient's basic information, chief complaint, doctor's interrogation process, etc.

[0035] Specifically, the server establishes a connection with the database of the second medical interaction scenario. According to a preset time interval or data update flag, the server performs a data pulling operation, and the pulled data is temporarily stored in the local database of the server for subsequent processing.

[0036] Step A120: Use the second graph generation model to perform graph feature representation on the to-be-predicted remote session trajectory data, generating second graph feature representation data. The second graph generation model is used to convert the remote session trajectory data from the second medical interaction scenario into the target medical service feature domain.

[0037] Different from the first graph generation model in the first medical interaction scenario, the second graph generation model is specifically used to process the remote session trajectory data in the second medical interaction scenario. This second graph generation model is also based on the graph neural network (GNN) architecture but has been optimized according to the data characteristics of the second medical interaction scenario.

[0038] Specifically, the server loads the pre-trained second graph generation model and inputs the to-be-predicted remote session trajectory data into the second graph generation model. Each piece of to-be-predicted remote session trajectory data is converted into a graph structure, and the definitions of nodes and edges are consistent with the training data. The second graph generation model extracts high-level features in the graph structure through multi-layer graph convolution operations, generating corresponding second graph feature representation data, which are also mapped into the target medical service feature domain to ensure consistency with the training data.

[0039] Step A130: Based on the second graph feature representation data, use the medical service knowledge point prediction network to perform knowledge point prediction, generating the medical service knowledge points of the to-be-predicted remote session trajectory data.

[0040] The second graph feature representation data after being converted by the second graph generation model is now input into the already trained medical service knowledge point prediction network, which can predict the corresponding medical service knowledge points based on the graph feature representation data.

[0041] Specifically, the server inputs the second graph feature representation data into the medical service knowledge point prediction network. The medical service knowledge point prediction network further processes the graph feature representation data through multi-layer convolutional layers or recurrent layers, extracting features related to medical service knowledge points. Using the attention mechanism to focus on the most relevant features, it generates predicted medical service knowledge points. The prediction results are output in the form of labels or texts, representing the predicted medical service knowledge points, such as "hypertension management", "diabetes diet guidance", etc.

[0042] For example, assume that there is a new remote session track data in the second medical interaction scenario, which is a regular follow-up record of a hypertensive patient. The server retrieves the follow-up record of this patient from the database, including blood pressure measurement results, the patient's self-reported symptoms, the doctor's interrogation process, etc. The second graph generation model converts the follow-up record into a graph structure and extracts key features, such as "persistently high blood pressure", "self-reported dizziness", etc., through graph convolution operations, generating corresponding second graph feature representation data. The medical service knowledge point prediction network receives the second graph feature representation data and predicts "hypertension management" as the main medical service knowledge point after processing. The server outputs the prediction result in text form, suggesting that the doctor strengthen the hypertension management of this patient, including adjusting the medication plan, increasing the monitoring frequency, etc.

[0043] In another example, the server processes a comprehensive care record of a chronic disease patient (such as a diabetic patient). The processing flow is similar to the above example, but the predicted medical service knowledge points may be more complex and comprehensive, such as "comprehensive diabetes management", which involves multiple aspects such as diet guidance, exercise advice, and blood glucose monitoring.

[0044] In a possible implementation manner, the first graph generation model includes a first interaction scenario model and a second interaction scenario model. The first interaction scenario model is used to convert the remote session track data from the first medical interaction scenario into the target medical service feature domain, and the second interaction scenario model is used to convert the remote session track data from the second medical interaction scenario into the target medical service feature domain.

[0045] Step S120 includes: using the first interaction scenario model to perform graph feature representation on the first remote session track training data, generating first graph feature representation data.

[0046] The method further includes:

[0047] Step B110, obtaining the to-be-predicted remote session track data from the second medical interaction scenario.

[0048] Step B120, using the second interaction scenario model to perform graph feature representation on the to-be-predicted remote session track data, generating second graph feature representation data.

[0049] Step B130, based on the second graph feature representation data, using the medical service knowledge point prediction network to perform knowledge point prediction, generating the medical service knowledge points of the to-be-predicted remote session track data.

[0050] In this embodiment, a first graph generation model is predefined and trained in the server. The first graph generation model includes two sub-models: a first interaction scenario model and a second interaction scenario model, and these two models are optimized according to the data characteristics of two different medical interaction scenarios respectively.

[0051] Specifically, in the training stage, the server uses a large amount of historical data from the first medical interaction scenario and the second medical interaction scenario to train the first interaction scenario model and the second interaction scenario model respectively. These models are based on the graph neural network (GNN) architecture, which can capture the complex relationship structure in the remote session trajectory and transform it into a unified target medical service feature domain. After training, the server integrates these two sub-models into the first graph generation model to enable flexible switching when processing data from different sources.

[0052] Thus, the server obtains a batch of remote session trajectory training data from the first medical interaction scenario and is ready to use the first interaction scenario model to perform graph feature representation on it.

[0053] Specifically, the server first preprocesses the first remote session trajectory training data, including operations such as cleaning, formatting, and standardization to ensure data quality. Then, each preprocessed first remote session trajectory data is converted into a graph structure, where the nodes represent the entities in the session (such as patients, doctors, medical events, etc.), and the edges represent the relationships between the entities (such as consultations, diagnoses, suggestions, etc.). Using the first interaction scenario model, the server performs multi-layer graph convolution operations on the constructed graph structure to extract the high-level features in the graph structure, and these features can characterize the key information in the remote session trajectory. After feature extraction, the server generates corresponding first graph feature representation data, and these first graph feature representation data are mapped into the target medical service feature domain for subsequent knowledge point prediction tasks.

[0054] As the system runs, the server needs to obtain new remote session trajectory data from the second medical interaction scenario in real time for knowledge point prediction. Specifically, the server establishes a real-time synchronization mechanism with the database of the second medical interaction scenario to ensure that the latest remote session trajectory data can be obtained in a timely manner. From the synchronized data, the server filters out the to-be-predicted remote session trajectory data that needs to be predicted for knowledge points, and these to-be-predicted remote session trajectory data may be filtered based on specific conditions or rules, such as only selecting session trajectories containing records of specific diseases or symptoms.

[0055] For the to-be-predicted remote session trajectory data selected, the server uses the second interaction scenario model to perform graph feature representation on it. Similarly, the server first converts the to-be-predicted remote session trajectory data into a graph structure. However, this time the second interaction scenario model is used, and this second interaction scenario model is optimized for the data characteristics of the second medical interaction scenario. Through multi-layer graph convolution operations, the server extracts high-level features in the graph structure. Similarly, these features are mapped into the target medical service feature domain to generate the second graph feature representation data.

[0056] After obtaining the second graph feature representation data, the server uses the trained medical service knowledge point prediction network to perform knowledge point prediction on it. For example, the second graph feature representation data can be input into the medical service knowledge point prediction network, and the input data is further processed through multi-layer convolutional layers or recurrent layers, and the attention mechanism is used to focus on the most relevant features. Finally, the network outputs the predicted medical service knowledge points, such as "respiratory infection management", "chronic disease follow-up guidance", etc. Thus, the prediction results are output in text or other forms for doctors or patients to refer to.

[0057] In a possible implementation manner, the method includes:

[0058] Step C110, obtaining a plurality of remote session trajectory training data combinations, where the remote session trajectory training data combinations include first member remote session trajectory training data from the first medical interaction scenario and second member remote session trajectory training data from the second medical interaction scenario, and the first member remote session trajectory training data and the second member remote session trajectory training data from the same remote session trajectory training data combination are the same medical service data.

[0059] In this embodiment, in this scenario, the server is responsible for processing and optimizing the remote session trajectory data of two different medical interaction scenarios (the first medical interaction scenario and the second medical interaction scenario). The server trains and optimizes two interaction scenario models (the first interaction scenario model and the second interaction scenario model) to ensure that these models can accurately convert the remote session trajectory data in their respective scenarios into a unified target medical service feature domain.

[0060] For example, remote session trajectory data can be collected from two different medical interaction scenarios and combined. Each combination contains two members: the first member remote session trajectory training data from the first medical interaction scenario and the second member remote session trajectory training data from the second medical interaction scenario, and these two members share the same medical service data, that is, they are involved in the same medical event or patient case.

[0061] Specifically, the server establishes connections with the databases of the first medical interaction scenario and the second medical interaction scenario respectively, and pulls remote session trajectory data therefrom. The data from different scenarios are matched according to medical service data (such as patient ID, medical event type, etc.) to form multiple remote session trajectory training data combinations. The data combinations after matching are stored in the local database of the server for subsequent processing.

[0062] Step C120, for the target remote session trajectory training data combination among the multiple remote session trajectory training data combinations, use the basic first interaction scenario model to perform graph feature representation on the first member remote session trajectory training data in the target remote session trajectory training data combination, and generate first member graph feature representation data.

[0063] For each remote session trajectory training data combination, the server first regards it as the target combination, and uses the basic first interaction scenario model and the basic second interaction scenario model respectively to perform graph feature representation on the first member and the second member in the combination.

[0064] For example, the server converts the first member remote session trajectory training data and the second member remote session trajectory training data in the first data combination into graph structures respectively. Use the basic first interaction scenario model to perform multi-layer graph convolution operations on the graph structure of the first member, and extract the first member graph feature representation data. Use the basic second interaction scenario model to perform multi-layer graph convolution operations on the graph structure of the second member, and extract the second member graph feature representation data.

[0065] Step C130, use the basic second interaction scenario model to perform graph feature representation on the second member remote session trajectory training data in the target remote session trajectory training data combination, and generate second member graph feature representation data.

[0066] Step C140, use the multiple remote session trajectory training data combinations as the target remote session trajectory training data combinations respectively, and generate multiple pieces of the first member graph feature representation data and multiple pieces of the second member graph feature representation data.

[0067] The server repeats the above process for all remote session trajectory training data combinations to generate corresponding first member graph feature representation data and second member graph feature representation data for each combination.

[0068] Specifically, the server traverses all remote session trajectory training data combinations and repeats the above graph structure construction and feature extraction operations until all combinations are processed.

[0069] Step C150: Update the network parameter information of the basic first interaction scenario model and the network parameter information of the basic second interaction scenario model according to the network optimization objective of minimizing the feature distance of the feature data combination of the target member graph and maximizing the feature distance of the feature data combination of other member graphs. Generate the first interaction scenario model and the second interaction scenario model. The first member graph feature representation data and the second member graph feature representation data covered by the feature data combination of the target member graph are generated based on the same remote session trajectory training data combination, and the first member graph feature representation data and the second member graph feature representation data covered by the feature data combination of other member graphs are not generated based on the same remote session trajectory training data combination.

[0070] In this embodiment, the server uses the generated multiple groups of first member graph feature representation data and second member graph feature representation data to optimize the network parameters of the basic first interaction scenario model and the basic second interaction scenario model. The optimization objective is to minimize the distance between the first member graph feature representation data and the second member graph feature representation data from the same remote session trajectory training data combination in the feature space, while maximizing the distance between the graph feature representation data from different combinations.

[0071] Specifically, the server defines a loss function that simultaneously considers minimizing the feature distance of the target member graph feature representation data combination and maximizing the feature distance of other member graph feature representation data combinations. Then, use gradient descent or other optimization algorithms to update the network parameters of the basic first interaction scenario model and the basic second interaction scenario model to minimize the loss function. The above process is repeated multiple times until the loss function converges to a stable value, indicating that the model has been sufficiently trained.

[0072] After multiple iterative trainings, the server obtains the optimized first interaction scenario model and second interaction scenario model. The first interaction scenario model and the second interaction scenario model can now better transform the remote session trajectory data in their respective scenarios into a unified target medical service feature domain. Save the trained models to a local or remote repository for use in subsequent knowledge point prediction tasks. At the same time, the server can also further evaluate and test the first interaction scenario model and the second interaction scenario model to ensure that their performance and accuracy meet the expected requirements.

[0073] In a possible implementation manner, if the remote session trajectory data from the first medical interaction scenario is sparse feature data and the remote session trajectory data from the second medical interaction scenario is dense feature data, then the method further includes:

[0074] Step D110: Obtain x pieces of second medical service knowledge point annotation data and a first guiding data sample including knowledge point placeholders. The second medical service knowledge point annotation data is used to describe the medical service knowledge points of the remote session trajectory data originating from the second medical interaction scenario. The first guiding data sample represents that a deep learning network generates multiple dense feature semantic data, and multiple medical image feature segments described by the multiple dense feature semantic data constitute target dense feature data related to the medical service knowledge point annotation data corresponding to the knowledge point placeholder, where x is a positive integer.

[0075] Step D120: Load the x pieces of second medical service knowledge point annotation data into the knowledge point placeholders in the first guiding data sample to generate first guiding data.

[0076] Step D130: Based on the first guiding data, use the deep learning network to generate the first remote session trajectory training data, where the first remote session trajectory training data is dense feature semantic data related to the x pieces of second medical service knowledge point annotation data.

[0077] Step D140: Output the x pieces of second medical service knowledge point annotation data as the first medical service knowledge point annotation data of the first remote session trajectory training data.

[0078] In this scenario, the server faces two medical interaction scenarios with significantly different data sparsity and density. To utilize the rich dense feature data in the second medical interaction scenario to enhance the training effect of the sparse feature data in the first medical interaction scenario, the server adopts a method of data enhancement and transformation.

[0079] For example, the server obtains x pieces of second medical service knowledge point annotation data from the database of the second medical interaction scenario. These data describe in detail the medical service knowledge points in the remote session trajectory, such as "diabetes management", "hypertension control", etc. At the same time, the server also has a predefined first guiding data sample, which is a template used to guide the deep learning network on how to generate dense feature semantic data related to specific knowledge points.

[0080] Specifically, the server extracts x pieces of second medical service knowledge point annotation data from the database of the second medical interaction scenario through SQL queries or other data retrieval methods. Then, it loads the first guiding data sample from local storage or a remote repository. The first guiding data sample contains multiple placeholders, and the knowledge point placeholder is used to insert specific medical service knowledge point annotation data later.

[0081] Next, the server loads the x second medical service knowledge point annotation data it has obtained into the knowledge point placeholder positions in the first guidance data sample, thereby generating the first guidance data containing specific knowledge point information.

[0082] For example, the server traverses the x second medical service knowledge point annotation data and inserts them one by one into the knowledge point placeholders in the first guidance data sample. After the insertion is completed, the server checks the format of the first guidance data to ensure that all placeholders have been filled correctly and the data format meets the requirements of subsequent processing.

[0083] Then, the server uses a pre-trained deep learning network (such as a generative adversarial network GAN, variational autoencoder VAE, etc.) to process the first guidance data, generating dense feature semantic data related to the x second medical service knowledge point annotation data. These data are richer and denser in features and are similar to the remote session trajectory data in the second medical interaction scenario.

[0084] For example, the server loads the pre-trained deep learning network model, takes the first guidance data as input and passes it to the deep learning network, and generates dense feature semantic data according to the input first guidance data. These data are similar in form to the remote session trajectory training data but are richer and denser in features. Thus, the generated dense feature semantic data is organized into the same format and structure as the first remote session trajectory training data.

[0085] Finally, the server directly outputs the x second medical service knowledge point annotation data originally used to generate the first guidance data as the first medical service knowledge point annotation data of the first remote session trajectory training data, and these annotation data are used in subsequent training and evaluation processes.

[0086] For example, the server associates the x second medical service knowledge point annotation data with the generated dense feature semantic data (i.e., the first remote session trajectory training data), and then outputs the associated data in a standard format to the local storage or remote repository of the server for subsequent use.

[0087] Through the above steps, the rich dense feature data knowledge in the second medical interaction scenario is introduced into the sparse remote session trajectory data training in the first medical interaction scenario, thereby enhancing the diversity and richness of the training data and hopefully improving the accuracy and generalization ability of the medical service knowledge point prediction model.

[0088] In a possible implementation, if the first guided data sample further characterizes that the deep learning network generates other medical service knowledge point annotation data related to the target dense feature data, step D140 includes: outputting the x second medical service knowledge point annotation data and the other medical service knowledge point annotation data as the first medical service knowledge point annotation data of the first remote session trajectory training data.

[0089] In this scenario, the server not only uses the dense feature data in the second medical interaction scenario to generate the first remote session trajectory training data related to specific knowledge points, but also the first guided data sample indicates that the deep learning network can generate additional medical service knowledge point annotation data associated with the target dense feature data. This means that in addition to the original x second medical service knowledge point annotation data, the deep learning network can also automatically discover and annotate other potential relevant knowledge points.

[0090] For example, the server first confirms the design of the first guided data sample, which not only contains placeholders for inserting specific knowledge point annotation data, but also has instructions or structures for guiding the deep learning network to generate additional relevant knowledge point annotation data.

[0091] Specifically, the server reviews the document or metadata of the first guided data sample to confirm that it can support the function of generating additional medical service knowledge point annotation data. If necessary, the server may fine-tune or configure the first guided data sample to ensure that it meets this requirement.

[0092] The server loads the x second medical service knowledge point annotation data into the first guided data sample and triggers the deep learning network to start processing. During the processing, the deep learning network not only generates the target dense feature data according to the given knowledge points, but also automatically explores and generates other medical service knowledge point annotation data associated with these data.

[0093] Specifically, the server fills the x second medical service knowledge point annotation data into the corresponding placeholders of the first guided data sample. The deep learning network is started, and the filled first guided data is passed to the network as input. When processing the input data, the deep learning network not only generates the target dense feature data (i.e., the first remote session trajectory training data) related to the given knowledge points, but also automatically discovers and annotates other potential relevant medical service knowledge points according to the patterns and features in the data.

[0094] After the deep learning network finishes processing, the server collects all the generated medical service knowledge point annotation data, including the original x second medical service knowledge point annotation data and the additional knowledge point annotation data automatically generated by the network. Then, the server outputs this data as the first medical service knowledge point annotation data of the first remote session trajectory training data.

[0095] For example, the server extracts all the generated medical service knowledge point annotation data from the output of the deep learning network, which may exist in the form of a list, an array, or other data structures. The server integrates the original x second medical service knowledge point annotation data with the additional knowledge point annotation data automatically generated by the network to form a complete first medical service knowledge point annotation data set. Finally, the server outputs the integrated first medical service knowledge point annotation data set in an appropriate format to local storage or a remote repository for subsequent training and evaluation processes.

[0096] Through this process, not only is the dense feature data in the second medical interaction scenario utilized to enhance the richness of the first remote session trajectory training data, but also additional potential knowledge points are discovered through the automatic exploration ability of the deep learning network, further improving the diversity and accuracy of the training data.

[0097] In a possible implementation manner, the first guidance data sample also represents that when there are contradictions in the x second medical service knowledge point annotation data, the deep learning network intercepts and generates the first remote session trajectory training data related to the target dense feature data. The method further includes: extracting the x second medical service knowledge point annotation data from y second medical service knowledge point annotation data, where y is an integer greater than x.

[0098] In this embodiment, in this scenario, the data processed by the server is more complex because there may be contradictions or inconsistencies among the second medical service knowledge point annotation data. To address this situation, the first guidance data sample is designed to be able to guide the deep learning network to pause generating the first remote session trajectory training data when detecting data contradictions, and the server needs to carefully select x non-contradictory annotation data from a larger data set (y annotation data) for training.

[0099] The server has a data set containing y second medical service knowledge point annotation data, where y is an integer greater than x. This data set is much larger than the original x annotation data used to generate the first guidance data, providing more selection space to avoid data contradictions. For example, the server retrieves the complete y annotation data set from the database of the second medical interaction scenario and stores it in local storage or a remote repository for subsequent processing.

[0100] Before extracting x labeled data from y labeled data, the server needs to clean and detect contradictions in the entire dataset to ensure that the x labeled data selected are logically consistent and do not conflict with each other. For example, the server performs a series of data cleaning operations, such as removing duplicates, correcting mislabeled data, filling in missing values, etc., to improve data quality. Then, specialized algorithms or logic are used to detect potential contradiction points in the dataset, which may involve comparing information between different labels to check if they conflict or are inconsistent with each other.

[0101] After confirming that there are no or all contradictions have been resolved in the y labeled dataset, the server begins to extract x labeled data from it for generating the first remote session trajectory training data.

[0102] For example, the server may adopt random sampling or sampling methods based on specific strategies (such as the importance of knowledge points, data diversity, etc.) to extract x from the y labeled data. After extraction, the server may verify the selected x labeled data again to ensure that there are no contradictions between them and that they meet the training requirements.

[0103] Using the extracted x non - contradictory labeled data, the server generates the first guiding data and triggers the deep - learning network for processing. During this process, the first guiding data sample will monitor the operations of the deep - learning network. Once a contradiction in the labeled data is detected (although theoretically it should not occur in this scenario, but as a preventive measure), the process of generating the first remote session trajectory training data will be intercepted.

[0104] For example, the server generates the first guiding data according to the previous steps and loads the x non - contradictory labeled data into it. The deep - learning network is started and the first guiding data is passed in for processing. During the process of generating the target dense feature data, the deep - learning network will monitor the consistency of the labeled data according to the instructions of the first guiding data sample.

[0105] Although in this specific scenario, the server has taken measures to ensure that there are no contradictions between the selected x labeled data, as a robust design, the first guiding data sample still contains the logic for contradiction interception. If (theoretically unlikely) the deep - learning network detects a contradiction during processing, the generation of the first remote session trajectory training data will be paused according to the instructions of the first guiding data sample.

[0106] If the deep learning network does detect a contradiction during the generation process (although this is unlikely to happen in the current scenario), it will pause the operation according to the instructions of the first guiding data sample and send an error report to the server. After receiving the error report, the server will conduct further investigation and processing, which may include rechecking the consistency of the labeled data, adjusting the parameters or logic of the deep learning network, etc.

[0107] However, in most cases, since the server has pre-cleaned and detected contradictions in y labeled data and selected x non-contradictory labeled data from them to generate the first guiding data, it is unlikely that the deep learning network will encounter contradictions in the labeled data during the processing. In this way, the server can smoothly use the deep learning network to generate high-quality first remote session trajectory training data for subsequent medical service knowledge point prediction tasks.

[0108] In a possible implementation manner, step S130 includes:

[0109] Step S131, based on the first graph feature representation data, use an adaptation network to perform transformation to generate first adapted graph feature representation data. The adaptation network is used to adapt the graph feature representation data from the target medical service feature domain to the network feature domain, and the network feature domain is the medical service feature domain adapted by the medical service knowledge point prediction network.

[0110] Step S132, based on the first adapted graph feature representation data, use the basic medical service knowledge point prediction network to perform knowledge point prediction to generate predicted medical service knowledge points.

[0111] The method further includes:

[0112] Step E110, obtain the remote session trajectory data to be predicted originating from the second medical interaction scenario.

[0113] Step E120, use the first graph generation model to perform graph feature representation on the remote session trajectory data to be predicted to generate second graph feature representation data.

[0114] Step E130, based on the second graph feature representation data, use the adaptation network to perform transformation to generate second adapted graph feature representation data.

[0115] Step E140, based on the second adapted graph feature representation data, use the medical service knowledge point prediction network to perform knowledge point prediction to generate the medical service knowledge points of the remote session trajectory data to be predicted.

[0116] In this scenario, the server has trained and optimized multiple models through previous steps, including a first graph generation model (including a first interaction scenario model and a second interaction scenario model), an adaptation network, and a basic medical service knowledge point prediction network. Now, the server needs to use these models to process new remote session trajectory data to generate predicted medical service knowledge points.

[0117] The server first processes the first remote session trajectory training data originating from the first medical interaction scenario, and through a series of transformation and prediction steps, generates predicted medical service knowledge points.

[0118] Specifically, the first interaction scenario model can be used to represent the first remote session trajectory training data in graph features, generating first graph feature representation data, which are in the target medical service feature domain. Then, the adaptation network is used to transform the first graph feature representation data, adapting it from the target medical service feature domain to the network feature domain. This step is to ensure that the data format matches the requirements of the basic medical service knowledge point prediction network. The adapted data is called the first adapted graph feature representation data. Finally, the first adapted graph feature representation data is input into the basic medical service knowledge point prediction network, and the basic medical service knowledge point prediction network outputs the predicted medical service knowledge points.

[0119] Next, the server obtains the remote session trajectory data to be predicted originating from the second medical interaction scenario and repeats similar processing steps to generate the predicted medical service knowledge points for this data.

[0120] For example, the second interaction scenario model can be used to represent the remote session trajectory data to be predicted in graph features, generating second graph feature representation data. Similar to the first remote session trajectory training data, these data are also in the target medical service feature domain. Similarly, the server uses the adaptation network to transform the second graph feature representation data, adapting it from the target medical service feature domain to the network feature domain, generating the second adapted graph feature representation data. Finally, the second adapted graph feature representation data is input into the basic medical service knowledge point prediction network, and the basic medical service knowledge point prediction network outputs the predicted medical service knowledge points for the remote session trajectory data to be predicted.

[0121] Suppose the server obtains a first remote session trajectory training data on "respiratory infection management". The server first uses the first interaction scenario model to convert it into graph feature representation data, and then uses the adaptation network to adapt these data from the target medical service feature domain to the network feature domain. Finally, the basic medical service knowledge point prediction network receives the adapted data and outputs the predicted medical service knowledge point "respiratory infection management".

[0122] Subsequently, the server obtains a new remote session trajectory data from the second medical interaction scenario, which involves the regular follow-up records of a hypertensive patient. The server uses the second interaction scenario model to represent it in graph features and adapts the data to the network feature domain through the adaptation network. Finally, the basic medical service knowledge point prediction network outputs the predicted medical service knowledge point "hypertensive management", providing valuable reference information for doctors.

[0123] Thereby, it is possible to flexibly process data from different medical interaction scenarios and use the trained model to generate accurate predicted medical service knowledge points, thus improving the efficiency and quality of telemedicine services.

[0124] In a possible implementation manner, the method further includes:

[0125] Step F110, obtaining second remote session trajectory training data sourced from the first medical interaction scenario, where the second remote session trajectory training data carries third medical service knowledge point annotation data.

[0126] Step F120, based on the second remote session trajectory training data, using the first graph generation model to represent the second remote session trajectory training data in graph features, generating third graph feature representation data.

[0127] Step F130, based on the third graph feature representation data, using the basic adaptation network for transformation to generate adapted graph feature representation training data.

[0128] Step F140, based on the adapted graph feature representation training data, using the basic medical service knowledge point prediction network for knowledge point prediction to generate knowledge point prediction data.

[0129] Step F150, based on the error between the knowledge point prediction data and the third medical service knowledge point annotation data, locking the network parameter information of the basic medical service knowledge point prediction network, updating the network parameter information of the basic adaptation network, and generating the adaptation network.

[0130] In this embodiment, the server retrieves new remote session trajectory data from the database of the first medical interaction scenario. These remote session trajectory data are called second remote session trajectory training data, and each data carries third medical service knowledge point annotation data.

[0131] For example, the server extracts the second remote session trajectory training data and its corresponding third medical service knowledge point annotation data from the database through a preset data retrieval interface or SQL query. The extracted data is temporarily stored in the local storage or memory of the server for subsequent processing.

[0132] The server uses the pre-trained first graph generation model (especially the first interaction scenario model therein, as the data is sourced from the first medical interaction scenario) to perform graph feature representation on the second remote session trajectory training data, generating the third graph feature representation data.

[0133] For example, the second remote session trajectory training data can be input into the first graph generation model. Through operations such as multi-layer graph convolution, the first graph generation model extracts the key features in the second remote session trajectory training data and generates the third graph feature representation data. These second remote session trajectory training data are now in the target medical service feature domain but have not been adapted to the network feature domain required by the basic medical service knowledge point prediction network.

[0134] Next, the basic adaptation network is used to transform the third graph feature representation data, adapting it from the target medical service feature domain to the network feature domain, generating the adapted graph feature representation training data.

[0135] For example, the third graph feature representation data can be input into the basic adaptation network. The adaptation network transforms the data from the target feature domain to the network feature domain through a series of transformation operations (such as linear transformation, non-linear activation, etc.). The transformed data (adapted graph feature representation training data) now matches the input requirements of the basic medical service knowledge point prediction network.

[0136] The server uses the basic medical service knowledge point prediction network to perform knowledge point prediction on the adapted graph feature representation training data, generating knowledge point prediction data. For example, the server inputs the adapted graph feature representation training data into the basic medical service knowledge point prediction network, and through processing such as multi-layer convolution, recurrent, or attention mechanisms, extracts the features related to the medical service knowledge points and generates the prediction results. The prediction results are recorded as knowledge point prediction data, that is, the predicted medical service knowledge points.

[0137] The server evaluates the prediction accuracy by comparing the error between the knowledge point prediction data and the third medical service knowledge point annotation data, and accordingly optimizes the network parameter information of the basic adaptation network, generating the final adaptation network.

[0138] For example, the server calculates the error (such as cross-entropy loss) between the knowledge point prediction data and the third medical service knowledge point annotation data. Using the backpropagation algorithm, the server backpropagates the error into the basic adaptation network, updating its network parameter information (such as weights and biases). This process may need to be iterated multiple times until the error converges to an acceptable range. Finally, the server generates the optimized adaptation network, which can more accurately adapt the graph feature representation data from the target medical service feature domain to the network feature domain.

[0139] Thus, not only new training data is utilized to enhance the generalization ability of the model, but also the accuracy of knowledge point prediction is improved by optimizing the adaptation network, which is of great significance for enhancing the intelligent level of telemedicine services and patient experience.

[0140] In a possible implementation manner, the method further includes:

[0141] Step G110, obtaining a second guiding data sample and an adaptation request, where the second guiding data sample includes a graph feature representation data placeholder and an adaptation request placeholder, and the adaptation request represents that the basic medical service knowledge point prediction network generates the knowledge point prediction data.

[0142] Step G120, loading the adaptation graph feature representation training data into the graph feature representation data placeholder in the second guiding data sample, and loading the adaptation request into the adaptation request placeholder in the second guiding data sample to generate second guiding data.

[0143] Step F140 includes: based on the second guiding data, using the basic medical service knowledge point prediction network to perform knowledge point prediction to generate the knowledge point prediction data.

[0144] In this embodiment, the server obtains a second guiding data sample from a configuration file, a database, or a remote service, and receives an adaptation request, which clearly indicates that the server should use the basic medical service knowledge point prediction network to generate the knowledge point prediction data.

[0145] For example, the server retrieves the second guiding data sample from local storage, a database, or a remote service through a predefined interface or method call. This sample is a template for guiding the subsequent data loading and prediction processes. At the same time, the server receives an adaptation request, which may be transmitted through an API, a message queue, command line parameters, or other communication mechanisms. The adaptation request contains the instructions and parameters required to execute the prediction task.

[0146] The server loads the adapted graph feature representation training data and the adaptation request into the corresponding placeholders in the second bootstrapping data example respectively, thereby generating the complete second bootstrapping data. For example, the server first parses the second bootstrapping data example and identifies the graph feature representation data placeholder and the adaptation request placeholder therein. Then, the server loads the adapted graph feature representation training data generated through a series of processing steps into the graph feature representation data placeholder. These data are obtained after the remote session trajectory data is transformed by the graph generation model and the adaptation network, and they are in the feature domain suitable for knowledge point prediction. At the same time, the server loads the received adaptation request into the adaptation request placeholder. The adaptation request contains the specific instructions and parameters required for performing the prediction task, such as the type of target knowledge point to be predicted, the confidence threshold for prediction, etc. After the above operations, the server generates the second bootstrapping data containing the adapted graph feature representation training data and the adaptation request, and this data structurally organizes all the information required for the prediction task.

[0147] Next, the server processes the adapted graph feature representation training data in the second bootstrapping data using the basic medical service knowledge point prediction network to generate knowledge point prediction data.

[0148] For example, the server passes the second bootstrapping data as input to the basic medical service knowledge point prediction network. After receiving the input, the basic medical service knowledge point network first parses the adapted graph feature representation training data and the adaptation request in the second bootstrapping data. Then, the basic medical service knowledge point network processes the adapted graph feature representation training data according to the instructions and parameters in the adaptation request, and this processing process can include various machine learning tasks such as feature extraction, classification, regression, etc.

[0149] Finally, the basic medical service knowledge point network outputs the prediction results, that is, the knowledge point prediction data, and these data may exist in forms such as text, labels, probability distributions, or others, depending on the requirements of the prediction task and the output design of the network.

[0150] Exemplarily, assume that the server is processing a remote session trajectory data about "follow-up of heart disease patients". After the graph feature representation and adaptation network transformation, the server obtains the adapted graph feature representation training data. At this time, the server receives an adaptation request, requiring the use of a specific basic medical service knowledge point prediction network to predict the medical service knowledge points related to this session trajectory.

[0151] The server obtained the second boot data sample and the adaptation request according to the above steps, and loaded the adaptation graph feature representation training data and the adaptation request into the sample to generate the second boot data. Then, the server input the second boot data into the basic medical service knowledge point prediction network. The network processed the adaptation graph feature representation training data according to the instructions in the adaptation request, and finally output the predicted medical service knowledge point "Precautions for Drug Treatment of Heart Disease" as the knowledge point prediction data. This result helps doctors or patients better understand the content of the conversation trajectory and take corresponding management measures.

[0152] In a possible implementation manner, step S140 includes: updating the network parameter information of the basic medical service knowledge point prediction network based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data to generate a medical service knowledge point prediction network, and updating the network parameter information of the adaptation network based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data to generate an updated adaptation network.

[0153] In this embodiment, the server first calculates the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data. This error is an important indicator for measuring the difference between the prediction result and the actual annotation. For example, the server traverses all training samples. For each sample, it compares the predicted medical service knowledge point with the corresponding first medical service knowledge point annotation data. It uses an appropriate error metric method (such as cross-entropy loss, mean squared error, etc.) to calculate the difference between the two to obtain the prediction error of this sample. The prediction errors of all samples are aggregated or averaged to obtain the overall training error.

[0154] Next, the calculated prediction error is used to update the network parameter information (such as weights and biases) of the basic medical service knowledge point prediction network to optimize the prediction performance of the network. For example, the backpropagation algorithm is adopted to propagate the prediction error layer by layer from the output layer of the network to the input layer. During the backpropagation process, the gradient of each layer of network parameters (i.e., the derivative of the error with respect to the parameters) is calculated according to the chain rule. The values of each layer of network parameters are updated using the gradient descent or other optimization algorithms to reduce the prediction error.

[0155] Repeat the above process multiple times (i.e., perform multiple training iterations) until the prediction error converges to an acceptable range or reaches the preset number of training epochs.

[0156] In addition to updating the basic medical service knowledge point prediction network, the server also needs to update the network parameter information of the adaptation network according to the prediction error. This is because the adaptation network is responsible for adapting the graph feature representation data from the target medical service feature domain to the network feature domain, and its performance directly affects the subsequent knowledge point prediction results.

[0157] Similar to updating the basic medical service knowledge point prediction network, the server first calculates the contribution error of the adaptation network during the training process, which can be obtained by backpropagating the error of the basic prediction network to the adaptation network layer. Then, the same backpropagation and optimization algorithms as those for updating the basic prediction network are used to update the network parameter information of the adaptation network.

[0158] It should be noted that when updating the adaptation network, the server may need to consider the combined effects of the adaptation error and the basic prediction error simultaneously to find the optimal parameter update strategy.

[0159] After multiple iterative trainings, the server generates an optimized medical service knowledge point prediction network and an updated adaptation network, and these two networks now have better prediction performance and generalization ability.

[0160] For example, the server saves the network parameter information of the trained medical service knowledge point prediction network and the updated adaptation network, and these network parameter information can be used in subsequent prediction tasks to generate accurate medical service knowledge point prediction results. The server can also deploy the trained model to the actual production environment to provide real-time medical service knowledge point prediction services for doctors and patients.

[0161] Exemplarily, assume that the server is processing remote session trajectory data regarding "diabetes management". After a series of processing steps, the server generates the predicted medical service knowledge point "diabetes diet guidance" and compares it with the corresponding first medical service knowledge point annotation data. The calculated prediction error indicates that there is a certain difference between the prediction result and the actual annotation.

[0162] To optimize the prediction performance, the server adopts the backpropagation algorithm and the gradient descent optimization algorithm to update the network parameter information of the basic medical service knowledge point prediction network and the adaptation network. After multiple iterative trainings, the server generates an optimized medical service knowledge point prediction network and an updated adaptation network, and these two networks can now more accurately predict medical service knowledge points related to diabetes management, providing more reliable reference information for doctors and patients.

[0163] In a possible implementation manner, the method further includes:

[0164] Step H110, obtaining auxiliary medical diagnosis data of the first remote session trajectory training data.

[0165] Step H120, performing feature extraction on the auxiliary medical diagnosis data to generate auxiliary medical diagnosis features.

[0166] Step S130 includes: Based on the first graph feature representation data and the auxiliary medical diagnosis features, use the basic medical service knowledge point prediction network to perform knowledge point prediction and generate the predicted medical service knowledge points.

[0167] In this embodiment, in this scenario, the server not only uses the graph feature representation data of the first remote session trajectory training data for knowledge point prediction, but also additionally obtains auxiliary medical diagnosis data associated with the graph feature representation data. These auxiliary medical diagnosis data provide additional context information, which helps to improve the accuracy and comprehensiveness of knowledge point prediction.

[0168] For example, retrieve auxiliary medical diagnosis data associated with the first remote session trajectory training data from the database of the first medical interaction scenario. These auxiliary medical diagnosis data may include the patient's medical history, laboratory test results, imaging examination reports, etc.

[0169] Next, the server extracts features from the obtained auxiliary medical diagnosis data and converts them into a feature representation form suitable for knowledge point prediction.

[0170] For example, for text-type auxiliary medical diagnosis data (such as medical history descriptions, diagnostic reports, etc.), the server may use natural language processing techniques for word segmentation, stop word removal, word embedding, etc., and convert the text into a numerical vector or feature matrix.

[0171] For image-type auxiliary medical diagnosis data (such as X-ray films, CT scan images, etc.), the server may use image processing and computer vision techniques for feature extraction, such as edge detection, texture analysis, convolutional neural network feature extraction, etc.

[0172] For numerical-type auxiliary medical diagnosis data (such as laboratory test results, vital sign monitoring data, etc.), the server may directly use them as features, or perform preprocessing operations such as normalization and standardization.

[0173] After feature extraction, the server generates auxiliary medical diagnosis features, which will be used together with the first graph feature representation data for subsequent knowledge point prediction.

[0174] Finally, use the first graph feature representation data and the auxiliary medical diagnosis features as inputs, and use the basic medical service knowledge point prediction network to perform knowledge point prediction and generate the predicted medical service knowledge points.

[0175] For example, the first graph feature representation data and the auxiliary medical diagnosis features can be combined or spliced to form a comprehensive feature representation containing rich context information, which is input into the basic medical service knowledge point prediction network. The network processes these features through mechanisms such as multi-layer convolution, recurrence, and attention, and extracts key information related to medical service knowledge points. Finally, the network outputs a prediction result, that is, the predicted medical service knowledge point, which comprehensively considers the information of the remote session trajectory data and the auxiliary medical diagnosis data, and is therefore more accurate and comprehensive.

[0176] Exemplarily, assume that the server is processing a first remote session trajectory training data regarding "hypertension management". In addition to the session trajectory data itself, the server also retrieves from the database the auxiliary medical diagnosis data associated with this patient, including the patient's medical history record, the latest blood pressure monitoring results, and the electrocardiogram report.

[0177] The server first extracts features from the auxiliary medical diagnosis data, converts the text description into a numerical vector, converts the image report into a feature matrix, and normalizes the numerical results. Then, the server combines these auxiliary medical diagnosis features with the first graph feature representation data to form a comprehensive feature representation.

[0178] Next, the server inputs this comprehensive feature representation into the basic medical service knowledge point prediction network. The network extracts key information through a multi-layer processing mechanism and outputs a prediction result of "adjustment of hypertension drug treatment plan", which comprehensively considers the information of the session trajectory data and the auxiliary medical diagnosis data and provides valuable reference opinions for doctors.

[0179] In a possible implementation manner, the method further includes:

[0180] Obtain a third guiding data sample and an extended training request, where the third guiding data sample includes a graph feature representation data placeholder and an extended training request placeholder, and the extended training request represents that the basic medical service knowledge point prediction network generates the predicted medical service knowledge point.

[0181] Load the first graph feature representation data into the graph feature representation data placeholder in the third guiding data sample, and load the extended training request into the extended training request placeholder in the third guiding data sample to generate third guiding data.

[0182] Step S130 includes: based on the third guiding data, use the basic medical service knowledge point prediction network to perform knowledge point prediction to generate the predicted medical service knowledge point.

[0183] In this embodiment, in this scenario, the server not only uses the basic data and models for knowledge point prediction, but also introduces a third guiding data sample to further guide and optimize the prediction process. This sample contains specific placeholders for receiving graph feature representation data and extended training requests, thus ensuring that the prediction process can be extended and optimized in the desired manner.

[0184] For example, the server obtains the third guiding data sample from a configuration file, a database, or a remote service, and receives an extended training request that instructs the server to consider certain additional training conditions or constraints when performing knowledge point prediction.

[0185] Next, the first graph feature representation data and the extended training request are respectively loaded into the corresponding placeholders in the third guiding data sample to generate the complete third guiding data.

[0186] For example, first, the third guiding data sample is parsed to identify the graph feature representation data placeholder and the extended training request placeholder therein. Then, the server loads the first graph feature representation data obtained through a series of processing steps into the graph feature representation data placeholder. These data are obtained by converting the remote session trace data through a graph generation model and contain rich session information. At the same time, the server loads the received extended training request into the extended training request placeholder, and this request may specify certain specific prediction targets, constraint conditions, or optimization strategies. After the above operations, the server generates the third guiding data containing the first graph feature representation data and the extended training request, which structurally organizes all the information and instructions required for the prediction task.

[0187] Finally, the basic medical service knowledge point prediction network is used to process the first graph feature representation data in the third guiding data, and considering the instructions and conditions in the extended training request, the predicted medical service knowledge points are generated.

[0188] For example, the third guiding data can be passed as input to the basic medical service knowledge point prediction network. After receiving the input, the network first parses the first graph feature representation data and the extended training request in the third guiding data. The network adjusts its prediction strategy according to the instructions and conditions in the extended training request, which can include changing the prediction target, applying specific optimization algorithms, adjusting model parameters, etc. Then, the network processes the first graph feature representation data, extracts the features related to the medical service knowledge points, and generates a prediction result based on these features and the conditions in the extended training request. Finally, the network outputs the predicted medical service knowledge points, and this result is not only based on the graph feature representation of the remote session trace data, but also fully considers the additional information and conditions in the extended training request.

[0189] Exemplarily, assume that the server is processing the first remote session trajectory data regarding "Follow-up of heart disease patients". After being transformed by the graph generation model, the server obtains the first graph feature representation data. At this time, the server receives an extended training request, requiring special attention to the aspect of "Drug treatment effect evaluation" when making knowledge point predictions.

[0190] The server obtains the third guiding data sample and the extended training request according to the above steps, and loads the first graph feature representation data and the extended training request into the sample to generate the third guiding data. Then, the server inputs this third guiding data into the basic medical service knowledge point prediction network. The network pays special attention to the features related to the drug treatment effect during the processing, and adjusts the prediction strategy according to the conditions in the extended training request. Finally, the network outputs the predicted medical service knowledge point "Evaluation and adjustment suggestions for the drug treatment effect of heart disease", and this result is more in line with the specific requirements in the extended training request.

[0191] In a possible implementation manner, the third guiding data sample further includes an auxiliary medical diagnosis data placeholder, and the method further includes:

[0192] Obtain the auxiliary medical diagnosis data of the first remote session trajectory training data.

[0193] Perform feature representation on the auxiliary medical diagnosis data to generate auxiliary medical diagnosis features.

[0194] The step of loading the first graph feature representation data into the graph feature representation data placeholder in the third guiding data sample, loading the extended training request into the extended training request placeholder in the third guiding data sample, and generating the third guiding data includes:

[0195] Load the first graph feature representation data into the graph feature representation data placeholder in the third guiding data sample, load the extended training request into the extended training request placeholder in the third guiding data sample, and load the auxiliary medical diagnosis features into the auxiliary medical diagnosis data placeholder in the third guiding data sample to generate the third guiding data.

[0196] In this embodiment, in this scenario, when the server processes the first remote session trajectory training data, it not only considers its graph feature representation data, but also additionally obtains the auxiliary medical diagnosis data associated with this session trajectory. To more effectively utilize this information, the server introduces a third guiding data sample containing multiple placeholders, which are respectively used to receive the graph feature representation data, the extended training request, and the auxiliary medical diagnosis features.

[0197] For example, the server retrieves auxiliary medical diagnosis data corresponding to the first remote session trajectory training data from the database of the first medical interaction scenario, which may include the patient's medical records, laboratory test results, imaging examination reports, etc.

[0198] Then, the server performs feature extraction and transformation on the obtained auxiliary medical diagnosis data to generate auxiliary medical diagnosis features suitable for knowledge point prediction.

[0199] For example, for text-type auxiliary medical diagnosis data (such as medical records, diagnosis reports, etc.), the server uses natural language processing techniques to perform operations such as word segmentation, stop word removal, and word embedding, and converts the text into a numerical vector or feature matrix.

[0200] For image-type auxiliary medical diagnosis data (such as X-ray films, CT scan images, etc.), the server uses image processing and computer vision techniques to extract features and generate image feature vectors.

[0201] For numerical-type auxiliary medical diagnosis data (such as laboratory test results, vital sign monitoring data, etc.), the server may directly use it as a feature or perform preprocessing operations such as normalization and standardization.

[0202] After the above processing, the server generates auxiliary medical diagnosis features, which will be used in the subsequent knowledge point prediction process.

[0203] Finally, the server loads the first graph feature representation data, the extended training request, and the auxiliary medical diagnosis features into the corresponding placeholders in the third bootstrap data sample respectively to generate the complete third bootstrap data.

[0204] For example, the server first parses the third bootstrap data sample to identify the graph feature representation data placeholder, the extended training request placeholder, and the auxiliary medical diagnosis data placeholder. Then, the server loads the first graph feature representation data into the graph feature representation data placeholder. These data are obtained after the remote session trajectory data is transformed by the graph generation model and contain the graph structure information of the session. At the same time, the server loads the extended training request into the extended training request placeholder, which may specify additional training conditions, optimization objectives, or constraint conditions to guide the knowledge point prediction process. Finally, the server loads the auxiliary medical diagnosis features into the auxiliary medical diagnosis data placeholder. These features provide additional information about the patient's health status and help improve the accuracy and comprehensiveness of the knowledge point prediction. After the above operations, the server generates the third bootstrap data containing all the necessary information, which structurally organizes all the data and instructions required for the prediction task.

[0205] Exemplarily, assume that the server is processing the first remote session trajectory training data regarding "diabetes management". In addition to the session trajectory data itself, the server also retrieves the auxiliary medical diagnosis data associated with this patient from the database, including the latest blood glucose monitoring results, glycated hemoglobin levels, and the doctor's diagnosis opinions.

[0206] The server first performs feature representation on the auxiliary medical diagnosis data, converts the text-type diagnosis opinions into numerical vectors, and normalizes the numerical-type blood glucose monitoring results and glycated hemoglobin levels. Then, the server loads these auxiliary medical diagnosis features together with the first graph feature representation data and the extended training request (for example, requiring special attention to the prediction of "blood glucose control strategy") into the third guiding data sample to generate the third guiding data.

[0207] Finally, the server processes the third guiding data using the basic medical service knowledge point prediction network, comprehensively considering the graph structure information of the session trajectory, the specific requirements in the extended training request, and the patient's health condition information in the auxiliary medical diagnosis data, and generates the predicted medical service knowledge point "personalized diabetes diet and exercise plan", providing valuable reference opinions for doctors and patients.

[0208] Figure 2 FIG. shows the hardware structure diagram of the medical Internet of Things service system 100 provided by the embodiments of the present application for implementing the above-mentioned medical sign data monitoring method based on the Internet of Things, as Figure 2 shown, the medical Internet of Things service system 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0209] In a possible design, the medical Internet of Things service system 100 may be a single server or a server group. The server group may be centralized or distributed (for example, the medical Internet of Things service system 100 may be a distributed system). In some embodiments, the medical Internet of Things service system 100 may be local or remote. For example, the medical Internet of Things service system 100 may access the information and / or data stored in the machine-readable storage medium 120 via a network. Also, for example, the medical Internet of Things service system 100 may be directly connected to the machine-readable storage medium 120 to access the stored information and / or data. In some embodiments, the medical Internet of Things service system 100 may be implemented on the medical Internet of Things service system. Only by way of example, the medical Internet of Things service system may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc. or any aggregation thereof.

[0210] The machine-readable storage medium 120 may store data and / or instructions. In some embodiments, the machine-readable storage medium 120 may store data obtained from an external terminal. In some embodiments, the machine-readable storage medium 120 may store the data and / or instructions that the medical Internet of Things service system 100 uses to execute or utilize to complete the exemplary methods described in this application.

[0211] In a specific implementation process, one or more processors 110 execute the computer-executable instructions stored in the machine-readable storage medium 120, so that the processors 110 can execute the medical IoT-based vital sign data monitoring method in the above method embodiments. The processors 110, the machine-readable storage medium 120, and the communication unit 140 are connected through the bus 130, and the processors 110 can be used to control the transceiver actions of the communication unit 140.

[0212] For the specific implementation process of the processors 110, reference may be made to the respective method embodiments executed by the medical Internet of Things service system 100 above. Their implementation principles and technical effects are similar, and will not be elaborated here in this embodiment.

[0213] In addition, an embodiment of the present application further provides a readable storage medium, in which computer-executable instructions are set. When a processor executes the computer-executable instructions, the medical IoT-based vital sign data monitoring method as described above is implemented.

[0214] It should be noted that, in order to simplify the presentation of the disclosure of the present application and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present application, sometimes multiple features are merged into one embodiment, drawing, or description thereof. Similarly, it should be noted that, in order to simplify the presentation of the disclosure of the present application and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present application, sometimes multiple features are merged into one embodiment, drawing, or description thereof.

Claims

1. A method for monitoring vital sign data based on medical Internet of Things, characterized in that: The method comprises: Acquire first remote conversation trajectory training data originating from a first medical interaction scenario, wherein the first remote conversation trajectory training data carries first medical service knowledge point annotation data, and the first medical service knowledge point annotation data represents actual medical service knowledge points of the first remote conversation trajectory training data; Using a first graph generation model to perform graph feature representation on the first remote conversation trajectory training data to generate first graph feature representation data, the first graph generation model is used to convert the remote conversation trajectory data originating from the first medical interaction scenario into a target medical service feature domain, the target medical service feature domain is a medical service feature domain where the graph feature representation data corresponding to the remote conversation trajectory data originating from the second medical interaction scenario is located, the first medical interaction scenario and the second medical interaction scenario are different medical interaction scenarios, and the collection density of the remote conversation trajectory data originating from the first medical interaction scenario is less than the collection density of the remote conversation trajectory data originating from the second medical interaction scenario; Based on the first graph feature representation data, using the basic medical service knowledge point prediction network to predict knowledge points and generate predicted medical service knowledge points; Based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data, the network parameter information of the basic medical service knowledge point prediction network is updated to generate a medical service knowledge point prediction network, so that the medical service knowledge point prediction network is used to predict the medical service knowledge points of the remote conversation trajectory data originating from the second medical interaction scenario, and then generates a corresponding vital sign data monitoring plan based on the medical service knowledge points of the remote conversation trajectory data originating from the second medical interaction scenario.

2. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: The method further comprises: Acquire remote conversation trajectory data to be predicted from the second medical interaction scenario; Performing graph feature representation on the remote conversation trajectory data to be predicted by using a second graph generation model to generate second graph feature representation data, wherein the second graph generation model is used to convert the remote conversation trajectory data originating from the second medical interaction scenario into the target medical service feature domain; Based on the second graph feature representation data, the medical service knowledge point prediction network is used to perform knowledge point prediction to generate the medical service knowledge points of the remote conversation trajectory data to be predicted.

3. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: The first graph generation model includes a first interaction scenario model and a second interaction scenario model, the first interaction scenario model is used to convert the remote conversation trajectory data originated from the first medical interaction scenario into the target medical service feature domain, and the second interaction scenario model is used to convert the remote conversation trajectory data originated from the second medical interaction scenario into the target medical service feature domain; The using the first graph generation model to perform graph feature representation on the first remote conversation trajectory training data to generate first graph feature representation data includes: Performing graph feature representation on the first remote conversation trajectory training data using the first interaction scenario model to generate first graph feature representation data; The method further comprises: Acquire remote conversation trajectory data to be predicted from the second medical interaction scenario; Using the second interaction scenario model to perform graph feature representation on the remote conversation trajectory data to be predicted, to generate second graph feature representation data; Based on the second graph feature representation data, using the medical service knowledge point prediction network to perform knowledge point prediction to generate the medical service knowledge points of the remote conversation trajectory data to be predicted; Wherein, the method comprises: Acquire multiple remote conversation trajectory training data combinations, wherein the remote conversation trajectory training data combination includes first member remote conversation trajectory training data originating from the first medical interaction scenario and second member remote conversation trajectory training data originating from the second medical interaction scenario, and the first member remote conversation trajectory training data and the second member remote conversation trajectory training data originating from the same remote conversation trajectory training data combination are the same medical service data; For a target remote conversation trajectory training data combination among the multiple remote conversation trajectory training data combinations, using a basic first interaction scenario model to perform graph feature representation on first member remote conversation trajectory training data in the target remote conversation trajectory training data combination to generate first member graph feature representation data; Using the basic second interaction scenario model, performing graph feature representation on the second member remote conversation trajectory training data in the target remote conversation trajectory training data combination to generate second member graph feature representation data; Using the plurality of remote conversation trajectory training data combinations as the target remote conversation trajectory training data combinations, respectively, to generate a plurality of first member graph feature representation data and a plurality of second member graph feature representation data; According to the network optimization goal of minimizing the feature distance of the target member graph feature representation data combination and maximizing the feature distance of other member graph feature representation data combinations, the network parameter information of the basic first interaction scenario model and the network parameter information of the basic second interaction scenario model are updated to generate the first interaction scenario model and the second interaction scenario model, the first member graph feature representation data and the second member graph feature representation data covered by the target member graph feature representation data combination are generated based on the same remote conversation trajectory training data combination, and the first member graph feature representation data and the second member graph feature representation data covered by the other member graph feature representation data combinations are not generated based on the same remote conversation trajectory training data combination.

4. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: If the remote conversation trajectory data originating from the first medical interaction scenario is sparse feature data, and the remote conversation trajectory data originating from the second medical interaction scenario is dense feature data, the method further includes: Obtain x second medical service knowledge point annotation data and first guide data samples including knowledge point placeholders, wherein the second medical service knowledge point annotation data is used to describe the medical service knowledge points of the remote conversation trajectory data originating from the second medical interaction scenario, and the first guide data samples represent that a deep learning network generates a plurality of dense feature semantic data, and the plurality of medical image feature fragments described by the plurality of dense feature semantic data constitute target dense feature data related to the medical service knowledge point annotation data corresponding to the knowledge point placeholders, and x is a positive integer; Loading the x second medical service knowledge point annotation data into the knowledge point placeholder in the first guide data sample to generate first guide data; Based on the first guide data, using the deep learning network to generate the first remote conversation trajectory training data, where the first remote conversation trajectory training data is dense feature semantic data related to the x second medical service knowledge point annotation data; The x second medical service knowledge point annotation data are output as the first medical service knowledge point annotation data of the first remote conversation trajectory training data.

5. The method for monitoring vital sign data based on medical Internet of Things according to claim 4, characterized in that: If the first guide data sample further represents that the deep learning network generates other medical service knowledge point annotation data related to the target dense feature data, then outputting the x second medical service knowledge point annotation data as first medical service knowledge point annotation data of the first remote conversation trajectory training data includes: Outputting the x second medical service knowledge point annotation data and the other medical service knowledge point annotation data as first medical service knowledge point annotation data of the first remote conversation trajectory training data; and The first guide data sample further represents that when the labeled data of the x second medical service knowledge points are inconsistent, the deep learning network intercepts and generates first remote conversation trajectory training data related to the target dense feature data, and the method further includes: The x second medical service knowledge point annotation data are extracted from y second medical service knowledge point annotation data, where y is an integer greater than x.

6. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: The step of predicting knowledge points based on the first graph feature representation data and using a basic medical service knowledge point prediction network to generate predicted medical service knowledge points includes: Based on the first graph feature representation data, transforming using an adaptation network to generate first adapted graph feature representation data, wherein the adaptation network is used to adapt the graph feature representation data from the target medical service feature domain to a network feature domain, wherein the network feature domain is a medical service feature domain adapted by the medical service knowledge point prediction network; Based on the first adaptation graph feature representation data, using the basic medical service knowledge point prediction network to perform knowledge point prediction to generate predicted medical service knowledge points; The method further comprises: Acquire remote conversation trajectory data to be predicted from the second medical interaction scenario; Using the first graph generation model to perform graph feature representation on the remote conversation trajectory data to be predicted, to generate second graph feature representation data; Based on the second graph feature representation data, transform using the adaptation network to generate second adapted graph feature representation data; Based on the second adaptation graph feature representation data, the medical service knowledge point prediction network is used to perform knowledge point prediction to generate the medical service knowledge points of the remote conversation trajectory data to be predicted.

7. The method for monitoring vital sign data based on medical Internet of Things according to claim 6, characterized in that: The method further comprises: Acquire second remote conversation trajectory training data originating from the first medical interaction scenario, wherein the second remote conversation trajectory training data carries third medical service knowledge point annotation data; Based on the second remote conversation trajectory training data, using the first graph generation model to perform graph feature representation on the second remote conversation trajectory training data to generate third graph feature representation data; Based on the third graph feature representation data, transforming using a basic adaptation network to generate adapted graph feature representation training data; Based on the adaptation graph feature representation training data, using the basic medical service knowledge point prediction network to perform knowledge point prediction and generate knowledge point prediction data; Based on the error between the knowledge point prediction data and the third medical service knowledge point annotation data, locking the network parameter information of the basic medical service knowledge point prediction network, updating the network parameter information of the basic adaptation network, and generating the adaptation network; Wherein, the method further comprises: Acquire a second guide data sample and an adaptation request, wherein the second guide data sample includes a graph feature representation data placeholder and an adaptation request placeholder, and the adaptation request represents that the basic medical service knowledge point prediction network generates the knowledge point prediction data; Loading the adaptation graph feature representation training data into a graph feature representation data placeholder in the second guide data sample, and loading the adaptation request into an adaptation request placeholder in the second guide data sample, to generate second guide data; The method of representing the training data based on the adaptation graph features and using the basic medical service knowledge point prediction network to perform knowledge point prediction and generate knowledge point prediction data includes: Based on the second guidance data, using the basic medical service knowledge point prediction network to perform knowledge point prediction to generate the knowledge point prediction data; The updating of network parameter information of the basic medical service knowledge point prediction network based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data to generate a medical service knowledge point prediction network includes: Based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data, the network parameter information of the basic medical service knowledge point prediction network is updated to generate a medical service knowledge point prediction network; and based on the error between the predicted medical service knowledge point and the first medical service knowledge point annotation data, the network parameter information of the adaptation network is updated to generate an updated adaptation network.

8. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: The method further comprises: Acquire auxiliary medical diagnosis data of the first remote conversation trajectory training data; Extracting features from the auxiliary medical diagnosis data to generate auxiliary medical diagnosis features; The step of predicting knowledge points based on the first graph feature representation data and using a basic medical service knowledge point prediction network to generate predicted medical service knowledge points includes: Based on the first graph feature representation data and the auxiliary medical diagnosis features, the basic medical service knowledge point prediction network is used to perform knowledge point prediction to generate the predicted medical service knowledge point.

9. The method for monitoring vital sign data based on medical Internet of Things according to claim 1, characterized in that: The method further comprises: Acquire a third guide data sample and an extended training request, wherein the third guide data sample includes a graph feature representation data placeholder and an extended training request placeholder, and the extended training request represents that the basic medical service knowledge point prediction network generates the predicted medical service knowledge point; Loading the first graph feature representation data into a graph feature representation data placeholder in the third guide data sample, and loading the extended training request into an extended training request placeholder in the third guide data sample, to generate third guide data; The step of predicting knowledge points based on the first graph feature representation data and using a basic medical service knowledge point prediction network to generate predicted medical service knowledge points includes: Based on the third guidance data, using the basic medical service knowledge point prediction network to perform knowledge point prediction and generate predicted medical service knowledge points; The third guide data sample further includes an auxiliary medical diagnosis data placeholder, and the method further includes: Acquire auxiliary medical diagnosis data of the first remote conversation trajectory training data; Performing feature representation on the auxiliary medical diagnosis data to generate auxiliary medical diagnosis features; The step of loading the first graph feature representation data into a graph feature representation data placeholder in the third guide data sample, and loading the extended training request into an extended training request placeholder in the third guide data sample, to generate third guide data, comprises: The first graph feature representation data is loaded into the graph feature representation data placeholder in the third guide data sample, the extended training request is loaded into the extended training request placeholder in the third guide data sample, and the auxiliary medical diagnosis feature is loaded into the auxiliary medical diagnosis data placeholder in the third guide data sample to generate third guide data.

10. A medical Internet of Things service system, characterized in that: The medical Internet of Things service system includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the vital sign data monitoring method based on the medical Internet of Things as described in any one of claims 1 to 9.

Images