Out-of-hospital follow-up visit management method and device, storage medium and electronic equipment

Abstract

The invention provides an out-of-hospital follow-up visit management method and device, a storage medium and electronic equipment, an initial follow-up visit plan path is generated by a pre-trained follow-up visit large model according to in-hospital treatment data of a patient, when the initial follow-up visit plan path is executed, health feature data of the patient is received and analyzed in real time, the initial follow-up visit plan path is adjusted and optimized, and the follow-up visit management efficiency is improved. And generating a personalized follow-up visit plan path according to the follow-up visit health data, processing the follow-up visit health data after each follow-up visit plan in the follow-up visit plan path is completed, and generating detailed follow-up visit notes for the patient. According to the method, the hospital treatment data and the pre-trained follow-up visit large model are combined to generate and adjust the follow-up visit plan, so that the workload of medical personnel during manual configuration of the follow-up visit plan can be remarkably reduced, the follow-up visit plan path is dynamically optimized in real time, personalized follow-up visit service is provided for patients, and the overall efficiency of out-of-hospital follow-up visit management is improved.

Description

Technical Field

[0001] The present invention relates to the field of big data processing technology, and in particular to an out-of-hospital follow-up management method, device, storage medium and electronic equipment. Background Art

[0002] At present, the progress of information technology has also brought about the innovation of out-of-hospital follow-up management. Medical staff can now use special follow-up management tools to effectively follow up patients out of the hospital through electronic communication means such as telephone or text messages.

[0003] In the traditional follow-up management process, medical staff need to manually set up follow-up plans, including detailed time, frequency, and content, etc. This process is often time-consuming and complicated. Taking the RESCUER (Registry for Improving the Quality of Cardiovascular Disease Management) follow-up project as an example, medical staff not only need to edit the follow-up plan in advance, but also need to deal with cumbersome forms and data element configuration, which undoubtedly increases their workload and affects the overall efficiency of outpatient follow-up management.

[0004] Therefore, how to optimize the outpatient follow-up management process and ensure the quality of outpatient follow-up services while reducing the workload of medical staff is the key to achieving efficient outpatient follow-up management. Summary of the invention

[0005] In view of the above problems, the present invention provides an out-of-hospital follow-up management method, device, storage medium and electronic device that overcome the above problems or at least partially solve the above problems. The technical solution is as follows:

[0006] An out-of-hospital follow-up management method comprises:

[0007] Obtain the patient's in-hospital visit data;

[0008] Processing the in-hospital visit data using a pre-trained follow-up large model to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule;

[0009] In the process of executing the initial follow-up plan path, obtaining the health characteristic data of the patient in real time;

[0010] Inputting the health characteristic data into the follow-up macromodel to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule;

[0011] When completing the initial follow-up plan path or any of the personalized follow-up plan paths, obtain follow-up health data associated with the follow-up plan, process the follow-up health data using the follow-up big model, and generate follow-up summary information of the patient under the follow-up plan.

[0012] Optionally, obtaining the patient's health characteristic data in real time during the execution of the initial follow-up plan path includes:

[0013] During the execution of the initial follow-up plan path, monitoring the patient's original health data in real time;

[0014] The original health data is preprocessed to obtain health characteristic data of the patient.

[0015] Optionally, the health characteristic data includes the patient's basic information, physical sign data, diet and exercise data, medication data, follow-up task data and risk assessment data.

[0016] Optionally, the step of inputting the health characteristic data into the follow-up macromodel to adjust the initial follow-up plan path and generate a personalized follow-up plan path includes:

[0017] Inputting the health characteristic data into the follow-up large model to obtain personalized follow-up adjustment suggestions output by the follow-up large model;

[0018] In response to an agreement operation on the personalized follow-up adjustment suggestion, the initial follow-up plan path is adjusted to a personalized follow-up plan path according to the personalized follow-up adjustment suggestion.

[0019] Optionally, the in-hospital medical treatment data include the patient's basic information, medical history, physical signs, auxiliary examination results, admission diagnosis, treatment process, surgical records, discharge diagnosis, discharge instructions and medical record front page.

[0020] Optionally, the training process of the follow-up large model includes:

[0021] Obtain historical follow-up medical records for multiple different types of diseases;

[0022] Standardizing each of the historical follow-up medical records to obtain standard follow-up medical records;

[0023] Using each of the standard follow-up medical records and their corresponding follow-up path simulation data, a standard follow-up medical record data set is constructed;

[0024] The standard follow-up medical record data set is used to perform machine learning on the follow-up large model: based on the comparison of input and output information of each round of the follow-up large model, the loss function of the follow-up large model is used to calculate the loss value of the follow-up large model, and the model parameters of the follow-up large model are adjusted using the loss value until the follow-up large model meets the predetermined training end conditions.

[0025] Optionally, the follow-up model is composed of a follow-up management prediction sub-model, a disease warning sub-model, a follow-up path sub-model, a form task sub-model, an entrustment task sub-model, a patient task sub-model and a health summary sub-model, and each sub-model follows the model parallel logic and feature data parallel logic.

[0026] An out-of-hospital follow-up management device comprises: an in-hospital medical treatment data acquisition unit, an initial follow-up plan path generation unit, a health characteristic data acquisition unit, a personalized follow-up plan path generation unit and a follow-up summary information generation unit.

[0027] The in-hospital medical treatment data obtaining unit is used to obtain the in-hospital medical treatment data of the patient;

[0028] The initial follow-up plan path generating unit is used to process the in-hospital visit data using a pre-trained follow-up large model to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule;

[0029] The health characteristic data obtaining unit is used to obtain the health characteristic data of the patient in real time during the execution of the initial follow-up plan path;

[0030] The personalized follow-up plan path generation unit is used to input the health feature data into the follow-up large model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule;

[0031] The follow-up summary information generating unit is used to obtain the follow-up health data associated with the follow-up plan when completing the initial follow-up plan path or any of the personalized follow-up plan paths, process the follow-up health data using the follow-up big model, and generate the follow-up summary information of the patient under the follow-up plan.

[0032] A computer-readable storage medium stores a program, wherein when the program is executed by a processor, the out-of-hospital follow-up management method described in any one of the above items is implemented.

[0033] An electronic device, comprising at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other via the bus; and the processor is used to call program instructions in the memory to execute any of the above-mentioned outpatient follow-up management methods.

[0034] By means of the above technical scheme, the present invention provides an out-of-hospital follow-up management method, device, storage medium and electronic device, which obtain the patient's in-hospital medical treatment data; use a pre-trained follow-up big model to process the in-hospital medical treatment data to generate the patient's initial follow-up plan path, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to the first follow-up plan schedule; in the process of executing the initial follow-up plan path, the patient's health characteristic data is obtained in real time; the health characteristic data is input into the follow-up big model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule; when any follow-up plan in the initial follow-up plan path or the personalized follow-up plan path is completed, the follow-up health data associated with the follow-up plan is obtained, the follow-up health data is processed by the follow-up big model, and the follow-up summary information of the patient under the follow-up plan is generated. The present invention combines in-hospital medical treatment data with a pre-trained follow-up large model to generate and adjust the follow-up plan, which can significantly reduce the workload of medical staff in manually configuring the follow-up plan, dynamically optimize the follow-up plan path in real time, and provide patients with personalized follow-up services, thereby improving the overall efficiency of out-of-hospital follow-up management.

[0035] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0037] Figure 1 A flow chart showing an implementation of an out-of-hospital follow-up management method provided by an embodiment of the present invention;

[0038] Figure 2 The overall framework schematic diagram of the outpatient follow-up management system based on the large model method provided by an embodiment of the present invention is shown;

[0039] Figure 3 A flow chart of the configuration of a follow-up plan path provided by an embodiment of the present invention is shown;

[0040] Figure 4 A flowchart of adjusting a follow-up visit plan path provided by an embodiment of the present invention is shown;

[0041] Figure 5 FIG. 4 shows a schematic diagram of the training process of the large follow-up model provided by an embodiment of the present invention;

[0042] Figure 6 A flowchart of the training process of the large follow-up model provided by an embodiment of the present invention is shown;

[0043] Figure 7 A schematic diagram showing the structure of an out-of-hospital follow-up management device provided by an embodiment of the present invention is shown;

[0044] Figure 8 A schematic structural diagram of an electronic device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0045] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0046] In the modern medical field, with the advancement of technology and the development of artificial intelligence, the medical health AI (Artificial Intelligence) big model has become one of the key technologies to promote the improvement of medical service quality. This model can generate new data instances by analyzing and learning a large amount of medical data, thus playing an important role in scenarios such as outpatient follow-up management.

[0047] At present, medical staff can follow up patients outside the hospital through follow-up management tools, and to a certain extent, realize the information management of follow-up outside the hospital. However, with the increase in the number of follow-up patients, the workload of medical staff has also increased. In addition, the large amount of accumulated follow-up data contains the potential to improve the decision-making support capabilities of follow-up management. If these data can be effectively collected, analyzed and deeply mined, it can not only provide more powerful information support for medical staff, but also further improve the automation and intelligence level of the outpatient follow-up management system. This can not only improve the overall efficiency of medical services, but also provide patients with more accurate and personalized follow-up services.

[0048] The traditional outpatient follow-up management process usually relies on medical staff to manually configure the follow-up plan path and content, which is not only time-consuming but also complicated to operate. For example, in the RESCUER (Registry for Improving the Quality of Cardiovascular Disease Management) follow-up project, medical staff need to edit and configure the follow-up plan in advance, including setting the follow-up time, frequency, content, etc. The form and data element configuration work involved is particularly cumbersome. In addition, the follow-up process also requires real-time monitoring of the patient's condition, manual adjustment of the follow-up path or addition of new follow-up tasks, which further increases the workload of medical staff.

[0049] However, although the existing outpatient follow-up management system includes modules such as information collection, knowledge base, plan configuration and data statistics, there are still some key problems. First, the process of manually configuring follow-up paths and content is time-consuming and complicated to operate, especially when managing multiple diseases. This problem is more prominent. Secondly, this configuration method requires high medical knowledge and system operation capabilities of medical staff, resulting in increased learning costs and increased error rates. Finally, the current system is not intelligent enough and cannot automatically adjust the follow-up path or add new follow-up tasks according to the personalized development of the patient's condition.

[0050] Therefore, with the continuous development of AI big models in medical health, using this technology to optimize the existing follow-up management system can not only reduce the workload of medical staff, but also improve the efficiency and quality of follow-up management. Through the application of AI big models, the automatic configuration and dynamic adjustment of follow-up plans can be realized, so as to better adapt to changes in patients' conditions, thereby promoting the intelligence and personalization of medical services.

[0051] Based on this, in an embodiment of the present invention, a method for follow-up management outside the hospital is provided. First, the patient's in-hospital medical treatment data is collected, and the pre-trained follow-up large model is used to analyze these data to develop an initial follow-up plan path. Secondly, in the process of implementing the initial follow-up plan path, the patient's real-time health feature data is continuously collected, and these real-time data are fed back to the follow-up large model to adjust and optimize the initial follow-up plan path and form a personalized follow-up plan path. At the same time, after each follow-up plan is completed, relevant health data is collected, and the follow-up large model is used to process these data to generate a detailed follow-up summary for the patient. It can be seen that the present invention ensures the dynamic optimization and personalization of the follow-up plan through real-time data feedback and follow-up plan path adjustment, improves the pertinence and effect of the follow-up service, and thus improves the overall efficiency of follow-up management outside the hospital.

[0052] like Figure 1 As shown, a flow chart of an implementation of an out-of-hospital follow-up management method provided by an embodiment of the present invention is provided, and the method may include:

[0053] S100. Obtain the patient's in-hospital medical treatment data.

[0054] Among them, in-hospital medical data refers to all relevant data and information generated when patients receive medical services in hospitals. Optionally, in-hospital medical data includes the patient's basic information, medical history, physical signs, auxiliary examination results, admission diagnosis, treatment process, surgical records, discharge diagnosis, discharge instructions and medical record homepage.

[0055] Optionally, the embodiment of the present invention can identify the patient's in-hospital medical treatment data by scanning the patient's medical records, and can also obtain the patient's in-hospital medical treatment data from the hospital's out-of-hospital follow-up management system.

[0056] The outpatient follow-up management system provided by the embodiment of the present invention may be integrated with a follow-up information collection module, which is responsible for collecting the patient's health data from multiple sources, including in-hospital medical data, data entered by patients or medical staff, and data collected by Internet of Things devices. These raw data are classified and stored in a database for further processing and analysis.

[0057] S110. Use a pre-trained follow-up large model to process the in-hospital medical treatment data to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule.

[0058] The follow-up big model is a machine learning model built using deep learning technology. The embodiment of the present invention can pre-train the follow-up big model so that the follow-up big model can effectively process a large amount of data sets and perform complex analysis tasks to achieve a high level of performance and accuracy.

[0059] Among them, the follow-up plan path is a continuous and structured medical tracking program designed for patients, which aims to systematically monitor the patient's health status and treatment effects and intervene when necessary. The follow-up plan path usually arranges one or more follow-up plans according to the schedule and frequency in the follow-up plan schedule. Each follow-up plan can include one or more follow-up tasks. Optionally, the follow-up task can be specifically a form task, an entrustment task, or a patient education task.

[0060] Optionally, the embodiment of the present invention may set a corresponding window period and follow-up reminder for each follow-up plan in the follow-up plan path, so as to remind relevant medical staff to complete the follow-up tasks under the follow-up plan.

[0061] Optionally, the embodiment of the present invention may limit the number of follow-up plans set in the follow-up plan path to ensure the patient's service experience of outpatient follow-up. For example, the embodiment of the present invention may limit the number of follow-up plans in the follow-up plan path to no more than 10.

[0062] Specifically, the follow-up model can identify disease type information from the patient's in-hospital medical data, generate a follow-up plan path based on the disease type information, and then determine the specific follow-up plan in the follow-up plan path based on other data information in the in-hospital medical data to ensure that the follow-up plan accurately matches the patient's basic condition.

[0063] S120. In the process of executing the initial follow-up plan path, the patient's health characteristic data is obtained in real time.

[0064] Among them, the health characteristic data is the characteristic data extracted based on the patient's health data and can be mapped to the follow-up large model. Optionally, the health characteristic data may include the patient's basic information, physical sign data, diet and exercise data, medication data, follow-up task data and risk assessment data.

[0065] Specifically, the embodiment of the present invention can obtain the patient's health characteristic data in real time from the out-of-hospital follow-up management system.

[0066] The outpatient follow-up management system provided by the embodiment of the present invention can be integrated with a real-time health monitoring module, which retrieves the health data in the follow-up information collection module in real time through an API (Application Programming Interface), and performs preprocessing including data classification, format unification and noise reduction, so that the processed health feature data can be mapped to a large model for more accurate subsequent analysis.

[0067] S130. Input the health characteristic data into the follow-up big model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule.

[0068] Specifically, the embodiment of the present invention inputs the health characteristic data into the follow-up big model, and the follow-up big model uses the health characteristic data to analyze the patient's condition and provide personalized follow-up suggestions to medical staff, including plan path adjustment and new follow-up plans, as well as related follow-up tasks.

[0069] The outpatient follow-up management system provided by the embodiment of the present invention can be integrated with a follow-up management module, and the health feature data is input into the follow-up big model by using feature loading and query analysis technology. The follow-up big model calculates the health feature data through parallel processing technology, and then generates decision-making content including disease analysis, follow-up management prediction, disease warning, follow-up path decision, form task decision, entrustment task decision, patient education task decision and health summary decision through prediction result caching technology and real-time generation technology, which is called by the follow-up management module, so as to adjust the initial follow-up plan path and generate a personalized follow-up plan path.

[0070] Among them, condition analysis refers to locating the patient's current follow-up path and follow-up plan by checking the time, frequency, parameter values, etc. of the patient's health characteristic data, and analyzing the reasons for changes in the patient's condition.

[0071] Among them, follow-up management prediction refers to the prediction of patient follow-up management based on the dynamic changes of multiple health characteristic parameters, such as the risk of complications, trend changes of certain indicators, prediction of disease outcomes, etc.

[0072] Among them, disease warning refers to the dynamic changes of multiple health characteristic parameters. According to the abnormal threshold logic, the risk level is divided into no obvious change, improvement, low risk, medium risk, and high risk. If the abnormality is at the low risk level or above, a message notification will be pushed to the user terminal in time, and the patient will be advised to go to the hospital immediately if necessary.

[0073] Among them, follow-up path decision refers to providing decision suggestions such as changing the follow-up plan path and adding a new follow-up plan. The plan includes forms, instructions, and patient education content. The follow-up plan schedule generated according to the follow-up path scenario is used to call the follow-up message access system, and the follow-up plan can be pushed to the patient and / or medical terminal.

[0074] Among them, form task decision refers to providing a form composed of a structured feature data set. The data element in the form is the smallest unit of the data set. The data element has standard form type and data type attributes and can be returned to the hospital data center so that medical staff can accurately grasp the recovery status and disease prognosis of discharged patients. In addition, it can support follow-up path decision calls.

[0075] Among them, the entrusted task decision includes medical text content such as humanized medical appointment reminders, medication guidance, nursing guidance, lifestyle guidance, etc. In addition, it can support follow-up path decision calls.

[0076] Among them, patient education task decisions include medical education content in the form of pictures, texts or videos, such as medication and disease knowledge, diet, exercise or rehabilitation exercises, psychological condition, quality of life, etc. In addition, it can support follow-up path decision calls.

[0077] Among them, the health summary decision refers to the generation of a patient follow-up health summary by the follow-up model based on the completion of the follow-up task and the changes in the patient's condition. It includes the patient's basic information, health history (personal history, allergy history, surgery, etc.), medication and examination and test information, key observation indicators and changing trends, disease development warnings and predictions, and other health information, which makes it easier for medical staff to quickly view the key content of the follow-up and improve the efficiency of the follow-up.

[0078] The embodiment of the present invention can regularly notify patients or medical staff through the out-of-hospital follow-up management system when the window period of any follow-up plan is reached, and can obtain the follow-up health data of the patient entered and updated by the patient or medical staff.

[0079] The outpatient follow-up management system provided by the embodiment of the present invention may be integrated with a message contact module, which sends notifications to medical staff, management personnel, patients and / or patients' families when the window period of any follow-up plan is reached, including management plan reminders, health record reminders, diet and exercise reminders, health report reminders, and risk warning reminders, etc. The notification method may include SMS, phone calls, and application notifications.

[0080] The outpatient follow-up management system provided by the embodiment of the present invention can also be integrated with a data server module and a terminal system, wherein the data server module provides data storage, management, access control, and backup and recovery functions to ensure the security and availability of data. The terminal system includes various user interfaces, such as APP, website, official account, and applets, which facilitate patients and medical staff to record follow-up conditions and access the follow-up management system, as well as to obtain knowledge related to the follow-up plan path. Through the collaborative work of multiple modules, the outpatient follow-up management system can provide patients with personalized health management services while improving the efficiency and quality of medical services.

[0081] S140. When any follow-up plan in the initial follow-up plan path or the personalized follow-up plan path is completed, the follow-up health data associated with the follow-up plan is obtained, and the follow-up health data is processed using the follow-up big model to generate the follow-up summary information of the patient under the follow-up plan.

[0082] Among them, follow-up health data includes the completion of the follow-up plan and various health information tracked after regular examinations or treatment of patients in the follow-up plan. Follow-up health data can be entered by medical staff or patients at specific time points in the follow-up plan.

[0083] It should be noted that, whether it is a follow-up plan in the initial follow-up plan path or a follow-up plan in the personalized follow-up plan path, after the follow-up plan is completed, the embodiment of the present invention can use the follow-up large model to generate the patient's follow-up summary information under the follow-up plan, so that the patient or medical staff can timely evaluate the treatment effect and monitor changes in the disease, which is beneficial for the doctor to adjust or formulate subsequent treatment plans according to the patient's current specific situation, and provide patients with more personalized medical services.

[0084] The embodiment of the present invention can input the follow-up health data into the follow-up big model, and the follow-up big model generates detailed follow-up summary information to improve the follow-up efficiency of medical staff. Optionally, the follow-up summary information can include the patient's basic information, health history, medication and examination information, key indicator trends and disease warnings.

[0085] It can be understood that the embodiment of the present invention can determine the end of follow-up after the last follow-up plan in the personalized follow-up plan path is completed, or it can use the follow-up large model to determine the patient's condition and then use the current condition to terminate the follow-up service of the follow-up plan path in advance.

[0086] The present invention provides an out-of-hospital follow-up management method, which obtains the patient's in-hospital medical treatment data; uses a pre-trained follow-up big model to process the in-hospital medical treatment data to generate the patient's initial follow-up plan path, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule; in the process of executing the initial follow-up plan path, the patient's health characteristic data is obtained in real time; the health characteristic data is input into the follow-up big model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to a second follow-up plan schedule; when any follow-up plan in the initial follow-up plan path or the personalized follow-up plan path is completed, the follow-up health data associated with the follow-up plan is obtained, the follow-up health data is processed by the follow-up big model, and the follow-up summary information of the patient under the follow-up plan is generated. The present invention combines in-hospital medical treatment data with a pre-trained follow-up large model to generate and adjust the follow-up plan, which can significantly reduce the workload of medical staff in manually configuring the follow-up plan, dynamically optimize the follow-up plan path in real time, and provide patients with personalized follow-up services, thereby improving the overall efficiency of out-of-hospital follow-up management.

[0087] In order to enhance the decision-making support capability of follow-up management and improve the automation and intelligence of the out-of-hospital follow-up management system, the embodiment of the present invention combines the follow-up big model with the out-of-hospital follow-up management system to provide an out-of-hospital follow-up management system based on the big model method. The overall framework is as follows: Figure 2As shown in the figure, the outpatient follow-up management system based on the big model method mainly includes the outpatient follow-up management system and the follow-up big model system. In the preparation stage of follow-up management, the standardized follow-up plan for specific diseases can be designed in advance by using historical follow-up medical record data and big data analysis technology. This plan provides a basic follow-up path and lays the foundation for subsequent personalized management. During the follow-up process, the system will analyze the health data submitted by the patient in real time and optimize the follow-up plan regularly based on this information. In this way, the system can generate personalized follow-up paths, including specific follow-up content and suggestions. These contents may include filled-in forms, medical advice and health education materials. In addition, the system will send relevant information to patients and medical staff regularly according to the set follow-up schedule to collect follow-up data. In the later stage of follow-up management, the system will generate a follow-up health summary for the patient based on the collected data. This summary makes it easy for medical staff to quickly understand the patient's follow-up status and changes in health status, and make corresponding medical decisions when necessary until the entire process of follow-up management is completed. This full-process follow-up management model not only improves the efficiency and accuracy of follow-up, but also provides patients with more personalized and continuous medical care.

[0088] Among them, the follow-up model can be composed of a follow-up management prediction sub-model, a disease warning sub-model, a follow-up path sub-model, a form task sub-model, an entrustment task sub-model, a patient task sub-model and a health summary sub-model. Each sub-model follows the model parallel logic and feature data parallel logic.

[0089] The follow-up management prediction sub-model is responsible for predicting and managing different stages of the follow-up process.

[0090] The disease warning sub-model is used to detect potential health problems of patients at an early stage and issue warnings.

[0091] The follow-up pathway sub-model is used to create a personalized follow-up plan for patients and guide the direction and focus of follow-up.

[0092] The Form Task sub-sub-model is used to automatically generate and process the forms and documents required for follow-up.

[0093] The entrustment task sub-model is used to plan tasks and reminders based on the doctor's instructions and the patient's condition.

[0094] The patient task sub-model is used to design specific tasks for patients, such as medication management, lifestyle adjustments, etc.

[0095] The health summary model is used to provide a summary report of the patient's health status after each follow-up visit.

[0096] The above sub-models work together according to the logic of model parallelism and feature data parallelism. The system receives the input of various feature parameters and applies advanced logic processing technology and algorithm operations to achieve in-depth analysis, prediction and early warning of the patient's condition. Based on these analysis results, the system can intelligently generate customized follow-up paths and follow-up content for patients, and provide a detailed health summary after the follow-up, thereby providing medical staff with a powerful auxiliary tool to make follow-up work more accurate and efficient, while also ensuring that patients receive continuous and personalized medical care.

[0097] The embodiment of the present invention realizes the real-time and seamless connection of patient data in and outside the hospital environment through the outpatient follow-up management system based on the big model method, providing a solid data foundation for post-hospital follow-up work. In addition, through the efficient big model system, the initial follow-up plan path is set in advance for discharged patients according to the characteristics of different diseases, ensuring the personalization and accuracy of the follow-up work.

[0098] In order to facilitate the understanding of the configuration process of the follow-up plan path of the outpatient follow-up management system based on the large model method, Figure 3 Provide explanation. Figure 3 The following is a flow chart of the configuration of the follow-up plan path. After the patient is discharged from the hospital, the standard data set of his / her in-hospital visit is automatically collected, including basic information, medical history, physical signs, auxiliary examination results, etc., and the patient is followed up in groups through information comparison. Based on the collected data, the follow-up model customizes personalized follow-up paths and content for each patient, and generates a corresponding follow-up plan schedule to ensure that the follow-up work is accurately matched with the patient's condition. During the execution of the follow-up plan, the patient's health data, including basic information, physical signs, diet, exercise, medication, etc., is monitored in real time, and the data is preprocessed. The follow-up model uses the preprocessed data to analyze the patient's condition and provide personalized follow-up suggestions to medical staff, including plan path adjustments and new follow-up plans, as well as related follow-up tasks such as forms, instructions, and patient education. Follow-up tasks regularly notify patients and medical staff through the message system to support them in entering and updating patients' follow-up health data. According to the completion of the follow-up tasks and the changes in the patient's condition, the follow-up model generates a detailed follow-up health summary, including basic information, health history, medication and examination information, key indicator trends and condition warnings, to improve the follow-up efficiency of medical staff. The follow-up plan is completed or the follow-up model determines the end of the follow-up based on the patient's condition.

[0099] Optional, in the above Figure 1 On the basis of one or more corresponding embodiments, in another optional embodiment provided by the embodiment of the present invention, in the process of executing the initial follow-up plan path, obtaining the patient's health characteristic data in real time may specifically include:

[0100] During the execution of the initial follow-up plan path, the patient's original health data is monitored in real time. The original health data is preprocessed to obtain the patient's health characteristic data.

[0101] Raw health data refers to physiological and biochemical information collected directly from patients without processing or analysis, which is used to reflect the patient's current health status. Optionally, raw health data may include the patient's physiological parameters, biochemical indicators, medical images, biomarkers, and activity behavior data.

[0102] Preprocessing refers to converting raw health data into a format suitable for analysis or modeling. The preprocessing provided by the embodiment of the present invention may include one or more processing methods including data classification processing, format unification processing, data cleaning processing and noise reduction processing.

[0103] Specifically, in the data classification process of the embodiment of the present invention, the raw health data can be classified according to the data type and health characteristics as the classification basis for subsequent processing. Optionally, the data type can include numerical type, text type and time series type. Health characteristics can include physiological indicators, medical history and drug use.

[0104] Specifically, in the process of format unification processing, the embodiments of the present invention may use the same unit and format for data of the same type in the original health data. For example, the unit of blood glucose level is unified as mg / dL or mmol / L). The date and time of the original health data may also be unified using a specific time standard format. For example: the date and time of the data are unified using the ISO8601 standard format (YYYY-MM-DD). The original health data may also be encoded using a unified coding system. For example: the disease is encoded using the International Classification of Diseases (ICD).

[0105] Specifically, the embodiments of the present invention can process missing values ​​and outliers in the original health data during data cleaning. For example, the embodiments of the present invention can delete samples with more missing data in the original health data, or use the mean, median or other statistical methods to fill in the missing values. Use statistical methods to identify and correct outliers.

[0106] Specifically, in the noise reduction process of the embodiment of the present invention, the original health data can be denoised by using moving average, exponential smoothing or interpolation to improve the data quality and signal-to-noise ratio, thereby making subsequent data analysis or modeling more accurate and effective.

[0107] In actual situations, the patient's health status may change at any time, and the patient or medical staff can enter or update the patient's original health data in the outpatient follow-up management system. In the implementation of the initial follow-up plan path related to the patient, the embodiment of the present invention monitors the patient's original health data in real time, obtains the patient's latest health data at regular intervals and performs preprocessing to obtain health feature data that can be mapped on the follow-up large model, so as to dynamically adjust the patient's follow-up plan and follow-up tasks based on the patient's real-time health status, and provide a personalized follow-up plan path, thereby improving the overall efficiency of outpatient follow-up management and providing patients with better follow-up services.

[0108] Optional, in the above Figure 1 On the basis of one or more corresponding embodiments, in another optional embodiment provided by the embodiment of the present invention, the health feature data is input into the follow-up big model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, which may specifically include:

[0109] The health characteristic data is input into the follow-up big model to obtain the personalized follow-up adjustment suggestion output by the follow-up big model. In response to the consent operation of the personalized follow-up adjustment suggestion, the initial follow-up plan path is adjusted to the personalized follow-up plan path according to the personalized follow-up adjustment suggestion.

[0110] Among them, personalized follow-up adjustment suggestions can be decision-making content output after performing condition analysis, follow-up management prediction, condition warning, follow-up path decision, form task decision, entrustment task decision, patient education task decision and / or health summary decision on health characteristic data based on the follow-up big model, and the output adjustment suggestion information of the follow-up plan path.

[0111] The embodiment of the present invention can present personalized follow-up adjustment suggestions to medical staff or managers, and provide corresponding option buttons, and the medical staff and / or managers can click the relevant option buttons related to the personalized follow-up adjustment suggestions to agree or reject the personalized follow-up adjustment suggestions. After receiving the instruction to agree to the personalized follow-up adjustment suggestions, the follow-up plan and / or follow-up tasks in the initial follow-up plan path are adjusted according to the personalized follow-up adjustment suggestions, and a reasonable window period and follow-up summary generation node are set for each follow-up plan.

[0112] It should be noted that the term "in response to" in the present invention is used to indicate the conditions or states on which the execution of an operation depends, and one or more operations can be executed when certain conditions or states are met. These operations can be real-time or have a certain delay.

[0113] In order to facilitate the understanding of the adjustment process of the follow-up plan path, Figure 4 Provide explanation. Figure 4The flowchart of the adjustment process of the follow-up plan path provided by the embodiment of the present invention is shown. Before the follow-up management is started, the initial follow-up plan path for a specific disease is pre-set using the follow-up large model. This path mainly includes the follow-up plan and follow-up tasks with set follow-up dates and times, and also covers other settings, such as window periods, follow-up reminders, etc. Each follow-up task may include forms, medical advice and patient education content. In order to optimize the user experience, the number of each task may not exceed 10 by default. During the follow-up management process, the follow-up large model processes and analyzes the patient's health data in real time, and intelligently adds follow-up tasks or adjusts the follow-up path for the patient based on the data results, thereby generating personalized follow-up paths, content and suggestions, and sends this information to patients and / or medical staff according to the set follow-up schedule through the message access module to collect follow-up data. After any follow-up plan in the follow-up management is completed, a follow-up health summary will be generated for the patient. The follow-up health summary will be automatically updated when the window period of the follow-up plan ends or the patient's condition changes, so that medical staff can easily check it until the entire follow-up management process is completed.

[0114] The embodiment of the present invention inputs health characteristic data into the follow-up model, and operates according to the personalized follow-up adjustment suggestions output by the model, automatically adjusting the follow-up plan path so that the adjusted follow-up plan path matches the patient's actual health status, thereby helping to provide patients with personalized follow-up services and improving the overall efficiency of outpatient follow-up management.

[0115] like Figure 5 As shown, the training process of the follow-up large model provided by the embodiment of the present invention may include:

[0116] S500. Obtain historical follow-up medical records of multiple different types of diseases.

[0117] Among them, historical follow-up medical records refer to medical documents recorded during regular or irregular follow-up observation and evaluation of patients. Historical follow-up medical records usually record in detail all relevant medical information of patients from the first visit to subsequent follow-up visits.

[0118] Specifically, the embodiment of the present invention can collect historical follow-up medical records of various different types of diseases, and the historical follow-up medical records can be follow-up medical records of common chronic diseases or follow-up medical records of rare diseases. The embodiment of the present invention collects historical follow-up medical records of a wide range of diseases to ensure the generalization and applicability of the subsequent follow-up large model.

[0119] Optionally, an embodiment of the present invention can screen out high-quality historical follow-up medical records from multiple historical follow-up medical records. Specifically, an embodiment of the present invention can screen out high-quality historical follow-up medical records from multiple historical follow-up medical records based on a pre-constructed dynamic screening strategy, wherein the dynamic screening strategy includes multiple screening dimensions. Optionally, the screening dimensions may include follow-up compliance rate, medication compliance rate, diet and exercise compliance rate, and disease outcome. By screening out high-quality historical follow-up medical records to participate in the training of the follow-up large model, the embodiment of the present invention can reduce the noise in subsequent model training, improve the quality of the training corpus, and thus enhance the prediction accuracy of the follow-up large model.

[0120] S510. Standardize each historical follow-up medical record to obtain a standard follow-up medical record.

[0121] Among them, standardization processing can include unifying data formats, cleaning missing or erroneous data, unifying medical terminology and classification methods, and removing patients' privacy-sensitive information.

[0122] The embodiment of the present invention can obtain standard follow-up medical records with higher data quality by standardizing each historical follow-up medical record, thereby providing high-quality input data for subsequent model training.

[0123] S520. Use each standard follow-up medical record and its corresponding follow-up path simulation data to construct a standard follow-up medical record data set.

[0124] Among them, the follow-up path simulation data includes the follow-up plan path actually adopted in the historical follow-up medical records, as well as the follow-up date, follow-up plan, follow-up task, follow-up summary and other data in the follow-up plan path.

[0125] Specifically, the embodiment of the present invention can use each standard follow-up medical record and its corresponding follow-up path simulation data as a set of training corpora to construct a standard follow-up medical record data set. The standard follow-up medical record data set includes multiple sets of training corpora. The expected input and output results contained in the training corpora help the follow-up large model understand the relationship between different follow-up medical records and follow-up plan paths.

[0126] S530. Use the standard follow-up medical record data set to perform machine learning on the follow-up large model: based on the information comparison between the input and output of each round of the follow-up large model, use the loss function of the follow-up large model to calculate the loss value of the follow-up large model, and use the loss value to adjust the model parameters of the follow-up large model until the follow-up large model meets the predetermined training end conditions.

[0127] Specifically, the embodiment of the present invention uses a well-constructed standard follow-up medical record data set to apply a machine learning algorithm to the follow-up large model. In each round of training, the follow-up large model will give a follow-up plan prediction path based on the input standard follow-up medical record, and compare the follow-up plan prediction path with the actual path of the follow-up plan corresponding to the standard follow-up medical record to obtain a comparison difference. The comparison difference is then used to calculate the loss function, that is, the accuracy index of the follow-up large model. The loss value calculated by the loss function of the follow-up large model is then used to adjust the model parameters (for example, weights and bias) of the follow-up large model to improve the prediction performance of the follow-up large model. By repeating multiple rounds of training, the performance of the follow-up large model meets the predetermined training end condition. Among them, the training end condition can be that the follow-up large model reaches a preset accuracy or has been trained for a specified number of rounds. The embodiment of the present invention trains the follow-up large model based on the classification of disease scenarios, so that it can intelligently adjust the path and content of the follow-up plan according to the specific condition and needs of the patient, thereby helping doctors to monitor the patient's recovery process more accurately.

[0128] The embodiment of the present invention collects historical follow-up medical records of various different types of diseases and performs standardized processing to provide a high-quality standard follow-up medical record data set for the training of the follow-up large model, thereby improving the accuracy and reliability of the follow-up large model. During the model training process, the loss value calculated by the loss function is used to continuously adjust the parameters of the follow-up large model, which can ensure that the model performance is continuously optimized and a high-performance follow-up large model is constructed, thereby improving the overall efficiency of outpatient follow-up management.

[0129] In order to facilitate the understanding of the training process of the follow-up large model, Figure 6 Provide explanation. Figure 6The following is a flowchart of the training process of the follow-up large model provided by an embodiment of the present invention, including follow-up medical record screening, standard follow-up medical record, follow-up corpus construction, follow-up large model fine-tuning, model evaluation and model application. In the follow-up medical record screening stage, high-quality follow-up medical records are first screened out from the historical disease follow-up data, using a dynamic screening strategy, mainly referring to factors such as follow-up compliance rate, medication compliance rate, diet and exercise compliance rate, and course of disease outcome. In the standard follow-up medical record stage, the screened medical records are further processed, and noise and abnormal data are removed according to dynamic data acquisition rules to ensure the quality of the selected medical record data so as to serve as input data for the follow-up large model. In the corpus construction stage, a training corpus is constructed using more than 10,000 standard follow-up medical record data sets. The loss value of the model is determined by comparing the input and output information of each round. In the follow-up large model fine-tuning stage, the follow-up large model is fine-tuned according to the loss value of the model, and training continues until the end condition is met, and an efficient follow-up large model is finally formed. The embodiment of the present invention trains the follow-up big model so that the follow-up big model can accurately provide personalized follow-up plans and contents according to the patient's specific condition and treatment response, thereby improving the follow-up effect and patient satisfaction.

[0130] Although operations are depicted in a particular order, this should not be understood as requiring that the operations be performed in the particular order shown or in a sequential order.Multitasking and parallel processing may be advantageous under certain circumstances.

[0131] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0132] Corresponding to the above method embodiment, the embodiment of the present invention also provides an outpatient follow-up management device, the structure of which is as follows: Figure 7 As shown, it may include: an in-hospital medical treatment data obtaining unit 10, an initial follow-up plan path generating unit 20, a health characteristic data obtaining unit 30, a personalized follow-up plan path generating unit 40 and a follow-up summary information generating unit 50.

[0133] The in-hospital medical treatment data obtaining unit 10 is used to obtain the in-hospital medical treatment data of the patient.

[0134] The initial follow-up plan path generation unit 20 is used to process the in-hospital medical treatment data using a pre-trained follow-up large model to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to the first follow-up plan schedule.

[0135] The health characteristic data obtaining unit 30 is used to obtain the patient's health characteristic data in real time during the execution of the initial follow-up plan path.

[0136] The personalized follow-up plan path generation unit 40 is used to input the health feature data into the follow-up large model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule.

[0137] The follow-up summary information generating unit 50 is used to obtain the follow-up health data associated with the follow-up plan when any follow-up plan in the initial follow-up plan path or the personalized follow-up plan path is completed, and to process the follow-up health data using the follow-up big model to generate the follow-up summary information of the patient under the follow-up plan.

[0138] Optionally, the health characteristic data obtaining unit 30 may be specifically used to monitor the original health data of the patient in real time during the execution of the initial follow-up plan path, and pre-process the original health data to obtain the health characteristic data of the patient.

[0139] Optionally, the health characteristic data includes the patient's basic information, physical sign data, diet and exercise data, medication data, follow-up task data and risk assessment data.

[0140] Optionally, the personalized follow-up plan path generation unit 40 can be specifically used to input health feature data into the follow-up big model to obtain personalized follow-up adjustment suggestions output by the follow-up big model; in response to the consent operation for the personalized follow-up adjustment suggestions, the initial follow-up plan path is adjusted to a personalized follow-up plan path according to the personalized follow-up adjustment suggestions.

[0141] Optionally, in-hospital medical data include the patient's basic information, medical history, physical signs, auxiliary examination results, admission diagnosis, treatment process, surgical records, discharge diagnosis, discharge instructions and medical record homepage.

[0142] Optionally, the out-of-hospital follow-up management device may also include: a historical follow-up medical record acquisition unit, a standard follow-up medical record acquisition unit, a standard follow-up medical record data set construction unit and a model training unit.

[0143] The historical follow-up medical record acquisition unit is used to acquire historical follow-up medical records of multiple different types of diseases.

[0144] The standard follow-up medical record acquisition unit is used to standardize various historical follow-up medical records and obtain standard follow-up medical records.

[0145] The standard follow-up medical record data set construction unit is used to construct a standard follow-up medical record data set by using each standard follow-up medical record and its corresponding follow-up path simulation data.

[0146] The model training unit is used to perform machine learning on the follow-up large model using the standard follow-up medical record data set: based on the information comparison between the input and output of each round of the follow-up large model, the loss function of the follow-up large model is used to calculate the loss value of the follow-up large model, and the model parameters of the follow-up large model are adjusted using the loss value until the follow-up large model meets the predetermined training end conditions.

[0147] The present invention provides an out-of-hospital follow-up management device, which obtains the patient's in-hospital medical treatment data; uses a pre-trained follow-up big model to process the in-hospital medical treatment data to generate the patient's initial follow-up plan path, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule; in the process of executing the initial follow-up plan path, the patient's health characteristic data is obtained in real time; the health characteristic data is input into the follow-up big model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to a second follow-up plan schedule; when any follow-up plan in the initial follow-up plan path or the personalized follow-up plan path is completed, the follow-up health data associated with the follow-up plan is obtained, the follow-up health data is processed by the follow-up big model, and the follow-up summary information of the patient under the follow-up plan is generated. The present invention combines in-hospital medical treatment data with a pre-trained follow-up large model to generate and adjust the follow-up plan, which can significantly reduce the workload of medical staff in manually configuring the follow-up plan, dynamically optimize the follow-up plan path in real time, and provide patients with personalized follow-up services, thereby improving the overall efficiency of out-of-hospital follow-up management.

[0148] Regarding the device in the above embodiment, the specific manner in which each unit performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0149] The out-of-hospital follow-up management device includes a processor and a memory. The in-hospital medical data acquisition unit 10, the initial follow-up plan path generation unit 20, the health characteristic data acquisition unit 30, the personalized follow-up plan path generation unit 40 and the follow-up summary information generation unit 50 are all stored in the memory as program units, and the processor executes the above program units stored in the memory to realize the corresponding functions.

[0150] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the kernel parameters are adjusted to generate the initial follow-up plan path based on the patient's in-hospital medical data by the pre-trained follow-up large model. When executing the initial follow-up plan path, the patient's health characteristic data is received and analyzed in real time, the initial follow-up plan path is adjusted and optimized, and a personalized follow-up plan path is generated. After each follow-up plan in the follow-up plan path is completed, the follow-up health data is processed to generate a detailed follow-up summary for the patient.

[0151] An embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the out-of-hospital follow-up management method is implemented.

[0152] An embodiment of the present invention provides a processor, which is used to run a program, wherein the out-of-hospital follow-up management method is executed when the program is run.

[0153] like Figure 8 As shown, an embodiment of the present invention provides an electronic device 1000, which includes at least one processor 1001, at least one memory 1002 connected to the processor 1001, and a bus 1003; wherein the processor 1001 and the memory 1002 communicate with each other through the bus 1003; the processor 1001 is used to call the program instructions in the memory 1002 to execute the above-mentioned outpatient follow-up management method. The electronic device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0154] The present invention also provides a computer program product, which, when executed on an electronic device, is suitable for executing a program that initializes the steps of the out-of-hospital follow-up management method.

[0155] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatuses, electronic devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable device to generate a machine, so that the instructions executed by the processor of the computer or other programmable device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0156] In a typical configuration, an electronic device includes one or more processors (CPU), a memory, and a bus. The electronic device may also include an input / output interface, a network interface, and the like.

[0157] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.

[0158] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0159] It should be noted that the patient information (including but not limited to patient device information, patient personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the patient or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0160] In the description of the present invention, it should be understood that the terms "up", "down", "front", "back", "left" and "right" etc. indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the positions or elements referred to must have specific directions, be constructed and operate in specific directions. Therefore, they should not be understood as limitations of the present invention.

[0161] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0162] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0163] The above are only embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for outpatient follow-up management, characterized in that: include: Obtain the patient's in-hospital visit data; Processing the in-hospital visit data using a pre-trained follow-up large model to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule; In the process of executing the initial follow-up plan path, obtaining the health characteristic data of the patient in real time; Inputting the health characteristic data into the follow-up macromodel to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule; When completing the initial follow-up plan path or any of the personalized follow-up plan paths, obtain follow-up health data associated with the follow-up plan, process the follow-up health data using the follow-up big model, and generate follow-up summary information of the patient under the follow-up plan.

2. The method according to claim 1, characterized in that The step of obtaining the patient's health characteristic data in real time during the execution of the initial follow-up plan path includes: During the execution of the initial follow-up plan path, monitoring the patient's original health data in real time; The original health data is preprocessed to obtain health characteristic data of the patient.

3. The method according to claim 1, characterized in that: The health characteristic data include the patient's basic information, physical sign data, diet and exercise data, medication data, follow-up task data and risk assessment data.

4. The method according to claim 1, characterized in that The step of inputting the health characteristic data into the follow-up model to adjust the initial follow-up plan path and generate a personalized follow-up plan path includes: Inputting the health characteristic data into the follow-up large model to obtain personalized follow-up adjustment suggestions output by the follow-up large model; In response to an agreement operation on the personalized follow-up adjustment suggestion, the initial follow-up plan path is adjusted to a personalized follow-up plan path according to the personalized follow-up adjustment suggestion.

5. The method according to claim 1, characterized in that The in-hospital medical treatment data includes the patient's basic information, medical history, physical signs, auxiliary examination results, admission diagnosis, treatment process, operation records, discharge diagnosis, discharge instructions and medical record homepage.

6. The method according to claim 1, characterized in that The training process of the follow-up large model includes: Obtain historical follow-up medical records for multiple different types of diseases; Standardizing each of the historical follow-up medical records to obtain standard follow-up medical records; Using each of the standard follow-up medical records and their corresponding follow-up path simulation data, a standard follow-up medical record data set is constructed; The standard follow-up medical record data set is used to perform machine learning on the follow-up large model: based on the comparison of input and output information of each round of the follow-up large model, the loss function of the follow-up large model is used to calculate the loss value of the follow-up large model, and the model parameters of the follow-up large model are adjusted using the loss value until the follow-up large model meets the predetermined training end conditions.

7. The method according to any one of claims 1 to 6, characterized in that The follow-up model is composed of a follow-up management prediction sub-model, a disease warning sub-model, a follow-up path sub-model, a form task sub-model, an entrustment task sub-model, a patient task sub-model and a health summary sub-model. Each sub-model follows the model parallel logic and the feature data parallel logic.

8. An outpatient follow-up management device, characterized in that: include: In-hospital visit data acquisition unit, initial follow-up plan path generation unit, health characteristic data acquisition unit, personalized follow-up plan path generation unit and follow-up summary information generation unit, The in-hospital medical treatment data obtaining unit is used to obtain the in-hospital medical treatment data of the patient; The initial follow-up plan path generating unit is used to process the in-hospital visit data using a pre-trained follow-up large model to generate an initial follow-up plan path for the patient, wherein the initial follow-up plan path includes at least one follow-up plan arranged according to a first follow-up plan schedule; The health characteristic data obtaining unit is used to obtain the health characteristic data of the patient in real time during the execution of the initial follow-up plan path; The personalized follow-up plan path generation unit is used to input the health feature data into the follow-up large model to adjust the initial follow-up plan path and generate a personalized follow-up plan path, wherein the personalized follow-up plan path includes at least one follow-up plan arranged according to the second follow-up plan schedule; The follow-up summary information generating unit is used to obtain the follow-up health data associated with the follow-up plan when completing the initial follow-up plan path or any of the personalized follow-up plan paths, process the follow-up health data using the follow-up big model, and generate the follow-up summary information of the patient under the follow-up plan.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the out-of-hospital follow-up management method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The electronic device includes at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other through the bus; the processor is used to call program instructions in the memory to execute the out-of-hospital follow-up management method according to any one of claims 1 to 7.

Images