Method, device, equipment, product and storage medium for generating resource scheduling task

Abstract

The application discloses a resource scheduling task generation method, device, equipment, product and storage medium. The method comprises the following steps: acquiring a target medical management scene corresponding to a medical rehabilitation state of a target object, and index data for monitoring the health state of the target object; inputting the index data into a follow-up demand assessment model corresponding to the target medical management scene to obtain an initial resource scheduling level corresponding to the target object; adjusting the initial resource scheduling level according to the deviation degree between the index data and the corresponding index threshold to obtain a target resource scheduling level; and generating a target resource scheduling task for follow-up resource scheduling of the target object according to the target resource scheduling level, the target medical management scene and the index data. The scheme provided in the application embodiment can specify a resource scheduling task for the target object, and reduces the waste of medical resources caused by invalid follow-up.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to a method, apparatus, equipment, product and storage medium for generating resource scheduling tasks. Background Technology

[0002] With social development and changes in lifestyle, the incidence of chronic diseases has increased. The core of chronic disease management lies in long-term, continuous monitoring and intervention to control disease progression and prevent complications.

[0003] In existing technologies, medical staff typically visit chronic disease patients regularly to understand their condition and recovery progress. However, current follow-up models use a uniform, fixed follow-up frequency regardless of the patient's condition, such as follow-ups or re-examinations one month, three months, and six months after discharge. In this model, high-risk patients may experience delays in diagnosis due to excessively long follow-up intervals, while low-risk patients may waste medical resources due to unnecessary and frequent follow-ups.

[0004] Therefore, it is of great significance to improve the flexibility of follow-up for patients with chronic diseases, thereby reducing the waste of medical resources. Summary of the Invention

[0005] This application provides a method, apparatus, device, product, and storage medium for generating resource scheduling tasks, which can specify resource scheduling tasks for target objects in a targeted manner, reducing the waste of medical resources caused by ineffective follow-up visits.

[0006] In a first aspect, embodiments of this application provide a method for generating resource scheduling tasks. The method includes: acquiring a target medical management scenario corresponding to the medical rehabilitation status of a target object, and indicator data for monitoring the health status of the target object; inputting the indicator data into a follow-up demand assessment model corresponding to the target medical management scenario to obtain an initial resource scheduling level corresponding to the target object; adjusting the initial resource scheduling level according to the deviation between the indicator data and the corresponding indicator threshold to obtain a target resource scheduling level; and generating a target resource scheduling task for follow-up resource scheduling of the target object based on the target resource scheduling level, the target medical management scenario, and the indicator data.

[0007] Secondly, embodiments of this application provide a resource scheduling task generation apparatus, which includes: a data acquisition module, used to acquire a target medical management scenario corresponding to the medical rehabilitation status of a target object, and indicator data for monitoring the health status of the target object under the target medical management scenario; a level assessment module, used to input the indicator data into a follow-up demand assessment model corresponding to the target medical management scenario to obtain an initial resource scheduling level corresponding to the target object; a level adjustment module, used to adjust the initial resource scheduling level according to the deviation between the indicator data and the corresponding indicator threshold to obtain a target resource scheduling level; and a task generation module, used to generate a target resource scheduling task for follow-up resource scheduling of the target object based on the target resource scheduling level, the target medical management scenario, and the indicator data.

[0008] Thirdly, embodiments of this application provide an electronic device, which includes: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the resource scheduling task generation method as described in the first aspect.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the resource scheduling task generation method as described in the first aspect.

[0010] Fifthly, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, cause the electronic device to perform the resource scheduling task generation method as described in the first aspect.

[0011] As described above, in this embodiment, the target medical management scenario is determined based on the target object's medical rehabilitation status. A follow-up needs assessment model adapted to this scenario is used to determine the initial resource allocation level for the target object. This initial resource allocation level dynamically changes based on the target object's medical rehabilitation status and indicator data. Consequently, the resource allocation tasks determined based on this initial resource allocation level are also compatible with the target object. Furthermore, the initial resource allocation level is adjusted based on the deviation between the indicator data and the corresponding indicator threshold. This ensures that the resource allocation level is no longer fixed to the initial value output by the follow-up needs assessment model but dynamically changes according to the deviation of the target object's actual indicators from the normal range. This allows follow-up resources to be accurately allocated to the target object that truly needs attention, reducing the waste of medical resources. Furthermore, the target resource scheduling task is generated based on the target resource scheduling level, medical management scenario, and indicator data to correspond to the actual status of the target object. This results in the final output target resource scheduling task being dynamically calculated based on three dimensions: scenario type, follow-up requirement level, and real-time indicator data, rather than a fixed node on a unified calendar. This allows for targeted allocation of resource scheduling tasks to the target object, reducing the waste of medical resources caused by ineffective follow-ups.

[0012] Therefore, it can be seen that the solution provided in this application can specify resource scheduling tasks for the target object in a targeted manner, reducing the waste of medical resources caused by ineffective follow-up. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart illustrating a method for generating resource scheduling tasks according to an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a resource scheduling task generation device provided in another embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in another embodiment of this application. Detailed Implementation

[0015] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0016] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0017] To facilitate understanding, before explaining the solution provided in this application, the background of the solution provided in this application will be explained first.

[0018] Currently, chronic diseases, such as coronary heart disease (CHD), are leading causes of death and disability worldwide, and post-hospital secondary prevention is crucial for improving long-term patient outcomes. However, current clinical practice regarding post-hospital follow-up management still has many shortcomings, hindering further improvements in patient management effectiveness.

[0019] First, existing follow-up models typically employ uniform, fixed time intervals, such as follow-up examinations at 1 month, 3 months, and 6 months after discharge, regardless of the patient's risk level. Under this model, high-risk patients may experience delays in recognizing changes in their condition due to excessively long follow-up intervals, while low-risk patients may waste medical resources due to unnecessary and frequent follow-ups. Therefore, existing follow-up models lack the ability to dynamically adjust the follow-up frequency based on the individual patient's condition.

[0020] Secondly, widely used clinical risk scoring models are primarily designed for risk assessment at the point of hospitalization, making them static snapshot tools. Taking the GRACE risk scoring model as an example, it is mainly used for risk assessment during hospitalization, but lacks a systematic technical solution for transforming it into continuous dynamic monitoring after discharge. Specifically, existing technologies have not yet solved problems such as how to classify model variables into dynamically updatable and fixed variables, how to establish mapping rules between each updatable variable and various external data sources, and how to conduct systematic fault-tolerant assessments when data collection is incomplete.

[0021] Furthermore, most existing mobile health mHealth intervention solutions only manage single risk factors, lacking a comprehensive decision-making framework that integrates baseline risk stratification with the control status of multiple risk factors into a unified management level. Meanwhile, existing AI (Artificial Intelligence) assisted management platforms typically use only a single risk model, failing to cover the diverse management scenarios throughout the entire course of chronic diseases. Moreover, the management scenarios for coronary artery disease patients are heterogeneous: high-risk individuals requiring primary prevention need to predict the probability of future first events; patients undergoing long-term management after coronary artery disease surgery need to assess the risk of short- to medium-term mortality and recurrence; and patients in the cardiac rehabilitation stage need to be risk-stratified based on exercise safety and functional recovery. Different models have different output dimensions, prediction time windows, and stratification criteria. Existing systems lack a technical solution to normalize the outputs of heterogeneous models into a unified risk level, making system architecture difficult to expand, and adding new models often requires modification of downstream decision-making logic.

[0022] Furthermore, even existing systems fail to disclose how to systematically categorize the original variables of the risk model into dynamically updatable variables and fixed variables, nor do they establish specific mapping rules between each updatable variable and external data sources. Moreover, they lack a mechanism for automatically switching and fully updating the risk model when patient management scenarios change. Under conditions of incomplete data, existing systems lack a systematic fault-tolerant mechanism for setting validity windows for various types of data, performing last observation advance for expired data and marking confidence downgrades, thereby driving conservative scheduling strategies.

[0023] Meanwhile, existing follow-up scheduling is based on whether the current value of the indicator meets the standard to trigger enhanced monitoring, without utilizing historical measurement series for trend analysis. It lacks the ability to predict risks based on the time-series trends of the indicators and cannot proactively identify deteriorating trends and intervene when the indicators are still within the acceptable range. At the scheduling architecture level, the existing system does not abstract the global follow-up frequency adjustment and the enhanced monitoring of individual indicators into a two-layer scheduling architecture that operates independently, does not interfere with each other, and is uniformly refreshed at the time of follow-up. It also does not define conflict arbitration rules and feedback triggering channels between the two layers. Furthermore, the existing platform only configures follow-up frequency schemes for a single management scenario and does not propose a scenario-based scheduling framework.

[0024] In summary, there is an urgent need for a follow-up resource scheduling method that can adaptively match risk models based on the management scenario of chronic disease patients, systematically transform static scoring tools into outpatient dynamic monitoring tools, integrate multi-dimensional indicator status to synthesize a unified management level, construct a two-layer collaborative scheduling architecture, have a fault-tolerant guarantee mechanism under data missing conditions, and have trend prediction capabilities, so as to achieve precise allocation of medical resources and continuous improvement of patient management effectiveness.

[0025] To address the problems of existing technologies, embodiments of this application provide a method, apparatus, device, product, and storage medium for generating resource scheduling tasks. In these embodiments, a resource scheduling platform can serve as the executing entity for the method provided. This resource scheduling platform can be used to acquire medical rehabilitation data and monitoring indicator data of the target object, and perform data processing and scheduling calculation operations such as model matching, level assessment and adjustment, and task generation.

[0026] For example, the resource scheduling platform can adopt a client-server architecture, the specific physical form of which includes, but is not limited to, servers and terminal devices. The server can be a standalone physical server, a cloud server (Elastic Compute Service, ECS), a virtual machine, or a clustered server system composed of multiple servers, possessing the high-performance computing and storage capabilities required for data processing, model computation, and task scheduling. The terminal device can be a smartphone, tablet, personal computer (PC), wearable device with communication capabilities (such as a smartwatch or wristband), or a doctor's workstation terminal deployed in a hospital information system, used to present follow-up tasks to patients or medical staff, collect indicator data, receive instruction input, and send reminder notifications, etc.

[0027] Furthermore, the server and terminal devices can establish a remote communication connection via a network to achieve data interaction. This network can be a wired or wireless network, including but not limited to one or more combinations of communication methods such as the Internet, Local Area Network (LAN), Wide Area Network (WAN), cellular mobile communication networks (such as 4G / 5G networks), Bluetooth, or Low Energy Wide Area Network (such as LoRa, NB-IoT). Through this communication connection, the terminal device can send the collected health monitoring data of the target object (such as physiological parameters collected by wearable devices, symptom information entered by the patient through an application, or uploaded laboratory reports) to the server. After completing the corresponding data processing steps such as model evaluation, level adjustment, and task generation, the server will send the generated follow-up time scheduling instructions and monitoring item lists to the terminal device to guide patients and / or medical staff in performing the corresponding follow-up tasks.

[0028] It should be noted that the functional division between the server and the terminal device described above is merely an illustrative example. Those skilled in the art should understand that all or part of the data processing steps of the method provided in the embodiments of this application can also be executed locally on the terminal device without departing from the protection scope of this application.

[0029] The following section describes the method for generating resource scheduling tasks provided in the embodiments of this application, using a resource scheduling platform as the execution entity.

[0030] Figure 1 A flowchart illustrating a method for generating resource scheduling tasks according to an embodiment of this application is shown. Figure 1 As shown, the method includes the following steps S101 to S104: Step S101: Obtain the target medical management scenario corresponding to the medical rehabilitation status of the target object, as well as the indicator data for monitoring the health status of the target object.

[0031] In step S101, the target is a patient with a chronic disease, such as a patient with coronary heart disease. The medical rehabilitation status of the target can be characterized by structured case data such as the target's discharge record, surgical record, rehabilitation plan, and diagnostic labels. In this embodiment, different medical rehabilitation statuses have corresponding medical management scenarios, which may include, but are not limited to, long-term postoperative management scenarios for coronary heart disease, cardiac rehabilitation management scenarios, and management scenarios for high-risk groups for primary prevention.

[0032] In step S101, the indicator data corresponding to the target object can be physiological parameters collected in real time through wearable devices (e.g., smart bracelets, blood pressure monitors, ECG patches, etc.), such as heart rate and blood pressure; the indicator data can also be report data entered through an application, such as symptoms and medication status; the indicator data can also be laboratory test data extracted by uploading test reports and combining OCR (Optical Character Recognition) and LLM (Large Language Model), such as blood lipids, blood glucose, and serum creatinine.

[0033] Step S102: Input the indicator data into the follow-up demand assessment model corresponding to the target medical management scenario to obtain the initial resource scheduling level corresponding to the target object.

[0034] In step S102, different medical management scenarios correspond to different follow-up needs assessment models. For example, for the long-term management scenario after coronary heart disease surgery, the GRACE (Global Registry of Acute Coronary Events) 6-month post-discharge mortality risk model can be used; for the cardiac rehabilitation management scenario, the AACVPR (American Association of Cardiovascular and Pulmonary Rehabilitation) risk stratification algorithm can be used; and for the management scenario of high-risk groups for primary prevention, the China-PAR (Prediction for ASCVD Risk in China) 10-year ASCVD risk model can be used.

[0035] In this embodiment, the follow-up needs assessment model has an open architecture and is scalable. In practical applications, the follow-up needs assessment model can be extended according to clinical needs. New models can be added by adding model entries (e.g., model identifiers, applicable scenario identification conditions, input variable specifications, and normalized threshold mapping tables) without modifying the downstream decision-making logic. For example, a dedicated model for patients with early-onset coronary heart disease or those with type 2 diabetes can be incorporated.

[0036] Furthermore, different follow-up needs assessment models differ in their output dimensions, prediction time windows, and stratification criteria. In this embodiment, evidence-based thresholds are set for different follow-up needs assessment models to map the original model output to a unified multi-level risk label, i.e., resource scheduling level, such as low, medium, and high levels. This provides a unified input interface for downstream adjustments and task generation. After normalization, a unified resource scheduling level can be obtained for any chronic disease patient, regardless of the underlying model used. The downstream management level determination and follow-up scheduling engine only receive this normalized risk label, without depending on the specific model type, thereby achieving decoupling between the model layer and the decision layer.

[0037] For example, in the GRACE model for long-term postoperative management, if the target group is NSTE-ACS patients and the 6-month mortality risk is less than 3%, the initial resource allocation level output by the follow-up needs assessment model is low; if the 6-month mortality risk is greater than or equal to 3% but less than 8%, the initial resource allocation level is medium; and if the 6-month mortality risk is greater than or equal to 8%, the initial resource allocation level is high. Similarly, if the target group is STE-ACS patients and the 6-month mortality risk is less than 4.4%, the initial resource allocation level is low; if the 6-month mortality risk is greater than or equal to 4.4% but less than 11%, the initial resource allocation level is medium; and if the 6-month mortality risk is greater than or equal to 11%, the initial resource allocation level is high. CCS patients follow the stratification thresholds corresponding to NSTE-ACS patients.

[0038] For example, the AACVPR model can directly output three levels: low, medium, and high. The AACVPR model is based on a comprehensive assessment of factors such as LVEF (Left Ventricle Ejection Fraction), the presence of complex ventricular arrhythmias, the occurrence of abnormal hemodynamic responses during exercise testing, and the recovery status after cardiac arrest or myocardial infarction.

[0039] For example, for the China-PAR model, if the 10-year ASCVD risk is less than 5%, the model outputs a low rating; if the 10-year ASCVD risk is greater than or equal to 5% and less than 10%, the model outputs a medium rating; and if the 10-year ASCVD risk is greater than or equal to 10%, the model outputs a high rating.

[0040] Step S103: Adjust the initial resource scheduling level according to the degree of deviation between the indicator data and the corresponding indicator threshold to obtain the target resource scheduling level.

[0041] In step S103, after obtaining the initial resource scheduling level, the resource scheduling platform further dynamically corrects the initial resource scheduling level based on the deviation between the obtained indicator data and the preset corresponding indicator threshold.

[0042] Specifically, the resource scheduling platform pre-sets corresponding thresholds for each monitoring indicator. These thresholds can include normal compliance thresholds and safety warning thresholds. When the indicator data is within the compliance threshold range, it indicates that the indicator is well controlled and no adjustment to the initial resource scheduling level is required. When the indicator data exceeds the compliance threshold but has not yet triggered a safety warning, it indicates that the target object has a certain degree of indicator anomaly and resource investment needs to be appropriately increased. When any core indicator triggers the safety warning threshold, the initial resource scheduling level is directly upgraded to the highest level to respond to potential risks immediately. Through the above deviation-driven adjustment mechanism, the final output target resource scheduling level not only depends on the baseline output of the follow-up demand assessment model but also reflects the dynamic deviation of the target object's current indicators from the normal range in real time, thereby realizing the transformation from static assessment to dynamic monitoring.

[0043] Step S104: Based on the target resource scheduling level, target medical management scenario, and indicator data, generate a target resource scheduling task for follow-up resource scheduling of the target object.

[0044] In step S104, the target resource scheduling task includes, but is not limited to, the resource scheduling frequency (e.g., the time interval between global follow-up visits), resource scheduling method (e.g., telephone follow-up, video follow-up, outpatient follow-up, home follow-up, etc.), and monitoring item list (e.g., blood lipid testing, blood pressure monitoring, electrocardiogram examination, etc.) for scheduling medical resources for the target object.

[0045] Specifically, in different medical management scenarios, the same resource scheduling level can correspond to different follow-up resource configuration schemes, such as different global follow-up visit intervals and monitoring item combinations. Furthermore, if the indicator data shows that a certain indicator has not met the standard or is under intensive monitoring, the final generated target resource scheduling task can also include intensive monitoring instructions for that indicator independent of the global follow-up visit cycle. This task is ultimately output to the terminal device in the form of time scheduling instructions and a list of monitoring items to guide patients and / or medical staff in performing the corresponding follow-up operations.

[0046] Based on the scheme defined in steps S101 to S104 above, it can be understood that in this embodiment, the target medical management scenario is determined according to the medical rehabilitation status of the target object, and a follow-up needs assessment model adapted to the target medical management scenario is used to determine the initial resource scheduling level of the target object. This initial resource scheduling level changes dynamically according to the medical rehabilitation status and indicator data of the target object. Consequently, the resource scheduling tasks determined based on this initial resource scheduling level can also be adapted to the target object. Moreover, the initial resource scheduling level is adjusted according to the deviation between the indicator data and the corresponding indicator threshold, so that the resource scheduling level is no longer fixed to the initial value output by the follow-up needs assessment model, but can dynamically change with the deviation of the target object's actual indicators from the normal range. This allows follow-up resources to be accurately allocated to the target objects that truly need attention, reducing the waste of medical resources. Furthermore, the target resource scheduling task is generated based on the target resource scheduling level, medical management scenario, and indicator data to correspond to the actual status of the target object. This results in the final output target resource scheduling task being dynamically calculated based on three dimensions: scenario type, follow-up requirement level, and real-time indicator data, rather than a fixed node on a unified calendar. This allows for targeted allocation of resource scheduling tasks to the target object, reducing the waste of medical resources caused by ineffective follow-ups.

[0047] Therefore, the embodiments of this application realize a complete data processing and scheduling calculation process from multi-source heterogeneous data input to safe adaptive follow-up scheduling instruction output, which enables follow-up resources to be dynamically configured according to the individualized status of the target object. While ensuring the follow-up density of high-risk patients, it actively reduces the unnecessary follow-up frequency of low-risk or stable patients, thereby effectively reducing the waste of medical resources.

[0048] The specific implementation process of the method provided in the embodiments of this application is described below.

[0049] In some embodiments, in the process of determining the resource scheduling task for the target object, the resource scheduling platform first needs to determine the medical management scenario corresponding to the target object.

[0050] Specifically, the resource scheduling platform acquires the medical rehabilitation data of the target object and determines the target medical management scenario corresponding to the target object from multiple medical management scenarios based on the preset correlation between the medical rehabilitation data and the medical management scenario.

[0051] In the above embodiments, medical rehabilitation data is used to characterize the medical rehabilitation status. The medical rehabilitation data may include, but is not limited to, the target subject's disease diagnosis type (e.g., acute coronary syndrome ACS, chronic coronary syndrome CCS), whether they have undergone percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG), whether they have been included in a cardiac rehabilitation program, whether they have cardiovascular risk factors (e.g., hypertension, diabetes, dyslipidemia, smoking history, etc.), and structured label information for cardiovascular risk assessment.

[0052] In this embodiment, the medical rehabilitation data can originate from the multi-dimensional structured medical record data generated by the target individual in a medical institution. For example, the resource scheduling platform can automatically obtain data such as the target individual's discharge records, surgical records, rehabilitation plans, diagnostic labels, and medical progress records through data interfaces with hospital information systems (HIS), electronic medical record systems (EMR), or rehabilitation management systems.

[0053] Through the aforementioned medical rehabilitation data, the resource scheduling platform can accurately identify the current disease stage and management needs of the target individuals, providing a data foundation for subsequent scenario matching.

[0054] In the above embodiments, the resource scheduling platform is pre-configured with multiple medical management scenarios, each corresponding to a different stage of disease management and clinical management goal. For example, these multiple medical management scenarios may include three categories: long-term post-coronary artery disease management, cardiac rehabilitation management, and management of high-risk groups for primary prevention.

[0055] The aforementioned pre-defined relationships can be used to define matching rules between various structured tags in medical rehabilitation data and medical management scenarios. Specifically, the resource scheduling platform uses a pre-defined set of judgment rules to perform conditional matching on various tags in the acquired medical rehabilitation data, thereby selecting the target medical management scenario corresponding to the current state of the target object from multiple medical management scenarios.

[0056] For example, if the medical record label in the medical rehabilitation data includes ACS (acute coronary syndrome) or CCS (chronic coronary syndrome), and the subject has undergone PCI (percutaneous coronary intervention) or CABG (coronary-artery-bypass-grafting), then the subject is determined to be in the long-term management scenario after coronary heart disease surgery; if the medical rehabilitation data indicates that the subject has been included in a cardiac rehabilitation program, then the subject is determined to be in the cardiac rehabilitation management scenario; if the medical rehabilitation data indicates that the subject has not yet experienced an ASCVD (atherosclerotic cardiovascular disease) event, but has multiple cardiovascular risk factors (e.g., hypertension, diabetes, dyslipidemia, smoking, obesity, etc.), then the subject is determined to be in the primary prevention high-risk population management scenario.

[0057] Through the above embodiments, the resource scheduling platform can accurately determine the target medical management scenario in which the target object is currently located based on the target object's medical rehabilitation data, thereby providing a scenario-based input basis for the subsequent selection of the corresponding follow-up needs assessment model, avoiding the problem of inaccurate assessment caused by improper model selection due to scenario misjudgment or single scenario configuration.

[0058] In some embodiments, each of the multiple medical management scenarios has a corresponding priority. When medical rehabilitation data matches multiple medical management scenarios, the resource scheduling platform can determine the medical management scenario with the highest priority among the multiple medical management scenarios as the target medical management scenario corresponding to the target object.

[0059] In the above embodiments, the resource scheduling platform can pre-configure a corresponding priority for each of the multiple medical management scenarios. This priority indicates the order in which each medical management scenario is selected. The correspondence between priority and medical management scenario can be manually set by medical personnel based on their management experience, or it can be based on the severity of the disease, the urgency of medical intervention, the clinical risk level of the management scenario, and the follow-up management priority recommended by evidence-based medicine guidelines. For example, in chronic disease management scenarios, because patients face a higher risk of recurrence and complications in the early postoperative stage, more rigorous management resource allocation is required. Therefore, early postoperative management usually has a higher priority than rehabilitation management; while primary prevention scenarios have a relatively lower management urgency compared to postoperative management scenarios.

[0060] When the resource scheduling platform matches acquired medical rehabilitation data with pre-defined relationships across scenarios, the data may simultaneously satisfy multiple medical management scenarios. For example, if a target patient's discharge record includes "ACS" and they underwent "PCI" surgery, and this patient is also included in a cardiac rehabilitation program, then the medical rehabilitation data matches both the long-term post-coronary artery disease management scenario and the cardiac rehabilitation management scenario. In this scenario, the resource scheduling platform determines the highest-priority medical management scenario among the multiple scenarios as the target medical management scenario for the target patient.

[0061] Taking the three management scenarios mentioned above as examples, their priorities can be set as follows: the priority of the long-term management scenario after coronary heart disease surgery is higher than that of the cardiac rehabilitation management scenario, and the priority of the cardiac rehabilitation management scenario is higher than that of the high-risk population management scenario for primary prevention. Therefore, when the medical rehabilitation data of the target individual matches both the long-term management scenario after coronary heart disease surgery and the cardiac rehabilitation management scenario, the resource scheduling platform will determine the long-term management scenario after coronary heart disease surgery, which has the highest priority, as the target medical management scenario, thereby ensuring that the high-risk management needs in the early postoperative period are met first. Meanwhile, the cardiac rehabilitation management scenario is registered as a backup scenario to trigger scenario switching when the assessment model corresponding to the main scenario experiences an anomaly such as a confidence level downgrade. The specific switching mechanism will be described in detail in subsequent embodiments.

[0062] Through the above embodiments, the resource scheduling platform can determine the most suitable target medical management scenario when there are multiple overlaps in patient medical rehabilitation data. This avoids the problem that the resource scheduling platform cannot make a decision or requires manual intervention due to matching conflicts, and ensures the automated execution of subsequent processes and the safety of clinical decision-making.

[0063] After determining the target medical management scenario corresponding to the target object, in this embodiment of the application, the resource scheduling platform can also dynamically adjust the target medical management scenario based on the confidence level of the indicator data, so that when the quality of the collected indicator data declines, the management scenario can be switched to an alternative scenario with more robust data requirements, so as to ensure the safety and reliability of subsequent evaluation and scheduling decisions.

[0064] In some embodiments, the resource scheduling platform obtains the data validity period corresponding to the indicator data in the target medical management scenario; then, based on the data duration and data validity period of the indicator data, it determines the confidence level of the indicator data; and if the confidence level of the indicator data is lower than the preset first level threshold, the medical management scenario with the second highest priority among multiple medical management scenarios is determined as the target medical management scenario corresponding to the target object.

[0065] In the above embodiments, the resource scheduling platform first obtains the data validity period corresponding to the indicator data under the target medical management scenario. The data validity period characterizes the time range within which various types of monitoring data can be considered valid, and its value can be set according to the clinical characteristics and collection methods of different data types. For example, for vital sign data (such as heart rate and blood pressure) collected in real time by wearable devices, since this type of data has high timeliness requirements, expired data may not accurately reflect the patient's current physiological state. Therefore, its data validity period can be set to a shorter time window, such as 7 days. For laboratory test data (such as blood lipids and serum creatinine), since the rate of change in the clinical significance of this type of data is relatively slow, appropriately extending the validity period can improve data utilization and avoid disturbing patients by triggering frequent data collection due to slight expiration. Therefore, its data validity period can be set to a relatively longer time window, such as 1.5 times the corresponding testing period.

[0066] After determining the data validity period, the resource scheduling platform further obtains the duration of the indicator data and determines the confidence level of the indicator data based on the comparison between the data duration and the data validity period. The data duration refers to the time span from the data collection time to the current evaluation time. When the data duration does not exceed the data validity period, it indicates that the indicator data is within the valid time window, and its confidence level is normal or not downgraded, meaning the data can be reliably used for subsequent evaluation calculations. When the data duration exceeds the data validity period, it indicates that the indicator data has expired, and its confidence level is correspondingly reduced. Furthermore, the resource scheduling platform can further subdivide the confidence level into multiple levels based on the degree to which the data exceeds the validity period. For example, expired data not exceeding 1 time the validity period is considered a slight downgrade; expired data exceeding 1 time but not more than 2 times the validity period is considered a moderate downgrade; and expired data exceeding 2 times the validity period is considered a severe downgrade.

[0067] After determining the confidence level, the resource scheduling platform compares the confidence level with a preset first-level threshold. If the confidence level is lower than the first-level threshold, for example, if the confidence level drops to a moderate or severe downgrade, it indicates that the input data on which the follow-up needs assessment model corresponding to the target medical management scenario relies is not reliable enough. If the follow-up needs model is continued to be used for assessment, the assessment results may be distorted due to low data quality, thereby generating unsafe scheduling instructions.

[0068] In this embodiment, when the confidence level of the indicator data is lower than a preset first-level threshold, the resource scheduling platform determines the second-highest priority medical management scenario among multiple medical management scenarios as the target medical management scenario corresponding to the target object. That is, when the indicator data corresponding to the main scenario (i.e., the initially matched highest-priority scenario) is severely outdated and its confidence level decreases, the resource scheduling platform switches the management scenario to the highest-priority scenario among the candidate scenarios and uses the follow-up need assessment model corresponding to that candidate scenario for subsequent level assessment and scheduling task generation. For example, when the target object was originally matched to the long-term management scenario after coronary heart disease surgery, but due to the long-term lack of updates to its core monitoring indicator data, its confidence level drops to a moderate downgrade and is lower than the first-level threshold, the resource scheduling platform switches the management scenario to the cardiac rehabilitation management scenario and uses the AACVPR model for subsequent assessment.

[0069] For example, a 55-year-old male patient was admitted for emergency PCI due to acute anterior ST-segment elevation myocardial infarction. Upon discharge, he was classified as a long-term post-coronary artery disease (CARD) patient, with a 6-month mortality risk of 6.8% calculated by GRACE (intermediate risk of STE-ACS), a moderate resource allocation level, and global follow-up every 3 months. Two months after discharge, the patient was enrolled in an outpatient cardiac rehabilitation program. After rehabilitation assessment data was entered, the resource allocation platform detected the new cardiac rehabilitation program record and switched the management scenario from long-term CARD management to cardiac rehabilitation management, and the follow-up needs assessment model from GRACE to AACVPR. AACVPR assessed the patient; LVEF was 42%, exercise testing showed no ischemia but low tolerance, and there were no complex ventricular arrhythmias, resulting in a moderate resource allocation level. The follow-up frequency for this patient was changed from every 3 months in the long-term CARD management scenario to every 4-8 weeks in the cardiac rehabilitation management scenario. Meanwhile, the monitoring frequency of the patient's indicators has also been adjusted. For example, the treatment rule for the patient's systolic blood pressure has been changed from the average of the home self-test over the past 7 days to the actual measurement value of the most recent rehabilitation assessment center, and the heart rate data source has been changed from wearable devices to assessment center measurements.

[0070] Through the above embodiments, in the case of insufficient data quality, management decisions can be switched from models with high requirements for data integrity to models with relatively robust data requirements or different evaluation dimensions. This avoids inaccurate evaluations or abnormal scheduling instructions due to data quality issues, ensuring the continuity of automated system operation while also ensuring the safety and robustness of clinical decisions.

[0071] In some embodiments, after determining the confidence level of the indicator data, the resource scheduling platform can also implement a differentiated treatment mechanism based on the confidence level. This mechanism executes different levels of intervention measures according to the severity of the confidence level, thereby forming a progressive treatment strategy from mild to severe. This approach avoids both overreacting and neglecting prevention when data quality declines, achieving a balance between the automatic operational stability of the resource scheduling platform and clinical safety.

[0072] Specifically, the resource scheduling platform pre-sets multiple confidence level thresholds, including a first-level threshold, a second-level threshold, and a third-level threshold. The first-level threshold is higher than the second-level threshold, and the second-level threshold is higher than the third-level threshold. These threshold values ​​are used to classify different levels of confidence level downgrade. A higher threshold value indicates a higher tolerance for that level of confidence, meaning a milder downgrade; a lower threshold value indicates that corresponding action measures will only be triggered when the confidence level drops to a more severe level.

[0073] If the confidence level of the indicator data is lower than the preset second-level threshold, the resource scheduling platform increases the data monitoring frequency of the indicator data; if the confidence level of the indicator data is lower than the preset third-level threshold, the resource scheduling platform generates a task review instruction, which is used to instruct the generated target resource scheduling task to be reviewed.

[0074] Specifically, when the confidence level of the indicator data decreases, the resource scheduling platform initiates a multi-channel data collection process, which is triggered sequentially in the order of application push, intelligent voice outbound call, instant messaging message, and human telephone, while maintaining a reasonable interval between adjacent channels to avoid disturbing patients.

[0075] In this embodiment, the confidence level can be divided into normal, mild, moderate, and severe degradation. The first level threshold corresponds to mild degradation, the second level threshold corresponds to moderate degradation, and the third level threshold corresponds to severe degradation. For different confidence levels, the resource scheduling platform adopts different task scheduling strategies. For example, for mildly degraded indicator data, the resource scheduling platform maintains the existing resource scheduling level and the existing resource scheduling tasks. For moderately degraded indicator data, the resource scheduling platform maintains the existing resource scheduling level and only performs conservative measures in the scheduling and monitoring dimensions. For example, it shortens the follow-up visit interval, increases the refresh frequency of the global follow-up visit schedule, and increases the monitoring frequency of the indicator data. Simultaneously, it extends the active data collection attempt window, recalculating and removing the degradation as soon as the indicator data is replenished. The initial resource scheduling level is only increased when the replenished real indicator data triggers the existing risk escalation rules. For severely downgraded indicator data, the resource scheduling platform triggers a manual review work order. Before the work order is completed, the automatic update of the existing global follow-up schedule is frozen, that is, the global follow-up schedule is not updated. The scheduling instructions are issued manually by clinical or operations personnel after evaluation.

[0076] For example, the monitoring frequency varies for different indicators. For blood pressure, it is typically collected every 7 days; if the blood pressure is below target, the measurement is uploaded every 3 days until it reaches the target. For heart rate, it is typically collected every 7 days; if the heart rate is below target, the resting heart rate is reported every 3 days until it returns to the target range. For blood glucose, it is typically collected every 7 days; if the blood glucose is below target, the measurement is uploaded every 3 days until it returns to the target range. For LDL-C, if the LDL-C is below target, the monitoring frequency is increased to once every 3 months until it reaches the target. For lifestyle indicators, such as smoking, weight, alcohol consumption, and physical activity, if the target is not met, personalized health education pushes and behavioral intervention reminders are triggered, but the monitoring frequency remains unchanged.

[0077] Furthermore, the monitoring frequency for indicators with the same confidence level varies across different medical management scenarios. For example, in the long-term management of coronary heart disease post-surgery, low-confidence indicators are monitored every 6 months, medium-confidence indicators every 3 months, and high-confidence indicators every month. In the cardiac rehabilitation management scenario, low-confidence indicators are monitored every 3 months, medium-confidence indicators every 4-8 weeks, and high-confidence indicators every 2-4 weeks. In the management of high-risk groups for primary prevention, low-confidence indicators are monitored every 12 months, medium-confidence indicators every 12 months, and high-confidence indicators every 6 months. In early warning scenarios, immediate remote consultation or recommendation for in-person emergency care is triggered, and a comprehensive follow-up examination is arranged within one month after the early warning indicators are eliminated.

[0078] For example, for a 58-year-old male patient, GRACE initially outputs a low resource scheduling level. Two months after the patient's discharge, the resource scheduling platform performs a monthly recalculation and finds that the patient has not submitted any new data. At this point, the resource scheduling platform implements a fault-tolerant strategy, recalculating the three dynamically updatable variables—heart rate, systolic blood pressure, and creatinine—using the last valid observation values. Since the vital signs data has expired for 14 days (twice the validity window), the confidence level is updated to a moderate downgrade. Simultaneously, the resource scheduling platform triggers a multi-channel active data collection process, for example, actively collecting data in the order of application push notifications, voice calls, and WeChat messages. On the third day, the patient responds to a voice call, reporting a blood pressure of 130 / 85 mmHg and a heart rate of 65 bpm. The resource scheduling platform updates the data and reassesses, resolving the confidence level downgrade and restoring the original scheduling schedule. This scheduling schedule includes resource scheduling time, which is determined by the previous resource scheduling time and frequency.

[0079] If there is no response within 3 days of actively collecting indicator data, the resource scheduling platform will treat the indicator data as a moderate downgrade. The existing target resource scheduling level will remain unchanged at the low level, only the resource scheduling frequency will be reduced, i.e., the interval between follow-up visits will be shortened, the full follow-up visits will be brought forward appropriately, and the refresh frequency of the follow-up visit schedule will be increased. At the same time, enhanced monitoring will be initiated for indicators with missing data, such as blood pressure and heart rate, i.e., the data monitoring frequency will be increased. Once the data is supplemented, the system will recalculate immediately. The resource scheduling level will only be upgraded if the actual data triggers the upgrade rules of the existing resource scheduling level. If the indicator data expires for more than 14 days (i.e., twice the validity period window), the resource scheduling platform will treat it as a severe downgrade, triggering a manual review work order. The automatic update of the global follow-up visit schedule will be frozen until the work order is completed. The scheduling instruction will be manually issued after evaluation by clinical or operations personnel.

[0080] In some embodiments, the indicator data includes first indicator data corresponding to multiple physiological indicators, and second indicator data corresponding to the historical diagnostic data of the target object.

[0081] In this embodiment, the indicator data is categorized into indicator data with dynamically updated variables (i.e., first indicator data) and indicator data with fixed, retained variables (i.e., second indicator data). This categorization of indicator data is not a simple binary labeling process, but rather achieved through multi-dimensional decision-making logic.

[0082] Specifically, firstly, an equivalence assessment of the data sources is conducted. The clinical equivalence of each variable under the conditions of alternative outpatient data sources versus the original inpatient data collection conditions is evaluated. Only when the alternative outpatient data source can reflect the pathophysiological information required by the follow-up needs assessment model within an acceptable error range is the variable classified as a dynamically updatable variable. Secondly, considering the physiological fluctuation cycle of the variable, the rate of change of clinical significance, and the feasibility of data collection, an effective data collection time window is set for each dynamically updatable variable, such as the last 7 days or the last 30 days. Thirdly, since different aggregation methods affect the numerical accuracy and clinical significance input into the follow-up needs assessment model, the appropriate data aggregation method must be selected to aggregate the indicator data according to the variable characteristics. For example, continuous vital sign data is aggregated using the mean within the time window, discrete laboratory data is aggregated using the last valid value, and event-type data is determined by the presence or absence of binary values. Furthermore, when multiple outpatient data sources exist for the same variable, the priority and conflicts between multiple data sources need to be resolved. Data can be selected according to a preset data source priority sequence. The priority ranking is based on the objectivity, continuity, and collection error of the data source, and conflict resolution rules are defined when there are differences between multiple data sources. Finally, by combining medical management scenarios with the aforementioned time windows, aggregation methods, and multi-source priorities, the classification of indicator data is achieved. That is, the processing rules for the same input variable can be different in different management scenarios. For example, in the post-coronary artery disease postoperative scenario, "systolic blood pressure" is recalculated using the average of the home self-measurements over the past 7 days, while in the cardiac rehabilitation scenario, the most recent measured value from the rehabilitation assessment center is used. "Heart rate" in the postoperative scenario prioritizes wearable device data, while in the rehabilitation scenario, it prioritizes the measurement value from the assessment center.

[0083] It should be noted that the contextual relevance of the indicator data allows the same methodological framework to adapt to the different requirements of data real-time performance, reliability, and clinical representativeness at different stages of disease progression.

[0084] In this embodiment, the dynamically updatable variables are those that, after the above-mentioned multidimensional decision evaluation, are confirmed to have clinically equivalent alternative data that can be obtained from data sources such as wearable devices, applications, or laboratory reports in the out-of-hospital environment; the fixed variables are those that, after evaluation, are confirmed to be accurately assessable only in the inpatient environment and do not have reliable alternative data sources in the out-of-hospital environment, and the values ​​of the most recent inpatient records are used during routine out-of-hospital management.

[0085] In this embodiment, an outpatient data source mapping, time window, and aggregation method are established for each dynamically updatable variable. Taking the GRACE model as an example, heart rate is mapped to the average resting heart rate collected by a wearable device over the past 7 days, or the most recent resting heart rate manually entered via an application or verbally reported via AI voice follow-up when the wearable device is not worn; systolic blood pressure is mapped to the arithmetic mean of home-measured systolic blood pressure over the past 7 days, with data sources including wearable blood pressure monitoring devices, application input, or AI voice follow-up; serum creatinine is mapped to the creatinine value extracted by OCR and LLM from the most recently uploaded laboratory report.

[0086] For fixed variables, age (which updates naturally over time), Killip functional classification, history of cardiac arrest at admission, elevated myocardial markers at admission, and ST segment changes on ECG at admission can be used as past event records and fixed. The assessment values ​​at discharge should be used and not updated in the out-of-hospital environment.

[0087] When a patient is readmitted, the resource scheduling platform can synchronously obtain the new hospitalization data through the hospital information system and refresh all input variables of the follow-up needs assessment model in one go. Simultaneously, it reassesses whether the patient needs a change in medical management scenario.

[0088] In this embodiment of the application, the resource scheduling platform can pre-set a period (e.g., monthly) to assess the patient's follow-up needs (i.e., risks) using a follow-up needs assessment model.

[0089] In some embodiments, after the indicator data is input into the follow-up needs assessment model, the model outputs an initial resource scheduling level for the target object. This initial resource scheduling level reflects the risk of the target object; the higher the risk, the higher the corresponding initial resource scheduling level, indicating a higher need for follow-up. To ensure that the final resource scheduling task is more aligned with the target object, the resource scheduling platform adjusts the initial resource scheduling level based on the degree of compliance of the indicator data after obtaining the initial resource scheduling level.

[0090] Specifically, the resource scheduling platform first obtains the relative values ​​of the first indicator data corresponding to multiple physiological indicators and the indicator thresholds of the corresponding physiological indicators; if the relative value of each physiological indicator is less than the first threshold of the corresponding physiological indicator, the initial resource scheduling level is determined as the target resource scheduling level; if the relative value of at least one physiological indicator is greater than the second threshold of the corresponding physiological indicator, the highest resource scheduling level is determined as the target resource scheduling level; if the number of first physiological indicators is greater than a preset number threshold, the initial resource scheduling level is increased to obtain the target resource scheduling level, wherein the first physiological indicator is a physiological indicator whose relative value is greater than the first threshold of the corresponding physiological indicator and less than the second threshold of the corresponding physiological indicator.

[0091] In this embodiment, the threshold values ​​corresponding to each physiological indicator may include one or more levels of thresholds. Each physiological indicator is configured with at least a first threshold and a second threshold. The first threshold is used to determine whether the physiological indicator is within the acceptable range, and the second threshold is used to determine whether the physiological indicator has triggered a safety alert. The relative value of the indicator is used to measure the direction and degree of deviation of the current measurement value of the physiological indicator relative to its indicator threshold. The relative value of the indicator may be the difference or ratio between the current measurement value of the indicator and the corresponding indicator threshold, or it may be a normalized relative deviation.

[0092] For example, different indicator data have different corresponding target ranges and safety warning ranges. For instance, for LDL-C, the target range is LDL-C < 1.8 mmol / L; for systolic blood pressure, the target range is systolic blood pressure < 130 mmHg, and the safety warning range is systolic blood pressure < 90 or > 160 mmHg; for diastolic blood pressure, the target range is diastolic blood pressure < 80 mmHg, and the safety warning range is diastolic blood pressure < 60 or > 120 mmHg; for heart rate, the target range is 55-6... 0 bpm is considered safe, with a heart rate <50 or >100 bpm. For fasting blood glucose, the target range is 4.4-7.2 mmol / L, with a safety warning range of ≤4 or ≥14 mmol / L. For glycated hemoglobin (HbA1c), the target range is <7.0%. For 2-hour postprandial blood glucose, the target range is <10 mmol / L. For antiplatelet drug adherence, the target range is ≥80%. For BMI (Body Mass Index), the target range is 20-25 kg / m². 2For moderate-intensity exercise frequency, the target range is 30-60 minutes / day, 5 days / week. Safety warnings may also include bleeding events or a significant decrease in hemoglobin; myalgia / muscle weakness with ALT >120 U / L or CK >1000 U / L; and persistent or spontaneous episodes of chest tightness, chest pain, palpitations, or dyspnea. In addition, diet, mental health, sleep quality, and follow-up adherence can be included as auxiliary assessment dimensions in health education intervention decisions, but will not trigger adjustments to the follow-up frequency. The above thresholds are set with reference to clinical guidelines and may be adjusted according to guideline updates.

[0093] In this embodiment, when the relative value of each physiological indicator is less than the first threshold of its corresponding physiological indicator, meaning all monitored physiological indicators are within the acceptable range, it indicates that the target subject's current health indicators are well controlled and no clinically significant abnormal deviations have occurred. In this scenario, the resource scheduling platform sets the initial resource scheduling level as the target resource scheduling level without making any adjustments to the initial resource scheduling level. This avoids unnecessary resource upgrades for patients in good health and helps conserve medical resources.

[0094] When the relative value of at least one of multiple physiological indicators exceeds the second threshold of the corresponding physiological indicator, meaning any monitored physiological indicator has triggered a safety warning interval, the target subject faces an immediate clinical safety risk and requires immediate response. In this scenario, the resource scheduling platform sets the highest resource scheduling level as the target resource scheduling level, directly maximizing the level to the highest level (e.g., "warning level" or "high-risk level"), regardless of the initial resource scheduling level. This achieves a "critical value direct access mechanism," ensuring that once a safety warning signal is issued, the resource scheduling platform can allocate follow-up resources to the patient with the highest priority, guaranteeing that clinical safety is not delayed.

[0095] Among multiple physiological indicators, there exists a category of physiological indicators whose relative values ​​are greater than the first threshold of the corresponding physiological indicator but less than the second threshold. In other words, this indicator is in the middle range of "not meeting the standard but not yet triggering a safety warning". This category of physiological indicators is called "first physiological indicator".

[0096] If the number of primary physiological indicators exceeds a preset threshold, it indicates that while no single indicator shows a critical condition, multiple indicators are borderline abnormal. This reflects a risk of deterioration in the target individual's overall health, necessitating a moderate increase in resource allocation. In this case, the resource scheduling platform raises the initial resource scheduling level to obtain the target resource scheduling level. For example, the initial resource scheduling level may be raised from medium to high, or from low to medium.

[0097] It should be noted that the preset quantity threshold can be configured differently according to the different initial resource scheduling levels. For example, the quantity threshold is negatively correlated with the initial resource scheduling level. When the initial resource scheduling level is low, at least 3 primary physiological indicators are required to trigger the upward adjustment. When it is medium, at least 2 are required. When it is high, only 1 is required to trigger the upward adjustment.

[0098] Through the above embodiments, the hierarchical response logic under different situations enables the adjustment of resource scheduling levels to simultaneously take into account the immediate response needs in the case of a single indicator in a critical situation and the gradual upgrade needs in the case of multiple indicators accumulating anomalies. This avoids the superimposed risks caused by focusing only on a single indicator and overlooking the marginal anomalies of multiple indicators, and also avoids the security risks of insufficient response when there are few non-compliant indicators but a single indicator is seriously abnormal.

[0099] In some embodiments, when the initial resource scheduling level is increased, and the relative values ​​of all physiological indicators among multiple physiological indicators are less than the first threshold of the corresponding physiological indicator, and the duration reaches the first preset duration, the resource scheduling platform shall reduce the increased initial resource scheduling level.

[0100] Specifically, after the initial resource scheduling level has been increased, the resource scheduling platform continuously monitors the relative values ​​of each of the multiple physiological indicators. When it is detected that the relative values ​​of all physiological indicators are less than the first threshold of the corresponding physiological indicator, it indicates that all monitored physiological indicators of the target object have recovered to the acceptable range, and there are no longer any conditions for triggering a level increase due to indicators not meeting the standards.

[0101] However, if the resource scheduling level is immediately reverted to the pre-increase level as soon as the indicator data recovers to the target level, the level may fluctuate frequently due to the short-term normal fluctuations of physiological indicators. For example, the measured value of a certain indicator may fluctuate repeatedly around the target line, causing the system to frequently switch between increasing and decreasing the level. This not only causes instability in scheduling instructions but may also confuse patients and medical staff and increase their operational burden.

[0102] To avoid the aforementioned problems, this embodiment introduces a delayed confirmation mechanism. Specifically, after all physiological indicators have returned to the target level, a level reversal operation is performed only after the duration of this target state reaches a first preset duration. Specifically, the resource scheduling platform starts timing from the moment all physiological indicators last fall within the target range. If the target state is maintained continuously for the first preset duration, such as three consecutive assessment cycles, or 14 or 21 consecutive days, it is confirmed that the target's health status has achieved stable improvement, rather than being merely a single measurement or a short-term, accidental achievement. In this scenario, the resource scheduling platform lowers the initial resource scheduling level after the initial increase, that is, it reverts the resource scheduling level to the level before the increase, or at least to a lower level that matches the current health status.

[0103] Taking a clinical scenario as an example, suppose a patient's initial resource allocation level is low, but is upgraded to medium because three indicators simultaneously fail to meet the standards. Subsequently, through lifestyle interventions and medication, these three indicators gradually return to the target level. Only if all indicators remain within the target range for three consecutive assessment cycles will the resource allocation platform revert the resource allocation level back from medium to low. If any indicator falls back into the unmet range during this period, the timer is reset, and the accumulation of consecutive target-compliant time must restart.

[0104] The aforementioned callback mechanism can prevent frequent fluctuations in scheduling levels due to short-term data volatility, thereby ensuring the continuity and stability of follow-up resource scheduling. It also makes the resource scheduling tasks output by the resource scheduling platform more reliable and reduces the chaos in clinical management caused by frequent changes.

[0105] In some embodiments, after determining the target resource scheduling level, the target medical management scenario, and the indicator data, the resource scheduling platform can generate a target resource scheduling task that includes the resource scheduling frequency.

[0106] Specifically, the resource scheduling platform determines an initial resource scheduling frequency that matches the target resource scheduling level and the target medical management scenario based on the correlation between the resource scheduling level, the medical management scenario, and the scheduling task. If the number of second physiological indicators is greater than a preset second quantity threshold, or the number of third physiological indicators is greater than a preset third quantity threshold, and the target collection number corresponding to the third physiological indicator is greater than a preset number threshold, the initial resource scheduling frequency is increased to obtain the target resource scheduling frequency.

[0107] In the above embodiments, the second physiological indicator is a physiological indicator whose relative value is greater than a first threshold among multiple physiological indicators, the third physiological indicator is a physiological indicator whose influence on the health status of the target object is greater than a preset degree threshold and whose relative value is greater than the first threshold among multiple physiological indicators, and the target collection number is the number of data collections corresponding to the third physiological indicator whose relative value is greater than the first threshold.

[0108] In this embodiment, the task scheduling system is divided into two levels: global re-examination scheduling and single-indicator enhanced monitoring scheduling, which operate independently.

[0109] For global follow-up appointment scheduling, the resource scheduling platform determines the frequency of resource scheduling for follow-up visits to target individuals based on the target resource scheduling level. The resource scheduling frequency varies for the same resource scheduling level across different medical management scenarios to match the clinical management goals of each scenario. The resource scheduling frequency can also be dynamically adjusted based on the resource scheduling level. Furthermore, the resource scheduling frequency will also be dynamically adjusted when a medical management scenario changes.

[0110] For enhanced monitoring and scheduling of individual indicators, the monitoring frequency of each indicator can be encrypted or restored independently of the resource scheduling frequency. When an indicator is identified as falling outside the acceptable range, the resource scheduling platform activates enhanced monitoring mode for that indicator, i.e., increases the monitoring frequency of that indicator; once the indicator recovers and meets the requirements, the enhanced monitoring mode is immediately exited, and the normal monitoring frequency is restored. This scheduling rule is applicable to all management scenarios.

[0111] In this embodiment, the two scheduling systems described above operate in coordination.

[0112] Specifically, in this embodiment, the completion record generated by the target resource scheduling task in the hospital information system is used as the unified refresh trigger point for the resource scheduling frequency and the data monitoring frequency of individual indicators. At this trigger point, the system completes data synchronization for all dynamically updatable variables at once, and then recalculates the normalized risk label, composite management level, global follow-up visit schedule, and enhanced monitoring status of various indicators in sequence, ensuring that both layers hold consistent data snapshots and evaluation results at this point in time.

[0113] When multiple indicators simultaneously enter the enhanced monitoring mode, and the time remaining before the next resource scheduling is too long, the resource scheduling platform can further detect if the number of indicators entering the enhanced monitoring mode (i.e., the number of second physiological indicators) exceeds a second threshold (e.g., 2). Alternatively, if any of these indicators is a preset critical indicator (e.g., including systolic blood pressure, heart rate, LDL-C, and fasting blood glucose) and has failed to meet the standard in multiple consecutive measurements, the next resource scheduling time will be brought forward, i.e., the initial resource scheduling frequency will be increased. If only one non-critical indicator enters the enhanced monitoring mode, the original resource scheduling frequency will remain unchanged, and only the data monitoring frequency of that non-critical indicator will be increased.

[0114] It should be noted that after the initial resource scheduling frequency is increased, that is, after the resource scheduling task is triggered in advance, the resource scheduling platform will not use the same arbitration strategy to trigger the resource scheduling task in advance again within the preset cooldown period (e.g., 14 days), in order to prevent repeated advances from causing scheduling oscillations.

[0115] Furthermore, within a second preset time period of increasing the initial resource scheduling frequency, if the relative value of the fourth physiological indicator is less than the threshold value of the corresponding physiological indicator, and the initial resource scheduling level is less than the preset level threshold, the resource scheduling platform reduces the initial resource scheduling frequency to obtain the target resource scheduling frequency. Here, the fourth physiological indicator can be either the second or third physiological indicator. That is, after the patient's abnormal indicators are effectively improved, the resource scheduling frequency is restored to a reasonable level in a timely manner, avoiding the prolonged maintenance of an encrypted scheduling state due to past abnormal events, thereby conserving medical resources while ensuring safety.

[0116] Specifically, after increasing the initial resource scheduling frequency (i.e., the target resource scheduling frequency is higher than the initial frequency), the resource scheduling platform continuously monitors the relative values ​​of various physiological indicators. When the following conditions are met, the resource scheduling platform executes a frequency reduction (i.e., a callback) operation: Condition 1, Time Condition: The current time point is within the second preset time period after the initial resource scheduling frequency was increased. This condition sets the effective time window for resource scheduling frequency callbacks, meaning that callbacks are only allowed to be executed within a preset time period after the resource scheduling frequency is increased.

[0117] It is worth noting that the increased frequency of resource allocation is usually triggered in response to the patient's abnormal condition. If the abnormal condition returns to normal quickly during the encrypted monitoring period, it indicates that the abnormality may be a short-term fluctuation or a transient event, and should be adjusted back in time to avoid over-monitoring. If it exceeds the preset time, it indicates that the abnormal condition has lasted for a long time and may require intervention at a higher level of clinical assessment, rather than simply adjusting the frequency.

[0118] Condition 2, indicator improvement condition: The relative value of the fourth physiological indicator of the target subject is less than the first threshold of the corresponding physiological indicator. That is, as long as any one of the previously non-compliant indicators that caused the increase in frequency has recovered to the standard state, the indicator improvement condition is met.

[0119] Condition 3, Level Condition: The initial resource allocation level is less than a preset level threshold. This condition ensures that frequency rollback is only performed when the patient's initial resource allocation level is within a controllable range. If the patient's initial resource allocation level is high, even if a certain indicator temporarily recovers to the target level, the frequency should not be immediately rolled back; encrypted monitoring should be maintained to ensure safety. This level condition adds a safety constraint to frequency rollback, avoiding hasty frequency reduction in patients with high underlying risk due to short-term improvement in a single indicator.

[0120] When all three conditions above are met, the resource scheduling platform determines that the abnormal indicators of the target object have improved within the effective time window of encrypted monitoring, and its resource scheduling level is within the allowable callback range. Therefore, it executes an operation to reduce the initial resource scheduling frequency to obtain the target resource scheduling frequency. That is, the scheduling frequency previously increased due to the abnormal state is reverted to the initial level or to a lower frequency level that matches the current improvement state.

[0121] It should be noted that if the target subject's fourth physiological indicator fails to recover to the target within the second preset time period after the initial resource scheduling frequency is increased, it indicates that the abnormality is not a short-term fluctuation but a persistent problem. In this case, the frequency reduction operation will not be executed, and the resource scheduling platform will continue to maintain the increased scheduling frequency until the problem is clinically confirmed to be resolved. If the problem persists, the resource scheduling platform can also trigger a higher level of intervention.

[0122] Through the aforementioned frequency callback mechanism, if the abnormal indicators recover to the target level within a preset time window and the initial resource scheduling level is less than the preset level threshold after the initial resource scheduling frequency is increased, a timely callback will be initiated to avoid resource waste. If the situation does not improve after the time window expires, the encrypted state will be maintained and further intervention will be triggered. This mechanism effectively balances clinical safety and resource efficiency, preventing the scheduling frequency from being over-configured due to a single abnormal event.

[0123] For example, a 72-year-old female patient initially had a medium resource scheduling level output by GRACE, with an initial follow-up frequency of once every 3 months. Six weeks after discharge, the patient reported intermittent chest tightness and shortness of breath after activity over the past week via AI-powered voice follow-up. These symptoms triggered a safety warning zone, at which point the resource scheduling level rose to the warning level, and the resource scheduling platform immediately pushed an alert suggesting remote consultation or emergency care. Simultaneously, the global level was immediately notified to arrange immediate consultation. After a doctor's evaluation ruled out an acute event, a comprehensive follow-up examination was arranged within one month of symptom resolution. If the follow-up examination was completed and the symptoms did not recur, the resource scheduling level was delayed and lowered to medium level after 3 cycles.

[0124] In another example, if a patient simultaneously enters enhanced monitoring mode for three indicators—blood pressure, heart rate, and LDL-C—and blood pressure and heart rate are among the preset key indicators that fail to meet the standards multiple times consecutively, then the frequency of the next resource scheduling will be increased and the resource scheduling time will be shortened according to the aforementioned conflict arbitration rules. For example, the resource scheduling time may be shortened from 3 months to 6 weeks. During the 14-day cooling-off period after the early trigger, similar arbitration requests will not be responded to.

[0125] In some embodiments, the resource scheduling platform can also use the historical measurement sequence of the target object to perform trend analysis, identify potential deterioration trends in advance when the indicators are still within the acceptable range, and proactively adjust the data monitoring frequency accordingly, thereby upgrading from passively responding to the current value to proactively predicting future trends.

[0126] Specifically, the resource scheduling platform first acquires indicator data within a third preset time period. Then, using time as the independent variable and indicator data as the dependent variable, it performs regression analysis on the indicator data to obtain the trend curve of the indicator data. When the relative value of the indicator corresponding to the indicator data is less than the first threshold, and the trend curve indicates that the indicator data is tending to deteriorate, the data monitoring frequency of the indicator data is increased. When the relative value of the indicator corresponding to the indicator data is greater than or equal to the first threshold and less than the second threshold, and the trend curve indicates that the indicator data is tending to be in a non-deteriorating state, the data monitoring frequency of the indicator data remains unchanged.

[0127] In this embodiment, the third preset duration is used to define a retrospective time window within which the resource scheduling platform collects a sequence of historical measurements of a specific physiological indicator. The specific value of the third preset duration can be configured according to the physiological fluctuation characteristics and clinical significance of the indicator, for example, it can be set to 30 days, 60 days, or a time span covering the most recent 3 to 5 valid measurements.

[0128] After acquiring the indicator data sequence within the third preset time period, the resource scheduling platform performs regression analysis on the indicator data, using time as the independent variable and the measured value of the indicator data as the dependent variable, thereby obtaining the trend curve of the indicator changing over time. For example, univariate linear regression analysis can be used to calculate the slope β of the trend of the indicator measurement value changing over time.

[0129] It should be noted that linear regression is only one implementation method of this application. In other implementations, other time series analysis methods such as exponentially weighted moving average (EWMA), moving average line, or nonlinear fitting can also be used to characterize the changing trend of the indicator, as long as they can quantify the direction and rate of change of the indicator within a certain time window. This application does not limit the specific trend calculation method.

[0130] After obtaining the trend curve, the resource scheduling platform combines the current relative value of the indicator with the trend direction. When the current relative value of a certain physiological indicator is less than the first threshold of the corresponding physiological indicator, and the trend curve indicates that the indicator is tending to deteriorate, the resource scheduling platform increases the data monitoring frequency of the indicator.

[0131] It should be noted that although the indicator is currently still within the acceptable range, it has shown a clear deteriorating trend. If the regular monitoring frequency is maintained, it will be impossible to detect in time when it exceeds the acceptable threshold, thus delaying intervention. By increasing the monitoring frequency in advance, the resource scheduling platform can obtain data updates when the indicator is about to exceed the threshold, providing data support for timely adjustments to the scheduling strategy.

[0132] When the current relative value of a certain physiological indicator is greater than or equal to the first threshold of the corresponding physiological indicator and less than the second threshold of the corresponding physiological indicator, and the indicator tends to be in a non-deteriorating state according to the trend curve, the resource scheduling platform maintains the data monitoring frequency of the indicator unchanged.

[0133] It should be noted that although the indicators are not currently up to standard, trend analysis shows that they are improving in a positive direction or at least not continuing to deteriorate. In this case, it is sufficient to maintain the current monitoring frequency. There is no need to increase the monitoring density because the indicators are not up to standard, so as to avoid unnecessary investment of monitoring resources when the patient is already on the path to improvement.

[0134] Through the above embodiments, when an indicator is still within the acceptable range but shows a deteriorating trend, monitoring can be intensified in advance to achieve early risk detection and response; when an indicator is not yet acceptable but tends to improve, unnecessary increases in frequency can be avoided, saving monitoring resources. This trend prediction mechanism ensures that the adjustment of monitoring frequency not only depends on the static compliance status of the indicator but also integrates the dynamic change information of the indicator, realizing an upgrade from state-driven to trend-driven, and overcoming the shortcomings of existing technologies that rely solely on current cross-sectional values ​​for decision-making and ignore time-series dynamic information.

[0135] To make the technical solution of this application clearer and easier to understand, the following uses the complete management process of a patient with coronary heart disease as an example to illustrate the execution process of the method described in this application in practical applications. It should be noted that the specific values, thresholds, and judgment results of each step in this embodiment are only used to assist in understanding the technical solution of this application and do not constitute any limitation on the scope of protection of this application.

[0136] A 63-year-old male patient was admitted to the hospital for unstable angina and underwent percutaneous coronary intervention (PCI). The discharge record showed a diagnosis of "ACS" (acute coronary syndrome) and clearly stated that he had undergone PCI.

[0137] Upon discharge, the resource scheduling platform automatically synchronizes the patient's inpatient data via a data interface, including discharge records, surgical records, diagnostic labels, and admission test results. Based on the discharge record label containing "ACS" and the patient undergoing PCI surgery, the platform matches the patient to the "long-term management scenario after coronary artery disease surgery" according to preset scenario priority rules, and selects the GRACE 6-month post-discharge mortality risk model as the follow-up needs assessment model for this scenario. The platform performs an initial GRACE score calculation based on the patient's discharge data, yielding a 6-month mortality risk of 2.5%. Based on the normalized threshold of the GRACE model in non-ST-segment elevation acute coronary syndrome (NSTE-ACS) patients, this score is mapped to a unified risk label (i.e., resource scheduling level) of "low level".

[0138] The patient's baseline parameters collected at this time included: low-density lipoprotein cholesterol (LDL-C) of 85 mg / dL, systolic blood pressure of 136 mmHg, normal fasting blood glucose, resting heart rate of 62 bpm, non-smoker, and body mass index (BMI) of 26.8 kg / m². 2 .

[0139] The resource scheduling platform further adjusts the resource scheduling level based on the compliance status of indicator data. Since the normalized risk label is "low level," the non-compliant indicators are LDL-C, blood pressure, and BMI (3 items in total). However, BMI is a lifestyle-related auxiliary indicator, and the number of core non-compliant indicators is 2, which is below the 3 thresholds required to trigger an upgrade from "low level," and no safety warning conditions have been triggered. Therefore, according to the basic inheritance rules, the target resource scheduling level remains "low level."

[0140] Based on the target resource scheduling level and target medical management scenario, the resource scheduling platform generates an initial resource scheduling task according to the scenario-based resource scheduling frequency configuration, namely, arranging a comprehensive follow-up visit after 6 months. Simultaneously, enhanced monitoring of two core indicators that have not met the standards is initiated: LDL-C is retested every 3 months; blood pressure is reported every 3 days. BMI, as a lifestyle indicator, triggers health education push notifications but does not change its monitoring frequency.

[0141] One month after the patient's discharge, the resource allocation platform performed a dynamic recalculation of the resource allocation level according to a preset cycle. At this time, the acquired out-of-hospital monitoring data were: average heart rate of 58 bpm over the past 7 days, average home-measured systolic blood pressure of 128 mmHg over the past 7 days, and serum creatinine had not been updated. Since the creatinine data had not been updated since discharge, its duration exceeded the validity window set by the platform for laboratory data, but the expiration was less than one full validity window. Therefore, the resource allocation platform recorded the confidence level as "mildly downgraded." According to the rules for handling mildly downgraded confidence levels, the original resource allocation level and schedule remained unchanged, and the last effective creatinine value of 1.0 mg / dL was continued for the score recalculation. After recalculation, the GRACE score was 2.3%, and after normalization, it remained "low level."

[0142] At the same time, the resource scheduling platform detected that the blood pressure had dropped to 128 / 76 mmHg (meeting the standard) and immediately exited the blood pressure intensive monitoring mode.

[0143] Three months after the patient's discharge, the resource scheduling platform sent a reminder to the patient for a follow-up laboratory test, according to the intensive monitoring plan of retesting LDL-C every three months. After the patient uploaded the test report, the resource scheduling platform used OCR and LLM combined extraction technology to extract the data from the report, finding that LDL-C was 62 mg / dL and serum creatinine was 0.9 mg / dL. The platform determined that LDL-C was within the target range and immediately exited the intensive LDL-C monitoring mode; the creatinine value was included in the data pool for the next monthly recalculation.

[0144] Five months after the patient's discharge, the patient's three most recent home blood pressure measurements were: 122 / 78 mmHg (first measurement), 126 / 80 mmHg (second measurement), and 132 / 82 mmHg (third measurement). The trend analysis module of the resource scheduling platform obtained the three most recent valid measurements and performed a univariate linear regression analysis with the measurement sequence number or measurement date as the independent variable and the systolic blood pressure measurement value as the dependent variable. The calculated systolic blood pressure rise slope was 0.67 mmHg / day, exceeding the deterioration warning threshold configured by the resource scheduling platform for this patient's management scenario, while the current systolic blood pressure of 132 mmHg was still within the target range. The resource scheduling platform determined that although the indicator was still within the target range, it showed a clear deterioration trend. Based on the trend prediction rules, it triggered enhanced blood pressure monitoring in advance, increasing the blood pressure monitoring frequency from the regular frequency to once every 3 days, initiating close monitoring.

[0145] Thus, the complete management process of this patient within 5 months after discharge demonstrates the complete closed-loop management process of the method described in this application, from scenario matching, baseline assessment, composite grade synthesis, initial scheduling generation, monthly dynamic recalculation, confidence downgrade handling, exit upon indicator achievement, to trend early warning and early intervention.

[0146] This application also provides a resource scheduling task generation apparatus, such as... Figure 2 As shown, the device 200 includes: a data acquisition module 201, a level evaluation module 202, a level adjustment module 203, and a task generation module 204.

[0147] The data acquisition module 201 is used to acquire the target medical management scenario corresponding to the medical rehabilitation status of the target object, and the indicator data for monitoring the health status of the target object under the target medical management scenario; The rating assessment module 202 is used to input indicator data into the follow-up demand assessment model corresponding to the target medical management scenario to obtain the initial resource scheduling level corresponding to the target object; The level adjustment module 203 is used to adjust the initial resource scheduling level according to the deviation between the indicator data and the corresponding indicator threshold to obtain the target resource scheduling level. The task generation module 204 is used to generate target resource scheduling tasks for follow-up resource scheduling of target objects based on the target resource scheduling level, target medical management scenario and indicator data.

[0148] In some embodiments, the data acquisition module includes: The first data acquisition module is used to acquire medical rehabilitation data of the target object, wherein the medical rehabilitation data is used to characterize the medical rehabilitation status. The scenario determination module is used to determine the target medical management scenario corresponding to the target object from multiple medical management scenarios based on the preset correlation between medical rehabilitation data and medical management scenarios.

[0149] In some embodiments, each of the multiple medical management scenarios has a corresponding priority. The scenario determination module is specifically used to determine the medical management scenario with the highest priority among the multiple medical management scenarios as the target medical management scenario corresponding to the target object when the medical rehabilitation data matches the multiple medical management scenarios.

[0150] In some embodiments, the resource scheduling task generation device further includes: a scenario adjustment module, used to obtain the data validity period corresponding to the indicator data in the target medical management scenario; determine the confidence level of the indicator data based on the data storage duration and the data validity period; and determine the medical management scenario with the second highest priority among multiple medical management scenarios as the target medical management scenario corresponding to the target object when the confidence level of the indicator data is lower than a preset first level threshold.

[0151] In some embodiments, the resource scheduling task generation apparatus further includes a first frequency adjustment module and a verification module. The first frequency adjustment module is used to increase the data monitoring frequency of the indicator data when the confidence level of the indicator data is lower than a preset second-level threshold, wherein the first-level threshold is greater than the second-level threshold. The verification module is used to generate a task verification instruction when the confidence level of the indicator data is lower than a preset third-level threshold, wherein the task verification instruction instructs the generated target resource scheduling task to be verified, and the third-level threshold is less than the second-level threshold.

[0152] In some embodiments, the indicator data includes first indicator data corresponding to multiple physiological indicators, and second indicator data corresponding to the historical diagnostic data of the target object.

[0153] In some embodiments, the level adjustment module includes: a first level adjustment module, configured to acquire relative values ​​between first indicator data corresponding to multiple physiological indicators and indicator thresholds of the corresponding physiological indicators; when the relative value of each physiological indicator among the multiple physiological indicators is less than the first threshold of the corresponding physiological indicator, determine the initial resource scheduling level as the target resource scheduling level; when the relative value of at least one physiological indicator among the multiple physiological indicators is greater than the second threshold of the corresponding physiological indicator, determine the highest level of resource scheduling as the target resource scheduling level; when the number of first physiological indicators among the multiple physiological indicators is greater than a preset number threshold, increase the initial resource scheduling level to obtain the target resource scheduling level, wherein the first physiological indicator is a physiological indicator whose relative value is greater than the first threshold of the corresponding physiological indicator and less than the second threshold of the corresponding physiological indicator.

[0154] In some embodiments, the resource scheduling task generation device further includes: a second level adjustment module, used to lower the initial resource scheduling level when the initial resource scheduling level is raised, provided that the relative values ​​of all physiological indicators among a plurality of physiological indicators are less than the first threshold of the corresponding physiological indicator and the duration reaches a first preset duration.

[0155] In some embodiments, the target resource scheduling task includes at least a resource scheduling frequency. The task generation module is specifically used to determine an initial resource scheduling frequency that matches the target resource scheduling level and the target medical management scenario based on the correlation between the resource scheduling level, the medical management scenario, and the scheduling task. When the number of second physiological indicators is greater than a preset second quantity threshold, or the number of third physiological indicators is greater than a preset third quantity threshold, and the target collection number corresponding to the third physiological indicator is greater than a preset number threshold, the initial resource scheduling frequency is increased to obtain the target resource scheduling frequency. Here, the second physiological indicator is a physiological indicator whose relative value is greater than a first threshold among multiple physiological indicators, the third physiological indicator is a physiological indicator whose influence on the health status of the target object is greater than a preset degree threshold, and the relative value of the indicator is greater than the first threshold among multiple physiological indicators, and the target collection number is the number of data collections whose relative value of the indicator corresponding to the third physiological indicator is greater than the first threshold.

[0156] In some embodiments, the resource scheduling task generation device further includes: a second frequency adjustment module, used to reduce the initial resource scheduling frequency to obtain a target resource scheduling frequency when, within a second preset duration of increasing the initial resource scheduling frequency, the relative value of the index corresponding to the fourth physiological index is less than the index threshold of the corresponding physiological index, and the initial resource scheduling level is less than a preset level threshold. The fourth physiological index is any one of the second physiological index or the third physiological index.

[0157] In some embodiments, the resource scheduling task generation device further includes: a third frequency adjustment module, configured to acquire indicator data within a third preset time period; perform regression analysis on the indicator data with time as the independent variable and indicator data as the dependent variable to obtain a trend curve of indicator data; increase the data monitoring frequency of indicator data when the relative value of the indicator data is less than a first threshold and the trend curve indicates that the indicator data is trending towards a deteriorating state; and maintain the data monitoring frequency of indicator data unchanged when the relative value of the indicator data is greater than or equal to the first threshold and less than a second threshold and the trend curve indicates that the indicator data is trending towards a non-deteriorating state.

[0158] The resource scheduling task generation apparatus provided in this application embodiment can implement the various processes implemented in the aforementioned method embodiments, and will not be described again here to avoid repetition.

[0159] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0160] Figure 3 A schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application is shown.

[0161] The electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0162] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0163] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0164] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0165] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the resource scheduling task generation methods in the above embodiments.

[0166] In one example, the electronic device may also include a communication interface 303 and a bus 310. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0167] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0168] Bus 310 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0169] Furthermore, in conjunction with the resource scheduling task generation method in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the resource scheduling task generation methods in the above embodiments.

[0170] Furthermore, in conjunction with the resource scheduling task generation method in the above embodiments, this application embodiment can provide a computer program product for implementation. When the instructions in this computer program product are executed by the processor of an electronic device, the electronic device performs the resource scheduling task generation method as described in any of the above embodiments.

[0171] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0172] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0173] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0174] The foregoing flowcharts and / or block diagrams describing methods, apparatuses, devices, products, and storage media for generating resource scheduling tasks according to embodiments of the present disclosure have described various aspects of this disclosure. It should be understood that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, 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, or other programmable data processing apparatus to create a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0175] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for generating resource scheduling tasks, characterized in that, include: Acquire the target medical management scenario corresponding to the medical rehabilitation status of the target object, as well as the indicator data for monitoring the health status of the target object; The indicator data is input into the follow-up demand assessment model corresponding to the target medical management scenario to obtain the initial resource scheduling level corresponding to the target object; Based on the degree of deviation between the indicator data and the corresponding indicator threshold, the initial resource scheduling level is adjusted to obtain the target resource scheduling level; Based on the target resource scheduling level, the target medical management scenario, and the indicator data, a target resource scheduling task for follow-up resource scheduling of the target object is generated. The step of generating a target resource scheduling task for follow-up resource scheduling of the target object based on the target resource scheduling level, the target medical management scenario, and the indicator data includes: Based on the correlation between resource scheduling level, medical management scenario and scheduling task, determine the initial resource scheduling frequency that matches the target resource scheduling level and the target medical management scenario; If the target indicator in the indicator data does not fall into the corresponding compliance range, the data monitoring frequency of the target indicator is increased, and the data monitoring frequency of the target indicator is adjusted independently of the initial resource scheduling frequency; If multiple indicators meet preset conditions or preset key indicators fail to meet the standards multiple times consecutively, the initial resource scheduling frequency is increased; if only one non-key indicator fails to meet the standards, the initial resource scheduling frequency is kept unchanged, and the data monitoring frequency of the non-key indicator is increased.

2. The method according to claim 1, characterized in that, The acquisition of the target medical management scenario corresponding to the medical rehabilitation status of the target object includes: Acquire the medical rehabilitation data of the target object, wherein the medical rehabilitation data is used to characterize the medical rehabilitation status; Based on the pre-defined correlation between medical rehabilitation data and medical management scenarios, the target medical management scenario corresponding to the target object is determined from multiple medical management scenarios.

3. The method according to claim 2, characterized in that, Each of the multiple medical management scenarios has a corresponding priority. The step of determining the target medical management scenario corresponding to the target object from the multiple medical management scenarios based on the preset association relationship between medical rehabilitation data and the medical management scenarios includes: When the medical rehabilitation data matches the multiple medical management scenarios, the medical management scenario with the highest priority among the multiple medical management scenarios is determined as the target medical management scenario corresponding to the target object.

4. The method according to claim 3, characterized in that, The method further includes: Under the target medical management scenario, obtain the data validity period corresponding to the indicator data; The confidence level of the indicator data is determined based on the data storage duration and the data validity period of the indicator data. If the confidence level of the indicator data is lower than the preset first level threshold, the medical management scenario with the second highest priority among the multiple medical management scenarios is determined as the target medical management scenario corresponding to the target object.

5. The method according to claim 4, characterized in that, The method further includes: If the confidence level of the indicator data is lower than the preset second-level threshold, the data monitoring frequency of the indicator data shall be increased, wherein the first-level threshold is greater than the second-level threshold. If the confidence level of the indicator data is lower than the preset third-level threshold, a task review instruction is generated. The task review instruction is used to instruct the generated target resource scheduling task to be reviewed. The third-level threshold is less than the second-level threshold.

6. The method according to any one of claims 1 to 5, characterized in that, The indicator data includes first indicator data corresponding to multiple physiological indicators, and second indicator data corresponding to the historical diagnostic data of the target object.

7. The method according to claim 6, characterized in that, The step of adjusting the initial resource scheduling level based on the deviation between the indicator data and the corresponding indicator threshold to obtain the target resource scheduling level includes: Obtain the relative values ​​of the first indicator data corresponding to the plurality of physiological indicators and the indicator thresholds of the corresponding physiological indicators; If the relative value of each of the plurality of physiological indicators is less than the first threshold of the corresponding physiological indicator, the initial resource scheduling level is determined as the target resource scheduling level. If the relative value of at least one of the multiple physiological indicators is greater than the second threshold of the corresponding physiological indicator, the highest level of resource scheduling will be determined as the target resource scheduling level. If the number of first physiological indicators among the plurality of physiological indicators is greater than a preset threshold, the initial resource scheduling level is increased to obtain the target resource scheduling level. The first physiological indicator is a physiological indicator whose relative value is greater than the first threshold of the corresponding physiological indicator and less than the second threshold of the corresponding physiological indicator.

8. The method according to claim 7, characterized in that, The method further includes: If the initial resource scheduling level is increased, and the relative values ​​of all physiological indicators among the plurality of physiological indicators are less than the first threshold of the corresponding physiological indicator, and the duration of the increase reaches the first preset duration, then the increased initial resource scheduling level is reduced.

9. The method according to claim 8, characterized in that, The method further includes: If, within a second preset time period during which the initial resource scheduling frequency is increased, the relative value of the fourth physiological indicator is less than the threshold value of the corresponding physiological indicator, and the initial resource scheduling level is less than a preset level threshold, the initial resource scheduling frequency is reduced to obtain the target resource scheduling frequency. Here, the fourth physiological indicator is either the second physiological indicator or the third physiological indicator.

10. The method according to claim 7, characterized in that, The method further includes: Obtain indicator data within the third preset time period; Using time as the independent variable and the indicator data as the dependent variable, a regression analysis is performed on the indicator data to obtain the trend curve of the indicator data. When the relative value of the indicator corresponding to the indicator data is less than the first threshold, and the indicator data is showing a deteriorating trend according to the trend curve, the data monitoring frequency of the indicator data is increased. If the relative value of the indicator corresponding to the indicator data is greater than or equal to the first threshold, less than the second threshold, and the indicator data tends to be in a non-deteriorating state according to the trend curve, the data monitoring frequency of the indicator data shall remain unchanged.

11. A resource scheduling task generation apparatus, characterized in that, include: The data acquisition module is used to acquire the target medical management scenario corresponding to the medical rehabilitation status of the target object, as well as the indicator data for monitoring the health status of the target object; The rating assessment module is used to input the indicator data into the follow-up demand assessment model corresponding to the target medical management scenario to obtain the initial resource scheduling rating corresponding to the target object; The level adjustment module is used to adjust the initial resource scheduling level according to the deviation between the indicator data and the corresponding indicator threshold to obtain the target resource scheduling level. The task generation module is used to generate a target resource scheduling task for follow-up resource scheduling of the target object based on the target resource scheduling level, the target medical management scenario, and the indicator data. Specifically, it is used to: determine an initial resource scheduling frequency that matches the target resource scheduling level and the target medical management scenario based on the correlation between the resource scheduling level, the medical management scenario, and the scheduling task; increase the data monitoring frequency of the target indicator when the target indicator in the indicator data does not fall into the corresponding compliance range, and the data monitoring frequency of the target indicator is adjusted independently of the initial resource scheduling frequency; increase the initial resource scheduling frequency when the abnormal state of multiple indicators meets preset conditions or when preset key indicators fail to meet the standards multiple times consecutively; and keep the initial resource scheduling frequency unchanged and increase the data monitoring frequency of the non-key indicator when only one non-key indicator fails to meet the standards.

12. An electronic device, characterized in that, The electronic device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the resource scheduling task generation method as described in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the method for generating resource scheduling tasks as described in any one of claims 1-10.

14. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device is able to perform the resource scheduling task generation method as described in any one of claims 1-10.

Images