Physical examination integrated management method and system based on intelligent public health

Abstract

The invention relates to the technical field of intelligent medical treatment and public health management, and discloses a physical examination integrated management method and system based on intelligent public health, and the method comprises the following steps: firstly constructing a quality control model and a personnel equipment initial capability portrait; then responding to the appointment, and executing quality perception type scheduling according to the portrait and the resource availability; collecting business data and process metadata through a mobile workstation, and dynamically updating a capability portrait according to the business data and the process metadata; then aggregating the data, evaluating the data confidence of the abnormal indexes in combination with the updated portraits, and generating a report with scores; and finally, according to the severity and confidence of the abnormal indexes, differentiated follow-up service pushing is executed. According to the method, capability portraits of operators and equipment are established, process metadata in the physical examination process is collected and dynamically updated, and in the resource scheduling stage, the system can accurately match health requirements of residents with the service quality of service resources in a feedforward mode.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent medical treatment and public health management, in particular to a physical examination integrated management method and system based on intelligent public health. BACKGROUND

[0002] Preventive health examination in the public health system is the core strategy to achieve early detection and intervention of diseases. With the increasing demand for health management of residents, the process of physical examination involves more and more complex links, from appointment scheduling, on-site execution to report interpretation and follow-up services, each link directly affects the final health management effect. Therefore, it is of great significance to build a physical examination integrated management method and system that can systematically integrate and optimize the whole process, in order to improve the accuracy and effectiveness of public health services.

[0003] Currently, various medical information systems have been widely used in physical examination management. These systems successfully convert traditional paper records into electronic data, realize online and automation of physical examination appointment, and significantly improve the efficiency of administrative management. By connecting the test and examination equipment of different departments within the hospital, the existing technology can collect scattered physical examination data and generate standardized electronic physical examination reports. This centralized data storage method enables doctors to easily access residents' previous examination results, providing basic data support for clinical diagnosis.

[0004] However, in actual application, the existing technical solutions have several deep-seated limitations. First, the resource scheduling mechanism and process quality control are disconnected. When the system allocates personnel or equipment, it mainly depends on the availability of time, and the intrinsic ability or recent work performance of the service provider (personnel or equipment) is unknown, resulting in blindness in resource allocation and inability to actively match high-quality medical resources to residents with higher health risks. Second, the existing system has deficiencies in evaluating the quality of physical examination data. Once the data passes the basic logical and range checks, it is accepted, and the contextual information in the production process, such as the operator's experience level or the smoothness of specific operations, is completely ignored, making all data appear equally reliable in the report, and downstream health managers cannot quantitatively distinguish the reliability of the data. Finally, the existing physical examination process presents a linear and one-way information flow between links. The quality performance in the physical examination process cannot be fed back to optimize future resource scheduling, and the intrinsic reliability of the data cannot be directly linked to the response level of subsequent services, which limits the accuracy of health management intervention. Therefore, the technical personnel in the field propose a physical examination integrated management method and system based on intelligent public health to solve the above problems. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a physical examination integrated management method and system based on intelligent public health, which solves the problems of allocation blindness caused by the disconnection between resource scheduling and process quality control, insufficient data quality assessment caused by the lack of context information of data generation process, and insufficient accuracy of subsequent service response caused by the fragmentation of process links in the prior art.

[0006] To achieve the above object, the present application is implemented by the following technical solutions: The present application provides a physical examination integrated management method based on intelligent public health, characterized by the following steps: first, in the integrated management platform, a preset quality control model is constructed and an initial capability profile is established. The preset quality control model includes business data quality control rules for checking physical examination data values and logic, and process metadata collection rules for recording operation associated information. The initial capability profile is quantitative data representing the initial service capability of each operating personnel and each physical examination device participating in the physical examination.

[0007] Secondly, in response to the physical examination appointment request transmitted by the external system, quality-aware resource scheduling is performed. This scheduling step generates a scheduling scheme containing specified operating personnel and specified physical examination devices for the appointment request according to the established capability profile and the real-time availability of medical resources obtained through system monitoring.

[0008] Subsequently, data collection is performed through a mobile workstation according to the generated scheduling scheme. This step collects physical examination business data and synchronously collects process metadata associated with the physical examination business data at the time of generation.

[0009] Then, based on the collected process metadata, the capability profile is dynamically updated using a preset update algorithm. This step enables the capability profile of the operating personnel and the physical examination device to reflect their recent work quality status.

[0010] After that, data confidence assessment is performed and a physical examination report is generated. This step aggregates all physical examination business data of the specified resident, and combines the updated capability profile to perform data confidence assessment on the abnormal indicators presented therein, and finally generates a structured physical examination report with a data confidence score.

[0011] Finally, according to the generated physical examination report, differentiated subsequent service push is performed. This step matches and executes preset service linkage strategies containing different response levels according to the clinical severity of abnormal indicators in the report and the data confidence score.

[0012] In one embodiment, the step of constructing the preset quality control model and the initial capability profile is performed by connecting the regional public health information system, the hospital information system, the inspection information system and various medical examination devices through a standard data interface to establish a unified data platform. The process metadata collection rules are used to record the unique identification of the personnel performing the operation, the unique identification of the device used, the start and end time stamp of the operation, the number of times the data is modified before submission and the log of the quality control events triggered during the operation.

[0013] Preferably, the step of performing quality-aware resource scheduling is performed by receiving a medical examination appointment request initiated by a resident service portal, obtaining real-time busy state information of the operation personnel and the medical examination device to form real-time availability of the medical resources, calculating a quality matching degree between the resident health risk level and the capability profile, and finally planning a medical examination path and generating the scheduling scheme through a target function with the optimization goal of reducing resident waiting time and improving service quality matching degree.

[0014] In one embodiment, the step of collecting data through a mobile workstation is performed by embedding a quality control engine in the mobile workstation, which performs real-time verification according to the business data quality control rules during data collection and sends an instant reminder to the operation personnel when data anomalies are found. When the mobile workstation is in a state of no network connection, all collected medical examination business data and process metadata are encrypted and stored locally, and after the network connection is restored, the data is automatically synchronized with the integrated management platform.

[0015] Preferably, the step of dynamically updating the capability profile is performed by the integrated management platform analyzing the process metadata returned from the mobile workstation to calculate a comprehensive quality observation value for the operation quality of each operation personnel and each medical examination device within a specified period. Then, a preset update algorithm is used to combine the comprehensive quality observation value with the capability profile score of the previous period to generate the capability profile score of the current period.

[0016] In one embodiment, the step of generating a medical examination report with a data confidence score is performed by extracting the associated operation personnel capability profile score, the device capability profile score, the quality log factor calculated according to the process metadata and the data internal consistency factor evaluating the logical coordination of the data point with other related examination results for the abnormal medical examination business data from the system. Finally, the four factors are weighted and summed according to the preset weights to obtain the data confidence score.

[0017] Further, the data confidence score DCS is calculated by the following formula: DCS = w p · F p + w d• F d + w q • F q + w c • F c ; wherein, DCS is a data confidence score; F p is a normalized operator capability profile score; F d is a normalized equipment capability profile score; F q is a quality log factor calculated according to process metadata; F c is a data internal consistency factor for evaluating logical consistency of the data point with other inspection results; w p , w d , w q , w c are preset weight coefficients of corresponding factors respectively.

[0018] Preferably, the preset updating algorithm is an exponential weighted moving average algorithm.

[0019] In one specific embodiment, the step of performing differentiated follow-up service pushing comprises: judging a clinical severity level of the abnormal indicator by a built-in medical knowledge base; then constructing a two-dimensional decision matrix with the clinical severity level and the data confidence score as coordinate axes; and finally, matching and executing a corresponding pushing strategy according to the coordinate position of the abnormal indicator in the two-dimensional decision matrix, the pushing strategy including pushing an emergency follow-up task to a family doctor, pushing a pre-warning with review suggestions, or being incorporated into a routine chronic disease management process.

[0020] The second aspect of the present application provides a physical examination integrated management system based on intelligent public health, comprising: a model construction module, configured to construct a preset quality control model containing business data quality control rules and process metadata collection rules in an integrated management platform, and establish initial capability profiles for operators and physical examination equipment; a resource scheduling module, configured to respond to physical examination reservation requests, perform quality-aware resource scheduling according to the capability profiles and real-time availability of medical resources, and generate a scheduling scheme containing specified operators and specified physical examination equipment; a data collection module, configured to collect physical examination business data through mobile workstations according to the scheduling scheme, and synchronously collect process metadata associated with the physical examination business data; a profile updating module, configured to dynamically update the capability profiles based on the process metadata using a preset updating algorithm; and a report generation module, configured to aggregate the physical examination business data, combine the updated capability profiles, perform data confidence evaluation on abnormal indicators, and generate physical examination reports with data confidence scores. The service pushing module is configured to perform differentiated follow-up service pushing according to the clinical severity of the abnormal index in the physical examination report and the data confidence score.

[0021] To sum up, the present application includes at least one of the following beneficial technical effects: 1. The present application establishes the ability profile of the operator and the equipment, and dynamically updates the process metadata during the physical examination process. In the resource scheduling stage, the system can accurately match the health needs of residents with the service quality of service resources in a feedforward manner. The prior art usually only schedules according to the busy state of resources, and the quality control link and the scheduling link are independent of each other. The present application solves the problem of strong blindness in scheduling in the prior art and the inability to use process quality as a scheduling decision basis, and realizes the transition from passive allocation to active optimization configuration.

[0022] 2. The present application performs real-time quality control at the source of data collected by the mobile workstation. It synchronously records the complete metadata of the operation process. Based on these process information, the system calculates the data confidence score for each abnormal index, so that the final physical examination report not only presents the results, but also reveals the reliability of the results. Compared with the prior art, the physical examination data generated by the prior art lacks internal quality measurement, and all data passing the basic verification are treated equally. The present application effectively solves the problem of data quality black box in the prior art and provides a key and quantifiable basis for subsequent clinical decision and health management.

[0023] 3. The present application deeply couples the dynamic update of the ability profile, the evaluation of the data confidence score, and the follow-up service pushing mechanism, and constructs a complete and adaptive closed-loop management system. The final service pushing is no longer a simple information notification, but a double decision based on data severity and data reliability. In the prior art, the various links of the physical examination process are usually fragmented, the data flow is unidirectional, and there is a lack of feedback and cooperation. The present application overcomes the defects of resource mismatch and insufficient service response accuracy caused by process fragmentation in the prior art, and ensures that the medical intervention resources can be accurately guided to the link that needs it most and has the most reliable data basis. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 Fig. 1 is a schematic diagram of the system structure of the present application; Figure 2 Fig. 2 is a schematic diagram of the method flow of the present application; Figure 3 Fig. 3 is a schematic diagram of the two-dimensional decision matrix of the present application; Mark explanation: 100, model construction module; 200, resource scheduling module; 300, data acquisition module; 400, profile updating module; 500, report generation module; 600, service pushing module. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the specification of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0026] Referring to the drawings Figure 1 , Figure 1 is a structural schematic diagram of a physical examination integrated management system according to an embodiment of the present application. The present application provides a physical examination integrated management system based on intelligent public health. The system is deployed and run on one or more servers. The system can include a model construction module 100, a resource scheduling module 200, a data acquisition module 300, an image updating module 400, a report generation module 500, and a service pushing module 600.

[0027] The system interacts with multiple external entities through a network, including but not limited to a public health information system storing resident health records, a hospital information system, a test information system, a resident service portal for residents to make appointments, and a mobile workstation for on-site physical examination by operating personnel.

[0028] The model construction module 100 is used to configure and manage the basic rules and initial data of the system. The module is connected with external information systems through a standard data interface to form a data interaction center. It receives and stores business data quality control rules input by managers, which define the numerical valid range of physical examination business data and the logical consistency requirements between data. The module also receives and stores process metadata acquisition rules, which clearly define the fields that need to be recorded synchronously during data acquisition, such as operating personnel identification, equipment identification, operation start and end time stamps, data modification times, and log of triggering quality control events.

[0029] In addition, the model construction module 100 is also used to establish the initial ability image of operating personnel and physical examination equipment. The module extracts static information of operating personnel such as title and years of service, or information of physical examination equipment such as specifications and historical calibration records, generates an initial, quantitative ability image score for each operating personnel and equipment, and stores it.

[0030] The resource scheduling module 200 is configured to receive the physical examination request and generate a scheduling scheme. The module is connected to the resident service portal and configured to receive the physical examination appointment request submitted by the resident. The module is further configured to monitor and acquire the working status of each operator and physical examination device in real time, and form real-time availability data of medical resources. After receiving the appointment request, the module calls the capability profile of all relevant operators and devices from the model construction module 100, and executes a preset scheduling algorithm with the objective of optimizing resource allocation. The input of the algorithm includes real-time availability of medical resources, capability profile, and health risk level of the resident, and the output of the algorithm is a scheduling scheme specifying the order of physical examination items, specific execution operators, and physical examination devices.

[0031] The data acquisition module 300 is configured to perform data acquisition and real-time quality control at the physical examination site. The module is deployed on the mobile workstation in the form of an application program. The module receives the scheduling scheme generated by the resource scheduling module 200 and presents the scheduling scheme to the operator in the form of a task list. When the operator performs the examination, the module acquires the physical examination business data through the connected physical examination device or manual input, and synchronously records the process metadata associated with the data acquisition operation.

[0032] The data acquisition module 300 internally integrates a quality control engine. The engine performs verification according to the business data quality control rules as soon as the data is input, and issues an instant reminder to the operator if the data is found to be inconsistent with the rules. The module further includes a local data storage unit and a data synchronization unit. When the network connection of the mobile workstation is interrupted, all newly acquired data is stored in the local area in an encrypted manner. When the network connection is restored, the data synchronization unit uploads the locally stored data to the integrated management platform.

[0033] The profile updating module 400 is configured to dynamically update the capability profile according to the actual working performance. The module receives and processes the process metadata uploaded by one or more data acquisition modules 300. The module statistically analyzes the process metadata of each operator and device within a specified period, and calculates a comprehensive quality observation value representing the recent working quality of the operator or device. Then, the module generates and stores the updated capability profile score by using a preset updating algorithm, such as the exponential weighted moving average algorithm, in combination with the capability profile score of the previous period and the newly calculated comprehensive quality observation value.

[0034] The report generation module 500 is configured to aggregate data and generate a physical examination report containing quality assessment. The module is triggered when all physical examination items of a resident are completed. It aggregates all business data of the resident in this physical examination. For the indicators marked as abnormal, the module calls the operator capability profile, equipment capability profile and process metadata associated with the abnormal indicators when they are generated, and calculates the data confidence score of the abnormal indicators according to the preset mathematical formula. Finally, the module generates a structured physical examination report containing the physical examination results, result interpretation and data confidence score of the corresponding abnormal indicators.

[0035] The service pushing module 600 is configured to perform differentiated follow-up service linkage according to the physical examination report. The module receives the physical examination report generated by the report generation module 500. It determines the clinical severity level of each abnormal indicator in the report by querying the built-in medical knowledge base. Then, the module takes the clinical severity level and the data confidence score as inputs and locates them in a two-dimensional decision matrix. According to the positioning result, the module automatically matches and executes the preset service pushing strategy, such as sending a high-priority follow-up task to the designated family doctor work system or sending a notification with review suggestions to the resident service portal.

[0036] The accompanying drawings are referred to in order to provide a better understanding of the present application. Figure 2 , Figure 2 FIG. 1 is a schematic diagram of a physical examination integrated management system according to an embodiment of the present application. The system is configured to perform the physical examination integrated management method according to an embodiment of the present application.

[0037] The flowchart of the method starts from step S1, in which system initialization is performed in the integrated management platform, including configuring multi-dimensional quality control models and establishing initial capability profiles of operators and physical examination equipment.

[0038] Then, in step S2, the system responds to the physical examination appointment request and performs quality-aware resource scheduling based on the established capability profiles and real-time availability of medical resources to generate a specific scheduling scheme.

[0039] Subsequently, in step S3, the operator collects physical examination business data through the mobile workstation according to the scheduling scheme, and synchronously collects process metadata associated with the business data.

[0040] Then, in step S4, the system dynamically updates the capability profiles established in step S1 based on the collected process metadata using a preset updating algorithm.

[0041] After that, in step S5, the system aggregates the physical examination business data of a specified resident and combines the updated capability profiles to perform data confidence assessment on abnormal indicators, and finally generates a physical examination report with data confidence scores.

[0042] Finally, in step S6, the system performs differentiated follow-up service pushing according to the clinical severity of the abnormal indicators in the physical examination report and the data confidence score thereof.

[0043] In order to further illustrate the specific implementation mode of the method of the present application, the following will be a detailed description of each step S1 to S6 in the process.

[0044] In step S1, the system performs initialization and construction of the multi-dimensional quality control model. This step establishes the data basis and quality control benchmark for all subsequent processes.

[0045] Firstly, the integrated management platform establishes a two-way data connection with external public health information systems, hospital information systems (HIS), laboratory information systems (LIS) and picture archiving and communication systems (PACS) through its internal data platform using pre-set standard data interfaces such as HL7 (Health Level Seven) or HTTP-based application programming interfaces (API). The purpose of this connection is to achieve automatic aggregation and sharing of physical examination related data.

[0046] Next, configure the multi-dimensional quality control model in the integrated management platform. The model is composed of two types of rule sets. The first type is business data quality control rules, which function to check the validity of the data itself when the data is input. This type of rules includes but is not limited to: numerical range rules, which are used to limit the numerical value of a specific physiological indicator to be within the pre-set minimum and maximum values; logical consistency rules, which are used to check the internal logical relationship between different data items, for example, the diastolic blood pressure value must be lower than the systolic blood pressure value; format rules, which are used to ensure that data such as ID card number or phone number conforms to specific format standards; and completeness rules, which are used to ensure that all mandatory fields under a specific physical examination item are not empty.

[0047] The second type of the model is the process metadata collection rule. This rule defines the accompanying information that the system needs to record in the background synchronously in each independent physical examination data collection operation. A process metadata set M contains the following fields: the operator's unique identifier ID p , which is used to record the personnel performing the operation; the device's unique identifier ID d , which is used to record the physical examination equipment used to perform the operation; the operation start timestamp T s and the operation end timestamp T e , which are used to record the start and end time of the operation; the data modification count N m , which is an integer counter recording the number of times the data item has been modified before final submission; and the quality control event log E q , which is a structured log recording the business data quality control rules triggered in this operation, the original data value at the time of triggering and the time of triggering.

[0048] Finally, the system establishes an initial capability profile for each registered operator and each connected medical examination device. The profile is a quantitative score representing the initial service capability. For an operator p, the initial capability profile score S p，0 can be calculated according to its static professional information, a non-limiting calculation example is as follows: S p，0 = w title · V title + w exp · V exp ; wherein V title is a value mapped according to the professional technical title of the person (e.g. chief physician, attending physician); V exp is a value converted according to the length of service of the person; w title and w exp are preset weight coefficients corresponding to the title and the length of service, respectively.

[0049] Similarly, for a medical examination device d, the initial capability profile score S d，0 can be calculated according to its device profile information, a non-limiting calculation example is as follows: S d，0 = w prec · V prec + w calib · V calib ; wherein V prec is a value mapped according to the precision level in the official technical specifications of the device; V calib is a value calculated according to the time interval since the last calibration of the device, which is inversely proportional to the time interval; w prec and w calib are preset weight coefficients corresponding to the device precision and the calibration status, respectively. All initial capability profile scores are stored in the integrated management platform for subsequent steps.

[0050] In step S2, the system performs AI resource scheduling based on quality perception. This step aims to generate a scheduling scheme that takes into account both efficiency and service quality for the received medical examination appointment request.

[0051] When the resource scheduling module 200 of the integrated management platform receives a medical examination appointment request from the resident service portal, the scheduling process is activated. The module first extracts the unique identifier of the resident from the request and retrieves the health risk level C u of the resident from the data platform according to the identifier. The health risk level C uis a quantitative index representing the risk of chronic disease or the need for specific examination, which is evaluated in advance according to the historical health records of the residents. Meanwhile, the resource scheduling module 200 queries the real-time state of all operating personnel and examination equipment in the system to obtain data constituting the real-time availability of medical resources.

[0052] The core of this step is to solve an optimization problem, i.e., to find an optimal scheduling scheme T to maximize a pre-set comprehensive objective function J(T). The scheduling scheme T is a data structure whose content includes the execution sequence of one or more examination items, as well as the specific operating personnel, examination equipment and planned start time allocated to each item. The objective function J(T) is defined as: J(T) = W wait · f wait (T) + W match · f match (T) ; wherein J(T) is the comprehensive evaluation value of the scheduling scheme T; f wait (T) is an efficiency function whose value is inversely proportional to the total waiting time and process time estimated by the scheme T; f match (T) is a quality matching function whose value represents the matching degree of resource allocation and resident health needs in the scheme T; W wait and W match are pre-set weight coefficients corresponding to the importance of efficiency and quality matching degree, respectively, and W wait + W match = 1.

[0053] The value of the quality matching function f match (T) is aggregated from the quality matching degrees of each examination item included in the scheme T. For the i-th item t i in the scheme T, if operating personnel p i and examination equipment d i are allocated to it, the quality matching degree q i of the item t i is calculated as: wherein s i is the current ability profile score of the operating personnel p p，i ; s d，i is the current ability profile score of the examination equipment d i ; w p，i and w d，i are pre-set weight coefficients for the item t d,i , representing the importance of personnel skills and equipment performance in the item, respectively. For example, for a palpation item that relies on operating skills and experience, the value of w p,i will be higher than w match; for imaging examination items that rely on the precision of the equipment, w d,i . p,i .

[0054] The value of the total quality matching function f match (T) of the entire scheduling scheme T is the arithmetic mean of the quality matching degrees q i of all items in the scheme.

[0055] Since finding the optimal scheduling scheme T is a combinatorial optimization problem, the resource scheduling module 200 uses a heuristic optimization algorithm, such as a genetic algorithm or a simulated annealing algorithm, to search for a solution that maximizes the value of the objective function J(T). The search process of the algorithm takes into account the real-time state and capability profile of each available resource and iteratively generates and evaluates different scheduling schemes. After the algorithm converges, the scheduling scheme T with the maximum J(T) value output is the final determined scheme. This scheme is then dispatched by the system to the mobile workstation of the first operator designated, to start the on-site physical examination process.

[0056] In step S3, the system performs the synchronized collection of mobile on-site physical examination and process metadata. This step is performed at the front line of the physical examination by the operator using the mobile workstation. The mobile workstation is a portable computing device, such as a tablet computer or a dedicated handheld terminal, on which the client application of the data collection module 300 is installed.

[0057] At the beginning of the process, the operator logs in to his mobile workstation, and the data collection module 300 receives and displays the scheduling scheme generated in step S2. The scheme is presented in the form of a task list, guiding the operator to perform each examination on the residents in the predetermined order.

[0058] For each examination in the scheduling scheme, the operator collects physical examination business data through the mobile workstation. There are two ways to collect data: one is to manually input numerical values on the interface of the application, such as height and weight; the other is to automatically obtain measurement results from portable physical examination equipment (such as an electronic sphygmomanometer, an electrocardiograph) connected to the mobile workstation through a wireless communication protocol, such as Bluetooth.

[0059] At the same time as collecting each physical examination business data, the data collection module 300 automatically performs the synchronized collection of process metadata in the background. The collected process metadata set M corresponds one-to-one to the physical examination business data, and contains fields consistent with the rules defined in step S1, i.e., the operator's unique identifier ID p , the equipment's unique identifier ID d , the operation start timestamp T s , the operation end timestamp T e , and the data modification count N mand quality control event log E q .

[0060] The data acquisition module 300 is integrated with a quality control engine. At any moment when a physical examination service data is inputted or automatically acquired, the quality control engine is triggered. It compares the inputted service data with the service data quality control rules configured in step S1. If the comparison result shows that the data violates a rule, for example, a blood pressure measurement value exceeds the preset physiological safety range, the quality control engine will immediately generate an instant reminder on the interface of the mobile workstation. The reminder will lock the current input interface and show the specific content of the abnormality to the operator, requiring him to re-measure or confirm the data before proceeding to the next step.

[0061] In addition, the data acquisition module 300 has offline operation capability. The module continuously monitors the network connection state of the mobile workstation. When it detects that the network connection is interrupted, it automatically switches to offline mode. In offline mode, all newly acquired physical examination service data and the process metadata associated therewith are written to an encrypted local data storage unit inside the mobile workstation. When the network connection is restored, the data synchronization unit in the module is activated, which reads all unsynchronized data stored locally and uploads it to the data center of the integrated management platform through the network. After successful uploading, the corresponding data stored locally is cleared.

[0062] In step S4, the system performs dynamic updating of the capability profile and closed-loop risk early warning. The purpose of this step is to make the capability profile established in step S1 adaptive, accurately reflecting the recent performance of the operator and the physical examination equipment in actual work.

[0063] This process is automatically performed periodically by the profile updating module 400 in the integrated management platform. The module first aggregates the entire process metadata set M uploaded by each data acquisition module 300 within a specified period.

[0064] For each independent operation occurring within the period, the profile updating module 400 first calculates a single operation quality observation value based on its process metadata. Then, the module aggregates the single quality observation values of all operations of a particular operator or equipment within the period to calculate a comprehensive quality observation value Q obs，t The calculation of this value aims to quantify the accuracy, efficiency and stability of the operation.

[0065] Subsequently, the profile updating module 400 updates the corresponding capability profile score based on the comprehensive quality observation value using the Exponential Weighted Moving Average (EWMA) algorithm. Taking the operator p as an example, its capability profile score S p，t is calculated by the following formula: Sp，t = β · Q obs，t + (1 - β) · S p，t-1 ; where S p，t-1 is the competence profile score of the operator in the last cycle t-1; Q obs，t is the integrated quality observation value of the operator in the current cycle t; and β is a preset smoothing coefficient between 0 and 1, which determines the influence weight of recent work performance on the final score. A higher β value means that recent performance accounts for a larger proportion. The competence profile score update of the equipment uses the same calculation method.

[0066] After the competence profile score update is completed, the system immediately performs a closed-loop risk warning check. This check contains two trigger conditions. The first is an absolute threshold trigger, that is, when an updated competence profile score is lower than a preset minimum acceptable score S min , a warning is triggered. The second is a change rate trigger, that is, when the decline rate of a competence profile score compared to the last cycle exceeds a preset maximum allowed decline rate Δ m ax, a warning is triggered.

[0067] Once any of the warning conditions is met, the profile update module 400 will automatically generate a risk warning information. The information contains the personnel or equipment identifier triggering the warning, the current profile score, and the specific condition triggering the warning. Subsequently, the warning information is pushed by the system to the pre-configured management personnel account or management terminal interface, so that the management personnel can take timely intervention measures such as equipment calibration and personnel training.

[0068] In step S5, the system performs a hierarchical report generation with data confidence. This step is triggered after all physical examination items of a resident are completed and all data are uploaded to the integrated management platform, and its purpose is to evaluate the abnormal indicators in the physical examination results, and generate a structured report containing the evaluation information.

[0069] The process is performed by the report generation module 500. The module first aggregates all business data of the resident's physical examination. Subsequently, the module compares each item of business data with the preset normal value range to identify all abnormal indicators.

[0070] For each identified abnormal indicator, the report generation module 500 calculates a data confidence score (DCS) for it. The score aims to quantify the reliability degree of the abnormal data itself. DCS is calculated by the following preset mathematical formula: DCS = w p · F p + w d• F d + w q • F q + w c • F c ; where DCS is the final calculated data confidence score, whose value range is [0, 1].

[0071] Each factor in the formula is defined as follows: F p is the operator capability factor. The module extracts the operator identification ID p according to the process metadata associated with the abnormal indicator, and queries the capability profile score S p of the operator at the time of the operation. This factor is obtained by normalizing S p , for example, F p = S p / S max , where S max is the maximum possible value of the capability profile score in the system.

[0072] F d is the equipment capability factor. The module extracts the equipment identification ID d according to the process metadata, queries the capability profile score S d of the equipment at the time of the operation, and obtains it by normalizing it in the same way as F p .

[0073] F q is the quality log factor. This factor is calculated according to the data modification times N m and the quality control event log E q in the process metadata, and is used to quantify the stability of the operation process. A non-limiting calculation example is as follows: F q = (1 + k m · N m + k e · N e ) -1 , where N e is the number of events recorded in the quality control event log, and k m and k e are preset penalty coefficients corresponding to the modification times and the number of events, respectively. In the ideal operation without modification and without quality control events, the value of F q is 1.

[0074] F cis a data internal consistency factor. The module runs an internal consistency check engine, which logically compares the current abnormality indicator with other relevant indicators of this resident's current physical examination. For example, if a fasting blood glucose indicator is an abnormally high value, the engine will check if its glycosylated hemoglobin indicator is also in the high-risk interval. The engine outputs a consistency score according to the comparison result, for example, 1 for complete consistency, and 0 for logical contradiction.

[0075] w p , w d , w q , w c are preset weight coefficients of the corresponding factors, and the sum is 1. These weight coefficients can be configured according to different examination items to reflect the influence degree of different factors on the result credibility of the item.

[0076] After the continuous DCS value is calculated, the report generation module 500 maps it to a discrete confidence level according to a preset threshold. For example, set the thresholds T high = 0.8 and T low = 0.5, then when DCS ≥ T high , the level is “high confidence”; when T low ≤ DCS < T high , the level is “medium confidence”; and when DCS < T low , the level is “low confidence”.

[0077] Finally, the report generation module 500 generates a structured physical examination report. The report contains the business data of all physical examination items. For each abnormality indicator, its corresponding confidence level (high, medium, or low) is explicitly marked next to it. This report with data confidence information is stored by the system and delivered to the service pushing module 600 for the next step.

[0078] Referring to the accompanying drawings Figure 3 , Figure 3 is a two-dimensional decision matrix diagram according to an embodiment of the present application. In step S6, the system performs intelligent service pushing based on severity and confidence. This step receives the physical examination report with data confidence score generated by step S5, and performs differentiated follow-up service linkage according to the preset decision logic.

[0079] This process is performed by the service pushing module 600. For each abnormality indicator in the physical examination report, the module first determines its clinical severity level C sev . The module completes this operation by querying an internal medical knowledge base pre-configured by medical experts. The knowledge base contains a series of rules that map specific abnormality indicators and their values to discrete severity levels, such as critical, severe, and general.

[0080] After determining the clinical severity level C sev The service push module 600 extracts the data confidence level (high, medium, low) corresponding to the abnormal indicator from the report.

[0081] Subsequently, the service push module 600 locates the abnormal indicator in a pre-set two-dimensional decision matrix with the clinical severity level C sev and the data confidence level as two coordinates, as shown in the accompanying drawings. Each region of the matrix is pre-associated with one or more specific service push strategies. According to the landing point of the abnormal indicator in the matrix, the system automatically executes the corresponding strategy. Figure 3

[0082] A non-limiting decision logic and push strategy implementation is as follows: When the positioning result of an abnormal indicator is high clinical severity (critical or severe) and high data confidence, the system determines it as a high-confidence major health risk. The service push module 600 will immediately generate an alarm message with the highest priority. The message contains the resident's identity information, abnormal indicator details, and physical examination report, and is pushed to the working system of the family doctor to which the resident subscribes, and at the same time, an emergency follow-up task is automatically created for the family doctor in the background.

[0083] When the positioning result of an abnormal indicator is high clinical severity but medium or low data confidence, the system determines it as a suspected major health risk, but the data basis is uncertain. The service push module 600 will generate a high-priority warning message. The message will be accompanied by a clear text label when pushed to the family doctor, which reads "low data confidence, suggest review". At the same time, the system can push a message to the resident service portal, suggesting that the resident review this indicator.

[0084] When the positioning result of an abnormal indicator is general clinical severity and high data confidence, the system determines it as a routine health problem. The service push module 600 includes this indicator and its health recommendations in the routine chronic disease management process of the resident, and pushes a standard notification containing this content to the resident and his family doctor.

[0085] When the positioning result of an abnormal indicator is general clinical severity and medium or low data confidence, the system determines it as a low-risk and data-uncertain situation. At this time, the service push module 600 can execute a lowest-priority strategy, such as only recording the case without active push, or only pushing a prompt message to the resident and explicitly stating its uncertainty.

[0086] ​By executing the differentiated service pushing strategy described above, the present application completes the whole closed-loop management process from data collection to service linkage.

[0087] In order to further illustrate the synergistic working process of the technical solutions of the present application, a specific working scenario will be described below.

[0088] The scenario of the present embodiment is set as a community health service center. In the physical examination integrated management system of the center, the initial capability portraits of the operating personnel and the physical examination equipment have been pre-recorded and established. Among them, operating personnel A is a chief physician with 15 years of working experience, whose initial capability portrait score is high; operating personnel B is a nurse with 1 year of working experience, whose initial capability portrait score is medium. The system also includes physical examination equipment 1, a newly calibrated electronic sphygmomanometer, whose initial capability portrait score is high; physical examination equipment 2, an electronic sphygmomanometer approaching the calibration cycle, whose initial capability portrait score is medium. In the business data quality control rules, it has been set that the normal value range of systolic pressure is 90-140 mmHg.

[0089] Resident A, a user with a history of borderline hypertension recorded in his health record, submits an annual physical examination appointment request through the resident service portal, and the blood pressure measurement is included in his physical examination items. According to his health record, the system determines that his health risk level is medium-high.

[0090] The resource scheduling module 200 of the integrated management platform receives this appointment request. When scheduling, the system obtains the real-time availability of medical resources: operating personnel A is currently handling other matters and is expected to be available in 15 minutes; operating personnel B is currently available; both physical examination equipment 1 and physical examination equipment 2 are currently available. In the objective function of the resource scheduling module 200, since the health risk level of resident A is medium-high, the quality matching degree weight W match is set to be higher than the efficiency weight W wait . After evaluation, the algorithm determines that the quality matching degree obtained by the service provided by the combination of operating personnel A and physical examination equipment 1 (although it needs to wait for 15 minutes) is higher than that of the instant service provided by the combination of operating personnel B and any equipment. Therefore, the system generates a scheduling scheme, which specifies that resident A's blood pressure will be measured by operating personnel A using physical examination equipment 1 in 15 minutes.

[0091] The scheduling scheme is dispatched to the mobile workstation held by operator A. At the scheduled time, operator A uses the physical examination device 1 to perform a measurement on resident A. The measurement result is transmitted to the mobile workstation via Bluetooth, showing a systolic pressure of 145 mmHg and a diastolic pressure of 95 mmHg. The quality control engine embedded in the data collection module 300 immediately compares the systolic pressure value with the preset business data quality control rules and determines that it exceeds the upper limit of 140 mmHg. A prompt is immediately displayed on the interface of the mobile workstation, reading: “Systolic pressure exceeds the preset range. Please confirm or re-measure.” Operator A performs a re-measurement, and the second measurement result is a systolic pressure of 142 mmHg and a diastolic pressure of 92 mmHg. Operator A confirms and submits the value. In the process, the data collection module 300 records in the background the process metadata associated with the blood pressure measurement data, including the identification of operator A, the identification of the physical examination device 1, a record of data modification, and a log of a quality control event trigger.

[0092] After all the physical examination items of resident A are completed, the report generation module 500 is triggered. For the abnormal indicator of 142 / 92 mmHg, the module begins to calculate its data confidence score (DCS). The module retrieves the high-score capability profiles of operator A and the physical examination device 1 during operation, so the F p and F d factor values are high. Since there is one quality control event with a reasonable explanation and data modification in the operation process, the F q factor value is slightly reduced but remains at a high level. At the same time, the system checks that the blood pressure value is logically consistent with the history of borderline hypertension of resident A, so the F c factor value is high. The final calculated DCS score is 0.88, which is mapped to the “high confidence” level according to the preset threshold. In the final physical examination report generated by the system, next to the entry “Blood pressure: 142 / 92 mmHg”, it is clearly marked “Data confidence level: high”.

[0093] The service pushing module 600 receives the report. The module queries the built-in medical knowledge base and determines the clinical severity level of the blood pressure value of 142 / 92 mmHg for the resident as “severe”. Subsequently, the module locates the indicator in the two-dimensional decision matrix, and its landing point is in the (severe, high confidence) region. According to the strategy associated with this region, the system automatically generates a high-priority follow-up task and pushes a message containing the task instruction and the physical examination report to the working system of the family doctor signed by resident A, to start the subsequent health intervention process.

Claims

1. A method for integrated management of physical examinations based on smart public health, characterized in that, Includes the following steps: In the integrated management platform, a preset quality control model is built that includes business data quality control rules and process metadata collection rules, and an initial capability profile is established for operators and physical examination equipment; In response to a physical examination appointment request, based on the aforementioned capability profile and real-time availability of medical resources, a quality-aware resource scheduling is executed to generate a scheduling plan that includes designated operators and designated physical examination equipment. According to the scheduling scheme, physical examination business data is collected through a mobile workstation, and process metadata associated with the physical examination business data is collected simultaneously. Based on the process metadata, the capability profile is dynamically updated using a preset update algorithm; Aggregate the physical examination business data and combine it with the updated capability profile to conduct a data confidence assessment on abnormal indicators and generate a physical examination report with a data confidence score. Based on the clinical severity of abnormal indicators in the physical examination report and the data confidence score, differentiated follow-up service recommendations will be implemented.

2. The integrated management method for physical examinations based on smart public health, as described in claim 1, is characterized in that... The steps for constructing the preset quality control model and initially establishing the capability profile include: A data platform is established by connecting public health information systems, hospital information systems, laboratory information systems, and physical examination equipment through standard data interfaces; Configure the business data quality control rules in the data platform. The business data quality control rules are used to verify the numerical range and logical consistency of the physical examination business data. Configure the process metadata collection rules in the data platform. The process metadata collection rules are used to record the personnel identification, equipment identification, operation start and end time, number of data modifications, and logs of triggered quality control events.

3. The integrated management method for physical examinations based on smart public health as described in claim 1, characterized in that, The steps for performing quality-aware resource scheduling include: Receive physical examination appointment requests initiated by the resident service portal; The real-time busy / idle status of the operators and the physical examination equipment is obtained to determine the real-time availability of the medical resources. The matching degree between residents' health risk levels and the aforementioned capability profiles is calculated to generate a quality matching degree; Based on the combined quality matching degree and the real-time availability of medical resources, a physical examination route is planned with the goal of reducing residents' waiting time and improving service quality matching degree, and the scheduling scheme is generated.

4. The integrated management method for physical examinations based on smart public health, as described in claim 1, is characterized in that... The steps of collecting physical examination business data through a mobile workstation and simultaneously collecting process metadata associated with the physical examination business data include: The scheduling plan is dispatched to the mobile workstation held by the operator; During data collection, the quality control engine embedded in the mobile workstation performs real-time verification of the collected physical examination business data according to the business data quality control rules, and issues an immediate reminder to the operator when data anomalies are detected. In a network-free environment, the mobile workstation encrypts and stores the collected physical examination business data and process metadata locally, and synchronizes it with the integrated management platform after the network is restored.

5. The integrated management method for physical examinations based on smart public health, as described in claim 1, is characterized in that... The steps for dynamically updating the capability profile include: The integrated management platform analyzes the process metadata transmitted back from the mobile workstation; Calculate the comprehensive quality observation value for the operational quality of each operator and each medical examination device; A preset update algorithm is used to generate the capability profile score for the current period by combining the comprehensive quality observation value with the capability profile score of the previous period, thereby realizing the dynamic update of the capability profile.

6. The integrated management method for physical examinations based on smart public health, as described in claim 5, is characterized in that... The step of dynamically updating the capability profile uses an exponentially weighted moving average algorithm, which calculates the comprehensive quality observation value for the current period and the capability profile score for the previous period by weighting the average value.

7. The integrated management method for physical examinations based on smart public health as described in claim 1, characterized in that, The steps for generating a medical examination report with accompanying data confidence scores include: Once a resident has completed all their physical examinations, the integrated management platform automatically aggregates all their physical examination data. For abnormal physical examination business data, extract the associated operator capability profile score, equipment capability profile score, quality log factor and internal data consistency factor in process metadata; The four factors are weighted and summed according to preset weights to obtain the data confidence score. The confidence scores of the data are marked next to the corresponding abnormal indicators in the physical examination report in the form of high, medium and low levels.

8. The integrated management method for physical examinations based on smart public health, as described in claim 7, is characterized in that... The data confidence score is calculated using the following formula: DCS=w p ·F p +w d ·F d +w q ·F q +w c ·F c ; Where DCS is the data confidence score; F p Scoring of operator competency profiles after normalization; F d To score the normalized equipment capability profile; F q The quality log factor calculated based on process metadata; F c The internal consistency factor is used to assess the logical consistency of this data point with other inspection results; p w d w q w c These are the preset weight coefficients for the corresponding factors.

9. The integrated management method for physical examinations based on smart public health as described in claim 1, characterized in that, The steps for implementing differentiated follow-up service push include: The clinical severity level of the abnormal indicators is determined by the built-in medical knowledge base. Construct a two-dimensional decision matrix with clinical severity level and data confidence score as dimensions; Based on the position of the abnormal indicator in the two-dimensional decision matrix, a preset push strategy is matched and executed. The push strategy includes pushing emergency follow-up tasks to family doctors, pushing early warnings with review suggestions, and incorporating into the routine chronic disease management process.

10. A smart public health-based integrated health checkup management system, applied to the smart public health-based integrated health checkup management method described in any one of claims 1-9, characterized in that, include: The model building module is used to build a preset quality control model in the integrated management platform, which includes business data quality control rules and process metadata collection rules, and to establish an initial capability profile for operators and physical examination equipment. The resource scheduling module is used to respond to physical examination appointment requests, and based on the capability profile and real-time availability of medical resources, to perform quality-aware resource scheduling and generate a scheduling plan that includes designated operators and designated physical examination equipment. The data acquisition module is used to collect physical examination business data through a mobile workstation according to the scheduling scheme, and simultaneously collect process metadata associated with the physical examination business data. The profile update module is used to dynamically update the capability profile based on the process metadata using a preset update algorithm. The report generation module is used to aggregate the physical examination business data, combine it with the updated capability profile, evaluate the data confidence of abnormal indicators, and generate a physical examination report with a data confidence score. The service push module is used to perform differentiated follow-up service pushes based on the clinical severity of abnormal indicators in the physical examination report and the data confidence score.

Images