TWI881529B - System and method for managing medical quality indicator - Google Patents

Abstract

Disclosed is a system for managing medical quality indicator, comprising a target medical information database, a data retrieving module, a data analyzing module, and a data visualization module. The data retrieving module and the data analyzing module is connected to the target medical information database. The data analyzing module performs data confirmation and data exclusion for multiple pieces of structured data based on an operational definition, and performs a computation on the multiple pieces of structured data based on a viewing request to obtain a set of statistical data. The data visualization module is connected to the data analyzing module and is used for displaying the set of statistical data through a user interface.

Description

Medical quality indicator management system and method

The present invention relates to a system and method for managing medical quality indicators.

Usually, hospitals calculate medical quality indicators with the help of programs, which include obtaining data from subsystem databases, frequently retrieving data, and reviewing cases in different forms.

However, data from different sources are scattered in different subsystem databases, and high-frequency data retrieval can easily lead to system instability. However, reducing the frequency of data retrieval can also result in a lack of real-time information.

Therefore, a more effective method and system is needed to manage and retrieve data.

In one aspect, the present invention provides a medical quality indicator management system, comprising: a target medical data database, a data acquisition module, a data analysis module and a data visualization module. The data acquisition module is connected to the target medical data database. The data acquisition module responds to a review request of a medical quality indicator, or according to an update frequency of the medical quality indicator set in a data update frequency control table, obtains the required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and after data processing, stores a plurality of structured data of the medical quality indicator in the target medical data database, each structured data A patient data of a patient, a consultation index data of the patient and an indicator element data of the medical quality indicator are stored in association; and the data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the hot data interval of the required data according to the update frequency. The data analysis module is connected to the target medical data database, and performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data. The data visualization module is connected to the data analysis module and is used to display the set of statistical data through a user interface.

On the other hand, the present invention provides a medical quality indicator management method, comprising: providing a target medical data database, a data acquisition module, a data analysis module and a data visualization module, wherein the data acquisition module is connected to the target medical data database, the data analysis module is connected to the target medical data database, and the data visualization module is connected to the data analysis module; The data acquisition module responds to a review request of a medical quality indicator, or an update frequency of the medical quality indicator set in a data update frequency control table, obtains required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and stores multiple records of the medical quality indicator in the target medical data database after data processing. Structured data, each of which stores patient data of a patient, consultation index data of the patient and indicator element data of the medical quality indicator in association; the data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator according to a hot data interval of the medical quality indicator set in the data update frequency control table, or The data in the hot data interval of the required data is obtained according to the update frequency; the data analysis module performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data; and the data visualization module displays the set of statistical data through a user interface.

According to some specific embodiments of the present invention, the operational definition defines the data items required for evaluating the medical quality indicator, and the data acquisition module obtains the required data from the corresponding original medical data database based on the data items.

According to some specific embodiments of the present invention, the patient data includes: medical record number, hospitalization number, test number, name, gender, birthday or a combination thereof.

According to some specific embodiments of the present invention, the visit index data includes: outpatient visit, emergency visit, hospitalization or a combination thereof.

According to some specific embodiments of the present invention, a cold data interval is further set in the data update frequency control table. The data acquisition module first determines whether the data in the hot data interval and the cold data interval in the required data have all been stored as structured data in the target medical data database. If not, the data that has not been stored is obtained and stored as structured data in the target medical data database. If so, only the data in the hot data interval in the required data is obtained and stored as structured data in the target medical data database, or the plurality of structured data stored in the target medical data database is updated accordingly.

According to some specific embodiments of the present invention, the medical quality indicator management system further includes a work scheduling module, which is connected to the data acquisition module and is used to arrange the data acquisition module to acquire or update data for multiple medical quality indicators. The work scheduling module has a work description list and a scheduling control unit.

According to some specific embodiments of the present invention, the work description list records the update frequency, retry frequency, hot data interval and upstream work name of each of the plurality of medical quality indicators.

According to some specific embodiments of the present invention, the scheduling control unit triggers the data acquisition module to acquire or update data according to the work description list.

Other purposes and advantages of the present invention are partially described in the following description, or can be understood through the embodiments of the present invention. It should be understood that the above invention content and the following embodiments are only exemplary and explanatory descriptions, and do not limit the present invention as the scope of the patent application.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

The terms "a" or "an" as used herein, unless otherwise specified, refer to a quantity of at least one (one or more than one).

The present invention provides a medical quality index management system, comprising: a target medical data database; a data acquisition module connected to the target medical data database, wherein: The data acquisition module responds to a review request of a medical quality indicator, or according to an update frequency of the medical quality indicator set in a data update frequency control table, obtains required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and stores a plurality of structured data of the medical quality indicator in the target medical data database after data processing, each structured data stores patient data of a patient, consultation index data of the patient and indicator element data of the medical quality indicator in association; and The data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the hot data interval of the required data according to the update frequency; a data analysis module is connected to the target medical data database, which performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data; and a data visualization module is connected to the data analysis module, which is used to display the set of statistical data through a user interface.

The present invention also provides a medical quality indicator management method, comprising: providing a target medical data database, a data acquisition module, a data analysis module and a data visualization module, wherein the data acquisition module is connected to the target medical data database, the data analysis module is connected to the target medical data database, and the data visualization module is connected to the data analysis module; The data acquisition module responds to a review request of a medical quality indicator or an update frequency of the medical quality indicator set in a data update frequency control table, obtains required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and stores a plurality of structured data of the medical quality indicator in the target medical data database after data processing, wherein each structured data stores patient data of a patient, consultation index data of the patient and indicator element data of the medical quality indicator in association; The data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the hot data interval in the required data according to the update frequency; the data analysis module performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the view request to obtain a set of statistical data; and the data visualization module displays the set of statistical data through a user interface.

Each module or database as mentioned above may independently include a storage device, a processor and/or a communication chip. Examples of storage devices include hard disks, flash drives, memories, memory cards, etc. Examples of processors include integrated circuits such as microcontrollers, microprocessors, digital signal processors, application-specific integrated circuits (ASICs), logic circuits or other similar components or combinations of the above components. The communication chip may be implemented as Long Term Evolution (LTE), Worldwide Interoperability for Microwave Access (WiMAX), Wireless Fidelity (Wi-Fi), Bluetooth, Global System for Mobile Communications (GSM) or Personal Handyphone System (PHS), etc.

In order to solve the problems in the previous technology, the present invention proposes a medical quality indicator management system that can manage a large amount of real-time report data on a single platform for use in various management scenarios in hospitals. The present invention simplifies the query process and reduces management costs through the sharing of "hot and cold data intervals" and "upstream and downstream" data, and optimizes the "task scheduling index mode" to manage a large amount of real-time report data on a single platform in various management scenarios in hospitals. Hot data allows users to update only important hot data (such as data within 1 week or 1 month) every time they view a report. Cold data is not updated every time a report is viewed, and this part of the data may be (data from half a year ago or a year ago). When downstream data needs to be captured, there is no need to re-create upstream data (already existing). It only needs to obtain the required data through the existing upstream data report for detailed data capture. There is no need to repeatedly run the host database resources, which can ensure the resources and security of the host database.

For example, the surgical list shows the data of the past three years. If all the data are queried for each report, it will take about 15 minutes to search all the data. For surgical pathology analysis, it will take about 3-4 hours to search for 3 years of data.

Taking surgical pathology physician order analysis as an example, through the upstream and downstream data structure, the departmental surgery list and surgical pathology physician order analysis can be performed. There is no need to search the entire hospital surgery list every time a report is searched. In other words, each user can save about 15 minutes of time to re-establish the hospital surgery list report data each time they use the report, and the host can reduce about 330,000 search resources.

Taking surgical pathology medical record analysis as an example, most of the data in the database is created at the time of surgery. Even if there is a waiting list or modified data, most of them will not change after 7 days of the record generation. Based on this logic, the hot data interval of this report is set to 7 days, which means that only seven days of data will be updated each time the data is updated. For example, if the data is updated on 2023-10-10, only the data from 2023-10-3 to 2023-10-10 will be updated.

Taking the response rate of critical values of patient examination results as an example, doctors are required to use their official mobile phones, the hospital website (SMSOT), or the medical operating system to report and query responses within 24 hours after receiving SMS notifications of abnormal or critical values of patient examination results. In other words, as long as more than 24 hours have passed since the SMS notification was sent, regardless of whether the doctor has responded, the record will be considered unanswered. At the same time, in hospital practice, a 7-day time monitoring will be opened to monitor records of responses within 7 days even though they were not responded within 24 hours. From the above definition, the operating characteristics of the indicator can be set: cold data refers to all inspection abnormality records recorded by the system seven days ago (data older than 7 days does not need to be updated every time), hot data refers to data records within 7 days (it is necessary to confirm whether there is a response), and the update frequency of hot data is once every 8 hours. For example, if data is updated on 2023-10-10, only the hot data from 2023-10-3 to 2023-10-10 will be updated, and the cold data before 2023-10-3 does not need to be updated.

Taking the surgery list as an example, it can be divided into three reports: the hospital-wide surgery list, the department-specific surgery list, and the surgical pathology medical report analysis. The hospital-wide surgery list is for senior managers of the hospital to understand the surgical situation of the entire hospital, and will not be detailed to the detailed level of each department. The department-specific surgery list is for managers of each department to understand the situation of the entire department. It only targets the department, and each department has its own independent report. The surgical pathology medical report analysis is for the quality management unit or managers of surgeries in each department. When encountering patient safety concerns or medical quality incidents, they need to use this report to understand the details of each surgical record. The hospital-wide surgical list is the highest level, covering all surgical record lists in the hospital, but this part does not require detailed information on each surgery (surgical method, cause, anesthesia method, etc.). The specialized surgical list and surgical pathology medical order analysis are both based on the hospital-wide surgical list for detailed analysis. The hospital-wide surgical list report is the upstream data of the other two reports. The specialized surgical list and surgical pathology medical order analysis are downstream data. When downstream data needs to be captured, there is no need to re-establish upstream data (already exists). It only needs to use the existing upstream data report to obtain the required data for detailed data capture. There is no need to repeatedly run the host database resources, which can ensure the resources and security of the host database.

For example, the surgical list shows the data of the past three years. If all the data are queried for each report, it will take about 15 minutes to search all the data. For surgical pathology analysis, it will take about 3-4 hours to search for 3 years of data.

Taking surgical pathology physician order analysis as an example, through the upstream and downstream data structure, the departmental surgery list and surgical pathology physician order analysis can be performed. Every time a report is searched, there is no need to search the entire hospital's surgery list. In other words, each user can save about 15 minutes of time to re-establish the hospital's surgery list report data each time they use the report, and the host can reduce about 330,000 search resources.

Take surgical pathology medical record analysis as an example. Most of the data in this database is created at the time of surgery. Even if there are waiting lists or modified data, most of them will not change within 7 days of the record generation. Based on this logic, the hot data interval of this report is set to 7 days, which means that each time the data is updated, only the data of 7 days before the time point is updated.

The hot and cold data mechanism means that each time the report data is updated, only 7 days of data needs to be updated, which takes only about 1-2 minutes to complete. Not only can the original 3-4 hours be shortened to 1-2 minutes to produce the most timely report, but it can also reduce the number of resource searches in the host database by millions (millions of times reduced to 2,000 data searches). The original report production method is limited by host resources and sensitive data security, which results in the inability of indicator reports to present data in real time, and is very limited in its application in the medical field. The present invention uses hot and cold data and upstream and downstream data mechanisms to shorten the report data collection time that originally took several hours (approximately one million search resources) to 1-2 minutes (approximately 2,000 search resources). This enables the rapid generation of real-time reports within a short period of time to meet the needs of actual clinical settings.

In one embodiment of the present invention, the quality control center refers to the scalable medical quality indicator management system and its operation method proposed by the present invention, which modularizes each system to give the entire medical quality indicator management system scalability. According to the indicator operational definition, the digitized data in the database is automatically captured and calculated, and the indicator results are automatically transmitted to the visual management platform (SASVA) for users to view in real time.

Therefore, the intelligent indicator function includes automatic acquisition, calculation, and presentation of indicator data; real-time presentation of data (indicator update frequency is once every 8 hours); graphical user interface; and presentation of data through point-and-click methods and charts.

In one embodiment of the present invention, the present invention quickly establishes a fully automated COVID-dedicated ward situation dashboard and case tracking mechanism, successfully achieving important projects such as standardized information, automated data processing, and visual presentation, and is quickly launched in a short time and can be adjusted at any time according to clinical needs.

As shown in FIG1 , the medical quality indicator management system 100 of the present invention includes a target medical data database 110, a data acquisition module 120, a data analysis module 130, a data visualization module 140, and a work scheduling module 150. The data acquisition module 120 is connected to the target medical data database 110, the data analysis module 130 is connected to the target medical data database 110, the data visualization module 140 is connected to the data analysis module 130, and the work scheduling module 150 is connected to the data acquisition module 120. The data acquisition module 120 includes an expandable database acquisition interface, and the data is acquired by interfacing with the read interface of different medical indicator databases. The read interface at least includes SQL syntax and API communication interface. At the same time, the data type is defined, the data content is calculated, the field naming is matched, and the data type integration is converted into a complete data. The data acquisition module 120 converts the complete data after the data acquisition module to establish a data set of "patient-visit index-indicator element", and at the same time, in conjunction with the needs of the subsequent analysis module, provides an expandable definition, and provides additional descriptive information other than the element for each data definition indicator, and outputs and writes it into the target database.

The data acquisition module 120 includes the following control items: data access method control table, data update frequency control table, data operation logic control table, and data output format comparison table. The data acquisition module 120 can interface with different data sources through the data access method control table, including relational databases such as Oracle, SqlServer, DB2, etc. that support ETL of data through SQL syntax, such as extract, transform, and load operations; or file type data sources such as Excel files, TXT files, XML files, etc. The data acquisition module can interface with multiple data sources and multiple terminal settings through multiple control items. The data source can be defined as cold data or hot data according to the usage requirements, and the data update frequency and the cold and hot data range are defined through the data update frequency control table. The data acquisition module 120 determines whether the data is new data each time the data is updated. If it is new data, regardless of whether it is cold data or hot data, it is output to the target database for storage through the data operation logic control table and the data output format comparison table. If it is old data and cold data, no data processing is performed to improve the efficiency of data processing. If it is old data and hot data, it is processed through the data operation logic control table and the data output format comparison table and then output to the target database for storage.

The data acquisition module 120 customizes multiple fields to name and consolidate data according to the data type provided by the data output format comparison table, and outputs the data in a standardized format to be stored in the target database.

The data acquisition module 120 responds to a review request of a medical quality indicator, or according to an update frequency of the medical quality indicator set in a data update frequency control table, and obtains required data from one or more original medical data databases 110 according to an operational definition of the medical quality indicator, wherein the operational definition defines data items required for evaluating the medical quality indicator, and the data acquisition module obtains the required data from the corresponding original medical data database according to the data items, and processes the data. After that, a plurality of structured data of the medical quality indicator in the target medical database 110 are stored, and each structured data is associated with a patient data (medical record number, hospitalization number, test number, name, gender, birthday, diagnosis code, medication record, treatment record, test record, outpatient emergency hospitalization visit record or a combination thereof), a visit index data of the patient (outpatient, emergency, hospitalization or a combination thereof) and an indicator element data of the medical quality indicator. The visit index can be flexibly designed, and the visit index includes a doctor, date, and diagnosis.

A cold data interval is further set in the data update frequency control table. The data acquisition module first determines whether the data in the hot data interval and the cold data interval in the required data have all been stored as structured data in the target medical data database 110. If not, the data that has not been stored is obtained and stored as structured data in the target medical data database 110. If so, only the data in the hot data interval in the required data is obtained and stored as structured data in the target medical data database 110, or the plurality of structured data stored in the target medical data database 110 is updated accordingly.

The data acquisition module 120 determines whether to obtain only the data in the required data that is within the hot data interval according to a hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the required data that is within the hot data interval according to the update frequency.

Data sources include: information room database (mainly relational tables), reporting system (abnormal events, indicators of various departments, etc.), clinical indicator event collection system. The data acquisition module 120 will design data reading interfaces (such as SQL syntax, API interface, etc.) according to the sources of each medical indicator database to interface and capture data. After the data is collected, the data acquisition module 120 can set the data conversion method according to the characteristic requirements of the indicator, convert data from different sources and different formats into the same format, and then hand it over to the subsequent model for use. In other words, the modules in the subsequent stages accept the converted and unified format data regardless of the source of the data collection. The data acquisition module 120 can adjust and determine the data acquisition and update frequency of the data acquisition module 120 in a modular manner according to the unique life cycle and timeliness of each indicator, taking into account the timeliness and accuracy of the data while maintaining a reasonable data maintenance cost.

After data collection, the data acquisition module 120 will convert the original scattered data into a complete set of data according to the operational definition of the indicator through data type definition, data content calculation, field name conversion and data type integration, and then output it to the target medical data database for storage after ensuring the consistency of the data.

In terms of quality management indicators, the data acquisition module 120 can establish a specific index: all indicator elements follow the "patient", and the patient's data in the hospital is indexed by "visit" (divided into outpatient, emergency, and inpatient). Therefore, establishing a data set of "patient-visit index-indicator element" is most conducive to subsequent data analysis and application.

When the data acquisition module 120 stores the information of "patient-visit index-indicator element", it will also attach the record creation time, change time, and operator to facilitate subsequent inspection and proofreading.

The data acquisition module 120 ensures the timeliness and accuracy of the data in a staging manner, and provides additional descriptive information other than the defined indicator elements for subsequent subgroup analysis (such as department analysis, age and gender stratification analysis, or analysis down to the patient level).

The data analysis module 130 performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data. The data analysis module 130 mainly performs two major tasks: data confirmation and operation. In data collection, different data databases may have different names and recording methods for the same data due to their respective establishment requirements, or have similar recording methods for different data. Therefore, the data analysis module 130 will integrate, confirm and exclude the collected data according to the operational definition of the medical quality control indicator, and only retain the required data and ensure its correctness before performing aggregation operations. In the integrated calculation, it is necessary to clearly define how the data is divided on the time axis. Traditional indicators are mostly in the form of monthly reports, that is, most indicators are calculated once every month, but the data analysis module 130 can plan integrated calculation methods for different periods such as days, weeks, months, quarters, and years according to different time and space backgrounds and the needs of indicator operational definitions, so as to give full play to the advantages of the scalable medical quality indicator management system and its operation method, and achieve the goal of the intelligent indicator system. The presentation method of the monthly report is mainly to meet the requirements of paper printing layout, and the medical department departments are divided into multiple custom groups, which at least include the whole hospital, internal medicine department, surgical department, gynecology and pediatrics, emergency and family medicine, ENT and other groups, and each group includes multiple clinical units of medical departments at all levels.

The data visualization module 140 displays the set of statistical data through a user interface. There are various modes of visualization platforms, such as Microsoft Power BI, SAS Visual Analytics, Tableau, QlikView, etc. Different visualization platforms have different interface requirements. The data visualization module 140 mainly serves as a communication interface for the interaction between data and users. It can be designed according to the user's purpose and the characteristics of the visualization platform, and the data terminal and the visualization platform are interfaced to present the data, so as to achieve the characteristics of scalability for different usage requirements and platform system requirements. In the future, if the indicators have been informatized to a certain extent, intelligent applications can be made in the visualization module, such as abnormal value warnings, etc., and if the content of the indicators can be viewed by the user and feedback information can be collected, the system itself can be further corrected and improved in the future.

The data visualization module 140 includes a plurality of interface means, which at least includes an API communication interface. The visualization platform includes a plurality of presentation methods, which at least includes a monthly report and a dashboard mode. The dashboard presentation method is mainly in line with computer screen operation, and provides a plurality of interactive query interfaces, which at least include monthly summary, recent trends, dynamics within the last 30 days, monthly query, quarterly query, hub analysis, indicator description and other items. Each item on the dashboard includes a drill-down function, which can query the detailed data of multiple "patient-visit index-indicator elements". Because the original data has been recoded and stored in the target medical database, each user's different needs (for example: overview by year/quarter/month or drill down to each case) can use the indicator element as an index to quickly obtain each related patient-visit index-indicator element data. The data and the indicator results generated by each drilled case at different time points. At the same time, the data also contains time information, so it can be quickly sorted and classified according to year/quarter/month.

In this way, the operation of the KPI indicator is coded as a new index for the automated operation of the management indicator in the information system through the basic indicator data - hot and cold data intervals - upstream data - execution records. Each indicator has its own unique corresponding single operation index, through which the operation of the current indicator can be quickly understood and managed. Through the definition and regulation of hot and cold data intervals, unnecessary data retrieval can be effectively reduced. Taking the response rate of critical values of patient examinations as an example, if each query of annual data originally requires the mobilization of nearly one million data records, through hot and cold data intervals, regardless of whether the query is annual or N years of data, only an average of about 2,000 data records from the past 7 days will be mobilized from the original database, which can effectively improve system performance and ensure system stability.

The work scheduling module 150 arranges the data acquisition module 120 to acquire or update data for a plurality of medical quality indicators. The work scheduling module 150 has a work description list and a scheduling control unit. The work description list records the update frequency, retry frequency, hot data interval and upstream work name of each of the plurality of medical quality indicators. The scheduling control unit triggers the data acquisition module 120 to acquire or update data according to the work description list. The content of the work is to query and refresh the data in the main server (such as the medical operating system), and then upload the data to the designated visual data platform. The name of such a work item is "the name of the indicator report for data search and organization". In the data update frequency control table, you can view and manage the update frequency and hot data range of all indicators.

The default frequency of automatic task data update is every 8 hours. Real-time data can be obtained through the automatic update frequency.

If a task fails to execute, it will try again after 1 hour by default.

The number of days that the job needs to continuously update data (hot data) is set to 7 days by default. Such a hot data period helps to quickly generate report data.

If the reference base date is set to 0, it means the base date is today; if it is set to -1, it means the base date is yesterday.

If the data collection start date is set to 2020-01-01, it means that data collection will start from January 1, 2020.

The upstream job name refers to the pointer name of the upstream data (job name), and the default value is null.

In each loop execution, the task scheduling module 150 checks whether there is a task that meets the execution condition. If so, the task is executed. The condition judgment process includes:

1. Load basic settings: Read the above work description list, load the work settings and create a work status file to record the work execution status. At the same time, perform basic data checks, including: data type definition (numeric type, Boolean type, date and time type), check the upstream indicator name (if there is a specified upstream indicator, check whether the upstream database exists; if not, first perform the upstream database creation work), and check the necessary parameters (check the indicator update frequency, retry frequency, time unit and other basic settings. If not filled in, the work will alarm abnormalities and record them in the work status file).

2. Calculate the data requirement interval: Calculate the time when the work should start and the data date interval that needs to be collected. According to the hot data period, the data collection start date and the date base offset days, obtain the data date that should exist in the target medical database.

3. Create a to-do list: Compare the data date in the target medical data database with the data date calculated above, and create a to-do list (a list of data dates that really need to be retrieved).

4. Perform data acquisition and record execution status: Perform data acquisition according to the above-mentioned agency list. At the same time, record the time of data acquisition, the time to calculate the start of the next loop, and whether the acquisition work is successfully completed, etc., and record it in the work status table.

5. Loop: If the system is not restarted, after completing the data acquisition and recording of the execution status, the system will wait until the time reaches the next expected work execution time, and then perform data acquisition and recording. If the system is restarted, it will start again from the step of loading basic settings.

For example, if there is work item 1, when the system starts to execute, it will first check whether the indicator update frequency, retry frequency and time unit are filled in. If there is no problem, it will proceed to the next step.

The system compares the data date in the target medical database with the date base offset days of 0 and the data collection start date of January 1, 2019 (2019-01-01), determines which data does not exist, and then creates a to-do list and performs data acquisition. If there are missing settings, the system will alarm and stop execution; if the settings are correct, a to-do list will be created.

Data is retrieved according to the to-do list. If the to-do list is empty, it means that all data already exists. Then proceed to the next step of hot data update. The hot data interval is set to 2, and the system will only update today's and yesterday's data each time (data within 2 days). Determine whether the system execution is successful. If successful, re-update the hot data (data within 2 days) every 8 hours. If the system execution fails, wait 2 hours and try to update the hot data (data within 2 days) again. Calculate the next start time, and then repeat the steps to determine whether the system execution is successful.

For example, there is now work item 3. If the indicator is executed successfully, it will be updated every 24 hours. If the execution fails, it will be re-executed after two hours. The hot data interval is 2, and only the data of the last two days will be updated each time. The date base offset days is 0, and the data collection start date is 2022-01-01. It means that the data in the database will have all the data from 2022-01-01 to today. The upstream data is work item 1, so when the data of work item 3 is captured, the data of work item 1 will be confirmed first.

For example, if there is work item 2, when the system starts to execute, it will first check whether the indicator update frequency and retry frequency and time unit are filled in. If there is no problem, it will proceed to the next step. If there is work item 1, execute the work of work item 1 first and check whether the data already exists.

The system sets the date base offset days to 0, and the data collection start date is 2022-01-01. Compare the data dates in the target medical data database to determine which data does not exist, and then create a to-do list and perform data acquisition. If there are missing settings, the system will alarm and stop execution. If the settings are correct, create a to-do list. Perform data acquisition based on the above to-do list. If the to-do list is empty, it means that all data already exists. Then proceed to the next step of hot data update. The hot data interval is set to 2, and the system will only update today's and yesterday's data each time (data within 2 days). Determine whether the system execution is successful. If successful, the hot data will be updated every 24 hours (data within 2 days). If the system fails to execute, it will try to update the hot data (data within 2 days) again after 2 hours. It will calculate the next start time and then repeat the steps to determine whether the system execution is successful.

Without using hot data and upstream data, Work Item 1 will capture data every 8 hours, and each time it needs to start from 2019. Work Item 2 will capture data every 24 hours, and each time it needs to first obtain the data of Work Item 1 starting in 2022, and then capture the data of Work Item 2 starting in 2022. Work Item 3 will capture data every 8 hours, and each time it needs to first obtain the data of Work Item 1 starting in 2023, and then capture the data of Work Item 3 starting in 2023. All data from the data start date to the present needs to be captured every time the data is updated.

When using hot data and upstream data, Work Item 1 only needs to be captured from 2019 for the first time (when there is no data in the destination medical database). After that, each update only needs to capture data for the missing data (to-do list). If there is no missing data, only hot data (2 days) needs to be captured each time. Work Item 2 and Work Item 3 both directly import the upstream data (Work Item 1 in the destination medical database) data, without re-establishing all the data of Work Item 1. For work items 2 and 3, only the first time (when there is no data in the target medical database) do you need to start capturing data from the set data start date (2022-01-01 for work item 2 and 2023-01-01 for work item 3). Each subsequent update only requires capturing data for the missing data (to-do list); if there is no missing data, only hot data (2 days) needs to be captured each time.

In order to manage the operation of multiple KPI indicators, the present invention provides a work management system and method that can be used to handle complex workflows and perform different types of work. The invention makes work management more convenient, controllable and efficient, while improving the performance and reliability of applications. This invention has broad application prospects and has important value in the fields of automation, data processing, workflow management, etc. Data from different sources are captured and merged into a data structure consisting of patient, visit index and indicator elements. This data structure integrates different related information and stores it in the target medical database for subsequent use.

The status and settings of all KPI indicators are managed through an index consisting of basic indicator data, hot and cold data ranges, upstream data, and execution records. The relevant information of different KPI indicator tasks is integrated through this index, including task name, status, settings, etc., so as to provide a unified index to manage tasks. At the same time, calculations are performed based on the settings and status of each KPI indicator task, taking into account the priority, dependency relationship, and other factors of the task to determine the best execution time and ensure that the task can be executed as expected. The status of the task is updated through the execution record and a Boolean value is returned, indicating whether the task is successfully executed and the updated task status.

Example 1: Scalable hospital TCM quality indicator data collection and management system

The scalable hospital traditional Chinese medicine quality indicator data collection and management system includes a data acquisition module, which includes an scalable database acquisition interface, and is connected to the read interface of different medical indicator databases to acquire data. The read interface at least includes SQL syntax and API communication interface. At the same time, at least define the data type, data content calculation, field naming comparison and data type integration, and convert it into a complete data; a data storage module, which converts the complete data after the data acquisition module to establish a "patient-visit index-indicator element" data set, and at the same time, in conjunction with the needs of the subsequent analysis module, provides extended definitions, and provides additional descriptive information other than the elements for each data definition indicator, outputs and writes it into the target database; a data analysis module, which performs analysis after confirming and excluding the data, at least including department analysis, age and gender analysis, hierarchical analysis, patient level analysis, etc. This module can perform integrated calculations and data time axis segmentation at the same time, and can perform integrated calculations according to the needs of different time and space backgrounds and indicator operation definitions; a data visualization module, which receives data transmitted by the data push module and presents the data through a visualization platform.

Taking the inpatient mortality rate as an example, its operational definition is as follows. The scope of data collection for the indicator includes (1) acute general hospital beds, (2) special hospital beds, intensive care beds, burn beds, infant beds, infant beds and subacute respiratory care beds (RCC), burn intensive care beds, bone marrow transplant beds and isolation beds. Indicator elements: The denominator of this indicator is the number of patients discharged (including transfers) from all the aforementioned beds during the monitoring month. Each patient should be classified into one of the following statuses: (1) discharged due to death, (2) discharged due to critical illness: the patient is diagnosed as critically ill and on the verge of death by a physician and the patient's family requests automatic discharge, (3) discharged due to violation of medical instructions (AAD), (4) discharged directly from the hospital (MBD) Exclusion rules: The "Filter by indicator case conditions" function can be used to switch to exclude cases from the nursery (department code: NB), psychiatric ward (department code: PSY, GPSY, PSYD), and hospice ward (department code: HOSP). Calculation formula: Inpatient mortality rate (including self-discharge due to critical illness) = number of deaths (including self-discharge due to critical illness) / total number of discharges Inpatient mortality rate (excluding self-discharge due to critical illness) = number of deaths (excluding self-discharge due to critical illness) / total number of discharges

As mentioned above, the data acquisition module includes 4 control items: data access method control table, data update frequency control table, data calculation logic control table, and data output format comparison table. The data acquisition module can interface with different data sources through the data access method control table, including relational databases such as Oracle, SqlServer, DB2, etc.) to support ETL of data through SQL syntax, such as extract, transform, and load operations; or file type data sources such as Excel files, TXT files, XML files, etc. The data acquisition module can interface with multiple data sources and multiple terminal settings through multiple means described in item 2. Taking the operational definition of inpatient mortality as an example, the data acquisition module will acquire at least four data sources including (1) patient basic data file, (2) patient hospitalization registration database, (3) physician data file, and (4) patient hospitalization history database to meet the requirements of the operational definition.

Data sources can be defined as cold and hot data according to usage requirements, and the data update frequency and cold and hot data ranges can be defined through the data update frequency control table. The following pseudocode method is used to determine whether the data is new data each time the data is updated. If it is new data, regardless of whether it is cold data or hot data, it is output to the target database for storage through the data calculation logic control table and the data output format comparison table. If it is old data, it is further determined whether it is cold data or hot data. If it is old data and cold data, data processing is not performed to improve data processing efficiency. If it is old data and hot data, it is processed through the data calculation logic control table and the data output format comparison table and then output to the target database for storage. Taking the mortality rate of inpatients as an example, the inpatient mortality data over a month should not change much in actual business, so the hot data interval is defined as the data within 30 days before the data acquisition date. Data over 30 days is defined as cold data.

The pseudocode example is as follows: Cold data range = 2020-01-01 ~ previous 30 days Hot data range = previous 30 days ~ today Required data date list = [day for day in date_range(2020-01-0, today)] Existing data date list = Read and list database existing file dates for data date in required data date list: if data date not in existing data date:    data type = new data    data needs to be processed and stored elif data date in existing data date:    data type = old data    if data date in cold data range:       data is not processed    elif data date hot data range:       data needs to be processed and stored     else:  raise Exception()

The data output format comparison table divides the data format definitions into three types: numeric, datetime, and string. All input data formats are mapped to the three data types using the following pseudocode method.

The pseudocode examples are as follows: Numerical data type list = ['n', 'number', 'float64'…...] Date data type list = ['d', 'dt', 'datetime', 'datetime64[ns]'……] String data type list = ['s', 'str', 'string', 'object'……] if input data format in date data type list: Output data format = pd.to_datetime (input data) à output in date data format elif input data format in numerical data type list: Output data format = pd.to_numeric (input data) à output in numeric data format elif input data format in string data type list: Output data format = input data.astype('object').fillna('') à string data format output else:    raise Exception()

The data storage module will name and consolidate the data in multiple custom fields according to the data type provided by the data output format comparison table, and output the data in a standardized format to the target database. The format of the data output format comparison table is "data field name = data type code", and at the same time, a data set structure of "patient-visit index-indicator element" is established.

Take the mortality rate of inpatients as an example: it is necessary to obtain the date of each hospitalization and discharge, the department of each hospitalization, the bed number, and the status of each discharge (death, survival, or dying, etc.) in order to calculate and exclude according to the operational definition. Therefore, the data set structure and instance of "patient-visit index-indicator element" are as follows: Patient data: at least should include medical record number, hospitalization number, name, gender, and birthday. Visit index: at least should include outpatient, inpatient, and emergency. Indicator elements: at least should include hospitalization date, hospitalization time, discharge date, discharge time, discharge status, attending physician code, attending physician name, department code, ward number, and bed number. Additional information: including data creation time, data change time, and operator.

Table 1: Dataset structure Dataset Architecture Data Field Name Data Type Code Example (a complete data record) patient Medical record number String(s) 10****40 Hospital Number String(s) 28****96 Name String(s) *** gender String(s) male Birthday Date(d) 19**/**/** Consultation Index Consultation Index String(s) Outpatient admission Pointer elements Date of hospitalization Date(d) 2022/08/29 Length of hospital stay Date(d) 16:04:00 Discharge date Date(d) 2022/08/30 Discharge time Date(d) 11:05:00 Discharge status String(s) Leaving hospital Discharge status String(s) Return Home Discharge details status String(s) Return home and continue to our outpatient clinic Attending physician code String(s) DOC2***** Name of attending physician String(s) appoint** Department Code String(s) GNS Ward No. String(s) A1** Bed number Value(n) 1** Pointer elements String(s) Mortality Additional Information Data creation time Date(d) 2022/08/30 Data change time Date(d) ---- Operator String(s) QAC8***

The data analysis module can perform data confirmation and exclusion functions for multiple complete data records based on multiple conditions, which at least includes conditional confirmation of indicators and exclusion of missing values. It can also perform multiple integrated operations, which at least includes integrated operations based on daily, weekly, monthly, quarterly, and annual periods. Taking the mortality rate of inpatients as an example, the data analysis module can perform the following operations on indicators: (1) Perform integrated operations by time period (year, quarter, month, week, day), etc., for example: present the mortality rate of a certain year, the mortality rate of a certain month, the mortality rate of inpatients in the past two years (2020~2022), etc. (2) Perform integrated operations by department, for example: present the mortality rate of inpatients in the Department of Internal Medicine or the mortality rate of inpatients in the Department of Orthopedics and Cardiology. (3) Perform integrated operations by attending physician, for example: the mortality rate of inpatients treated and cared for by a certain physician. (4) Perform multiple integration operations on the above different conditions, for example: the mortality rate of inpatients in the Department of Orthopedics in 2022 (multiple integrations by time and department), the mortality rate of inpatients treated and cared for by Doctor A in the past two years and presented by month (multiple integrations by time and doctor)

The following table shows an example of the integration operation of the ENT head and neck department in a certain year and month (multiple integrations are performed based on time and department).

Table 2: Integrated calculation results Year moon Department Discharge Department Number of deaths (excluding critically ill AAD) Mortality rate (excluding critically ill AAD)* Number of deaths (including critically ill AAD) Mortality (including critically ill AAD)* Number of discharged patients 20** ** ENT Head and Neck Rhinology *** *.**% *** *.**% *** 20** ** ENT Head and Neck Laryngology *** *.**% *** *.**% *** 20** ** ENT Head and Neck otology *** *.**% *** *.**% ***

The data visualization module includes multiple interface methods, which at least include an API communication interface. The visualization platform includes multiple presentation methods, which at least include monthly reports and dashboard modes. The presentation method of the monthly report is mainly based on the requirements of paper printing layout, and the medical department departments are divided into multiple customized groups, which at least include the whole hospital, internal medicine department, surgical department, gynecology and pediatrics, emergency and family medicine, ENT and others. Each group includes multiple clinical units of various levels of medical departments. Example: The internal medicine department group includes first-level units such as the internal medicine department and the thoracic department, and the internal medicine department includes second-level units such as general internal medicine and nephrology. The dashboard is presented in a way that is compatible with computer screen operation and provides multiple interactive query interfaces, which at least include monthly summary, recent trends, dynamics within the last 30 days, monthly query, quarterly query, hub analysis, indicator description, etc. Each item on the dashboard includes a drill-down function, which can query the detailed data of multiple "patient-visit index-indicator elements". Taking the mortality rate of inpatients as an example, the drill-down function provides detailed data for querying each data field in the data set table, allowing users to trace back the detailed records of each data.

Example 2: Specific KPI - Patient test critical value response rate

To ensure "hospital medical quality and patient safety", our hospital requires that doctors should report and respond to inquiries within 24 hours using their official mobile phones, the hospital website (SMSOT), or the medical operating system after receiving SMS notifications of abnormal or critical test results from patients.

When the laboratory completes the test report, if the test result is abnormal, the data will be recorded in the Abnormal Report Registration (ARR); then through a one-to-many method, it will correspond to the patient's attending physician, resident physician, ordering physician and other doctors, and the data will be recorded in the Abnormal Report Notification (ARN). Then, a text message will be sent through the SMS sending system to the official mobile phones of all doctors recorded in the ARN, and the time and content of the SMS will be recorded in the SMS paging log file (PHSLOG). Finally, after the doctor receives the SMS, the content and time of the reply SMS through the official mobile phone will be recorded in the reply SMS content file (PHSREPLY). If the doctor responds through the hospital website (SMSOT) or medical operating system, it will be recorded in the abnormal value notification record (ORDREPLY) data file.

To monitor the response rate of critical values of patients' examinations, data must be obtained from at least five databases: Abnormal Report Registration (ARR), Order Abnormal Report Notification (ARN), PHSLOG, PHSREPLY, and ORDREPLY. If you want to know the doctor's and patient's data, you need to add at least two more databases: doctor's basic data file and patient's basic data file.

That is to say, our hospital has an average of more than 10,000 abnormal reports per month. Each report corresponds to at least 3 doctors, and each doctor has three possible responses, which means that the amount of data is 90,000. If it is expanded to quarterly or annual data, the amount of data retrieved each time will exceed several million. If it is further classified according to the department to which the doctor belongs and the basic information of the patient, at least 2 more databases will be connected in series, and the amount of data will also increase several times. Such a large amount of data, if frequently retrieved from the information system, will cause a high burden on the operation of the system. Therefore, most of them currently calculate the data of the current month once a month to maintain the stability of the system. However, many patients with abnormal test results may suffer irreversible consequences if they are not properly treated within a short period of time. Therefore, in order to ensure medical quality and patient safety, this indicator should be able to reflect the current situation in real time rather than an outdated indicator that is only counted once a month.

According to the definition of the indicator element, data is extracted from the original database and converted into the content required by the indicator element to establish the patient-visit index-indicator element. That is, data is extracted from the original medical database through the patient's medical record number (such as the data obtained from the above 7 databases), and the patient's medical record number is used as the main combination with the visit index (doctor, date, examination order number) and other record data, and it is linked to the indicator element (test abnormal value response rate) according to the examination item order number and this file is stored in the target medical database.

The original data was scattered across seven original medical data databases, and each database could not be connected from beginning to end, and needed to be connected through different indexes. This also means that every query needs to search all the contents of the seven databases from beginning to end, which also limits the system performance and prevents frequent queries in real time. After the test order is issued, REQNO will be generated, and this index can correspond to the test abnormal value registration table (ARR). The test abnormal value registration table (ARR) is then mapped to the test abnormal value notification table (ARN) through ARRID, and then mapped to the abnormal position response record table (ORDREPLY) through ARNID. The patient basic data file and the doctor basic data file correspond to the test order through HISID as an index. The inspection form is linked to the SMS notification record through REQNO, and then mapped to the SMS reply content file through MSGID as an index.

According to the present invention, the data of the above 7 databases are reorganized and concatenated, and then recoded and stored in the target medical database. The data will be as follows, using the patient-visit index-indicator element as a new index code for the data in the target medical database: [Patient] Medical record number (Patient basic data file)-Test order number (Patient basic data file)-[Visit index] Test date (Patient basic data file)-Doctor (Doctor basic data file)- [Indicator elements] Test report results (abnormal value registration table) - report date (abnormal value registration table) - abnormal order number (abnormal value registration table) - abnormal content (abnormal value notification table) - SMS notification time (SMS paging record file) - SMS content (SMS paging record file) - doctor's reply method (SMS reply content file/abnormal value reply record table) - reply content (SMS reply content file/abnormal value reply record table) - reply time (SMS reply content file/abnormal value reply record table) - reply time difference.

At the same time, because the original data has been recoded and stored in the target medical database, even if each user has different needs (for example, overview by year/quarter/month or drilling down to each case), they can use the indicator element as an index to quickly obtain each related patient-visit index-indicator element data, and the data represents the indicator results generated by each drilled case at different time points. At the same time, the data also contains time information, so it can be quickly sorted and classified according to year/quarter/month.

There are many KPI indicators required in hospital management, and each indicator has its own particularity, which determines the update frequency of each indicator and its correlation with other indicators. Taking the response rate of critical values of patient examinations as an example, the hospital can stipulate that after receiving the SMS notification of abnormal or critical values of patient examination results, the doctor should use the official mobile phone, the hospital website (SMSOT), or the medical operating system report to query and reply within 24 hours. In other words, as long as the SMS notification is sent more than 24 hours later, regardless of whether the doctor has replied, the record will be regarded as unanswered. At the same time, in hospital practice, a 7-day time monitoring will be opened to monitor the records of replying within 7 days even if there is no reply within 24 hours.

From the above definition, we can set the operating characteristics of the indicator: 1.   Cold data refers to all inspection abnormality records recorded by the system so far (more than 7 days, no need to update every time). 2.   Hot data refers to data records within 7 days (must confirm whether there is a response). 3.   The update frequency of hot data (once every 8 hours).

The present invention uses the basic data of the indicator - hot and cold data intervals - upstream data - execution records as a new index code for the automatic operation of the management indicator in the information system: [Basic data of the indicator] Name of the indicator - Whether the indicator is activated - Current activation status [Hot and cold data intervals] Cold data interval - Hot data interval - Update frequency - [Upstream data] Whether there is an upstream indicator - Upstream indicator name [Execution record] Last start execution time - Last end execution time - Last execution result - Next start execution time - Other notes

Each indicator has its own unique corresponding single operation index, through which the operation of the current indicator can be quickly understood and managed. Through the definition and regulation of hot and cold data intervals, unnecessary data retrieval can be effectively reduced. For example, if the original query of annual data requires the retrieval of nearly one million data records each time, through the hot and cold data intervals, regardless of whether the query is annual or N years of data, only the data of the past 7 days will be retrieved from the original database, an average of about 2,000 data records, which can effectively improve system performance and ensure system stability.

In order to manage the operation of multiple KPI indicators, the present invention provides a work management system and method that can be used to handle complex workflows and perform different types of work, making work management more convenient, controllable and efficient, while improving the performance and reliability of applications. The present invention has broad application prospects and has important value in the fields of automation, data processing, workflow management, etc. For example: (1) Capturing and merging data from different sources into a data structure of patient-visit index-indicator elements. This data structure integrates different related information and stores it in the target medical database for subsequent use. (2) Managing the status and settings of all KPI indicators through the indexing method of indicator basic data-cold and hot data intervals-upstream data-execution records. The index integrates the relevant information of different KPI indicator tasks, including the task name, status, settings, etc., to provide a unified index to manage tasks. At the same time, calculations are performed based on the settings and status of each KPI indicator task, taking into account the priority, dependency relationship and other factors of the task to determine the best execution time and ensure that the task can be executed at the expected time. The status of the task is updated through the execution record and a Boolean value is returned, indicating whether the task is successfully executed and the updated task status.

The medical quality indicator management system and method provided by the present invention can efficiently share information in different reports, thereby saving a large amount of computing resources, and appropriately optimize computing resources to increase the frequency of automatically updating reports, thereby significantly reducing the cost of managing reports.

100   Medical quality indicator management system 110   Medical data database 120   Data acquisition module 130   Data analysis module 140   Data visualization module 150   Work scheduling module

FIG1 is a block diagram of the medical quality indicator management system of the present invention.

100: Medical quality indicator management system

110: Medical Data Database

120: Data acquisition module

130:Data analysis module

140:Data visualization module

150: Task Scheduling Module

Claims (7)

A medical quality indicator management system, comprising: a target medical data database; a data acquisition module, connected to the target medical data database, wherein: The data acquisition module responds to a review request of a medical quality indicator, or an update frequency of the medical quality indicator set in a data update frequency control table, obtains required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and stores multiple structured data of the medical quality indicator in the target medical data database after data processing, each structured data stores patient data of a patient, consultation index data of the patient and indicator element data of the medical quality indicator in association; and The data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the hot data interval of the required data according to the update frequency; A data analysis module, connected to the target medical data database, performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data; and A data visualization module, connected to the data analysis module, is used to display the set of statistical data through a user interface, The operational definition defines the data items required to evaluate the medical quality indicator, and the data acquisition module obtains the required data from the corresponding original medical data database based on the data items. The data update frequency control table further sets a cold data interval. The data acquisition module first determines whether the data in the hot data interval and the cold data interval in the required data have all been stored as structured data in the target medical data database. If not, the data that has not been stored is obtained and stored as structured data in the target medical data database. If so, only the data in the hot data interval in the required data is obtained and stored as structured data in the target medical data database, or the plurality of structured data stored in the target medical data database is updated accordingly. The system as described in claim 1, wherein the patient data includes: medical record number, hospitalization number, test order number, name, gender, birthday or a combination thereof. The system as claimed in claim 1, wherein the visit index data includes: outpatient visit, emergency visit, hospitalization or a combination thereof. The system as described in claim 1 further includes a work scheduling module connected to the data acquisition module and used to arrange the data acquisition module to acquire or update data for multiple medical quality indicators. The work scheduling module has a work description list and a scheduling control unit. A system as described in claim 4, wherein the work description list records the update frequency, retry frequency, hot data interval and upstream work name of each of the plurality of medical quality indicators. A system as described in claim 4 or 5, wherein the scheduling control unit triggers the data acquisition module to acquire or update data based on the work description list. A medical quality indicator management method, comprising: Providing a target medical data database, a data acquisition module, a data analysis module and a data visualization module, wherein the data acquisition module is connected to the target medical data database, the data analysis module is connected to the target medical data database, and the data visualization module is connected to the data analysis module; The data acquisition module responds to a review request of a medical quality indicator, or according to an update frequency of the medical quality indicator set in a data update frequency control table, obtains required data from one or more original medical data databases according to an operational definition of the medical quality indicator, and after data processing, stores a plurality of structured data of the medical quality indicator in the target medical data database, each structured data stores patient data of a patient, consultation index data of the patient and indicator element data of the medical quality indicator in association; The data acquisition module determines whether to obtain only the data in the hot data interval of the medical quality indicator set in the data update frequency control table, or to obtain the data in the hot data interval of the required data according to the update frequency; The data analysis module performs data confirmation and data exclusion on the plurality of structured data according to the operational definition, and performs an operation on the plurality of structured data according to the viewing request to obtain a set of statistical data; and The data visualization module displays the set of statistical data through a user interface, The operational definition defines the data items required to evaluate the medical quality indicator, and the data acquisition module obtains the required data from the corresponding original medical data database based on the data items. The data update frequency control table further sets a cold data interval. The data acquisition module first determines whether the data in the hot data interval and the cold data interval in the required data have all been stored as structured data in the target medical data database. If not, the data that has not been stored is obtained and stored as structured data in the target medical data database. If so, only the data in the hot data interval in the required data is obtained and stored as structured data in the target medical data database, or the plurality of structured data stored in the target medical data database is updated accordingly.

Images