WO2022166859A1 - 一种医疗数据治理系统 - Google Patents

一种医疗数据治理系统 Download PDF

Info

Publication number
WO2022166859A1
WO2022166859A1 PCT/CN2022/074873 CN2022074873W WO2022166859A1 WO 2022166859 A1 WO2022166859 A1 WO 2022166859A1 CN 2022074873 W CN2022074873 W CN 2022074873W WO 2022166859 A1 WO2022166859 A1 WO 2022166859A1
Authority
WO
WIPO (PCT)
Prior art keywords
governance
data
node
medical
model
Prior art date
Application number
PCT/CN2022/074873
Other languages
English (en)
French (fr)
Inventor
张建中
王列
Original Assignee
无锡慧方科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 无锡慧方科技有限公司 filed Critical 无锡慧方科技有限公司
Publication of WO2022166859A1 publication Critical patent/WO2022166859A1/zh

Links

Classifications

    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

  • the invention relates to the field of medical technology, in particular to a medical data management system.
  • each hospital has its own medical information system.
  • the composition of the medical information system is very complex, and the amount of raw medical data accumulated in daily operation is very large.
  • the original medical data generated by different hospitals and different medical information systems often do not follow the same data standards. Therefore, it is usually necessary to perform corresponding data governance operations on the original medical data to mine the required data before it can be used.
  • the current method of data governance is usually to establish a corresponding data governance model for data application scenarios to achieve corresponding functions. Due to the high complexity of the original medical data, the complexity of the data governance model is often high, which leads to modeling difficulties. It is large and error-prone, making it difficult to guarantee the efficiency and quality of data governance.
  • a medical data governance system including:
  • Model building module the model building module is used to create several governance nodes corresponding to business requirements, and establish the connection relationship between different governance nodes to obtain a data governance model corresponding to the governance process of business requirements; each governance node has corresponding node type and call a governance function in the governance function library.
  • the node type includes input node, intermediate node and output node. The data output terminal of the input node is connected to the corresponding intermediate node, and the data input terminal of the output node is connected to the intermediate node. Each The data input end and data output end of the intermediate node are respectively connected to other governance nodes;
  • Model execution module the model execution module is used to obtain the data to be managed and input the data input terminal of the input node of the data governance model corresponding to the business requirements, and schedule and execute each governance node in turn.
  • Each governance node uses the medical knowledge base to call the corresponding
  • the governance function processes the data input from the data input terminal, and outputs the processed data through the data output terminal until the output node outputs data through the data output terminal after the execution of the output node to generate the processed data corresponding to the business requirements;
  • governance process management module the governance process management module is used to record the data governance process data of each governance node and feed back an error message when the governance node performs an error.
  • the medical data governance system is used to perform data governance on the data to be governed to obtain post-governance data corresponding to N business requirements
  • the model building module is used to build a standardized governance model and N different
  • the in-depth governance model corresponding to the business requirements, the standardized governance model and an in-depth governance model constitute the data governance model corresponding to the business requirements, N ⁇ 2;
  • the model execution module is used to input the data to be managed into the standardized governance model, and input the data output by the standardized governance model into N deep governance models respectively, and each deep governance model outputs the post-governance data corresponding to the business requirements.
  • a standardized model building unit which is used to create a standardized governance model corresponding to the standardized governance process shared by N business requirements;
  • the deep governance model building unit is used to create N deep governance models, each of which corresponds to an in-depth governance process required by a business.
  • the medical stage division module the medical stage division module is used to determine the medical stage division result corresponding to the business requirement, and the medical stage division result includes several medical stages continuously divided according to time;
  • the business data screening module filters out the data of the index items whose data generation time is within the corresponding target time period from all the data output by the data governance model corresponding to the business requirement, and obtains the post-governance data corresponding to the business requirement;
  • the target time period corresponding to each indicator item matches the business requirement, different indicator items under the same business requirement have the same or different target time period, and the same indicator item has the same or different target time period under different business requirements,
  • the target time period is a medical stage in the medical stage division result corresponding to the business requirement.
  • the medical stage division module is used to determine the data extraction instructions of each medical stage corresponding to the business needs, and extract the data of the index items in each medical stage according to the data extraction instructions, and according to the data in each medical stage.
  • the data generation time of the index item determines the time period of each medical stage to obtain the medical stage division result.
  • the medical stage division module is used to determine the start time and the end time of each medical stage according to the data generation time of the index item in each medical stage, and adjust the current medical stage according to the start time of the next medical stage. At the end time, the time period of each medical stage is continuous and not repeated, and the medical stage division result is obtained.
  • governance process management module is also used for executing errors in the governance node, and when it is detected that the corresponding execution error data is not stored in the problem database, the corresponding execution error data is stored in the problem database, and the problem database uses for updating the medical knowledge base.
  • the intermediate node in the data governance model includes a processing node and/or a sorting node
  • the processing node includes a data input terminal and a data output terminal
  • the sorting node includes a data input terminal and a plurality of data output terminals.
  • the sorting node divides the input data into multiple channels of data with different medical index attributes and outputs them through the data output terminal respectively.
  • one data output terminal of each other governance node except the output node is connected to one or more other governance nodes as its lower node, when one data output terminal of the governance node is connected to multiple lower layers.
  • the governance node outputs the data stream output by the data output terminal to each lower node respectively;
  • each governance node is connected to one or more other governance nodes as its upper-level node.
  • the data input terminal of the governance node inputs
  • the data includes the output data of all upper-level nodes.
  • the present application discloses a medical data governance system.
  • the medical data governance system uses a plurality of governance nodes that call different governance functions to build and form a data governance model to govern raw medical data according to business requirements.
  • the governance function has simple functions, low complexity, and is easy to write, which can effectively reduce the difficulty of modeling and improve the efficiency of data governance.
  • each governance node can record the data of the data governance process and report errors, so that the governance process can be evaluated, and errors can be fed back in time, which is also easy to implement. Issue tracking, thereby contributing to improving the quality of data governance and feedback improvements to the governance process.
  • each governance node When building a data governance model, each governance node directly calls the functions in the governance function library to implement corresponding functions, thereby improving the reuse rate of governance functions, reducing repeated operations in the modeling process, and further improving data governance. s efficiency.
  • the pre-defined governance functions are also more normative and standard, which ultimately improves the quality of data governance.
  • the system When building a data governance model, the system builds a standardized governance model for the same governance operation part of multiple different business requirements, and builds its own in-depth governance model for different personalized demand parts, which can meet different business needs. On the basis of reducing the repeated operation process, improve the efficiency of data governance.
  • the system sets up a medical stage division module and a business data screening module in the system to further screen the data that has completed data governance from the perspective of time attributes, so that it can be more realistic. Use in-demand, practical data.
  • FIG. 1 is a schematic diagram of the data processing flow of each module in the medical data governance system disclosed in the present application.
  • FIG. 2 is a schematic diagram of the connection relationship between governance nodes constructed during the execution of the medical data governance system of the present application.
  • FIG. 1 discloses a medical data management system, please refer to FIG. 1, the system includes:
  • Model building module The model building module is used to obtain a data governance model corresponding to the governance process of business requirements.
  • the business requirements in this application are the requirements generated during the process of data mining and analysis for the data to be governed.
  • the business requirements need to extract specific types of data in the data to be governed and perform specific operations, according to the governance process of the business requirements.
  • specific data that meets business needs can be extracted.
  • the system can be directly connected to different medical information systems, such as HIS, LIS, CIS and EMR, etc., and the connection technologies used include DBLINK, WEBSERVICE and KETTLE, etc. Therefore, the data to be managed in this application includes sources from one or more medical information System's raw medical data.
  • the governance process for each business requirement is usually pre-configured and determined, and the governance process indicates the operation steps of the governance process and their logical relationships.
  • the governance process includes: extracting preoperative data - unit conversion of the red blood cell count value in the preoperative data to A Units. Convert the white blood cell count values in the preoperative data to B units.
  • the model building module creates several governance nodes corresponding to the business requirements, and establishes the connection relationship between different governance nodes to obtain a data governance model corresponding to the governance process of the business requirements.
  • Each governance node in this application has a corresponding node type and calls a governance function in the governance function library, and optionally also includes function parameters corresponding to the governance function, and the function parameters can be configured by the user.
  • the node type includes an input node (InoutNode), an intermediate node and an output node (OutputNode).
  • Each input node is connected to the output node after passing through several intermediate nodes, and the data output terminal of the input node is connected to the corresponding intermediate node and the output node.
  • the data input terminal of each intermediate node is connected to the intermediate nodes, and the data input terminal and data output terminal of each intermediate node are respectively connected to other governance nodes.
  • the node types of nodes A, B, and C are input nodes
  • the node types of nodes D, E, F, G, and H are intermediate nodes
  • the node types of nodes I, J, and K are output nodes.
  • One data output end of each governance node except the output node is connected to one or more other governance nodes as its lower nodes.
  • the governance node passes the data The data stream output by the output end is output to each lower node respectively, for example, the data output end of node E is connected to node I, and the data output end of node D is connected to nodes F and G.
  • the data input terminal of each governance node is connected to one or more other governance nodes as its upper-level node.
  • the data input terminal of the governance node inputs The data includes the output data of all upper-level nodes. For example, node D is connected to three upper-layer nodes A, B, and C, and node H is connected to one upper-layer node F.
  • the intermediate node includes a processing node (ProcessNode) and/or a sorting node (CondNode), and the processing node includes a data input terminal and a data output terminal.
  • processing node includes a data input terminal and a data output terminal.
  • nodes D, E, G, and H in FIG. 2 are processing nodes.
  • Node F is a sorting node.
  • the sorting node includes a data input terminal and a plurality of data output terminals. The sorting node divides the input data into multiple channels of data with different medical index attributes and outputs them through the data output terminals respectively.
  • the sorting node divides the input data into one channel of male data and output through one data output terminal, and one channel of female data through another data output terminal.
  • the sorting node divides the input data into three channels of data to output through the data output terminal, one channel outputs the data with the item name "white blood cell count”, and one channel outputs the project.
  • the governance function library is a pre-maintained function library that contains governance functions commonly used in the field of medical data governance. Different governance nodes call the same or different governance functions.
  • the data governance model is constructed by a plurality of governance nodes that call different governance functions, which is equivalent to splitting the prior art practice of completing governance functions by one complex function into multiple simple functions jointly completing governance Compared with complex functions, simple functions are easier to construct and have lower complexity, so it can effectively reduce the difficulty of modeling.
  • the same governance function can be reused in different governance nodes in a data governance model, and can also be reused in governance nodes in different data governance models, which improves the reuse rate of governance functions and reduces the complexity of modeling. Spend.
  • the governance functions are pre-maintained in the governance function library, the uncertainty caused by the user's own writing of governance functions is avoided, which is beneficial to improve the standardization and standardization of governance functions, and ultimately improve the standardization of governance results.
  • Each governance node can be represented as (Type, Function, Argument), Type is the node type, Function is the governance function called by the governance node, and Argument is the function parameter used.
  • Type is the node type
  • Function is the governance function called by the governance node
  • Argument is the function parameter used.
  • the upper and lower nodes of each governance node can also be determined.
  • the corresponding relationship between the governance nodes and their connection relationships and the governance process required by the business is pre-configured.
  • the operation step of "extracting preoperative data" in the above example can be implemented by using the input node that calls the GetValue function.
  • the actual application scenario may be that the model building module creates governance nodes and their connection relationships according to the user's instructions, that is, the user creates governance nodes according to the governance process of business requirements and connects their data input/output ports, and selects each governance node.
  • the medical data governance system is used to perform different data governance on the same group of data to be governed to obtain post-governance data corresponding to N business requirements, for example, to perform different data governance on the same batch of data to be governed.
  • Post-treatment application in different studies.
  • some governance steps in these business requirements overlap, so it can be considered that the governance process corresponding to each business requirement can be divided into standardized governance processes and in-depth governance processes.
  • the governance process is the same, but the in-depth governance process is different to meet different business needs.
  • Common standardized governance processes include name standardization, data format standardization, etc.; in-depth governance processes have different content according to different business needs, and may process data. Logical operations, etc., then generate new data. Therefore, the model building module is used to build a standardized governance model and N in-depth governance models corresponding to different business requirements.
  • the standardized governance model and an in-depth governance model constitute a data governance model corresponding to the business requirements, N ⁇ 2.
  • the model building module includes a standardized model building unit and an in-depth governance model building unit.
  • the standardized model building unit is used to create a standardized governance model corresponding to the standardized governance process shared by the N business requirements.
  • the in-depth governance model building unit is used to create N in-depth governance models, each of which corresponds to an in-depth governance process required by a business.
  • the specific methods for creating a standardized governance model and an in-depth governance model are also obtained by creating governance nodes and connection relationships as described above, which will not be repeated in this application.
  • Model execution module The model execution module is used to execute the data governance model constructed above.
  • the model execution module obtains the data to be managed and inputs the data input terminal of the input node of the data governance model corresponding to the business requirement, and schedules and executes each governance node in turn.
  • each governance node uses the medical knowledge base to call the corresponding governance function to process the data input from the data input terminal during execution, and outputs the processed data through the data output terminal until the output node outputs data through the data output terminal after execution Generate post-governance data corresponding to business requirements.
  • the process of scheduling and executing each governance node is as follows: Determine the weight of the governance node according to the number of upper-level nodes connected to each governance node. For example, nodes A, B, and C in Figure 2 are input nodes without upper-level nodes, and the corresponding weights are 0, and node D has three three-layer nodes and the corresponding weight is 3. To deal with governance nodes with a weight of 0, when there are multiple governance nodes with equal weights, one can be randomly selected for execution. After the execution is completed, the weight of the connected lower-level nodes is reduced by 1, so that each governance node with a weight of 0 is executed in turn. . For example, in Figure 2, the weights of nodes A, B, and C are all 0.
  • a governance node can be randomly selected to execute, such as executing node A. After the execution is completed, the weight of its lower node D is reduced by 1 to 2; continue Execute node B, after the execution is completed, reduce the weight of its lower node D by 1 to 1; continue to execute node B, after the execution is completed, reduce the weight of its lower node D by 1 to 0; then execute node D, and so on.
  • the model execution module will input the data to be governed into the standardized governance model during execution, and input the data output by the standardized governance model into N pieces of data respectively.
  • each in-depth governance model outputs the post-governance data corresponding to the business requirements.
  • This approach in this application can greatly improve the efficiency of data governance.
  • N times of standardized governance processes and N times of in-depth governance processes need to be completed in turn for N business requirements.
  • a total of only one standardized governance process and N deep governance processes need to be completed for each business requirement, which greatly simplifies repetitive operations and improves the efficiency of data governance.
  • governance process management module the governance process management module is used to record the data governance process data of each governance node, and feed back an error message when the governance node performs an error.
  • the data governance process data includes the type, content and data volume of the data to be governed.
  • the recorded data governance process data and error messages can be displayed visually, so that the data governance process and the effect of data governance can be seen intuitively.
  • each governance node During execution of each governance node, if it is detected that the data format of the input data does not conform to the predetermined rules; Governance rules corresponding to the input data are not stored in the knowledge base; and/or, if it is detected that the input data is missing, it is determined that there is an error in execution.
  • the medical knowledge base is a pre-maintained library that stores the governance rules and required parameter data that each governance function depends on when executing.
  • the drug standardization function needs to rely on the drug standardization information stored in the medical knowledge base. When there is no standardized drug information for a certain type of drug in the library, it will lead to governance errors.
  • the governance process management module will not only feedback an error message, but also store the corresponding execution error data in the problem database when it detects that the corresponding execution error data is not stored in the problem database.
  • the problem database is also a pre-maintained database, which stores various execution error data in the governance process.
  • the medical knowledge base can be updated by manual intervention to fix execution errors in the problem base, so the problem base is used to update the medical knowledge base.
  • This application uses multiple governance nodes to complete the required governance requirements. At the same time, recording the data governance process data of each governance node is also conducive to evaluating the governance results. Therefore, errors can be found in time, which can effectively improve the efficiency of data governance. It is reliable, and it is easy to locate the governance node where governance errors are made, which is convenient for improvement.
  • This application does not directly regard the data output by the data governance model as the post-governance data corresponding to the final business requirements, but has a further screening process, because the applicant considers that in the actual medical field, the time of data generation is very important for medical data
  • An attribute has different meanings under different business requirements. For example, a patient may visit a doctor multiple times to generate data corresponding to multiple blood routines. In one study, the first blood routine data of the patient may be focused on, and in another study, the blood routine data of the patient after treatment may need to be focused on. Routine data to judge treatment effect. For another example, when studying drug-induced liver injury, it is also necessary to define the data of the first relevant index items, and the subsequent data do not have the medical research value of the first data. Therefore the system also includes:
  • the medical stage division module is used to determine the medical stage division result corresponding to the business requirement, and the medical stage division result includes several medical stages divided continuously according to time.
  • the medical stage division module is used to determine the data extraction instructions of each medical stage corresponding to the business requirements.
  • the data extraction instructions include the name of the index to be extracted and the data extraction conditions.
  • the data extraction instructions of each medical stage corresponding to the business requirements are usually pre-configured. of.
  • the data of the index items in each medical stage is extracted according to the data extraction instruction, and the time period of each medical stage is determined according to the data generation time of the index items in each medical stage to obtain the medical stage division result.
  • start time and end time of each medical stage determined according to the data generation time of the index items in each medical stage.
  • the start time and end time of the stage are adjusted, and the end time of the current medical stage is adjusted according to the start time of the next medical stage, so that the time periods of each medical stage are continuous and not repeated, and the medical stage division result is obtained.
  • the business data screening module filters out the data of the index items whose data generation time is within the corresponding target time period from all the data output by the data governance model corresponding to the business requirements, and obtains the post-governance data corresponding to the business requirements , that is, the data in a specific time period corresponding to the business requirement can be obtained.
  • the target time period corresponding to each indicator item matches the business requirement
  • different indicator items under the same business requirement have the same or different target time period
  • the same indicator item has the same or different target time period under different business requirements
  • the target time period is a medical stage in the medical stage division result corresponding to the business requirement.

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

一种医疗数据治理系统,涉及医疗技术领域,该医疗数据治理系统根据业务需求、利用多个调用不同治理函数的治理节点搭建形成数据治理模型对原始医疗数据进行治理,每个治理节点所调用的治理函数功能简单、复杂度低、易于编写,可以有效降低建模难度,从而有利于提高数据治理的效率,由于对整体化治理功能分成治理节点逐步执行完成,因此使得每个治理节点可以实现数据治理过程数据的记录以及报错提示,实现了治理过程可评估化,出现了错误可以及时反馈,也便于实现问题追踪,从而有利于提高数据治理的质量以及对治理过程的反馈改进。

Description

一种医疗数据治理系统 技术领域
本发明涉及医疗技术领域,尤其是一种医疗数据治理系统。
背景技术
为了满足各自医疗场景的复杂化、多样化以及个性化的需求,各医院均具有各自的医疗信息系统,医疗信息系统的构成非常复杂、日常运行中积累的原始医疗数据的数据量非常庞大,而且不同医院、不同医疗信息系统所产生的原始医疗数据往往不会遵循相同的数据标准,因此目前通常需要对原始医疗数据执行相应的数据治理操作以挖掘到所需的数据后才能使用。而当前数据治理的方法通常是面向数据应用场景建立相应的数据治理模型实现相应的功能,由于原始医疗数据的复杂度高,因此数据治理模型的复杂度往往也较高,这就导致建模难度大而且容易出错,导致数据治理的效率和质量往往都难以保证。
技术问题
由于原始医疗数据的复杂度高,因此数据治理模型的复杂度往往也较高,这就导致建模难度大而且容易出错,导致数据治理的效率和质量往往都难以保证。
技术解决方案
本发明人针对上述问题及技术需求,提出了一种医疗数据治理系统,本发明的技术方案如下:
一种医疗数据治理系统,该系统包括:
模型搭建模块,模型搭建模块用于创建若干个与业务需求对应的治理节点,并建立不同治理节点之间的连接关系,得到与业务需求的治理流程对应的数据治理模型;每个治理节点具有相应的节点类型并调用治理函数库中的一个治理函数,节点类型包括输入节点、中间节点和输出节点,输入节点的数据输出端连接相应的中间节点,输出节点的数据输入端连接中间节点,每个中间节点的数据输入端和数据输出端分别连接其他治理节点;
模型执行模块,模型执行模块用于获取待治理数据并输入业务需求对应的数据治理模型的输入节点的数据输入端并依次调度执行各个治理节点,每个治理节点在执行时利用医学知识库调用对应的治理函数对数据输入端输入的数据进行处理,并将处理后的数据通过数据输出端输出,直至输出节点执行后通过数据输出端输出数据产生业务需求对应的治理后数据;
治理过程管理模块,治理过程管理模块用于记录各个治理节点的数据治理过程数据并在治理节点执行出错时反馈报错提示。
其进一步的技术方案为,医疗数据治理系统用于对待治理数据进行数据治理分别得到与N个业务需求对应的治理后数据,则模型搭建模块用于搭建得到一个标准化治理模型和N个分别与不同的业务需求对应的深度治理模型,标准化治理模型和一个深度治理模型构成对应的业务需求的数据治理模型,N≥2;
则模型执行模块用于将待治理数据输入标准化治理模型,并将标准化治理模型输出的数据分别输入N个深度治理模型,每个深度治理模型输出对应的业务需求的治理后数据。
其进一步的技术方案为,模型搭建模块包括:
标准化模型搭建单元,标准化模型搭建单元用于创建与N个业务需求共用的标准化治理流程对应的标准化治理模型;
深度治理模型搭建单元,深度治理模型搭建单元用于创建N个深度治理模型,每个深度治理模型分别对应一个业务需求的深度治理流程。
其进一步的技术方案为,该系统还包括:
医疗阶段划分模块,医疗阶段划分模块用于确定与业务需求对应的医疗阶段划分结果,医疗阶段划分结果包括若干个按照时间连续划分的医疗阶段;
业务数据筛选模块,业务数据筛选模块从业务需求对应的数据治理模型输出的所有数据中筛选出数据产生时间在对应的目标时间段内的指标项的数据,得到业务需求对应的治理后数据;
其中,各个指标项所对应的目标时间段与业务需求匹配,相同业务需求下的不同指标项具有相同或不同的目标时间段,同一指标项在不同业务需求下具有相同或不同的目标时间段,目标时间段是业务需求对应的医疗阶段划分结果中的一个医疗阶段。
其进一步的技术方案为,医疗阶段划分模块用于确定业务需求对应的各个医疗阶段的数据抽取指令,并根据数据抽取指令抽取得到各个医疗阶段内的指标项的数据,根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的时间段得到医疗阶段划分结果。
其进一步的技术方案为,医疗阶段划分模块用于根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的开始时刻和结束时刻,并根据下一个医疗阶段的开始时刻调整当前医疗阶段的结束时刻,使得各个医疗阶段的时间段连续而不重复,得到医疗阶段划分结果。
其进一步的技术方案为,治理过程管理模块还用于在治理节点执行出错,且检测到问题库中未存储对应的执行错误数据时,将对应的执行错误数据存储到问题库中,问题库用于更新医学知识库。
其进一步的技术方案为,每个治理节点在执行时,若检测到输入的数据的数据格式不符合预定规则;和/或,在调用治理函数利用医学知识库中与输入的数据对应的治理规则进行治理时,检测到医学知识库中未存储与输入的数据对应的治理规则;和/或,检测到输入的数据存在缺失;则确定执行出错。
其进一步的技术方案为,数据治理模型中的中间节点包括处理节点和/或分拣节点,处理节点包括一个数据输入端和一个数据输出端,分拣节点包括一个数据输入端和多个数据输出端,分拣节点将输入的数据分成多路具有不同医疗指标属性的数据后分别通过数据输出端输出。
其进一步的技术方案为,除输出节点之外的其他每个治理节点的一个数据输出端连接至一个或多个其他的治理节点作为其下层节点,当治理节点的一个数据输出端连接多个下层节点时,治理节点通过数据输出端输出的数据流分别输出给各个下层节点;
除输入节点之外的其他每个治理节点的数据输入端连接一个或多个其他的治理节点作为其上层节点,当治理节点的数据输入端连接多个上层节点时,治理节点的数据输入端输入的数据包括所有上层节点输出的数据。
有益效果
1、本申请公开了一种医疗数据治理系统,该医疗数据治理系统根据业务需求、利用多个调用不同治理函数的治理节点搭建形成数据治理模型对原始医疗数据进行治理,每个治理节点所调用的治理函数功能简单、复杂度低、易于编写,可以有效降低建模难度,从而有利于提高数据治理的效率。而且由于对整体化治理功能分成治理节点逐步执行完成,因此使得每个治理节点可以实现数据治理过程数据的记录以及报错提示,实现了治理过程可评估化,出现了错误可以及时反馈,也便于实现问题追踪,从而有利于提高数据治理的质量以及对治理过程的反馈改进。
2、搭建形成数据治理模型时,每个治理节点直接调用治理函数库中的函数实现相应的功能,从而提高了治理函数的复用率,减少了建模过程中的重复操作,进一步提高数据治理的效率。预先定义好的治理函数也更具有规范性和标准性,最终提高数据治理的质量。
3、该系统在构建数据治理模型时,针对多个不同的业务需求的相同治理操作部分构建一个标准化治理模型,对不同的个性化需求部分构建各自的深度治理模型,可以在满足不同业务需求的基础上减少重复操作过程,提高数据治理效率。
4、该系统从医疗领域数据的实际使用需求和特点出发,在系统中设置了医疗阶段划分模块和业务数据筛选模块,对完成数据治理的数据从时间属性角度进一步筛选,从而可以得到更符合实际使用需求、具有实用价值的数据。
附图说明
图1是本申请公开的医疗数据治理系统中各个模块的数据处理流程示意图。
图2是本申请的医疗数据治理系统执行过程中构建得到的治理节点之间的连接关系的示意图。
本发明的实施方式
下面结合附图对本发明的具体实施方式做进一步说明。
本申请公开了一种医疗数据治理系统,请参考图1,该系统包括:
一、模型搭建模块,模型搭建模块用于得到与业务需求的治理流程对应的数据治理模型。
本申请中的业务需求是在针对待治理数据进行数据挖掘和分析过程中所产生的需求,该业务需求需要提取待治理数据中特定类型的数据并进行特定的操作,按照业务需求的治理流程进行数据治理时即能提取到满足业务需求的特定的数据。该系统可以直接对接不同的医疗信息系统,比如HIS、LIS、CIS和EMR等,采用的对接技术包括DBLINK、WEBSERVICE和KETTLE等,因此本申请中的待治理数据包括来源于一个或多个医疗信息系统的原始医疗数据。
每个业务需求的治理流程通常预先配置确定,治理流程指示治理过程的操作步骤及其逻辑关系,比如治理流程包括:提取术前数据——对术前数据中的红细胞计数值进行单位转换为A单位、对术前数据中的白细胞计数值进行单位转换为B单位。
模型搭建模块创建若干个与业务需求对应的治理节点,并建立不同治理节点之间的连接关系得到与业务需求的治理流程对应的数据治理模型。本申请中的每个治理节点具有相应的节点类型并调用治理函数库中的一个治理函数,可选的还包括治理函数对应的函数参数,该函数参数可以由用户进行配置。
其中,节点类型包括输入节点(InoutNode)、中间节点和输出节点(OutputNode),每个输入节点经过若干个中间节点后连接到输出节点,则输入节点的数据输出端连接相应的中间节点,输出节点的数据输入端连接中间节点,每个中间节点的数据输入端和数据输出端分别连接其他治理节点。请参考图2所示,节点A、B、C的节点类型为输入节点,节点D、E、F、G、H的节点类型为中间节点,节点I、J、K的节点类型为输出节点。
除输出节点之外的其他每个治理节点的一个数据输出端连接至一个或多个其他的治理节点作为其下层节点,当治理节点的一个数据输出端连接多个下层节点时,治理节点通过数据输出端输出的数据流分别输出给各个下层节点,比如节点E的数据输出端连接节点I,节点D的数据输出端连接节点F和G。除输入节点之外的其他每个治理节点的数据输入端连接一个或多个其他的治理节点作为其上层节点,当治理节点的数据输入端连接多个上层节点时,治理节点的数据输入端输入的数据包括所有上层节点输出的数据。比如节点D连接三个上层节点A、B、C,节点H连接一个上层节点F。
进一步的,中间节点包括处理节点(ProcessNode)和/或分拣节点(CondNode),处理节点包括一个数据输入端和一个数据输出端,比如图2中节点D、E、G、H为处理节点,节点F为分拣节点。分拣节点包括一个数据输入端和多个数据输出端,分拣节点将输入的数据分成多路具有不同医疗指标属性的数据后分别通过数据输出端输出。其中,当医疗指标项的项目名称不同,和/或,医疗指标项的项目数据不同,和/或,身份信息项不同时,确定两组数据有不同的医疗指标属性。其中医疗指标项用于指示患者所做的医疗检查的相关信息,医疗指标项的项目名称为医学指标中文名称和/或医学指标英文简写。身份信息项用于指示患者的基础身份信息,主要包括患者的性别和/或年龄。比如当不同医疗指标属性为患者的不同性别时,分拣节点将输入的数据分成性别为男的一路数据通过一个数据输出端输出以及性别为女的一路数据通过另一个数据输出端。再比如当不同医疗指标属性为不同的医疗指标项的项目名称时,分拣节点将输入的数据分成三路数据通过数据输出端输出,一路输出项目名称为“白细胞计数”的数据,一路输出项目名称为“红细胞压积”的数据,一路输出项目名称为“红细胞计数”的数据。
治理函数库是一个预先维护的包含医疗数据治理领域常用的治理函数的函数库,不同的治理节点调用相同或不同的治理函数。基于本申请的系统,数据治理模型由多个调用不同治理函数的治理节点搭建而成,相当于将现有技术中由一个复杂函数完成治理功能的做法拆分成由多个简单函数共同完成治理功能,简单函数相比于复杂函数来说更容易构建、复杂度更低,因此可以有效降低建模难度。而且同一个治理函数可以被重复使用在一个数据治理模型中的不同治理节点中,也可以被重复使用在不同数据治理模型的治理节点中,提高了治理函数的复用率,降低了建模复杂度。而且由于治理函数都是预先维护在治理函数库中的,因此避免了用户自行编写治理函数所带来的不确定性,有利于提高治理函数的规范化和标准化,最终提高治理结果的标准化。
每个治理节点可以表示为(Type,Function,Argument),Type即为节点类型,Function是该治理节点调用的治理函数,Argument是使用的函数参数。同时还可以确定每个治理节点的上层节点和下层节点。治理节点及其连接关系与业务需求的治理流程之间的对应关系是预先配置的,比如可以利用调用GetValue函数的输入节点实现上述举例中“提取术前数据”这一操作步骤。实际的应用场景可能是模型搭建模块根据用户的指令创建治理节点及其连接关系,也即用户根据业务需求的治理流程创建治理节点并对其数据输入/输出端口进行连接,并选择每个治理节点所使用的治理函数。
可选的,在本申请中,该医疗数据治理系统用于对同一组待治理数据进行不同的数据治理分别得到与N个业务需求对应的治理后数据,比如将同一批待治理数据做不同的治理后应用于不同的研究中。通过情况下,这些业务需求中有一些治理步骤是重合的,因此可以认为每个业务需求对应的治理流程可以被拆分为标准化治理流程和深度治理流程,而多个不同的业务需求中的标准化治理流程相同、但深度治理流程不同以满足不同业务需求,常见的标准化治理流程比如有名称标准化、数据格式标准化等等;深度治理流程则根据不同的业务需求有不同的内容,可能会对数据进行逻辑运算等再生成新的数据。因此模型搭建模块用于搭建得到一个标准化治理模型和N个分别与不同的业务需求对应的深度治理模型,标准化治理模型和一个深度治理模型构成对应的业务需求的数据治理模型,N≥2。
则模型搭建模块包括标准化模型搭建单元和深度治理模型搭建单元。标准化模型搭建单元用于创建与N个业务需求共用的标准化治理流程对应的标准化治理模型。深度治理模型搭建单元用于创建N个深度治理模型,每个深度治理模型分别对应一个业务需求的深度治理流程。具体创建标准化治理模型和深度治理模型的方法也是如上所述通过创建治理节点及连接关系得到,本申请不再赘述。
二、模型执行模块,模型执行模块用于执行上述构建的数据治理模型,则模型执行模块获取待治理数据并输入业务需求对应的数据治理模型的输入节点的数据输入端并依次调度执行各个治理节点,每个治理节点在执行时利用医学知识库调用对应的治理函数对数据输入端输入的数据进行处理,并将处理后的数据通过数据输出端输出,直至输出节点执行后通过数据输出端输出数据产生业务需求对应的治理后数据。
调度执行各个治理节点的过程如下:根据每个治理节点所连接的上层节点的数量确定该治理节点的权重,比如图2中节点A、B、C均为输入节点没有上层节点、对应的权重为0,而节点D有三个三层节点则对应的权重为3。处理权重为0的治理节点,当有多个权重相等的治理节点时可以随机选取一个执行,执行完成后对其所连接的下层节点的权重减1,如此循环依次执行各个权重为0的治理节点。比如在图2中,节点A、B、C的权重均为0,此时可以随机选取一个治理节点执行,比如执行节点A,执行完成后,对其下层节点D的权重减1为2;继续执行节点B,执行完成后,对其下层节点D的权重减1为1;继续执行节点B,执行完成后,对其下层节点D的权重减1为0;然后执行节点D,如此循环。
由于本申请会针对多个业务需求构建一个标准化治理模型和多个深度治理模型,因此模型执行模块在执行时,将待治理数据输入标准化治理模型,并将标准化治理模型输出的数据分别输入N个深度治理模型,每个深度治理模型输出对应的业务需求的治理后数据。本申请的这种做法可以很好地提高数据治理的效率,按照常规的治理思路,对N个业务需求需要依次分别完成N次标准化治理流程和N次深度治理流程,而在本申请中对N个业务需求只需完成总计一次标准化治理流程和N次深度治理流程即可,大大简化了重复性操作,提高了数据治理的效率。
三、治理过程管理模块,治理过程管理模块用于记录各个治理节点的数据治理过程数据并在治理节点执行出错时反馈报错提示。其中数据治理过程数据包括治理的数据的类型、内容和数据量等。记录的数据治理过程数据以及报错提示都可以可视化显示,从而直观的看到数据治理过程以及数据治理的效果。
每个治理节点在执行时,若检测到输入的数据的数据格式不符合预定规则;和/或,在调用治理函数利用医学知识库中与输入的数据对应的治理规则进行治理时,检测到医学知识库中未存储与输入的数据对应的治理规则;和/或,检测到输入的数据存在缺失,则确定执行出错。其中,医学知识库是预先维护的库,存储各个治理函数在执行时所依赖的治理规则和所需的参数数据等,比如药品标准化函数需要依赖医学知识库中存储的药品标准化信息,当医学知识库中未有某一类药品的药品标准化信息时,则会导致治理出错。
此时治理过程管理模块不仅会反馈报错提示,还会在检测到问题库中未存储对应的执行错误数据时,将对应的执行错误数据存储到问题库中。问题库也是一个预先维护的数据库,存储治理过程中出现的各种执行错误数据,在将新的执行错误数据存储进问题库中时会有去重处理,避免相同的执行错误数据被重复记录。后续在质量管理环节,可以人工介入修复问题库中出现的执行错误更新医学知识库,因此该问题库用于更新医学知识库。
本申请由多个治理节点来完成所需的治理需求,同时记录每个治理节点的数据治理过程数据也有利于对治理结果进行评估,因此当出错时可以及时发现,从而可以有效提高数据治理的可靠性,而且可以方便定位到治理出错的治理节点,便于改进。
本申请并不直接将数据治理模型输出的数据作为最终业务需求对应的治理后数据,而有进一步的筛选过程,这是因为申请人考虑到在实际医疗领域,数据产生时间是医疗数据非常重要的一个属性,在不同业务需求下有不同的意义。比如患者可能会就诊多次产生多个血常规对应的数据,在一个研究中可能重点需要研究患者第一次的血常规数据,在另一项研究中则可能需要重点研究患者在治疗后的血常规数据以判断治疗效果。再比如在研究药性肝损伤时,也是需要定义到第一次相关指标项的数据,而后续的数据则没有第一次的数据所具有的医疗研究价值。因此该系统还包括:
四、医疗阶段划分模块,医疗阶段划分模块用于确定与业务需求对应的医疗阶段划分结果,医疗阶段划分结果包括若干个按照时间连续划分的医疗阶段。
具体的,医疗阶段划分模块用于确定业务需求对应的各个医疗阶段的数据抽取指令,数据抽取指令包括待抽取指标名称和数据抽取条件,业务需求对应的各个医疗阶段的数据抽取指令通常是预先配置的。根据数据抽取指令抽取得到各个医疗阶段内的指标项的数据,根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的时间段得到医疗阶段划分结果。
实际根据每个医疗阶段内的指标项的数据产生时间所确定的各个医疗阶段的开始时刻和结束时刻之间可能存在重合,则在根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的开始时刻和结束时刻,并根据下一个医疗阶段的开始时刻调整当前医疗阶段的结束时刻,使得各个医疗阶段的时间段连续而不重复,得到医疗阶段划分结果。
五、业务数据筛选模块,业务数据筛选模块从业务需求对应的数据治理模型输出的所有数据中筛选出数据产生时间在对应的目标时间段内的指标项的数据,得到业务需求对应的治理后数据,也即能得到与该业务需求对应的特定的时间段内的数据。其中,各个指标项所对应的目标时间段与业务需求匹配,相同业务需求下的不同指标项具有相同或不同的目标时间段,同一指标项在不同业务需求下具有相同或不同的目标时间段,目标时间段是业务需求对应的医疗阶段划分结果中的一个医疗阶段。
以上所述的仅是本申请的优选实施方式,本发明不限于以上实施例。可以理解,本领域技术人员在不脱离本发明的精神和构思的前提下直接导出或联想到的其他改进和变化,均应认为包含在本发明的保护范围之内。

Claims (10)

  1. 一种医疗数据治理系统,其特征在于,所述系统包括:
    模型搭建模块,所述模型搭建模块用于创建若干个与业务需求对应的治理节点,并建立不同治理节点之间的连接关系,得到与所述业务需求的治理流程对应的数据治理模型;每个治理节点具有相应的节点类型并调用治理函数库中的一个治理函数,所述节点类型包括输入节点、中间节点和输出节点,输入节点的数据输出端连接相应的中间节点,输出节点的数据输入端连接中间节点,每个中间节点的数据输入端和数据输出端分别连接其他治理节点;
    模型执行模块,所述模型执行模块用于获取待治理数据并输入业务需求对应的数据治理模型的输入节点的数据输入端并依次调度执行各个治理节点,每个治理节点在执行时利用医学知识库调用对应的治理函数对数据输入端输入的数据进行处理,并将处理后的数据通过数据输出端输出,直至输出节点执行后通过数据输出端输出数据产生所述业务需求对应的治理后数据;
    治理过程管理模块,所述治理过程管理模块用于记录各个治理节点的数据治理过程数据并在治理节点执行出错时反馈报错提示。
  2. 根据权利要求1所述的系统,其特征在于,
    所述医疗数据治理系统用于对所述待治理数据进行数据治理分别得到与N个业务需求对应的治理后数据,则所述模型搭建模块用于搭建得到一个标准化治理模型和N个分别与不同的业务需求对应的深度治理模型,所述标准化治理模型和一个深度治理模型构成对应的业务需求的数据治理模型,N≥2;
    则所述模型执行模块用于将所述待治理数据输入所述标准化治理模型,并将所述标准化治理模型输出的数据分别输入N个深度治理模型,每个深度治理模型输出对应的业务需求的治理后数据。
  3. 根据权利要求2所述的系统,其特征在于,所述模型搭建模块包括:
    标准化模型搭建单元,所述标准化模型搭建单元用于创建与N个业务需求共用的标准化治理流程对应的标准化治理模型;
    深度治理模型搭建单元,所述深度治理模型搭建单元用于创建所述N个深度治理模型,每个深度治理模型分别对应一个业务需求的深度治理流程。
  4. 根据权利要求1所述的系统,其特征在于,所述系统还包括:
    医疗阶段划分模块,所述医疗阶段划分模块用于确定与业务需求对应的医疗阶段划分结果,所述医疗阶段划分结果包括若干个按照时间连续划分的医疗阶段;
    业务数据筛选模块,所述业务数据筛选模块从所述业务需求对应的数据治理模型输出的所有数据中筛选出数据产生时间在对应的目标时间段内的指标项的数据,得到所述业务需求对应的治理后数据;
    其中,各个指标项所对应的目标时间段与业务需求匹配,相同业务需求下的不同指标项具有相同或不同的目标时间段,同一指标项在不同业务需求下具有相同或不同的目标时间段,所述目标时间段是所述业务需求对应的医疗阶段划分结果中的一个医疗阶段。
  5. 根据权利要求4所述的系统,其特征在于,
    所述医疗阶段划分模块用于确定业务需求对应的各个医疗阶段的数据抽取指令,并根据所述数据抽取指令抽取得到各个医疗阶段内的指标项的数据,根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的时间段得到所述医疗阶段划分结果。
  6. 根据权利要求4所述的系统,其特征在于,
    所述医疗阶段划分模块用于根据每个医疗阶段内的指标项的数据产生时间确定各个医疗阶段的开始时刻和结束时刻,并根据下一个医疗阶段的开始时刻调整当前医疗阶段的结束时刻,使得各个医疗阶段的时间段连续而不重复,得到所述医疗阶段划分结果。
  7. 根据权利要求1-6任一所述的系统,其特征在于,
    所述治理过程管理模块还用于在治理节点执行出错,且检测到问题库中未存储对应的执行错误数据时,将对应的执行错误数据存储到问题库中,所述问题库用于更新所述医学知识库。
  8. 根据权利要求1-6任一所述的系统,其特征在于,
    每个治理节点在执行时,若检测到输入的数据的数据格式不符合预定规则;和/或,在调用治理函数利用医学知识库中与输入的数据对应的治理规则进行治理时,检测到所述医学知识库中未存储与输入的数据对应的治理规则;和/或,检测到输入的数据存在缺失;则确定执行出错。
  9. 根据权利要求1-6任一所述的系统,其特征在于,
    所述数据治理模型中的中间节点包括处理节点和/或分拣节点,所述处理节点包括一个数据输入端和一个数据输出端,所述分拣节点包括一个数据输入端和多个数据输出端,所述分拣节点将输入的数据分成多路具有不同医疗指标属性的数据后分别通过数据输出端输出。
  10. 根据权利要求1-6任一所述的系统,其特征在于,
    除输出节点之外的其他每个治理节点的一个数据输出端连接至一个或多个其他的治理节点作为其下层节点,当治理节点的一个数据输出端连接多个下层节点时,所述治理节点通过所述数据输出端输出的数据流分别输出给各个下层节点;
    除输入节点之外的其他每个治理节点的数据输入端连接一个或多个其他的治理节点作为其上层节点,当治理节点的数据输入端连接多个上层节点时,所述治理节点的数据输入端输入的数据包括所有上层节点输出的数据。
PCT/CN2022/074873 2021-02-07 2022-01-29 一种医疗数据治理系统 WO2022166859A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110169148.8 2021-02-07
CN202110169148.8A CN112802607B (zh) 2021-02-07 2021-02-07 一种医疗数据治理系统

Publications (1)

Publication Number Publication Date
WO2022166859A1 true WO2022166859A1 (zh) 2022-08-11

Family

ID=75814734

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/074873 WO2022166859A1 (zh) 2021-02-07 2022-01-29 一种医疗数据治理系统

Country Status (2)

Country Link
CN (1) CN112802607B (zh)
WO (1) WO2022166859A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116453637A (zh) * 2023-03-20 2023-07-18 杭州市卫生健康事业发展中心 一种基于区域大数据的健康数据治理方法和系统

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112802607B (zh) * 2021-02-07 2022-07-08 无锡慧方科技有限公司 一种医疗数据治理系统
CN115223674B (zh) * 2022-08-16 2023-11-03 无锡慧方科技有限公司 一种适用于临床研究的医疗数据服务平台

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106777970A (zh) * 2016-12-15 2017-05-31 北京锐软科技股份有限公司 一种医疗信息系统数据模板化的集成系统及方法
CN110875095A (zh) * 2019-09-27 2020-03-10 长沙瀚云信息科技有限公司 一种标准化临床大数据中心系统
CN111161815A (zh) * 2019-12-27 2020-05-15 深圳中兴网信科技有限公司 医疗数据检测方法、装置、终端和计算机可读存储介质
CN111367969A (zh) * 2020-03-19 2020-07-03 北京三维天地科技股份有限公司 一种数据挖掘方法和系统
CN111651460A (zh) * 2020-06-11 2020-09-11 上海德易车信息科技有限公司 一种数据治理方法、装置、电子设备及可读存储介质
CN111881136A (zh) * 2020-07-29 2020-11-03 山东健康医疗大数据有限公司 一种实现医疗行业增量数据治理的方法
CN112802607A (zh) * 2021-02-07 2021-05-14 无锡慧方科技有限公司 一种医疗数据治理系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9083762B2 (en) * 2010-05-28 2015-07-14 Greg Saunders System and method for providing hybrid on demand services to a work unit
CN105467953B (zh) * 2015-11-11 2018-01-09 中国科学院软件研究所 一种面向工业大数据的知识表示及其自动化应用方法
CN110750540A (zh) * 2019-10-18 2020-02-04 中国人民解放军军事科学院军事医学研究院 构建医疗业务知识库的方法、获得医疗业务语义模型的方法及系统、介质
CN110781236A (zh) * 2019-10-29 2020-02-11 山西云时代技术有限公司 一种构建政务大数据治理体系的方法
CN112100451B (zh) * 2020-09-14 2023-11-17 上海飞机制造有限公司 基于图数据库搭建工业神经网络的方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106777970A (zh) * 2016-12-15 2017-05-31 北京锐软科技股份有限公司 一种医疗信息系统数据模板化的集成系统及方法
CN110875095A (zh) * 2019-09-27 2020-03-10 长沙瀚云信息科技有限公司 一种标准化临床大数据中心系统
CN111161815A (zh) * 2019-12-27 2020-05-15 深圳中兴网信科技有限公司 医疗数据检测方法、装置、终端和计算机可读存储介质
CN111367969A (zh) * 2020-03-19 2020-07-03 北京三维天地科技股份有限公司 一种数据挖掘方法和系统
CN111651460A (zh) * 2020-06-11 2020-09-11 上海德易车信息科技有限公司 一种数据治理方法、装置、电子设备及可读存储介质
CN111881136A (zh) * 2020-07-29 2020-11-03 山东健康医疗大数据有限公司 一种实现医疗行业增量数据治理的方法
CN112802607A (zh) * 2021-02-07 2021-05-14 无锡慧方科技有限公司 一种医疗数据治理系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116453637A (zh) * 2023-03-20 2023-07-18 杭州市卫生健康事业发展中心 一种基于区域大数据的健康数据治理方法和系统
CN116453637B (zh) * 2023-03-20 2023-11-07 杭州市卫生健康事业发展中心 一种基于区域大数据的健康数据治理方法和系统

Also Published As

Publication number Publication date
CN112802607A (zh) 2021-05-14
CN112802607B (zh) 2022-07-08

Similar Documents

Publication Publication Date Title
WO2022166859A1 (zh) 一种医疗数据治理系统
CN107239665B (zh) 医疗信息查询系统及方法
CN106295807A (zh) 一种信息处理的方法及装置
CN109741826B (zh) 麻醉评估决策树构建方法及设备
CN106250543A (zh) 一种自动化数据查询同步存储方法
CN103218540A (zh) 一种可视化交互式临床试验和临床随访的系统及方法
CN112331282B (zh) 基于随访项目对患者进行分组的方法和系统
CN106776986B (zh) 一种可配置的表单扩展信息传递方法
CN112506917A (zh) 一种主数据的字典映射方法、装置、系统、设备和介质
CN112307216A (zh) 药品知识图谱的构建方法和装置
CN118093632B (zh) 基于大语言模型和图结构的图数据库查询方法和装置
CN115599840A (zh) 一种复杂业务数据治理方法和系统
US10192031B1 (en) System for extracting information from DICOM structured reports
CN110334001A (zh) 一种批量自动生成回声测试的方法和装置
CN109509515A (zh) 电子病历管理方法、电子设备、存储介质、系统
CN112700878A (zh) 一种基于过程挖掘的临床路径优化方法
CN105933396B (zh) 一种协同各渠道系统接口实现长流程的方法及系统
CN107563128A (zh) 一种基于元数据的构建智能化区域急救医疗知识库方法
CN107610760A (zh) 一种基于软件定义的智能化区域急救医疗集成数据中心系统架构
CN115080594A (zh) 一种对数据进行多维度分析方法、系统和电子设备
CN109558403A (zh) 数据聚合方法及装置、计算机装置及计算机可读存储介质
JPS6285336A (ja) 表変換型ソフトウエア構成方式
CN108899067A (zh) 一种心电结构化报告智能生成方法
CN112233781A (zh) 一种手术室护士自动排班系统及方法
CN110931088A (zh) 一种电子病历模板数据元重用的系统和方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22749138

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22749138

Country of ref document: EP

Kind code of ref document: A1