Lifecycle management-based medical data set maintenance method and system

Abstract

The application discloses a kind of based on life cycle management's medical data set maintenance method and system, the method includes: obtaining the current version of the medical data set to be maintained and the operation to be executed;According to the current version, determine the current maintenance stage corresponding to the medical data set;According to the data operation rule corresponding to the current maintenance stage, the operation to be executed is verified to obtain the verification result;According to the verification result, determine the execution strategy of the medical data set to the operation to be executed;The execution strategy is used to limit execution supervision mode, execution supervision terminal and execution environment. Visible, the application can realize the security control of differentiation according to dynamic maintenance cycle, effectively reduce the risk of misoperation of high-sensitive medical data in the maintenance process, significantly improve the system stability and compliance audit efficiency of cross-stage data management.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for maintaining medical datasets based on lifecycle management. Background Technology

[0002] With the deepening advancement of digital healthcare and precision medicine, the scale and complexity of medical datasets are constantly increasing. Ensuring the security and compliance of large-scale medical data during maintenance has become a core requirement for healthcare IT infrastructure development. Existing technologies typically employ fixed security control rules or generic access control methods to manage the data maintenance process. Because they lack differentiated security controls based on dynamic maintenance cycles, they struggle to effectively address the risks of misoperation during the maintenance of highly sensitive medical data. This leads to insufficient system stability and compliance auditing efficiency in cross-stage data management, limiting the data security protection level and management reliability of healthcare big data platforms in complex maintenance scenarios. Clearly, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for maintaining medical datasets based on lifecycle management, which can realize differentiated security control according to the dynamic maintenance cycle, effectively reduce the risk of misoperation of highly sensitive medical data during the maintenance process, and significantly improve the system stability and compliance audit efficiency of cross-stage data management.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for maintaining medical datasets based on lifecycle management, the method comprising: Get the current version and pending operations of the medical dataset to be maintained; Based on the current version, determine the current maintenance phase corresponding to the medical dataset; The operation to be executed is verified according to the data operation rules corresponding to the current maintenance phase to obtain the verification result; Based on the verification result, an execution strategy is determined for performing the operation to be performed on the medical dataset; the execution strategy is used to limit the execution supervision method, execution supervision terminal, and execution environment.

[0005] As an optional implementation, in the first aspect of the present invention, the operation to be performed is a data acquisition operation, a data update operation, a data backup operation, a data recovery operation, or a data audit operation.

[0006] As an optional implementation, in the first aspect of the present invention, determining the current maintenance stage corresponding to the medical dataset based on the current version includes: Obtain data update information corresponding to multiple previous versions of the current version; Based on the stage prediction model, the current maintenance stage corresponding to the medical dataset is determined according to the data update information.

[0007] As an optional implementation, in the first aspect of the present invention, determining the current maintenance stage corresponding to the medical dataset based on the data update information using the stage prediction model includes: Each of the aforementioned data update information is input into the trained stage prediction model to obtain the corresponding prediction maintenance stage; Based on the predicted maintenance phase corresponding to all the data update information, determine the current maintenance phase corresponding to the medical dataset.

[0008] As an optional implementation, in the first aspect of the present invention, determining the current maintenance stage corresponding to the medical dataset based on the predicted maintenance stage corresponding to all the data update information includes: The prediction maintenance stages are sorted from earliest to latest according to the average update time of the corresponding data update information to obtain a stage sequence; The prediction maintenance stage that is closest to the current time point in the stage sequence is identified as a candidate stage; Calculate the rate of change parameter corresponding to the stage sequence; the rate of change parameter is the ratio between the total type data and the total time span of all the predicted maintenance stages in the stage sequence; Determine whether the rate of change parameter is greater than a preset parameter threshold to obtain a first determination result; If the first judgment result is yes, the next stage in the preset stage process of the candidate stage is determined as the current maintenance stage corresponding to the medical dataset; If the first determination result is negative, the candidate stage is determined as the current maintenance stage corresponding to the medical dataset.

[0009] As an optional implementation, in the first aspect of the present invention, the step of verifying the operation to be performed according to the data operation rules corresponding to the current maintenance phase to obtain a verification result includes: Determine the trained data update prediction model corresponding to the current maintenance phase; the data update prediction model is trained on a training dataset that includes multiple training execution operations performed in the current maintenance phase and corresponding data change annotations; Each sub-operation of the operation to be executed is input into the data update prediction model to obtain the predicted update data corresponding to each sub-operation; Based on the LLM model corresponding to the current maintenance phase and the predicted update data, determine the compliance level of each sub-operation; The compliance level of all the sub-operations is determined as the verification result of the operation to be executed.

[0010] As an optional implementation, in the first aspect of the present invention, determining the compliance level of each sub-operation based on the LLM model corresponding to the current maintenance phase and the predicted update data includes: For each sub-operation, the operation text description corresponding to the sub-operation and the predicted update data are combined to form the corresponding operation update text prompt. The operation update text prompt is input into the trained compliance judgment LLM corresponding to the current maintenance phase to obtain the compliance level of the sub-operation.

[0011] As an optional implementation, in the first aspect of the present invention, determining the execution strategy for performing the operation to be performed on the medical dataset based on the verification result includes: For each sub-operation, determine the operation data object corresponding to that sub-operation; Identify multiple historical execution terminals of the operation data object; All the historical execution terminals are sorted from high to low according to their corresponding terminal security levels to obtain a terminal sequence; The historical execution terminals in the target order of the terminal sequence are identified as the execution supervision terminals corresponding to the sub-operation; the target order is obtained by calculating the product of the total number of terminals in the terminal sequence and the target security ratio; the target security ratio is obtained by calculating the ratio between the compliance level of the sub-operation and the preset compliance standard level; The sandbox environment and data encryption verification algorithm established on the execution monitoring terminal are determined as the execution environment and execution monitoring method corresponding to this sub-operation.

[0012] A second aspect of this invention discloses a medical dataset maintenance system based on lifecycle management, the system comprising: The acquisition module is used to obtain the current version and pending operations of the medical dataset to be maintained. The determination module is used to determine the current maintenance stage corresponding to the medical dataset based on the current version. The verification module is used to verify the operation to be executed according to the data operation rules corresponding to the current maintenance stage in order to obtain the verification result; The execution module is used to determine the execution strategy for performing the operation to be performed on the medical dataset based on the verification result; the execution strategy is used to limit the execution supervision method, execution supervision terminal and execution environment.

[0013] As an optional implementation, in the second aspect of the present invention, the operation to be performed is a data acquisition operation, a data update operation, a data backup operation, a data recovery operation, or a data audit operation.

[0014] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method by which it determines the current maintenance phase corresponding to the medical dataset based on the current version, including: Obtain data update information corresponding to multiple previous versions of the current version; Based on the stage prediction model, the current maintenance stage corresponding to the medical dataset is determined according to the data update information.

[0015] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method of the current maintenance stage corresponding to the medical dataset based on a stage prediction model and according to the data update information, including: Each of the aforementioned data update information is input into the trained stage prediction model to obtain the corresponding prediction maintenance stage; Based on the predicted maintenance phase corresponding to all the data update information, determine the current maintenance phase corresponding to the medical dataset.

[0016] As an optional implementation, in a second aspect of the invention, the determining module determines the current maintenance stage corresponding to the medical dataset based on the predicted maintenance stage corresponding to all the data update information in a specific manner, including: The prediction maintenance stages are sorted from earliest to latest according to the average update time of the corresponding data update information to obtain a stage sequence; The prediction maintenance stage that is closest to the current time point in the stage sequence is identified as a candidate stage; Calculate the rate of change parameter corresponding to the stage sequence; the rate of change parameter is the ratio between the total type data and the total time span of all the predicted maintenance stages in the stage sequence; Determine whether the rate of change parameter is greater than a preset parameter threshold to obtain a first determination result; If the first judgment result is yes, the next stage in the preset stage process of the candidate stage is determined as the current maintenance stage corresponding to the medical dataset; If the first determination result is negative, the candidate stage is determined as the current maintenance stage corresponding to the medical dataset.

[0017] As an optional implementation, in a second aspect of the present invention, the specific method by which the verification module verifies the operation to be performed according to the data operation rules corresponding to the current maintenance phase to obtain a verification result includes: Determine the trained data update prediction model corresponding to the current maintenance phase; the data update prediction model is trained on a training dataset that includes multiple training execution operations performed in the current maintenance phase and corresponding data change annotations; Each sub-operation of the operation to be executed is input into the data update prediction model to obtain the predicted update data corresponding to each sub-operation; Based on the LLM model corresponding to the current maintenance phase and the predicted update data, determine the compliance level of each sub-operation; The compliance level of all the sub-operations is determined as the verification result of the operation to be executed.

[0018] As an optional implementation, in a second aspect of the invention, the verification module determines the specific method by which it determines the compliance level of each sub-operation based on the LLM model corresponding to the current maintenance phase and the predicted update data, including: For each sub-operation, the operation text description corresponding to the sub-operation and the predicted update data are combined to form the corresponding operation update text prompt. The operation update text prompt is input into the trained compliance judgment LLM corresponding to the current maintenance phase to obtain the compliance level of the sub-operation.

[0019] As an optional implementation, in a second aspect of the invention, the execution module determines, based on the verification result, the specific method of the execution strategy for performing the operation to be performed on the medical dataset, including: For each sub-operation, determine the operation data object corresponding to that sub-operation; Identify multiple historical execution terminals of the operation data object; All the historical execution terminals are sorted from high to low according to their corresponding terminal security levels to obtain a terminal sequence; The historical execution terminals in the target order of the terminal sequence are identified as the execution supervision terminals corresponding to the sub-operation; the target order is obtained by calculating the product of the total number of terminals in the terminal sequence and the target security ratio; the target security ratio is obtained by calculating the ratio between the compliance level of the sub-operation and the preset compliance standard level; The sandbox environment and data encryption verification algorithm established on the execution monitoring terminal are determined as the execution environment and execution monitoring method corresponding to this sub-operation.

[0020] A third aspect of this invention discloses another medical dataset maintenance system based on lifecycle management, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the lifecycle management-based medical dataset maintenance method disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the lifecycle management-based medical dataset maintenance method disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention obtains the current version and pending operations of the medical dataset to be maintained, determines the maintenance stage based on the current version, and verifies the pending operations according to the data operation rules corresponding to the maintenance stage to determine the execution strategy. This enables differentiated security control based on the dynamic maintenance cycle, effectively reduces the risk of misoperation of highly sensitive medical data during the maintenance process, and significantly improves the system stability and compliance audit efficiency of cross-stage data management. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a medical dataset maintenance method based on lifecycle management disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a medical dataset maintenance system based on lifecycle management disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another medical dataset maintenance system based on lifecycle management disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for maintaining medical datasets based on lifecycle management. By acquiring the current version and pending operations of the medical dataset to be maintained, and determining the maintenance stage based on the current version, the method verifies the pending operations according to the data operation rules corresponding to the maintenance stage to determine the execution strategy. This enables differentiated security control based on the dynamic maintenance cycle, effectively reducing the risk of misoperation of highly sensitive medical data during maintenance, and significantly improving the system stability and compliance audit efficiency of cross-stage data management. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a medical dataset maintenance method based on lifecycle management, as disclosed in an embodiment of the present invention. Wherein, Figure 1 The described lifecycle-management-based medical dataset maintenance method can be applied to data processing systems / data processing devices / data processing servers (including local processing servers or cloud processing servers). For example... Figure 1 As shown, this lifecycle-management-based medical dataset maintenance method may include the following operations: 101. Obtain the current version and pending operations of the medical dataset to be maintained.

[0032] Optionally, the operation to be performed can be a data acquisition operation, a data update operation, a data backup operation, a data recovery operation, or a data audit operation.

[0033] Optionally, the medical dataset can be a relational database based on MySQL or PostgreSQL, a non-relational database based on MongoDB or HBase, a data warehouse based on Hadoop, or a cloud-based medical object storage system; this invention does not impose any limitations.

[0034] Optionally, the data acquisition operation can be real-time data synchronization based on the HL7 protocol, image data retrieval based on the DICOM interface, vital sign data pass-through based on sensor network (BSN), or offline data extraction based on ETL tools. This invention does not limit the scope of the data acquisition operation.

[0035] 102. Based on the current version, determine the current maintenance phase corresponding to the medical dataset.

[0036] 103. Based on the data operation rules corresponding to the current maintenance phase, verify the operation to be executed to obtain the verification result.

[0037] 104. Based on the verification results, determine the execution strategy for performing the pending operations on the medical dataset.

[0038] Optionally, the execution strategy is used to limit the execution supervision method, execution supervision terminal, and execution environment.

[0039] As can be seen, the above-described embodiments of the invention obtain the current version and the operation to be performed of the medical dataset to be maintained, and determine the maintenance stage based on the current version. Then, the operation to be performed is verified according to the data operation rules corresponding to the maintenance stage to determine the execution strategy. This enables differentiated security control based on the dynamic maintenance cycle, effectively reduces the risk of misoperation of highly sensitive medical data during the maintenance process, and significantly improves the system stability and compliance audit efficiency of cross-stage data management.

[0040] As an optional embodiment, the step above, determining the current maintenance stage corresponding to the medical dataset based on the current version, includes: Retrieve data update information for multiple previous versions of the current version; Based on the phase prediction model, the current maintenance phase of the medical dataset is determined according to the data update information.

[0041] Optionally, the data update information can be an operation log stream (OpLog), metadata change records, data incremental index, database commit log, or transaction rollback frequency data; this invention does not impose any limitations.

[0042] As can be seen, by using the above optional embodiments, the current maintenance stage of the medical dataset is determined based on the data update information corresponding to multiple previous versions and by using a stage prediction model. This achieves automated identification and precise positioning of the dataset lifecycle. By using historical update trajectories as feature inputs, the objectivity and accuracy of stage judgment are improved, providing a reliable data-driven decision-making basis for the formulation of subsequent targeted maintenance strategies.

[0043] As an optional embodiment, the step above, determining the current maintenance stage of the medical dataset based on the stage prediction model and data update information, includes: Each data update is input into the trained stage prediction model to obtain the corresponding prediction maintenance stage; Based on the predicted maintenance phases corresponding to all data update information, determine the current maintenance phase for the medical dataset.

[0044] Optionally, the stage prediction model can be a Hidden Markov Model (HMM), a Recurrent Neural Network (RNN), a Long Short-Term Memory Network (LSTM), or a stage classifier based on logistic regression; this invention does not limit the specific model.

[0045] As can be seen, by inputting each data update information into the trained stage prediction model to output the corresponding maintenance stage prediction result and determine the current stage through the above optional embodiments, the deep learning model is used to perform deep feature extraction and pattern matching on heterogeneous update records, which effectively solves the problem of inconsistent traditional manual judgment standards and greatly improves the efficiency of automatic identification of the current system state in the scenario of large-scale dataset maintenance.

[0046] As an optional embodiment, the step above, determining the current maintenance stage of the medical dataset based on the predicted maintenance stage corresponding to all data update information, includes: The stage sequence is obtained by sorting all prediction and maintenance stages from earliest to latest according to the average update time of the corresponding data update information; The prediction maintenance stage that is closest to the current time point in the stage sequence is identified as a candidate stage; Calculate the rate of change parameter corresponding to the phase sequence; optionally, the rate of change parameter is the ratio between the total type data and the total time span of all predicted maintenance phases in the phase sequence; Determine whether the rate of change parameter is greater than a preset parameter threshold to obtain the first determination result; If the first judgment result is yes, the next stage in the preset stage process of the candidate stage will be determined as the current maintenance stage corresponding to the medical dataset. If the first judgment result is negative, the candidate stage will be determined as the current maintenance stage corresponding to the medical dataset.

[0047] Optionally, the rate of change parameter can also be the slope value of the first derivative of the prediction and maintenance phase expressed in numerical or data form, the rate of change of information entropy, the cosine similarity decay coefficient, or the time decay weight value based on the weighted average. This invention does not limit the parameter.

[0048] Optionally, the preset stage process can be in the following order: data initialization stage, rapid growth stage, stable maintenance stage, historical archiving stage, and data cancellation stage. This invention does not limit this process.

[0049] As can be seen, by performing time-series sorting on the prediction stage and combining the rate of change parameter to determine whether the candidate stage evolves into the next stage of the preset process, and by quantitatively analyzing the type density and time span of data updates, the trend of dataset evolution can be dynamically captured, ensuring that the stage determination result conforms to the actual data flow speed, thereby optimizing the logical rigor and prediction sensitivity of the maintenance process switching.

[0050] As an optional embodiment, the step above, verifying the operation to be performed according to the data operation rules corresponding to the current maintenance phase to obtain the verification result, includes: Determine the trained data corresponding to the current maintenance phase to update the prediction model; Each sub-operation of the operation to be executed is input into the data update prediction model to obtain the prediction update data corresponding to each sub-operation; Based on the LLM model corresponding to the current maintenance phase and the predicted update data, determine the compliance level of each sub-operation; The compliance level of all sub-operations is determined as the verification result of the operation to be executed.

[0051] Optionally, the data update prediction model is trained using a training dataset that includes multiple training execution operations performed in the current maintenance phase and corresponding data change annotations.

[0052] Optionally, the data update prediction model can be an autoregressive model (ARIMA), a gradient boosting decision tree (GBDT), a graph convolutional neural network (GCN), or a spatiotemporal graph neural network; this invention does not limit the model.

[0053] Optionally, this sub-operation can be a batch insertion of data records, an atomic update of a single field, a schema change of the data table structure, or an access control permission escalation request; this invention does not limit the scope of the sub-operation.

[0054] Optionally, the LLM model can be a fine-tuned model for the medical field (such as Med-LLaMA), a general pre-trained language model (such as GPT-4, Claude), or a locally deployed lightweight quantization model; this invention does not limit the scope of the model.

[0055] As can be seen, by using the above optional embodiments, the data update prediction model corresponding to the current maintenance phase is used to simulate the data changes of sub-operations and combine it with a large language model to determine the compliance level. This constructs a real-time monitoring mechanism based on behavior prediction and semantic analysis, which significantly enhances the system's accuracy in predicting and intercepting illegal or abnormal operations. It reduces the resource overhead of manual review while ensuring the security of medical data computing tasks.

[0056] As an optional embodiment, the above steps, determining the compliance level of each sub-operation based on the LLM model and predicted update data corresponding to the current maintenance phase, include: For each sub-operation, the corresponding operation text description and predicted update data are combined to form the corresponding operation update text prompt. The operation update text prompt is input into the trained compliance judgment LLM corresponding to the current maintenance phase to obtain the compliance level of the sub-operation.

[0057] As can be seen, through the above optional embodiments, the operation text description of the sub-operation and the predicted update data are combined into prompt words and input into the compliance judgment big language model to obtain the compliance level. This realizes the cross-modal semantic association between maintenance instructions and expected execution results. The logical reasoning ability of the big model is used to enhance the deep understanding of complex maintenance actions, thereby providing a higher-dimensional automated processing capability for refined compliance verification.

[0058] As an optional embodiment, the step above, determining the execution strategy for performing the operation on the medical dataset based on the verification result, includes: For each sub-operation, determine the operation data object corresponding to that sub-operation; Identify multiple historical execution terminals of the data object being manipulated; All historical execution terminals are sorted from high to low according to their corresponding terminal security levels to obtain a terminal sequence; The historical execution terminal that is in the target order on the terminal sequence is identified as the execution supervision terminal corresponding to this sub-operation; The sandbox environment and data encryption verification algorithm established on the execution monitoring terminal will be determined as the execution environment and execution monitoring method corresponding to this sub-operation.

[0059] Specifically, the execution monitoring terminal can be used to: process the operation data based on the data encryption verification algorithm before executing the sub-operation and send it to the sandbox environment of the execution monitoring terminal for execution; then judge the security and rationality of the sub-operation based on the security of the execution result, and determine to execute the sub-operation when the execution result is safe and reasonable.

[0060] Optionally, the target order is obtained by calculating the product of the total number of terminals in the terminal sequence and the target security ratio.

[0061] Optionally, the target safety ratio is obtained by calculating the ratio between the compliance level of the sub-operation and the preset compliance standard level.

[0062] Optionally, the historical execution terminal can be a mobile medical PDA, a ward nurse station PC, a doctor's mobile workstation, a hospital IDC computer room management terminal, or a remote audit handheld device; this invention does not impose any limitations.

[0063] Optionally, the sandbox environment can be a Docker-based containerized isolation environment, a KVM-based virtual machine environment, a hardware-isolated Trusted Execution Environment (TEE), or a WebAssembly sandbox; this invention does not limit the scope of the sandbox environment.

[0064] Optionally, the data encryption and verification algorithm can be a national cryptographic algorithm (SM2 / SM3 / SM4), an RSA asymmetric encryption algorithm, a SHA-256 digest algorithm, an elliptic curve cryptography (ECC) algorithm, or a blind signature algorithm; this invention does not limit the specific algorithm.

[0065] As can be seen, through the above optional embodiments, the execution supervision terminal, sandbox environment and encryption verification algorithm are automatically matched according to the compliance level of the sub-operation and the security level of the historical execution terminal. By dynamically quantifying the adaptation ratio of compliance level and terminal security capability, the adaptive alignment of supervision resources and operational risks is achieved. Under the premise of ensuring data privacy, the scheduling weight of hardware resources is optimized and the environmental defense strength is improved.

[0066] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a medical dataset maintenance system based on lifecycle management, as disclosed in an embodiment of the present invention. Figure 2 The described lifecycle-management-based medical dataset maintenance system can be applied to data processing systems / data processing devices / data processing servers (including local processing servers or cloud processing servers). For example... Figure 2 As shown, the lifecycle-management-based medical dataset maintenance system may include: The acquisition module 201 is used to acquire the current version and operations to be performed of the medical dataset to be maintained.

[0067] Module 202 is used to determine the current maintenance stage corresponding to the medical dataset based on the current version.

[0068] The verification module 203 is used to verify the operation to be executed according to the data operation rules corresponding to the current maintenance stage in order to obtain the verification result.

[0069] The execution module 204 is used to determine the execution strategy for performing the operation to be performed on the medical dataset based on the verification results.

[0070] Optionally, the execution strategy is used to limit the execution supervision method, execution supervision terminal, and execution environment.

[0071] As can be seen, the above-described embodiments of the invention obtain the current version and the operation to be performed of the medical dataset to be maintained, and determine the maintenance stage based on the current version. Then, the operation to be performed is verified according to the data operation rules corresponding to the maintenance stage to determine the execution strategy. This enables differentiated security control based on the dynamic maintenance cycle, effectively reduces the risk of misoperation of highly sensitive medical data during the maintenance process, and significantly improves the system stability and compliance audit efficiency of cross-stage data management.

[0072] As an optional embodiment, the operation to be performed is a data acquisition operation, a data update operation, a data backup operation, a data recovery operation, or a data audit operation.

[0073] As can be seen, the above optional embodiments limit the types of operations to be performed, so that the invention can be comprehensively applied to different operation verification scenarios of medical datasets, and assist in achieving more comprehensive and intelligent maintenance of medical datasets.

[0074] As an optional implementation, the determining module determines the specific method for the current maintenance phase corresponding to the medical dataset based on the current version, including: Retrieve data update information for multiple previous versions of the current version; Based on the phase prediction model, the current maintenance phase of the medical dataset is determined according to the data update information.

[0075] As can be seen, by using the above optional embodiments, the current maintenance stage of the medical dataset is determined based on the data update information corresponding to multiple previous versions and by using a stage prediction model. This achieves automated identification and precise positioning of the dataset lifecycle. By using historical update trajectories as feature inputs, the objectivity and accuracy of stage judgment are improved, providing a reliable data-driven decision-making basis for the formulation of subsequent targeted maintenance strategies.

[0076] As an optional embodiment, the determining module determines the specific method for the current maintenance stage of the medical dataset based on the stage prediction model and data update information, including: Each data update is input into the trained stage prediction model to obtain the corresponding prediction maintenance stage; Based on the predicted maintenance phases corresponding to all data update information, determine the current maintenance phase for the medical dataset.

[0077] As can be seen, by inputting each data update information into the trained stage prediction model to output the corresponding maintenance stage prediction result and determine the current stage through the above optional embodiments, the deep learning model is used to perform deep feature extraction and pattern matching on heterogeneous update records, which effectively solves the problem of inconsistent traditional manual judgment standards and greatly improves the efficiency of automatic identification of the current system state in the scenario of large-scale dataset maintenance.

[0078] As an optional embodiment, the determining module determines the specific method for determining the current maintenance stage of the medical dataset based on the predicted maintenance stage corresponding to all data update information, including: The stage sequence is obtained by sorting all prediction and maintenance stages from earliest to latest according to the average update time of the corresponding data update information; The prediction maintenance stage that is closest to the current time point in the stage sequence is identified as a candidate stage; Calculate the rate of change parameter corresponding to the phase sequence; optionally, the rate of change parameter is the ratio between the total type data and the total time span of all predicted maintenance phases in the phase sequence; Determine whether the rate of change parameter is greater than a preset parameter threshold to obtain the first determination result; If the first judgment result is yes, the next stage in the preset stage process of the candidate stage will be determined as the current maintenance stage corresponding to the medical dataset. If the first judgment result is negative, the candidate stage will be determined as the current maintenance stage corresponding to the medical dataset.

[0079] As can be seen, by performing time-series sorting on the prediction stage and combining the rate of change parameter to determine whether the candidate stage evolves into the next stage of the preset process, and by quantitatively analyzing the type density and time span of data updates, the trend of dataset evolution can be dynamically captured, ensuring that the stage determination result conforms to the actual data flow speed, thereby optimizing the logical rigor and prediction sensitivity of the maintenance process switching.

[0080] As an optional embodiment, the verification module verifies the operation to be executed according to the data operation rules corresponding to the current maintenance phase to obtain the verification result in the following specific ways: Determine the pre-trained data update prediction model corresponding to the current maintenance phase; optionally, the data update prediction model is trained using a training dataset that includes multiple training execution operations performed in the current maintenance phase and corresponding data change annotations. Each sub-operation of the operation to be executed is input into the data update prediction model to obtain the prediction update data corresponding to each sub-operation; Based on the LLM model corresponding to the current maintenance phase and the predicted update data, determine the compliance level of each sub-operation; The compliance level of all sub-operations is determined as the verification result of the operation to be executed.

[0081] As can be seen, by using the above optional embodiments, the data update prediction model corresponding to the current maintenance phase is used to simulate the data changes of sub-operations and combine it with a large language model to determine the compliance level. This constructs a real-time monitoring mechanism based on behavior prediction and semantic analysis, which significantly enhances the system's accuracy in predicting and intercepting illegal or abnormal operations. It reduces the resource overhead of manual review while ensuring the security of medical data computing tasks.

[0082] As an optional implementation, the verification module determines the specific method for the compliance level of each sub-operation based on the LLM model corresponding to the current maintenance phase and the predicted update data, including: For each sub-operation, the corresponding operation text description and predicted update data are combined to form the corresponding operation update text prompt. The operation update text prompt is input into the trained compliance judgment LLM corresponding to the current maintenance phase to obtain the compliance level of the sub-operation.

[0083] As can be seen, through the above optional embodiments, the operation text description of the sub-operation and the predicted update data are combined into prompt words and input into the compliance judgment big language model to obtain the compliance level. This realizes the cross-modal semantic association between maintenance instructions and expected execution results. The logical reasoning ability of the big model is used to enhance the deep understanding of complex maintenance actions, thereby providing a higher-dimensional automated processing capability for refined compliance verification.

[0084] As an optional embodiment, the execution module determines, based on the verification results, the specific method of the execution strategy for performing the operation to be performed on the medical dataset, including: For each sub-operation, determine the operation data object corresponding to that sub-operation; Identify multiple historical execution terminals of the data object being manipulated; All historical execution terminals are sorted from high to low according to their corresponding terminal security levels to obtain a terminal sequence; The historical execution terminals in the target order on the terminal sequence are identified as the execution supervision terminals corresponding to the sub-operation; optionally, the target order is obtained by calculating the product of the total number of terminals in the terminal sequence and the target security ratio; the target security ratio is obtained by calculating the ratio between the compliance level of the sub-operation and the preset compliance standard level; The sandbox environment and data encryption verification algorithm established on the execution monitoring terminal will be determined as the execution environment and execution monitoring method corresponding to this sub-operation.

[0085] As can be seen, through the above optional embodiments, the execution supervision terminal, sandbox environment and encryption verification algorithm are automatically matched according to the compliance level of the sub-operation and the security level of the historical execution terminal. By dynamically quantifying the adaptation ratio of compliance level and terminal security capability, the adaptive alignment of supervision resources and operational risks is achieved. Under the premise of ensuring data privacy, the scheduling weight of hardware resources is optimized and the environmental defense strength is improved.

[0086] Example 3 Please see Figure 3 , Figure 3 This is another medical dataset maintenance system based on lifecycle management disclosed in the embodiments of the present invention. Figure 3 The described lifecycle-management-based medical dataset maintenance system is applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 3 As shown, the lifecycle-management-based medical dataset maintenance system may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the lifecycle management-based medical dataset maintenance method described in Embodiment 1.

[0087] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the lifecycle management-based medical dataset maintenance method described in Embodiment 1.

[0088] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the lifecycle management-based medical dataset maintenance method described in Embodiment 1.

[0089] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0090] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0091] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0092] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0093] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0094] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0095] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0096] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0097] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0098] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0099] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0100] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0101] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0102] Finally, it should be noted that the medical dataset maintenance method and system based on lifecycle management disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for maintaining a medical dataset based on lifecycle management, characterized in that, The method includes: Get the current version and pending operations of the medical dataset to be maintained; Based on the current version, determine the current maintenance phase corresponding to the medical dataset; Based on the data operation rules corresponding to the current maintenance phase, the operation to be executed is validated to obtain a validation result, including: Determine the trained data update prediction model corresponding to the current maintenance phase; the data update prediction model is trained on a training dataset that includes multiple training execution operations performed in the current maintenance phase and corresponding data change annotations; Each sub-operation of the operation to be executed is input into the data update prediction model to obtain the predicted update data corresponding to each sub-operation; For each sub-operation, the operation text description corresponding to the sub-operation and the predicted update data are combined to form the corresponding operation update text prompt. The operation update text prompt is input into the trained compliance judgment LLM corresponding to the current maintenance phase to obtain the compliance level of the sub-operation. The compliance level of all the sub-operations is determined as the verification result of the operation to be executed. Based on the verification result, an execution strategy is determined for performing the operation to be performed on the medical dataset, including: For each sub-operation, determine the operation data object corresponding to that sub-operation; Identify multiple historical execution terminals of the operation data object; All the historical execution terminals are sorted from high to low according to their corresponding terminal security levels to obtain a terminal sequence; The historical execution terminals in the target order of the terminal sequence are identified as the execution supervision terminals corresponding to the sub-operation; the target order is obtained by calculating the product of the total number of terminals in the terminal sequence and the target security ratio; the target security ratio is obtained by calculating the ratio between the compliance level of the sub-operation and the preset compliance standard level; The sandbox environment and data encryption verification algorithm established on the execution monitoring terminal are determined as the execution environment and execution monitoring method corresponding to the sub-operation; the execution strategy is used to limit the execution monitoring method, execution monitoring terminal and execution environment.

2. The method for maintaining a medical dataset based on lifecycle management according to claim 1, characterized in that, The operation to be performed is a data acquisition operation, a data update operation, a data backup operation, a data recovery operation, or a data audit operation.

3. The method for maintaining a medical dataset based on lifecycle management according to claim 1, characterized in that, The step of determining the current maintenance stage corresponding to the medical dataset based on the current version includes: Obtain data update information corresponding to multiple previous versions of the current version; Based on the stage prediction model, the current maintenance stage corresponding to the medical dataset is determined according to the data update information.

4. The method for maintaining a medical dataset based on lifecycle management according to claim 3, characterized in that, The stage-based prediction model determines the current maintenance stage of the medical dataset based on the data update information, including: Each of the aforementioned data update information is input into the trained stage prediction model to obtain the corresponding prediction maintenance stage; Based on the predicted maintenance phase corresponding to all the data update information, determine the current maintenance phase corresponding to the medical dataset.

5. The method for maintaining a medical dataset based on lifecycle management according to claim 4, characterized in that, The step of determining the current maintenance stage of the medical dataset based on the predicted maintenance stage corresponding to all the data update information includes: The prediction maintenance stages are sorted from earliest to latest according to the average update time of the corresponding data update information to obtain a stage sequence; The prediction maintenance stage that is closest to the current time point in the stage sequence is identified as a candidate stage; Calculate the rate of change parameter corresponding to the stage sequence; the rate of change parameter is the ratio between the total type data and the total time span of all the predicted maintenance stages in the stage sequence; Determine whether the rate of change parameter is greater than a preset parameter threshold to obtain a first determination result; If the first judgment result is yes, the next stage in the preset stage process of the candidate stage is determined as the current maintenance stage corresponding to the medical dataset; If the first determination result is negative, the candidate stage is determined as the current maintenance stage corresponding to the medical dataset.

6. A medical dataset maintenance system based on lifecycle management, characterized in that, The system executes the medical dataset maintenance method based on lifecycle management as described in any one of claims 1-5, the system comprising: The acquisition module is used to obtain the current version and pending operations of the medical dataset to be maintained. The determination module is used to determine the current maintenance stage corresponding to the medical dataset based on the current version. The verification module is used to verify the operation to be executed according to the data operation rules corresponding to the current maintenance stage in order to obtain the verification result; The execution module is used to determine the execution strategy for performing the operation to be performed on the medical dataset based on the verification result; the execution strategy is used to limit the execution supervision method, execution supervision terminal and execution environment.

7. A medical dataset maintenance system based on lifecycle management, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the medical dataset maintenance method based on lifecycle management as described in any one of claims 1-5.

Images