Process-oriented multi-version brain image data processing method and system

Abstract

The invention discloses a process-oriented multi-version brain image data processing method and system, and the method comprises the steps: firstly determining each calculation process of brain image data according to a brain image analysis task, and generating a calculation process document according to all calculation processes; then, according to the sequence of the calculation processes of the analysis task, comparing the calculation processes in the calculation process document with the calculation processes in the historical calculation process document one by one, and when the calculation process j in the calculation process document is inconsistent with the calculation process j in the historical calculation process document, judging whether the calculation process j is consistent with the calculation process j in the historical calculation process document; if yes, modifying the first to (j-1) th calculation processes in the calculation process document according to the calculation processes in the historical calculation process document; calculating the calculation processes with empty results one by one in sequence to complete a brain image analysis task; according to the method, different versions of the brain image data can be efficiently managed, stored and retrieved, and the efficiency of brain science research work is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a process-oriented multi-version brain image data processing method and system. Background Art

[0002] With the rapid development of medical imaging technology, brain imaging data is increasingly used in neuroscience research, clinical diagnosis, and surgical planning. Brain imaging data usually contains a large amount of three-dimensional image information, which plays a vital role in the diagnosis of diseases, the formulation of treatment plans, and the evaluation of treatment effects. Therefore, the accuracy, completeness, and traceability of data are of great significance to the formulation of brain science research plans.

[0003] In the process of analyzing brain imaging data, brain imaging data samples are first collected from the brain imaging database. Due to the influence of scanning technology, scanning equipment, patient cooperation and analysis task requirements, the brain imaging data samples need to be preprocessed, and then a research plan is formulated based on the preprocessed brain imaging data samples. When the sample size of brain imaging data is relatively large, the analysis efficiency of the brain imaging data is seriously affected; especially in the process of training artificial intelligence models, the number of samples directly affects the prediction results and at the same time affects the training efficiency of the model.

[0004] In addition, each time an analysis task is performed, the brain imaging data samples need to be preprocessed once. In particular, when multiple researchers perform different analysis tasks, they need to preprocess the brain imaging data samples separately. At the same time, multiple different versions of processing files are formed in the database. Each processing file records the preprocessing process data of the brain imaging data samples in detail. As time and analysis tasks increase, it will bring a heavy data storage burden and complex data management challenges.

[0005] Since the preprocessing process of brain imaging data samples is highly similar, there is a lot of redundancy in the preprocessing process data stored in the database. How to use the historical preprocessing data in the database to improve the efficiency of brain imaging analysis tasks has become an urgent problem to be solved. Researchers have proposed using big data sharing platforms to share the results of preprocessing, but these shared data are often the final results, and lack the intermediate process. Once the needs of other researchers are not completely consistent with the preprocessing process and parameters, they must be calculated from scratch. In addition, the existing big data sharing platform adopts a result-oriented storage method, which is particularly slow in retrieving and comparing data. This method not only increases the difficulty for brain science researchers to locate and obtain specific versions of data, but also leads to low operational efficiency and is prone to errors. Especially in emergency scenarios, this inefficient retrieval method may significantly delay the formulation of research plans. Therefore, there is an urgent need for a method for processing brain imaging data to improve the efficiency of brain imaging analysis tasks. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention provides a process-oriented multi-version brain image data processing method and system, which can efficiently manage, store and retrieve different versions of brain image data, thereby improving the efficiency of brain science research.

[0007] The present invention is achieved through the following technical solutions:

[0008] A process-oriented multi-version brain image data processing method comprises the following steps:

[0009] Step 1: determine each calculation process of brain image data according to the brain image analysis task, and generate a calculation process document according to all the calculation processes;

[0010] Step 2: Compare each calculation process in the calculation process document with the calculation process in the historical calculation process document one by one according to the calculation process sequence of the analysis task. When the calculation process j in the calculation process document is inconsistent with the calculation process j in the historical calculation process document, modify the calculation processes 1 to j-1 in the calculation process document according to the calculation process in the historical calculation process document.

[0011] Step 3: Mark the results of all calculation processes after calculation process j-1 in the calculation process document as "empty";

[0012] Step 4: Calculate the calculation processes with "empty" results one by one in order to complete the brain image analysis task.

[0013] Preferably, the calculation process of brain image data in step 1 includes at least one of conversion of DICOM to Nifti, removal of the first n images, Slice Timing, head motion correction, standardization, spatial smoothing, removal of linear drift, regression covariate, frequency domain filtering and Scrubbing.

[0014] Preferably, the calculation process document includes a document number ID, original data used data, calculation steps steps and the name of each calculation process name, configured parameter names and parameter values ​​params, status status, and calculation result result.

[0015] Preferably, in step 2, the calculation process document is compared with all historical calculation process documents in the database;

[0016] When the historical calculation process document d i If the first several or all calculation processes in the calculation process document M are consistent with all the calculation processes in the calculation process document M, then according to the historical calculation process document d iThe calculation process modifies the calculation process document M to complete the current analysis task.

[0017] Preferably, the modification of the calculation process document M includes modifying the calculation result and status of the calculation process.

[0018] Preferably, the comparison method between the calculation process document in step 2 and the historical calculation process document is as follows:

[0019] Compare the data of the calculation process document M with each historical calculation process document d in the historical calculation process document set D in the database. i , when the historical calculation process document d with the same data is traversed i , the calculation process of the calculation process document M and the historical calculation process document d i The calculation process is compared in sequence;

[0020] When no historical calculation process document with the same data is traversed in the historical calculation process document collection in the database, the results of all calculation processes in the calculation process document M are marked as "empty".

[0021] Preferably, the comparison of the calculation process includes the name of the calculation process, calculation steps, configured parameters and parameter values.

[0022] Preferably, in step 4, the marked calculation process is calculated based on the data of the previous calculation process of the calculation process whose calculation result is marked as "empty", and the calculation result and state of the calculation process are modified.

[0023] A process-oriented multi-version brain image data processing system, comprising:

[0024] A process planning module is used to determine each calculation process of brain image data according to the brain image analysis task, and generate a calculation process document according to all the calculation processes;

[0025] A retrieval module is used to compare each calculation process in the calculation process document with the calculation process in the historical calculation process document one by one according to the calculation process sequence of the analysis task. When the calculation process j in the calculation process document is inconsistent with the calculation process j in the historical calculation process document, the calculation processes 1 to j-1 in the calculation process document are modified according to the calculation process in the historical calculation process document.

[0026] An iterative retrieval module is used to repeatedly execute the retrieval module, compare all historical calculation process documents with the calculation process documents in turn, and mark the results of the calculation process that have not been traversed to the calculation process in the calculation process document as "null";

[0027] The output module is used to calculate the calculation processes whose results are marked as "empty" one by one in sequence to complete the brain image analysis task.

[0028] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the process-oriented multi-version brain image data processing method when executing the computer program.

[0029] Compared with the prior art, the present invention has the following beneficial technical effects:

[0030] The present invention provides a process-oriented multi-version brain image data processing method, which adopts a data sharing method to disclose the calculation process documents of historical analysis tasks, traverse the historical calculation process documents in the database according to the calculation process of the current analysis task, and in the comparison process of the calculation process documents, compare each historical calculation process document according to the order of the calculation process, and use the historical calculation document to update the calculation process of the current calculation process document according to the comparison result, thereby avoiding the process of repeated calculation and greatly improving the processing efficiency of brain image data. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flowchart of the process-oriented multi-version brain image data processing method of the present invention;

[0032] Figure 2 The figure is a calculation flow chart of the calculation process document of the present invention. DETAILED DESCRIPTION

[0033] The present invention will be further described in detail below in conjunction with the accompanying drawings, which are intended to explain the present invention rather than to limit it.

[0034] See also Figure 1 and 2 , a process-oriented multi-version brain image data processing method, comprising the following steps:

[0035] Step 1: Generate a calculation process document corresponding to the calculation process according to the calculation process of the brain image analysis task.

[0036] According to a specific brain image analysis task, several corresponding calculation processes are planned, and then the corresponding calculation process document is generated from the entire calculation process. The content of the calculation process document at least includes the ID number of this calculation process, the original data used, and the calculation process (each calculation process consists of several calculation steps, and each calculation step includes the calculation step name, parameters, status, calculation results and other information). The content of the calculation process document is stored in JSON format.

[0037] In this embodiment, according to a specific brain image analysis task, brain image data is collected, the corresponding calculation process of the analysis task is determined, and each calculation process is organized into a calculation process document M and stored in JSON format.

[0038] like Figure 2 As shown, the calculation process document content includes the processing process document ID of the brain image analysis task, the original data used, and the calculation steps. Construct a calculation step set A = {act i |i∈[1,N]}, where N represents the number of calculation steps. Take the calculation process act from the set A one by one i , clearly define the name of the process name, the configured parameter name and parameter value params, the status status, and the calculation result result of the process, and write this information in sequence into the calculation process document M.

[0039] The computational process of brain imaging data analysis tasks includes but is not limited to: converting DICOM to Nifti, removing the first n images, Slice Timing, head motion correction, standardization, spatial smoothing, removing linear drift, regression covariates, frequency domain filtering, and Scrubbing.

[0040] DICOM conversion of brain images to Nifti. The image format generated by medical imaging equipment is DICOM, but the format of image analysis tools is Nifti. Therefore, in the process of performing image tasks, the image format must be converted first, and the image format must be converted into a format that can be recognized by the analysis tool, that is, the image in Nifti format.

[0041] Removing the first n images refers to removing the first n images in the brain imaging data.

[0042] Slice Timing is an important step in fMRI (functional magnetic resonance imaging) data preprocessing, which is used to correct the problem of inconsistent scanning time points of data in different layers of the brain. During the fMRI scanning process, due to technical limitations, the scanning time of different slices is often not strictly synchronized, which will lead to inconsistency in the time of the data, and then affect the subsequent data analysis and interpretation. Therefore, through Slice Timing correction, the scanning time of adjacent layers can be made relatively consistent, thereby improving the quality and reliability of the data.

[0043] Head motion correction of brain images is a key step in fMRI data preprocessing, which is used to reduce or eliminate the impact of head movement of subjects during scanning on data quality. During fMRI scanning, although subjects are asked to keep their heads as still as possible, they may still have slight head movements due to physiological activities such as breathing and heartbeat, as well as discomfort caused by long-term scanning. These movements will cause the position of the image to not match over time, resulting in motion artifacts, which will affect subsequent data analysis and result interpretation.

[0044] Image standardization of brain images makes brain image data from different individuals or under different acquisition conditions consistent in morphology, position or characteristics.

[0045] Spatial smoothing of brain images is used to reduce image noise, improve signal-to-noise ratio, and enhance image uniformity.

[0046] Brain imaging delinearization is used to remove linear trends in time series data, thereby more accurately analyzing other potential changes or relationships. Delinearization is particularly critical when analyzing brain imaging data (such as fMRI) because linear drift may be caused by a variety of factors, including but not limited to subject head movement, instrument temperature changes, and small fluctuations in the magnetic field.

[0047] Regression covariates of brain imaging. In the process of brain imaging processing and analysis, regression covariates may be implemented through statistical methods (such as linear regression), in which brain imaging data (such as functional connectivity, gray matter volume, etc.) are used as dependent variables, and some external factors (i.e., covariates) are used as independent variables. By performing regression analysis on these covariates, their impact on the dependent variable can be estimated and considered or removed in subsequent statistical analysis, so as to more accurately evaluate the relationship between the research variables (such as brain area activity, brain network connectivity, etc.) and the results of interest.

[0048] Frequency domain filtering of brain images converts data from the time domain (or spatial domain) to the frequency domain (frequency domain), performs filtering in the frequency domain, and then converts the processed signal back to the time domain (or spatial domain). In brain image processing, frequency domain filtering is mainly used to remove noise, smooth images, or perform feature extraction.

[0049] Scrubbing of brain images is used to eliminate the influence of whole-brain signals or data at time points with excessive head movement. Scrubbing can identify and eliminate noise data caused by excessive head movement or abnormal whole-brain signals, thereby reducing interference with subsequent data analysis.

[0050] Step 2: Compare each calculation process in the current calculation process document with the calculation process in the historical calculation process document one by one according to the calculation process sequence of the analysis task. If the calculation process j in the current calculation process document is inconsistent with the calculation process j in the historical calculation process document, modify the 1st to j-1th calculation processes in the calculation process document according to the calculation process in the historical calculation process document.

[0051] Repeat the above process, and compare all historical calculation process documents with the current calculation process document in turn. If the j-th calculation process has not been traversed, the results of all calculation processes from the j-th calculation process onwards are marked as "empty";

[0052] When all the calculation processes are traversed according to the historical calculation process documents, the current calculation process document is modified according to the historical calculation process documents to complete the brain image analysis task.

[0053] In the above comparison process, it is necessary to follow the order of the calculation process determined by the brain image analysis task, that is, the order of the calculation process of the current calculation process document must be consistent with the order of the calculation process of the historical calculation process document. If they are different, the comparison is stopped. During the comparison process, each calculation process of the two process documents is compared one by one. If the calculation data of a certain calculation process is consistent, it means that the same analysis has been performed on the brain image sample in the database before, and the analysis result can be directly used as the current analysis result.

[0054] The brain image data database pre-stores the brain image data and the calculation process documents generated in the past during the analysis of the brain image data. In this embodiment, the calculation process documents in the database are defined as historical calculation process documents.

[0055] The historical computing process document set D = {d i |i∈[1,K]}

[0056] Where K represents the number of historical calculation process documents in the database.

[0057] When performing brain image analysis tasks, a certain amount of sample data is first collected from the database, and the sample data is used as input data for the analysis task. Each input data is processed. For example, the input data is sequentially subjected to four processing steps: head motion correction, standardization, spatial smoothing, and removal of linear drift. This means that four calculation processes are performed on the input data. The above are just examples, and the specific number of calculation processes needs to be designed according to the specific analysis task.

[0058] The calculation process document includes data and steps. Data represents input data, and steps represents each calculation process, including the calculation process name, configuration parameter name and corresponding parameter value params, status and result. Status represents the status, and result represents the calculation result. The historical calculation process document also includes the actual calculation results of each calculation process.

[0059] The following is a detailed description of the traversal process of the calculation process document, as follows:

[0060] S2.1. Compare the data of the computation process document M with each historical computation process document d in the historical computation process document set D in the database. i , when the historical calculation process document d with the same data is traversed i , then execute S2.2;

[0061] When no historical calculation process document with the same data is traversed in the historical calculation process document collection in the database, the results of all calculation processes in the calculation process document M are marked as "empty".

[0062] S2.2, according to the steps of the calculation process document M, the calculation process of the calculation process document M is compared one by one with the calculation process of the historical calculation process document D;

[0063] When the historical calculation process document d i If the previous several or all calculation processes are consistent with the calculation process in the calculation process document M, then according to the historical calculation process document d i The calculation process modifies the status and calculation results of each calculation process in the calculation process document M to complete the current analysis task.

[0064] When the historical calculation process document d i The first j calculation processes of are the same as the first j calculation processes of the calculation process document M, but the j+1th calculation process is different. Then, according to the historical calculation process document d i Modify the first j calculation processes of the calculation process document M;

[0065] During the comparison process, the name of each calculation process, the configuration parameter name and the corresponding parameter value params need to be compared one by one until all are the same, then the two calculation processes are the same.

[0066] S2.3. Retrieve the historical calculation process document from the database i+1, i + 1 < K, and repeat steps S2.1 and S2.2 to traverse the (j + 1)-th to the last calculation process of the calculation process document M. When all historical calculation process documents in the database are traversed, mark the calculation process results in the calculation process document M that have not been traversed as "empty".

[0067] Step 3: Calculate the calculation processes marked as "empty" one by one in sequence to complete the brain image analysis task.

[0068] According to the comparison result of the previous step, recalculate the calculation processes with "empty" calculation process results in the calculation process document generated in step 2. After the calculation is completed, modify the "calculation result" in the current calculation step and set the corresponding "status".

[0069] According to the comparison result between the calculation process document M and the historical document set D, if the result of a certain calculation process in the steps of M is empty, it indicates that this calculation process has not been calculated in the set D. Then, find the corresponding data based on the result of the previous calculation process in the calculation process document M, calculate according to this step of the calculation process, and modify the status and result of this step of the calculation process. When the results of all calculation processes in M are not empty, add M to the set D.

[0070] On the other hand, the present invention also provides a processing system for process-oriented brain image data, including the following steps:

[0071] A process planning module, configured to determine each calculation process of the brain image data according to the brain image analysis task, and generate a calculation process document based on all calculation processes;

[0072] A retrieval module, configured to compare each calculation process in the calculation process document with the calculation processes in the historical calculation process document one by one in the order of the calculation process of the analysis task. When the j-th calculation process in the calculation process document is inconsistent with the j-th calculation process in the historical calculation process document, modify the 1st to the (j - 1)-th calculation processes in the calculation process document according to the calculation process in the historical calculation process document;

[0073] An iterative retrieval module, configured to repeatedly execute the retrieval module, compare all historical calculation process documents with the calculation process document in sequence, and mark the calculation process results that have not been traversed in the calculation process document as "empty";

[0074] An output module, configured to calculate the calculation processes marked as "empty" one by one in sequence to complete the brain image analysis task.

[0075] The process-oriented multi-version brain image data processing method provided by the present invention assigns a unique number ID to each calculation process document, and the corresponding content of the same calculation process in different calculation process documents is completely consistent, which reduces the repetitive data preprocessing process in the analysis task, and the calculation process does not need to be repeatedly calculated, which reduces the problem of repeated data occupying a large amount of storage space in the traditional management method. Secondly, in each calculation process document, the original data used, the specific calculation process of data processing, the parameters used, the parameter values, and the calculation results in each calculation process are very clear. Researchers can call the calculation results of the corresponding steps at any time according to their needs, avoid repeated calculations, and facilitate data exchange and sharing. In addition, the JSON format has good readability and writability, realizes the unification of the recording method in the brain image data processing process, and is convenient for users to expand and maintain data as needed, which can effectively increase the retrieval efficiency of brain science researchers for data.

[0076] In another embodiment of the present invention, a computer device is provided, the computer device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the process-oriented multi-version brain image data processing method.

[0077] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, an extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the process-oriented multi-version brain image data processing method in the above embodiment.

[0078] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may 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.

[0079] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0082] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A process-oriented multi-version brain image data processing method, characterized in that: The following steps are involved: Step 1: determine each calculation process of brain image data according to the brain image analysis task, and generate a calculation process document according to all the calculation processes; Step 2: Compare each calculation process in the calculation process document with the calculation process in the historical calculation process document one by one according to the calculation process sequence of the analysis task. When the calculation process j in the calculation process document is inconsistent with the calculation process j in the historical calculation process document, modify the calculation processes 1 to j-1 in the calculation process document according to the calculation process in the historical calculation process document. Step 3: Mark the results of all calculation processes after calculation process j-1 in the calculation process document as "empty"; Step 4: Calculate the calculation processes with "empty" results one by one in order to complete the brain image analysis task.

2. A process-oriented multi-version brain image data processing method according to claim 1, characterized in that: The calculation process of brain image data in step 1 includes at least one of converting DICOM to Nifti, removing the first n images, Slice Timing, head motion correction, standardization, spatial smoothing, removing linear drift, regression covariates, frequency domain filtering and Scrubbing.

3. The process-oriented multi-version brain image data processing method according to claim 1, characterized in that: The calculation process document includes the document number ID, the original data used, the calculation steps, the name of each calculation process, the configured parameter name and parameter value params, the status status, and the calculation result result.

4. The process-oriented multi-version brain image data processing method according to claim 1, characterized in that: In step 2, the calculation process document is compared with all historical calculation process documents in the database; When the historical calculation process document d i If the first several or all calculation processes in the calculation process document M are consistent with all the calculation processes in the calculation process document M, then according to the historical calculation process document d i The calculation process modifies the calculation process document M to complete the current analysis task.

5. The process-oriented multi-version brain image data processing method according to claim 4, characterized in that: The modification of the calculation process document M includes modifying the calculation result and status of the calculation process.

6. The process-oriented multi-version brain image data processing method according to claim 1, characterized in that: The comparison method between the calculation process document described in step 2 and the historical calculation process document is as follows: Compare the data of the calculation process document M with each historical calculation process document d in the historical calculation process document set D in the database. i , when the historical calculation process document d with the same data is traversed i , the calculation process of the calculation process document M and the historical calculation process document d i The calculation process is compared in sequence; When no historical calculation process document with the same data is traversed in the historical calculation process document collection in the database, the results of all calculation processes in the calculation process document M are marked as "empty".

7. The process-oriented multi-version brain image data processing method according to claim 6, characterized in that: The comparison of the calculation process includes the name of the calculation process, calculation steps, configured parameters and parameter values.

8. The process-oriented multi-version brain image data processing method according to claim 1, characterized in that: In step 4, according to the data of the previous calculation process of the calculation process whose calculation result is marked as "empty", the marked calculation process is calculated, and the calculation result and state of the calculation process are modified.

9. A process-oriented multi-version brain image data processing system, characterized in that: include: A process planning module is used to determine each calculation process of brain image data according to the brain image analysis task, and generate a calculation process document according to all the calculation processes; A retrieval module is used to compare each calculation process in the calculation process document with the calculation process in the historical calculation process document one by one according to the calculation process sequence of the analysis task. When the calculation process j in the calculation process document is inconsistent with the calculation process j in the historical calculation process document, the calculation processes 1 to j-1 in the calculation process document are modified according to the calculation process in the historical calculation process document. An iterative retrieval module is used to repeatedly execute the retrieval module, compare all historical calculation process documents with the calculation process documents in turn, and mark the results of the calculation process that have not been traversed to the calculation process in the calculation process document as "empty"; The output module is used to calculate the calculation processes whose results are marked as "empty" one by one in sequence to complete the brain image analysis task.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the process-oriented multi-version brain image data processing method according to any one of claims 1 to 8 are implemented.

Images