Pretreatment method and electronic equipment for 7t ultra-high field magnetic resonance multi-time point functional brain image data, and storage medium

Abstract

This invention discloses a preprocessing method, electronic device, and storage medium for multi-timepoint functional brain imaging data from 7T ultra-high field magnetic resonance imaging (MRI). The method involves preprocessing functional images to determine a 3D functional image sequence and craniotomized mean functional images; preprocessing structural images to determine craniotomized structural images, gray matter images, and white matter images; converting the 3D functional image sequence into a registered functional image sequence based on the spatial transformation relationship between the craniotomized mean functional images and the craniotomized structural images; determining gray matter and white matter group templates based on the gray matter and white matter images of different target objects at the initial time point; registering the gray matter and white matter images of each target object at other time points to obtain a nonlinear differential homeomorphic flow field; and standardizing the registered functional image sequence using a preset standard spatial template to obtain the target functional image data. This invention improves the spatial alignment accuracy between functional and structural images and mitigates the instability problem of standardization under high field strength.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing, neuroimaging computation and computer-aided analysis technology, and in particular relates to a preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data, as well as electronic equipment and storage media. Background Technology

[0002] Magnetic resonance imaging (MRI) of the brain can reflect the brain's anatomical structure, functional activities, and related physiological and pathological states, and has been widely used in brain science, cognitive neuroscience, and clinical research. With the development of ultra-high field MRI equipment, 7T ultra-high field MRI has gradually become an important technical means for high-precision brain function research due to its higher spatial resolution and signal-to-noise ratio.

[0003] However, while 7T ultra-high field magnetic resonance imaging improves resolution, it also introduces more pronounced inhomogeneities between the B0 and B1 fields, enhanced local geometric distortion, increased cross-modal image differences, and a significant increase in data volume, especially when the functional image is an EPI sequence (echo plane imaging). This issue is particularly prominent in planar imaging. Existing neuroimage preprocessing typically relies on a single software or toolkit, such as SPM (Statistical Parametric Mapping), FSL (fMRIB Software Library), AFNI (Analysis of Functional NeuroImages), or their upper-level encapsulation tools. Although these software programs each support some preprocessing and registration operations, they still suffer from insufficient automation and unstable registration accuracy when dealing with 7T ultra-high field and multi-timepoint longitudinal data. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this invention provides a preprocessing method, electronic device, and storage medium for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data.

[0005] The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, the present invention provides a preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data, comprising: Acquire raw DICOM data of 7T high-field magnetic resonance structural images and raw 7T high-field magnetic resonance functional images of multiple target objects; Preprocessing was performed on the original DICOM data of 7T high-field magnetic resonance structural images and the original DICOM data of 7T high-field magnetic resonance functional images to obtain preprocessed structural images and preprocessed functional images. Based on the preprocessed functional images, 4D functional images and cranium-removed mean functional images of each target object are determined; and the 4D functional images are split into 3D functional image sequences along the time dimension. The preprocessed structural images are de-skeletalized to obtain de-skeletalized structural images of each target object; the de-skeletalized structural images are then segmented to obtain gray matter and white matter images of each target object. Based on the craniotomized mean functional image and craniotomized structural image, the spatial transformation parameters are determined; the 3D functional image sequence is spatially transformed based on the spatial transformation parameters to obtain the registered functional image sequence of each target object; Based on the gray matter and white matter images of different target objects at the initial time point, gray matter and white matter group templates are determined; based on the gray matter and white matter group templates, the gray matter and white matter images of each target object at other time points are registered to obtain the nonlinear differential homeomorphic flow field. Based on the preset standard space template, the registration function image sequence of each target object, and the nonlinear differential homeomorphic flow field, the target function image data of each target object in the standard space are determined.

[0006] In a second aspect, the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor, when executing a computer program stored in memory, implements the steps of the first aspect of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data.

[0007] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data of the first aspect.

[0008] This invention provides a preprocessing method, electronic device, and storage medium for 7T ultra-high field magnetic resonance imaging (MRI) multi-timepoint functional brain imaging data. The method involves preprocessing functional images to determine a 3D functional image sequence and craniotomized mean functional images; preprocessing structural images to determine craniotomized structural images, gray matter images, and white matter images; converting the 3D functional image sequence into a registration functional image sequence based on the spatial transformation relationship between the craniotomized mean functional images and the craniotomized structural images; determining gray matter and white matter group templates based on the gray matter and white matter images of different target objects at the initial time point; registering the gray matter and white matter images of each target object at other time points based on the gray matter and white matter group templates to obtain a nonlinear differential homeomorphic flow field; and finally, standardizing the registration functional image sequence to a standard space using a preset standard spatial template to obtain the target functional image data. This invention improves the spatial alignment accuracy between functional and structural images by employing a cross-modal joint high-precision registration strategy for 7T ultra-high field data, and mitigates the standardization instability caused by magnetic field inhomogeneity and local geometric distortion under high field strength conditions.

[0009] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data provided in an embodiment of the present invention. Figures 2A to 2C This is a schematic diagram of a preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data provided in an embodiment of the present invention; Figure 3 This is a structural block diagram of a preprocessing device for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data provided in an embodiment of the present invention; Figures 4A to 4C This is a schematic diagram of the graphical user interface provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0012] This invention provides a preprocessing method for 7T ultra-high field magnetic resonance imaging (MRI) multi-timepoint functional brain imaging data. See also... Figure 1 as well as Figures 2A to 2C The method includes the following steps: S10. Acquire raw DICOM data of 7T high-field magnetic resonance structural images and raw DICOM data of 7T high-field magnetic resonance functional images of multiple target objects.

[0013] For example, structural images (sMRI) are scans designed to visualize brain anatomy, with clear boundaries between gray matter, white matter, and cerebrospinal fluid. They are characterized by high spatial resolution and are single, static images without a time series. Their purpose is to provide anatomical references, segmentation, and spatial normalization for functional images. Functional images (fMRI) are scans designed to observe changes in brain blood oxygenation, reflecting neural activity (resting state / task state). They are characterized by low spatial resolution and are a complete time series. DICOM (Digital Imaging and Communications in Medicine) is an international standard for medical digital imaging and communication (ISO 12052). Its core is the unified storage, transmission, and display of CT / MRI (Magnetic Resonance Imaging) / X-ray / ultrasound images and associated metadata.

[0014] The raw input directories for the raw DICOM data of 7T high-field magnetic resonance structural images and the raw DICOM data of 7T high-field magnetic resonance functional images are specified by the user. The raw input directories must include at least anat_data folder for storing structural image DICOM data and func_data folder for storing functional image DICOM data. The raw data does not need to be pre-organized into a fixed-level directory or into the BIDS standard directory structure.

[0015] S20. Preprocess the original DICOM data of the 7T high-field magnetic resonance structural image and the original DICOM data of the 7T high-field magnetic resonance functional image to obtain the preprocessed structural image and the preprocessed functional image.

[0016] Optionally, step S20 may specifically include: S201. Based on the preset recombination rules, perform BIDS catalog conversion on the original DICOM data of the 7T high-field magnetic resonance structural image and the original DICOM data of the 7T high-field magnetic resonance functional image to obtain the first structural image and the first functional image.

[0017] For example, the raw DICOM data of 7T high-field magnetic resonance structural images and raw DICOM data of 7T high-field magnetic resonance functional images are automatically identified and their paths are parsed, and then reorganized into a data directory structure conforming to the BIDS standard according to preset reorganization rules. Specifically, the preset reorganization rules are as follows: The subfolders at all levels under the original input directory are traversed, and the target object identification information and acquisition date information in the header information of the raw DICOM data of 7T high-field magnetic resonance structural images and raw DICOM data of 7T high-field magnetic resonance functional images are read. The target object number is extracted, and multiple time point data of the same target object are identified. Then, the corresponding session number is automatically assigned according to the order of acquisition date, and the raw DICOM data of 7T high-field magnetic resonance structural images and raw DICOM data of 7T high-field magnetic resonance functional images are organized into the sub-xx / ses-xx / anat and sub-xx / ses-xx / func directories, respectively. For the raw DICOM data of 7T high-field magnetic resonance functional images, it can be further organized into task nodes in the form of sub-xx_task-xx_bold according to task type. Among them, the BIDS standard (Brain Imaging Data Structure) is a globally recognized open-source data organization standard in the field of neuroimaging, which standardizes folder structure, file names, data formats, and metadata.

[0018] S202. Perform NIfTI format conversion on the first structural image and the first functional image to obtain the preprocessed structural image and the preprocessed functional image.

[0019] For example, the NIfTI format (Neuroimaging Informatics Technology Initiative) is a universal standard format in the field of neuroimaging / medical imaging and a core data format that is mandated by the BIDS standard. It is specifically designed to store 3D / 4D brain MRI, fMRI, diffusion imaging, and other data, and solves the problems of lack of spatial orientation, file separation, and poor compatibility of the traditional Analyze format.

[0020] S30. Based on the preprocessed functional images, determine the 4D functional images and the cranial mean-reduced functional images of each target object; and split the 4D functional images into a 3D functional image sequence along the time dimension.

[0021] Optionally, in step S30, based on the preprocessed functional images, the 4D functional images and the craniotomized mean functional images of each target object are determined. Specifically, this may include: calling the DPABI process to perform time-level correction on the preprocessed functional images to obtain initial corrected functional images; performing head motion correction on the initial corrected functional images to obtain the 4D functional images and mean functional images of each target object; and performing craniotomy on the mean functional images to obtain the craniotomized mean functional images of each target object.

[0022] For example, calling DPABI (Data Processing & Analysis for (Resting) The State (BrainImaging, Brain Imaging Data Processing and Analysis Toolkit) workflow, based on user-configured parameters such as TR (Repetition Time), number of slices, reference slices, and slice order, sequentially performs temporal correction and head motion correction to obtain 4D functional images and mean functional images of each target object; then, craniotomy is performed on the mean functional image to obtain the reference functional image required for subsequent cross-modal registration, i.e., the craniotomy mean functional image.

[0023] Furthermore, the 4D functional image is split into multiple independent 3D functional image volumes along the time dimension to obtain a 3D functional image sequence, so that spatial transformation parameters can be applied in batches later.

[0024] It should be noted that the present invention does not restrict the execution order of steps S30 and S40, and they can also be executed simultaneously.

[0025] S40. Perform craniotomy on the preprocessed structural images to obtain craniotomy structural images of each target object; perform segmentation on the craniotomy structural images to obtain gray matter and white matter images of each target object.

[0026] Optionally, in step S40, the preprocessed structural image is subjected to craniotomy to obtain craniotomy structural images of each target object, which may specifically include: S401. Call the CAT12 process to perform de-craniography on the preprocessed structural image to obtain de-craniography structural images of each target object.

[0027] For example, the CAT12 related processing flow is invoked to perform high-precision craniotomy on the preprocessed structural image to obtain a craniotomy structural image containing only brain tissue, thereby reducing the impact of non-brain tissues such as skull and scalp on subsequent processing.

[0028] CAT12 (Computational Anatomy Toolbox 12) is a free and open-source brain imaging analysis toolbox based on SPM12, used for automated preprocessing and morphological measurement of structural MRI. SPM (Statistical Parametric Mapping) is the most classic and mainstream MATLAB toolbox in the field of neuroimaging, used for fMRI / MRI preprocessing and statistical analysis.

[0029] Optionally, in step S40, the decranialized image is segmented to obtain gray matter and white matter images of each target object, which may specifically include: S402. Perform bias field correction on the decranialized structural image to obtain the corrected decranialized structural image of each target object.

[0030] Specifically, due to the high magnetic field strength under 7T MRI, the bias field exhibits strong inhomogeneity. Therefore, this embodiment performs bias field correction on the craniotomy structure image to improve image quality, thereby resulting in better brain tissue segmentation. This step can be achieved using SPM or other tools; this embodiment does not impose any limitations on this.

[0031] S403. Call the SPM segmentation process to perform tissue segmentation on the corrected craniotomy image to obtain the original spatial gray matter probability map and the original spatial white matter probability map of each target object.

[0032] S404. Perform rigid body registration on the original spatial gray matter probability map and the original spatial white matter probability map to obtain the gray matter image and white matter image of each target object.

[0033] For example, the rigid registration process only performs translation and rotation operations, without scaling, stretching, or shearing operations. The gray matter image rc1 is the original spatial gray matter probability map (c1) after resampling and rigid alignment. The white matter image rc2 is the original spatial white matter probability map (c2) after the same registration process.

[0034] S50. Determine the spatial transformation parameters based on the craniotomy mean functional image and the craniotomy structural image; perform spatial transformation on the 3D functional image sequence based on the spatial transformation parameters to obtain the registration functional image sequence of each target object.

[0035] For example, using the cranial mean-removed functional image as the floating image and the cranial structural image at the corresponding time point as the reference image, the align_epi_anat.py script in the neuroimaging analysis tool AFNI (Analysis of Functional NeuroImages) is first called, and local cortical... Global affine registration is performed using the local Pearson correlation (LPC) cost function for soft membrane contrast, outputting a global affine transformation matrix and a global affine transformation mean functional image. The de-cranialized image and the global affine transformation mean functional image are then input into the nonlinear deformable registration tool 3dQwarp in AFNI, and processed through multi-scale B... Nonlinear local deformation registration is performed between splines and sinusoidal basis functions to estimate the local nonlinear deformation field. The global affine transformation matrix and the local nonlinear deformation field constitute the spatial transformation parameters. Then, the obtained spatial transformation parameters are applied in batches to all 3D functional image sequences to obtain a registered functional image sequence that is consistent with the decranial structural image space.

[0036] S60. Based on the gray matter and white matter images of different target objects at the initial time point, determine the gray matter group template and the white matter group template; based on the gray matter group template and the white matter group template, register the gray matter images and white matter images of each target object at other time points to obtain the nonlinear differential homeomorphic flow field.

[0037] Optionally, in step S60, based on the gray matter and white matter images of different target objects at the initial time point, gray matter group templates and white matter group templates are determined, which may specifically include: The DARTEL algorithm for differential homeomorphic anatomical registration was used to iteratively register gray matter and white matter images of different target objects at the initial time point to construct gray matter and white matter population templates.

[0038] For example, gray matter and white matter images corresponding to the initial time points of different target objects are extracted as input. The DARTEL (Diffeomorphic Anatomical Registration using Exponentiated Lie algebra) algorithm is used for multiple iterations to generate gray matter and white matter population templates, which serve as a common reference space for subsequent longitudinal standardization. The DARTEL algorithm is an advanced nonlinear registration tool in SPM12 for implementing brain morphological analysis.

[0039] Subsequently, the gray matter and white matter images of each target object at other time points are no longer used in group template reconstruction. Instead, based on the gray matter and white matter group templates, the gray matter and white matter images of the target objects at other time points are registered to the corresponding group templates. The flow field corresponding to the gray matter and white matter images at each time point is calculated to obtain the nonlinear differential homeomorphic flow field. Specifically, in the SPM working interface, open the DARTEL (or Shoot) toolbox and select the Run DARTEL (Create Templates) module. Import the gray matter and white matter images of all time points and all target objects into the corresponding input lists (Images Channels). Through multi-band co-registration and exponentiated Lie Algebra parameterization techniques, the iterative process of "calculating individual flow field" and "updating group template" is executed alternately. After six rounds of progressive nonlinear deformation iterations from coarse to fine, nonlinear differential homeomorphic flow fields are obtained and saved in the form of u_rc1*_Template.nii (DARTEL format) or y_*.nii (Shoot format). These flow fields accurately encode the local nonlinear deformation displacement vector from the individual anatomical structure to the population mean space at each time point.

[0040] S70. Based on the preset standard space template, the registration function image sequence of each target object, and the nonlinear differential homeomorphic flow field, determine the target function image data of each target object in the standard space.

[0041] Optionally, the default standard space template is the MNI152 template.

[0042] For example, the application spatial transformation workflow of ANTs (Advanced Normalization Tools) can be invoked to perform spatial mapping transformation on the registered functional image sequences of each target object based on the nonlinear differential homeomorphic flow field, obtaining the mapped registered functional image sequences. The resampling module in FSL (FMRIB Software Library) flirt or ANTs can be invoked to resample the mapped registered functional image sequences to the standard resolution of the MNI standard space, obtaining the target functional image data of each target object in the standard space. This invention can be implemented using the nipype framework, a Python-based development framework in neuroimaging that seamlessly integrates various mainstream neuroimaging analysis software (such as FSL, SPM, AFNI, ANTs, FreeSurfer, etc.) into a single Python workflow, thereby achieving automated scheduling and automatic connection of files and parameter information.

[0043] The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data of the present invention has the following significant advantages compared with the prior art: (1) In view of the differentiated needs of 7T ultra-high field magnetic resonance data in different stages such as structural image processing, functional image preprocessing, cross-modal registration and standardization, this invention constructs an integrated automatic processing flow composed of multiple advantageous algorithms. Among them, SPM software is used for structural image segmentation and structural image registration to MNI standard template, DPABI software is used to realize head motion correction and time layer correction of functional images, and AFNI can effectively realize the registration of functional images to structural images, avoiding the limitation of a single software in the whole process.

[0044] (2) By establishing a unified automated scheduling mechanism, this invention enables the automatic connection of input and output files, reference space and parameter transfer between heterogeneous software, reducing the complexity of researchers manually organizing processes and repetitive operations, and improving the efficiency of batch processing and the consistency of data processing for different subjects and multiple time points.

[0045] (3) This invention solves the problem of data organization caused by inconsistent original directory levels and naming methods in different experimental projects by automatically identifying the original DICOM data and reorganizing it according to the BIDS standard, and reduces the probability of human error.

[0046] (4) By adopting a cross-modal joint high-precision registration strategy for 7T ultra-high field data, this invention improves the spatial alignment accuracy between functional images and structural images and mitigates the standardization instability caused by magnetic field inhomogeneity and local geometric distortion under high field strength conditions.

[0047] Secondly, referring to Figure 3 This invention also provides a preprocessing device for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data, comprising: The acquisition module is used to acquire raw DICOM data of 7T high-field magnetic resonance structural images and raw DICOM data of 7T high-field magnetic resonance functional images of multiple target objects; The preprocessing module is used to preprocess the raw DICOM data of 7T high-field magnetic resonance structural images and the raw DICOM data of 7T high-field magnetic resonance functional images to obtain preprocessed structural images and preprocessed functional images. The functional image preprocessing module is used to determine the 4D functional image and the cranium-removed mean functional image of each target object based on the preprocessed functional image; The time-point splitting module is used to split 4D functional images into 3D functional image sequences along the time dimension; The structural image desquamation module is used to desquamate the preprocessed structural image to obtain the desquamated structural image of each target object. The structural image segmentation module is used to segment craniotomized structural images to obtain gray matter and white matter images of each target object. The cross-modal registration module is used to determine spatial transformation parameters based on the craniotomized mean functional image and the craniotomized structural image; and to perform spatial transformation on the 3D functional image sequence based on the spatial transformation parameters to obtain the registered functional image sequence of each target object. The DARTEL template creation module is used to determine the gray matter population template and the white matter population template based on the gray matter images and white matter images of different target objects at the initial time point. There is an existing template flow field generation module, which is used to register the gray matter and white matter images of each target object at other time points based on the gray matter population template and the white matter population template to obtain a nonlinear differential homeomorphic flow field. The standardization module is used to determine the target functional image data of each target object in the standard space based on the preset standard space template, the registration functional image sequence of each target object, and the nonlinear differential homeomorphic flow field.

[0048] For example, refer to Figure 3 The preprocessing module includes a BIDS catalog conversion module and an NIfTI format conversion module. The BIDS catalog conversion module and the NIfTI format conversion module are sequentially connected; the NIfTI format conversion module is connected to both the structural image de-skullization module and the functional image preprocessing module; the structural image de-skullization module is connected to the structural image segmentation module; the functional image preprocessing module is connected to the time-point splitting module; the structural image de-skullization module, the functional image preprocessing module, and the time-point splitting module are all connected to the cross-modal registration module; the structural image segmentation module is connected to the DARTEL template creation module and the existing template flow field generation module; the DARTEL template creation module is connected to the existing template flow field generation module; and the cross-modal registration module and the existing template flow field generation module are all connected to the normalization module.

[0049] Reference Figure 4A The preprocessing device for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data provided in this embodiment of the invention also includes a graphical user interface. The graphical user interface includes at least a project path configuration area, a global parameter configuration area, a subject data structure display area, and a workflow operation area in one-click processing mode.

[0050] The project path configuration area is used to receive the raw data input path and the result output path; the global parameter configuration area is used to configure the parameters required for processing; the subject data structure display area is used to display the identified target objects, time points and modal hierarchical structure; and the workflow operation area is used to execute the complete preprocessing process in a one-click manner.

[0051] like Figure 4B As shown, in the modular processing mode, the workflow operation area includes the BIDS catalog conversion module, the NIfTI format conversion module, the structural image de-skullization module, the structural image segmentation module, the functional image preprocessing module, the time point splitting module, the cross-modal registration module, the DARTEL template creation module, the existing template flow field generation module, and the standardization module.

[0052] Users can execute any module individually as needed, or execute all modules in the complete order.

[0053] like Figure 4C As shown, the global parameter configuration interface includes a basic parameter configuration page, an advanced parameter configuration page, and a software configuration page.

[0054] The basic parameter configuration page is used to set parameters related to functional image preprocessing, including the number of CPU cores, TR, number of slices, reference slices, and slice order; the advanced parameter configuration page is used to set parameters related to structural image processing and normalization, including voxel size, craniotomy parameters, and normalization output parameters; and the software configuration page is used to configure the external software path and template file path.

[0055] For details regarding the device, please refer to the steps of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data provided in the first aspect, which will not be repeated here.

[0056] Thirdly, embodiments of the present invention also provide an electronic device, such as... Figure 5 As shown, it includes a processor 501, a communication interface 502, a memory 503, and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other through the communication bus 504. Memory 503 is used to store computer programs; The processor 501, when executing the program stored in the memory 503, implements a preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data, as described in the first aspect.

[0057] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0058] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0059] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0060] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0061] The method provided in this invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc. No limitation is made herein; any electronic device that can implement this invention falls within the protection scope of this invention.

[0062] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data according to the first aspect.

[0063] For the embodiments of the device / electronic device / storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to in the description of the method embodiments.

[0064] It should be noted that the device, electronic device and storage medium in the embodiments of the present invention are respectively the device, electronic device and storage medium for applying the above-mentioned preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data. Therefore, all embodiments of the above-mentioned preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data are applicable to the device, electronic device and storage medium, and can achieve the same or similar beneficial effects.

[0065] It should be noted that the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention.

[0066] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0067] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings and the disclosure in carrying out the claimed invention. In the description of the invention, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.

[0068] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A preprocessing method for 7T ultra-high field magnetic resonance imaging (MRI) multi-timepoint functional brain imaging data, characterized in that, include: Acquire raw DICOM data of 7T high-field magnetic resonance structural images and raw 7T high-field magnetic resonance functional images of multiple target objects; Preprocessing is performed on the original DICOM data of the 7T high-field magnetic resonance structural image and the original DICOM data of the 7T high-field magnetic resonance functional image to obtain preprocessed structural images and preprocessed functional images. Based on the preprocessed functional images, determine the 4D functional images and cranial mean-reduced functional images of each target object; The 4D functional image is then split into a 3D functional image sequence along the time dimension. The preprocessed structural images are subjected to craniotomy to obtain craniotomy structural images of each target object; the craniotomy structural images are then segmented to obtain gray matter and white matter images of each target object. Based on the craniotomy mean functional image and the craniotomy structural image, spatial transformation parameters are determined; based on the spatial transformation parameters, the 3D functional image sequence is spatially transformed to obtain the registration functional image sequence of each target object; Based on the gray matter images and white matter images of different target objects at the initial time point, determine the gray matter population template and the white matter population template. Based on the gray matter population template and the white matter population template, the gray matter images and white matter images of each target object at other time points are registered to obtain a nonlinear differential homeomorphic flow field. Based on the preset standard space template, the registration functional image sequence of each target object, and the nonlinear differential homeomorphic flow field, the target functional image data of each target object in the standard space are determined.

2. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 1, characterized in that, The preprocessing of the raw DICOM data of the 7T high-field magnetic resonance structural image and the raw DICOM data of the 7T high-field magnetic resonance functional image to obtain preprocessed structural images and preprocessed functional images includes: Based on preset recombination rules, the original DICOM data of the 7T high-field magnetic resonance structural image and the original DICOM data of the 7T high-field magnetic resonance functional image are converted into BIDS catalog to obtain the first structural image and the first functional image. The first structural image and the first functional image are converted to NIfTI format to obtain a preprocessed structural image and a preprocessed functional image.

3. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 2, characterized in that, The process of determining the 4D functional image and craniotomized mean functional image of each target object based on the preprocessed functional image includes: The DPABI process is invoked to perform temporal correction on the preprocessed functional image to obtain the initial corrected functional image. Head motion correction is performed on the initial corrected functional images to obtain 4D functional images and mean functional images of each target object; The mean functional image is subjected to decranialization to obtain the decranialized mean functional image of each target object.

4. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 2, characterized in that, The segmentation process of the decranialized structural image to obtain gray matter and white matter images of each target object includes: The decranial structural images are subjected to bias field correction to obtain corrected decranial structural images of each target object; The SPM segmentation process is invoked to perform tissue segmentation on the corrected craniotomy image, resulting in the original spatial gray matter probability map and the original spatial white matter probability map of each target object. Rigid body registration is performed on the original spatial gray matter probability map and the original spatial white matter probability map to obtain gray matter images and white matter images of each target object.

5. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 1, characterized in that, The step of determining the gray matter population template and the white matter population template based on the gray matter images and white matter images of different target objects at the initial time point includes: The DARTEL algorithm for differential homeomorphic anatomical registration was used to iteratively register the gray matter images and white matter images of different target objects at the initial time point to construct gray matter population templates and white matter population templates.

6. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 1, characterized in that, The process of de-craniating the preprocessed structural image to obtain de-craniated structural images of each target object includes: The CAT12 workflow is invoked to perform decranialization on the preprocessed structural images, resulting in decranialized structural images of each target object.

7. The preprocessing method for 7T ultra-high field magnetic resonance multi-timepoint functional brain imaging data according to claim 1, characterized in that, The preset standard space template is the MNI152 template.

8. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes the computer program stored in the memory, it implements the steps of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the preprocessing method for 7T ultra-high field magnetic resonance multi-time point functional brain imaging data as described in any one of claims 1-7.

Images