Brain Functional Sub-region Localization Method, Device, Computer Equipment and Storage Medium

Abstract

The present application relates to a method, device, computer device and storage medium for localizing functional sub-regions of the brain. Among them, the method for localizing functional sub-regions of the brain includes: constructing a first three-dimensional skull model of a target object based on the brain imaging data of the target object; determining the characteristic information of the brain target region of the target object based on the brain imaging data; the characteristic information includes a multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region; dividing the brain target region according to the characteristic information of the brain target region to obtain multiple functional sub-regions; determining the first coordinate information of different functional sub-regions, and localizing the functional sub-regions on the first three-dimensional skull model based on the first coordinate information. Through the present application, the problem of inability to accurately localize the target functional region is solved, and the accurate localization of the target functional region is achieved.

Description

Technical Field

[0001] This application relates to the field of medical image processing technology, and particularly to a method, device, computer device, and storage medium for localizing functional sub-regions of the brain. Background Art

[0002] In neuroscience research, non-human primates (such as macaques, marmosets, etc.) are key models for studying the prefrontal cortex in higher cognitive functions. Due to the complex anatomical structure of the prefrontal cortex, existing methods for localizing functional regions of the prefrontal cortex usually rely on regional division based on anatomical atlases. However, affected by factors such as individual differences and rough partitioning, the above methods cannot accurately localize the target functional region.

[0003] Regarding the problem that the target functional region cannot be accurately localized in the related art, no effective solution has been proposed yet. Summary of the Invention

[0004] In this embodiment, a method, device, computer device, and storage medium for localizing functional sub-regions of the brain are provided to solve the problem that the target functional region cannot be accurately localized in the related art.

[0005] In a first aspect, in this embodiment, a method for localizing functional sub-regions of the brain is provided. The method includes:

[0006] Based on the brain image data of the target object, construct a first three-dimensional skull model of the target object;

[0007] Based on the brain image data, determine the characteristic information of the brain target region of the target object; the characteristic information includes the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region;

[0008] According to the characteristic information of the brain target region, divide the brain target region to obtain multiple functional sub-regions;

[0009] Determine the first coordinate information of different functional sub-regions, and localize the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0010] In some of these embodiments, the constructing a first three-dimensional skull model of the target object based on the brain image data of the target object includes:

[0011] Obtain the brain image data of the target object; the brain image data includes the magnetic resonance image of the brain and the computed tomography scan image of the brain of the target object;

[0012] Based on the computed tomography scan image of the brain, construct a second three-dimensional skull model of the target object;

[0013] Fuse the registered brain magnetic resonance image and the brain computed tomography image;

[0014] Optimize the second three-dimensional skull model based on the fusion result to obtain the first three-dimensional skull model of the target object.

[0015] In some embodiments, based on the brain image data, determining a multimodal fusion connection matrix corresponding to the brain target region includes:

[0016] Based on the brain magnetic resonance image of the target object in the brain image data, construct a functional connection network and a structural connection network corresponding to the brain target region;

[0017] Based on the functional connection network and the structural connection network, generate the multimodal fusion connection matrix corresponding to the brain target region.

[0018] In some embodiments, the brain magnetic resonance image includes a functional magnetic resonance image and a structural magnetic resonance image; based on the brain magnetic resonance image of the target object in the brain image data, constructing a functional connection network and a structural connection network corresponding to the brain target region includes:

[0019] Analyze the functional magnetic resonance image and the structural magnetic resonance image of the target object in the brain image data to obtain the functional connection strength within the brain target region;

[0020] Based on the functional connection strength within the brain target region, construct the functional connection network corresponding to the brain target region;

[0021] Analyze the structural magnetic resonance image of the target object in the brain image data to obtain the structural connection pattern within the brain target region;

[0022] Based on the structural connection pattern within the brain target region, construct the structural connection network corresponding to the brain target region.

[0023] In some embodiments, dividing the brain target region according to the feature information of the brain target region to obtain a plurality of functional sub-regions includes:

[0024] Based on the feature information of the brain target region, identify the brain target region through a community discovery algorithm to obtain each functional-structure joint module within the brain target region;

[0025] Based on the functional connectivity heterogeneity within the functional-structure combined module and the similarity between adjacent functional-structure combined modules, the optimized functional-structure combined module is subdivided to obtain multiple functional sub-regions.

[0026] In some embodiments, the determining the first coordinate information of different functional sub-regions includes:

[0027] Obtaining the second coordinate information of each functional sub-region in the current image space;

[0028] Through a spatial transformation algorithm, the second coordinate information of each functional sub-region is transformed into the target space to obtain the first coordinate information of each functional sub-region; the target space is the space where the first three-dimensional skull model is located.

[0029] In some embodiments, the positioning of each functional sub-region on the first three-dimensional skull model based on the first coordinate information includes:

[0030] Inputting the first coordinate information of the functional sub-region into a target device to control the target device to position the functional sub-region on the first three-dimensional skull model based on the first coordinate information.

[0031] In a second aspect, a device for positioning brain functional sub-regions is provided in this embodiment. The device includes:

[0032] A construction module for constructing the first three-dimensional skull model of the target object based on the brain image data of the target object;

[0033] An analysis module for determining the characteristic information of the brain target region of the target object based on the brain image data; the characteristic information includes the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region;

[0034] A division module for dividing the brain target region according to the characteristic information of the brain target region to obtain multiple functional sub-regions;

[0035] A positioning module for determining the first coordinate information of different functional sub-regions and positioning the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0036] In a third aspect, a computer device is provided in this embodiment, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method for positioning brain functional sub-regions described in the first aspect above is implemented.

[0037] Fourthly, in this embodiment, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the brain functional sub-region localization method described in the first aspect above is implemented.

[0038] Compared with the related art, the brain functional sub-region localization method, device, computer device and storage medium provided in this embodiment construct a first three-dimensional skull model of the target object based on the brain image data of the target object; determine the characteristic information of the brain target region of the target object based on the brain image data; the characteristic information includes a multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region; divide the brain target region according to the characteristic information of the brain target region to obtain a plurality of functional sub-regions; determine the first coordinate information of different functional sub-regions, and localize the functional sub-regions on the first three-dimensional skull model based on the first coordinate information, solving the problem of being unable to accurately localize the target functional region and achieving accurate localization of the target functional region.

[0039] Details of one or more embodiments of the present application are set forth in the following drawings and description, so that other features, objects, and advantages of the present application become more concise and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0041] Figure 1 is a hardware structure block diagram of a terminal device for the brain functional sub-region localization method provided in an embodiment of the present application;

[0042] Figure 2 is a flowchart of the brain functional sub-region localization method provided in an embodiment of the present application;

[0043] Figure 3 is a flowchart of a three-dimensional skull model construction method provided in an embodiment of the present application;

[0044] Figure 4 is a flowchart of a functional sub-region division method provided in an embodiment of the present application;

[0045] Figure 5 is a flowchart of the brain functional sub-region localization method provided in a preferred embodiment of the present application;

[0046] Figure 6 is a structure block diagram of a brain functional sub-region localization device provided in an embodiment of the present application.

[0047] In the figure: 102, a processor; 104, a memory; 106, a transmission device; 108, an input / output device; 10, a construction module; 20, an analysis module; 30, a partitioning module; 40, a positioning module. Detailed implementation

[0048] To understand the purpose, technical solution, and advantages of the present application more clearly, the present application is described and explained below in conjunction with the accompanying drawings and embodiments.

[0049] Unless otherwise defined, the technical terms or scientific terms involved in the present application shall have the general meaning understood by those with ordinary skills in the technical field to which the present application belongs. In the present application, words such as "a", "one", "a kind of", "the", "these", etc. do not indicate a limitation in quantity, and they can be singular or plural. The terms "including", "containing", "having" and any variations thereof involved in the present application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device including a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent in these processes, methods, products, or devices. The terms "connected", "coupled", etc. involved in the present application do not limit to physical or mechanical connections, but may include electrical connections, whether directly connected or indirectly connected. The term "a plurality of" involved in the present application refers to two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, and B exists alone. Usually, the character " / " indicates that the associated objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in the present application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0050] The method embodiment provided in this embodiment can be executed on a terminal, a computer, or a similar computing device. For example, running on a terminal, Figure 1 is the hardware structure block diagram of the terminal for the method of localizing brain functional sub-regions in this embodiment. As Figure 1 shown, the terminal may include one or more ( Figure 1 only one is shown in the figure) processors 102 and a memory 104 for storing data. Among them, the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA. The above terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand, Figure 1The structure shown is only schematic and does not limit the structure of the above terminal. For example, the terminal may further include more or fewer components than those shown in Figure 1 or different configurations from those shown in Figure 1 .

[0051] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the brain functional sub-region localization method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implements the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely provided with respect to the processor 102, and these remote memories can be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and their combinations.

[0052] The transmission device 106 is used to receive or send data via a network. The above network includes a wireless network provided by the communication provider of the terminal. In one instance, the transmission device 106 includes a network adapter (abbreviated as NIC), which can be connected to other network devices through a base station and thus can communicate with the Internet. In one instance, the transmission device 106 may be a radio frequency (abbreviated as RF) module, which is used to communicate with the Internet wirelessly.

[0053] In this embodiment, a method for localizing brain functional sub-regions is provided. Figure 2 is a flowchart of the method for localizing brain functional sub-regions in this embodiment, as shown in Figure 2 . The process includes the following steps:

[0054] Step S210, based on the brain image data of the target object, construct the first three-dimensional skull model of the target object;

[0055] Step S220, based on the brain image data, determine the characteristic information of the brain target region of the target object; the characteristic information includes the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region;

[0056] Step S230, according to the characteristic information of the brain target region, divide the brain target region to obtain multiple functional sub-regions;

[0057] Step S240: Determine the first coordinate information of different functional sub-regions, and locate the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0058] Specifically, collect the brain imaging data of the target object. The brain imaging data includes magnetic resonance imaging (MRI) images of the target object's brain, computed tomography (CT) images of the brain, etc. The brain MRI images include high-resolution T1-weighted structural image data, T2-weighted structural image data, and resting-state functional magnetic resonance (fMRI) data, and preprocess the brain imaging data. The preprocessing operations include but are not limited to denoising, normalization, and alignment processing. For example, perform head motion correction, field strength uniformity correction, brain tissue segmentation, and skull stripping on the brain MRI image data; perform noise suppression, bone tissue segmentation, and enhancement processing on the brain CT image data; use a standardization method to perform spatial normalization on the brain MRI image data and the brain CT image data to ensure the comparability of different modality data.

[0059] Furthermore, based on the preprocessed brain imaging data, construct the first three-dimensional skull model of the target object. For example, analyze the brain CT images of the target object, accurately segment the skull image data from the image data, and convert the skull image data into the corresponding first three-dimensional skull model through a surface reconstruction algorithm, so as to construct an individualized three-dimensional skull model using the CT image data and provide an accurate spatial reference. Extract the functional magnetic resonance image and the structural magnetic resonance image of the target object from the brain imaging data, analyze and calculate the functional magnetic resonance image and the structural magnetic resonance image of the target object to obtain the functional connection strength in the target region of the brain. Based on the functional connection strength in the target region of the brain, construct a functional connection network corresponding to the target region of the brain, identify the interaction relationship between different functional regions in the target region of the brain, and combine the structural magnetic resonance image of the target object to obtain the structural connection pattern in the target region of the brain. Based on the structural connection pattern in the target region of the brain, construct a structural connection network corresponding to the target region of the brain. Then, based on the functional connection network and the structural connection network, generate a multi-modal fusion connection matrix corresponding to the target region of the brain. It should be noted that the target region of the brain is the target region to be divided and located, and the target region of the brain can be set according to actual application requirements, such as the prefrontal cortex, hippocampus, or cerebellum, etc., which are not limited here.

[0060] Based on the characteristic information of the brain target region, the brain target region is identified through a community discovery algorithm to obtain each functional-structure joint module within the brain target region. When the functional connectivity heterogeneity within the functional-structure joint module is greater than the preset coefficient of variation, the optimized functional-structure joint module is subdivided, and the similarity between adjacent functional-structure joint modules is judged. The sub-regions with similarity greater than the preset threshold are merged to obtain multiple finally divided functional sub-regions.

[0061] In addition, the brain network group atlas method can be used to finely divide the functional sub-regions of the brain target region based on the individual functional connectivity pattern and anatomical structure information. Specifically, multi-modal image processing technology is used to perform standardized preprocessing on diffusion magnetic resonance images. The preprocessing methods include denoising, eddy current distortion correction, calculating the voxel-level fiber orientation distribution probability, and cortical surface reconstruction; white matter fiber bundle parsing is performed based on the preprocessed magnetic resonance images, and whole-brain probabilistic tractography is implemented with the gray-white matter junction surface as the seed region to establish a vertex-fiber bundle connection probability model and generate a vertex-level structural connection matrix; based on the reconstructed cortical surface mesh model, the minimum path distance between vertices, i.e., the geodesic distance, is calculated, and based on the geodesic distance and the preset attenuation coefficient, a spatial neighborhood matrix is constructed through an exponential decay model; when multi-modal feature fusion is performed, the similarity network fusion algorithm is used to achieve the robust fusion of the bimodal matrix. Taking the structural connection matrix and the spatial neighborhood matrix as inputs, a vertex similarity network is constructed for each modality, and bimodal information fusion is achieved through a cross-diffusion mechanism; a non-normalized Laplacian matrix is constructed based on the weighted adjacency matrix, and a symmetric normalized form of the Laplacian matrix is constructed to define the spectral analysis framework of the weighted adjacency matrix. The Laplacian matrix is eigen-decomposed, and the first k non-trivial eigenvectors are selected from the obtained eigenvector matrix to construct a feature space to achieve low-dimensional embedding; the similarity matrix between the individual feature matrix and the group feature matrix is calculated. Based on the similarity matrix, the individual feature matrix, and the group feature matrix, a corresponding extended matrix is constructed, and the extended matrix is subjected to graph Laplacian decomposition to obtain a low-dimensional spectral space representation. Furthermore, with the individual vertex as the center, a spherical neighborhood containing multiple group-level reference vertices is established, and probability modeling is performed through the prior probability and the likelihood function to quantify the possibility that the individual vertex belongs to different labels, and the maximum a posteriori probability label is selected according to Bayes' formula to assign the optimal label to the individual vertex, thereby achieving the precise redefinition of the individual brain region by constructing a group-individual joint feature space. After the partition is completed, quantitative verification is performed from three dimensions: partition quality assessment, topological consistency, and method superiority. The partition quality assessment includes quantifying the difference degree of the internal and external connection patterns of the brain region through the structural consistency index and evaluating the spatial gradient change of the partition boundary through the boundary sharpness. The topological consistency verification can use the multi-scale Dice coefficient and the topological preservation rate, while the method superiority verification can use the improvement gain and the statistical significance test.

[0062] After that, the second coordinate information of each functional sub-region in the current image space is obtained. Through a spatial transformation algorithm, the second coordinate information of each functional sub-region is transformed into the target space to obtain the first coordinate information of each functional sub-region, and a functional sub-region atlas corresponding to the brain target region is generated. The functional sub-region atlas usually contains multiple sub-regions. For example, a functional sub-region atlas of the prefrontal cortex is generated, which contains sub-regions such as the rostral part of the superior frontal gyrus, the rostral middle part of the superior frontal gyrus, and the dorsocaudal part of the inferior frontal gyrus.

[0063] Among them, the current image space is determined by the image data used for sub-region division. For example, in the case of using MRI images for division, the current image space is the MRI space, and the target space is the space where the first three-dimensional skull model is located. For example, in the case of constructing the first three-dimensional skull model based on brain CT images, the target space is the CT space.

[0064] When performing the positioning operation, the first coordinate information of the target functional sub-region is input into the target device to control the target device to position the target functional sub-region on the first three-dimensional skull model based on the first coordinate information. Exemplarily, the first coordinate information of the target functional sub-region is input into the surgical robot control system, and the high-precision robotic arm of the robot is used to achieve automatic positioning of the target functional sub-region. At the same time, combined with infrared tracking technology or an optical positioning system, minute displacements caused by the head fixation method or environmental factors are corrected in real time to ensure navigation accuracy; the first coordinate information of the target functional sub-region is input into stereotactic devices such as a stereotactic frame system to perform precise targeted positioning of microelectrodes, fiber optic probes, or drug injection devices; in prefrontal cortex research, according to the first coordinate information of the target functional sub-region, the optimal placement position and angle of the transcranial magnetic stimulation coil on the skull surface are calculated to ensure that the magnetic field can target a specific fine sub-region of the prefrontal cortex to the greatest extent, and an electric field modeling method is used to simulate the transcranial magnetic stimulation effect to optimize the coil parameters and improve the targeting and repeatability of the stimulation.

[0065] It should be noted that the above-mentioned functional sub-region positioning method can be applied to a variety of studies, including advanced cognitive function analysis, neural regulation and brain disease research, etc. For example, in the research on advanced cognitive function of macaques such as transcranial magnetic stimulation intervention experiments, electrophysiological recording experiments and pharmacological research, surgical robots or stereotactic equipment can be used to achieve precise navigation of the target brain area, providing a high-precision positioning method for the research on the prefrontal cortex function of macaques. Among them, in transcranial magnetic stimulation intervention experiments, the above-mentioned positioning method is used to accurately locate the functional sub-regions of the prefrontal cortex, so that transcranial magnetic stimulation can be accurately applied during the macaque's cognitive tasks (such as delayed matching tasks and attention switching tasks), and the regulatory effects of different prefrontal cortex fine sub-regions on cognitive functions can be studied; in electrophysiological recording experiments, combined with microelectrode recording technology, the effects of transcranial magnetic stimulation or other intervention methods on the neural activity of the target prefrontal cortex sub-region are analyzed to provide data support for understanding the network function of the prefrontal cortex; in pharmacological research, stereotactic equipment is used to accurately inject neuromodulators into specific prefrontal cortex fine sub-regions to study the effects of neuromodulators on macaque behavior and neural activity.

[0066] In neuroscience research, non-human primates (such as macaques and marmosets) are key models for studying the role of the prefrontal cortex in advanced cognitive functions. Due to the complex anatomical structure of the prefrontal cortex, existing methods for locating functional areas of the prefrontal cortex usually rely on regional divisions based on anatomical maps, but due to factors such as individual differences and rough partitioning, the above methods cannot accurately locate the target functional area.

[0067] Compared with the prior art, the present application constructs a first three-dimensional skull model of the target object based on the brain image data of the target object; determines the characteristic information of the target brain area of the target object based on the brain image data; the characteristic information includes the multimodal fusion connection matrix and anatomical structure information corresponding to the target brain area; divides the target brain area according to the characteristic information of the target brain area to obtain multiple functional sub-areas; determines the first coordinate information of different functional sub-areas, and locates the functional sub-areas on the first three-dimensional skull model based on the first coordinate information. Based on this, by combining the multimodal brain image data of the target object, analyzing the multimodal connection pattern and anatomical structure information of the target brain area, individualized fine functional partitioning of the prefrontal cortex can be achieved, so that accurate positioning can be performed based on the partitioning results, solving the problem of being unable to accurately locate the target functional area, achieving accurate positioning of the target functional area, avoiding low positioning accuracy, ignoring individual differences, etc., and helping to improve the accuracy and repeatability of fine functional research.

[0068] In some of the embodiments, Figure 3 As shown, constructing a first three-dimensional skull model of the target object based on the brain image data of the target object in step S210 includes the following steps:

[0069] Step S211: Obtain the brain image data of the target object; the brain image data includes the magnetic resonance imaging (MRI) of the target object's brain and the computed tomography (CT) scan of the target object's brain.

[0070] Step S212: Based on the computed tomography scan of the brain, construct the second three-dimensional skull model of the target object.

[0071] Step S213: Fuse the registered magnetic resonance imaging of the brain and the computed tomography scan of the brain.

[0072] Step S214: Optimize the second three-dimensional skull model based on the fusion result to obtain the first three-dimensional skull model of the target object.

[0073] Specifically, analyze the brain CT image of the target object, accurately segment the skull image data from the image data, and convert the skull image data into the corresponding second three-dimensional skull model through a surface reconstruction algorithm.

[0074] Furthermore, adopt a rigid registration based on feature points and a non-rigid registration algorithm based on morphological features to register the brain MRI image and the brain CT image. For example, extract the positioning marks of the head bracket in the brain MRI image and the brain CT image, initially align the brain MRI image and the brain CT image through rigid body transformation, and perform non-linear registration of the brain MRI image and the brain CT image through an improved symmetric normalization algorithm.

[0075] Among them, when performing non-linear registration, mutual information is used as the similarity metric, and the prior knowledge of the target object's skull anatomy is combined for constraint. For example, the deformation field is constrained according to features such as the contour of the temporal bone and the frontal sinus, making the registration process more conform to the actual anatomical structure of the target object's skull. At the same time, a multi-resolution registration strategy is adopted to optimize the parameters, realizing the gradual refinement from 1 / 4 resolution to full resolution, and by setting reasonable regularization weights, local excessive deformation is suppressed to ensure the stability and reliability of the registration result. Thus, the registration effect is continuously optimized through multiple iterations, and finally high-precision non-linear refined registration is achieved.

[0076] It should be noted that the verification methods of the registration results include quantitative evaluation, visual verification, etc. For example, during registration verification, the skull surface in the brain CT image and the meningeal boundary in the brain MRI image are respectively extracted, and the Dice coefficient between the skull surface and the meningeal boundary is calculated. If the calculated Dice coefficient is greater than the preset threshold, it indicates that there is a high degree of overlap in the spatial positions between the skull surface in the brain CT image and the meningeal boundary in the brain MRI image, and the current registration effect is good. Or the brain CT image and the brain MRI image are superimposed and displayed to determine whether the key landmark points (such as the optic canal, cribriform plate) in the image are aligned. If the key landmark points are accurately aligned, it indicates that the registration effect is good.

[0077] After that, through the image fusion technology, the functional information of the registered brain MRI image and the structural information of the brain CT image are fused, the second three-dimensional skull model is optimized based on the fusion result, and the first three-dimensional skull model of the target object is obtained, and a visual display is provided to optimize the navigation accuracy.

[0078] Exemplarily, a dual-branch U-Net network is constructed, and the functional information of the registered brain MRI image and the brain CT image are respectively input. Among them, the encoder of the dual-branch U-Net network uses a backbone network to extract multi-scale features of the input image; in the fusion layer, a channel attention mechanism is used to perform weighted fusion on the bony features extracted from the CT data and the soft tissue features extracted from the MRI data; after being processed by the network, the fused three-dimensional volume data is output, one channel contains CT bone density information, and the other channel contains MRI gray matter signal information. When the fusion is completed, morphological closing operation is used to eliminate the fusion gap between the inner table of the skull and the meninges, and multi-planar reconstruction is performed on the fused three-dimensional volume data to generate coronal and sagittal fusion views, so as to highlight the spatial relationship between the brain target area and the skull from different angles.

[0079] Through this embodiment, the brain image data of the target object is obtained. The brain image data includes the brain magnetic resonance image and the brain computed tomography image of the target object. Based on the brain computed tomography image, the second three-dimensional skull model of the target object is constructed, the registered brain magnetic resonance image and the brain computed tomography image are fused, and the second three-dimensional skull model is optimized based on the fusion result to obtain the first three-dimensional skull model of the target object, so as to integrate the functional information and the structural information and improve the accuracy and reliability of the model.

[0080] In some of these embodiments, based on the brain image data, determining the multi-modal fusion connection matrix corresponding to the brain target area includes the following steps:

[0081] Based on the brain magnetic resonance image of the target object in the brain image data, a functional connection network and a structural connection network corresponding to the brain target area are constructed;

[0082] Generate a multimodal fusion connection matrix corresponding to the target brain region based on the functional connection network and the structural connection network.

[0083] Specifically, based on the magnetic resonance imaging (MRI) of the brain of the target object, perform a multimodal connection pattern analysis on the target brain region of the target object, obtain the functional connection strength within the target brain region through functional connection pattern analysis, and obtain the structural fiber density data within the target brain region through structural connection pattern analysis.

[0084] Furthermore, construct a functional connection network corresponding to the target brain region according to the functional connection strength within the target brain region, and construct a corresponding structural connection network according to the structural fiber density data. Perform feature fusion on the functional connection network and the structural connection network to obtain a multimodal fusion connection matrix corresponding to the target brain region. Among them, perform Z-score normalization on the obtained functional connection strength and structural fiber density data, and perform linear superposition on the normalized functional connection strength and structural fiber density data based on a preset weight ratio to generate a multimodal fusion connection matrix.

[0085] Through this embodiment, based on the magnetic resonance imaging (MRI) of the brain of the target object in the brain image data, construct a functional connection network and a structural connection network corresponding to the target brain region, and generate a multimodal fusion connection matrix corresponding to the target brain region based on the functional connection network and the structural connection network, so as to realize the multimodal connection pattern analysis of the target brain region, which helps the subsequent fine sub-region division of the target brain region to identify and define different functional sub-regions.

[0086] In some of these embodiments, the magnetic resonance imaging (MRI) of the brain includes functional magnetic resonance imaging (fMRI) and structural magnetic resonance imaging (sMRI); constructing a functional connection network and a structural connection network corresponding to the target brain region based on the magnetic resonance imaging (MRI) of the brain of the target object in the brain image data includes the following steps:

[0087] Analyze the functional magnetic resonance imaging (fMRI) and structural magnetic resonance imaging (sMRI) of the target object in the brain image data to obtain the functional connection strength within the target brain region;

[0088] Construct a functional connection network corresponding to the target brain region based on the functional connection strength within the target brain region;

[0089] Analyze the structural magnetic resonance imaging (sMRI) of the target object in the brain image data to obtain the structural connection pattern within the target brain region;

[0090] Construct a structural connection network corresponding to the target brain region based on the structural connection pattern within the target brain region.

[0091] Specifically, based on the geometric network partitioning method of structural magnetic resonance imaging, the target brain region is divided into multiple initial seed points. These seed points serve as the nodes of the functional connectivity network. Time series corresponding to various seed points are extracted from the resting-state functional magnetic resonance imaging data, and the Pearson correlation coefficients are calculated pairwise for the time series of various seed points obtained, resulting in a corresponding correlation coefficient matrix. Each element in the correlation coefficient matrix represents the functional connectivity strength between two seed points and is used to describe the functional connectivity relationship between various seed points, thereby completing the establishment of the functional connectivity network. Among them, during the process of extracting the time series, the data is subjected to band-pass filtering to limit the frequency range, such as 0.01 to 0.1 Hz.

[0092] Furthermore, diffusion tensor imaging data of the target object's brain is obtained, and the diffusion tensor imaging data of the brain is processed through a fiber tracking algorithm to reconstruct the trajectory of the white matter fiber bundle in the target brain region, so as to obtain the fiber bundle distribution of each sub-region in the target brain region and other regions of the whole brain. Based on the fiber tracking results, the fiber density between each sub-region in the target brain region and other regions of the whole brain is statistically analyzed, and the number of fiber bundles between each connection pair is calculated to complete the quantification of the fiber bundle, so as to realize the construction of the structural connectivity network.

[0093] Through this embodiment, the functional magnetic resonance imaging and structural magnetic resonance imaging of the target object in the brain imaging data are analyzed to obtain the functional connectivity strength within the target brain region. Based on the functional connectivity strength within the target brain region, a functional connectivity network corresponding to the target brain region is constructed, and the structural magnetic resonance imaging of the target object in the brain imaging data is analyzed to obtain the structural connectivity pattern within the target brain region. Based on the structural connectivity pattern within the target brain region, a structural connectivity network corresponding to the target brain region is constructed, thereby realizing the multi-modal connectivity pattern analysis of the target brain region.

[0094] In some of these embodiments, as Figure 4 shown, the step of dividing the target brain region according to the characteristic information of the target brain region to obtain multiple functional sub-regions in step S230 includes the following steps:

[0095] Step S231, based on the characteristic information of the target brain region, the target brain region is identified through a community discovery algorithm to obtain each functional-structural joint module within the target brain region;

[0096] Step S232, based on the functional connectivity heterogeneity within the functional-structural joint module and the similarity between adjacent functional-structural joint modules, the boundary-optimized functional-structural joint module is subdivided to obtain multiple functional sub-regions.

[0097] Specifically, based on the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region, the brain target region is identified by a community detection algorithm to obtain each functional-structural joint module within the brain target region. The community detection algorithm includes, but is not limited to, the Louvain community detection algorithm, spectral clustering algorithm, and stochastic block algorithm. Among them, a reasonable resolution parameter γ is preset. For example, γ is taken as 1.0 to balance the module size and quantity. At the same time, the number of iterations of the community detection algorithm is set. After reaching the preset number of iterations, the Consensus Clustering method is used to extract a partition result with higher consistency from different partition results obtained from multiple iterations.

[0098] Furthermore, the corresponding gray matter gradient is calculated based on the structural magnetic resonance imaging data, and the boundary of the functional-structural joint module is smoothed based on the gray matter gradient. In this way, through spatial constraints, the sub-region boundary is more spatially consistent with the actual anatomical structure. If it is detected that the functional connection heterogeneity within the functional-structural joint module is greater than the preset coefficient of variation, the functional-structural joint module with optimized boundary is subdivided to obtain multiple dynamically adjusted functional sub-regions. At the same time, the similarity between adjacent functional-structural joint modules is calculated, and the sub-regions with similarity greater than the preset threshold are merged to obtain each finally divided functional sub-region. Among them, the similarity calculation methods include cosine similarity, edit distance algorithm, Jaccard similarity coefficient algorithm, etc., which are not limited here.

[0099] Through this embodiment, based on the characteristic information of the brain target region, the brain target region is identified by a community detection algorithm to obtain each functional-structural joint module within the brain target region. And based on the functional connection heterogeneity within the functional-structural joint module and the similarity between adjacent functional-structural joint modules, the functional-structural joint module with optimized boundary is subdivided to obtain multiple functional sub-regions, realizing the accurate division of the functional sub-regions of the brain target region, which is helpful for the subsequent precise positioning of the target functional region.

[0100] In some of these embodiments, determining the first coordinate information of different functional sub-regions in step S240 includes the following steps:

[0101] Obtain the second coordinate information of each functional sub-region in the current image space;

[0102] Through a spatial transformation algorithm, the second coordinate information of each functional sub-region is transformed into the target space to obtain the first coordinate information of each functional sub-region; the target space is the space where the first three-dimensional skull model is located.

[0103] Specifically, determine the second coordinate information of each functional sub-region in the current image space. Through a spatial transformation algorithm, use the registration transformation matrix to transform the second coordinate information of each functional sub-region into the target space, obtain the first coordinate information of each functional sub-region, and through three-dimensional reconstruction technology, superimpose and display the functional sub-regions after coordinate transformation on the first three-dimensional skull model. Among them, the current image space is determined by the image data used to divide the sub-regions, and the target space is the space where the constructed three-dimensional skull model is located.

[0104] Exemplarily, in the case of constructing a three-dimensional skull model based on brain CT images and using MRI images for functional sub-region division, the current image space is the MRI space, and the target space is the CT space.

[0105] Through this embodiment, obtain the second coordinate information of each functional sub-region in the current image space. Through a spatial transformation algorithm, transform the second coordinate information of each functional sub-region into the target space, obtain the first coordinate information of each functional sub-region. The target space is the space where the first three-dimensional skull model is located, so as to ensure the coordinate consistency between different modality image data, which helps to improve the subsequent navigation accuracy.

[0106] In some of the embodiments, the positioning of each functional sub-region on the first three-dimensional skull model based on the first coordinate information in step S240 includes the following steps:

[0107] Input the first coordinate information of the functional sub-region into the target device to control the target device to position the functional sub-region on the first three-dimensional skull model based on the first coordinate information.

[0108] Specifically, when performing the positioning operation, input the first coordinate information of the target functional sub-region into the target device to control the target device to position the target functional sub-region on the first three-dimensional skull model based on the first coordinate information. Exemplarily, input the first coordinate information of the target functional sub-region into the surgical robot control system, and use the high-precision robotic arm of the robot to achieve automatic positioning of the target functional sub-region. At the same time, combine infrared tracking technology or an optical positioning system to correct the tiny displacements caused by the head fixation method or environmental factors in real time to ensure the navigation accuracy; input the first coordinate information of the target functional sub-region into a stereotactic device such as a stereotactic frame system, and perform precise targeted positioning on microelectrodes, fiber optic probes or drug injection devices based on the first coordinate information; in the study of the prefrontal cortex, according to the first coordinate information of the target functional sub-region, calculate the optimal placement position and angle of the transcranial magnetic stimulation coil on the skull surface to ensure that the magnetic field can target a specific fine sub-region of the prefrontal cortex to the greatest extent, and use the electric field modeling method to simulate the transcranial magnetic stimulation effect to optimize the coil parameters and improve the targeting and repeatability of the stimulation.

[0109] The above functional sub-region localization method can be applied to a variety of studies, including the analysis of advanced cognitive functions, neuromodulation, and brain disease research. For example, in studies on the advanced cognitive functions of macaques such as transcranial magnetic stimulation (TMS) intervention experiments, electrophysiological recording experiments, and pharmacological studies, with the help of a surgical robot or stereotactic equipment, precise navigation of the target brain region can be achieved, providing a high-precision localization method for the study of the function of the prefrontal cortex of macaques. Among them, in the TMS intervention experiment, the above localization method is used to accurately locate the functional sub-regions of the prefrontal cortex, so that TMS can be precisely applied during the macaque's execution of cognitive tasks (such as delayed matching tasks, attention switching tasks) to study the regulatory effects of different fine sub-regions of the prefrontal cortex on cognitive functions; in the electrophysiological recording experiment, combined with microelectrode recording technology, the effects of TMS or other intervention means on the neural activities of the target prefrontal cortex sub-region are analyzed, providing data support for understanding the network function of the prefrontal cortex; in the pharmacological study, using stereotactic equipment, the neuromodulator is precisely injected into specific fine sub-regions of the prefrontal cortex to study the effects of the neuromodulator on the behavior and neural activities of macaques.

[0110] In this embodiment, the first coordinate information of the functional sub-region is input into the target device to control the target device to locate the functional sub-region on the first three-dimensional skull model based on the first coordinate information. On this basis of constructing an individualized functional sub-region, with the help of a surgical robot or stereotactic equipment, precise navigation of the target brain region is achieved, providing a high-precision localization method for the functional study of the target brain region, thereby improving the reliability and accuracy of the research results.

[0111] The following describes and illustrates this embodiment through specific examples.

[0112] Taking the localization of the fine functional sub-regions of the prefrontal cortex of macaques as an example, brain MRI images and brain CT images of macaques are collected. The brain MRI images include structural magnetic resonance data and functional magnetic resonance data. The brain MRI image data is preprocessed, including bias field correction, skull stripping, and cortical surface reconstruction for the structural magnetic resonance data to extract the prefrontal cortex mask, head motion correction and denoising for the resting-state functional magnetic resonance data, and eddy current correction and fiber tracking for diffusion tensor imaging to generate a whole-brain white matter connection matrix. At the same time, the iterative reconstruction algorithm is used to remove metal artifacts from the brain CT image data, correct the beam hardening artifacts caused by the head fixation screws, and the spatial normalization of the brain MRI image data and brain CT image data is performed using a standardization method to ensure the comparability of different modality data.

[0113] Based on brain CT images, a second three-dimensional skull model of the macaque is generated through surface reconstruction algorithms. The positioning markers of the head stent in the brain MRI images and brain CT images are extracted. The brain MRI images and brain CT images are initially aligned through rigid body transformation, and the non-linear registration of the brain MRI images and brain CT images is performed through an improved symmetric normalization algorithm. Subsequently, through image fusion technology, the functional information of the registered brain MRI images and the structural information of the brain CT images are fused. Based on the fusion results, the second three-dimensional skull model is optimized to obtain the first three-dimensional skull model of the target object, and a visual display is provided to optimize the navigation accuracy.

[0114] Furthermore, based on the geometric network partitioning method of structural magnetic resonance images, the prefrontal cortex is divided into multiple initial seed points. These seed points serve as the nodes of the functional connection network. The time series corresponding to each seed point are extracted from the resting-state functional magnetic resonance imaging data, and the Pearson correlation coefficients are calculated pairwise for the extracted time series of each seed point to obtain the corresponding correlation coefficient matrix. Each element in the correlation coefficient matrix represents the functional connection strength between two seed points, which is used to describe the functional connection relationship between each seed point, thereby completing the establishment of the functional connection network. At the same time, the brain diffusion tensor imaging data of the target object is obtained, and the brain diffusion tensor imaging data is processed through a fiber tracking algorithm to reconstruct the running trajectories of the white matter fiber bundles in the prefrontal cortex, so as to obtain the fiber bundle distribution of each sub-region in the prefrontal cortex and other regions of the whole brain. Based on the fiber tracking results, the fiber density between each sub-region in the prefrontal cortex and other regions of the whole brain is statistically calculated, and the number of fiber bundles between each connection pair is calculated to complete the fiber bundle quantification, so as to realize the construction of the structural connection network.

[0115] After that, the obtained functional connection strength and structural fiber density data are subjected to Z-score standardization. Based on a preset weight ratio, the standardized functional connection strength and structural fiber density data are linearly superimposed to generate a multimodal fusion connection matrix. Based on the multimodal fusion connection matrix corresponding to the prefrontal cortex and the anatomical structure information, the prefrontal cortex is identified through the Louvain community detection algorithm to obtain the functional-structural joint modules within the prefrontal cortex. The corresponding gray matter gradient is calculated based on the structural magnetic resonance imaging data, and the boundaries of the functional-structural joint modules are smoothed based on the gray matter gradient. If it is detected that the functional connection heterogeneity within the functional-structural joint module is greater than the preset coefficient of variation, the functional-structural joint module with optimized boundaries is subdivided to obtain multiple dynamically adjusted functional sub-regions. At the same time, the similarity between adjacent functional-structural joint modules is calculated, and the sub-regions with similarity greater than the preset threshold are merged to obtain multiple finally divided functional sub-regions.

[0116] Obtain the second coordinate information of each functional sub-region in the magnetic resonance imaging space. Through a spatial transformation algorithm, convert the second coordinate information of each functional sub-region to the CT space to obtain the first coordinate information of each functional sub-region, and generate a prefrontal sub-region atlas including sub-regions such as the dorsolateral prefrontal cortex, orbital frontal cortex, and anterior cingulate cortex. Input the first coordinate information of the functional sub-region into target devices such as the surgical robot control system and stereotactic equipment to control the target devices to locate the functional sub-region on the first three-dimensional skull model based on the first coordinate information.

[0117] The following describes and illustrates this embodiment through preferred embodiments.

[0118] Figure 5 is the flowchart of the brain functional sub-region localization method of this preferred embodiment, as Figure 5 shown, the brain functional sub-region localization method includes the following steps:

[0119] Step S510, based on the brain computed tomography image of the target object, construct the second three-dimensional skull model of the target object;

[0120] Step S520, fuse the registered brain magnetic resonance image and brain computed tomography image;

[0121] Step S530, optimize the second three-dimensional skull model based on the fusion result to obtain the first three-dimensional skull model of the target object;

[0122] Step S540, based on the brain magnetic resonance image of the target object in the brain image data, construct a functional connection network and a structural connection network corresponding to the brain target region;

[0123] Step S550, generate a multi-modal fusion connection matrix corresponding to the brain target region based on the functional connection network and the structural connection network;

[0124] Step S560, based on the multi-modal fusion connection matrix and anatomical structure information of the brain target region, identify the brain target region through a community discovery algorithm to obtain each functional-structure joint module within the brain target region;

[0125] Step S570, based on the functional connection heterogeneity within the functional-structure joint module and the similarity between adjacent functional-structure joint modules, perform a subdivision process on the functional-structure joint module with optimized boundaries to obtain multiple functional sub-regions;

[0126] Step S580, determine the first coordinate information of different functional sub-regions, and locate the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0127] Through this embodiment, based on the brain computed tomography (CT) image of the target object, a second three-dimensional skull model of the target object is constructed. The registered brain magnetic resonance imaging (MRI) and brain CT images are fused, and the second three-dimensional skull model is optimized based on the fusion result to obtain the first three-dimensional skull model of the target object.

[0128] Further, based on the brain MRI of the target object in the brain image data, a functional connection network and a structural connection network corresponding to the brain target region are constructed. Based on the functional connection network and the structural connection network, a multimodal fusion connection matrix corresponding to the brain target region is generated. Based on the multimodal fusion connection matrix and anatomical structure information of the brain target region, the brain target region is identified by a community detection algorithm to obtain each functional-structural joint module within the brain target region. And based on the functional connection heterogeneity within the functional-structural joint module and the similarity between adjacent functional-structural joint modules, the optimized functional-structural joint module is subdivided to obtain multiple functional subregions. The first coordinate information of different functional subregions is determined, and the functional subregions on the first three-dimensional skull model are located based on the first coordinate information, solving the problem of inability to accurately locate the target functional region and achieving the accurate location of the target functional region.

[0129] It should be noted that the steps shown in the above process or the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0130] In this embodiment, a device for localizing functional subregions of the brain is also provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated here. The following terms "module", "unit", "sub-unit", etc. can be a combination of software and / or hardware that can achieve a predetermined function. Although the device described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0131] Figure 6 is the structural block diagram of the device for localizing functional subregions of the brain in this embodiment. As Figure 6 shown, the device includes:

[0132] A construction module 10, configured to construct a first three-dimensional skull model of the target object based on the brain image data of the target object;

[0133] An analysis module 20, configured to determine the characteristic information of the brain target region of the target object based on the brain image data; the characteristic information includes a multimodal fusion connection matrix and anatomical structure information corresponding to the brain target region;

[0134] A partitioning module 30, configured to partition a brain target region according to feature information of the brain target region, so as to obtain a plurality of functional sub-regions;

[0135] A positioning module 40, configured to determine first coordinate information of different functional sub-regions, and position the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0136] Through the device provided in this embodiment, based on the brain image data of the target object, a first three-dimensional skull model of the target object is constructed; based on the brain image data, feature information of the brain target region of the target object is determined; the feature information includes a multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region; the brain target region is partitioned according to the feature information of the brain target region to obtain a plurality of functional sub-regions; first coordinate information of different functional sub-regions is determined, and the functional sub-regions on the first three-dimensional skull model are positioned based on the first coordinate information, thereby solving the problem of being unable to accurately position the target functional region and achieving accurate positioning of the target functional region.

[0137] In some embodiments, a construction module 10 is configured to obtain brain image data of the target object; the brain image data includes magnetic resonance images and computed tomography images of the brain of the target object; based on the computed tomography images of the brain, a second three-dimensional skull model of the target object is constructed; the registered magnetic resonance images and computed tomography images of the brain are fused; and the second three-dimensional skull model is optimized based on the fusion result to obtain the first three-dimensional skull model of the target object.

[0138] In some embodiments, an analysis module 20 is configured to construct a functional connection network and a structural connection network corresponding to the brain target region based on the magnetic resonance images of the brain of the target object in the brain image data; and generate a multi-modal fusion connection matrix corresponding to the brain target region based on the functional connection network and the structural connection network.

[0139] In some embodiments, the analysis module 20 is configured to analyze the functional magnetic resonance images and structural magnetic resonance images of the target object in the brain image data to obtain the functional connection strength within the brain target region; construct a functional connection network corresponding to the brain target region based on the functional connection strength within the brain target region; analyze the structural magnetic resonance images of the target object in the brain image data to obtain the structural connection pattern within the brain target region; and construct a structural connection network corresponding to the brain target region based on the structural connection pattern within the brain target region.

[0140] In some of these embodiments, a partitioning module 30 is configured to identify a brain target region through a community discovery algorithm based on the feature information of the brain target region, so as to obtain each functional-structure joint module within the brain target region; and perform a subdivision process on the functional-structure joint module with optimized boundaries based on the functional connection heterogeneity within the functional-structure joint module and the similarity between adjacent functional-structure joint modules, so as to obtain a plurality of functional sub-regions.

[0141] In some of these embodiments, a positioning module 40 is configured to obtain second coordinate information of each functional sub-region in the current image space; convert the second coordinate information of each functional sub-region to a target space through a space conversion algorithm to obtain first coordinate information of each functional sub-region; the target space is the space where the first three-dimensional skull model is located.

[0142] In some of these embodiments, the positioning module 40 is configured to input the first coordinate information of the functional sub-region into a target device, so as to control the target device to position the functional sub-region on the first three-dimensional skull model based on the first coordinate information.

[0143] It should be noted that the above-mentioned various modules can be functional modules or program modules, and can be implemented either by software or by hardware. For the modules implemented by hardware, the above-mentioned various modules can be located in the same processor; or the above-mentioned various modules can also be located in different processors in any combination form.

[0144] In this embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0145] Optionally, the above-mentioned computer device may further include a transmission device and an input / output device, wherein the transmission device is connected to the above-mentioned processor, and the input / output device is connected to the above-mentioned processor.

[0146] Optionally, in this embodiment, the above-mentioned processor may be configured to execute the following steps through a computer program:

[0147] S1, construct a first three-dimensional skull model of a target object based on the brain image data of the target object;

[0148] S2, determine the feature information of the brain target region of the target object based on the brain image data; the feature information includes a multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region;

[0149] S3, divide the brain target region according to the feature information of the brain target region to obtain a plurality of functional sub-regions;

[0150] S4. Determine the first coordinate information of different functional sub-regions, and locate the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

[0151] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and alternative embodiments, and will not be repeated in this embodiment.

[0152] In addition, in combination with the brain functional sub-region localization method provided in the above embodiments, a storage medium can also be provided to implement it in this embodiment. A computer program is stored on the storage medium; when the computer program is executed by a processor, any one of the brain functional sub-region localization methods in the above embodiments is implemented.

[0153] It should be understood that the specific embodiments described here are only used to explain this application, rather than to limit it. According to the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.

[0154] Obviously, the drawings are only some examples or embodiments of this application. For those of ordinary skill in the art, this application can also be applied to other similar situations based on these drawings without creative work. In addition, it can be understood that although the work done during this development process may be complex and time-consuming, for those of ordinary skill in the art, some design, manufacturing, or production changes based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient disclosure of this application.

[0155] The term "embodiment" in this application means that the specific features, structures, or characteristics described in combination with the embodiment may be included in at least one embodiment of this application. The phrase appears in various positions in the specification does not necessarily mean the same embodiment, nor does it mean being independent or alternative to other embodiments and mutually exclusive. Those of ordinary skill in the art can clearly or implicitly understand that the embodiments described in this application can be combined with other embodiments without conflict.

[0156] The above-described embodiments only represent several implementation manners of this application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of patent protection. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of this application, several deformations and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application should be subject to the appended claims.

Claims

1. A method for localizing sub-regions of brain functions, characterized in that, The method includes: Based on the brain image data of the target object, constructing a first three-dimensional skull model of the target object; Based on the brain image data, determining the characteristic information of the brain target region of the target object; the characteristic information includes the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region; Wherein, based on the brain image data, determining the multi-modal fusion connection matrix corresponding to the brain target region includes: based on the magnetic resonance imaging of the brain of the target object in the brain image data, constructing a functional connection network and a structural connection network corresponding to the brain target region; based on the functional connection network and the structural connection network, generating the multi-modal fusion connection matrix corresponding to the brain target region; Wherein, the magnetic resonance imaging of the brain includes functional magnetic resonance imaging and structural magnetic resonance imaging; based on the magnetic resonance imaging of the brain of the target object in the brain image data, constructing a functional connection network and a structural connection network corresponding to the brain target region includes: analyzing the functional magnetic resonance imaging and the structural magnetic resonance imaging of the target object in the brain image data to obtain the functional connection intensity within the brain target region; based on the functional connection intensity within the brain target region, constructing the functional connection network corresponding to the brain target region; analyzing the structural magnetic resonance imaging of the target object in the brain image data to obtain the structural connection pattern within the brain target region; based on the structural connection pattern within the brain target region, constructing the structural connection network corresponding to the brain target region; According to the characteristic information of the brain target region, dividing the brain target region to obtain multiple functional sub-regions; Determining the first coordinate information of different functional sub-regions, and positioning the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

2. The brain functional sub-region localization method according to claim 1, characterized in that, The constructing a first three-dimensional skull model of the target object based on the brain image data of the target object includes: Obtaining the brain image data of the target object; the brain image data includes the magnetic resonance imaging of the brain and the computed tomography imaging of the brain of the target object; Based on the computed tomography imaging of the brain, constructing a second three-dimensional skull model of the target object; Fusing the registered magnetic resonance imaging of the brain and the computed tomography imaging of the brain; Optimizing the second three-dimensional skull model based on the fusion result to obtain the first three-dimensional skull model of the target object.

3. The brain functional sub-region localization method according to claim 1, wherein The dividing the brain target region to obtain multiple functional sub-regions according to the characteristic information of the brain target region includes: Based on the characteristic information of the brain target region, identifying the brain target region through a community discovery algorithm to obtain each functional-structure joint module within the brain target region; Based on the functional connection heterogeneity within the functional-structure joint module and the similarity between adjacent functional-structure joint modules, performing a subdivision process on the functional-structure joint module after boundary optimization to obtain multiple functional sub-regions.

4. The brain functional sub-region localization method according to claim 1, characterized in that, The determination of the first coordinate information of different functional sub-regions includes: Obtaining the second coordinate information of each functional sub-region in the current image space; Converting the second coordinate information of each functional sub-region to the target space through a spatial transformation algorithm to obtain the first coordinate information of each functional sub-region; the target space is the space where the first three-dimensional skull model is located.

5. The brain functional sub-region localization method according to claim 1, wherein The positioning of each functional sub-region on the first three-dimensional skull model based on the first coordinate information includes: Inputting the first coordinate information of the functional sub-region into a target device to control the target device to position the functional sub-region on the first three-dimensional skull model based on the first coordinate information.

6. A brain functional sub-region localization device, characterized in that, The device includes: A construction module for constructing the first three-dimensional skull model of the target object based on the brain image data of the target object; An analysis module for determining the characteristic information of the brain target region of the target object based on the brain image data; the characteristic information includes the multi-modal fusion connection matrix and anatomical structure information corresponding to the brain target region; The analysis module is further configured to construct a functional connection network and a structural connection network corresponding to the brain target region based on the magnetic resonance imaging of the brain of the target object in the brain image data; and generate the multi-modal fusion connection matrix corresponding to the brain target region based on the functional connection network and the structural connection network; The analysis module is further configured to analyze the functional magnetic resonance imaging and structural magnetic resonance imaging of the target object in the brain image data to obtain the functional connection strength in the brain target region; construct the functional connection network corresponding to the brain target region based on the functional connection strength in the brain target region; analyze the structural magnetic resonance imaging of the target object in the brain image data to obtain the structural connection pattern in the brain target region; construct the structural connection network corresponding to the brain target region based on the structural connection pattern in the brain target region; A division module for dividing the brain target region according to the characteristic information of the brain target region to obtain a plurality of functional sub-regions; A positioning module for determining the first coordinate information of different functional sub-regions and positioning the functional sub-regions on the first three-dimensional skull model based on the first coordinate information.

7. A computer device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps of the brain functional sub-region positioning method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the brain functional sub-region positioning method according to any one of claims 1 to 5 are implemented.

Images