A method and device for processing neuroimaging data

Abstract

The present invention provides a method and apparatus for processing neuroimaging data. The method includes: obtaining a subcortical structural partition corresponding to the brain structural imaging data based on the brain structural imaging data and a brain structure segmentation model; obtaining a level set corresponding to the brain structural imaging data based on the brain structural imaging data, the white matter partition part included in the subcortical structural partition corresponding to the brain structural imaging data, and a cerebral cortex reconstruction model, and reconstructing a cerebral cortex image based on the level set corresponding to the brain structural imaging data to obtain reconstructed cerebral cortex imaging data; performing spherical registration between the surface of the cerebral cortex corresponding to the reconstructed cerebral cortex imaging data of the brain structural imaging data and the surface of the cerebral cortex in the first standard space based on a cerebral cortex registration model to obtain a spherical surface after registration of the surface of the cerebral cortex corresponding to the brain structural imaging data. The method and apparatus for processing neuroimaging data provided by the present invention improve the accuracy of neuroimaging data processing.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to a method and device for processing neuroimaging data. Background Art

[0002] Neuroimaging is an important technology involving multiple disciplines. Brain structure and function data can be analyzed through neuroimaging. Through the processing of neuroimaging, we can clearly see the different regions, tissues and structures of the brain as well as the changes in blood flow in the brain. Based on the changes in blood flow, we can analyze the activity state of the brain under different task conditions.

[0003] At present, the neuroimaging data of some patients are quite different from those of normal people. For example, patients with brain damage, tumors, cerebral hemorrhage, patients with unclear consciousness, etc. Their brain tissue structure is different from that of normal people, which leads to the inability of the neuroimaging processing methods in the existing technology to perform correct structural segmentation, cortical reconstruction, and surface registration of the cerebral cortex; or the inability to perform correct spatial normalization. Therefore, how to propose a neuroimaging data processing method that can be applicable to the processing of patients' neuroimaging data and improve the success rate of neuroimaging data processing has become an important issue that needs to be solved in this field. Summary of the invention

[0004] In view of the problems in the prior art, the embodiments of the present invention provide a method and device for processing neuroimaging data, which can at least partially solve the problems in the prior art.

[0005] In a first aspect, the present invention provides a method for processing neuroimaging data, comprising:

[0006] Based on the brain structure image data and the brain structure segmentation model, obtaining the structural partition under the cerebral cortex corresponding to the brain structure image data; wherein the brain structure segmentation model is obtained by pre-training;

[0007] Based on the brain structure image data, the white matter partition part included in the structural partition under the cerebral cortex corresponding to the brain structure image data, and the cerebral cortex reconstruction model, a level set corresponding to the brain structure image data is obtained, and a cerebral cortex image is reconstructed based on the level set corresponding to the brain structure image data to obtain reconstructed cerebral cortex image data corresponding to the brain structure image data; wherein the cerebral cortex reconstruction model is obtained by pre-training;

[0008] Performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and the cerebral cortex registration model, to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data; wherein, the cerebral cortex registration model is pre-trained.

[0009] In a second aspect, the present invention provides a processing device for neuroimage data, comprising:

[0010] A segmentation unit, configured to obtain the structural partitions under the cerebral cortex corresponding to the brain structure image data based on the brain structure image data and the brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained;

[0011] A reconstruction unit, configured to obtain the level set corresponding to the brain structure image data based on the brain structure image data, the white matter partition part included in the structural partitions under the cerebral cortex corresponding to the brain structure image data, and the cerebral cortex reconstruction model, and reconstruct the cerebral cortex image based on the level set corresponding to the brain structure image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structure image data; wherein, the cerebral cortex reconstruction model is pre-trained;

[0012] A registration unit, configured to perform spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and the cerebral cortex registration model, to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data; wherein, the cerebral cortex registration model is pre-trained.

[0013] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored on the memory, and the processor executes the program to implement the processing method for neuroimage data described in any one of the above embodiments.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores a computer program / instructions, and when the computer program / instructions are executed by a processor, the processing method for neuroimage data described in any one of the above embodiments is implemented.

[0015] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instructions, and when the computer program / instructions are executed by a processor, the processing method for neuroimage data described in any one of the above embodiments is implemented.

[0016] The processing method and device for neuroimaging data provided by the embodiments of the present invention obtain the subcortical structural partitions corresponding to the brain structural imaging data based on the brain structural imaging data and the brain structural segmentation model; obtain the level set corresponding to the brain structural imaging data based on the brain structural imaging data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural imaging data, and the cerebral cortex reconstruction model, and reconstruct the cerebral cortex image based on the level set corresponding to the brain structural imaging data to obtain the reconstructed cerebral cortex imaging data corresponding to the brain structural imaging data; perform spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex imaging data corresponding to the brain structural imaging data and the cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural imaging data, which has higher robustness and adaptability, effectively overcomes the deficiencies of the traditional technology in processing the neuroimaging data of specific patients, and improves the accuracy of neuroimaging data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0018] Figure 1 is a schematic flowchart of the processing method for neuroimaging data provided by the first embodiment of the present invention.

[0019] Figure 2 is a schematic flowchart of the processing method for neuroimaging data provided by the second embodiment of the present invention.

[0020] Figure 3 is a schematic flowchart of the processing method for neuroimaging data provided by the third embodiment of the present invention.

[0021] Figure 4 is a schematic flowchart of the processing method for neuroimaging data provided by the fourth embodiment of the present invention.

[0022] Figure 5 is a schematic flowchart of the processing method for neuroimaging data provided by the fifth embodiment of the present invention.

[0023] Figure 6 is a schematic flowchart of the processing method for neuroimaging data provided by the sixth embodiment of the present invention.

[0024] Figure 7 is a schematic flowchart of the processing method for neuroimaging data provided by the seventh embodiment of the present invention.

[0025] Figure 8 It is a schematic structural diagram of a processing device for neuroimaging data provided by the eighth embodiment of the present invention.

[0026] Figure 9 It is a schematic structural diagram of a processing device for neuroimaging data provided by the ninth embodiment of the present invention.

[0027] Figure 10 It is a schematic structural diagram of a processing device for neuroimaging data provided by the tenth embodiment of the present invention.

[0028] Figure 11 It is a schematic structural diagram of a processing device for neuroimaging data provided by the eleventh embodiment of the present invention.

[0029] Figure 12 It is a schematic structural diagram of a processing device for neuroimaging data provided by the twelfth embodiment of the present invention.

[0030] Figure 13 It is a schematic structural diagram of a processing device for neuroimaging data provided by the thirteenth embodiment of the present invention.

[0031] Figure 14 It is a schematic structural diagram of a processing device for neuroimaging data provided by the fourteenth embodiment of the present invention.

[0032] Figure 15 It is a schematic physical structure diagram of an electronic device provided by the fifteenth embodiment of the present invention. Detailed implementation manners

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the following further describes the embodiments of the present invention in detail with reference to the accompanying drawings. Herein, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but not to limit the present invention. It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be arbitrarily combined with each other.

[0034] In the technical solutions of this application, the acquisition, storage, use, processing, etc. of data all comply with the relevant regulations of laws and regulations. The user information in the embodiments of this application is all obtained through legal and compliant channels, and the acquisition, storage, use, processing, etc. of the user information have obtained the authorization and consent of the customers.

[0035] To facilitate the understanding of the technical solutions provided by this application, the following first explains the relevant content of the technical solutions of this application.

[0036] Neuroimaging data is data that reflects brain structure and function information obtained through various neuroimaging techniques. It mainly includes structural magnetic resonance imaging (sMRI) data, functional magnetic resonance imaging (fMRI) data, etc.

[0037] In the field of medical imaging, especially in the analysis of brain structural imaging data, traditional processing techniques often show significant limitations when dealing with specific patient groups. Especially for patients with brain injuries, tumors, intracerebral hemorrhages, and those with symptoms such as unconsciousness, the generated brain structural imaging data is significantly different from that of normal people. These differences are not only reflected in the overall structure of brain tissue but also in multiple aspects such as the contrast, noise level, and morphology of brain neuroimaging. These factors cause difficulties in key processes such as structural segmentation, cortical reconstruction, and spatial standard normalization in the existing technology, resulting in a significant reduction in the accuracy and reliability of processing results.

[0038] To address the above challenges, the present invention introduces a variety of deep learning-based processing algorithms, including brain structure segmentation algorithms, cerebral cortex reconstruction algorithms, cerebral cortex registration algorithms, and spatial normalization algorithms. Through the design and performance optimization of these algorithms, the success rate and accuracy of preprocessing patient brain structural imaging data have been greatly improved, thus providing a more accurate and reliable basis for clinical decision-making.

[0039] Taking the server as the execution entity as an example below, the specific implementation process of the neuroimaging data processing method provided by the embodiments of the present invention will be described. It can be understood that the execution entity of the neuroimaging data processing method provided by the embodiments of the present invention is not limited to the server.

[0040] Figure 1 It is a schematic flowchart of the neuroimaging data processing method provided by the first embodiment of the present invention. As Figure 1 shown, the neuroimaging data processing method provided by the embodiments of the present invention includes:

[0041] S101. Based on the brain structural imaging data and the brain structure segmentation model, obtain the structural partitions under the cerebral cortex corresponding to the brain structural imaging data; wherein, the brain structure segmentation model is pre-trained.

[0042] Specifically, brain structure image data can be obtained through a Magnetic Resonance Imaging (MRI) device or a CT imaging device. For example, the brain structure image data includes various modal data such as T1-weighted data, T2-weighted data, diffusion tensor imaging, and CT. Based on the brain structure image data and the brain structure segmentation model, the server obtains the subcortical structure partitions and the extracortical structure partitions corresponding to the brain structure image data. To compare and analyze the data on the same scale, the brain structure image data is normalized so that the intensity of the brain structure image data is converted to between 0 and 1. The normalized brain structure image data is input into the brain structure segmentation model, and the subcortical structure partitions and the extracortical structure partitions corresponding to the brain structure image data are output. The subcortical structure partitions corresponding to the brain structure image data include a white matter partition part. The brain structure segmentation model can distinguish the cerebral cortex and subcortical structures in the brain structure image data. Among them, the brain structure image data can adopt T1-weighted data.

[0043] Traditional structure segmentation algorithms often rely on some assumptions, such as the shape, size, and local features of the structure. These assumptions are usually effective when dealing with normal brain structures, but when faced with patients with abnormal brain structures (such as those with brain injuries or tumors), these assumptions will fail, resulting in inaccurate segmentation. For example, the damaged area may be mislabeled as normal brain tissue, while the actual normal area may be ignored, thus affecting subsequent analysis. Different from traditional generation algorithms, the brain structure segmentation model of the present application adopts a non-generation model, enabling abnormal brain tissues (such as lesions) to be accurately labeled as non-brain tissues, while normal brain tissues are retained. The brain structure segmentation model of the present application not only improves the recognition rate of abnormal tissues but also reduces the probability of mislabeling, ensuring the accuracy of subsequent processing.

[0044] S102. Based on the brain structure image data, the white matter partition part included in the subcortical structure partitions corresponding to the brain structure image data, and the cerebral cortex reconstruction model, obtain the level set corresponding to the brain structure image data, and reconstruct the cerebral cortex image based on the level set corresponding to the brain structure image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structure image data; wherein, the cerebral cortex reconstruction model is pre-trained.

[0045] Specifically, based on the brain structural image data, the white matter partition part included in the structural partition under the cerebral cortex corresponding to the brain structural image data, and the cerebral cortex reconstruction model, the server can obtain the level set corresponding to the brain structural image data. To compare and analyze the data on the same scale, the brain structural image data is normalized so that the intensity of the brain structural image data is converted to between 0 and 1. The normalized brain structural image data and the white matter partition part included in the structural partition under the cerebral cortex corresponding to the brain structural image data are input into the cerebral cortex reconstruction model, and the level set corresponding to the brain structural image data can be output. Among them, the cerebral cortex reconstruction model is obtained by pre-training.

[0046] The server reconstructs the cerebral cortex image based on the level set corresponding to the brain structural image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structural image data. The specific process of reconstructing the cerebral cortex image based on the level set corresponding to the brain structural image data is a prior art and will not be elaborated here. The traditional cerebral cortex segmentation algorithm is very slow, and the input of the cerebral cortex reconstruction model in this application is obtained based on the output of the aforementioned brain structure segmentation model, which can reconstruct the cerebral cortex relatively quickly. Therefore, the reconstruction of the cerebral cortex in this application is faster than that using the traditional cerebral cortex segmentation algorithm, and the reconstruction efficiency of the cerebral cortex in this application is improved.

[0047] Further, to obtain a more accurate level set, the white matter partition part included in the structural partition under the cerebral cortex corresponding to the brain structural image data can be filled, and the filled white matter partition part and the normalized brain structural image data are used as the input data of the cerebral cortex reconstruction model.

[0048] The reconstruction of the cerebral cortex is an important step in understanding brain function. Traditional reconstruction algorithms are usually complex and highly dependent on the morphology of the input data. When the input data has structural abnormalities, these algorithms may require long iterative calculations, and the results are often unreliable. For some extreme cases, the reconstruction process may even be unable to complete, resulting in the valuable clinical data not being effectively utilized. The cerebral cortex reconstruction model of this application can adapt to different brain structural morphologies and sizes. Compared with traditional algorithms, it can handle more complex topological structures, quickly reconstruct a reasonable cerebral cortex, and greatly improve the processing speed and effect.

[0049] S103. Perform spherical registration of the cerebral cortex surface and the cerebral cortex surface of the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural image data and the cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural image data; among them, the cerebral cortex registration model is obtained by pre-training.

[0050] Specifically, the server performs registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and the cerebral cortex registration model, and can obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data. By performing inflation and spherical parameterization processing on the reconstructed cerebral cortex image data corresponding to the brain structure image data, the original unregistered spherical surface and sulcus and gyrus depth information corresponding to the brain structure image data can be obtained; sampling the original unregistered sulcus and gyrus depth information corresponding to the brain structure image data into the first standard space can obtain the sulcus and gyrus depth information of the cerebral cortex surface after sampling in the first standard space corresponding to the brain structure image data; inputting the spherical surface of the cerebral cortex surface in the first standard space, the sulcus and gyrus depth information of the cerebral cortex surface after sampling in the first standard space corresponding to the brain structure image data, and the template sulcus and gyrus depth information in the first standard space into the cerebral cortex registration model can obtain the deformation matrix for registering the individual space to the first standard space; according to the deformation matrix for registering the individual space to the first standard space, the original unregistered spherical surface of the cerebral cortex surface corresponding to the brain structure image data, and the spherical surface of the cerebral cortex surface in the first standard space, the deformation matrix of the original unregistered spherical surface space of the cerebral cortex surface can be obtained, and then applying the deformation matrix of the original unregistered spherical surface space of the cerebral cortex surface to the original unregistered spherical surface of the cerebral cortex surface corresponding to the brain structure image data can obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data. Among them, the first standard space is a general, standardized reference space, and the fsaverage standard space can be used. In the analysis of cerebral cortex image data, mapping the brain images of different individuals into the first standard space can achieve comparison and statistical analysis between different individuals. The individual space is opposite to the first standard space and refers to the unique spatial structure of the brain of a specific individual. The cerebral cortex registration model is obtained by pre-training.

[0051] The cerebral cortex registration model of the present application can significantly improve the robustness of the model in registration between different individuals, avoid the deviation and distortion caused by the fixed template, and thus improve the accuracy of the registration result.

[0052] The processing method for neuroimaging data provided by the embodiments of the present invention is based on brain structural imaging data and a brain structure segmentation model to obtain the subcortical structural partitions corresponding to the brain structural imaging data; based on the brain structural imaging data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural imaging data, and a cerebral cortex reconstruction model, to obtain the level set corresponding to the brain structural imaging data, and based on the level set corresponding to the brain structural imaging data, reconstruct a cerebral cortex image to obtain the reconstructed cerebral cortex image data corresponding to the brain structural imaging data; perform spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural imaging data and a cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex corresponding to the brain structural imaging data, which has higher robustness and adaptability, effectively overcomes the deficiencies of traditional technologies in processing neuroimaging data of specific patients, and improves the accuracy of neuroimaging data processing.

[0053] Figure 2 is a schematic flowchart of the processing method for neuroimaging data provided by the second embodiment of the present invention. As Figure 2 shown, on the basis of the above embodiments, further, the performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural imaging data and a cerebral cortex registration model to obtain the spherical surface after registration of the brain corresponding to the brain structural imaging data includes:

[0054] S201. Perform dilation and spherical parameterization processing on the reconstructed cerebral cortex image data corresponding to the brain structural imaging data to obtain the original unregistered spherical surface of the cerebral cortex and the gyrus and sulcus depth information corresponding to the brain structural imaging data;

[0055] Specifically, the server performing dilation and spherical parameterization processing on the reconstructed cerebral cortex image data corresponding to the brain structural imaging data can obtain the original unregistered spherical surface of the cerebral cortex and the gyrus and sulcus depth information corresponding to the brain structural imaging data. By moderately dilating the cerebral cortex, the surface details of the cortex can be presented more clearly. The cerebral cortex has a complex gyrus and sulcus structure. When no dilation processing is performed, some deeper gyri and sulci may be blocked, making it difficult to observe comprehensively. While the dilated cortex can fully unfold the gyri and sulci, enabling a more intuitive view of each region of the cerebral cortex and obtaining the gyrus and sulcus depth information. Spherical parameterization refers to the process of gradually adjusting the shape of the cerebral cortex to be approximately spherical. Through spherical parameterization processing, the shape of the cerebral cortex can be made more regular, facilitating subsequent analysis and processing.

[0056] S202. Perform first standard space sampling based on the spherical surface of the cerebral cortex in the first standard space, the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the depth information of the originally unregistered sulci and gyri on the cerebral cortex surface corresponding to the brain structure image data, to obtain the depth information of the sulci and gyri on the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data;

[0057] Specifically, the server performs first standard space sampling based on the spherical surface of the cerebral cortex in the first standard space, the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the depth information of the originally unregistered sulci and gyri on the cerebral cortex surface corresponding to the brain structure image data, and maps the depth information of the originally unregistered sulci and gyri on the cerebral cortex surface corresponding to the brain structure image data to the first standard space, so as to obtain the depth information of the sulci and gyri on the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data. Among them, the first standard space can adopt the fsaverage standard space.

[0058] S203. Obtain the deformation matrix for registering the individual space to the first standard space based on the spherical surface of the cerebral cortex in the first standard space, the depth information of the sulci and gyri on the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data, the depth information of the template sulci and gyri in the first standard space, and the cerebral cortex registration model.

[0059] Specifically, the server inputs the spherical surface of the cerebral cortex in the first standard space, the depth information of the sulci and gyri on the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data, and the depth information of the template sulci and gyri in the first standard space into the cerebral cortex registration model, and can output the deformation matrix for registering the individual space to the first standard space.

[0060] S204. Upsample to obtain the deformation matrix in the space of the originally unregistered spherical surface of the cerebral cortex according to the deformation matrix for registering the individual space to the first standard space, the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the spherical surface of the cerebral cortex in the first standard space; obtain the registered spherical surface of the cerebral cortex corresponding to the brain structure image data according to the deformation matrix in the space of the originally unregistered spherical surface of the cerebral cortex and the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data.

[0061] Specifically, based on the deformation matrix of the individual space registered to the first standard space, the original unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the spherical surface of the cerebral cortex in the first standard space, the server can upsample to obtain the deformation matrix of the original unregistered spherical surface space of the cerebral cortex, and then apply the deformation matrix of the original unregistered spherical surface space of the cerebral cortex to the original unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, so as to obtain the registered spherical surface of the cerebral cortex corresponding to the brain structure image data.

[0062] Figure 3 FIG. 4 is a schematic flowchart of a method for processing neuroimage data provided in the third embodiment of the present invention. As Figure 3 shown, on the basis of the above embodiments, further, the step of training the brain structure segmentation model includes:

[0063] S301. Obtain first training data according to the first brain structure image sample data, the cerebral cortex structure and the subcortical structure partition corresponding to the first brain structure image sample data;

[0064] Specifically, historical brain structure image data can be collected, and the first preset number of historical brain structure image data can be obtained therefrom as the first brain structure image sample data. The first brain structure image sample data is normalized, and the intensity of the first brain structure image sample data can be converted to between 0 and 1. Using a neuroimage preprocessing tool to perform structural partition data annotation processing on the first brain structure image sample data, the cerebral cortex structure partition and the subcortical structure partition of the first brain structure image sample data can be obtained. The normalized first brain structure image sample data, the cerebral cortex structure partition and the subcortical structure partition of the first brain structure image sample data are used as the first training data.

[0065] S302. Train the brain structure segmentation model according to the first training data and the first original model.

[0066] Specifically, the server can train the first original model through the first training data to obtain the brain structure segmentation model. Among them, the first original model can adopt FastSurferCNN. FastSurferCNN is a segmentation model based on deep learning and uses a U-Net segmentation network. FastSurferCNN can quickly process a large amount of neuroimage data. Compared with traditional methods, it can complete the segmentation and analysis of brain structures in a shorter time, improving the efficiency of research and clinical applications.

[0067] Since the first training data includes the cerebral cortex structure partitions corresponding to the first brain structure image sample data and the subcortical structure partitions, and uses brain structure image data of different modalities and different resolutions, it can enhance the generalization ability and robustness of the brain structure segmentation model.

[0068] Figure 4 It is a schematic flowchart of the method for processing neuroimaging data provided in the fourth embodiment of the present invention. As Figure 4 shown, on the basis of the above embodiments, further, the steps of training the cerebral cortex reconstruction model include:

[0069] S401. Obtain second training data according to the second brain structure image sample data, the white matter partition part in the subcortical structure partition corresponding to the second brain structure image sample data, and the label corresponding to the white matter partition part;

[0070] Specifically, historical brain structure image data can be collected, and a second preset number of historical brain structure image data can be obtained therefrom as the second brain structure image sample data. The second brain structure image sample data is normalized, and the intensity of the second brain structure image sample data can be converted to between 0 and 1.

[0071] Traditional neuroimaging preprocessing tools can be used to perform structural partition data annotation processing on the second brain structure image sample data, and the subcortical structure partition corresponding to the second brain structure image sample data can be obtained. The subcortical structure partition includes the white matter partition part and the label corresponding to the white matter partition part. In addition, after the brain structure segmentation model is trained, according to the second brain structure image sample data and the brain structure segmentation model, the white matter partition part in the subcortical structure partition corresponding to the second brain structure image sample data can be obtained. The normalized second brain structure image sample data, the white matter partition part in the subcortical structure partition corresponding to the second brain structure image sample data, and the label corresponding to the white matter partition part are used as the second training data.

[0072] S402. Train and obtain the cerebral cortex reconstruction model according to the second training data and the second original model.

[0073] Specifically, the server trains the second original model through the second training data, and can train and obtain the cerebral cortex reconstruction model. Among them, the second original model can adopt a deep learning model with a 3D-U-Net structure.

[0074] Figure 5 It is a schematic flowchart of the method for processing neuroimaging data provided in the fifth embodiment of the present invention. As Figure 5As shown, on the basis of the above embodiments, further, the steps of training the cerebral cortex registration model include:

[0075] S501. Obtain cerebral cortex image sample data;

[0076] Specifically, historical cerebral cortex image data can be collected, and a third preset number of cerebral cortex image data can be obtained therefrom as the cerebral cortex image sample data.

[0077] S502. Perform inflation and spherical parameterization processing on the cerebral cortex image sample data to obtain the original unregistered spherical surface and gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data;

[0078] Specifically, when the server performs inflation and spherical parameterization processing on the cerebral cortex image sample data, the original unregistered spherical surface of the cerebral cortex corresponding to the cerebral cortex image sample data and the original unregistered gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data can be obtained.

[0079] S503. Perform first standard space sampling according to the spherical surface of the cerebral cortex surface in the first standard space, the original unregistered spherical surface of the cerebral cortex corresponding to the cerebral cortex image sample data, and the original unregistered gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data, to obtain the gyrus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the cerebral cortex image sample data; and use the spherical surface of the cerebral cortex surface in the first standard space, the template gyrus depth information in the first standard space, the gyrus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the cerebral cortex image sample data, and the corresponding partition label data as the third training data;

[0080] Specifically, when the server maps the original unregistered gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data to the first standard space, the left and right brain sampling information after the first standard space sampling corresponding to the cerebral cortex image sample data can be obtained. The server obtains the labels corresponding to the left and right brain sampling information after the standard space sampling corresponding to the cerebral cortex image sample data, and uses the spherical surface of the cerebral cortex surface in the first standard space, the template gyrus depth information in the first standard space, the gyrus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the cerebral cortex image sample data, and the corresponding partition label data as the third training data.

[0081] S504. Train and obtain the cerebral cortex registration model according to the third training data and the third original model; wherein, in the process of training and obtaining the cerebral cortex registration model, the loss value corresponding to the third training data is calculated based on an improved loss function.

[0082] Specifically, the server trains the third original model with the third training data to obtain the cerebral cortex reconstruction model. The third original model can adopt spherical geometric neural network deep learning models based on the U-Net structure, such as S-GAT and Sphrical U-Net.

[0083] During the process of training to obtain the cerebral cortex registration model, the loss value corresponding to the third training data can be calculated based on an improved loss function. The improved loss function is used to reduce deformation and improve registration accuracy. The improved loss function improves the regularization term and adds a supervision term compared with the prior art. By improving the calculation of the area loss and angle loss in the regularization term, it is beneficial to reduce distortion. By adding the supervision term and increasing the contrast loss between coordinates, the local deformation in large areas is reduced, making the surface of the registration result more uniform and improving the stability of the model.

[0084] Based on the above embodiments, further, the sphere of the cerebral cortex surface in the first standard space, the template sulcus and gyrus depth information in the first standard space, and the sulcus and gyrus depth information after sampling in the first standard space corresponding to the cerebral cortex image sample data are used as the third training data to train the third original model, and the cerebral cortex registration model can be obtained through training.

[0085] At this time, an unsupervised model training method is adopted. During the model training process, the loss function that can be used is

[0086] Based on the above embodiments, further, the improved loss function is:

[0087]

[0088] Among them, represents the loss value, represents the similarity term, represents the regularization term, represents the segmentation term, represents the supervision term, represents corresponding weight, λ parc represents corresponding weight, λ comp represents corresponding weight; represents the area loss, represents the angle loss, represents the distance loss, represents the folding loss, λ area represents corresponding weight, λangle denote the corresponding weight, λ dist denote the corresponding weight, λ fold denote the corresponding weight; N represents the number of vertices for non-rigid registration, M represents the deformed moving sphere, Φ represents the rotation tensor, C represents the comparison sphere; T represents the number of triangles, denote the original area of the triangle, A t is the area of the deformed triangle, t is a positive integer and t is less than or equal to T; denote the angle value of the τ-th angle of the triangle, ∠t τ denote the angle value of the τ-th angle of the deformed triangle, τ is a positive integer and τ is less than or equal to 3.

[0089] Specifically, since the surface of the sphere after algorithm registration in the prior art will be distorted in local areas due to some extreme values, the area loss is improved for the problem of distortion extreme values and the angle loss are expressed in exponential form, increasing the attention to extreme values and being able to reduce distortion. At the same time, the uniformity and accuracy of the surface of the registration sphere are optimized, and a supervision term is added increasing the contrast loss between coordinates, reducing the places with large local deformations, making the surface of the registration result more uniform, and also improving the stability of the model.

[0090] Among them, is the similarity term, and the mean square error is used to estimate the similarity of rigid and non-rigid registration.

[0091]

[0092] N represents the number of vertices for non-rigid registration, M represents the deformed moving sphere, Φ represents the rotation tensor, F represents the fixed sphere.

[0093]

[0094] denote the area loss, denote the angle loss, denote the distance loss, denote the folding loss, λ area denote the corresponding weight, λ angle denote the corresponding weight, λ dist denote the corresponding weight, λ fold denote the corresponding weight.

[0095]

[0096] T represents the number of triangles, represents the original area of the triangle, A t is the area of the deformed triangle, where t is a positive integer and t ≤ T;

[0097]

[0098] represents the angular value of the τ-th angle of the triangle, ∠t τ represents the angular value of the τ-th angle of the deformed triangle, where τ is a positive integer and τ ≤ 3;

[0099]

[0100] d(·) represents the Euclidean distance, represents the original coordinates of the centroid of the 1-hop neighborhood, represents the original coordinates of the barycenter of the 1-hop neighborhood, v′ C represents the updated coordinates of the centroid of the 1-hop neighborhood after deformation, v′ B represents the updated coordinates of the barycenter of the 1-hop neighborhood after deformation, ∈ l represents the average side length between two points of the standard sphere in the current level.

[0101]

[0102] Δ t represents the current area of the t-th triangle, represents the original area of the t-th triangle, v i,j represents the edge vector from vertex i to vertex j of the t-th triangle, v i,k represents the edge vector from vertex i to vertex k of the t-th triangle, where n represents the unit normal vector of the t-th triangle, and n points outward from the origin of the sphere.

[0103]

[0104] P represents the number of partitions, where p is a positive integer and p ≤ P, represents the p-th partition on the moving sphere, F p represents the p-th partition on the fixed sphere;

[0105]

[0106] N represents the number of vertices for non-rigid registration, M represents the deformed moving sphere, Φ represents the rotation tensor, and C represents the comparison sphere. C can be the spherical surface after registration using the MSM_strain method.

[0107] Figure 6 is a schematic flowchart of a method for processing neuroimaging data provided by the sixth embodiment of the present invention. As Figure 6 described above, on the basis of the above embodiments, further, the method for processing neuroimaging data provided by the embodiments of the present invention further includes:

[0108] S601. Based on the individual's brain structural imaging data, the white matter partition part included in the subcortical structural partition corresponding to the brain structural imaging data, and the cerebral cortex reconstruction model, obtain the level set corresponding to the individual's brain structural imaging data, and reconstruct the cerebral cortex image based on the level set corresponding to the individual's brain structural imaging data to obtain the reconstructed cerebral cortex imaging data corresponding to the individual's brain structural imaging data;

[0109] Specifically, the server can obtain the level set corresponding to the individual's brain structural imaging data based on the individual's brain structural imaging data, the white matter partition part included in the subcortical structural partition corresponding to the individual's brain structural imaging data, and the cerebral cortex reconstruction model. The server reconstructs the cerebral cortex image according to the level set corresponding to the individual's brain structural imaging data to obtain the reconstructed cerebral cortex imaging data corresponding to the individual's brain structural imaging data.

[0110] S602. Obtain the white matter surface under the cerebral cortex corresponding to the individual's brain structural imaging data based on the reconstructed cerebral cortex imaging data corresponding to the individual's brain structural imaging data;

[0111] Specifically, the server can obtain the white matter surface (white surface) under the cerebral cortex corresponding to the individual's brain structural imaging data according to the reconstructed cerebral cortex imaging data corresponding to the individual's brain structural imaging data. The white matter surface under the cerebral cortex corresponding to the individual's brain structural imaging data is regarded as the white matter surface under the cerebral cortex corresponding to the individual's brain structural imaging data in the first individual space.

[0112] S603. Perform spatial registration based on the individual's brain structural imaging data in the first individual space, the white matter surface under the cerebral cortex corresponding to the individual's brain structural imaging data in the first individual space, and the individual's neuroimaging data in the second individual space to obtain the volume data of the individual's neuroimaging data in the first individual space of the individual;

[0113] Specifically, based on the brain structural image data of the individual in the first individual space, the white matter surface under the cerebral cortex corresponding to the brain structural image data of the individual in the first individual space, and the neuroimage data of the individual in the second individual space, spatial registration can be performed to obtain the volume data of the neuroimage data of the individual in the second individual space in the first body space of the individual. The brain structural image data of the individual in the first individual space can adopt T1-weighted imaging (T1w) data, and the neuroimage data of the individual in the second space can adopt blood oxygenation level-dependent functional magnetic resonance imaging (BOLD) data, etc. The neuroimage data of the individual in the second individual space can be structural neuroimage data or neuroimage data of any modality. Among them, the bbregister method can be used to register the volume data of the individual in other individual spaces to the brain structural image data of the first individual space to obtain the volume data of the neuroimage data of the individual in other individual spaces in the first space. Neuroimage data can be divided into brain structural image data and brain functional image data.

[0114] S604. Obtain the deformation field of registering the first individual space of the individual to the second standard template space based on the brain structural image data of the individual in the first individual space, the neuroimage data of the second standard template space, and the brain space normalization model; wherein, the brain space normalization model is pre-trained.

[0115] Specifically, the server inputs the brain structural image data of the individual in the first individual space and the neuroimage data of the second standard template space into the brain space normalization model, and can output the deformation field of registering the first individual space of the individual to the second standard template space. Among them, the brain neuroimage data of the standard template space can adopt T1w data of the second standard template space. The second standard template space can adopt the MNI standard template space.

[0116] Furthermore, downsampling can be performed on the brain structural image data of the individual in the first individual space and the neuroimage data of the second standard template space, and an appropriate sampling rate can be set so that the sampling rate can not only reduce the occupancy of video memory but also not affect the accuracy.

[0117] S605. Apply the deformation field of registering the first individual space of the individual to the second standard template space to the volume data of the neuroimage data of the individual in the second individual space in the first body space of the individual to obtain the volume data of the neuroimage data of the individual in the second individual space on the second standard template space.

[0118] Specifically, the server applies the deformation field that registers the first individual space of the individual to the second standard template space to the volume data of the neuroimaging data of the individual in the second individual space in the first individual space of the individual, and can obtain the volume data of the neuroimaging data of the individual in the second individual space on the second standard template space.

[0119] The accuracy of registering the different modality data of the individual to the data in the second standard template space through the above steps S601 to S605 is better than the result of directly using the brain space normalization model for cross-modal registration.

[0120] Figure 7 It is a schematic flowchart of the processing method of neuroimaging data provided by the seventh embodiment of the present invention. As Figure 7 shown, on the basis of the above embodiments, further, training the brain space normalization model includes:

[0121] S701. Obtain the brain structure image sample data and the neuroimaging sample data in the second standard template space as the fourth training data;

[0122] Specifically, historical brain structure image data can be collected, and based on the historical brain structure image data, the brain structure image data of a fourth preset number of individuals in the individual space is obtained as the brain structure image sample data, and the neuroimaging sample data in the second standard template space is obtained. The brain structure image sample data and the neuroimaging sample data in the second standard template space are used as the fourth training data. The brain structure image data can be the volume data of the individual in the weighted imaging space, and is selected according to actual needs, and is not limited in the embodiments of the present invention.

[0123] S702. Train and obtain the brain space normalization model according to the fourth training data and the fourth original model.

[0124] Specifically, the server trains the fourth original model through the fourth training data, and can train and obtain the brain space normalization model. Among them, the fourth original model can adopt a U-Net network structure model.

[0125] It can be understood that based on the neuroimaging data of the individual in the first individual space, the neuroimaging data in the second standard template space, and the brain space normalization model, the deformation field for registering the first individual space of the individual to the second standard template space can be obtained, so that the neuroimaging data of any modality of the individual can be configured to the second standard template space. At this time, the brain space normalization model should be trained and obtained based on the training data composed of the neuroimaging sample data and the neuroimaging sample data in the second standard template space.

[0126] The processing method for neuroimaging data provided by the embodiments of the present invention effectively overcomes the deficiencies of traditional technologies in processing neuroimaging data of specific patients with higher robustness and adaptability, and provides a more accurate and efficient preprocessing solution. The application of the processing method for neuroimaging data provided by the embodiments of the present invention not only improves the accuracy of medical imaging analysis, but also provides a more solid scientific basis for clinical decision-making, laying a better foundation for the diagnosis and treatment of patients.

[0127] Figure 8 is a schematic structural diagram of the processing device for neuroimaging data provided by the eighth embodiment of the present invention, as Figure 8 shown, the processing device for neuroimaging data provided by the embodiments of the present invention includes a segmentation unit 801, a reconstruction unit 802, and a registration unit 803, where:

[0128] The segmentation unit 801 is used to obtain the subcortical structural partitions corresponding to the brain structure imaging data based on the brain structure imaging data and the brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained; the reconstruction unit 802 is used to obtain the level set corresponding to the brain structure imaging data based on the brain structure imaging data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structure imaging data, and the cerebral cortex reconstruction model, and reconstruct the cerebral cortex imaging based on the level set corresponding to the brain structure imaging data to obtain the reconstructed cerebral cortex imaging data corresponding to the brain structure imaging data; wherein, the cerebral cortex reconstruction model is pre-trained; the registration unit 803 is used to perform spherical registration of the cerebral cortex surface and the cerebral cortex surface of the first standard space based on the reconstructed cerebral cortex imaging data corresponding to the brain structure imaging data and the cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure imaging data; wherein, the cerebral cortex registration model is pre-trained.

[0129] Specifically, brain structure image data can be obtained through a Magnetic Resonance Imaging (MRI) device or a CT imaging device. For example, the brain structure image data includes various modal data such as T1-weighted data, T2-weighted data, diffusion tensor imaging, and CT. The segmentation unit 801 obtains the subcortical structure partitions and the extracortical structure partitions corresponding to the brain structure image data based on the brain structure image data and the brain structure segmentation model. To compare and analyze the data on the same scale, the brain structure image data is normalized so that the intensity of the brain structure image data is converted to between 0 and 1. The normalized brain structure image data is input into the brain structure segmentation model, and the subcortical structure partitions and the extracortical structure partitions corresponding to the brain structure image data are output. The subcortical structure partitions corresponding to the brain structure image data include the white matter partition part. The brain structure segmentation model can distinguish the cerebral cortex and subcortical structures in the brain structure image data. Among them, the brain structure image data can adopt T1-weighted data.

[0130] Based on the brain structure image data, the white matter partition part included in the subcortical structure partitions corresponding to the brain structure image data, and the cerebral cortex reconstruction model, the reconstruction unit 802 can obtain the level set corresponding to the brain structure image data. To compare and analyze the data on the same scale, the brain structure image data is normalized so that the intensity of the brain structure image data is converted to between 0 and 1. The normalized brain structure image data and the white matter partition part included in the subcortical structure partitions corresponding to the brain structure image data are input into the cerebral cortex reconstruction model together, and the level set corresponding to the brain structure image data can be output. Among them, the cerebral cortex reconstruction model is obtained by pre-training. The reconstruction unit 802 reconstructs the cerebral cortex image according to the level set corresponding to the brain structure image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structure image data.

[0131] The registration unit 803 performs registration of the cerebral cortex surface and the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and the cerebral cortex registration model, and can obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data. By performing inflation and spherical parameterization on the reconstructed cerebral cortex image data corresponding to the brain structure image data, the original unregistered spherical surface and gyrus-sulcus depth information corresponding to the brain structure image data can be obtained; sampling the original unregistered gyrus-sulcus depth information corresponding to the brain structure image data into the first standard space can obtain the gyrus-sulcus depth information of the cerebral cortex surface after sampling in the first standard space corresponding to the brain structure image data; inputting the spherical surface of the cerebral cortex surface in the first standard space, the gyrus-sulcus depth information of the cerebral cortex surface after sampling in the first standard space corresponding to the brain structure image data, and the gyrus-sulcus depth information of the template in the first standard space into the cerebral cortex registration model can obtain the deformation matrix for registering the individual space to the first standard space; according to the deformation matrix for registering the individual space to the first standard space, the original unregistered spherical surface of the cerebral cortex surface corresponding to the brain structure image data, and the spherical surface of the cerebral cortex surface in the first standard space, the deformation matrix of the original unregistered spherical surface space of the cerebral cortex surface can be obtained, and then applying the deformation matrix of the original unregistered spherical surface space of the cerebral cortex surface to the original unregistered spherical surface of the cerebral cortex surface corresponding to the brain structure image data can obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure image data. Among them, the first standard space is a general, standardized reference space, and the fsaverage standard space can be used. In the analysis of cerebral cortex image data, mapping the brain images of different individuals into the first standard space can achieve comparison and statistical analysis between different individuals. The individual space is opposite to the first standard space and refers to the unique spatial structure of the brain of a specific individual. The cerebral cortex registration model is obtained by pre-training.

[0132] The processing device for neuroimaging data provided by an embodiment of the present invention obtains the subcortical structural partitions corresponding to the brain structural imaging data based on the brain structural imaging data and the brain structure segmentation model; obtains the level set corresponding to the brain structural imaging data based on the brain structural imaging data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural imaging data, and the cerebral cortex reconstruction model, and reconstructs the cerebral cortex image based on the level set corresponding to the brain structural imaging data to obtain the reconstructed cerebral cortex imaging data corresponding to the brain structural imaging data; performs spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex imaging data corresponding to the brain structural imaging data and the cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural imaging data, which has higher robustness and adaptability, effectively overcomes the deficiencies of the traditional technology in processing the neuroimaging data of specific patients, and improves the accuracy of neuroimaging data processing.

[0133] Figure 9 is a schematic structural diagram of the processing device for neuroimaging data provided by the ninth embodiment of the present invention, as Figure 9 shown, on the basis of the above embodiments, further, the registration unit 803 includes a first obtaining subunit 8031, a standard space sampling subunit 8032, a second obtaining subunit 8033, and an individual space sampling subunit 8034, where:

[0134] The first acquisition subunit 8031 is configured to perform inflation and spherical parameterization processing on the reconstructed cerebral cortex image data corresponding to the brain structure image data, so as to obtain the spherical surface and sulcus depth information of the original unregistered cerebral cortex corresponding to the brain structure image data; the standard space sampling subunit 8032 is configured to perform first standard space sampling according to the spherical surface of the cerebral cortex in the first standard space, the original unregistered spherical surface of the brain corresponding to the brain structure image data, and the original unregistered sulcus depth information of the cerebral cortex corresponding to the brain structure image data, so as to obtain the sulcus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data; the second acquisition subunit 8033 is configured to obtain a deformation matrix for registering the individual space to the first standard space according to the spherical surface of the cerebral cortex surface in the first standard space, the sulcus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the brain structure image data, the template sulcus depth information of the first standard space, and the cerebral cortex registration model; the individual space sampling subunit 8034 is configured to upsample according to the deformation matrix for registering the individual space to the first standard space, the original unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the spherical surface of the cerebral cortex surface in the first standard space to obtain a deformation matrix in the spherical surface space of the original unregistered cerebral cortex; and according to the deformation matrix in the spherical surface space of the original unregistered cerebral cortex and the original unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, obtain the registered spherical surface of the cerebral cortex corresponding to the brain structure image data.

[0135] Figure 10 is a schematic structural diagram of a neural image data processing device provided in the tenth embodiment of the present invention, as Figure 10 shown. On the basis of the above embodiments, further, the neural image data processing device provided in the embodiment of the present invention further includes a first acquisition unit 804 and a first training unit 805, wherein:

[0136] The first acquisition unit 804 is configured to obtain first training data according to the first brain structure image sample data, the cerebral cortex structure and subcutaneous structure partition corresponding to the first brain structure image sample data; the first training unit 805 is configured to train and obtain the brain structure segmentation model according to the first training data and the first original model.

[0137] Figure 11 is a schematic structural diagram of a neural image data processing device provided in the eleventh embodiment of the present invention, as Figure 11 shown. On the basis of the above embodiments, further, the neural image data processing device provided in the embodiment of the present invention further includes a second acquisition unit 806 and a second training unit 807, wherein:

[0138] The second acquisition unit 806 is configured to obtain second training data according to the second subcortical brain structure image sample data, the white matter partition part in the subcortical brain structure partition corresponding to the second subcortical brain structure image sample data, and the label corresponding to the white matter partition part; the second training unit 807 is configured to train and obtain the cerebral cortex reconstruction model according to the second training data and the second original model.

[0139] Figure 12 FIG. is a schematic structural diagram of a neural image data processing device provided in the twelfth embodiment of the present invention, as Figure 12 shown. On the basis of the above embodiments, further, the neural image data processing device provided in the embodiments of the present invention further includes a first acquisition unit 808, a third acquisition unit 809, a sample sampling unit 810, and a third training unit 811, where:

[0140] The first acquisition unit 808 is configured to acquire cerebral cortex image sample data; the third acquisition unit 809 is configured to perform inflation and spherical parameterization processing on the cerebral cortex image sample data to obtain the spherical surface and gyrus depth information of the original unregistered cerebral cortex surface corresponding to the cerebral cortex image sample data; the sample sampling unit 810 is configured to perform first standard space sampling according to the spherical surface of the first standard space cerebral cortex surface, the spherical surface of the original unregistered cerebral cortex surface corresponding to the cerebral cortex image sample data, and the gyrus depth information of the original unregistered cerebral cortex surface corresponding to the cerebral cortex image sample data to obtain the gyrus depth information of the first standard space sampled cerebral cortex surface corresponding to the cerebral cortex image sample data; and use the spherical surface of the first standard space cerebral cortex surface, the template gyrus depth information of the first standard space, the gyrus depth information of the first standard space sampled corresponding to the cerebral cortex image sample data, and the corresponding partition label data as the third training data; the third training unit 811 is configured to train and obtain the cerebral cortex registration model according to the third training data and the third original model; wherein, in the process of training and obtaining the cerebral cortex registration model, the loss value corresponding to the third training data is calculated based on an improved loss function.

[0141] Figure 13 FIG. is a schematic structural diagram of a neural image data processing device provided in the thirteenth embodiment of the present invention, as Figure 13 shown. On the basis of the above embodiments, further, the neural image data processing device provided in the embodiments of the present invention further includes a reconstruction unit 812, a fourth acquisition unit 813, an individual registration unit 814, a fifth acquisition unit 815, and a sixth acquisition unit 816, where:

[0142] The reconstruction unit 812 is configured to obtain a level set corresponding to the individual's brain structural image data based on the individual's brain structural image data, the white matter partition part included in the subcortical structural partition corresponding to the brain structural image data, and the cerebral cortex reconstruction model, and reconstruct a cerebral cortex image based on the level set corresponding to the individual's brain structural image data to obtain reconstructed cerebral cortex image data corresponding to the individual's brain structural image data; the fourth obtaining unit 813 is configured to obtain a white matter surface under the cerebral cortex corresponding to the individual's brain structural image data based on the reconstructed cerebral cortex image data corresponding to the individual's brain structural image data; the individual registration unit 814 is configured to perform spatial registration based on the brain structural image data of the individual in the first individual space, the white matter surface under the cerebral cortex corresponding to the brain structural image data of the individual in the first individual space, and the neuroimage data of the individual in the second individual space, to obtain volume data of the neuroimage data of the individual in the second individual space in the first body space of the individual; the fifth obtaining unit 815 is configured to obtain a deformation field for registering the first individual space of the individual to the second standard template space based on the brain structural image data of the individual in the first individual space, the neuroimage data of the second standard template space, and the brain space normalization model; wherein, the brain space normalization model is obtained by pre-training; the sixth obtaining unit 816 is configured to apply the deformation field for registering the first individual space of the individual to the second standard template space to the volume data of the neuroimage data of the individual in the second individual space in the first body space of the individual, to obtain volume data of the neuroimage data of the individual in the second individual space on the second standard template space.

[0143] Figure 14 It is a schematic structural diagram of a processing device for neuroimage data provided in the fourteenth embodiment of the present invention. As Figure 14 shown, on the basis of the above embodiments, further, the processing device for neuroimage data provided in the embodiments of the present invention further includes a second obtaining unit 817 and a fourth training unit 818, where:

[0144] The second obtaining unit 817 is configured to obtain brain structural image sample data and neuroimage sample data of the second standard template space as fourth training data; the fourth training unit 818 is configured to train and obtain the brain space normalization model according to the fourth training data and the fourth original model.

[0145] The embodiments of the device provided in the embodiments of the present invention can specifically be used to execute the processing procedures of the above method embodiments, and its functions will not be elaborated here. Reference can be made to the detailed descriptions of the above method embodiments.

[0146] Figure 15 It is a schematic physical structure diagram of an electronic device provided in the fifteenth embodiment of the present invention. As Figure 15As shown, the electronic device may include: a processor 1501, a communications interface 1502, a memory 1503, and a communication bus 1504. Among them, the processor 1501, the communications interface 1502, and the memory 1503 complete communication with each other through the communication bus 1504. The processor 1501 may call the logical instructions in the memory 1503 to execute the methods provided in the above method embodiments, for example, including: obtaining the subcortical structural partitions corresponding to the brain structural image data based on the brain structural image data and the brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained; obtaining the level set corresponding to the brain structural image data based on the brain structural image data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural image data, and the cerebral cortex reconstruction model, and reconstructing the cerebral cortex image based on the level set corresponding to the brain structural image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structural image data; wherein, the cerebral cortex reconstruction model is pre-trained; performing spherical registration of the cerebral cortex surface with the cerebral cortex surface of the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural image data and the cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural image data; wherein, the cerebral cortex registration model is pre-trained.

[0147] In addition, when the logical instructions in the above-mentioned memory 1503 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0148] This embodiment discloses a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided in the above method embodiments. For example, it includes: obtaining the subcortical structural partitions corresponding to the brain structural image data based on the brain structural image data and a brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained; obtaining the level set corresponding to the brain structural image data based on the brain structural image data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural image data, and a cerebral cortex reconstruction model, and reconstructing a cerebral cortex image based on the level set corresponding to the brain structural image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structural image data; wherein, the cerebral cortex reconstruction model is pre-trained; performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural image data and a cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural image data; wherein, the cerebral cortex registration model is pre-trained.

[0149] This embodiment provides a computer-readable storage medium. The computer-readable storage medium stores computer programs / instructions. The computer programs / instructions cause the computer to execute the methods provided in the above method embodiments. For example, it includes: obtaining the subcortical structural partitions corresponding to the brain structural image data based on the brain structural image data and a brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained; obtaining the level set corresponding to the brain structural image data based on the brain structural image data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structural image data, and a cerebral cortex reconstruction model, and reconstructing a cerebral cortex image based on the level set corresponding to the brain structural image data to obtain the reconstructed cerebral cortex image data corresponding to the brain structural image data; wherein, the cerebral cortex reconstruction model is pre-trained; performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structural image data and a cerebral cortex registration model to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structural image data; wherein, the cerebral cortex registration model is pre-trained.

[0150] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0151] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0152] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realizes the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or the combination of blocks.

[0154] In the description of this specification, the descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0155] The above - described specific embodiments have further elaborated on the object, technical solution, and beneficial effects of the present invention. It should be understood that the above - described are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing neuroimaging data, characterized in that, Including: Based on brain structure image data and a brain structure segmentation model, obtaining a subcortical structure partition corresponding to the brain structure image data; wherein, the brain structure segmentation model is pre-trained; Based on the brain structure image data, the white matter partition part included in the subcortical structure partition corresponding to the brain structure image data, and a cerebral cortex reconstruction model, obtaining a level set corresponding to the brain structure image data, and reconstructing a cerebral cortex image based on the level set corresponding to the brain structure image data to obtain reconstructed cerebral cortex image data corresponding to the brain structure image data; wherein, the cerebral cortex reconstruction model is pre-trained; Performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and a cerebral cortex registration model, to obtain a registered spherical surface of the cerebral cortex corresponding to the brain structure image data; wherein, the cerebral cortex registration model is pre-trained based on an improved loss function; Wherein, the improved loss function includes: Among them, represents the loss value, represents the similarity term, represents the regularization term, represents the segmentation term, represents the supervision term, λ sim represents the corresponding weight, λ parc represents the corresponding weight, λ comp represents the corresponding weight.

2. The method according to claim 1, wherein The performing spherical registration of the cerebral cortex surface with the cerebral cortex surface in the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure image data and the cerebral cortex registration model, to obtain a registered spherical surface of the brain corresponding to the brain structure image data includes: Performing inflation and spherical parameterization processing on the reconstructed cerebral cortex image data corresponding to the brain structure image data, to obtain an originally unregistered spherical surface of the cerebral cortex and gyrus depth information corresponding to the brain structure image data; Performing first standard space sampling according to the spherical surface of the cerebral cortex in the first standard space, the originally unregistered spherical surface of the brain corresponding to the brain structure image data, and the originally unregistered gyrus depth information of the cerebral cortex corresponding to the brain structure image data, to obtain gyrus depth information of the cerebral cortex surface after first standard space sampling corresponding to the brain structure image data; According to the spherical surface of the cerebral cortex surface in the first standard space, the gyrus depth information of the cerebral cortex surface after first standard space sampling corresponding to the brain structure image data, the template gyrus depth information of the first standard space, and the cerebral cortex registration model, obtaining a deformation matrix for registering the individual space to the first standard space; According to the deformation matrix for registering the individual space to the first standard space, the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, and the spherical surface of the cerebral cortex surface in the first standard space, upsampling to obtain a deformation matrix in the spherical surface space of the originally unregistered cerebral cortex surface; according to the deformation matrix in the spherical surface space of the originally unregistered cerebral cortex surface and the originally unregistered spherical surface of the cerebral cortex corresponding to the brain structure image data, obtaining the registered spherical surface of the cerebral cortex corresponding to the brain structure image data.

3. The method according to claim 1, characterized in that The steps of training the brain structure segmentation model include: According to first brain structure image sample data, the cerebral cortex structure and subcortical structure partition corresponding to the first brain structure image sample data, obtaining first training data; The brain structure segmentation model is trained and obtained according to the first training data and the first original model.

4. The method according to claim 1, wherein The steps of training the cerebral cortex reconstruction model include: Second training data is obtained according to second brain structure image sample data, the white matter partition part in the subcortical structure partition corresponding to the second brain structure image sample data, and the label corresponding to the white matter partition part; The cerebral cortex reconstruction model is trained and obtained according to the second training data and the second original model.

5. The method according to claim 1, wherein The steps of training the cerebral cortex registration model include: Cerebral cortex image sample data is acquired; The cerebral cortex image sample data is subjected to dilation and spherical parameterization processing to obtain the original unregistered spherical surface and gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data; First standard space sampling is performed according to the spherical surface of the cerebral cortex in the first standard space, the original unregistered spherical surface of the cerebral cortex corresponding to the cerebral cortex image sample data, and the original unregistered gyrus depth information of the cerebral cortex corresponding to the cerebral cortex image sample data, to obtain the gyrus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the cerebral cortex image sample data; and the spherical surface of the cerebral cortex in the first standard space, the template gyrus depth information in the first standard space, the gyrus depth information of the cerebral cortex surface after the first standard space sampling corresponding to the cerebral cortex image sample data, and the corresponding partition label data are used as the third training data; The cerebral cortex registration model is trained and obtained according to the third training data and the third original model; wherein, in the process of training and obtaining the cerebral cortex registration model, the loss value corresponding to the third training data is calculated based on an improved loss function.

6. The method according to claim 5, wherein The improved loss function further includes: Among them, represents the area loss, represents the angle loss, represents the distance loss, represents the folding loss, λ area represents the corresponding weight, λ angle represents the corresponding weight, λ dist represents the corresponding weight, λ fold denotes the corresponding weight; N represents the number of vertices of non-rigid registration, M represents the deformed moving sphere, Φ represents the rotation tensor, C represents the comparison sphere; T represents the number of triangles, represents the original area of the triangle, A t is the area of the deformed triangle, t is a positive integer and t is less than or equal to T; represents the angular value of the τ-th angle of the triangle, ∠t τ represents the angular value of the τ-th angle of the deformed triangle, τ is a positive integer and τ is less than or equal to 3.

7. The method according to any one of claims 1 to 6, characterized in that, It further includes: Based on the individual's brain structure image data, the white matter partition part included in the subcortical structure partition corresponding to the brain structure image data, and the cerebral cortex reconstruction model, the level set corresponding to the individual's brain structure image data is obtained, and based on the level set corresponding to the individual's brain structure image data, the cerebral cortex image is reconstructed to obtain the reconstructed cerebral cortex image data corresponding to the individual's brain structure image data; Based on the reconstructed cerebral cortex image data corresponding to the individual's brain structure image data, the white matter surface under the cerebral cortex corresponding to the individual's brain structure image data is obtained; Spatial registration is performed based on the individual's brain structure image data in the first individual space, the white matter surface under the cerebral cortex corresponding to the individual's brain structure image data in the first individual space, and the individual's neuroimage data in the second individual space, to obtain the volume data of the individual's neuroimage data in the first body space of the individual; Based on the individual's brain structure image data in the first individual space, the neuroimage data in the second standard template space, and the brain space normalization model, the deformation field of the individual's first individual space registered to the second standard template space is obtained; wherein, the brain space normalization model is pre-trained and obtained. Apply the deformation field that registers the first individual space of the individual to the second standard template space to the volume data of the individual's neuroimaging data in the second individual space on the volume data of the individual in the first individual space, to obtain the volume data of the individual's neuroimaging data in the second individual space on the second standard template space.

8. The method according to claim 7, characterized in that, Training the brain space normalization model includes: Obtain brain structure imaging sample data and neuroimaging sample data of the second standard template space as the fourth training data; Train to obtain the brain space normalization model according to the fourth training data and the fourth original model.

9. A processing device for neuroimaging data, characterized in that, Includes: A segmentation unit for obtaining the subcortical structural partitions corresponding to the brain structure imaging data based on the brain structure imaging data and the brain structure segmentation model; wherein, the brain structure segmentation model is pre-trained; A reconstruction unit for obtaining the level set corresponding to the brain structure imaging data based on the brain structure imaging data, the white matter partition part included in the subcortical structural partitions corresponding to the brain structure imaging data, and the cerebral cortex reconstruction model, and reconstructing the cerebral cortex image based on the level set corresponding to the brain structure imaging data to obtain the reconstructed cerebral cortex image data corresponding to the brain structure imaging data; wherein, the cerebral cortex reconstruction model is pre-trained; A registration unit for performing spherical registration of the cerebral cortex surface and the cerebral cortex surface of the first standard space based on the reconstructed cerebral cortex image data corresponding to the brain structure imaging data and the cerebral cortex registration model, to obtain the spherical surface after registration of the cerebral cortex surface corresponding to the brain structure imaging data; wherein, the cerebral cortex registration model is pre-trained based on an improved loss function; Wherein, the improved loss function includes: Among them, represents the loss value, represents the similarity term, represents the regularization term, represents the segmentation term, represents the supervision term, λ sim represents the corresponding weight, λ parc represents the corresponding weight, λ comp represents the corresponding weight.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program / instructions, and when the computer program / instructions are executed by the processor, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer program product comprising computer programs / instructions, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images