Brain image processing method and related equipment

Abstract

The embodiment of the invention discloses a brain image processing method and related equipment, which are used for improving the image processing efficiency. The method comprises the following steps: acquiring a brain anatomical structure image and at least one brain function state image of the same brain; inputting the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result of a plurality of brain regions including the brain; inputting the brain region segmentation result into a brain volume algorithm to obtain a brain volume index value of each brain region; and for each brain function state image, inputting the brain function state image and the brain region segmentation result into a brain index model corresponding to the brain function state image to obtain a brain function index value of each brain region.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of medical image processing, and particularly to a brain image processing method and related devices. Background Art

[0002] The brain is the center of human nerve activities, governing all life activities of the human body. However, the brain structure is extremely complex. Anatomically, it can be divided into hundreds of brain regions, intertwined with tens of millions of nerve fibers. Such a complex structure leads to very limited human understanding of the brain, and as a result, it is difficult to detect various brain tissue injury diseases, cognitive diseases, or brain function disorders and other brain health problems in a timely manner, seriously affecting their health and quality of life. With the development of Magnetic Resonance Imaging (MRI), it has become one of the important technical means for brain examinations. Especially the introduction of DTI and fMRI, the analysis of brain connection networks and brain functional connections has become an important means for various brain cognitive and functional diseases. The quantitative segmentation of brain regions through MRI, the analysis of brain connection networks through DTI, and the analysis of brain functional connections through fMRI are of great significance for the diagnosis and treatment of various brain injury diseases and various cognitive diseases, greatly promoting the diagnosis and treatment of brain diseases.

[0003] However, due to the huge differences in the image semantic features of different brain images, there is currently no image processing method for determining the index values of various brain metrics by combining different brain images. The index values for various brain metrics (brain volume and brain function status) are mainly determined by doctors through comprehensive decision-making, which largely depends on doctors' medical experience and has a low processing efficiency. Summary of the Invention

[0004] Embodiments of the present application provide a brain image processing method and related devices for improving the image processing efficiency.

[0005] A first aspect of an embodiment of the present application provides a brain image processing method, including:

[0006] Obtaining a brain anatomical structure image and at least one brain function status image of the same brain;

[0007] Inputting the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain;

[0008] Inputting the brain region segmentation result into a brain volume algorithm to obtain the brain volume index value of each brain region;

[0009] For each of the brain functional state images, input the brain functional state image and the brain region segmentation result into the brain index model corresponding to the brain functional state image to obtain the brain functional index values of each of the brain regions.

[0010] In a specific implementation manner, the at least one brain functional state image includes: a brain network connection image and / or a brain functional connection image; the brain network connection image is acquired based on diffusion tensor imaging technology, and the brain functional connection image is acquired based on resting-state functional magnetic resonance imaging technology.

[0011] In a specific implementation manner, the method further includes:

[0012] Construct a three-dimensional model of the brain based on the brain anatomical structure image;

[0013] In the three-dimensional model of the brain, display each of the brain regions in the brain region segmentation result in different styles.

[0014] In a specific implementation manner, after obtaining the three-dimensional model of the brain, the method further includes:

[0015] In response to a user's operation of selecting a target brain region in the three-dimensional model, display the target brain region in a style different from that of other brain regions outside the target brain region; where the target brain region is any one of the multiple brain regions;

[0016] Display the brain volume index value and / or the brain functional index value of the target brain region.

[0017] In a specific implementation manner, the method further includes:

[0018] If the at least one brain functional state image includes a brain network connection image, display the multiple bundles of cerebral white matter fibers included in the brain network connection image in the three-dimensional model;

[0019] In response to a user's operation of selecting a target brain region in the three-dimensional model, display the cerebral white matter fibers associated with the target brain region in a style different from that of other cerebral white matter fibers outside the cerebral white matter fibers associated with the target brain region.

[0020] In a specific implementation manner, the method further includes:

[0021] According to the age of the object to which the brain belongs, obtain the statistical values of each brain region for each brain index at the age from a brain index database, where the brain index includes the brain function and the brain volume;

[0022] Based on the degree of difference between the index value and the statistical value of each brain region for each brain index, determine the health status of each brain region for each brain index respectively.

[0023] In a specific implementation manner, the method further includes:

[0024] Based on the statistical values of each brain region for each brain metric at different ages and the metric values of each brain region for each brain metric, determine the predicted brain age of each brain region for each brain metric;

[0025] Based on the predicted brain age of each brain region for each brain metric, determine the predicted brain age of the brain.

[0026] The second aspect of the embodiments of the present application provides a computer device, including:

[0027] An acquisition unit, configured to acquire a brain anatomical structure image and at least one brain functional state image of the same brain;

[0028] A segmentation unit, configured to input the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain;

[0029] A calculation unit, configured to input the brain region segmentation result into a brain volume algorithm to obtain the brain volume metric value of each brain region;

[0030] The calculation unit is further configured to, for each brain functional state image, input the brain functional state image and the brain region segmentation result into the brain metric model corresponding to the brain functional state image to obtain the brain functional metric value of each brain region.

[0031] In a specific implementation manner, the at least one brain functional state image includes: a brain network connection image and / or a brain functional connection image; the brain network connection image is acquired based on diffusion tensor imaging technology, and the brain functional connection image is acquired based on resting-state functional magnetic resonance imaging technology.

[0032] In a specific implementation manner, the device further includes: a construction unit and a display unit;

[0033] The construction unit is configured to construct a three-dimensional model of the brain based on the brain anatomical structure image;

[0034] The display unit is configured to display each brain region in the brain region segmentation result in different styles in the three-dimensional model of the brain.

[0035] In a specific implementation manner, the device further includes: a display unit;

[0036] The display unit is configured to, in response to a user's operation of selecting a target brain region in the three-dimensional model, display the target brain region in a style different from other brain regions outside the target brain region; wherein the target brain region is any one of the multiple brain regions;

[0037] The display unit is further configured to display the brain volume index value and / or the brain function index value of the target brain region.

[0038] In a specific implementation manner, the device further includes: a display unit;

[0039] The display unit is configured to, if the at least one brain function state image includes a brain network connection image, display multiple bundles of cerebral white matter fibers included in the brain network connection image on the three-dimensional model;

[0040] The display unit is further configured to, in response to a user's operation of selecting a target brain region in the three-dimensional model, display the cerebral white matter fibers associated with the target brain region in a style different from other cerebral white matter fibers outside the cerebral white matter fibers associated with the target brain region.

[0041] In a specific implementation manner, the device further includes: a determination unit;

[0042] The acquisition unit is further configured to, according to the age of the object to which the brain belongs, obtain the statistical values of each brain region for each brain index at the age from a brain index database, where the brain indexes include the brain function and the brain volume;

[0043] The determination unit is configured to respectively determine the health status of each brain region for each brain index based on the degree of difference between the index value and the statistical value of each brain region for each brain index.

[0044] In a specific implementation manner, the device further includes: a determination unit;

[0045] The determination unit is configured to determine the predicted brain age of each brain region for each brain index based on the statistical values of each brain region for each brain index at different ages and the index values of each brain region for each brain index.

[0046] The determination unit is further configured to determine the predicted brain age of the brain based on the predicted brain age of each brain region for each brain index.

[0047] A third aspect of the embodiments of the present application provides a computer device, including:

[0048] A central processing unit, a memory, and an input / output interface;

[0049] The memory is a transient storage memory or a persistent storage memory;

[0050] The central processing unit is configured to communicate with the memory and execute the instruction operations in the memory to execute the method described in the first aspect.

[0051] A computer program product containing instructions is provided in the fourth aspect of the embodiments of the present application. When the computer program product runs on a computer, the computer is caused to execute the method described in the first aspect.

[0052] A computer storage medium is provided in the fifth aspect of the embodiments of the present application. Instructions are stored in the computer storage medium. When the instructions are executed on a computer, the computer is caused to execute the method described in the first aspect.

[0053] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages: obtaining a brain anatomical structure image and at least one brain functional state image of the same brain; inputting the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain; inputting the brain region segmentation result into a brain volume algorithm to obtain a brain volume index value of each brain region; for each brain functional state image, inputting the brain functional state image and the brain region segmentation result into the brain index model corresponding to the brain functional state image to obtain a brain functional index value of each brain region. Through the above processing architecture, the index values of different brain indexes (brain volume and brain functional state) can be determined flexibly, and the image processing efficiency is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a system architecture diagram of a brain image processing method disclosed in an embodiment of the present application;

[0055] Figure 2 It is a flowchart of a brain image processing method disclosed in an embodiment of the present application;

[0056] Figure 3 It is a structural diagram of a computer device disclosed in an embodiment of the present application;

[0057] Figure 4 It is another structural diagram of a computer device disclosed in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0059] Embodiments of the present application provide a brain image processing method and related devices for improving the efficiency of image processing.

[0060] To better implement the brain image processing method provided by the embodiments of the present application, the present application provides a brain image processing architecture as Figure 1 shown. To analyze each brain region in the brain, first, an anatomical brain structure image needs to be input into a pre-trained brain region segmentation model to obtain the brain region segmentation result of the brain. Then, the brain region segmentation result is input into a brain volume algorithm, and the brain volume index value (i.e., the volume value of each brain region) of each brain region in the brain can be obtained. In addition, for each type of brain functional image, the brain region segmentation result and the brain functional image are input into the corresponding brain index model, and the brain functional index value output by the corresponding brain index model can be obtained. Among them, the brain index includes brain volume and various brain functions. It should be noted that the anatomical brain structure image and each brain functional state image can each be a three-dimensional image, or multiple two-dimensional images that can form a complete three-dimensional image, which can be specifically configured as needed.

[0061] Please refer to Figure 2 , based on the foregoing brain image processing architecture, the brain image processing method of the embodiments of the present application includes the following steps:

[0062] , 201. Obtain an anatomical brain structure image and at least one brain functional state image of the same brain.

[0063] Generally, any image of the brain structure captured by any imaging technique that needs to be used for brain region segmentation can be called an anatomical brain structure image. For example, an image acquired through T1-weighted imaging technology in magnetic resonance imaging.

[0064] In addition, a brain functional state image refers to an image that can reflect the conditions of different brain functions in the brain. Brain functions can include but are not limited to brain network connection and / or brain functional connection, which are not specifically limited here. For example, a brain network connection image acquired based on diffusion tensor imaging technology (reflecting the conditions of the brain network connection in the brain), and a brain functional connection image acquired based on resting-state functional magnetic resonance imaging technology (reflecting the conditions of the brain functional connection in the brain).

[0065] 202. Input the anatomical brain structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain.

[0066] Since each brain function is analyzed from the perspective of brain regions, such as the brain volume of each brain region, the brain network connections between each brain region and other brain regions, and the brain functional connections between each brain region and other brain regions. Therefore, in order to obtain the brain metrics of each brain region, first, a brain region segmentation operation should be performed on the brain to obtain the positions of each brain region included in the brain. Specifically, brain region segmentation can be achieved through a pre-trained brain region segmentation model.

[0067] Among them, the brain region segmentation model is a deep learning model, which can be a convolutional neural network or other network architectures that can be constructed by learnable parameters, and is not limited here. And before performing brain region segmentation, it is necessary to perform an image normalization operation on the brain anatomical structure image, dividing the pixel value of each pixel in the image by its maximum pixel value, so as to map the widely distributed image signal values to a unified interval and adapt to the input of the brain region segmentation model.

[0068] It should be noted that since there are various brain region segmentation criteria in medicine, the brain region segmentation model in the embodiments of the present application is not limited to a specific brain region segmentation method. In actual applications, by providing sample images using any brain region segmentation criterion to the brain region segmentation model during the training process of the brain region segmentation model, the trained brain region segmentation model can be made to learn to use this brain region segmentation criterion to perform brain region segmentation on the brain anatomical structure image.

[0069] 203. Input the brain region segmentation result into the brain volume algorithm to obtain the brain volume metric value of each brain region.

[0070] After obtaining the brain region segmentation result of the brain, the brain volume of each brain region can be calculated through the brain volume algorithm to obtain the brain volume metric value of each brain region.

[0071] 204. For each brain functional state image, input the brain functional state image and the brain region segmentation result into the brain metric model corresponding to the brain functional state image to obtain the brain functional metric value of each brain region.

[0072] In addition to calculating the brain volume, the metric values of corresponding brain functions can also be calculated through other brain functional state images. For example, for the brain network connection model, it can trace the brain fibers included in the brain network connection image and, in combination with the brain region segmentation result, determine the number of brain fibers connecting each brain region to each other brain region except this brain region as the brain network connection metric value of this brain region; for the brain functional connection model, it can determine the correlation strength between each brain region and each other brain region except this brain region according to the activity status of each part of the brain included in the brain functional connection image over a period of time and in combination with the brain region segmentation result (for example, when the activity intensity of brain region A increases, the activity intensity of brain region B also increases, indicating a relatively high correlation strength between brain region A and brain region B).

[0073] In addition, images acquired through different imaging techniques can also reflect different brain function conditions of the brain. Therefore, each brain function state image may correspond to one or more brain index algorithms, which are not specifically limited herein.

[0074] In the embodiments of the present application, through the above processing architecture, the index values of different brain indexes (brain volume and brain function state) can be flexibly determined, and the image processing efficiency is higher.

[0075] On the basis of the foregoing embodiments, in some specific implementation manners, by constructing brain index data, quantitative analysis of the brain can be realized from the perspective of different types of brain indexes. The details are as follows: according to the age of the object to which the brain belongs, obtain the statistical values of each brain region for each brain index in the brain index database, where the brain indexes include brain function and brain volume; based on the degree of difference between the index value and the statistical value of each brain region for each brain index, determine the health status of each brain region for each brain index respectively.

[0076] In fact, there is a large correlation between the brain indexes of the brain and the age of the object to which the brain belongs. That is to say, with the change of age, the index values of various brain indexes of each brain region of the brain will also change accordingly. Based on the above content, the embodiments of the present application construct a brain index database. Specifically, based on the index values of various brain indexes of the brains of objects of different ages (it should be noted that the objects of different ages should not suffer from diseases that affect the index values of various brain indexes), cluster analysis can be performed to obtain the statistical values of various brain indexes of the brains of objects of different ages or different age groups. Among them, the statistical value can be a specific value or a value range, which is not specifically limited herein.

[0077] In addition, since the brain index database records the statistical values of various brain indices of subjects of different ages, the present application can find the index values of various brain indices of the brain of the subject to which the brain belongs at the age of the subject, and compare the difference between the index value of each brain index in each brain region and the statistical value. Specifically, for each brain index, if the index value is inconsistent with the statistical value (that is, not equal to the statistical value or not within the value range of the statistical value), it is necessary to judge the health status in combination with the degree of difference between the index value and the statistical value. Specifically, the inconsistency between the index value and the statistical value may include but is not limited to the following situations: 1. Premature aging, that is, the index value is close to the statistical value of a much older age; 2. Dysplasia, that is, the subject to which the brain belongs is a child, the index value is close to the statistical value of a much younger age, and the much younger age is much less than the age of the subject to which the brain belongs; 3. Good condition, that is, the subject to which the brain belongs is an adult, the index value is close to the statistical value of a slightly younger age, and the slightly younger age is slightly less than the age of the subject to which the brain belongs. As can be seen from the above, there are various ways to judge the health status in combination with the degree of difference between the index value and the statistical value, and the embodiments of the present application do not make specific limitations.

[0078] Further, it is also possible to predict the brain age of the brain based on the index value and the statistical value of each brain index. For details, see the following steps: Determine the predicted brain age of each brain region for each brain index based on the statistical values of each brain region for each brain index at different ages and the index values of each brain region for each brain index; Determine the predicted brain age of the brain based on the predicted brain age of each brain region for each brain index.

[0079] First, for each brain index, determine the first change trend (i.e., rising or falling) of the statistical value of this (age of the subject to which the brain belongs) relative to the statistical value of a much younger age, and the second change trend (i.e., rising or falling) of the statistical value of this age relative to the statistical value of a much older age. Then, determine the third change trend of the statistical value relative to the index value, whether it is more in line with the first change trend or the second change trend, and find the target statistical value close to the index value from the statistical values of the corresponding age according to the change trend that is in line (for example, if it is in line with the first change trend, find it from the statistical values of a much younger age than this age). Finally, determine the target statistical value as the predicted brain age of the brain for the corresponding brain index. Or, it is also possible to directly find the target statistical value close to the index value and determine the age corresponding to the target statistical value as the predicted brain age of the brain for this brain index. The embodiments of the present application do not limit the determination of the predicted brain age of the brain for various brain indices.

[0080] It should be noted that the proximity referred to in the foregoing embodiments of the present application means that the absolute value between the two values does not exceed a preset threshold (if the statistical value is a value range, it is the absolute value between the minimum value, the median value or the maximum value), or the two values are the same, and no limitation is made here.

[0081] In addition to the above embodiments for quantitatively analyzing brain metrics, the present application also provides a visualization interface as described below to facilitate users in finding the metric values of various brain metrics of the brain. First, the embodiments of the present application perform three-dimensional reconstruction of the brain in the following manner: constructing a three-dimensional model of the brain based on brain anatomical structure images; in the three-dimensional model of the brain, each brain region in the brain region segmentation result is displayed in a different style.

[0082] Specifically, the brain anatomical structure images contain all the brain regions of the brain. Therefore, a three-dimensional model of the brain can be accurately constructed based on the brain anatomical structure images. Similarly, the brain region segmentation result is obtained based on the brain anatomical structure images. Therefore, further according to the position information of each brain region in the brain in the brain region segmentation result, each brain region is divided in the three-dimensional model, and each brain region in the brain region segmentation result is displayed in a different style, so as to display the brain region segmentation result in a more intuitive manner. Among them, the different styles can be any way of filling the brain regions with different colors, or filling the brain regions with different lines, etc., which can distinguish different brain regions, and are not limited here.

[0083] Furthermore, after obtaining the three-dimensional model of the brain, brain metric display can be performed in combination with the three-dimensional model and user operations, which can refer to the following steps: in response to the operation of the user selecting a target brain region in the three-dimensional model, the target brain region is displayed in a style different from other brain regions outside the target brain region; where the target brain region is any one of multiple brain regions; displaying the brain volume metric value and / or the brain function metric value of the target brain region.

[0084] That is to say, when the user selects the target brain region, a way can be selected to display other brain regions in the brain except the target brain region, and another way is used to display the target brain region; or, the colors of other brain regions except the target brain region are grayscale processed, and no processing is done to the color filling of the target brain region, etc., which are display methods for distinguishing the target brain region and all other brain regions. And, obtaining the metric values of various brain metrics of the target brain region, that is, the brain volume metric value and / or the brain function metric value.

[0085] Even further, the brain network connection images collected based on diffusion tensor imaging technology contain all the brain white matter fibers of the brain. Therefore, multiple bundles of brain white matter fibers included in the brain network connection images can also be displayed in the three-dimensional model. After the user selects the target brain region, the brain white matter fibers associated with the target brain region are displayed in a style different from other brain white matter fibers outside the brain white matter fibers associated with the target brain region. Specifically, the display method can refer to the aforementioned display method for distinguishing the target brain region and all other brain regions, and will not be elaborated here.

[0086] The foregoing described various embodiments in the solution of this application. Below, the application of the brain image processing method of this application will be described in a specific business scenario.

[0087] Based on the image processing method of the embodiments of this application, a brain quantitative analysis system can be constructed. This system includes an input module, a processing module, and an output module. The functions of each module will be described below.

[0088] 1. The input module is used to receive brain anatomical structure images and brain functional state images. The brain anatomical structure images and brain functional state images can be received simultaneously or differently. Among them, the images can be directly or indirectly sent by the corresponding acquisition devices. The format of each image can be DICOM and / or Mosaic, etc., which is not limited here.

[0089] 2. The processing module includes a brain region segmentation model, a brain volume algorithm, and each brain index model.

[0090] 3. The output module includes the brain region segmentation result (the specific format can refer to the foregoing image format and will not be elaborated here), the health status (document) of the brain in various brain indexes, and the predicted brain age (Wen Jing) of the brain, etc. In addition, the output module can provide a visualization interface through the front end of the computer device to interact with the user operation and display the brain three-dimensional model, etc.

[0091] Specifically, the brain network connection image can be a nuclear magnetic resonance image sequence collected based on the Diffusion Tensor Imaging (DTI) technology. This sequence is sensitive to the diffusion movement of water molecules. When many nerve fibers are arranged in the same direction, the diffusion movement of water molecules in the direction perpendicular to this direction is relatively difficult, and the diffusion movement in the direction parallel to this direction is relatively easy. Therefore, the DTI image can reflect the orientation of the white matter fiber bundle through the relevant average diffusion pattern of water molecules. For the input brain network connection image (i.e., the DTI image), this application first obtains a label map through the position information of the white matter in the brain region segmentation result, and then estimates the direction distribution function of each voxel in the white matter as the water molecule diffusion direction distribution function. The peak of the direction distribution function can well estimate the bundle segment direction at a certain point in the image, so it can be used to calculate the fiber directions in all voxels of the brain white matter. Next, each voxel of the brain white matter is specified as the starting point of fiber tracking, and then a fast algorithm EuDX is used to generate the streamline of each tracking. These streamlines form the result of brain white matter fiber tracking. It should be noted that fiber bundle tracking can use including but not limited to deterministic tracking algorithms and / or probabilistic tracking algorithms.

[0092] Brain functional connectivity images are magnetic resonance image sequences acquired based on resting-state functional magnetic resonance imaging (fMRI) technology. fMRI measures the oxygen content in the blood through magnetic signals. When a brain region is active, it requires more oxygen and nutrients to support metabolic activities, and the blood flow in the brain will also increase. Therefore, fMRI images can measure the minute changes in blood flow in the brain, thereby indirectly reflecting the activities of various regions in the brain. For the input brain functional connectivity images (i.e., fMRI images), by combining the position information of each brain region in the brain segmentation results, the mean value of the voxel values corresponding to each brain region in each time series in the fMRI images is calculated. Then, the Pearson correlation coefficient is used to calculate the correlation coefficients between each brain region and all other brain regions, which serves as the correlation strength between each brain region (i.e., the brain index of the brain functional connectivity image).

[0093] Utilize artificial intelligence algorithms to complete the quantitative analysis of brain indices, quickly and accurately provide brain segmentation results, and based on the brain segmentation results, give a quantitative analysis result report of brain regions (i.e., the health status of the brain volume), give a quantitative analysis result report of brain network connections based on fiber tracking (i.e., the health status of the brain network connections in the brain), and give a quantitative analysis result report of brain functional connectivity based on brain functional connectivity (i.e., the health status of the brain functional connectivity in the brain).

[0094] Brain quantitative analysis is the most accurate method for analyzing the brain's health, diagnosing brain diseases, and studying the brain's microscopic structure. However, quantitative analysis is difficult to achieve. On the one hand, because the structure of the brain is complex, it is difficult for doctors to give analysis results on the volume or fiber connection or functional connection in a short time. On the other hand, there is currently no automatic quantitative analysis method applied clinically. The currently commonly used technical means is to observe the brain region volume by simply segmenting images, or calculate various correlation indices through brain network connection images to replace fiber tracking. It is very difficult to quantitatively analyze the brain structure or connection information based on simple segmentation or just calculating indices, which is not conducive to the analysis of the brain's health status or the diagnosis of diseases. Therefore, this method establishes a rapid and automatic quantitative analysis workflow of brain indices based on deep learning algorithms. Through this method, accurate automatic segmentation of brain regions can be achieved, the volume of each brain region can be automatically calculated, and automatic fiber tracking and brain functional connectivity analysis can be completed, ultimately providing a brain quantitative analysis report. At the same time, an innovative interactive front-end display interface for brain quantitative analysis is provided, which can very conveniently display the quantitative analysis results.

[0095] It should be noted that the relevant data such as the brain images (including but not limited to brain anatomical structure images, brain functional state images, and sample images) adopted in the embodiments of the present application are all obtained after being authorized by the affiliated object. Moreover, when the embodiments of the present application are applied to specific products or technologies, the data involved need to obtain the permission or consent of the corresponding object, and the collection, use, and processing of the relevant data need to comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0096] Please refer to Figure 3 , the second aspect of the embodiments of the present application provides a computer device, including:

[0097] An acquisition unit 301, configured to acquire a brain anatomical structure image and at least one brain functional state image of the same brain;

[0098] A segmentation unit 302, configured to input the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain;

[0099] A calculation unit 303, configured to input the brain region segmentation result into a brain volume algorithm to obtain a brain volume index value of each brain region;

[0100] The calculation unit 303 is further configured to, for each brain functional state image, input the brain functional state image and the brain region segmentation result into the brain index model corresponding to the brain functional state image to obtain a brain function index value of each brain region.

[0101] In a specific implementation manner, the at least one brain functional state image includes: a brain network connection image and / or a brain functional connection image; the brain network connection image is acquired based on diffusion tensor imaging technology, and the brain functional connection image is acquired based on resting-state functional magnetic resonance imaging technology.

[0102] In a specific implementation manner, the device further includes: a construction unit and a display unit;

[0103] The construction unit is configured to construct a three-dimensional model of the brain based on the brain anatomical structure image;

[0104] The display unit is configured to display each brain region in the brain region segmentation result in different styles in the three-dimensional model of the brain.

[0105] In a specific implementation manner, the device further includes: a display unit;

[0106] The display unit is configured to, in response to an operation of a user selecting a target brain region in the three-dimensional model, display the target brain region in a style different from that of other brain regions outside the target brain region; where the target brain region is any one of the multiple brain regions;

[0107] A display unit, which is further configured to display the brain volume index value and / or the brain function index value of the target brain region.

[0108] In a specific implementation manner, the device further includes: a display unit;

[0109] The display unit is configured to, if at least one brain function state image includes a brain network connection image, display multiple bundles of cerebral white matter fibers included in the brain network connection image on a three-dimensional model;

[0110] The display unit is further configured to, in response to a user's operation of selecting a target brain region in the three-dimensional model, display the cerebral white matter fibers associated with the target brain region in a style different from that of other cerebral white matter fibers associated with the target brain region.

[0111] In a specific implementation manner, the device further includes: a determination unit;

[0112] The acquisition unit 301 is further configured to, according to the age of the object to which the brain belongs, acquire the statistical values of each brain region for each brain index in each brain index from a brain index database, where the brain index includes brain function and brain volume;

[0113] The determination unit is configured to respectively determine the health status of each brain region for each brain index based on the degree of difference between the index value and the statistical value of each brain region for each brain index.

[0114] In a specific implementation manner, the device further includes: a determination unit;

[0115] The determination unit is configured to determine the predicted brain age of each brain region for each brain index based on the statistical values of each brain region for each brain index at different ages and the index values of each brain region for each brain index.

[0116] The determination unit is further configured to determine the predicted brain age of the brain based on the predicted brain age of each brain region for each brain index.

[0117] Figure 4 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. The computer device 400 may include one or more central processing units (CPUs) 401 and a memory 405, and one or more application programs or data are stored in the memory 405.

[0118] Among them, the memory 405 may be volatile storage or persistent storage. The program stored in the memory 405 may include one or more modules, and each module may include a series of instruction operations on the computer device. Further, the central processor 401 may be configured to communicate with the memory 405 and execute a series of instruction operations in the memory 405 on the computer device 400.

[0119] The computer device 400 may also include one or more power supplies 402, one or more wired or wireless network interfaces 403, one or more input / output interfaces 404, and / or one or more operating systems, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0120] The central processing unit 401 may perform the operations executed by the image processing method in the foregoing Figures 1 to 3 illustrated embodiments, and details thereof will not be elaborated herein.

[0121] It should be noted that although the steps in the flowcharts involved in the embodiments are sequentially drawn according to the indications of the arrows, unless otherwise clearly stated in this article, the execution of these steps is not strictly limited in order, and these steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but may be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but may be executed alternately or alternately with at least some of the steps or stages in other steps or other steps.

[0122] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0123] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, indirect couplings or communication connections of devices or units, and can be in electrical, mechanical or other forms.

[0124] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] In addition, in each embodiment of the present application, each functional unit can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0126] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, read-only memory), random access memories (RAM, random access memory), magnetic disks, or optical discs that can store program codes.

[0127] The embodiments of the present application also provide a computer program product containing instructions. When the computer program product runs on a computer, it enables the computer to execute the brain image processing method as described above.

Claims

1. A method for brain image processing, characterized in that, it includes: obtaining a brain anatomical structure image and at least one brain functional state image of the same brain; inputting the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including multiple brain regions of the brain; inputting the brain region segmentation result into a brain volume algorithm to obtain a brain volume index value for each of the brain regions; for each of the brain functional state images, inputting the brain functional state image and the brain region segmentation result into the brain index model corresponding to the brain functional state image to obtain a brain function index value for each of the brain regions.

2. The method according to claim 1, characterized in that, the at least one brain functional state image includes: a brain network connection image and / or a brain functional connection image; the brain network connection image is acquired based on diffusion tensor imaging technology, and the brain functional connection image is acquired based on resting-state functional magnetic resonance imaging technology.

3. The method according to claim 1, characterized in that, the method further includes: constructing a three-dimensional model of the brain based on the brain anatomical structure image; in the three-dimensional model of the brain, displaying each of the brain regions in the brain region segmentation result in different styles.

4. The method according to claim 3, characterized in that, after obtaining the three-dimensional model of the brain, the method further includes: responding to an operation in which a user selects a target brain region in the three-dimensional model, and displaying the target brain region in a style different from that of other brain regions outside the target brain region; wherein the target brain region is any one of the multiple brain regions; displaying the brain volume index value and / or the brain function index value of the target brain region.

5. The method according to claim 4, characterized in that, the method further includes: if the at least one brain functional state image includes a brain network connection image, displaying multiple bundles of cerebral white matter fibers included in the brain network connection image in the three-dimensional model; responding to an operation in which a user selects a target brain region in the three-dimensional model, and displaying the cerebral white matter fibers associated with the target brain region in a style different from that of other cerebral white matter fibers outside the cerebral white matter fibers associated with the target brain region.

6. The method according to claim 1, characterized in that, the method further includes: according to the age of the object to which the brain belongs, obtaining the statistical value of each brain region under each brain index in the brain index database, where the brain index includes the brain function and the brain volume; based on the degree of difference between the index value and the statistical value of each brain region for each brain index, respectively determining the health status of each brain region for each brain index.

7. The method according to claim 1, characterized in that, the method further includes: based on the statistical value of each brain region for each brain index at different ages and the index value of each brain region for each brain index, determining the predicted brain age of each brain region for each brain index; based on the predicted brain age of each brain region for each brain index, determining the predicted brain age of the brain.

8. A computer device, characterized in that, it includes: An acquisition unit, configured to acquire a brain anatomical structure image and at least one brain functional state image of the same brain; A segmentation unit, configured to input the brain anatomical structure image into a pre-trained brain region segmentation model to obtain a brain region segmentation result including a plurality of brain regions of the brain; A calculation unit, configured to input the brain region segmentation result into a brain volume algorithm to obtain a brain volume index value of each of the brain regions; The calculation unit is further configured to, for each of the brain functional state images, input the brain functional state image and the brain region segmentation result into a brain index model corresponding to the brain functional state image to obtain a brain functional index value of each of the brain regions.

9. A computer device, Characterized in that, It includes: A central processing unit, a memory, and an input / output interface; The memory is a transient storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instruction operations in the memory to execute the method according to any one of claims 1 to 7.

10. A computer storage medium, Characterized in that, Instructions are stored in the computer storage medium, and when the instructions are executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images