Method, device, equipment and medium for extracting brain area from head and neck CT images

Abstract

The present application relates to the technical field of brain region extraction methods in head and neck CT images, and in particular to a method, device, equipment and medium for extracting brain regions in head and neck CT images, wherein the method comprises: obtaining an actual CT image of the head and neck to be extracted; establishing a three-dimensional change relationship between a reference CT image and the actual CT image, and based on the three-dimensional change relationship, changing the brain region annotation result of the reference CT image to the actual CT image to obtain the brain region annotation result of the actual CT image; extracting the actual brain region image of the head and neck to be extracted from the actual CT image according to the brain region annotation result of the actual CT image. Therefore, the embodiment of the present application can use a watering algorithm to extract all brain tissue regions in the skull, and the extraction result is complete, and operations such as expansion and corrosion can be used to remove extra-brain tissue in the case of skull defects and thin skulls, thereby improving robustness.

Description

Technical Field

[0001] The present application relates to the technical field of methods for extracting brain regions in head and neck CT (Computed Tomography) images, and in particular to a method, device, equipment and medium for extracting brain regions in head and neck CT images. Background Art

[0002] CT is an important medical image for diagnosing brain diseases, and plays an important role in diagnosing brain diseases including cerebral hemorrhage, cerebral infarction, tumors, etc. However, it is labor-intensive and slow for doctors to diagnose head and neck CT images. With the development of intelligent algorithms such as neural networks, many automatic disease diagnosis methods for head and neck CT have emerged, which can help doctors diagnose brain diseases more quickly and accurately.

[0003] However, head and neck CT is a three-dimensional image, which generally includes various parts of the head and neck, including images of the brain, skull, maxillofacial region, neck, etc. As a result, when performing automatic diagnosis of brain diseases on head and neck CT, it is easily interfered by areas outside the brain, thereby reducing the effectiveness of automatic diagnosis. Summary of the invention

[0004] The present application provides a method, device, electronic device and storage medium for extracting brain areas in head and neck CT images. All brain tissue areas in the skull can be extracted using a watering algorithm, and the extraction results are complete. In addition, operations such as expansion and corrosion can also be used to remove extra-brain tissue in cases of skull defects and thin skulls, thereby improving robustness.

[0005] The first aspect of the present application provides a method for extracting a brain region in a head and neck CT image, comprising the following steps: obtaining an actual CT image of the head and neck to be extracted; establishing a three-dimensional change relationship between a reference CT image and the actual CT image, and based on the three-dimensional change relationship, changing the brain region annotation result of the reference CT image to the actual CT image to obtain the brain region annotation result of the actual CT image; and extracting the actual brain region image of the head and neck to be extracted from the actual CT image according to the brain region annotation result of the actual CT image.

[0006] Furthermore, the extracting the image of the brain region from the actual CT image according to the brain region annotation result of the actual CT image includes: according to the brain region annotation result of the actual CT image, removing the image other than the brain region in each layer of the two-dimensional image of the actual CT image to obtain the brain region image; and generating the head and neck brain region image to be extracted according to all brain region images.

[0007] Furthermore, according to the brain region annotation result of the actual CT image, the image outside the brain region in each layer of the two-dimensional image of the actual CT image is removed to obtain a brain region image, including: identifying the brain region and the region outside the brain region in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and setting the pixel value corresponding to the brain region to 1 and the pixel value corresponding to the region outside the brain region to 0 to obtain a first image; corroding the first image through a first convolution kernel to obtain a second image; diffusing the pixel values ​​of the brain region in the second image that are not within the brain region annotation result using a preset watering algorithm to obtain a diffused region, and merging the diffused region into the brain region annotation result to obtain a third image; dilating the third image through a second convolution kernel to obtain a fourth image, and multiplying the fourth image by the first image to obtain a brain region image.

[0008] Furthermore, the establishing of the three-dimensional change relationship between the reference CT image and the actual CT image includes: transforming the reference CT image onto the actual CT image through affine transformation and B-spline to obtain the three-dimensional change relationship.

[0009] The second aspect of the present application provides a device for extracting the brain area in a head and neck CT image, including: an acquisition module, used to acquire the actual CT image of the head and neck to be extracted; a change module, used to establish a three-dimensional change relationship between a reference CT image and the actual CT image, and based on the three-dimensional change relationship, change the brain area annotation result of the reference CT image to the actual CT image to obtain the brain area annotation result of the actual CT image; an extraction module, used to extract the actual brain area image of the head and neck to be extracted from the actual CT image according to the brain area annotation result of the actual CT image.

[0010] Furthermore, the extraction module is used to: remove the images outside the brain area in each layer of the two-dimensional image of the actual CT image according to the brain area labeling result of the actual CT image to obtain the brain area image; and generate the head and neck brain area image to be extracted according to all brain area images.

[0011] Furthermore, the extraction module is further used to: identify the brain area and the area outside the brain area in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and set the pixel value corresponding to the brain area to 1, and the pixel value corresponding to the area outside the brain area to 0, to obtain a first image; corrode the first image through a first convolution kernel to obtain a second image; use a preset watering algorithm to diffuse the pixel values ​​of the brain area in the second image that are not within the brain area annotation result to obtain a diffused area, and merge the diffused area into the brain area annotation result to obtain a third image; expand the third image through a second convolution kernel to obtain a fourth image, and multiply the fourth image by the first image to obtain a brain area image.

[0012] Furthermore, the change module is used to: transform the vertical reference CT image to the actual CT image through affine transformation and B-spline to obtain the three-dimensional change relationship.

[0013] A third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for extracting the brain region in the head and neck CT image as described in the above embodiment.

[0014] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the method for extracting the brain area in the head and neck CT image as described in the above embodiment.

[0015] Therefore, this application has at least the following beneficial effects:

[0016] The water filling algorithm can be used to extract all brain tissue areas in the skull, and the extraction result is complete. In addition, operations such as expansion and corrosion can also be used to remove extra-brain tissue in cases of skull defects and thin skulls to improve robustness. Therefore, automatic brain extraction of head and neck CT is achieved using operations including three-dimensional alignment, water filling algorithm, expansion, corrosion, etc.

[0017] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0019] Figure 1 A schematic diagram of a process for extracting a brain region from a head and neck CT image according to an embodiment of the present application;

[0020] Figure 2 This is an example diagram of the extraction result of the brain area in the head and neck CT image provided according to the embodiment of the present application;

[0021] Figure 3 This is an example diagram of a device for extracting a brain region from a head and neck CT image according to an embodiment of the present application;

[0022] Figure 4 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0024] With the development of intelligent algorithms such as neural networks, automatic disease diagnosis methods for head and neck CT can help doctors diagnose diseases more quickly and accurately, while some automatic diagnosis algorithms only diagnose brain diseases, such as cerebral hemorrhage and cerebral infarction, and do not require information from other parts of the brain. Therefore, in order to improve the effectiveness of these methods, parts other than the brain can be deleted from the image, and only the brain image can be retained, which is conducive to eliminating interference from other parts.

[0025] In order to extract the brain area from the head and neck CT images, the embodiment of the present application proposes a method, device, electronic device and storage medium for extracting the brain area from the head and neck CT images. The method, device, electronic device and storage medium for extracting the brain area from the head and neck CT images of the embodiment of the present application will be described with reference to the accompanying drawings.

[0026] Specifically, Figure 1 A flowchart of a method for extracting the brain region from a head and neck CT image provided in an embodiment of the present application.

[0027] like Figure 1 As shown, the method for extracting the brain region in the head and neck CT image comprises the following steps:

[0028] In step S101, an actual CT image of the head and neck to be extracted is obtained.

[0029] In step S102, a three-dimensional change relationship is established between the reference CT image and the actual CT image, and based on the three-dimensional change relationship, the brain region annotation result of the reference CT image is changed to the actual CT image to obtain the brain region annotation result of the actual CT image.

[0030] In an embodiment of the present application, a three-dimensional change relationship between a reference CT image and an actual CT image is established, including: transforming the reference CT image to the actual CT image through affine transformation and B-spline to obtain the three-dimensional change relationship.

[0031] It can be understood that the embodiment of the present application can obtain the three-dimensional change relationship through three-dimensional registration. Specifically, a three-dimensional CT image of the head and neck is obtained, which is called the reference CT, and the brain area is annotated on each two-dimensional layer thereon to prepare for reference annotation; and the reference CT and the brain area annotation are transformed to the actual CT through three-dimensional registration, and the brain area annotation on the actual CT is obtained to achieve three-dimensional registration and transformation.

[0032] It should be noted that the affine transformation is obtained through registration, where registration is an optimization process, which allows the reference CT to undergo two transformations in sequence and the result obtained to be as similar as possible to the target CT. The definition of "similar" can be the pixel-by-pixel difference between the two three-dimensional images or other definition methods.

[0033] Specifically, (1) Reference annotation preparation. A head and neck CT image is obtained, which is called the reference CT. At the same time, the brain region of each layer is manually annotated. The brain regions of all layers are used as 3D brain region annotation masks. The annotated brain regions must be completely surrounded by the skull.

[0034] (2) Three-dimensional registration and transformation. The reference CT is registered to the actual CT image to be subjected to the brain extraction operation through three-dimensional transformation, where the three-dimensional transformation includes an affine transformation and a B-spline transformation. The brain region mask on the reference CT is then annotated using the same three-dimensional transformation to obtain the initial brain mask on the actual CT.

[0035] It should be noted that an important feature of brain CT is that the brain is surrounded by the skull, the pixel value of the brain is low, while the pixel value of the skull is very high. At the same time, the brain structures of different people are relatively similar; therefore, three-dimensional alignment of the brain is easier to achieve good results.

[0036] In step S103, the actual brain region image of the head and neck to be extracted is extracted from the actual CT image according to the brain region labeling result of the actual CT image.

[0037] It can be understood that the embodiments of the present application can realize automatic extraction of brain area images based on the brain area labeling results, so as to eliminate interference from areas outside the brain when automatically diagnosing brain diseases on head and neck CT and achieve better diagnostic results.

[0038] In an embodiment of the present application, an image of the brain region is extracted from the actual CT image according to the brain region annotation result of the actual CT image, including: according to the brain region annotation result of the actual CT image, removing the image other than the brain region in each layer of the two-dimensional image of the actual CT image to obtain a brain region image; generating a head and neck brain region image to be extracted based on all brain region images.

[0039] It is understandable that the embodiments of the present application can remove the content of areas outside the brain (such as the skull, face, neck, etc.) and retain only the brain area to achieve brain extraction, so that when automatically diagnosing brain diseases on head and neck CT, interference from areas outside the brain can be eliminated to achieve better diagnostic results.

[0040] In an embodiment of the present application, according to the brain region annotation result of the actual CT image, the image outside the brain region in each layer of the two-dimensional image of the actual CT image is removed to obtain a brain region image, including: identifying the brain region and the region outside the brain region in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and setting the pixel value corresponding to the brain region to 1 and the pixel value corresponding to the region outside the brain region to 0 to obtain a first image; corroding the first image through a first convolution kernel to obtain a second image; diffusing the pixel values ​​of the brain region in the second image that are not in the brain region annotation result by using a preset watering algorithm to obtain a diffused region, and merging the diffused region into the brain region annotation result to obtain a third image; dilating the third image through a second convolution kernel to obtain a fourth image, and multiplying the fourth image by the first image to obtain a brain region image.

[0041] Among them, the second convolution kernel is larger than the first convolution kernel; the preset watering algorithm can be selected according to actual needs and is not specifically limited to this.

[0042] It is understandable that the embodiment of the present application can use the brain region annotation on the actual CT to process each layer of the actual CT separately, and obtain the final brain extraction result through operations such as floodfill, expansion, and corrosion. Figure 2 As shown, the left side is a certain layer of the CT image (under the soft tissue window), and the right side is the result after extracting the brain area. The result of extracting the brain area image in the embodiment of the present application is complete and has good robustness.

[0043] Specifically, since the initial brain mask obtained by the registration transformation does not necessarily completely cover the brain area, the embodiment of the present application processes each layer of the actual CT two-dimensional image separately. The processing of each layer of the two-dimensional image includes:

[0044] (1) Based on the knowledge of imaging, the lower limit T1 and upper limit T2 of the pixel value of brain tissue in CT are determined. For each layer of the 2D CT image, the pixel value between T1 and T2 is set to 1, and the other positions are set to 0 to obtain a mask, which is called M1. M1 includes not only the brain area, but also the soft tissue outside the brain, such as the soft tissue of the maxillofacial region and the neck. Therefore, M1 cannot be directly regarded as the brain area.

[0045] (2) Use a smaller kernel K1 to erode M1 to obtain M2. The purpose is to reduce its scope and prevent the brain tissue in M1 from connecting with the soft tissue outside the brain due to the lack of skull or thin skull.

[0046] Specifically: the first image M1 is a binary image, the area with a pixel value of 1 is the brain tissue or soft tissue outside the brain, and the area with a pixel value of 0 is other tissue. The corrosion operation is performed using a two-dimensional convolution kernel. Taking a rectangular kernel of size K*K as an example, for each position of M1, its pixel value becomes the minimum value of all the original M1 pixel values ​​in the nearby K*K range. Because the corrosion operation of the image obtains different results depending on the image value, therefore, in the embodiment of the present application, the position of the brain tissue (or soft tissue) is 1, and the other positions are 0 to ensure the correctness of the results obtained by the corrosion operation.

[0047] (3) Traverse each pixel of the two-dimensional CT image in turn. If the pixel position is in the initial brain mask and is not currently classified as a brain region, use the floodfill algorithm to diffuse from this pixel as the starting point, and classify all diffused pixels as brain regions. However, diffusion can only be performed at positions where the M2 value is 1. By setting all pixels classified as brain regions to 1 and other pixels to 0, a mask is obtained, which is called M3. M3 roughly represents the brain region, but because M2 is obtained through corrosion operations, M3 does not completely cover the brain region at the edge.

[0048] (4) M3 is expanded through a kernel K2 that is slightly larger than K1 to obtain M4, and M4 is multiplied by M1 to obtain the final brain region mask M. By multiplying M with the original CT two-dimensional image, the complete brain region can be retained while removing the image of the area outside the brain.

[0049] Specifically, the dilation process is similar to the above-mentioned erosion, except that the final value of each pixel in the erosion operation is the minimum value of all pixels covered by the convolution kernel centered on it, while the dilation operation is the maximum value.

[0050] According to the method for extracting the brain area in the head and neck CT images proposed in the embodiment of the present application, all brain tissue areas in the skull can be extracted using a water filling algorithm, and the extraction result is complete. In addition, operations such as expansion and corrosion can also be used to remove extra-brain tissue in cases of skull defects and thin skulls, thereby improving robustness. Therefore, automatic brain extraction of head and neck CT is achieved using operations including three-dimensional alignment, water filling algorithm, expansion, corrosion, etc.

[0051] Next, a device for extracting the brain region in the head and neck CT image according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0052] Figure 3 It is a block diagram of a device for extracting a brain region from a head and neck CT image according to an embodiment of the present application.

[0053] like Figure 3 As shown, the device 10 for extracting the brain region in the head and neck CT image includes: an acquisition module 100, a change module 200 and an extraction module 300.

[0054] Among them, the acquisition module 100 is used to acquire the actual CT image of the head and neck to be extracted; the change module 200 is used to establish a three-dimensional change relationship between the reference CT image and the actual CT image, and based on the three-dimensional change relationship, change the brain area annotation result of the reference CT image to the actual CT image to obtain the brain area annotation result of the actual CT image; the extraction module 300 is used to extract the actual brain area image of the head and neck to be extracted from the actual CT image according to the brain area annotation result of the actual CT image.

[0055] Furthermore, the extraction module 300 is used to: remove the images other than the brain area in each layer of the two-dimensional image of the actual CT image according to the brain area labeling result of the actual CT image to obtain the brain area image; and generate the head and neck brain area image to be extracted according to all the brain area images.

[0056] Furthermore, the extraction module 300 is further used to: identify the brain area and the area outside the brain area in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and set the pixel value corresponding to the brain area to 1, and the pixel value corresponding to the area outside the brain area to 0, to obtain a first image; corrode the first image through a first convolution kernel to obtain a second image; use a preset watering algorithm to diffuse the pixel values ​​of the brain area in the second image that are not in the brain area annotation result to obtain a diffused area, and merge the diffused area into the brain area annotation result to obtain a third image; expand the third image through a second convolution kernel to obtain a fourth image, and multiply the fourth image by the first image to obtain a brain area image.

[0057] Furthermore, the change module 200 is used to transform the three-dimensional reference CT image to the actual CT image through affine transformation and B-spline to obtain a three-dimensional change relationship.

[0058] It should be noted that the above explanation of the embodiment of the method for extracting the brain region in the head and neck CT image is also applicable to the device for extracting the brain region in the head and neck CT image of this embodiment, and will not be repeated here.

[0059] According to the device for extracting the brain area in the head and neck CT image proposed in the embodiment of the present application, all brain tissue areas in the skull can be extracted by using a water filling algorithm, and the extraction result is complete. In addition, operations such as expansion and corrosion can also be used to remove extra-brain tissue in the case of skull defects and thin skulls, thereby improving robustness. Therefore, automatic brain extraction of head and neck CT is achieved by using operations including three-dimensional alignment, water filling algorithm, expansion, corrosion, etc.

[0060] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0061] Memory 401 , processor 402 , and a computer program stored in the memory 401 and executable on the processor 402 .

[0062] When the processor 402 executes the program, the method for extracting the brain region in the head and neck CT image provided in the above embodiment is implemented.

[0063] Furthermore, the electronic device further comprises:

[0064] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0065] The memory 401 is used to store computer programs that can be executed on the processor 402 .

[0066] The memory 401 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0067] If the memory 401, the processor 402 and the communication interface 403 are implemented independently, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0068] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0069] The processor 402 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0070] The embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for extracting the brain area in the head and neck CT image.

[0071] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0072] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0073] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0074] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array, a field programmable gate array, etc.

[0075] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0076] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for extracting brain regions in head and neck CT images, characterized in that: The following steps are involved: Obtain the actual CT image of the head and neck to be extracted; Establishing a three-dimensional change relationship between the reference CT image and the actual CT image, and based on the three-dimensional change relationship, changing the brain region annotation result of the reference CT image to the actual CT image to obtain the brain region annotation result of the actual CT image; as well as Extracting the actual brain region image of the head and neck to be extracted from the actual CT image according to the brain region labeling result of the actual CT image; The step of extracting an image of the brain region from the actual CT image according to the brain region labeling result of the actual CT image includes: According to the brain region labeling result of the actual CT image, remove the image other than the brain region in each layer of the two-dimensional image of the actual CT image to obtain a brain region image; Generating the head and neck brain region image to be extracted according to all brain region images; The step of removing the image outside the brain region in each layer of the two-dimensional image of the actual CT image according to the brain region labeling result of the actual CT image to obtain the brain region image includes: Identifying the brain region and the region outside the brain region in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and setting the pixel value corresponding to the brain region to 1 and the pixel value corresponding to the region outside the brain region to 0, to obtain a first image; Performing corrosion processing on the first image by using a first convolution kernel to obtain a second image; Using a preset watering algorithm, the pixel values ​​of the brain region in the second image that are not in the brain region annotation result are diffused to obtain a diffused region, and the diffused region is incorporated into the brain region annotation result to obtain a third image; The third image is expanded by a second convolution kernel to obtain a fourth image, and the fourth image is multiplied by the first image to obtain a brain region image.

2. The method according to claim 1, characterized in that The establishing of a three-dimensional change relationship between the reference CT image and the actual CT image comprises: The three-dimensional reference CT image is transformed onto the actual CT image through affine transformation and B-spline transformation to obtain the three-dimensional change relationship.

3. A device for extracting brain regions from head and neck CT images, characterized in that: include: An acquisition module, used to acquire the actual CT image of the head and neck to be extracted; a change module, used to establish a three-dimensional change relationship between the reference CT image and the actual CT image, and based on the three-dimensional change relationship, change the brain region annotation result of the reference CT image to the actual CT image to obtain the brain region annotation result of the actual CT image; as well as An extraction module, used for extracting the actual brain region image of the head and neck to be extracted from the actual CT image according to the brain region annotation result of the actual CT image; The extraction module is used to: According to the brain region labeling result of the actual CT image, remove the image other than the brain region in each layer of the two-dimensional image of the actual CT image to obtain a brain region image; Generating the head and neck brain region image to be extracted according to all brain region images; The extraction module is further used for: Identifying the brain region and the region outside the brain region in each layer of the two-dimensional image according to the pixel value range of the brain tissue, and setting the pixel value corresponding to the brain region to 1 and the pixel value corresponding to the region outside the brain region to 0, to obtain a first image; Performing corrosion processing on the first image by using a first convolution kernel to obtain a second image; Using a preset watering algorithm, the pixel values ​​of the brain region in the second image that are not in the brain region annotation result are diffused to obtain a diffused region, and the diffused region is incorporated into the brain region annotation result to obtain a third image; The third image is expanded by a second convolution kernel to obtain a fourth image, and the fourth image is multiplied by the first image to obtain a brain region image.

4. The device according to claim 3, characterized in that The change module is used to: The three-dimensional reference CT image is transformed onto the actual CT image through affine transformation and B-spline transformation to obtain the three-dimensional change relationship.

5. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for extracting the brain region in the head and neck CT image as described in any one of claims 1 to 2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for extracting the brain area in the head and neck CT image as described in any one of claims 1-2.

Images