A bronchial extraction method, device and computer equipment based on CT images

Abstract

The present application provides a bronchial extraction method, device and computer equipment based on CT images, which relates to the field of CT image processing technology. The method includes: obtaining a lung electronic computer tomography CT image to be segmented; segmenting the CT image to obtain a pulmonary interstitial mask image, a bronchial mask image and a pulmonary artery mask image; segmenting the bronchial mask image to obtain at least three bronchial segment mask images, segmenting the pulmonary artery mask image to obtain at least three pulmonary artery segment masks, determining the classification result of the bronchial mask image based on the adjacent and docking relationship of each bronchial segment, and determining the classification result of the pulmonary artery mask image based on the adjacent and docking relationship of the pulmonary artery; fusing the classification result of the bronchial mask image and the classification result of the pulmonary artery mask image, and generating the final bronchial mask image based on the accompanying relationship of the pulmonary artery corresponding to the bronchi.

Description

Technical Field

[0001] The present application relates to the technical field of CT image processing, and in particular to a bronchial extraction method, device and computer equipment based on CT images. Background Art

[0002] Segmenting bronchial mask images based on lung CT images helps doctors analyze lung diseases, plan surgeries, and reduce treatment risks for patients. However, since the bronchial tubes in the lungs have complex branching structures and topological morphologies, the branching levels can be as high as 10 or more depending on the size of the tubes, which makes it difficult to complete manually.

[0003] Due to the complex distribution of pulmonary bronchi, it is difficult to classify pulmonary bronchi, and the existing pulmonary bronchi classification technology has poor classification accuracy and low efficiency. Therefore, how to improve the accuracy and efficiency of pulmonary bronchi classification has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the technical problems in the related art, the present application provides a bronchial extraction method, device and computer equipment based on CT images.

[0005] In order to achieve the above objectives, the technical solutions adopted in this application include:

[0006] According to a first aspect of the present application, a bronchial extraction method based on CT images is provided, comprising:

[0007] Acquire lung computer tomography (CT) images to be segmented;

[0008] The image segmentation model is used to segment the CT images to obtain the pulmonary interstitial mask image GJ, the bronchial mask image GZ and the pulmonary artery mask image GD;

[0009] The bronchial mask image is segmented to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3; the pulmonary artery mask image is segmented to obtain at least three pulmonary artery segment masks GD-1, GD-2 and GD-3; the classification result of the bronchial mask image GZ is determined according to the adjacent and docking relationship of each bronchial segment; the classification result of the pulmonary artery mask image GD is determined according to the adjacent and docking relationship of the pulmonary artery;

[0010] Then, the classification results of the confirmed bronchial mask image GZ and the confirmed pulmonary artery mask image GD are fused, and the final bronchial mask image is generated in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

[0011] Optionally, segmenting the CT image using the image segmentation model specifically includes:

[0012] Obtaining the interstitial lung mask image: firstly, the threshold segmentation method is used to separate the lung parenchyma from other structures of the thoracic cavity, and then the connected left and right lungs are obtained based on the region growing method, the main trachea and the main bronchi extending into the lungs are removed, and then the holes in the parenchyma are filled to obtain the lung parenchyma image. The interstitial lung mask image GJ is obtained using the CT image and the lung parenchyma image;

[0013] Obtaining a lung interstitial mask image GJ: performing bronchi segmentation on the lung bronchi in the CT image to obtain a bronchial mask image GZ;

[0014] Obtaining a pulmonary artery vascular mask image GD: Obtain and fuse the arteriovenous segmentation result map and the vascular distribution map corresponding to the pulmonary vessels in the CT image respectively to obtain a vascular comprehensive map, segment the vascular comprehensive map based on the connected area threshold, and obtain a pulmonary artery vascular mask image GD.

[0015] Optionally, segmenting the bronchial mask image specifically includes: performing secondary segmentation on the obtained bronchial mask image GZ to obtain left and right main bronchial segment mask images GZ-1, left and right secondary bronchial segment mask images GZ-2, and left and right lobular bronchial segment images GZ-3 respectively;

[0016] Then, the classification result of the bronchial mask image GZ is determined according to the adjacent and docking relationship of each bronchial segment.

[0017] Optionally, segmenting the pulmonary artery vascular mask image specifically includes: performing secondary segmentation on the obtained pulmonary artery vascular mask image GD to obtain left and right pulmonary artery main vessel segment mask images GD-1, left and right pulmonary artery secondary vessel segment mask images GD-2, and left and right pulmonary artery capillary vessel segment mask images GD-3 respectively;

[0018] Then, according to the adjacent and docking relationship of the pulmonary artery vessels, the classification result of the pulmonary artery vessel mask image GD is determined.

[0019] Optionally, according to the accompanying relationship between the bronchi and pulmonary arteries corresponding to GZ-1 and GD-1, GZ-2 and GD-2, and GZ-3 and GD-3, respectively, the classification result of the confirmed bronchial mask image GZ and the classification result of the confirmed pulmonary artery mask image GD are fused;

[0020] Based on the fused image, a final bronchial mask image is generated in the lung interstitial mask image GJ.

[0021] Optionally, in the process of fusing the classification result of the confirmed bronchial mask image GZ and the classification result of the confirmed pulmonary artery mask image GD,

[0022] If the connected area corresponding to a certain left or right lobular bronchiole segment is connected to the area of ​​the arterial blood vessel corresponding to a certain main blood vessel segment mask image in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right lobular bronchiole segment is not connected to the area of ​​the arterial blood vessel corresponding to all the main blood vessel segment mask images in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is not regarded as the confirmed classification result of the bronchial mask image GZ;

[0023] If the connected area corresponding to a certain left or right secondary bronchus segment is connected to the area of ​​the arterial blood vessel corresponding to a certain secondary blood vessel segment mask image in the corresponding left or right pulmonary artery secondary blood vessel segment mask image GD-2, then the left or right secondary bronchus segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right secondary bronchus segment is not connected to the area of ​​the arterial blood vessel corresponding to all the secondary blood vessel segment mask images in the corresponding left or right pulmonary artery secondary blood vessel segment mask image GD-2, then the left or right secondary bronchus segment is not regarded as the confirmed classification result of the bronchial mask image GZ;

[0024] If the connected area corresponding to a certain left or right main bronchus segment is connected to the area of ​​the arterial blood vessel corresponding to a main blood vessel segment mask image in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right main bronchus segment is not connected to the area of ​​the arterial blood vessel corresponding to all the main blood vessel segment mask images in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is not regarded as the confirmed classification result of the bronchial mask image GZ.

[0025] According to the second aspect of the present application, a bronchial extraction device based on CT images is also provided, characterized in that the bronchial extraction method based on CT images described in any technical solution in the first aspect of the present application is applied, and the device comprises:

[0026] Image acquisition module, used to obtain lung computer tomography CT images to be segmented

[0027] The first image processing module is used to segment the CT image to obtain a pulmonary interstitial mask image GJ, a bronchial mask image GZ and a pulmonary artery mask image GD;

[0028] The second image processing module is used to segment the bronchial mask image to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3, segment the pulmonary artery mask image to obtain at least three pulmonary artery segment masks GD-1, GD-2 and GD-3, determine the classification result of the bronchial mask image GZ according to the adjacent and docking relationship of each bronchial segment, and determine the classification result of the pulmonary artery mask image GD according to the adjacent and docking relationship of the pulmonary artery;

[0029] The third image processing module is used to fuse the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD, and generate a final bronchial mask image in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

[0030] According to the third aspect of the present application, a computer device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the bronchial extraction method based on CT images described in any one of the technical solutions in the first aspect of the present application can be implemented.

[0031] According to the fourth aspect of the present application, a computer-readable storage medium is also provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, it can implement the steps of the bronchial extraction method based on CT images described in any technical solution in the first aspect of the present application.

[0032] Beneficial effects:

[0033] 1. Through the above technical scheme, the present application first divides the CT image into a pulmonary interstitial mask image GJ, a bronchial mask image GZ and a pulmonary artery mask image GD, and then performs secondary segmentation on the segmented bronchial mask image GZ and the pulmonary artery mask image GD. The classification result of the bronchial mask image GZ and the classification result of the pulmonary artery mask image GD are determined respectively through the adjacent and docking relationship of each bronchial segment and the adjacent and docking relationship of the pulmonary artery. Finally, the confirmed classification result of the bronchial mask image GZ and the confirmed classification result of the pulmonary artery mask image GD are fused, and the final bronchial mask image is generated in the pulmonary interstitial mask image GJ according to the accompanying relationship between the bronchi and the pulmonary arteries. In this way, it is not only convenient to verify and confirm the bronchial mask image according to the accompanying relationship between the bronchus and the pulmonary artery vessels, but also convenient for doctors to compare and analyze the bronchial mask image with the lung interstitial mask image, which is conducive to improving the classification accuracy and efficiency of the bronchial mask image.

[0034] 2. Other beneficial effects or advantages of the present application will be described in detail in the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative labor.

[0036] in:

[0037] Figure 1 It is a schematic flow chart of the steps of a bronchial extraction method based on CT images provided in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0039] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0040] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes other steps or units that are not listed, or optionally includes other steps or units that are inherent to these processes, methods, products or devices. Among them, it should also be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or design solutions. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0041] The technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0042] Example 1

[0043] like Figure 1 As shown, according to the first aspect of the present application, this embodiment provides a bronchial extraction method based on CT images, comprising:

[0044] Acquire lung computer tomography (CT) images to be segmented;

[0045] The image segmentation model is used to segment the CT images to obtain the pulmonary interstitial mask image GJ, the bronchial mask image GZ and the pulmonary artery mask image GD;

[0046] The bronchial mask image is segmented to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3; the pulmonary artery mask image is segmented to obtain at least three pulmonary artery segment masks GD-1, GD-2 and GD-3; the classification result of the bronchial mask image GZ is determined according to the adjacent and docking relationship of each bronchial segment; the classification result of the pulmonary artery mask image GD is determined according to the adjacent and docking relationship of the pulmonary artery;

[0047] Then, the classification results of the confirmed bronchial mask image GZ and the confirmed pulmonary artery mask image GD are fused, and the final bronchial mask image is generated in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

[0048] Through the above technical scheme, the present application first divides the CT image into a pulmonary interstitial mask image GJ, a bronchial mask image GZ and a pulmonary artery mask image GD, and then performs secondary segmentation on the segmented bronchial mask image GZ and the pulmonary artery mask image GD. The classification result of the bronchial mask image GZ and the classification result of the pulmonary artery mask image GD are determined respectively through the adjacent and docking relationship of each bronchial segment and the adjacent and docking relationship of the pulmonary artery. Finally, the confirmed classification result of the bronchial mask image GZ and the confirmed classification result of the pulmonary artery mask image GD are fused, and the final bronchial mask image is generated in the pulmonary interstitial mask image GJ according to the accompanying relationship between the bronchi and the pulmonary arteries. In this way, it is not only convenient to verify and confirm the bronchial mask image according to the accompanying relationship between the bronchus and the pulmonary artery vessels, but also convenient for doctors to compare and analyze the bronchial mask image with the lung interstitial mask image, which is conducive to improving the classification accuracy and efficiency of the bronchial mask image.

[0049] Specifically, first, since there is a companion relationship between the bronchi and the pulmonary arteries, by segmenting the bronchial mask image into at least three bronchial segment mask images, and through the adjacent and docking relationship of each bronchial segment, the classification result of the bronchial mask image GZ can be quickly and accurately determined, and by similarly segmenting the pulmonary artery mask image to obtain at least three pulmonary artery segment mask images, and through the adjacent and docking relationship of the pulmonary artery vessels, the classification result of the pulmonary artery mask image GD can be quickly and accurately determined. In this way, based on the one-to-one correspondence between the three groups of images, the classification result of the confirmed bronchial mask image GZ and the confirmed pulmonary artery mask image GD can be accurately and quickly fused to generate the final bronchial mask image.

[0050] Second, since the interstitial lung mask image GJ includes connective tissue, lymphatic vessels, nerve fibers and blood vessels, it can serve as the analysis basis for the pulmonary artery vascular image. In this way, combined with the above-mentioned fusion image, the final bronchial mask image can be quickly and accurately generated directly in the interstitial lung mask image GJ, so that doctors can make more accurate and rapid judgments and analyses.

[0051] In one embodiment of the present application, segmenting a CT image using an image segmentation model specifically includes: obtaining a lung interstitial mask image: first using a threshold segmentation method to separate the lung parenchyma from other structures of the chest cavity, then obtaining a connected left and right lung based on a region growing method, removing the main trachea and the main bronchi extending into the lungs, and then filling the holes in the parenchyma to obtain a lung parenchyma image, and using the CT image and the lung parenchyma image to obtain a lung interstitial mask image GJ; obtaining a lung interstitial mask image GJ: performing bronchi segmentation on the lung bronchi in the CT image to obtain a bronchial mask image GZ; obtaining a pulmonary artery vascular mask image GD: respectively obtaining and fusing the arteriovenous segmentation result map and the vascular distribution map corresponding to the pulmonary blood vessels in the CT image to obtain a vascular comprehensive map, segmenting the vascular comprehensive map based on the connected region threshold, and obtaining a pulmonary artery vascular mask image GD.

[0052] Specifically, first, the image of the lung parenchyma can be segmented from the images of other irrelevant structures through the threshold segmentation method (the gray levels of the connected regions of the two are quite different, and they can be segmented by the gray level threshold. Specifically, the global threshold can be used for iterative segmentation. For example, the parameter T0 can be set, and an initial threshold T1 is selected. The image is segmented using the initial threshold T1 into a first image with a gray value greater than T1 and a second image with a gray value less than or equal to T1. Then, the average gray values U1 and U2 of the first image and the second image are calculated, as well as the new threshold T2 = (U1 + U2) / 2. If |T2 - T1| < T0, the new threshold T2 is the optimal threshold; otherwise, T2 is assigned to T1, and the image segmentation is performed again until the absolute value of the difference between the new threshold and the previous threshold is less than the set parameter). Then, inversion is performed to eliminate the backing plate (such as the bedplate) and holes (for example, after performing closing operation with a circular structural element with a radius of 6 and then taking the inverse). Then, based on the size of the pixels occupied by the connected regions, all the backing plates are completely removed. In this way, the mask of the torso can be obtained. Then, the new mask is obtained by filling the holes in this mask, and the two masks are subtracted to obtain the lung parenchyma mask. Then, by removing the lung parenchyma mask in the original CT image, the lung interstitial mask image can be obtained.

[0053] Second, by separating the bronchi in the CT image of the lungs, the bronchial mask image GZ can be directly obtained.

[0054] Third, the arterial and venous segmentation result maps and the vascular distribution map corresponding to the pulmonary vessels in the CT image are respectively obtained and fused to obtain the vascular synthesis map. Since the vascular synthesis map includes arterial vessels, venous vessels, and some vessels of unknown types, based on the connected region threshold, this vascular synthesis map can be segmented (specifically, for this unknown vessel, the connected region corresponding to it can be calculated. If the connected region of this unknown vessel is connected to the connected region corresponding to the arterial vessel, it is determined as an arterial vessel. In this way, the vascular synthesis map can be segmented based on the connected region threshold), and the pulmonary artery vessel mask image GD is obtained.

[0055] In an implementation manner of the present application, segmenting the bronchial mask image specifically includes: performing secondary segmentation on the obtained bronchial mask image GZ to respectively obtain the left and right main bronchus segment mask images GZ-1, the left and right secondary bronchus segment mask images GZ-2, and the left and right lobular bronchiole segment images GZ-3; then, according to the adjacent and docking relationships of each bronchial segment, the classification result of the bronchial mask image GZ is determined.

[0056] In one embodiment of the present application, segmenting the pulmonary artery vascular mask image specifically includes: performing secondary segmentation on the obtained pulmonary artery vascular mask image GD to obtain the left and right pulmonary artery main vessel segment mask images GD-1, the left and right pulmonary artery secondary vessel segment mask images GD-2, and the left and right pulmonary artery capillary segment mask images GD-3 respectively; and then determining the classification result of the pulmonary artery vascular mask image GD based on the adjacent and docking relationship of the pulmonary artery vessels.

[0057] In one embodiment of the present application, the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD are fused according to the accompanying relationship between the bronchi and pulmonary arteries corresponding to GZ-1 and GD-1, GZ-2 and GD-2, and GZ-3 and GD-3, respectively; and the final bronchial mask image is generated in the interstitial lung mask image GJ according to the fused image. In this way, the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD are quickly and accurately fused according to the accompanying relationship between the corresponding bronchi and pulmonary arteries of the three groups of images (GZ-1 and GD-1, GZ-2 and GD-2, and GZ-3 and GD-3), so that the final bronchial mask image can be quickly and accurately generated in the interstitial lung mask image GJ.

[0058] In one embodiment of the present application, in the process of fusing the classification results of the confirmed bronchial mask image GZ and the confirmed pulmonary artery mask image GD, if the connected area corresponding to a certain left or right lobular bronchiole segment is connected to the area of ​​the arterial vessel corresponding to a certain main vessel segment mask image in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is regarded as the classification result of the confirmed bronchial mask image GZ; if the connected area corresponding to a certain left or right lobular bronchiole segment is not connected to the area of ​​the arterial vessel corresponding to all the main vessel segment mask images in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is not regarded as the classification result of the confirmed bronchial mask image GZ; if the connected area corresponding to a certain left or right secondary bronchial segment is connected to the area of ​​the arterial vessel corresponding to a certain primary vessel segment mask image in the corresponding left or right pulmonary artery secondary vessel segment mask image GD-2 If the connected area corresponding to a certain left or right secondary bronchus segment is connected with the area of ​​the arterial blood vessels corresponding to all the secondary blood vessel segment mask images in the corresponding left or right pulmonary artery secondary blood vessel segment mask image GD-2, then the left or right secondary bronchus segment is not considered as the classification result of the confirmed bronchial mask image GZ; if the connected area corresponding to a certain left or right main bronchus segment is connected with the area of ​​the arterial blood vessels corresponding to a certain main blood vessel segment mask image in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is considered as the classification result of the confirmed bronchial mask image GZ; if the connected area corresponding to a certain left or right main bronchus segment is not connected with the area of ​​the arterial blood vessels corresponding to all the main blood vessel segment mask images in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is not considered as the classification result of the confirmed bronchial mask image GZ.

[0059] In this way, through this implementation, the accompanying relationship between the bronchi and the pulmonary artery vessels can be utilized to more quickly and accurately confirm the left and right main bronchial segment mask images, the left and right secondary bronchial segment mask images, and the left and right lobular bronchiolar segment images, respectively, so as to quickly and relatively accurately generate a bronchial mask image.

[0060] According to the second aspect of the present application, a bronchial extraction device based on CT images is also provided, characterized in that the bronchial extraction method based on CT images described in any technical solution in the first aspect of the present application is applied, and the device comprises:

[0061] Image acquisition module, used to obtain lung computer tomography CT images to be segmented

[0062] The first image processing module is used to segment the CT image to obtain a pulmonary interstitial mask image GJ, a bronchial mask image GZ and a pulmonary artery mask image GD;

[0063] The second image processing module is used to segment the bronchial mask image to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3, segment the pulmonary artery mask image to obtain at least three pulmonary artery segment masks GD-1, GD-2 and GD-3, determine the classification result of the bronchial mask image GZ according to the adjacent and docking relationship of each bronchial segment, and determine the classification result of the pulmonary artery mask image GD according to the adjacent and docking relationship of the pulmonary artery;

[0064] The third image processing module is used to fuse the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD, and generate a final bronchial mask image in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

[0065] According to the third aspect of the present application, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the bronchial extraction method based on CT images in any one of the technical solutions in the first aspect of the present application can be implemented.

[0066] It is understood that in this embodiment, the memory may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk or a solid state drive; in addition, the memory may also include a combination of the above-mentioned types of memory. This application does not specifically limit this.

[0067] Similarly, the processor may be a processor that implements or executes various exemplary logic steps described in conjunction with the disclosure of the present application. The processor may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. It may implement or execute various exemplary logic steps described in conjunction with the disclosure of the present application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0068] According to the fourth aspect of the present application, a computer-readable storage medium is also provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, it can implement the steps of the bronchial extraction method based on CT images in any technical solution in the first aspect of the present application.

[0069] In this embodiment, the computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may be located in an application-specific integrated circuit (ASIC). In the embodiments of the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0070] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A bronchial extraction method based on CT images, characterized in that: include: Acquire lung computer tomography (CT) images to be segmented; The image segmentation model is used to segment the CT images to obtain the pulmonary interstitial mask image GJ, the bronchial mask image GZ and the pulmonary artery mask image GD; The use of image segmentation models to segment CT images specifically includes: Obtaining the interstitial lung mask image: firstly, the threshold segmentation method is used to separate the lung parenchyma from other structures of the thoracic cavity, and then the connected left and right lungs are obtained based on the region growing method, the main trachea and the main bronchi extending into the lungs are removed, and then the holes in the parenchyma are filled to obtain the lung parenchyma image. The interstitial lung mask image GJ is obtained using the CT image and the lung parenchyma image; Obtaining a lung interstitial mask image GJ: performing bronchi segmentation on the lung bronchi in the CT image to obtain a bronchial mask image GZ; Obtaining a pulmonary artery vascular mask image GD: respectively obtaining and fusing the arteriovenous segmentation result map and the vascular distribution map corresponding to the pulmonary vessels in the CT image to obtain a vascular comprehensive map, segmenting the vascular comprehensive map based on the connected area threshold, and obtaining a pulmonary artery vascular mask image GD; Segmenting the bronchial mask image to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3, segmenting the pulmonary artery mask image to obtain at least three pulmonary artery segment mask images GD-1, GD-2 and GD-3, determining the classification result of the bronchial mask image GZ according to the adjacent and docking relationship of each bronchial segment, and determining the classification result of the pulmonary artery mask image GD according to the adjacent and docking relationship of the pulmonary artery; The segmentation of the bronchial mask image specifically includes: performing secondary segmentation on the obtained bronchial mask image GZ to obtain left and right main bronchial segment mask images GZ-1, left and right secondary bronchial segment mask images GZ-2, and left and right lobular bronchial segment images GZ-3 respectively; then determining the classification result of the bronchial mask image GZ according to the adjacent and docking relationship of each bronchial segment; Segmenting the pulmonary artery vascular mask image specifically includes: performing secondary segmentation on the obtained pulmonary artery vascular mask image GD to obtain left and right pulmonary artery main vessel segment mask images GD-1, left and right pulmonary artery secondary vessel segment mask images GD-2, and left and right pulmonary artery capillary vessel segment mask images GD-3 respectively; then determining the classification result of the pulmonary artery vascular mask image GD according to the adjacent and docking relationship of the pulmonary artery vessels; Then, the classification results of the confirmed bronchial mask image GZ and the confirmed pulmonary artery mask image GD are fused, and the final bronchial mask image is generated in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

2. The bronchial extraction method based on CT images according to claim 1, characterized in that: According to the accompanying relationship between the bronchi and pulmonary arteries corresponding to GZ-1 and GD-1, GZ-2 and GD-2, and GZ-3 and GD-3, respectively, the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD are fused; Based on the fused image, a final bronchial mask image is generated in the lung interstitial mask image GJ.

3. The bronchial extraction method based on CT images according to claim 1, characterized in that: In the process of fusing the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD, If the connected area corresponding to a certain left or right lobular bronchiole segment is connected to the area of ​​the arterial blood vessel corresponding to a certain main blood vessel segment mask image in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right lobular bronchiole segment is not connected to the area of ​​the arterial blood vessel corresponding to all the main blood vessel segment mask images in the corresponding left or right pulmonary artery capillary segment mask image GD-3, then the left or right lobular bronchiole segment is not regarded as the confirmed classification result of the bronchial mask image GZ; If the connected area corresponding to a certain left or right secondary bronchus segment is connected to the area of ​​the arterial blood vessel corresponding to a certain secondary blood vessel segment mask image in the corresponding left or right pulmonary artery secondary blood vessel segment mask image GD-2, then the left or right secondary bronchus segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right secondary bronchus segment is not connected to the area of ​​the arterial blood vessel corresponding to all the secondary blood vessel segment mask images in the corresponding left or right pulmonary artery secondary blood vessel segment mask image GD-2, then the left or right secondary bronchus segment is not regarded as the confirmed classification result of the bronchial mask image GZ; If the connected area corresponding to a certain left or right main bronchus segment is connected to the area of ​​the arterial blood vessel corresponding to a main blood vessel segment mask image in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is regarded as the confirmed classification result of the bronchial mask image GZ; if the connected area corresponding to a certain left or right main bronchus segment is not connected to the area of ​​the arterial blood vessel corresponding to all the main blood vessel segment mask images in the corresponding left or right pulmonary artery main blood vessel segment mask image GD-1, then the left or right main bronchus segment is not regarded as the confirmed classification result of the bronchial mask image GZ.

4. A bronchial extraction device based on CT images, characterized in that: The bronchial extraction method based on CT images applied to any one of claims 1 to 3, the device comprising: An image acquisition module, used for acquiring lung electronic computer tomography CT images to be segmented; The first image processing module is used to segment the CT image to obtain a pulmonary interstitial mask image GJ, a bronchial mask image GZ and a pulmonary artery mask image GD; The second image processing module is used to segment the bronchial mask image to obtain at least three bronchial segment mask images GZ-1, GZ-2 and GZ-3, segment the pulmonary artery mask image to obtain at least three pulmonary artery segment masks GD-1, GD-2 and GD-3, determine the classification result of the bronchial mask image GZ according to the adjacent and docking relationship of each bronchial segment, and determine the classification result of the pulmonary artery mask image GD according to the adjacent and docking relationship of the pulmonary artery; The third image processing module is used to fuse the classification results of the confirmed bronchial mask image GZ and the classification results of the confirmed pulmonary artery mask image GD, and generate a final bronchial mask image in the pulmonary interstitial mask image GJ according to the accompanying relationship of the pulmonary artery vessels corresponding to the bronchi.

5. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the bronchus extraction method based on CT images described in any one of claims 1 to 3 can be implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the bronchus extraction method based on CT images described in any one of claims 1 to 3 can be implemented.

Images