Image processing methods, apparatus, electronic devices and storage media

Abstract

This application discloses an image processing method, apparatus, electronic device, and storage medium. The method includes: acquiring a cardiac computed tomography angiography (CCTA) image; determining the intravascular lumen image of the coronary artery based on the CCTA image; performing edge extraction on the intravascular lumen image of the coronary artery to obtain an outer contour image of the intravascular lumen; determining a feature image of the coronary artery based on the CCTA image, the feature image indicating the outer contour of the coronary artery; processing the outer contour image and the feature image of the intravascular lumen based on a preset algorithm to obtain a vascular region image of the coronary artery; and performing a set operation on the vascular region image and the intravascular lumen image to obtain a vascular wall image of the coronary artery. The method effectively improves the efficiency of vascular wall extraction by employing a technique that combines vascular image enhancement processing with vascular region image extraction.

Description

Technical Field

[0001] This application belongs to the field of image processing technology, and in particular relates to an image processing method, apparatus, electronic device and storage medium. Background Technology

[0002] Cardiovascular disease is primarily caused by atherosclerotic lesions in the coronary arteries, leading to narrowing or blockage of the blood vessels and resulting in ischemic or hemorrhagic diseases of the heart. It has the highest mortality rate among all diseases. Current medical research demonstrates that the analysis of coronary atherosclerotic plaque burden and plaque radiomics characteristics is of significant value in assessing the risk of coronary heart disease. Since coronary atherosclerotic plaques reside within the vessel wall, coronary artery wall segmentation is fundamental for plaque extraction and analysis. Furthermore, current medical practice tends to base coronary heart disease diagnosis and risk assessment on cardiac computed tomography angiography (CCTA) images; therefore, coronary artery wall segmentation based on CCTA images has significant clinical implications.

[0003] With the development of deep learning technology, medical image segmentation based on deep learning has been widely used, and blood vessel wall segmentation is also frequently achieved using deep learning technology. However, the extraction of blood vessel walls using deep learning technology requires manually defining the initial contour of the blood vessel and manually annotating a large amount of blood vessel wall data, which is complex and time-consuming, resulting in low efficiency of blood vessel wall extraction. Summary of the Invention

[0004] This application provides an implementation scheme that differs from related technologies, in order to solve the technical problem that the extraction of blood vessel walls in related technologies requires manually defining the initial contour of the blood vessel and manually annotating a large amount of blood vessel wall data, which is complex and time-consuming, resulting in low efficiency in blood vessel wall extraction.

[0005] In a first aspect, this application provides an image processing method, comprising:

[0006] Acquiring cardiac computed tomography (CCTA) angiography images;

[0007] The intravascular images of the coronary arteries were determined based on the CCTA images;

[0008] Edge extraction is performed on the intravascular image of the coronary artery to obtain the outer contour image of the intravascular lumen of the coronary artery;

[0009] The characteristic image of the coronary artery is determined based on the CCTA image, and the characteristic image is used to indicate the outer contour of the coronary artery.

[0010] The outer contour image of the blood vessel lumen and the feature image are processed based on a preset algorithm to obtain a vascular region image of the coronary artery.

[0011] A set operation is performed on the image of the blood vessel region and the image of the blood vessel lumen to obtain the image of the blood vessel wall of the coronary artery.

[0012] Secondly, this application provides an image processing apparatus, comprising:

[0013] The acquisition module is used to acquire cardiac computed tomography (CCTA) angiography images;

[0014] The first determining module is used to determine the intravascular lumen image of the coronary artery based on the CCTA image;

[0015] The edge extraction module is used to extract the edges of the intravascular image of the coronary artery to obtain the outer contour image of the intravascular image of the coronary artery.

[0016] The second determining module is used to determine a feature image of the coronary artery based on the CCTA image, wherein the feature image is used to indicate the outer contour of the coronary artery.

[0017] The processing module is used to process the outer contour image of the blood vessel lumen and the feature image based on a preset algorithm to obtain a vascular region image of the coronary artery.

[0018] The set operation module is used to perform set operations on the vascular region image and the vascular lumen image to obtain the vascular wall image of the coronary artery.

[0019] Thirdly, this application provides an electronic device, comprising:

[0020] Processor; and

[0021] Memory for storing the executable instructions of the processor;

[0022] The processor is configured to execute the first aspect or any of the possible implementations of the first aspect by executing the executable instructions.

[0023] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the methods in the first aspect or any of the possible implementations of the first aspect.

[0024] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect or any of the possible implementations of the first aspect.

[0025] The solution provided in this application involves acquiring cardiac computed tomography angiography (CCTA) images; determining the intravascular lumen image of the coronary arteries based on the CCTA images; extracting edges from the intravascular lumen image of the coronary arteries to obtain the outer contour image of the intravascular lumen; determining feature images of the coronary arteries based on the CCTA images, the feature images indicating the outer contour of the coronary arteries; processing the outer contour image of the intravascular lumen and the feature images based on a preset algorithm to obtain a vascular region image of the coronary arteries; and performing a set operation on the vascular region image and the intravascular lumen image to obtain a vascular wall image of the coronary arteries. This method employs a fusion of vascular image enhancement processing and vascular region image extraction to automatically extract the coronary artery wall from the CCTA images. This solves the technical problem that vascular wall extraction requires manually defining the initial vascular contour and manually annotating a large amount of vascular wall data, which is complex and time-consuming, resulting in low efficiency. Therefore, this method effectively improves the efficiency of vascular wall extraction. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0027] Figure 1 A schematic flowchart of an image processing method provided for an exemplary embodiment of this application;

[0028] Figure 2a A schematic diagram of a cardiac region image provided for an exemplary embodiment of this application;

[0029] Figure 2b A schematic diagram of a coronary enhanced image of the cardiac region provided for an exemplary embodiment of this application;

[0030] Figure 2c A schematic diagram showing the location of the lumen of a coronary artery in a cardiac region image, provided as an exemplary embodiment of this application;

[0031] Figure 2d A schematic diagram of a CTA image provided for an exemplary embodiment of this application;

[0032] Figure 2e A schematic diagram of a gradient magnitude image corresponding to a CCTA image provided for an exemplary embodiment of this application;

[0033] Figure 2f A schematic diagram of a feature image of a coronary artery provided for an exemplary embodiment of this application;

[0034] Figure 2g A schematic diagram showing the location of a coronary artery region in a heart region image, provided as an exemplary embodiment of this application;

[0035] Figure 3 A schematic diagram of the structure of an image processing apparatus provided for an exemplary embodiment of this application;

[0036] Figure 4 This is a schematic block diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0037] The embodiments of this application are described in detail below, with examples of these embodiments illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0038] The terms "first" and "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the present application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0039] First, some terms used in the embodiments of this application will be explained below to facilitate understanding by those skilled in the art.

[0040] Coronary artery wall: refers to the area between the outer contour of the inner lumen (blood part) and the outer contour of the vessel. The lesion area is located in the vessel wall.

[0041] CCTA: Cardiac Computed Tomography Angiograph.

[0042] CTA: Computed Tomography Angiograph.

[0043] Voxel: short for volume pixel, is the smallest unit of digital data in three-dimensional space, conceptually similar to the smallest unit of two-dimensional space, the pixel.

[0044] CT stands for Computed Tomography, which uses precisely collimated X-ray beams, gamma rays, ultrasound waves, etc., along with highly sensitive detectors to scan a specific part of the human body one section after another. It features fast scanning time and clear images and can be used to examine a variety of diseases.

[0045] CT value: A unit of measurement used to determine the density of a local tissue or organ in the human body. It is usually called the Hounsfield unit (HU). Air is -1000 and dense bone is +1000.

[0046] The Hessian matrix, also known as the Hessian matrix, is a square matrix composed of the second-order partial derivatives of a multivariable function, describing the local curvature of the function. The Hessian matrix is ​​commonly used in Newton's method to solve optimization problems and can be used to determine the extrema of multivariable functions. In the optimization design of practical engineering problems, the objective function is often very complex. To simplify the problem, the objective function is often expanded into a Taylor polynomial in the neighborhood of a certain point to approximate the original function. In this case, the matrix form of the Taylor expansion of the function at that point will involve the Hessian matrix.

[0047] The confidence-connected region growing algorithm works as follows: First, the algorithm calculates the mean and standard deviation of the brightness values ​​of all pixels contained within the region. The user provides a factor to multiply by the standard deviation and define a range for the mean. Neighboring pixels whose brightness values ​​fall within this range are included in the region. The algorithm terminates its first iteration when no more pixels meet this standard. The mean and standard deviation of the brightness values ​​are recalculated using all pixels contained within the region. This mean and standard deviation define a new brightness range, used to examine the neighborhood of the current region and evaluate whether its brightness falls within this range. This iterative process is repeated until no new pixels are added or the maximum number of iterators has been reached.

[0048] Contrast agents, also known as contrast media, are chemical products injected (or ingested) into human tissues or organs to enhance the effect of image observation. These products have a higher or lower density than the surrounding tissues, creating a contrast used by certain instruments to display images. Examples include iodine preparations and barium sulfate commonly used in X-ray observation.

[0049] Multi-threshold segmentation: If there are multiple regions with different gray values ​​in an image, a series of thresholds can be selected to classify each pixel into the appropriate category. This method of segmenting with multiple thresholds is called multi-threshold segmentation.

[0050] Morphological operations: new images are generated one by one through convolution kernels. Factors affecting the final image include: kernel radius (size), kernel stride, weight values ​​in the kernel, and kernel method.

[0051] The main difference between dilation and erosion lies in the convolution kernel method;

[0052] Inflation: Similar to using the maximum value method;

[0053] Corrosion: Similar to using the minimum method.

[0054] Morphological closing operation: First dilation, then erosion (appearing to close two finely connected patches together). Closing can fill small lakes (i.e., holes) and mend small cracks, while the overall position and shape remain unchanged. Closing filters the image by filling the concave corners. Different structuring element sizes will result in different filtering effects. Different choices of structuring elements lead to different segmentation.

[0055] Otsu's method: also known as the Otsu algorithm, uses "maximum inter-class variance" as a standard. It uses the distribution information of the image histogram to calculate a threshold and divides the image into foreground and background based on whether the pixels exceed the threshold.

[0056] CTA (Computed Tomography Angiography) is a crucial part of clinical CT applications. Because of the poor natural contrast between blood vessels and their background soft tissues, conventional CT scans often fail to visualize blood vessels. During a CTA examination, a contrast agent is introduced to alter the image contrast between the blood vessels and the background tissues, thereby highlighting the vessels.

[0057] Finding the gradient of an image generally refers to operations on grayscale or color images. Digital images are discrete point value spectra, which can also be called finding the derivative of a two-dimensional discrete function.

[0058] The Sigmoid function is a common sigmoid function in biology, also known as an sigmoid growth curve. In information science, due to its monotonically increasing properties and the monotonically increasing properties of its inverse function, the Sigmoid function is often used as an activation function in neural networks, mapping variables to the range of 0 and 1.

[0059] Active contour models, also known as "Snakes," are an architecture for extracting object contours from potentially noisy images. The process begins by creating an initial curve in the image; its shape is flexible, but it must enclose the target object's contour. Next, an "energy equation" is established, including "internal energy" to regulate the curve's shape and "external energy" to regulate the curve's closeness to the target object's contour. During computation, minimizing the internal energy causes the curve to continuously contract inward and remain smooth; while minimizing the external energy causes the curve to continuously approach the target object's contour until they become identical.

[0060] Cardiovascular disease is primarily caused by atherosclerotic lesions in the coronary arteries, leading to narrowing or blockage of the blood vessels and resulting in ischemic or hemorrhagic diseases of the heart. It has the highest mortality rate among all diseases. Current medical research demonstrates that the analysis of coronary atherosclerotic plaque burden and plaque radiomics characteristics is of significant value in assessing the risk of coronary heart disease. Since coronary atherosclerotic plaques reside within the vessel wall, coronary artery wall segmentation is fundamental for plaque extraction and analysis. Furthermore, current medical practice tends to rely on a one-stop inference method based on cardiac computed tomography angiography (CCTA) images for the diagnosis and risk assessment of coronary heart disease; therefore, coronary artery wall segmentation based on CCTA images has significant clinical implications.

[0061] With the development of deep learning technology, medical image segmentation based on deep learning has been widely used, and blood vessel wall segmentation is also frequently achieved using deep learning technology. However, the extraction of blood vessel walls using deep learning technology requires manually defining the initial contour of the blood vessel and manually annotating a large amount of blood vessel wall data, which is complex and time-consuming, resulting in low efficiency of blood vessel wall extraction.

[0062] Therefore, this application provides an image processing method, apparatus, electronic device, and storage medium to solve the technical problem in the related art that the extraction of blood vessel walls requires manual definition of the initial contour of the blood vessel and manual annotation of a large amount of blood vessel wall data, which is complicated and time-consuming, resulting in low efficiency of blood vessel wall extraction.

[0063] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0064] Figure 1A flowchart illustrating an exemplary embodiment of this application provides an image processing method. The execution subject of this method can be a terminal device, such as a computer. The method includes at least the following steps S1-S5:

[0065] S1. Acquire cardiac computed tomography (CCTA) angiography images.

[0066] Optionally, the CCTA image may represent a 3D image obtained by post-processing (e.g., reconstruction) of CCTA cross-sectional (e.g., axial, sagittal, coronal, etc.) images acquired by a cardiac computed tomography angiography imaging device.

[0067] S2. Determine the intravascular image of the coronary artery based on the CCTA image.

[0068] Due to the contrast agent in CCTA images, the inner lumen of the coronary arteries has a high density and is more prominent. Therefore, determining the inner lumen of the coronary arteries is the foundation for extracting images of the coronary artery walls.

[0069] In some optional embodiments of this application, the CCTA image includes cardiac region information, wherein the cardiac region information is a portion of the image region in the CCTA image.

[0070] In some optional embodiments of this application, determining the intravascular lumen image of the coronary artery based on the CCTA image in step S2 includes steps S21-S25:

[0071] S21. Extract the heart region information from the CCTA image to obtain a heart region image.

[0072] CCTA images typically include a portion of the lung region. Since the lung region involves many blood vessels, such as pulmonary veins and pulmonary arteries, if subsequent vascular enhancement steps are performed directly on the CCTA images, the blood vessels in the lung region may interfere with the coronary artery enhancement process. Therefore, it is advisable to extract the cardiac region first.

[0073] In some optional embodiments of this application, in step S21, the cardiac region information is extracted from the CCTA image to obtain a cardiac region image, including steps S211-S216:

[0074] S211. The CCTA image is processed according to the multi-threshold segmentation algorithm to obtain a binary image of the lung region.

[0075] Compared to surrounding tissues, lung tissue has a lower density, meaning a lower CT value. The CT value is equal to the voxel value of the corresponding voxel in a CCTA image. Therefore, this low density of lung tissue can be utilized to obtain a binary image of the lung region using a multi-threshold segmentation method. This binary image of the lung region can then be used as the basis for extracting the heart region.

[0076] Optionally, the CCTA image is processed according to a multi-threshold segmentation algorithm to obtain a binary image of the lung region, including:

[0077] Obtain a target threshold range, wherein the target threshold range is a preset range of voxel values ​​representing the lung region image;

[0078] For each voxel in the CCTA image, if the voxel value is within the target threshold range, the voxel is identified as a voxel of the lung region; if the voxel value is not within the target threshold range, the voxel is identified as not a voxel of the lung region.

[0079] By iterating through multiple voxels in the CCTA image, the gray values ​​of voxels in the lung region are set to 1, and the gray values ​​of other voxels are set to 0, thus obtaining a binary image of the lung region.

[0080] Specifically, the target threshold range can be determined based on the density values ​​of lung tissue under normal conditions. For example, the target threshold range can be [-900, -350], in Hu.

[0081] S212. Based on morphological closure operation, the binary image of the lung region is filled with holes to obtain a complete binary image of the lung region.

[0082] Binary images of the lung region often contain numerous holes, which are typically pulmonary blood vessels and soft tissue. Since the grayscale values ​​of pulmonary blood vessels and soft tissue are much larger than those of the lung itself, the threshold segmentation of the CCTA image using the multi-threshold segmentation algorithm may segment pulmonary blood vessels and soft tissue into different threshold ranges. To fill these holes, morphological closing operations can be performed on the binary image of the lung region. In practice, a larger kernel radius can be chosen as needed to ensure that the holes within the lung region are filled, thus obtaining a complete binary image of the lung region.

[0083] S213. Based on the binary image of the complete lung region, determine the complete lung region image corresponding to the binary image of the complete lung region from the CCTA image.

[0084] In some optional embodiments of this application, in S213, based on the binary image of the complete lung region, the complete lung region image corresponding to the binary image of the complete lung region is determined from the CCTA image, including the following S01-S03:

[0085] S01. Based on the complete binary image of the lung region, determine multiple first position information corresponding to multiple voxels in the lung region; the first position information is the coordinates of the voxels in the first coordinate system corresponding to the complete binary image of the lung region;

[0086] S02. Based on the plurality of first location information, determine the plurality of second location information of a plurality of voxels in the lung region; the second location information is the coordinates of the voxels in the second coordinate system corresponding to the CCTA image;

[0087] The first location information and the second location information correspond one-to-one;

[0088] The coordinate transformation between the first and second coordinate systems is known.

[0089] The method for determining multiple second location information based on the multiple first location information can be found in related technologies, and will not be elaborated here.

[0090] S03. The image region composed of voxels at the multiple second location information points in the CCTA image is determined as the complete lung region image corresponding to the complete lung region binary image.

[0091] S214. Display the complete lung region image.

[0092] Optionally, in S214, the complete lung region image is displayed, including:

[0093] Based on the complete lung region image, determine the coordinates of the initial vertices of the circumscribed cuboid of the complete lung region image, and the length, width, and height of the circumscribed cuboid; display the circumscribed cuboid according to the coordinates of the initial vertices and the length, width, and height of the circumscribed cuboid.

[0094] S215. Obtain the user-defined location information for extracting the heart region image from the complete lung region image.

[0095] In some optional embodiments of this application, the user can set the cropping location information by triggering a click operation on a complete lung region image displayed on the interface, or within the circumscribed cuboid. Specifically, the user's click location is the cropping location information. The cropping location information may include multiple location information (i.e., multiple three-dimensional coordinates).

[0096] In some optional embodiments of this application, the user can also drag the eight vertices of the aforementioned circumscribed cuboid. The method further includes: determining the cutoff position information based on the position information of the eight vertices after the user drags them. Specifically, the position information of the eight vertices after dragging can be used as the cutoff position information.

[0097] Users can flexibly select the location information to capture based on the content displayed on the screen, and can observe the selected result in real time, namely the image of the heart region, thus improving the user experience.

[0098] S216. Extract the heart region image from the complete lung region image based on the extraction location information.

[0099] In step S216, extracting the heart region image from the complete lung region image based on the cropping location information includes:

[0100] The image region corresponding to the aforementioned cropping location information in the complete lung region image is used as the heart region map.

[0101] Optionally, the aforementioned cropping location information includes the coordinates of eight vertices, and the image region corresponding to the cropping location information is a cuboid image region enclosed by the cropping location information.

[0102] Figure 2a This is a schematic diagram of a cardiac region image provided for an exemplary embodiment of this application.

[0103] Specifically, the aforementioned circumscribed cuboid can be the circumscribed cuboid with the largest volume intersection-to-union ratio with the lung region. The axes of the three edges of the circumscribed cuboid intersecting at the same vertex are the X, Y, and Z axes, respectively. Users can flexibly set the above-mentioned cut-off position information based on experience. For example, in the X-axis direction, the heart region is relatively far from the sides, so in the X-axis direction, the two cut-off positions selected by the user on the same edge A are far from the endpoint of edge A. In the Z-axis direction, because the capillaries at the end of the coronary arteries extend relatively long, in the Z-axis direction, the two cut-off positions selected by the user on the same edge B are close to the endpoint of edge B, or even coincide.

[0104] S22. According to the recursive filtering algorithm, the heart region image is filtered to obtain a second image to be processed corresponding to the heart region image. The second image to be processed is used to highlight the vascular region of the coronary artery in the heart region.

[0105] Optionally, the heart region image is filtered according to a recursive filtering algorithm to obtain a second image to be processed corresponding to the heart region image, including:

[0106] The Hessian image of the heart region is obtained using the hessian_recursive_gaussian image filtering function, and the Hessian image is used as the second image to be processed corresponding to the heart region image.

[0107] In the process of solving the Hessian image of the heart region using the hessian_recursive_gaussian image filtering function, the value of the Sigma parameter will affect the prominence of the coronary artery region. Therefore, users can adjust the value of the Sigma parameter to obtain the desired prominence of the vessel region.

[0108] S23. The second image to be processed is processed according to the Frangi filter enhancement algorithm to obtain the third image to be processed.

[0109] Optionally, the Frangi filter enhancement algorithm is used to process the second image to be processed (Hessian image) and further enhance it. Figure 2b This is a schematic diagram of a coronary enhanced image of the cardiac region provided as an exemplary embodiment of this application. The aforementioned third image to be processed can be a coronary enhanced image of the cardiac region, such as... Figure 2b As shown, the dashed area represents the vascular region of the coronary arteries, which can be highlighted in the image after the above-mentioned further enhancement processing.

[0110] S24. According to the Otsu method, the third image to be processed is segmented to obtain a binary image of the tubular region of the coronary artery.

[0111] The principle of Otsu's method is to traverse the entire range of voxels in the image (from 0 to 255 for an 8-bit grayscale image), find a suitable threshold, and segment the original image into a foreground image and a background image with the largest class variance between them.

[0112] According to the Otsu method, the third image to be processed is segmented into a foreground image and a background image. The foreground image is a binary image of the tubular region of the coronary artery.

[0113] S25. Based on the region growing algorithm, process the binary image of the tubular region of the coronary artery to obtain the intravascular image of the coronary artery.

[0114] In some optional embodiments of this application, S25, processing the binary image of the tubular region of the coronary artery according to the region growing algorithm includes steps S251-S256:

[0115] S251. Obtain a first preset parameter, wherein the first preset parameter includes at least: a preset adjacent kernel radius, a preset scaling factor, and a preset number of iterations.

[0116] The first preset parameter can be a parameter input by the user, and the user can flexibly determine the first preset parameter based on experience.

[0117] S252. For each of the multiple vascular segment regions in the binary image of the tubular region, at least one seed point is selected as the seed point corresponding to the vascular segment region. The aforementioned multiple vascular segment regions can be pre-labeled by the user.

[0118] Specifically, the aforementioned multiple vascular segment regions are included in the left and right branch regions of the coronary artery in the binary image of the tubular region. The left branch region may include a portion of the multiple vascular segment regions, and the right branch region may include another portion of the multiple vascular segment regions.

[0119] Alternatively, since the two branches are different, the extraction of the vascular lumen can be performed branch by branch when performing regional growth extraction.

[0120] Optionally, the seed point can be a randomly selected voxel or a voxel that meets certain conditions, such as a voxel whose voxel value is greater than a preset voxel value threshold.

[0121] S253. Based on the preset adjacent kernel radius, determine the seed point region in the blood vessel segment region based on the seed point corresponding to the blood vessel segment region.

[0122] Each seed point corresponds one-to-one with a seed point region, and the seed point region in the vascular segment region contains the seed point regions corresponding to various sub-points in the vascular segment region.

[0123] Specifically, determining the seed point region corresponding to the seed point includes: taking the spherical region formed with the seed point as the center and the radius of the adjacent kernel as the sphere radius as the sphere radius, as the seed point region corresponding to the seed point.

[0124] S254. Based on the preset scaling factor and preset iteration number, perform region growth processing on the seed point region in the blood vessel segment region to obtain the grown region corresponding to the blood vessel segment region.

[0125] When there are multiple seed point regions, the seed point regions in the blood vessel segment region are subjected to region growth processing to obtain the grown region corresponding to the blood vessel segment region. This includes: performing region growth processing on the seed point regions for various sub-point regions in the blood vessel segment region to obtain the grown region corresponding to the seed point regions; and using the multiple grown regions corresponding to the multiple seed point regions as the grown region corresponding to the blood vessel segment region.

[0126] Optionally, based on the preset scaling factor and preset iteration number, the seed point region is subjected to region growing processing to obtain the grown region corresponding to the seed point region, including:

[0127] Determine the mean and standard deviation of each voxel value in the seed point region;

[0128] Based on the preset scaling factor and the average and standard deviation of each voxel value, the target voxel value range is determined.

[0129] Based on the target voxel value range and the preset number of iterations, the seed point region is subjected to region growth processing to obtain the grown region corresponding to the seed point region.

[0130] Specifically, the above target voxel value range can be expressed as:

[0131] I(X)∈[mf·σ,m+f·σ]

[0132] Where I(X) is the voxel value, X is the coordinate of the adjacent voxels of the seed point, m and σ are the mean and standard deviation of the voxel values ​​in the seed point region, respectively, and f is the scaling factor.

[0133] The region growth process for the seed point region includes multiple iterations. In each iteration, if the voxel values ​​of adjacent voxels of the seed point are within the target voxel value range, then that voxel is added to the seed point region corresponding to the seed point. If no more voxel values ​​are within the target voxel value range, then the iteration ends and the next iteration begins, gradually increasing the region. When a preset termination condition is met, the region growth terminates, and the current seed point region is taken as the grown region corresponding to the seed point region. The preset termination condition can be: the number of iterations is greater than the preset number of iterations.

[0134] S255. Generate a binary image of the blood vessel segment region corresponding to the blood vessel segment region based on the grown region corresponding to the blood vessel segment region, and obtain multiple binary images of multiple blood vessel segment regions corresponding to multiple blood vessel segment regions.

[0135] S256. Calculate the union of the binary images of the multiple vascular segment regions to obtain the intravascular image of the coronary artery.

[0136] In some optional embodiments of this application, S256, obtaining the intravascular image of the coronary artery by taking the union of the binary images of the plurality of vascular segment regions, includes:

[0137] The union of the binary images of the multiple vascular segment regions is used to obtain the binary image of the lumen of the coronary artery.

[0138] The binary image of the coronary artery lumen is multiplied with the image of the heart region to obtain the image of the coronary artery lumen.

[0139] Figure 2c This is a schematic diagram showing the location of the lumen of a coronary artery in a heart region image, as provided in an exemplary embodiment of this application. Figure 2c The area marked by the dashed line represents the lumen of the coronary artery.

[0140] In the aforementioned binary image of the tubular region of the coronary artery, the coronary artery is tubular, containing the lumen and plaque (such as calcification areas). Due to uneven contrast agent distribution and artifacts, the voxel values ​​in the lumen region may be uneven. By using the region growing algorithm, not only can plaque be removed from the binary image of the tubular region of the coronary artery, but the integrity of the lumen can also be maintained when the voxel values ​​in the lumen region are uneven.

[0141] S3. Extract the edges of the inner lumen image of the coronary artery to obtain the outer contour image of the inner lumen of the coronary artery.

[0142] Specifically, an active contour recognition model or other preset edge extraction algorithms can be used to extract the edges of the inner lumen image of the coronary artery to obtain the outer contour image of the inner lumen of the coronary artery.

[0143] S4. Determine a feature image of the coronary artery based on the CCTA image, the feature image being used to indicate the outer contour of the coronary artery.

[0144] In some optional embodiments of this application, in S4, determining the characteristic image of the coronary artery based on the CCTA image includes steps S41-S43:

[0145] S41. Perform gradient calculation based on the CCTA image to obtain the gradient magnitude image corresponding to the CCTA image.

[0146] The voxel values ​​of the boundaries of coronary arteries in CCTA images vary considerably. To highlight the boundaries of coronary arteries, gradient calculations can be performed on CCTA images to obtain the gradient amplitude images corresponding to the CCTA images.

[0147] In some optional embodiments of this application, in S41, gradient calculation is performed based on the CCTA image, including:

[0148] The CCTA image is preprocessed to obtain a preprocessed target image, wherein the preprocessing includes image magnification and / or image cropping;

[0149] Gradient calculation is performed on the target image to obtain the gradient magnitude image corresponding to the CCTA image.

[0150] Specifically, the target image mentioned above can be a CTA image. Figure 2d This is a schematic diagram of a CTA image provided for an exemplary embodiment of this application.

[0151] Figure 2e This is a schematic diagram of a gradient magnitude image corresponding to a CCTA image, provided as an exemplary embodiment of this application.

[0152] S42. Perform Gaussian filtering on the gradient magnitude image to obtain the first image to be processed after filtering.

[0153] In the process of Gaussian filtering the gradient amplitude image, the value of the parameter Sigma (standard deviation) in the Gaussian filtering function can affect the boundary thickness of the coronary artery in the gradient amplitude image. The larger the value of the parameter Sigma, the thicker the boundary of the coronary artery in the gradient amplitude image. Therefore, the value of the parameter Sigma can be flexibly adjusted according to actual needs to control the boundary thickness of the coronary artery in the gradient amplitude image.

[0154] S43. Adjust the contrast of the first image to be processed to obtain a feature image of the coronary artery.

[0155] In some optional embodiments of this application, in S43, adjusting the contrast of the first image to be processed includes:

[0156] According to the Sigmoid function, the voxel values ​​of each voxel in the first image to be processed are mapped to a preset range.

[0157] Specifically, the aforementioned preset interval can be [0,1], and the Sigmoid function can be expressed as:

[0158]

[0159] In this process, y also requires a linear transformation, i.e., y = αx + β, where α is the magnification factor, β is the transformation amplitude, and x is the voxel value of the original image. In this application, the gradient amplitude transformation is performed, therefore α is negative and β is the mean of the gradient values ​​at the blood vessel boundary. Users can adjust these values ​​flexibly according to their specific needs.

[0160] Solving the Sigmoid function yields an image where voxel values ​​are approximately 0 or 1. This maps the voxel values ​​of each voxel in the first image to be processed to a preset range, thereby adjusting the contrast of the first image and obtaining a feature image of the coronary arteries. The feature image can be found in [reference needed]. Figure 2f As shown.

[0161] S5. Based on a preset algorithm, the outer contour image of the blood vessel lumen and the feature image are processed to obtain a vascular region image of the coronary artery.

[0162] In some optional embodiments of this application, in step S5, the outer contour image of the blood vessel lumen and the feature image are processed based on a preset algorithm to obtain a vascular region image of the coronary artery, including steps S51-S52:

[0163] S51. Obtain the second preset parameters, which include: extension direction information and iteration termination condition information of the loop corresponding to the preset algorithm. The iteration termination condition includes: the number of loops is greater than the preset iteration number threshold, or the relevant loss information is less than the preset root mean square error threshold.

[0164] Specifically, the aforementioned direction of extension can be the direction from the outer contour of the lumen of the coronary artery to the boundary of the vascular region of the coronary artery.

[0165] S52. Based on the preset algorithm and the second preset parameters, process the outer contour image and the feature image to obtain a vascular region image of the coronary artery.

[0166] Optionally, the preset algorithm is a level set algorithm.

[0167] Optionally, the preset algorithm may also be other algorithms that process the outer contour image and the feature image based on the aforementioned second preset parameters to obtain a vascular region image of the coronary artery.

[0168] In some optional embodiments of this application, in S52, the outer contour image and the feature image are processed according to the level set algorithm and the second preset parameters to obtain a vascular region image of the coronary artery, including steps S521-S523:

[0169] S521. According to the second preset parameters, perform a level set operation on the outer contour image and the feature image to obtain a level set image.

[0170] S522. Set the voxel values ​​of voxels in the horizontal set image that are greater than or equal to the first threshold to 1, and set the voxel values ​​of voxels in the horizontal set image that are less than the first threshold to 0, to obtain the image to be merged.

[0171] S523. Calculate the product of the image to be merged and the CCTA image to obtain the vascular region image.

[0172] In some other optional embodiments of this application, in S52, the outer contour image and the feature image are processed according to the level set algorithm and the second preset parameters to obtain a vascular region image of the coronary artery, including:

[0173] Obtain the preset root mean square error threshold;

[0174] Based on the second preset parameters and the function corresponding to the preset algorithm, the outer contour image and the feature image are processed to obtain the vascular region image of the coronary artery.

[0175] Specifically, this can be achieved through ITK (Insight Segmentation and Registration Toolkit) software.

[0176] Optionally, the aforementioned extension direction information can be determined based on the default extension direction information corresponding to the function of the preset algorithm.

[0177] Optionally, users can pre-set: a preset iteration number threshold and a preset root mean square error threshold.

[0178] Specifically, the preset root mean square error threshold (0.01) can be set using the following function:

[0179] geodesicActiveContour.SetMaximumRMSError(0.01)

[0180] Specifically, the preset iteration count threshold (1500) can be set using the following function:

[0181] geodesicActiveContour.SetNumberOfIterations(1500);

[0182] Specifically, the following function can be executed:

[0183] levelset=geodesicActiveContour.Execute(initial_img,featureImage).

[0184] Here, initial_img is the outer contour image, featureImage is the feature image, levelset is the vascular region image of the coronary artery, and geodesicActiveContour.Execute() is the function used to execute the preset algorithm.

[0185] Figure 2g This is a schematic diagram showing the location of a coronary artery region in a heart region image, provided as an exemplary embodiment of this application. (See diagram below.) Figure 2g As shown, region 201 is the vascular region of the coronary artery.

[0186] S6. Perform a set operation on the image of the blood vessel region and the image of the blood vessel lumen to obtain the image of the blood vessel wall of the coronary artery.

[0187] In some optional embodiments of this application, in S6, a set operation is performed on the vascular region image and the vascular lumen image to obtain the vessel wall image of the coronary artery, including:

[0188] The set of voxels representing the vascular region in the vascular region image is taken as the first voxel set;

[0189] The voxels representing the vascular lumen region in the image of the vascular lumen region are used as the second voxel set;

[0190] The complement of the second voxel set in the first voxel set is determined as the third voxel set;

[0191] The image region composed of multiple voxels in the third voxel set in the vascular region image is used as the vascular wall image of the coronary artery.

[0192] The solution provided in this application involves acquiring cardiac computed tomography angiography (CCTA) images; determining the intravascular lumen image of the coronary arteries based on the CCTA images; extracting edges from the intravascular lumen image of the coronary arteries to obtain the outer contour image of the intravascular lumen; determining feature images of the coronary arteries based on the CCTA images, the feature images indicating the outer contour of the coronary arteries; processing the outer contour image of the intravascular lumen and the feature images based on a preset algorithm to obtain a vascular region image of the coronary arteries; and performing a set operation on the vascular region image and the intravascular lumen image to obtain a vascular wall image of the coronary arteries. By employing a technique that integrates vascular image enhancement processing and vascular region image extraction, the coronary artery wall is automatically extracted from CCTA images. This solves the technical problem that vascular wall extraction requires manually defining the initial vascular contour and manually annotating a large amount of vascular wall data, which is complex and time-consuming, resulting in low efficiency. Therefore, the solution effectively improves the efficiency of vascular wall extraction.

[0193] Figure 3 A schematic diagram of an image processing apparatus provided for an exemplary embodiment of this application; wherein the apparatus includes:

[0194] The acquisition module is used to acquire cardiac computed tomography (CCTA) angiography images;

[0195] The first determining module is used to determine the intravascular lumen image of the coronary artery based on the CCTA image;

[0196] The edge extraction module is used to extract the edges of the intravascular image of the coronary artery to obtain the outer contour image of the intravascular image of the coronary artery.

[0197] The second determining module is used to determine a feature image of the coronary artery based on the CCTA image, wherein the feature image is used to indicate the outer contour of the coronary artery.

[0198] The processing module is used to process the outer contour image of the blood vessel lumen and the feature image based on a preset algorithm to obtain a vascular region image of the coronary artery.

[0199] The set operation module is used to perform set operations on the vascular region image and the vascular lumen image to obtain the vascular wall image of the coronary artery.

[0200] In some optional embodiments of this application, the characteristic image of the coronary artery is determined based on the CCTA image. Specifically, when the second determining module is used to determine the characteristic image of the coronary artery based on the CCTA image, it is used for:

[0201] Gradient calculation is performed on the CCTA image to obtain the gradient magnitude image corresponding to the CCTA image;

[0202] The gradient magnitude image is subjected to Gaussian filtering to obtain the first image to be processed after filtering;

[0203] The contrast of the first image to be processed is adjusted to obtain a feature image of the coronary artery.

[0204] In some optional embodiments of this application, the CCTA image contains cardiac region information, and when the first determining module is used to determine the intravascular lumen image of the coronary artery based on the CCTA image, it is specifically used for:

[0205] The information about the heart region is extracted from the CCTA image to obtain a heart region image;

[0206] According to the recursive filtering algorithm, the heart region image is filtered to obtain a second image to be processed corresponding to the heart region image. The second image to be processed is used to highlight the vascular region of the coronary artery in the heart region.

[0207] The second image to be processed is processed according to the Frangi filter enhancement algorithm to obtain the third image to be processed;

[0208] According to the Otsu method, the third image to be processed is segmented to obtain a binary image of the tubular region of the coronary artery.

[0209] The binary image of the tubular region of the coronary artery is processed according to the region growing algorithm to obtain the intravascular image of the coronary artery.

[0210] In some optional embodiments of this application, the first determining module, when processing the binary image of the tubular region of the coronary artery according to a region growing algorithm, includes:

[0211] Obtain a first preset parameter, which includes at least: a preset adjacent kernel radius, a preset scaling factor, and a preset number of iterations;

[0212] For each blood vessel segment region in the multiple blood vessel segment regions in the binary image of the tubular region, at least one seed point is selected as the seed point corresponding to the blood vessel segment region.

[0213] Based on the preset adjacent kernel radius, the seed point region in the blood vessel segment region is determined based on the seed point corresponding to the blood vessel segment region.

[0214] Based on the preset scaling factor and preset number of iterations, the seed point region in the blood vessel segment region is subjected to region growth processing to obtain the grown region corresponding to the blood vessel segment region.

[0215] Based on the grown region, generate a binary image of the blood vessel segment region corresponding to the blood vessel segment region, and obtain multiple binary images of multiple blood vessel segment regions corresponding to multiple blood vessel segment regions.

[0216] The intraluminal image of the coronary artery is obtained by taking the union of the binary images of the multiple vascular segment regions.

[0217] In some optional embodiments of this application, when the second determining module is used to adjust the contrast of the first image to be processed, it is specifically used for:

[0218] According to the Sigmoid function, the voxel values ​​of each voxel in the first image to be processed are mapped to a preset range.

[0219] In some optional embodiments of this application, when the first determining module is used to extract the cardiac region information from the CCTA image to obtain a cardiac region image, it is specifically used for:

[0220] The CCTA image is processed using a multi-threshold segmentation algorithm to obtain a binary image of the lung region.

[0221] Based on morphological closure operation, the binary image of the lung region is filled with holes to obtain a complete binary image of the lung region;

[0222] Based on the binary image of the complete lung region, the complete lung region image corresponding to the binary image of the complete lung region is determined from the CCTA image;

[0223] The complete lung region is displayed;

[0224] Obtain the user-defined location information for extracting the heart region image from the complete lung region image;

[0225] The heart region image is extracted from the complete lung region image based on the cropping location information.

[0226] In some optional embodiments of this application, when the processing module processes the outer contour image of the blood vessel lumen and the feature image based on a preset algorithm to obtain a vascular region image of the coronary artery, it is specifically used for:

[0227] Obtain a second preset parameter, which includes: extension direction information and iteration termination condition information of the loop corresponding to the preset algorithm. The iteration termination condition includes: the number of loops is greater than a preset iteration number threshold, or the relevant loss information is less than a preset root mean square error threshold.

[0228] Based on the preset algorithm and the second preset parameters, the outer contour image and the feature image are processed to obtain a vascular region image of the coronary artery.

[0229] It should be understood that the device embodiments and method embodiments can correspond to each other, and similar descriptions can be referred to the method embodiments. To avoid repetition, they will not be repeated here. Specifically, the device can execute the above method embodiments, and the foregoing and other operations and / or functions of each module in the device correspond to the corresponding processes in the various methods in the above method embodiments, which will not be repeated here for the sake of brevity.

[0230] The apparatus of this application embodiment has been described above from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that this functional module can be implemented in hardware, in software instructions, or in a combination of hardware and software modules. Specifically, the steps of the method embodiments in this application can be completed by integrated logic circuits in the processor's hardware and / or by software instructions. The steps of the method disclosed in this application embodiment can be directly embodied as being executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. Optionally, the software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps in the above method embodiments.

[0231] Figure 4 This is a schematic block diagram of an electronic device provided in an embodiment of this application. The electronic device may include:

[0232] The system includes a memory 301 and a processor 302. The memory 301 stores computer programs and transfers the program code to the processor 302. In other words, the processor 302 can retrieve and run the computer programs from the memory 301 to implement the methods described in the embodiments of this application.

[0233] For example, the processor 302 can be used to execute the above-described method embodiments according to instructions in the computer program.

[0234] In some embodiments of this application, the processor 302 may include, but is not limited to:

[0235] General-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0236] In some embodiments of this application, the memory 301 includes, but is not limited to:

[0237] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0238] In some embodiments of this application, the computer program may be divided into one or more modules, which are stored in the memory 301 and executed by the processor 302 to perform the method provided in this application. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0239] like Figure 4 As shown, the electronic device may also include:

[0240] Transceiver 303, which can be connected to processor 302 or memory 301.

[0241] The processor 302 can control the transceiver 303 to communicate with other devices; specifically, it can send information or data to other devices or receive information or data sent by other devices. The transceiver 303 may include a transmitter and a receiver. The transceiver 303 may further include antennas, and the number of antennas may be one or more.

[0242] It should be understood that the various components in the electronic device are connected through a bus system, which includes a data bus, a power bus, a control bus, and a status signal bus.

[0243] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a computer, enables the computer to perform the methods of the above-described method embodiments. Alternatively, embodiments of this application also provide a computer program product containing instructions that, when executed by a computer, cause the computer to perform the methods of the above-described method embodiments.

[0244] When implemented using software, it can be implemented entirely or partially as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0245] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0246] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0247] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. For example, the functional modules in the various embodiments of this application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0248] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An image processing method, characterized in that, include: Acquiring cardiac computed tomography (CCTA) angiography images; The intravascular images of the coronary arteries were determined based on the CCTA images; Edge extraction is performed on the intravascular image of the coronary artery to obtain the outer contour image of the intravascular lumen of the coronary artery; The characteristic image of the coronary artery is determined based on the CCTA image, and the characteristic image is used to indicate the outer contour of the coronary artery. The outer contour image of the blood vessel lumen and the feature image are processed based on a preset algorithm to obtain a vascular region image of the coronary artery. A set operation is performed on the vascular region image and the vascular lumen image to obtain the vessel wall image of the coronary artery. Specifically, the set of voxels representing the vascular region in the vascular region image is designated as the first voxel set, the voxels representing the vascular lumen in the vascular lumen image are designated as the second voxel set, the complement of the second voxel set in the first voxel set is determined as the third voxel set, and the image region composed of multiple voxels in the third voxel set in the vascular region image is taken as the vessel wall image of the coronary artery.

2. The method according to claim 1, characterized in that, Determining the characteristic images of the coronary arteries based on the CCTA images includes: Gradient calculation is performed on the CCTA image to obtain the gradient magnitude image corresponding to the CCTA image; The gradient magnitude image is subjected to Gaussian filtering to obtain the first image to be processed after filtering; The contrast of the first image to be processed is adjusted to obtain a feature image of the coronary artery.

3. The method according to claim 1, characterized in that, The CCTA image contains information about the cardiac region. Determining the intravascular lumen image of the coronary arteries based on the CCTA image includes: The information about the heart region is extracted from the CCTA image to obtain a heart region image; According to the recursive filtering algorithm, the heart region image is filtered to obtain a second image to be processed corresponding to the heart region image. The second image to be processed is used to highlight the vascular region of the coronary artery in the heart region. The second image to be processed is processed according to the Frangi filter enhancement algorithm to obtain the third image to be processed; According to the Otsu method, the third image to be processed is segmented to obtain a binary image of the tubular region of the coronary artery. The binary image of the tubular region of the coronary artery is processed according to the region growing algorithm to obtain the intravascular image of the coronary artery.

4. The method according to claim 3, characterized in that, The binary image of the tubular region of the coronary artery is processed according to the region growing algorithm, including: Obtain a first preset parameter, which includes at least: a preset adjacent kernel radius, a preset scaling factor, and a preset number of iterations; For each blood vessel segment region in the multiple blood vessel segment regions in the binary image of the tubular region, at least one seed point is selected as the seed point corresponding to the blood vessel segment region. Based on the preset adjacent kernel radius, the seed point region in the blood vessel segment region is determined based on the seed point corresponding to the blood vessel segment region. Based on the preset scaling factor and preset number of iterations, the seed point region in the blood vessel segment region is subjected to region growth processing to obtain the grown region corresponding to the blood vessel segment region. Based on the grown region corresponding to the blood vessel segment region, a binary image of the blood vessel segment region corresponding to the blood vessel segment region is generated, resulting in multiple binary images of multiple blood vessel segment regions corresponding to multiple blood vessel segment regions. The intraluminal image of the coronary artery is obtained by taking the union of the binary images of the multiple vascular segment regions.

5. The method according to claim 2, characterized in that, Adjusting the contrast of the first image to be processed includes: According to the Sigmoid function, the voxel values ​​of each voxel in the first image to be processed are mapped to a preset range.

6. The method according to claim 3, characterized in that, The information about the heart region is extracted from the CCTA image to obtain a heart region image, including: The CCTA image is processed using a multi-threshold segmentation algorithm to obtain a binary image of the lung region. Based on morphological closure operation, the binary image of the lung region is filled with holes to obtain a complete binary image of the lung region; Based on the binary image of the complete lung region, the complete lung region image corresponding to the binary image of the complete lung region is determined from the CCTA image; The complete lung region image is displayed; Obtain the user-defined location information for extracting the heart region image from the complete lung region image; The heart region image is extracted from the complete lung region image based on the cropping location information.

7. The method according to claim 1, characterized in that, Based on a preset algorithm, the outer contour image of the blood vessel lumen and the feature image are processed to obtain a vascular region image of the coronary artery, including: Obtain a second preset parameter, which includes: extension direction information and iteration termination condition information of the loop corresponding to the preset algorithm. The iteration termination condition includes: the number of loops is greater than a preset iteration number threshold, or the relevant loss information is less than a preset root mean square error threshold. Based on the preset algorithm and the second preset parameters, the outer contour image and the feature image are processed to obtain a vascular region image of the coronary artery.

8. An image processing apparatus, characterized in that, include: The acquisition module is used to acquire cardiac computed tomography (CCTA) angiography images; The first determining module is used to determine the intravascular lumen image of the coronary artery based on the CCTA image; The edge extraction module is used to extract the edges of the intravascular image of the coronary artery to obtain the outer contour image of the intravascular image of the coronary artery. The second determining module is used to determine a feature image of the coronary artery based on the CCTA image, wherein the feature image is used to indicate the outer contour of the coronary artery. The processing module is used to process the outer contour image of the blood vessel lumen and the feature image based on a preset algorithm to obtain a vascular region image of the coronary artery. The set operation module is used to perform set operations on the vascular region image and the vascular lumen image to obtain the vascular wall image of the coronary artery. Specifically, the set of voxels representing the vascular region in the vascular region image is designated as the first voxel set, the voxels representing the vascular lumen in the vascular lumen image are designated as the second voxel set, the complement of the second voxel set in the first voxel set is determined as the third voxel set, and the image region composed of multiple voxels in the third voxel set in the vascular region image is taken as the vascular wall image of the coronary artery.

9. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1-7 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-7.

Images