Vascular image segmentation method, device, equipment, storage medium and program product

Abstract

The present invention provides a vascular image segmentation method, apparatus, device, storage medium, and program product, relating to the field of image processing technology. The method comprises: obtaining an original vascular image and its corresponding initial vascular segmentation image; extracting the vascular skeleton structure based on the vascular segmentation image and determining endpoints on the vascular skeleton structure; superimposing the original vascular image and the vascular segmentation image to obtain a dual-channel image; determining, for each endpoint, a local segmentation window corresponding to the endpoint, performing iterative vascular segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and updating the vascular segmentation image based on the iterative vascular segmentation results; and traversing all endpoints to obtain a final vascular segmentation image. The present invention improves segmentation accuracy while conserving computing resources.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a blood vessel image segmentation method, device, equipment, storage medium and program product. Background Art

[0002] In the field of medical image analysis, accurate vascular segmentation plays an indispensable role in the diagnosis, treatment planning, and surgical planning of a variety of diseases. For example, in the diagnosis of cardiovascular disease, clearly displaying the morphology, distribution, and degree of stenosis of blood vessels can help doctors accurately judge the condition and formulate appropriate treatment strategies. In tumor treatment, clarifying the distribution of blood vessels around the tumor helps determine the extent of surgical resection or assess the feasibility of interventional therapy.

[0003] Common vascular image segmentation methods mainly rely on deep learning models to perform global optimization of the original vascular images. That is, the basic model is used to perform preliminary segmentation of medical images to obtain the binary or semantic segmentation results of the blood vessels. Then, the initial segmentation results are input into the overall optimization model for multiple iterative optimizations to gradually improve the quality of vascular segmentation.

[0004] However, this holistic iterative optimization method, which performs iterative optimization across the entire image, cannot accurately locate unsegmented areas and is easily affected by background interference. This in turn neglects the local features of small blood vessels, making it difficult to identify vessel segments in low-contrast or small vessel areas. This can cause some vascular branches to remain incorrectly segmented, thereby reducing segmentation accuracy. Furthermore, this holistic iterative optimization method requires repeated iterative calculations for already correctly segmented areas, resulting in a waste of computing resources. Summary of the Invention

[0005] The embodiments of the present invention provide a blood vessel image segmentation method, apparatus, device and storage medium to solve the problems of low segmentation accuracy and waste of computing resources in the blood vessel image segmentation process.

[0006] In a first aspect, an embodiment of the present invention provides a blood vessel image segmentation method, comprising:

[0007] Obtaining an original blood vessel image and its corresponding initial blood vessel segmentation image;

[0008] Extracting a vascular skeleton structure based on the vascular segmentation image and determining endpoints on the vascular skeleton structure; the vascular skeleton structure includes a plurality of vascular centerlines, and the endpoints are endpoints of the vascular centerlines;

[0009] superimposing the original blood vessel image and the blood vessel segmentation image to obtain a dual-channel image;

[0010] For each endpoint, determining a local segmentation window corresponding to the endpoint, performing iterative vascular segmentation on a local image corresponding to the local segmentation window in the dual-channel image, and updating the vascular segmentation image according to the iterative vascular segmentation result;

[0011] Traverse all endpoints to obtain the final blood vessel segmentation image.

[0012] In a possible implementation, performing iterative blood vessel segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and updating the blood vessel segmentation image according to the iterative blood vessel segmentation result, includes:

[0013] Initialize the number of iterations;

[0014] Determining a local image corresponding to the local segmentation window in the dual-channel image, and inputting the local image into a blood vessel segmentation model to obtain a local blood vessel segmentation image;

[0015] determining whether an iteration termination condition is satisfied according to the local blood vessel segmentation image and the number of iterations;

[0016] If the iteration termination condition is not met, determining a target endpoint in the local blood vessel segmentation image; the target endpoint is an endpoint farthest from the blood vessel skeleton structure in the local image among all endpoints in the local blood vessel segmentation image;

[0017] updating the local segmentation window according to the target endpoint, updating the blood vessel segmentation image and the dual-channel image based on the local blood vessel segmentation image, and updating the current number of iterations;

[0018] Jump to the step of determining the local image corresponding to the local segmentation window in the dual-channel image until the iteration termination condition is met, determine the current local blood vessel segmentation image as the iterative blood vessel segmentation result, and update the blood vessel segmentation image according to the iterative blood vessel segmentation result.

[0019] In a possible implementation, the iteration termination condition includes:

[0020] During N consecutive iterations, the newly added blood vessel segmentation area in the local blood vessel segmentation image is smaller than a preset value; N is an integer greater than or equal to 1;

[0021] or

[0022] The current number of iterations has reached the maximum limit.

[0023] In a possible implementation, determining the target endpoint in the local blood vessel segmentation image includes:

[0024] extracting a blood vessel skeleton structure from the local blood vessel segmentation image, and determining endpoints in the blood vessel skeleton structure;

[0025] An original blood vessel skeleton structure in a local image corresponding to the local blood vessel segmentation image is determined, and an endpoint farthest from the original blood vessel skeleton structure among all endpoints of the blood vessel skeleton structure in the local blood vessel segmentation image is determined as a target endpoint.

[0026] In a possible implementation, determining, for each endpoint, a local segmentation window corresponding to the endpoint includes:

[0027] For each endpoint, extracting the actual blood vessel diameter at the endpoint in the blood vessel segmentation image;

[0028] respectively obtaining a preset standard window size and a preset standard blood vessel diameter, and determining a proportional relationship between the standard window size and the standard blood vessel diameter;

[0029] determining a local window size corresponding to the endpoint based on the proportional relationship and the actual blood vessel diameter;

[0030] The endpoint is used as the center of the local segmentation window, and the local segmentation window is determined according to the local window size.

[0031] In a possible implementation, extracting a vascular skeleton structure based on the vascular segmentation image and determining endpoints on the vascular skeleton structure includes:

[0032] For each foreground pixel in the blood vessel segmentation image, detecting whether there is a background pixel in the neighborhood of the foreground pixel; the foreground pixel is a pixel used to represent the blood vessel, and the background pixel is the remaining pixel except the foreground pixel;

[0033] If there is a background pixel in the neighborhood of the foreground pixel, the foreground pixel is determined as a candidate pixel;

[0034] Detect whether the topological structure of the blood vessel segmentation image changes before and after each candidate pixel is deleted;

[0035] Determine the candidate pixels whose topological structures in the blood vessel segmentation image before and after deletion have not changed as target candidate pixels, and delete them to obtain a new blood vessel segmentation image;

[0036] Jumping to the step of detecting whether there is a background pixel in the neighborhood of each foreground pixel in the blood vessel segmentation image until the target candidate pixel does not exist in the current blood vessel segmentation image, thereby obtaining the blood vessel skeleton structure;

[0037] For each pixel point on the blood vessel skeleton structure, if the number of foreground pixels in the neighborhood of the pixel point is 1, the pixel point is determined to be an endpoint.

[0038] In a second aspect, an embodiment of the present invention provides a blood vessel image segmentation device, comprising:

[0039] An acquisition module, used for acquiring an original blood vessel image and its corresponding initial blood vessel segmentation image;

[0040] an extraction module, configured to extract a vascular skeleton structure based on the vascular segmentation image and determine endpoints on the vascular skeleton structure; the vascular skeleton structure includes a plurality of vascular centerlines, and the endpoints are endpoints of the vascular centerlines;

[0041] Segmentation module for:

[0042] superimposing the original blood vessel image and the blood vessel segmentation image to obtain a dual-channel image;

[0043] For each endpoint, determining a local segmentation window corresponding to the endpoint, performing iterative vascular segmentation on a local image corresponding to the local segmentation window in the dual-channel image, and updating the vascular segmentation image according to the iterative vascular segmentation result;

[0044] Traverse all endpoints to obtain the final blood vessel segmentation image.

[0045] In a third aspect, an embodiment of the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method in the first aspect or any possible implementation of the first aspect is implemented.

[0046] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method in the first aspect or any possible implementation of the first aspect.

[0047] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements the method in the first aspect or any possible implementation of the first aspect.

[0048] In an embodiment of the present invention, endpoints can serve as critical points between segmented and unsegmented areas in a vascular segmentation image. By extracting the vascular skeleton structure from the vascular segmentation image and then determining the endpoints in the vascular skeleton structure, the unsegmented areas in the vascular segmentation image can be accurately located to reduce background interference in the vascular segmentation image. On this basis, by determining the local image corresponding to the local segmentation window corresponding to each endpoint in the dual-channel image, iterative vascular segmentation can be performed. Local iterative segmentation can be performed on the unsegmented areas corresponding to each endpoint, thereby improving the segmentation accuracy of small blood vessels and thus improving the segmentation precision, while also saving computing resources. In addition, when performing local vascular segmentation in an embodiment of the present invention by superimposing a dual-channel image of the original vascular image and the vascular segmentation image, the segmented areas in the vascular segmentation image can guide the current local vascular segmentation work, thereby improving the robustness of vascular segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flow chart for implementing the blood vessel image segmentation method provided by an embodiment of the present invention;

[0050] Figure 2 is a schematic diagram of a blood vessel segmentation image provided by an embodiment of the present invention;

[0051] Figure 3 is a schematic diagram of a blood vessel skeleton structure provided by an embodiment of the present invention;

[0052] Figure 4 This is a flowchart for implementing iterative blood vessel segmentation on a local image provided by an embodiment of the present invention;

[0053] Figure 5 is a schematic diagram of determining a target endpoint provided by an embodiment of the present invention;

[0054] Figure 6 is a structural diagram of a blood vessel image segmentation device provided by an embodiment of the present invention;

[0055] Figure 7 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] Existing vascular segmentation techniques primarily rely on deep learning models for global optimization. This involves using a basic model to perform preliminary segmentation on vascular images to obtain an initial vascular segmentation result. This vascular segmentation result is then fed into an overall optimization model for multiple iterative optimizations to gradually refine the initial vascular segmentation result and obtain the final vascular segmentation result. However, this method cannot accurately locate the unsegmented areas within the initial vascular segmentation result and can only perform overall iterative optimization across the entire image. This approach is susceptible to interference from the image background and can ignore the local features of small blood vessels, leading to reduced vascular segmentation accuracy. Furthermore, during the overall iterative optimization process for the entire image, even if certain areas have been correctly segmented, the overall optimization model will still perform repeated calculations on those areas, wasting computing resources.

[0058] To improve segmentation accuracy and conserve computing resources, embodiments of the present invention extract endpoints from a segmented vessel image to precisely locate unsegmented regions within the image. Furthermore, by iteratively segmenting the local image corresponding to each endpoint, each segmented region can be precisely segmented, thereby improving the accuracy and precision of small vessel segmentation. Furthermore, this eliminates the need to resegment already segmented regions, thus conserving computing resources.

[0059] See also Figure 1 , which shows a flow chart of the implementation of the blood vessel image segmentation method provided by an embodiment of the present invention, and is described in detail as follows:

[0060] Step 101, obtaining an original blood vessel image and its corresponding initial blood vessel segmentation image;

[0061] In the embodiment of the present invention, an original blood vessel image can be obtained by using angiography medical imaging technology. The original blood vessel image is usually a single-channel grayscale image. By inputting the original blood vessel image into an image segmentation model such as U-Net or DeepLabV3+, an initial blood vessel segmentation image can be obtained. For example, see Figure 2 The initial blood vessel segmentation image is usually a binary image, in which the grayscale value of the pixel points in the blood vessel area is usually 1, and the grayscale value of the pixel points in the remaining area is usually 0.

[0062] To ensure the accuracy of subsequent image processing, the embodiment of the present invention can pre-process the original blood vessel image and the blood vessel segmentation image respectively.

[0063] For the original vascular image, normalization processing can be performed on the original vascular image to eliminate the changes in pixel intensity distribution caused by different imaging devices, imaging parameters or individual physiological differences. The embodiment of the present invention adopts the Min-Max Normalization method to perform normalization processing. The specific calculation formula is as follows:

[0064]

[0065] in, Represents the normalized pixel value, whose value range is constrained to be between [0,1]. I Represents the original pixel value, and Represent the minimum and maximum pixel values in the original vascular image, respectively. This normalization process can improve data consistency, enhance the convergence of the model, and avoid the problem of unstable gradient updates caused by pixel value ranges that are too large or too small.

[0066] To further enhance blood vessel edge information and improve the contrast between blood vessels and the background, embodiments of the present invention employ, on top of normalization, a contrast-limited adaptive histogram equalization (CLAHE) method to optimize the image's grayscale distribution in a locally adaptive manner. The CLAHE method divides the image into multiple small blocks (typically an 8×8 or 16×16 grid) and performs histogram equalization within each block, while also applying a contrast limit to prevent excessive contrast enhancement in local areas, which can lead to noise amplification. Its mathematical expression is as follows:

[0067]

[0068] in, Indicates that the pixel value is The pixel point, the pixel value after equalization, i Indicates from 0 to The gray level index between Indicates grayscale i The frequency of occurrence in the current block, Indicates the total number of pixels in the current tile.

[0069] Through CLAHE processing, the separability of vascular structures in the input data is enhanced, which helps to improve the accuracy of subsequent vascular segmentation.

[0070] For the initial segmented blood vessel image, noise or local low contrast may cause discontinuity or missing small areas of blood vessel structure. Therefore, the embodiment of the present invention uses morphological closing operation to post-process the segmented blood vessel image. The closing operation consists of dilation and erosion operations, and its mathematical definition is as follows:

[0071]

[0072] in, A represents a binary image, i.e., a blood vessel segmentation image, B Represents a structural element, represents the expansion operation, Represents an erosion operation.

[0073] This process can fill small holes in the blood vessel segments and enhance the connectivity of the vascular structure.

[0074] In addition, in order to reduce the boundary noise in the blood vessel segmentation image and improve the smoothness of the blood vessel contour, the embodiment of the present invention also uses Gaussian filtering to smooth the blood vessel segmentation image. The kernel function of Gaussian filtering is defined as follows:

[0075]

[0076] in, The coordinates are Gaussian filtering result of the pixel at , represents the standard deviation of the Gaussian kernel.

[0077] Gaussian filtering can effectively reduce high-frequency noise by performing weighted averaging on pixel values while maintaining the overall shape of the blood vessel boundary, thereby improving segmentation accuracy.

[0078] Step 102: extract the vascular skeleton structure based on the vascular segmentation image and determine the endpoints on the vascular skeleton structure. The vascular skeleton structure includes multiple vascular centerlines, and the endpoints are the endpoints of the vascular centerlines.

[0079] The purpose of extracting the vascular skeleton structure is to convert the vascular region in the segmented image into a thinned structure consisting only of the vascular centerline, facilitating subsequent vascular morphology analysis, branch tracking, and diameter measurement. This embodiment of the present invention uses a morphological thinning algorithm to skeletonize the segmented image, removing redundant pixels and retaining only the vascular centerline structure.

[0080] In general, the process of extracting the vascular skeleton structure is carried out in an iterative manner. Each iteration scans the pixels in the image and deletes the boundary pixels that meet the requirements according to the preset removal conditions until the skeleton is stable.

[0081] Specifically, the main steps of the morphological thinning algorithm include: for each foreground pixel in the vessel segmentation image, checking whether there are background pixels in the neighborhood of the foreground pixel; if there are background pixels in the neighborhood of the foreground pixel, then the foreground pixel is determined as a candidate pixel. Then, each candidate pixel is checked for changes in the topological structure of the vessel segmentation image before and after deletion; candidate pixels whose topological structure has not changed before and after deletion are determined as target candidate pixels and deleted to obtain a new vessel segmentation image;

[0082] Jump to the step of detecting whether there are background pixels in the neighborhood of each foreground pixel in the blood vessel segmentation image until there are no target candidate pixels in the current blood vessel segmentation image, and obtain the blood vessel skeleton structure.

[0083] Here, the foreground pixels refer to the pixels used to represent blood vessels, and the background pixels refer to the remaining pixels except the foreground pixels.

[0084] For each foreground pixel, if there is at least one background pixel within its eight-neighborhood region, the foreground pixel is marked as a candidate pixel. Connectivity analysis is performed on the candidate pixels to check whether their deletion will not disrupt the vascular structure, that is, whether the topology of the original image remains unchanged after deletion. If the connectivity constraint is met (i.e., the topology remains unchanged after deletion), the pixel is deleted and the next iteration begins. When there are no more deletable pixels in the image, the iteration ends and the final skeletonization result is output.

[0085] Mathematically, the above process of extracting the vascular skeleton structure can be expressed as:

[0086]

[0087] in, Indicates the n The vascular skeleton structure extracted after +1 iteration, Indicates the n The vascular skeleton structure extracted after iterations, Indicates the n The set of pixel points that meet the connectivity constraints in the vascular skeleton structure after iterations.

[0088] For example, Figure 2 After extracting the vascular skeleton structure from the vascular segmentation image, we can get the following Figure 3 The above method can effectively remove redundant pixels in the vascular area while ensuring that the topological structure of the blood vessels remains unchanged.

[0089] Based on the extraction of the blood vessel skeleton structure, the embodiment of the present invention further performs endpoint detection on the blood vessel skeleton structure.

[0090] according to Figure 3 As can be seen, the vascular skeleton structure contains multiple vascular centerlines. The pixels at the endpoints of each vascular centerline are referred to as endpoints in this embodiment of the present invention. Endpoint detection can identify broken vascular regions and accurately locate unsegmented areas within the segmented vascular image. In this embodiment of the present invention, endpoints are defined as those pixels on a vascular centerline that are connected to only one other pixel.

[0091] Here, endpoint detection can be performed based on an eight-neighborhood pixel counting method. Specifically, for each pixel on the vascular skeleton structure, if the number of foreground pixels in its neighborhood is 1, the pixel is determined to be an endpoint. Foreground pixels are pixels that represent blood vessels.

[0092] That is, for each pixel on each vascular skeleton structure, the number of foreground pixels in its eight-neighborhood region is calculated. If the number is equal to 1, the pixel is determined to be an endpoint. The mathematical expression is as follows:

[0093]

[0094] in, Represents pixel points p The number of foreground pixels in the eight-neighborhood of Represents pixel points p The first in the eight neighborhoods of The pixel value of the pixel point, here, for the foreground pixel point, , for background pixels, .

[0095] Step 103, superimposing the original blood vessel image and the blood vessel segmentation image to obtain a dual-channel image;

[0096] The original vascular image and the segmented vascular image can be superimposed and fused by channel superposition to obtain a dual-channel image. According to the above, the original vascular image is a single-channel grayscale image, and the segmented vascular image is a binary image, that is, it is also a single-channel grayscale image.

[0097] This embodiment of the present invention aligns the original vessel image and the segmented vessel image at the same spatial scale and splices them in the channel dimension to form a dual-channel image. Specifically, the original vessel image and the segmented vessel image are resized to the same size, and the pixels at corresponding locations are channel-superimposed.

[0098] It can be understood that the dual-channel image has the same image size as the original blood vessel image and the blood vessel segmentation image, and each pixel in the dual-channel image corresponds to the pixel value of two channels.

[0099] During the channel superposition process, the geometric consistency between the original vascular grayscale distribution and the segmentation result is maintained, so as to enhance the attention of the subsequent vascular segmentation model to the image edge information.

[0100] Step 104 : for each endpoint, determine the local segmentation window corresponding to the endpoint, perform iterative blood vessel segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and update the blood vessel segmentation image according to the iterative blood vessel segmentation result.

[0101] The local segmentation window corresponding to the endpoint is actually the local area where the endpoint is located, that is, the fracture area of the current blood vessel segment. The local image corresponding to the local segmentation window in the dual-channel image is the unsegmented local area in the dual-channel image.

[0102] The embodiment of the present invention performs iterative vascular segmentation on an unsegmented local area, thereby further improving and expanding the vascular segmentation result in the local area based on the initial vascular segmentation image. After the iterative segmentation of the local area is completed, the overall vascular segmentation image is updated according to the iterative vascular segmentation result of the local area.

[0103] To ensure that blood vessels of different scales maintain the same proportional relationship during iterative blood vessel segmentation, an embodiment of the present invention adopts a method based on adaptive window adjustment to dynamically adjust the size of the local segmentation window to adapt to blood vessel structures of different diameters.

[0104] In general, the local segmentation window is centered on the endpoint, and the window size is determined by the vessel diameter in the overall vessel segmentation image. The size of the local segmentation window is adjusted by determining the vessel diameter at the endpoint.

[0105] Specifically, when determining the local segmentation window corresponding to each endpoint, for each endpoint, the actual blood vessel diameter at the endpoint position is extracted in the blood vessel segmentation image; then, the preset standard window size and the preset standard blood vessel diameter are obtained respectively, and the proportional relationship between the standard window size and the standard blood vessel diameter is determined; then, based on the proportional relationship and the actual blood vessel diameter, the local window size corresponding to the endpoint is determined; finally, the endpoint is used as the center of the local segmentation window, and the local segmentation window is determined according to the local window size.

[0106] In the vascular segmentation image, for each endpoint of the vascular skeleton structure, the width of continuous foreground pixels (i.e., the segmented vascular area) is determined along a cross-section perpendicular to the vascular direction with the endpoint as the center to obtain the actual vascular diameter at the endpoint.

[0107] According to the actual blood vessel diameter, the local window size of the local segmentation window is adjusted to ensure that the proportion of the blood vessel in the standard window size is consistent. The specific adjustment formula is as follows:

[0108]

[0109] in, Indicates the local window size, Indicates the preset standard window size, Indicates the preset standard blood vessel diameter, Indicates the actual blood vessel diameter.

[0110] Here, the standard window size and standard blood vessel diameter can be determined according to actual conditions and are not specifically limited in the embodiment of the present invention. For example, the standard window can be a square, 128×128 pixels, the standard window size is 128 pixels, and the standard blood vessel diameter can be 10 pixels.

[0111] The adjusted local window size can ensure that blood vessels of different scales maintain the same morphological features in the standard window, thereby improving the adaptability of subsequent blood vessel segmentation models at different blood vessel scales.

[0112] The embodiment of the present invention determines a local segmentation window based on the size of the local window with the endpoint as the center, and determines the local image corresponding to the local segmentation window in the dual-channel image based on cropping.

[0113] After determining the local image, further normalization can be performed. Specifically, for endpoints near the image edge, the image is expanded using mirror padding or zero padding to ensure the integrity of the extraction window. Bicubic interpolation is used to scale the local image to a standard window size of 128×128 pixels. The pixel values of the scaled image are normalized to a value between [0, 1] to improve the stability of vessel segmentation.

[0114] Traditional methods use global optimization, which tends to ignore the local features of small blood vessels, resulting in blurred blood vessel edges and missing branches. The present invention improves window extraction by adaptively adjusting the local segmentation window, focusing on local segmentation and improving the segmentation accuracy of small blood vessels.

[0115] Step 105: traverse all endpoints to obtain the final blood vessel segmentation image.

[0116] After completing the local iterative segmentation of an endpoint, the overall segmented vessel image is updated based on the iterative vessel segmentation results of that endpoint, and then the dual-channel image is updated. Based on the update, iterative vessel segmentation is performed on the next endpoint until all endpoints are traversed to obtain the final segmented vessel image.

[0117] Compared to the prior art, the embodiments of the present invention can accurately locate the unsegmented areas in the vascular segmentation image by extracting the vascular skeleton structure from the vascular segmentation image and then determining the endpoints in the vascular skeleton structure, thereby reducing background interference in the vascular segmentation image. The endpoints can serve as critical points between the segmented and unsegmented areas in the vascular segmentation image and are used to locate the unsegmented areas in the vascular segmentation image. On this basis, by determining the local image corresponding to the local segmentation window corresponding to each endpoint in the dual-channel image, iterative vascular segmentation is performed. Local iterative segmentation can be performed on the unsegmented areas corresponding to each endpoint, thereby improving the segmentation accuracy of small blood vessels and thus improving segmentation precision, while also saving computing resources. Furthermore, when the embodiments of the present invention perform local vascular segmentation on a dual-channel image that superimposes the original vascular image and the vascular segmentation image, the segmented areas in the vascular segmentation image can guide the current local vascular segmentation work, thereby improving the robustness of vascular segmentation.

[0118] In order to supplement and improve the incompletely segmented vascular regions, improve the integrity and accuracy of the segmentation results, and at the same time, ensure the connectivity and integrity of the vascular structure, the embodiment of the present invention determines a local segmentation window based on the endpoints and adopts a gradual expansion strategy based on the local segmentation window to supplement the incompletely segmented vascular regions.

[0119] Here, endpoint detection is used to determine the current segmented vessel's broken region. A local image is extracted centered on the endpoint, and the vessel segmentation model is used to finely segment the local image. Based on this fine segmentation, the vessel segmentation results within the local image are updated, and new endpoints are extracted based on the vessel segmentation results within the new local image. The iteration continues until the termination condition is met.

[0120] In the embodiment of the present invention, the core idea of iterative expansion is: in each round of iteration, the skeleton structure of the currently segmented blood vessel is used to guide the local segmentation, and the endpoints are continuously updated until the blood vessel structure is complete.

[0121] Optional, see Figure 4 ,The specific steps of iterative blood vessel segmentation are as follows:

[0122] Step 401: Initialize the number of iterations.

[0123] During the initialization phase, the number of iterations can be set to zero.

[0124] Step 402 : Determine a local image corresponding to the local segmentation window in the dual-channel image, and input the local image into a blood vessel segmentation model to obtain a local blood vessel segmentation image.

[0125] The blood vessel segmentation model is used to perform blood vessel segmentation on an input local image and output a blood vessel segmentation result corresponding to the local image, that is, a local blood vessel segmentation image.

[0126] The blood vessel segmentation model here can be any deep learning segmentation model, such as U-Net, AttentionU-Net, DeepLabV3+, TransUNet, and a segmentation network based on the Transformer structure. The blood vessel segmentation model in the embodiment of the present invention is based on the U-Net structure, and the specific structure is as follows:

[0127] The input to the vascular segmentation model consists of two channels. The first channel receives a local image corresponding to the original vascular image. The original vascular image provides grayscale or color structural information of the vessels, enabling the model to learn the vascular morphology and boundary characteristics in the original image. The second channel receives a local image corresponding to the segmented vascular image. The segmented vascular image contains information about the segmented vascular regions, guiding the model to focus on unsegmented areas and suppressing background noise, thereby improving segmentation accuracy.

[0128] The vascular segmentation model uses a dual-channel input strategy, comprehensively utilizing the original vascular image and the initial vascular segmentation image to enhance the model's ability to recognize small vascular structures and reduce background interference.

[0129] The model adopts an encoder-decoder framework and combines it with an attention mechanism to enhance the ability to focus on blood vessel edges. Its main components include:

[0130] Encoder: Consisting of multiple convolutional and pooling layers, the encoder extracts multi-scale features layer by layer. Each convolutional layer is followed by batch normalization to stabilize gradients, and the ReLU (Rectified Linear Unit) activation function is used to enhance nonlinear representation. Appropriate stride and padding strategies are employed to ensure that the feature map size is gradually reduced while preserving key vascular characteristics.

[0131] Skip Connection: Skip connections are introduced between the encoder and decoder to transfer low-level features from the encoding stage directly to the decoding stage, preserving fine-grained edge information and preventing blurred blood vessel boundaries. A weighted fusion strategy is used to adaptively fuse features at different scales to optimize segmentation results.

[0132] Decoder: Upsampling is performed using deconvolution or bilinear interpolation to gradually restore the image's original resolution. During the decoding process, attention mechanisms (such as SE-Net, CBAM, and Transformer-based Attention) are introduced to enhance the ability to focus on microvascular structures and improve segmentation accuracy.

[0133] Output Layer: A 1×1 convolutional layer is used as the final output layer to map the high-dimensional features extracted in the previous layer to a single-channel segmentation probability map. This 1×1 convolution compresses multi-channel features into a single channel through a linear transformation while preserving spatial information, ensuring that the output pixel-level prediction has accurate segmentation boundaries. A sigmoid activation function is used to normalize the output of the 1×1 convolution, mapping pixel values to the [0, 1] range. This output represents the probability that each pixel belongs to a blood vessel region. The specific calculation formula is as follows:

[0134]

[0135] in, Represents a pixel ( x , y ), Represents the original output value of the 1×1 convolution.

[0136] The output layer also uses a fixed threshold T=0.5 to binarize the probability map to generate the final segmentation mask (Binary Segmentation Mask). The specific calculation method is as follows:

[0137]

[0138] in, Indicates the final segmentation result. When the value is 1, it means that the pixel belongs to the blood vessel area, and when the value is 0, it means that the pixel belongs to the background area.

[0139] It can be understood that the local blood vessel segmentation image output by the model is equivalent to further expanding the blood vessel area based on the local image input by the model to further improve the blood vessel skeleton structure.

[0140] Step 403: judging whether an iteration termination condition is satisfied based on the local blood vessel segmentation image and the number of iterations;

[0141] On the one hand, an iteration termination condition can be created based on the vessel expansion result in the local vessel segmentation and highlighting to determine whether to terminate the iteration. On the other hand, an iteration termination condition can be created based on the current number of iterations to determine whether to terminate the iteration.

[0142] Optionally, the iteration termination condition may include: during N consecutive iterations, the newly segmented blood vessel area in the local blood vessel segmentation image is less than a preset value; or the current number of iterations reaches a maximum limit. Here, N is an integer greater than or equal to 1.

[0143] Here, the newly added blood vessel segmentation area refers to the area difference between the blood vessel area in the local blood vessel segmentation image output by the model and the blood vessel area in the local image input to the model.

[0144] The preset value, N value, and maximum limit here can be determined according to actual conditions. The preset value can be 1% of the blood vessel area in the local blood vessel segmentation image. The value of N can be 3. The maximum limit can be 10.

[0145] That is, if the newly added blood vessel area accounts for less than 1% in three consecutive iterations, the segmentation is considered converged and no further expansion is performed. Alternatively, if the current number of iterations reaches 10, the iteration is terminated to prevent the model from repeatedly calculating invalid areas and improve computational efficiency.

[0146] Step 404: If the iteration termination condition is not met, then determine the target endpoint in the local blood vessel segmentation image.

[0147] Here, the target endpoint is the endpoint farthest from the vascular skeleton structure in the local image among all endpoints of the local blood vessel segmentation image.

[0148] Optionally, the specific steps of determining the target endpoint in the local blood vessel segmentation image are as follows:

[0149] First, the vascular skeleton structure in the local vascular segmentation image is extracted and each endpoint in the vascular skeleton structure is determined. Then, the original vascular skeleton structure in the local image corresponding to the local vascular segmentation image is determined, and the endpoint farthest from the original vascular skeleton structure among all the endpoints of the vascular skeleton structure in the local vascular segmentation image is determined as the target endpoint.

[0150] For ease of description, the vascular skeleton structure in the local vascular segmentation image output by the vascular segmentation model is defined as the first vascular skeleton structure. The vascular skeleton structure in the local vascular image input to the vascular segmentation model is defined as the second vascular skeleton structure. The target endpoint is the endpoint farthest from the second vascular skeleton structure among all endpoints of the first vascular skeleton structure.

[0151] For example, see Figure 5 , Figure 5 The left side shows the vessel skeleton structure in the local image input to the vessel segmentation model, that is, the second vessel skeleton structure. According to the second vessel skeleton structure, the corresponding second vessel segment region can be determined. Figure 5 The right side shows the vessel skeleton structure in the local vessel segmentation image output by the vessel segmentation model, namely the first vessel skeleton structure. Based on the first vessel skeleton structure, the corresponding first vessel segment region can be determined. Among all endpoints of the first vessel skeleton structure, the endpoint farthest from the second vessel skeleton structure / second vessel segment region can be determined, namely the target endpoint.

[0152] Here, by determining the endpoint farthest away as the target endpoint, it can be ensured that the blood vessel can be gradually expanded in the correct direction.

[0153] Step 405 : updating the local segmentation window according to the target endpoint, updating the blood vessel segmentation image and the dual-channel image based on the local blood vessel segmentation image, and updating the current number of iterations.

[0154] First, based on the target endpoint, the local segmentation window is redefined with the target endpoint as the center. Second, based on the local vessel segmentation image, the overall vessel segmentation image is updated, and the dual-channel image is updated accordingly. In addition, the current iteration count can be updated to prepare for the next iterative segmentation.

[0155] When merging the local vessel segmentation images into the overall vessel segmentation image to update the image, the local vessel segmentation results are aligned according to spatial coordinates and accumulated into the overall vessel segmentation image. Furthermore, morphological closing operations can be used to remove small segmentation gaps to maintain the smoothness of vessel boundaries.

[0156] Step 406 jumps to the step of determining the local image corresponding to the local segmentation window in the dual-channel image until the iteration termination condition is met, determines the current local vessel segmentation image as the iterative vessel segmentation result, and updates the vessel segmentation image according to the iterative vessel segmentation result.

[0157] Based on the updated local segmentation window and dual-channel image, the local image is re-determined and the subsequent vascular segmentation steps are continued until the iteration termination condition is met. The local vascular segmentation image currently output by the vascular segmentation model is used as the iterative vascular segmentation result, and the overall vascular segmentation image is updated based on the local iterative vascular segmentation result to complete the local iterative segmentation of an endpoint.

[0158] The above local iterative segmentation process ensures the gradual advancement of blood vessel segmentation, while avoiding repeated calculation of invalid areas, thereby improving the efficiency and accuracy of segmentation.

[0159] Traditional methods, due to overall optimization, struggle to accurately focus on unsegmented regions of blood vessels, potentially leading to problems such as broken vessels and missing side branches in the segmentation results. This embodiment of the present invention utilizes vascular skeleton structure extraction and endpoint iterative tracking segmentation to gradually expand the segmented region. This effectively complements the segmentation results for low-contrast and small vessels, ensuring the integrity and connectivity of the vascular network.

[0160] Traditional methods are prone to vascular discontinuity and serious missegmentation in low-contrast, noisy medical images (such as CTA and MRA). This embodiment of the present invention combines endpoint local segmentation, dual-channel image input, and endpoint tracking iterative segmentation to enable the model to more stably identify low-contrast vessels and improve segmentation robustness.

[0161] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0162] The following are device embodiments of the present invention. For details not fully described therein, reference may be made to the corresponding method embodiments described above.

[0163] Figure 6 The following is a schematic diagram showing the structure of a blood vessel image segmentation device provided by an embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown, which are described in detail as follows:

[0164] like Figure 6 As shown, the blood vessel image segmentation device 6 includes: an acquisition module 61 , an extraction module 62 and a segmentation module 63 .

[0165] An acquisition module 61 is used to acquire an original blood vessel image and its corresponding initial blood vessel segmentation image;

[0166] An extraction module 62 is configured to extract a vascular skeleton structure based on the vascular segmentation image and determine endpoints on the vascular skeleton structure; the vascular skeleton structure includes multiple vascular centerlines, and the endpoints are the endpoints of the vascular centerlines;

[0167] The segmentation module 63 is used to:

[0168] Superimpose the original vascular image and the vascular segmentation image to obtain a dual-channel image;

[0169] For each endpoint, determine the local segmentation window corresponding to the endpoint, perform iterative vascular segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and update the vascular segmentation image according to the iterative vascular segmentation result;

[0170] Traverse all endpoints to obtain the final blood vessel segmentation image.

[0171] In a possible implementation, the segmentation module 63 is specifically configured to:

[0172] Initialize the number of iterations;

[0173] Determine the local image corresponding to the local segmentation window in the dual-channel image, and input the local image into the blood vessel segmentation model to obtain a local blood vessel segmentation image;

[0174] According to the local blood vessel segmentation image and the number of iterations, whether the iteration termination condition is met is determined;

[0175] If the iteration termination condition is not met, the target endpoint in the local blood vessel segmentation image is determined; the target endpoint is the endpoint farthest from the blood vessel skeleton structure in the local image among all the endpoints in the local blood vessel segmentation image;

[0176] According to the target endpoint, the local segmentation window is updated, the vascular segmentation image and the dual-channel image are updated based on the local vascular segmentation image, and the current number of iterations is updated;

[0177] Jump to the step of determining the local image corresponding to the local segmentation window in the dual-channel image until the iteration termination condition is met, determine the current local blood vessel segmentation image as the iterative blood vessel segmentation result, and update the blood vessel segmentation image according to the iterative blood vessel segmentation result.

[0178] In a possible implementation, the iteration termination condition includes:

[0179] During N consecutive iterations, the newly added blood vessel segmentation area in the local blood vessel segmentation image is smaller than a preset value; N is an integer greater than or equal to 1;

[0180] or

[0181] The current number of iterations has reached the maximum limit.

[0182] In a possible implementation, the segmentation module 63 is specifically configured to:

[0183] Extracting the vascular skeleton structure in the local vascular segmentation image and determining each endpoint in the vascular skeleton structure;

[0184] An original blood vessel skeleton structure in a local image corresponding to the local blood vessel segmentation image is determined, and an endpoint farthest from the original blood vessel skeleton structure among all endpoints of the blood vessel skeleton structure in the local blood vessel segmentation image is determined as a target endpoint.

[0185] In a possible implementation, the segmentation module 63 is specifically configured to:

[0186] For each endpoint, extract the actual blood vessel diameter at the endpoint in the blood vessel segmentation image;

[0187] respectively obtaining a preset standard window size and a preset standard blood vessel diameter, and determining a proportional relationship between the standard window size and the standard blood vessel diameter;

[0188] Based on the proportional relationship and the actual blood vessel diameter, the local window size corresponding to the endpoint is determined;

[0189] The endpoint is used as the center of the local split window, and the local split window is determined according to the local window size.

[0190] In a possible implementation, the extraction module 62 is specifically configured to:

[0191] For each foreground pixel in the blood vessel segmentation image, detect whether there are background pixels in the neighborhood of the foreground pixel; the foreground pixel is the pixel used to represent the blood vessel, and the background pixel is the remaining pixel except the foreground pixel;

[0192] If there is a background pixel in the neighborhood of the foreground pixel, the foreground pixel is determined as a candidate pixel;

[0193] Detect whether the topological structure of the blood vessel segmentation image changes before and after each candidate pixel is deleted;

[0194] Determine the candidate pixels whose topological structures in the blood vessel segmentation image before and after deletion have not changed as target candidate pixels, and delete them to obtain a new blood vessel segmentation image;

[0195] Jump to the step of detecting whether there are background pixels in the neighborhood of each foreground pixel in the blood vessel segmentation image, until there are no target candidate pixels in the current blood vessel segmentation image, thereby obtaining the blood vessel skeleton structure;

[0196] For each pixel point on the blood vessel skeleton structure, if the number of foreground pixels in the neighborhood of the pixel point is 1, the pixel point is determined to be an endpoint.

[0197] This device embodiment can be used to execute the above method embodiment. Its technical principles and technical effects are the same as those of the above method embodiment, and will not be repeated here.

[0198] Figure 7 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 7 As shown, the electronic device 7 of this embodiment includes a processor 70 and a memory 71. The memory 71 stores a computer program 72. When the processor 70 executes the computer program 72, the steps of the above-described method embodiments are implemented. Alternatively, when the processor 70 executes the computer program 72, the functions of the modules / units in the above-described device embodiments are implemented.

[0199] Exemplarily, the computer program 72 may be divided into one or more modules / units, which are stored in the memory 71 and executed by the processor 70 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 72 in the electronic device 7.

[0200] The electronic device 7 may include, but is not limited to, a processor 70 and a memory 71. Those skilled in the art will appreciate that Figure 7 It is only an example of the electronic device 7 and does not constitute a limitation of the electronic device 7. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device 7 may also include input and output devices, network access devices, buses, etc.

[0201] The processor 70 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0202] The memory 71 can be an internal storage unit of the electronic device 7, such as the hard drive or memory of the electronic device 7. The memory 71 can also be an external storage device of the electronic device 7, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the electronic device 7. Furthermore, the memory 71 can include both the internal storage unit of the electronic device 7 and an external storage device. The memory 71 is used to store the computer program 72 and other programs and data required by the electronic device 7. The memory 71 can also be used to temporarily store data that has been output or is about to be output.

[0203] For the sake of convenience and brevity, the division of the above functional modules / units is only used as an example. In actual applications, the above functions can be assigned to different functional modules / units as needed. The above modules / units can be implemented in the form of hardware, software, or a combination of hardware and software.

[0204] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in the above-mentioned method embodiments.

[0205] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the methods in the above-mentioned method embodiments.

[0206] The term "computer program" includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. Computer-readable media may include any entity or device capable of carrying computer program code, recording media, USB flash drives, removable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media.

[0207] In the above embodiments, the descriptions of each embodiment have their own focus. For parts not described or recorded in detail in one embodiment, please refer to the relevant descriptions of other embodiments. Unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other. The technical features of different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0208] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A blood vessel image segmentation method, characterized in that: include: Obtaining an original blood vessel image and its corresponding initial blood vessel segmentation image; Extracting a vascular skeleton structure based on the vascular segmentation image and determining endpoints on the vascular skeleton structure; the vascular skeleton structure includes a plurality of vascular centerlines, and the endpoints are endpoints of the vascular centerlines; the endpoints are used to locate unsegmented areas in the vascular segmentation image; superimposing the original blood vessel image and the blood vessel segmentation image to obtain a dual-channel image; For each endpoint, determining a local segmentation window corresponding to the endpoint, performing iterative vascular segmentation on a local image corresponding to the local segmentation window in the dual-channel image, and updating the vascular segmentation image according to the iterative vascular segmentation result; Traverse all endpoints to obtain the final blood vessel segmentation image; The iteratively performing blood vessel segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and updating the blood vessel segmentation image according to the iterative blood vessel segmentation result, comprises: Initialize the number of iterations; Determining a local image corresponding to the local segmentation window in the dual-channel image, and inputting the local image into a blood vessel segmentation model to obtain a local blood vessel segmentation image; determining whether an iteration termination condition is satisfied according to the local blood vessel segmentation image and the number of iterations; If the iteration termination condition is not met, determining a target endpoint in the local blood vessel segmentation image; the target endpoint is an endpoint farthest from the blood vessel skeleton structure in the local image among all endpoints in the local blood vessel segmentation image; updating the local segmentation window according to the target endpoint, updating the blood vessel segmentation image and the dual-channel image based on the local blood vessel segmentation image, and updating the current number of iterations; Jumping to the step of determining the local image corresponding to the local segmentation window in the dual-channel image until the iteration termination condition is satisfied, determining the current local blood vessel segmentation image as the iterative blood vessel segmentation result, and updating the blood vessel segmentation image according to the iterative blood vessel segmentation result; The iteration termination condition includes: During N consecutive iterations, the newly added blood vessel segmentation area in the local blood vessel segmentation image is smaller than a preset value; N is an integer greater than or equal to 1; or The current number of iterations has reached the maximum limit.

2. The blood vessel image segmentation method according to claim 1, characterized in that: Determining the target endpoint in the local blood vessel segmentation image includes: extracting a blood vessel skeleton structure from the local blood vessel segmentation image, and determining endpoints in the blood vessel skeleton structure; An original blood vessel skeleton structure in a local image corresponding to the local blood vessel segmentation image is determined, and an endpoint farthest from the original blood vessel skeleton structure among all endpoints of the blood vessel skeleton structure in the local blood vessel segmentation image is determined as a target endpoint.

3. The blood vessel image segmentation method according to claim 1, characterized in that: The step of determining, for each endpoint, a local segmentation window corresponding to the endpoint includes: For each endpoint, extracting the actual blood vessel diameter at the endpoint in the blood vessel segmentation image; respectively obtaining a preset standard window size and a preset standard blood vessel diameter, and determining a proportional relationship between the standard window size and the standard blood vessel diameter; determining a local window size corresponding to the endpoint based on the proportional relationship and the actual blood vessel diameter; The endpoint is used as the center of the local segmentation window, and the local segmentation window is determined according to the local window size.

4. The blood vessel image segmentation method according to claim 1, wherein: The extracting of the blood vessel skeleton structure based on the blood vessel segmentation image and determining endpoints on the blood vessel skeleton structure includes: For each foreground pixel in the blood vessel segmentation image, detecting whether there is a background pixel in the neighborhood of the foreground pixel; the foreground pixel is a pixel used to represent the blood vessel, and the background pixel is the remaining pixel except the foreground pixel; If there is a background pixel in the neighborhood of the foreground pixel, the foreground pixel is determined as a candidate pixel; Detect whether the topological structure of the blood vessel segmentation image changes before and after each candidate pixel is deleted; Determine the candidate pixels whose topological structures in the blood vessel segmentation image before and after deletion have not changed as target candidate pixels, and delete them to obtain a new blood vessel segmentation image; Jumping to the step of detecting whether there is a background pixel in the neighborhood of each foreground pixel in the blood vessel segmentation image until the target candidate pixel does not exist in the current blood vessel segmentation image, thereby obtaining the blood vessel skeleton structure; For each pixel point on the blood vessel skeleton structure, if the number of foreground pixels in the neighborhood of the pixel point is 1, the pixel point is determined to be an endpoint.

5. A blood vessel image segmentation device, characterized in that: include: An acquisition module, used for acquiring an original blood vessel image and its corresponding initial blood vessel segmentation image; an extraction module, configured to extract a vascular skeleton structure based on the vascular segmentation image and determine endpoints on the vascular skeleton structure; the vascular skeleton structure includes a plurality of vascular centerlines, and the endpoints are endpoints of the vascular centerlines; the endpoints are used to locate unsegmented areas in the vascular segmentation image; Segmentation module for: superimposing the original blood vessel image and the blood vessel segmentation image to obtain a dual-channel image; For each endpoint, determining a local segmentation window corresponding to the endpoint, performing iterative vascular segmentation on a local image corresponding to the local segmentation window in the dual-channel image, and updating the vascular segmentation image according to the iterative vascular segmentation result; Traverse all endpoints to obtain the final blood vessel segmentation image; The iteratively performing blood vessel segmentation on the local image corresponding to the local segmentation window in the dual-channel image, and updating the blood vessel segmentation image according to the iterative blood vessel segmentation result, comprises: Initialize the number of iterations; Determining a local image corresponding to the local segmentation window in the dual-channel image, and inputting the local image into a blood vessel segmentation model to obtain a local blood vessel segmentation image; determining whether an iteration termination condition is satisfied according to the local blood vessel segmentation image and the number of iterations; If the iteration termination condition is not met, determining a target endpoint in the local blood vessel segmentation image; the target endpoint is an endpoint farthest from the blood vessel skeleton structure in the local image among all endpoints in the local blood vessel segmentation image; updating the local segmentation window according to the target endpoint, updating the blood vessel segmentation image and the dual-channel image based on the local blood vessel segmentation image, and updating the current number of iterations; Jumping to the step of determining the local image corresponding to the local segmentation window in the dual-channel image until the iteration termination condition is satisfied, determining the current local blood vessel segmentation image as the iterative blood vessel segmentation result, and updating the blood vessel segmentation image according to the iterative blood vessel segmentation result; The iteration termination condition includes: During N consecutive iterations, the newly added blood vessel segmentation area in the local blood vessel segmentation image is smaller than a preset value; N is an integer greater than or equal to 1; or The current number of iterations has reached the maximum limit.

6. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

8. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 4 when the computer program is executed by a processor.

Images