Methods, devices, ultrasound equipment, and storage media for extracting the intima contour of blood vessels.

Abstract

This application discloses a method, apparatus, ultrasound device, and computer-readable storage medium for extracting the intima contour of a blood vessel lumen. The method includes: acquiring an ultrasound image of a cross-section of a blood vessel and preprocessing the ultrasound image to obtain a target image; determining the lumen wall region in the target image and determining the initial intima contour line corresponding to the lumen wall region; fitting the initial intima contour line to obtain the final intima contour line; and adding the final intima contour line to the ultrasound image. Therefore, the method for extracting the intima contour of a blood vessel lumen provided in this application first finds the initial intima contour line, and then accurately fits the intima of the blood vessel lumen to the ultrasound image of the cross-section of the blood vessel, thus improving the accuracy of intima contour extraction.

Description

Technical Field

[0001] This application relates to the field of ultrasound imaging technology, and more specifically, to a method and apparatus for extracting the contour of the endothelium of a blood vessel lumen, an ultrasound device, and a computer-readable storage medium. Background Technology

[0002] Intravenous ultrasound (IVUS) is a medical imaging technique that combines non-invasive ultrasound technology with invasive catheter technology, using a special catheter with an ultrasound probe attached to its end. The intima and endothelium of the coronary artery wall are crucial for the diagnosis and quantitative analysis of various parameters in coronary artery disease; therefore, accurate detection of these intima and endothelium edges in IVUS images is of great significance for the clinical diagnosis and treatment of coronary artery disease. In related technologies, image clustering algorithms can be used to statistically cluster the grayscale distribution of IVUS images to segment the intima and endothelium of the vessel lumen. However, due to the complexity of IVUS images, including varying degrees of lesions, artifacts, and plaques, the accuracy of segmenting the intima contour of the vessel lumen is relatively low.

[0003] Therefore, improving the accuracy of vascular intima contour extraction is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, ultrasound device, and computer-readable storage medium for extracting the contour of the vascular lumen, thereby improving the accuracy of vascular lumen contour extraction.

[0005] To achieve the above objectives, this application provides a method for extracting the intima contour of a blood vessel lumen, comprising:

[0006] Acquire an ultrasound image of the cross-section inside the blood vessel, and preprocess the ultrasound image to obtain the target image;

[0007] In the target image, a lumen wall region is determined, and an initial intima contour line corresponding to the lumen wall region is determined.

[0008] The initial intima profile is fitted to obtain the final lumen intima profile;

[0009] The final endothelial contour is added to the ultrasound image.

[0010] The step of preprocessing the ultrasound image to obtain the target image includes:

[0011] The ultrasound image is downsampled and filtered;

[0012] Accordingly, fitting the initial intima profile to obtain the final luminal intima profile includes:

[0013] The initial intima profile is fitted to obtain the fitted lumen intima profile, and the fitted lumen intima profile is upsampled to obtain the final lumen intima profile.

[0014] The step of preprocessing the ultrasound image to obtain the target image includes:

[0015] Remove the conduit effect from the image.

[0016] The removal of the duct effect in the image includes:

[0017] The target location is determined by identifying the first non-zero gray value pixel in each direction starting from the center of the image.

[0018] Calculate the average distance between each target location and the center location of the image, and calculate the sum of the average distance and the preset offset distance as the superimposed distance;

[0019] Set the grayscale values ​​of all pixels whose distance from the center of the image is less than the superposition distance to the lower limit value.

[0020] The step of preprocessing the ultrasound image to obtain the target image includes:

[0021] Saliency enhancement is performed on regions of interest in the image.

[0022] The step of preprocessing the ultrasound image to obtain the target image includes:

[0023] Normalize the grayscale values ​​of pixels in the image.

[0024] The step of determining the lumen wall region in the target image includes:

[0025] The category of each pixel in the target image is determined in order to segment the target image into different cluster blocks;

[0026] The lumen wall region is identified in the cluster block, and external and internal noise of the lumen wall region is removed.

[0027] The step of determining the category of each pixel in the target image to segment the target image into different clustering blocks includes:

[0028] The fuzzy C-means clustering algorithm based on grayscale histograms performs clustering operations on the pixels in the target image to segment the target image into different cluster blocks.

[0029] The fuzzy C-means clustering algorithm based on grayscale histograms performs clustering operations on pixels in the target image to segment the target image into different cluster blocks, including:

[0030] Morphological processing is used to perform erosion reconstruction on the target image to obtain an intermediate image, and clustering parameters are determined; wherein, the clustering parameters include the number of cluster centers, kernel processing scale, and fuzziness factor;

[0031] Different gray values ​​are determined as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers. The gray value membership function is determined based on the fuzziness factor and the Euclidean clustering of the gray values ​​in the gray-level histogram with the initial cluster centers.

[0032] The initial cluster centers and the gray value membership function are iteratively updated until the fuzzy C-mean objective function is optimized or the number of iterations reaches a preset number.

[0033] The maximum value in the updated cluster centers is determined as the second grayscale threshold. The grayscale values ​​of all pixels in the intermediate image whose grayscale values ​​are greater than or equal to the second grayscale threshold are set to the upper limit value, and the grayscale values ​​of all pixels whose grayscale values ​​are less than the second grayscale threshold are set to the lower limit value, so as to segment the target image into different cluster blocks.

[0034] The step of determining the lumen wall region in the cluster includes:

[0035] The area and center position of each cluster are determined, and the cluster with the largest area and the closest distance between its center position and the center position of the image is determined as the lumen wall region.

[0036] The removal of external noise from the lumen wall region includes:

[0037] Calculate the Euclidean distance between the region center positions of all the clustered blocks and the region center positions of the lumen wall region;

[0038] Calculate the overlap area between the bounding rectangles corresponding to all the clustered blocks and the bounding rectangles corresponding to the lumen wall region;

[0039] Clusters whose Euclidean distance is greater than a preset distance threshold and whose overlapping area is less than a first preset area threshold are identified as external noise in the lumen wall region, and the external noise is removed.

[0040] The removal of internal noise in the lumen wall region includes:

[0041] The first polar coordinate image is obtained by performing polar coordinate transformation on the image after removing external noise.

[0042] In the first polar coordinate image, determine the region area of ​​each cluster block, the average gray level of the pixels, and the number of pixels overlapping with the lumen wall region;

[0043] Clustered blocks whose area is less than a second preset area threshold, whose average grayscale value is greater than a third grayscale threshold, and whose number of overlapping pixels is greater than a preset number are identified as internal noise in the cavity wall region, and the internal noise is removed.

[0044] Wherein, determining the initial intima contour line corresponding to the lumen wall region includes:

[0045] A second polar coordinate image is obtained by performing polar coordinate transformation on the image after removing external and internal noise. The first pixel with a non-zero gray value in each column of the first polar coordinate image is taken as the contour point.

[0046] The missing parts of the lines connecting the contour points are filled in using interpolation, and a Cartesian coordinate transformation is performed to obtain the initial inner membrane contour line.

[0047] The process of removing external and internal noise from the lumen wall region further includes:

[0048] Determine the maximum lateral width of the lumen wall region, and determine whether the ratio of the maximum lateral width to the width of the image after removing external and internal noise is greater than a preset value;

[0049] If so, proceed to the step of performing polar coordinate transformation on the image after removing external and internal noise to obtain a second polar coordinate image;

[0050] If not, update the clustering parameters and re-enter the step of determining different gray values ​​as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers.

[0051] The fitting of the initial intima contour includes:

[0052] The initial inner membrane contour is fitted using a vector field convolution active contour model.

[0053] To achieve the above objectives, this application provides a device for extracting the intima contour of a blood vessel lumen, comprising:

[0054] The preprocessing module is used to acquire ultrasound images of the intravascular cross-section and preprocess the ultrasound images to obtain the target image.

[0055] The determination module is used to determine the lumen wall region in the target image and determine the initial intima contour line corresponding to the lumen wall region;

[0056] The fitting module is used to fit the initial intima contour line to obtain the final lumen intima contour line.

[0057] An add module is used to add the final endothelial contour to the ultrasound image.

[0058] To achieve the above objectives, this application provides an ultrasonic device, comprising:

[0059] Memory, used to store computer programs;

[0060] A processor is configured to implement the steps of the above-described method for extracting the vascular lumen intima contour when executing the computer program.

[0061] A display for showing an ultrasound image of a cross-section within a blood vessel and the intima outline in the ultrasound image.

[0062] To achieve the above objectives, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for extracting the endothelial contour of a blood vessel lumen.

[0063] As can be seen from the above scheme, the method for extracting the endothelial contour of a blood vessel lumen provided in this application includes: acquiring an ultrasound image of a cross-section of a blood vessel and preprocessing the ultrasound image to obtain a target image; determining a lumen wall region in the target image and determining an initial endothelial contour line corresponding to the lumen wall region; fitting the initial endothelial contour line to obtain a final endothelial contour line; and adding the final endothelial contour line to the ultrasound image.

[0064] Therefore, the vascular endothelial contour extraction method provided in this application first identifies the initial contour line of the endothelial lining, and then accurately fits the endothelial lining of the vascular cross-section to the ultrasound image, thereby improving the accuracy of vascular endothelial contour extraction. This application also discloses a vascular endothelial contour extraction device, an ultrasound device, and a computer-readable storage medium, which can achieve the same technical effects.

[0065] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this application. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. The drawings are used to provide a further understanding of this disclosure and constitute a part of the specification. They are used together with the following detailed description to explain this disclosure, but do not constitute a limitation of this disclosure. In the drawings:

[0067] Figure 1 This is a flowchart illustrating a method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment;

[0068] Figure 2a and Figure 2b This is a schematic diagram illustrating the removal of the duct effect before and after, according to an exemplary embodiment.

[0069] Figure 3a and Figure 3b This is a schematic diagram illustrating saliency enhancement and normalization before and after processing according to an exemplary embodiment;

[0070] Figure 4a This is a schematic diagram illustrating an ellipse after fitting, according to an exemplary embodiment.

[0071] Figure 4b This is a schematic diagram of an edge image according to an exemplary embodiment;

[0072] Figure 4c This is a schematic diagram illustrating the final luminal intima profile according to an exemplary embodiment;

[0073] Figure 5 This is a flowchart illustrating another method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment;

[0074] Figure 6a This is a schematic diagram illustrating morphological processing according to an exemplary embodiment;

[0075] Figure 6b This is a schematic diagram illustrating fuzzy clustering according to an exemplary embodiment;

[0076] Figure 6c This is a schematic diagram illustrating threshold segmentation according to an exemplary embodiment;

[0077] Figure 7 This is a schematic diagram illustrating a connected region according to an exemplary embodiment;

[0078] Figure 8aThis is a schematic diagram illustrating the removal of the external candidate lumen wall region according to an exemplary embodiment;

[0079] Figure 8b This is a schematic diagram of a first polar coordinate image according to an exemplary embodiment;

[0080] Figure 8c This is a schematic diagram illustrating the removal of the internal candidate lumen wall region according to an exemplary embodiment;

[0081] Figure 9a This is a schematic diagram of a second polar coordinate image according to an exemplary embodiment;

[0082] Figure 9b This is a schematic diagram of contour points in an initial endometrial contour line according to an exemplary embodiment;

[0083] Figure 9c This is a schematic diagram illustrating contour point interpolation according to an exemplary embodiment;

[0084] Figure 9d This is a schematic diagram illustrating the initial intima contour according to an exemplary embodiment;

[0085] Figure 10 This is a flowchart illustrating yet another method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment;

[0086] Figure 11 This is a structural diagram of a vascular lumen intima contour extraction device according to an exemplary embodiment;

[0087] Figure 12 This is a structural diagram of an ultrasonic device according to an exemplary embodiment. Detailed Implementation

[0088] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Furthermore, in the embodiments of this application, "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0089] This application discloses a method for extracting the contour of the vascular lumen intima, which improves the accuracy of vascular lumen intima contour extraction.

[0090] See Figure 1 A flowchart illustrating a method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment is shown below. Figure 1 As shown, it includes:

[0091] S101: Acquire an ultrasound image of the cross-section inside the blood vessel, and preprocess the ultrasound image to obtain the target image;

[0092] The subject of this application can be an ultrasound device, and the purpose is to extract the endothelium of a blood vessel lumen from an ultrasound image of a cross-section within the blood vessel. In this step, the ultrasound image of the cross-section within the blood vessel is first preprocessed to obtain the target image. As a possible implementation, the preprocessing of the ultrasound image may include downsampling and filtering the ultrasound image.

[0093] In practice, given the large size of ultrasound images of intravascular cross-sections, downsampling is necessary to improve the speed of subsequent image processing. For example, a sampling rate of 0.5 can be set, taking one data point from every two data points in both the horizontal and vertical directions, effectively halving the image size. The downsampled image is then filtered to eliminate some noise affecting the extraction of the intima edge. For instance, the filter coefficient can be set to Filter = [1,4,6,4,1] / 16. During filtering, the original image data can be horizontally convolved first, followed by a vertical convolution operation on the result to complete the filtering process.

[0094] Preprocessing ultrasound images may also include removing ductal effects from the image. It is understood that since the center of an ultrasound image of a cross-section within a blood vessel is the duct, and the vicinity of the image center is a halo artifact, this can easily interfere with subsequent contour extraction. Therefore, during preprocessing, the bright halo artifact near the duct can be removed. As a feasible implementation, removing ductal effects from the image includes: determining the first pixel with a non-zero grayscale value in each direction starting from the image center as the target location; calculating the average distance between each target location and the image center location; calculating the sum of the average distance and a preset offset distance as the superposition distance; setting the grayscale values ​​of all pixels whose distance from the image center location is less than the superposition distance to a lower limit value to obtain a second intermediate image. In this embodiment, the first pixel with a non-zero grayscale value is searched for in a 360-degree circumferential direction from the image center location as the target location; the average distance between each target location in the 360-degree circumference and the image center location is calculated; then the sum of the average distance and a preset offset distance is determined as the superposition distance. The preset offset distance is set empirically, for example, it can be set to 15 pixels. Finally, all pixels with a grayscale value less than the superposition distance from the image center are set to the lower limit value, for example, all are set to zero. (Before removing the conduit effect...) Figure 2a As shown, after removing the duct effect, as Figure 2b As shown.

[0095] Preprocessing ultrasound images may further include: significantly enhancing the region of interest (ROI) in the image to increase the grayscale value of the tissue near the vessel wall and highlight the contour of the endothelial lining. As a feasible implementation, significantly enhancing the ROI in the image includes: determining the ROI in the image; wherein the grayscale value of each pixel in the ROI is greater than a first grayscale threshold; calculating the average pixel value of all pixels in the ROI, and determining a mapping curve based on the average pixel value; and processing the grayscale value of each pixel in the image using the mapping curve. It is understood that the purpose of significant enhancement is to highlight the ROI of the tissue near the vessel wall and suppress non-ROI regions within the vessel lumen. In this embodiment, the number of grayscale values ​​in the image is counted to calculate the average pixel value of all pixels in the ROI, establishing a mapping curve y = k × log2x. This mapping curve is used to reset the grayscale value of each pixel in the image, where x is the original grayscale value of the pixel, y is the updated grayscale value of the pixel, and k is the aforementioned average pixel value. Simultaneously, the grayscale values ​​of the highlighted areas are calculated and a mapping curve is established. Finally, the original IVUS is significantly enhanced through the mapping curve.

[0096] Furthermore, preprocessing of ultrasound images can also include normalizing the grayscale values ​​of pixels in the image. In specific implementations, normalizing the grayscale values ​​of pixels in the image yields the target image, avoiding grayscale values ​​exceeding the range of 0-255. For example... Figure 3a As shown, after significance enhancement and normalization, as Figure 3b As shown.

[0097] It should be noted that the order of the above preprocessing steps is not strictly limited in this embodiment, nor is the number of times each step is performed. These steps can be flexibly adjusted according to the actual imaging conditions of the ultrasound image. As a preferred embodiment, the ultrasound image is first downsampled and filtered; secondly, the duct effect in the image is removed; then, the region of interest in the image is significantly enhanced; and finally, the grayscale values ​​of the pixels in the image are normalized.

[0098] S102: Determine the lumen wall region in the target image, and determine the initial intima contour line corresponding to the lumen wall region;

[0099] The purpose of this step is to identify the lumen wall region in the target image and use the contour line of this lumen wall region as the initial intima contour line. In a preferred embodiment, identifying the lumen wall region in the target image includes: determining the category of each pixel in the target image to segment the target image into different clusters; identifying the lumen wall region within each cluster; and removing external and internal noise from the lumen wall region. In a specific implementation, a clustering algorithm is used to cluster the pixels in the target image, dividing the target image into different clusters. The lumen wall region is then identified within all clusters, and other clusters are removed as noise.

[0100] Preferably, determining the category of each pixel in the target image to segment the target image into different cluster blocks includes: performing a fuzzy C-means clustering algorithm based on gray-level histograms to cluster the pixels in the target image into different cluster blocks. It can be understood that the fuzzy C-means clustering algorithm based on gray-level histograms uses different gray-level values ​​as cluster centers, thereby dividing the target image into classes that are continuous in both gray-level and position, and further extracting and segmenting different cluster blocks of the blood vessel lumen wall.

[0101] S103: Fit the initial intima profile to obtain the final lumen intima profile;

[0102] S104: Add the final endothelial contour to the ultrasound image.

[0103] Since the intima contour determined by fuzzy clustering deviates from the actual vascular lumen, the initial intima contour is subjected to evolutionary fitting to extract the final luminal intima contour, which is then displayed in the ultrasound image of the intravascular cross-section.

[0104] In a preferred embodiment, fitting the initial intima contour line includes: fitting the initial intima contour line using a vector field convolution active contour model. In a specific implementation, the initial intima contour line is first fitted with an ellipse to reduce the computational complexity of the contour line fitting, such as... Figure 4a As shown, the initial intima contour v(s)=(x(s),y(s)) can be expressed as:

[0105]

[0106] Where (x) c ,y c ) represents the coordinates of the center position of the target image, and r, θ represents the distance radius and angle between the contour point in the initial inner membrane contour and the center position of the image, respectively.

[0107] In the contour model, the contour lines satisfy the following equation:

[0108]

[0109] Where E int The initial external force field for the intima contour includes two external force field parameters, α and β. Parameter α controls the contour tension; a larger value makes the curve less prone to stretching and causes faster contraction. Parameter β controls the curve rigidity; a larger value makes the curve less prone to bending and smoother. E ext [v(s)] represents the external force term of the image features, which can be expressed as the F-value of the convolution of the vector field kernel with the image edges. vfc To represent F vfc (x,y)=f(x,y)*K(x,y)=[f(x,y)*u(x,y),f(x,y)*v(x,y)], where f(x,y) is the edge image detected by phase consistency of the ultrasound image of the intravascular cross-section, such as... Figure 4b As shown, K(x,y) is the vector field convolution kernel, which can be defined as K(x,y)=[u(x,y),v(x,y)]=m(x,y)n(x,y), where u(x,y) and v(x,y) are the horizontal and vertical method components of the vector field convolution kernel, respectively, and n(x,y) is the unit vector pointing to the kernel origin (0,0), and m(x,y) is the magnitude of each vector. Minimize the above energy formula E. Snake Then the following Euler formula must be satisfied: If we consider v(s) as a function of time, it can be expressed as v(s,t). Therefore, by minimizing the above formula, we can obtain an iterative function with v(s,t) as the variable. Through multiple iterations, we can find the optimal endothelial profile, such as... Figure 4c As shown. The iterative function with v(s,t) as the variable is:

[0110]

[0111] It should be noted that, because the ultrasound image of the intravascular cross-section has been preprocessed, the final intimal contour is located on the downsampled image. To accurately reproduce the intimal contour from the ultrasound image of the intravascular cross-section, the final intimal contour needs to be upsampled according to the previously set sampling rate, thus locating the final intimal contour onto the original intravascular cross-section ultrasound image. That is, the step of fitting the initial intimal contour to obtain the final intimal contour includes: fitting the initial intimal contour to obtain a fitted intimal contour, and upsampling the fitted intimal contour to obtain the final intimal contour.

[0112] Therefore, the vascular lumen intima contour extraction method provided in this application first finds the initial intima contour line, and then accurately fits the intima of the vascular lumen to the ultrasound image of the vascular cross section, thereby improving the accuracy of vascular lumen intima contour extraction.

[0113] This application discloses a method for extracting the intima contour of a blood vessel lumen. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution. Specifically:

[0114] See Figure 5 A flowchart illustrating another method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment is shown below. Figure 5 As shown, it includes:

[0115] S201: Acquire an ultrasound image of the cross-section inside the blood vessel, and preprocess the ultrasound image to obtain the target image;

[0116] S202: Morphological processing is used to perform erosion reconstruction on the target image to obtain an intermediate image, and clustering parameters are determined; wherein, the clustering parameters include the number of cluster centers, kernel processing scale, and fuzziness factor;

[0117] In practice, due to the presence of speckle noise in the target image, morphological processing is needed to perform erosion reconstruction to obtain an intermediate image, such as... Figure 6a As shown. In the process of performing fuzzy C-means clustering based on gray-level histogram on intermediate images, the clustering parameters are first determined, including the number of cluster centers, kernel processing scale, and fuzziness factor.

[0118] S203: Determine different gray values ​​as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers, and determine the gray value membership function based on the fuzziness factor and the Euclidean clustering of the gray values ​​in the gray-level histogram with the initial cluster centers;

[0119] In this step, different gray values ​​can be determined by the range of the gray-level histogram of the intermediate image and the number of cluster centers, and these values ​​are used as the initial cluster centers. Furthermore, the gray-level membership function can be derived based on the membership function matrix in the standard fuzzy C-means clustering algorithm:

[0120]

[0121] Where C is the number of cluster centers, k = 1, 2, 3, ..., C, m is the fuzzy factor, and d ij =||x i -v i || is the Euclidean distance from the gray value of the j-th pixel to the i-th cluster center.

[0122] S204: Iteratively update the initial cluster centers and the gray value membership function until the fuzzy C-mean objective function is optimized or the number of iterations reaches a preset number;

[0123] In this step, the initial cluster centers and gray-level membership functions are updated iteratively until the fuzzy C-means objective function is optimized or the iteration reaches a preset number, at which point the iteration terminates. The final gray-level membership functions and cluster centers are then obtained, calculated using the following formulas:

[0124] Cluster center:

[0125] Objective function:

[0126] Among them, v i Let x be the i-th cluster center. i Let be the grayscale value of the j-th pixel, and n be the total number of pixels in the fourth intermediate image. The category of each pixel in the fourth intermediate image is determined based on the final grayscale value membership function, and the fourth intermediate image is then segmented into different parts, such as... Figure 6b As shown.

[0127] S205: The maximum value in the updated cluster center is determined as the second gray threshold. The gray values ​​of all pixels in the intermediate image whose gray values ​​are greater than or equal to the second gray threshold are set to the upper limit value, and the gray values ​​of all pixels whose gray values ​​are less than the second gray threshold are set to the lower limit value, so as to segment the target image into different cluster blocks.

[0128] In practice, the maximum value in the final cluster centers is determined as the second grayscale threshold. For pixels in the intermediate image, the grayscale values ​​of pixels with grayscale values ​​greater than or equal to the second grayscale threshold are set as the upper limit, which can be 255. The grayscale values ​​of pixels with grayscale values ​​less than the second grayscale threshold are set as the lower limit, which can be 0. This process segments the vascular lumen region. Figure 6c As shown.

[0129] S206: Determine the lumen wall region in the clustered block and remove the external and internal noise of the lumen wall region;

[0130] As a possible implementation, determining the lumen wall region in the clustered blocks includes: determining the area and center position of each clustered block, and determining the clustered block with the largest area and the closest distance between its center position and the image center position as the lumen wall region.

[0131] Understandably, an image after fuzzy C-means clustering is divided into different regions, and image connectivity methods are used to identify these different image regions, such as... Figure 7 As shown. The lumen wall region is determined based on the feature that the lumen wall region is closest to the image center. That is, the area and center position of each cluster block are determined, and the cluster block with the largest area and the closest distance between its center position and the image center position is determined as the final lumen wall region.

[0132] In this step, in order to eliminate the influence of other interfering blocks such as pericardial membrane and artifacts, it is necessary to remove other clustered blocks except for the final lumen wall region, that is, to remove external and internal noise in the lumen wall region.

[0133] Understandably, since external noise is far from the final lumen wall region, it can be removed by calculating the Euclidean distance between the center of the clustered blocks and the center of the final lumen wall region in a Cartesian coordinate system. Internal noise, however, is mostly caused by guide wire artifacts, strong reflections from calcified plaques, and other factors, and is located inside the final lumen wall region. It cannot be removed using the above method. Therefore, polar coordinate transformation can be performed on the image to remove internal noise. Of course, external noise can also be removed in polar coordinate images, but removing external and internal noise separately in Cartesian and polar coordinate images avoids confusion between the two and achieves better removal results.

[0134] As a feasible implementation method, removing external noise from the lumen wall region includes: calculating the Euclidean distance between the center position of all clustered blocks and the center position of the lumen wall region; calculating the overlap area between the bounding rectangles corresponding to all clustered blocks and the bounding rectangles corresponding to the lumen wall region; identifying clustered blocks whose Euclidean distance is greater than a preset distance threshold and whose overlap area is less than a first preset area threshold as external noise from the lumen wall region, and removing the external noise.

[0135] In this embodiment, noise is removed using preset distance thresholds and region overlap area thresholds. First, the Euclidean distance between the center of the final lumen wall region and the centers of all clustered blocks is calculated, and the bounding rectangles corresponding to each clustered block are determined. The overlap area between the bounding rectangle of each clustered block and the bounding rectangle of the final lumen wall region is calculated. When the overlap area is less than a first preset area threshold and the Euclidean distance is greater than a preset distance threshold, the clustered block is defined as an external candidate lumen wall region, and this external noise is removed. Figure 8a As shown.

[0136] As a feasible implementation method, removing internal noise from the lumen wall region includes: performing polar coordinate transformation on the image after removing external noise to obtain a first polar coordinate image; determining the region area, average gray value of each pixel, and number of pixels overlapping with the lumen wall region for each cluster in the first polar coordinate image; identifying clusters whose region area is less than a second preset area threshold, whose average gray value is greater than a third gray value threshold, and whose number of overlapping pixels is greater than a preset number as internal noise from the lumen wall region, and removing the internal noise.

[0137] In this embodiment, the image after removing external noise is transformed into polar coordinates to obtain a first polar coordinate image, such as... Figure 8b As shown. The conversion between rectangular coordinates (x, y) and polar coordinates (r, θ) is as follows:

[0138]

[0139] In the first polar coordinate image, the area of ​​the clustered blocks and the average grayscale value of the pixels are calculated, and the internal noise is determined by combining this with the number of overlapping pixels with the final lumen wall region. In a specific implementation, the final lumen wall region and all clustered blocks can be projected onto the horizontal axis of the second polar coordinate image, and the number of overlapping pixels is determined based on the overlap length between the projection of each clustered block and the projection of the final lumen wall region. When the area is less than a second preset area threshold, the average grayscale value is greater than a third grayscale threshold, and the number of overlapping pixels is greater than a preset number, the clustered block is defined as internal noise, and this internal noise is removed. Figure 8c As shown.

[0140] S207: Perform polar coordinate transformation on the image after removing external and internal noise to obtain a second polar coordinate image, and take the first pixel with a non-zero gray value in each column of the first polar coordinate image as a contour point.

[0141] S208: Use interpolation to fill in the missing parts of the lines connecting the contour points and perform rectangular coordinate transformation to obtain the initial inner membrane contour line;

[0142] In practice, the image after removing external and internal noise is transformed into polar coordinates to obtain a second polar coordinate image, such as... Figure 9a As shown, the edge map of the lumen region is detected through phase consistency. Since the lumen endothelial contour line is located in the upper part of the image after polar coordinate transformation, the first pixel with a non-zero gray value in each column of the second polar coordinate image is taken as the contour point in the initial endothelial contour line, as shown below. Figure 9b As shown. For the missing portion, i.e., the part with a ordinate of zero, perform cubic spline interpolation to find points where the overall intima edge is smooth, such as... Figure 9cAs shown. Finally, the calculated contour points are transformed in coordinates and output as initial inner membrane contour lines connected by line segments, as shown. Figure 9d As shown.

[0143] S209: Fit the initial intima contour line in the ultrasound image using the vector field convolution active contour model to obtain the final lumen intima contour line;

[0144] S210: Add the final luminal endothelial contour to the ultrasound image.

[0145] The vascular intima contour extraction method provided in this application combines fuzzy clustering and an active contour model to extract the vascular intima contour. First, the ultrasound image of the intravascular cross-section is preprocessed. Then, a fuzzy C-means clustering algorithm based on gray-level histograms is used to find the initial intima contour line. Finally, a vector field convolution active contour model is used to evolve and fit the vascular intima of the ultrasound image of the intravascular cross-section based on the initial intima contour line. Therefore, the vascular intima contour extraction method provided in this application improves the accuracy of vascular intima contour extraction by finding the initial intima contour line based on fuzzy C-means clustering of gray-level histograms and accurately fitting the vascular intima of the ultrasound image of the intravascular cross-section based on the initial intima contour line using a vector field convolution active contour model.

[0146] This application discloses a method for extracting the intima contour of a blood vessel lumen. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution. Specifically:

[0147] See Figure 10 A flowchart illustrating another method for extracting the intima contour of a blood vessel lumen according to an exemplary embodiment is shown below. Figure 10 As shown, it includes:

[0148] S301: Acquire an ultrasound image of the cross-section inside the blood vessel, and preprocess the ultrasound image to obtain the target image;

[0149] S302: Morphological processing is used to perform erosion reconstruction on the target image to obtain an intermediate image, and clustering parameters are determined; wherein, the clustering parameters include the number of cluster centers, kernel processing scale, and fuzziness factor;

[0150] S303: Determine different gray values ​​as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers, and determine the gray value membership function based on the fuzziness factor and the Euclidean clustering of the gray values ​​in the gray-level histogram with the initial cluster centers;

[0151] S304: Iteratively update the initial cluster centers and the gray value membership function until the fuzzy C-mean objective function is optimized or the number of iterations reaches a preset number;

[0152] S305: The maximum value in the updated cluster center is determined as the second gray threshold. The gray values ​​of all pixels in the intermediate image whose gray values ​​are greater than or equal to the second gray threshold are set to the upper limit value, and the gray values ​​of all pixels whose gray values ​​are less than the second gray threshold are set to the lower limit value, so as to segment the target image into different cluster blocks.

[0153] S306: Determine the area and center position of each cluster block, and determine the cluster block with the largest area and the closest distance between its center position and the image center position as the lumen wall region;

[0154] S307: Calculate the Euclidean distance between the region center position of all the clustered blocks and the region center position of the lumen wall region;

[0155] S308: Calculate the overlap area between the bounding rectangles corresponding to all the clustered blocks and the bounding rectangles corresponding to the lumen wall region;

[0156] S309: The clustered blocks whose Euclidean distance is greater than a preset distance threshold and whose overlapping area is less than a first preset area threshold are identified as external noise in the lumen wall region, and the external noise is removed;

[0157] S310: Perform polar coordinate transformation on the image after removing external noise to obtain the first polar coordinate image;

[0158] S311: Determine the area of ​​each cluster block, the average gray value of the pixels, and the number of pixels overlapping with the lumen wall region in the first polar coordinate image;

[0159] S312: The clustered blocks whose area is less than a second preset area threshold, whose average gray value is greater than a third gray value threshold, and whose number of overlapping pixels is greater than a preset number are identified as internal noise of the cavity wall region, and the internal noise is removed.

[0160] S313: Determine the maximum lateral width of the lumen wall region, and determine whether the ratio of the maximum lateral width to the width of the image after removing external and internal noise is greater than a preset value; if yes, proceed to S315; if no, proceed to S314.

[0161] S314: Update the clustering parameters and re-enter S303.

[0162] It should be noted that due to the complexity of coronary arteries, the resulting IVUS images exhibit diverse shapes and are subject to significant noise and artifact interference. Therefore, the clustering parameters need to be adjusted according to different tissue shapes. In this embodiment, after extracting the final lumen wall region, its maximum lateral width is determined. The ratio of this maximum lateral width to the width of the sixth intermediate image is then checked against a preset value. If the ratio is greater than a preset value, it indicates good fuzzy clustering, effectively segmenting the vascular lumen region and facilitating subsequent initial intimal contour localization. If not, the clustering parameters need to be updated, and operations S303 to S316 are repeated. The preset value is not limited here; for example, it can be set to 0.7.

[0163] S315: Perform polar coordinate transformation on the image after removing external and internal noise to obtain a second polar coordinate image, and take the first pixel with a non-zero gray value in each column of the first polar coordinate image as a contour point.

[0164] S316: Use interpolation to fill in the missing parts of the lines connecting the contour points, and perform rectangular coordinate transformation to obtain the initial inner membrane contour line;

[0165] S317: Fit the initial intima contour line in the ultrasound image using the vector field convolution active contour model to obtain the final lumen intima contour line;

[0166] S318: Add the final luminal endothelial contour to the ultrasound image.

[0167] Therefore, this embodiment discloses a clustering parameter update strategy and a fuzzy clustering effect judgment criterion, which improves the accuracy of fuzzy clustering in extracting the initial intima contour, thereby improving the accuracy of vascular lumen intima contour extraction.

[0168] The following describes a vascular lumen endometrial contour extraction device provided in the embodiments of this application. The vascular lumen endometrial contour extraction device described below and the vascular lumen endometrial contour extraction method described above can be referred to each other.

[0169] See Figure 11 A structural diagram of a vascular lumen endometrial contour extraction device according to an exemplary embodiment is shown, as follows: Figure 11 As shown, it includes:

[0170] The preprocessing module 100 is used to acquire ultrasound images of the intravascular cross-section and preprocess the ultrasound images to obtain a target image.

[0171] The determining module 200 is used to determine the lumen wall region in the target image and determine the initial intima contour line corresponding to the lumen wall region;

[0172] The fitting module 300 is used to fit the initial intima contour line to obtain the final lumen intima contour line.

[0173] Add module 400 is used to add the final endothelial contour to the ultrasound image.

[0174] Therefore, the vascular lumen intima contour extraction device provided in this application first finds the initial intima contour line, and then accurately fits the intima of the vascular lumen in the ultrasound image of the vascular cross section, thereby improving the accuracy of vascular lumen intima contour extraction.

[0175] Based on the above embodiments, as a preferred embodiment, the preprocessing module 100 is specifically a module that acquires an ultrasound image of the intravascular cross-section and performs downsampling and filtering processing on the ultrasound image;

[0176] Accordingly, the fitting module 300 is specifically a module that fits the initial intima contour line to obtain a fitted lumen intima contour line, and performs upsampling processing on the fitted lumen intima contour line to obtain the final lumen intima contour line.

[0177] Based on the above embodiments, as a preferred embodiment, the preprocessing module 100 is specifically a module that acquires an ultrasound image of the intravascular cross section and removes catheter effects from the image.

[0178] Based on the above embodiments, as a preferred embodiment, the preprocessing module 100 includes:

[0179] Acquisition unit, used to acquire ultrasound images of the intravascular cross-section;

[0180] The first determining unit is used to determine the first pixel with a non-zero gray value in each direction starting from the center position of the image as the target position;

[0181] The calculation unit is used to calculate the average distance between each target location and the center location of the image, and to calculate the sum of the average distance and the preset offset distance as the superimposed distance;

[0182] The setting unit is used to set the grayscale values ​​of all pixels whose distance from the center of the image is less than the superposition distance to the lower limit value.

[0183] Based on the above embodiments, as a preferred embodiment, the preprocessing module 100 is specifically a module that acquires an ultrasound image of the intravascular cross-section and significantly enhances the region of interest in the image.

[0184] Based on the above embodiments, as a preferred embodiment, the preprocessing module 100 is specifically a module that acquires an ultrasound image of the intravascular cross-section and normalizes the gray values ​​of the pixels in the image.

[0185] Based on the above embodiments, as a preferred embodiment, the determining module 200 includes:

[0186] The segmentation submodule is used to determine the category of each pixel in the target image so as to segment the target image into different cluster blocks;

[0187] The first determining submodule is used to determine the lumen wall region in the cluster block and remove external and internal noise from the lumen wall region;

[0188] The second determining submodule is used to determine the initial inner membrane contour line corresponding to the lumen wall region.

[0189] Based on the above embodiments, as a preferred implementation, the segmentation submodule is specifically a unit that performs clustering operations on the pixels in the target image using a fuzzy C-means clustering algorithm based on grayscale histograms, so as to segment the target image into different cluster blocks.

[0190] Based on the above embodiments, as a preferred implementation, the segmentation sub-module includes:

[0191] The reconstruction unit is used to perform erosion reconstruction on the target image using morphological processing to obtain an intermediate image and determine clustering parameters; wherein, the clustering parameters include the number of cluster centers, kernel processing scale, and blur factor;

[0192] The second determining unit is used to determine different gray values ​​as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers, and to determine the gray value membership function based on the fuzzing factor and the Euclidean clustering of the gray values ​​in the gray-level histogram with the initial cluster centers.

[0193] The update unit is used to iteratively update the initial cluster centers and the gray value membership function until the fuzzy C-mean objective function is optimized or the number of iterations reaches a preset number.

[0194] The setting unit is used to determine the maximum value in the updated cluster center as the second grayscale threshold, and to set the grayscale values ​​of all pixels in the intermediate image whose grayscale values ​​are greater than or equal to the second grayscale threshold to the upper limit value, and the grayscale values ​​of all pixels whose grayscale values ​​are less than the second grayscale threshold to the lower limit value, so as to segment the target image into different cluster blocks.

[0195] Based on the above embodiments, as a preferred implementation, the first determining submodule includes:

[0196] The third determining unit is used to determine the area and center position of each cluster block, and to determine the cluster block with the largest area and the closest distance between the center position and the image center position as the lumen wall region.

[0197] The noise removal unit is used to remove external and internal noise from the lumen wall region.

[0198] Based on the above embodiments, as a preferred embodiment, the removal unit includes:

[0199] A calculation subunit is used to calculate the Euclidean distance between the region center position of all the clustered blocks and the region center position of the lumen wall region;

[0200] The statistical subunit is used to calculate the overlapping area between the bounding rectangles corresponding to all the clustered blocks and the bounding rectangles corresponding to the lumen wall region.

[0201] The first removal subunit is used to identify clusters whose Euclidean distance is greater than a preset distance threshold and whose overlapping area is less than a first preset area threshold as external noise in the lumen wall region, and to remove the external noise.

[0202] Based on the above embodiments, as a preferred embodiment, the removal unit includes:

[0203] The transformation subunit is used to perform polar coordinate transformation on the image after removing external noise to obtain the first polar coordinate image;

[0204] A subunit is defined for determining the region area, average gray value of pixels, and number of pixels overlapping with the lumen wall region for each cluster in the first polar coordinate image.

[0205] The second removal subunit is used to identify clustered blocks whose area is less than a second preset area threshold, whose average gray value is greater than a third gray value threshold, and whose number of overlapping pixels is greater than a preset number as internal noise of the cavity wall region, and to remove the internal noise.

[0206] Based on the above embodiments, as a preferred implementation, the second determining submodule includes:

[0207] The conversion unit is used to perform polar coordinate transformation on the image after removing external and internal noise to obtain a second polar coordinate image, and to take the first non-zero gray value pixel in each column of the first polar coordinate image as a contour point.

[0208] The completion unit is used to complete the missing parts of the connection line of the contour points using interpolation and perform rectangular coordinate transformation to obtain the initial inner membrane contour line.

[0209] Based on the above embodiments, as a preferred implementation, the second determining submodule further includes:

[0210] The judgment unit is used to determine the maximum lateral width of the lumen wall region and to determine whether the ratio of the maximum lateral width to the width of the image after removing external and internal noise is greater than a preset value; if yes, the workflow of the conversion unit is started; if no, the clustering parameters are updated and the workflow of the second determination unit is restarted.

[0211] Based on the above embodiments, as a preferred embodiment, the fitting module 300 is specifically a module that uses a vector field convolution active contour model to fit the initial intima contour line to obtain the final lumen intima contour line.

[0212] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0213] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiments of this application, the embodiments of this application also provide an ultrasonic device. Figure 12 This is a structural diagram illustrating an ultrasonic device according to an exemplary embodiment, such as... Figure 12 As shown, the ultrasound equipment includes:

[0214] Communication interface 1 enables information exchange with other devices, such as network devices;

[0215] Processor 2, connected to communication interface 1, enables information exchange with other devices and, when running a computer program, executes the vascular lumen intima contour extraction method provided by one or more of the above-mentioned technical solutions. The computer program is stored in memory 3.

[0216] Display 4 is used to display an ultrasound image of a cross-section inside the blood vessel and the intima outline in the ultrasound image.

[0217] Of course, in practical applications, the various components of the ultrasonic equipment are coupled together through bus system 5. It can be understood that bus system 5 is used to achieve communication and connection between these components. In addition to the data bus, bus system 5 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 12 The general will label all buses as Bus System 5.

[0218] The memory 3 in this embodiment is used to store various types of data to support the operation of the ultrasound device. Examples of such data include any computer program used to operate the ultrasound device.

[0219] It is understood that memory 3 can be volatile memory or non-volatile memory, or both. 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), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. 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 Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 3 described in the embodiments of this application is intended to include, but is not limited to, these and any other suitable types of memory.

[0220] The methods disclosed in the embodiments of this application can be applied to processor 2, or implemented by processor 2. Processor 2 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 2 or by instructions in the form of software. The processor 2 may be a general-purpose processor, DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 2 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in memory 3. Processor 2 reads the program in memory 3 and completes the steps of the aforementioned method in combination with its hardware.

[0221] When processor 2 executes the program, it implements the corresponding processes in the various methods of the embodiments of this application. For the sake of brevity, these will not be described in detail here.

[0222] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a memory 3 that stores a computer program, which can be executed by a processor 2 to complete the steps described in the aforementioned method. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM.

[0223] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0224] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an ultrasound device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0225] The above description is merely a specific embodiment 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. A method for extracting the intima contour of a blood vessel lumen, characterized in that, include: Acquire an ultrasound image of the cross-section inside the blood vessel, and preprocess the ultrasound image to obtain the target image; In the target image, a lumen wall region is determined, and an initial intima contour line corresponding to the lumen wall region is determined. The initial intima profile is fitted to obtain the final lumen intima profile; The final endothelial contour is added to the ultrasound image; The step of preprocessing the ultrasound image to obtain the target image includes: Remove the duct effect from the image; The removal of the duct effect in the image includes: The target location is determined by identifying the first non-zero gray value pixel in each direction starting from the center of the image. Calculate the average distance between each target location and the center location of the image, and calculate the sum of the average distance and the preset offset distance as the superimposed distance; Set the grayscale values ​​of all pixels whose distance from the center of the image is less than the superposition distance to the lower limit value.

2. The method for extracting the endothelial contour of a blood vessel lumen according to claim 1, characterized in that, The preprocessing of the ultrasound image to obtain the target image includes: The ultrasound image is downsampled and filtered; Accordingly, fitting the initial intima profile to obtain the final luminal intima profile includes: The initial intima profile is fitted to obtain the fitted lumen intima profile, and the fitted lumen intima profile is upsampled to obtain the final lumen intima profile.

3. The method for extracting the endothelial contour of a blood vessel lumen according to claim 1, characterized in that, The preprocessing of the ultrasound image to obtain the target image includes: Saliency enhancement is performed on regions of interest in the image.

4. The method for extracting the vascular intima contour according to claim 3, characterized in that, Saliency enhancement of regions of interest in an image includes: Determine the region of interest in the image; wherein the gray value of each pixel in the region of interest is greater than a first gray value threshold; Calculate the average pixel value of all pixels in the region of interest, and determine the mapping curve based on the average pixel value; The grayscale value of each pixel in the image is processed using the mapping curve.

5. The method for extracting the endothelial contour of a blood vessel lumen according to claim 1, characterized in that, The preprocessing of the ultrasound image to obtain the target image includes: Normalize the grayscale values ​​of pixels in the image.

6. The method for extracting the endothelial contour of a blood vessel lumen according to claim 1, characterized in that, Determining the lumen wall region in the target image includes: The category of each pixel in the target image is determined in order to segment the target image into different cluster blocks; The lumen wall region is identified in the cluster block, and external and internal noise of the lumen wall region is removed.

7. The method for extracting the intima contour of a blood vessel lumen according to claim 6, characterized in that, The step of determining the category of each pixel in the target image to segment the target image into different cluster blocks includes: The fuzzy C-means clustering algorithm based on grayscale histograms performs clustering operations on the pixels in the target image to segment the target image into different cluster blocks.

8. The method for extracting the endothelial contour of a blood vessel lumen according to claim 7, characterized in that, The fuzzy C-means clustering algorithm based on grayscale histograms performs clustering operations on the pixels in the target image to segment the target image into different cluster blocks, including: Morphological processing is used to perform erosion reconstruction on the target image to obtain an intermediate image, and clustering parameters are determined; wherein, the clustering parameters include the number of cluster centers, kernel processing scale, and fuzziness factor; Different gray values ​​are determined as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers. The gray value membership function is determined based on the fuzziness factor and the Euclidean distance between the gray values ​​in the gray-level histogram and the initial cluster centers. The initial cluster centers and the gray value membership function are iteratively updated until the fuzzy C-mean objective function is optimized or the number of iterations reaches a preset number. The maximum value in the updated cluster centers is determined as the second grayscale threshold. The grayscale values ​​of all pixels in the intermediate image whose grayscale values ​​are greater than or equal to the second grayscale threshold are set to the upper limit value, and the grayscale values ​​of all pixels whose grayscale values ​​are less than the second grayscale threshold are set to the lower limit value, so as to segment the target image into different cluster blocks.

9. The method for extracting the intima contour of a blood vessel lumen according to claim 6, characterized in that, Determining the lumen wall region within the clustered blocks includes: The area and center position of each cluster are determined, and the cluster with the largest area and the closest distance between its center position and the center position of the image is determined as the lumen wall region.

10. The method for extracting the intima contour of a blood vessel lumen according to claim 6, characterized in that, Removing external noise from the lumen wall region includes: Calculate the Euclidean distance between the region center positions of all the clustered blocks and the region center positions of the lumen wall region; Calculate the overlap area between the bounding rectangles corresponding to all the clustered blocks and the bounding rectangles corresponding to the lumen wall region; Clusters whose Euclidean distance is greater than a preset distance threshold and whose overlapping area is less than a first preset area threshold are identified as external noise in the lumen wall region, and the external noise is removed.

11. The method for extracting the intima contour of a blood vessel lumen according to claim 10, characterized in that, Removing internal noise from the lumen wall region includes: The first polar coordinate image is obtained by performing polar coordinate transformation on the image after removing external noise. In the first polar coordinate image, determine the region area of ​​each cluster block, the average gray level of the pixels, and the number of pixels overlapping with the lumen wall region; Clustered blocks whose area is less than a second preset area threshold, whose average grayscale value is greater than a third grayscale threshold, and whose number of overlapping pixels is greater than a preset number are identified as internal noise in the cavity wall region, and the internal noise is removed.

12. The method for extracting the intima contour of a blood vessel lumen according to claim 6, characterized in that, Determining the initial intima contour line corresponding to the lumen wall region includes: The image after removing external and internal noise is transformed into polar coordinates to obtain a second polar coordinate image. The first pixel with a non-zero gray value in each column of the second polar coordinate image is taken as the contour point. The missing parts of the lines connecting the contour points are filled in using interpolation, and a Cartesian coordinate transformation is performed to obtain the initial inner membrane contour line.

13. The method for extracting the intima contour of a blood vessel lumen according to claim 8, characterized in that, After removing external and internal noise from the lumen wall region, the method further includes: Determine the maximum lateral width of the lumen wall region, and determine whether the ratio of the maximum lateral width to the width of the image after removing external and internal noise is greater than a preset value; If so, proceed to the step of performing polar coordinate transformation on the image after removing external and internal noise to obtain a second polar coordinate image; If not, update the clustering parameters and re-enter the step of determining different gray values ​​as initial cluster centers based on the range of the gray-level histogram of the intermediate image and the number of cluster centers.

14. The method for extracting the endothelial contour of a blood vessel lumen according to claim 1 or 2, characterized in that, The fitting of the initial intima contour includes: The initial inner membrane contour is fitted using a vector field convolution active contour model.

15. A device for extracting the intima contour of a blood vessel lumen, characterized in that, include: The preprocessing module is used to acquire ultrasound images of the intravascular cross-section and preprocess the ultrasound images to obtain the target image. The determination module is used to determine the lumen wall region in the target image and determine the initial intima contour line corresponding to the lumen wall region; The fitting module is used to fit the initial intima contour line to obtain the final lumen intima contour line. An add module is used to add the final endothelial contour to the ultrasound image; Specifically, the preprocessing module is a module that acquires ultrasound images of the intravascular cross-section and removes catheter effects from the images. The preprocessing module includes: Acquisition unit, used to acquire ultrasound images of the intravascular cross-section; The first determining unit is used to determine the first pixel with a non-zero gray value in each direction starting from the center position of the image as the target position; The calculation unit is used to calculate the average distance between each target location and the center location of the image, and to calculate the sum of the average distance and the preset offset distance as the superimposed distance; The setting unit is used to set the grayscale values ​​of all pixels whose distance from the center of the image is less than the superposition distance to the lower limit value.

16. An ultrasonic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the vascular lumen intima contour extraction method as described in any one of claims 1 to 14 when executing the computer program; A display for showing an ultrasound image of a cross-section within a blood vessel and the intima outline in the ultrasound image.

17. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the vascular lumen intima contour extraction method as described in any one of claims 1 to 14.

Images