A blood vessel centerline correction method, device, storage medium and equipment

Abstract

The application provides a blood vessel center line correction method and device, a storage medium and equipment. The method is applied to the technical field of blood vessel center line extraction. The method obtains an abnormal blood vessel mask area in a target blood vessel based on a blood vessel reconstruction view of the target blood vessel, performs abnormal type determination based on morphological features of the abnormal blood vessel mask area, and then corrects the abnormal center line of the abnormal blood vessel mask area based on a corresponding correction strategy. After the initial center line extraction of the blood vessel, the method increases the blood vessel abnormality determination process, assigns different correction strategies to different abnormal types, and corrects the abnormal center line caused by the blood vessel abnormality in a targeted manner, thereby improving the efficiency of the center line correction and the accuracy of the center line extraction.

Description

Technical Field

[0001] This application relates to the field of vascular centerline extraction technology, and more specifically, to a method, apparatus, storage medium and device for vascular centerline correction in the field of vascular centerline extraction technology. Background Technology

[0002] With the development of medical imaging technology, vascular analysis has become a key task in clinical medicine. Centerline extraction, as an important step in vascular analysis, provides a benchmark framework. However, in clinical practice, centerlines extracted based on traditional algorithms such as topology refinement and distance transformation often deviate from the actual lumen center or have pathological errors when dealing with complex pathological structures such as vascular adhesions or aneurysms. Related techniques rely on manual correction of the extracted centerline by physicians, resulting in low efficiency and high subjectivity. Summary of the Invention

[0003] This application provides a method, apparatus, storage medium, and device for correcting the centerline of blood vessels. The method can automatically correct the centerline, improving the efficiency of centerline correction and the accuracy of centerline extraction.

[0004] In a first aspect, a method for correcting the centerline of a blood vessel is provided. The method includes: acquiring a reconstructed view of a target blood vessel; acquiring an abnormal blood vessel mask region of the target blood vessel based on the reconstructed view; determining the abnormal type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region; and correcting the abnormal centerline of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormal type. The abnormal centerline is the centerline determined by extracting the centerline of the abnormal blood vessel mask region.

[0005] The above technical solution obtains abnormal vessel mask regions within the target vessel based on a reconstructed vascular view. Anomaly types are determined based on the morphological characteristics of these mask regions, and then the abnormal centerlines of these mask regions are corrected using corresponding correction strategies. This method, after initial centerline extraction, adds a vessel anomaly determination process, assigning different correction strategies to different anomaly types. This targeted correction of abnormal centerlines caused by vascular anomalies improves both the efficiency of centerline correction and the accuracy of centerline extraction.

[0006] In conjunction with the first aspect, in some possible implementations, the step of obtaining the abnormal blood vessel mask region of the target blood vessel based on the blood vessel reconstruction view includes: determining the abnormal blood vessel segment of the target blood vessel based on the blood vessel reconstruction view; mapping the abnormal blood vessel segment to the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region.

[0007] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the step of determining the abnormality type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region includes: if the morphological characteristics of the abnormal blood vessel mask region are the characteristics of a closed loop after bifurcation, then the abnormality type of the abnormal blood vessel mask region is determined to be the straight segment adhesion type.

[0008] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the step of determining the abnormality type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region includes: if the abnormal blood vessel mask region does not include loop features, then obtaining the cross-sectional area change curve of the blood vessel in the abnormal blood vessel mask region; if the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then determining the convex blood vessel region corresponding to the single-peak shape cross-sectional area change curve segment in the abnormal blood vessel mask region; obtaining the three-dimensional convexity measure of the convex blood vessel region; if the three-dimensional convexity measure is greater than the convexity threshold, then determining the abnormality type of the abnormal blood vessel mask region as an aneurysm type; if the three-dimensional convexity measure is less than or equal to the convexity threshold, then determining the abnormality type of the abnormal blood vessel mask region as a tortuous segment adhesion type.

[0009] Combining the first aspect and the above implementation methods, in some possible implementation methods, the step of correcting the abnormal centerline of the abnormal vessel mask region based on the correction strategy corresponding to the abnormal type includes: if the abnormal type is a straight segment adhesion type, then based on each voxel point in the first abnormal centerline of the abnormal vessel mask region, a first inscribed ball sequence is generated in the abnormal vessel mask region; a first diameter change curve of the first inscribed ball sequence is obtained; the first diameter change curve is matched with a second diameter change curve to determine the abnormal curve segment in the first diameter change curve that does not match the second diameter change curve, the second diameter change curve is the diameter change curve of the second inscribed ball sequence, and the second inscribed ball sequence is the inscribed ball sequence in the normal vessel mask region adjacent to the abnormal vessel mask region in the target vessel; a second abnormal centerline corresponding to the abnormal curve segment in the first abnormal centerline is obtained; the second abnormal centerline is removed from the first abnormal centerline to obtain the corrected target centerline.

[0010] Combining the first aspect and the above implementation methods, in some possible implementation methods, the step of correcting the abnormal centerline of the abnormal vessel mask region based on the correction strategy corresponding to the abnormality type includes: if the abnormality type is a tortuous segment adhesion type, then based on each voxel point in the third abnormal centerline of the abnormal vessel mask region, a third inlined ball sequence is generated in the abnormal vessel mask region; in the third inlined ball sequence, a fourth inlined ball sequence is determined corresponding to the single-peak segment in the cross-sectional area change curve, the single-peak segment including the rising line segment and the falling line segment in the cross-sectional area change curve; in the third inlined ball sequence... In this sequence, the upstream sequence of the fourth inner ball sequence is determined as the fifth inner ball sequence, and the downstream sequence of the fourth inner ball sequence is determined as the sixth inner ball sequence; the sphere adjacent to the fourth inner ball sequence in the fifth inner ball sequence is determined as the first sphere, and the sphere adjacent to the fourth inner ball sequence in the sixth inner ball sequence is determined as the second sphere; the first sphere is controlled to roll along the vessel wall opposite to the adhesion area in the abnormal vessel mask region until the first sphere overlaps with the second sphere; the third abnormal center line of the abnormal vessel mask region is corrected based on the first center line of the first sphere during the rolling process.

[0011] In conjunction with the first aspect, in some possible implementations, the step of correcting the abnormal centerline of the abnormal vessel mask region based on the correction strategy corresponding to the abnormality type includes: if the abnormality type is an aneurysm, then based on each voxel point in the fourth abnormal centerline of the abnormal vessel mask region, a seventh inline sequence is generated in the abnormal vessel mask region; in the seventh inline sequence, an eighth inline sequence is determined corresponding to the single-peak segment in the cross-sectional area change curve, the single-peak segment including the rising and falling segments in the cross-sectional area change curve; in the seventh inline sequence, The upstream sequence of the eighth inner ball sequence is determined as the ninth inner ball sequence, and the downstream sequence of the eighth inner ball sequence is determined as the tenth inner ball sequence; the sphere adjacent to the eighth inner ball sequence in the ninth inner ball sequence is determined as the third sphere, and the sphere adjacent to the eighth inner ball sequence in the tenth inner ball sequence is determined as the fourth sphere; the third sphere is controlled to roll along the vessel wall opposite to the aneurysm in the abnormal vessel mask region until the third sphere overlaps with the fourth sphere; the abnormal center line of the abnormal vessel mask region is corrected based on the second sphere center line of the third sphere during the rolling process.

[0012] Secondly, a vascular centerline correction device is provided, the device comprising: An abnormal region determination unit is used to obtain a vascular reconstruction view of the target blood vessel and obtain the abnormal blood vessel mask region of the target blood vessel based on the vascular reconstruction view. An anomaly type determination unit is used to determine the anomaly type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region. The correction unit is used to correct the abnormal center line of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormal type. The abnormal center line is the center line determined by center line extraction of the abnormal blood vessel mask region.

[0013] Thirdly, a computer device is provided, the computer device comprising: a memory for storing executable program code; A processor for calling and running executable program code from memory to perform the methods in the first aspect or any possible implementation of the first aspect described above.

[0014] Fourthly, a computer program product is provided, comprising: computer program code, which, when run on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0015] Fifthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation thereof. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a scenario for a method of correcting the centerline of a blood vessel provided in an embodiment of this application; Figure 2 This is a flowchart illustrating a method for correcting the centerline of a blood vessel provided in an embodiment of this application; Figure 3 This is a flowchart illustrating a method for correcting the centerline of a blood vessel provided in an embodiment of this application; Figure 4 This is a schematic diagram illustrating an example of a blood vessel with a straight segment adhesion, provided in an embodiment of this application. Figure 5 This is a flowchart illustrating a method for correcting the centerline of a blood vessel provided in an embodiment of this application; Figure 6 This is an example schematic diagram of a cross-sectional area variation curve provided in an embodiment of this application; Figure 7 This is an example schematic diagram of a convex vascular region provided in an embodiment of this application; Figure 8 This is a schematic diagram illustrating an example of a tortuous segment adhesion type of blood vessel provided in an embodiment of this application; Figure 9 This is a flowchart illustrating a method for correcting the centerline of a blood vessel provided in an embodiment of this application; Figure 10This is a schematic diagram illustrating an example of an aneurysm vessel provided in an embodiment of this application; Figure 11 This is a schematic diagram of the structure of a vascular centerline correction device provided in an embodiment of this application; Figure 12 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0018] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0019] The vascular centerline correction method provided in this application embodiment is applicable to post-processing scenarios of centerline extraction. If the blood vessel has complex pathological structures such as vascular adhesion or aneurysm, it may cause the extracted centerline to deviate from the real lumen center or have path topology errors, requiring correction of the extracted initial centerline.

[0020] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating a method for correcting the centerline of blood vessels according to an embodiment of this application. Figure 1 As shown, the blue dashed line represents the correct centerline, while the red dashed line represents the abnormal centerline caused by vascular pathological structures. Vascular pathological structures include adhesions in tortuous segments at bifurcation points, adhesions in straight segments at non-bifurcation points, and aneurysms. The centerline extracted from a straight segment adhesion area will have an extra red portion, resulting in a loop shape that does not conform to the vascular morphology characteristics; the centerline extracted from a tortuous segment adhesion area will deviate significantly from the outer wall of the vessel. An aneurysm refers to a localized abnormal dilation or bulging of the arterial wall. Due to the bulging of the vessel wall, the centerline extracted from the aneurysm area will deviate significantly from the side of the bulging vessel wall.

[0021] To address the aforementioned issues, this application provides a method for correcting the centerline of a blood vessel. The method involves obtaining an abnormal blood vessel mask region from a reconstructed view of the target blood vessel, determining the abnormality type based on the morphological characteristics of the mask region, and then correcting the abnormal centerline of the mask region using a corresponding correction strategy. This method, after initial centerline extraction, adds a blood vessel abnormality determination process, assigning different correction strategies to different abnormality types. This targeted correction of abnormal centerlines caused by blood vessel abnormalities improves the efficiency of centerline correction and the accuracy of centerline extraction.

[0022] based on Figure 1 The scene diagram shown below will be combined with... Figures 2-10 This application provides a detailed description of the vascular centerline correction method provided in the embodiments.

[0023] Please see Figure 2 , Figure 2 This is a schematic flowchart of a method for correcting the centerline of blood vessels provided in an embodiment of this application. Figure 2 As shown, the method in this application embodiment may include the following steps S101-S103.

[0024] S101, Obtain the vascular reconstruction view of the target blood vessel, and obtain the abnormal blood vessel mask region of the target blood vessel based on the vascular reconstruction view; Specifically, the process involves acquiring a vascular image of the target blood vessel, performing binarized segmentation on the image to obtain a segmentation mask for the target blood vessel, determining a reconstructed view of the target blood vessel based on the segmentation mask, and then identifying abnormal vascular segments within the reconstructed view to determine the abnormal vascular mask regions in the acquired segmentation mask.

[0025] The vascular images refer to three-dimensional vascular images such as computed tomography angiography (CTA) or magnetic resonance angiography (MRA). Binarized vascular segmentation refers to classifying each voxel in the vascular image, marking which voxels belong to blood vessels and which belong to the background. Voxels are points in three-dimensional space. The vascular segmentation mask is the image obtained by binary partitioning voxels in the vascular image based on the gray-level differences between blood vessels and the background.

[0026] Vascular reconstruction views refer to three-dimensional visualization models obtained by transforming vascular structures in vascular image post-processing techniques. These models are used to visualize the morphology, orientation, and pathological features of blood vessels. Vascular reconstruction views include Curved Planar Reformation (CPR) views and Multi-Planar Reformation (MPR) views. A CPR view involves "cutting and flattening" the vessel along its centerline onto a two-dimensional plane. Obtaining a CPR view requires extracting the centerline from the vessel segmentation mask to obtain an initial centerline. The CPR view is then obtained based on the vessel segmentation mask and the initial centerline. An MPR view involves cutting the vessel segmentation mask along any plane (axial, sagittal, coronal, or oblique) to generate a two-dimensional cross-sectional image. The initial centerline is the vascular topology obtained by extracting the centerline of the vessel portion from the vessel segmentation mask, representing the width of a single pixel. The centerline is obtained based on the centerline extraction algorithm, which is used to iteratively "erode" the blood vessel segmentation mask until the blood vessel segmentation mask is reduced to a skeleton with the width of a single element. The centerline extraction algorithm includes, but is not limited to, algorithms such as the two-dimensional thinning algorithm (Zhang-Suen) and the three-dimensional thinning algorithm (Palágyi) that thin objects with width into a set of lines of single elements.

[0027] S102, Determine the abnormality type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region; Specifically, the abnormality type of the abnormal vessel mask region is determined based on its morphological characteristics. Abnormality types in blood vessels include straight segment adhesions, tortuous segment adhesions, and aneurysms. Straight segment adhesions will result in loop morphology within the vessel; tortuous segment adhesions will cause adhesions at the tortuous segments corresponding to bifurcation points; and aneurysms will cause protrusions at the straight segments or tortuous segments corresponding to bifurcation points, forming spherical protrusions.

[0028] S103, Correct the abnormal center line of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormal type. The abnormal center line is the center line determined by extracting the center line of the abnormal blood vessel mask region.

[0029] Specifically, in this application embodiment, different correction strategies are set for different abnormality types. After determining the abnormality type of the abnormal blood vessel mask region, the abnormal center line of the abnormal blood vessel mask region is corrected based on the correction strategy corresponding to the abnormality type. The abnormal center line is the center line determined by extracting the center line of the abnormal blood vessel mask region.

[0030] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological features of the mask region, and then the abnormal centerline of the mask region is corrected using a corresponding correction strategy. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. This targeted correction of abnormal centerlines caused by vessel abnormalities improves the efficiency of centerline correction and the accuracy of centerline extraction.

[0031] Please see Figure 3 , Figure 3 This is a schematic flowchart of a method for correcting the centerline of blood vessels provided in an embodiment of this application. Figure 3 As shown, the method in this application embodiment may include the following steps S201-S207.

[0032] S201, Obtain a vascular reconstruction view of the target blood vessel, and determine the abnormal vascular segments of the target blood vessel based on the vascular reconstruction view; Specifically, the process involves acquiring vascular images of the target blood vessel, performing binarized segmentation on the images to obtain a segmentation mask for the target blood vessel, determining a reconstructed view of the target blood vessel based on the segmentation mask, and then identifying abnormal vascular segments based on the reconstructed view.

[0033] The vascular images refer to three-dimensional vascular images such as CTA or MRA images. Binarized vascular segmentation refers to classifying each voxel in the vascular image, marking which voxels belong to blood vessels and which belong to the background. Voxels are points in three-dimensional space. The vascular segmentation mask is an image obtained by binary partitioning voxels in the vascular image based on the gray-level differences between blood vessels and the background.

[0034] Vascular reconstruction views refer to three-dimensional visualization models obtained by transforming vascular structures in vascular image post-processing techniques. These models are used to visualize the morphology, orientation, and pathological features of blood vessels. Vascular reconstruction views include CPR views and MPR views. CPR views involve "cutting and flattening" the blood vessel along its centerline onto a two-dimensional plane. Obtaining a CPR view requires extracting the centerline from the vascular segmentation mask to obtain an initial centerline. The CPR view is then obtained based on the vascular segmentation mask and the initial centerline. MPR views involve cutting the vascular segmentation mask along any plane (axial, sagittal, coronal, or oblique) to generate a two-dimensional cross-sectional image. The initial centerline is the vascular topology obtained by extracting the centerline from the vascular portion of the vascular segmentation mask, representing the width of a single pixel. The centerline is obtained based on a centerline extraction algorithm, which iteratively "erodes" the vascular segmentation mask until only a skeleton of single pixel width remains. Centerline extraction algorithms include, but are not limited to, algorithms such as the Zhang-Suen algorithm and the Palágyi algorithm, which refine objects with width into a set of single pixel lines.

[0035] It should be noted that the vascular centerline correction device or computer device provided in this application embodiment is equipped with a deep learning model. By inputting multiple sample vascular reconstruction views with different types of vascular abnormalities and normal sample vascular reconstruction views into the deep learning model, the deep learning model is trained so that it learns the distinguishing features between normal and abnormal vascular segments and outputs the abnormal vascular segments in the vascular reconstruction view.

[0036] In one feasible implementation, since the vascular reconstruction view is a visual output, the embodiments of this application can also output the vascular reconstruction view after obtaining the data vascular reconstruction view, so as to receive the doctor's annotations on abnormal vascular segments in the vascular reconstruction view.

[0037] S202, Map the abnormal blood vessel segment to the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region; Specifically, the abnormal blood vessel segment is a three-dimensional visualization in the blood vessel reconstruction view. The abnormal blood vessel segment is mapped to the blood vessel segmentation mask of the target blood vessel, and the blood vessel mask area in the blood vessel segmentation mask that corresponds to the abnormal blood vessel segment is determined as the abnormal blood vessel mask area.

[0038] In one feasible implementation, if the vessel reconstruction view is an MPR view, centerline extraction is not required when acquiring the MPR view. Therefore, after obtaining the abnormal vessel mask region, centerline extraction is performed on the abnormal vessel mask region to determine the abnormal centerline within the abnormal vessel mask region. If the vessel reconstruction view is a CPR view, centerline extraction is performed on the vessel segmentation mask when acquiring the CPR view. While mapping the abnormal vessel segment to the vessel segmentation mask, the abnormal vessel segment is also mapped to the initial centerline to obtain the abnormal centerline corresponding to the abnormal vessel mask region.

[0039] S203, If the morphological characteristics of the abnormal blood vessel mask region are those of a closed loop after bifurcation, then the abnormality type of the abnormal blood vessel mask region is determined to be the straight segment adhesion type. Specifically, the morphological features of the abnormal vessel mask region are obtained. If the morphological features are a loop characteristic of first bifurcation and then closure, then the abnormality type of the abnormal vessel mask region is determined to be a straight segment adhesion type. For example... Figure 1 As shown in the straight segment adhesion type, similar to the vascular morphology characteristics, the abnormal center line in the abnormal vascular mask area of ​​the straight segment adhesion type also has the loop feature of first bifurcation and then closure.

[0040] S204, If the abnormality type is straight segment adhesion type, then based on each voxel point in the first abnormal center line in the abnormal blood vessel mask region, generate the first inscribed ball sequence in the abnormal blood vessel mask region. Specifically, when the abnormality type is straight segment adhesion, the abnormal center line in the abnormal blood vessel mask region is the first abnormal center line. Based on each voxel point in the first abnormal center line, a first inscribed sphere sequence is generated in the abnormal blood vessel mask region. Each sphere in the first inscribed sphere sequence is the largest inscribed sphere in the blood vessel, and the center of each sphere is a voxel point in the first abnormal center line.

[0041] S205, Obtain the first diameter variation curve of the first inscribed ball sequence; Specifically, the diameter of each sphere in the first inscribed sphere sequence is obtained, and a first diameter variation curve is generated according to the arrangement order of the spheres in the first inscribed sphere sequence. Please refer to [link to relevant documentation]. Figure 4 , Figure 4 This is an example schematic diagram of a blood vessel with a straight segment adhesion provided in an embodiment of this application. For example... Figure 4 As shown, Figure 4 The red dashed line in the diagram represents the center line of the first anomaly. The inscribed sphere corresponding to the red dashed line is a gray sphere, which has a different diameter than the blue inscribed sphere corresponding to the blue dashed line. Therefore, according to... Figure 4 From the first inscribed sphere sequence, it can be seen that the diameter of the sphere in the vascular adhesion segment is different from the diameter of the sphere in the other vascular segments.

[0042] S206, Match the first diameter change curve with the second diameter change curve to identify the abnormal curve segments in the first diameter change curve that do not match the second diameter change curve; Specifically, the first diameter change curve is matched with the second diameter change curve to identify abnormal curve segments in the first diameter change curve that do not match the second diameter change curve. The second diameter change curve is the diameter change curve of the second inscribed ball sequence, which is the inscribed ball sequence in the normal blood vessel mask region adjacent to the abnormal blood vessel mask region in the target blood vessel. The diameter of the largest inscribed ball in the normal blood vessel mask region remains stable within a certain range. Therefore, the second diameter change curve can be approximated as a straight line segment, while the first diameter change curve is a curve containing changing line segments. The straight line segments in the first diameter change curve are the same as those in the second diameter curve. The changing line segments in the first diameter change curve are the abnormal curve segments that do not match the second diameter change curve.

[0043] S207, obtain the second abnormal center line corresponding to the abnormal curve segment in the first abnormal center line, remove the second abnormal center line from the first abnormal center line, and obtain the corrected target center line.

[0044] Specifically, the abnormal curve segment in the first diameter change curve corresponds to the inscribed ball sequence of the vascular adhesion segment. The center line of the vascular adhesion segment is the second abnormal center line in the first abnormal center line. The method for correcting the center line of the straight segment adhesion is to separate the adhesion area, that is, to remove the second abnormal center line of the vascular adhesion segment from the first abnormal center line to obtain the corrected target center line.

[0045] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological characteristics of the mask region. When the abnormality type is a straight segment adhesion, the second abnormal centerline of the adhesion segment in the mask region is removed from the first abnormal centerline, thus achieving centerline correction. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. This targeted correction of abnormal centerlines caused by vessel abnormalities improves the efficiency of centerline correction and the accuracy of centerline extraction.

[0046] Please see Figure 5 , Figure 5 This is a schematic flowchart of a method for correcting the centerline of blood vessels provided in an embodiment of this application. Figure 5 As shown, the method in this application embodiment may include the following steps S301-S311.

[0047] S301, Obtain a vascular reconstruction view of the target blood vessel, and determine the abnormal vascular segments of the target blood vessel based on the vascular reconstruction view; Please refer to step S201 for the specific process, which will not be repeated here.

[0048] S302, Map the abnormal blood vessel segment to the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region; Please refer to step S202 for the specific process, which will not be repeated here.

[0049] S303, If the abnormal blood vessel mask region does not include loop features, then obtain the cross-sectional area change curve of the blood vessel in the abnormal blood vessel mask region. Specifically, the morphological features of the abnormal vessel mask region are obtained. If these morphological features do not include loop features, the cross-sectional area variation region of the vessel within the abnormal vessel mask region is obtained. If the morphological features of the abnormal vessel mask region do not include loop features, the abnormal type of the abnormal vessel mask region is determined to be either a tortuous segment adhesion type or an aneurysm. Both tortuous segment adhesion type vessels and aneurysm vessels include abnormal convexity features. The cross-sectional area of ​​the vessel segment corresponding to the abnormal convexity feature is greater than or equal to the cross-sectional area of ​​its adjacent vessel segment. Therefore, the cross-sectional area variation curve is obtained.

[0050] S304. If the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then determine the convex blood vessel region corresponding to the single-peak cross-sectional area change curve segment in the abnormal blood vessel mask region. Please see Figure 6 , Figure 6 This is an example schematic diagram of a cross-sectional area variation curve provided in an embodiment of this application. For example... Figure 6 As shown, if the cross-sectional area change curve exhibits a single-peak shape that first rises and then falls, the vessel corresponding to the single-peak segment is identified as a convex vessel region corresponding to abnormal convexity characteristics. Both aneurysm types and tortuous segment adhesion types include convex vessel regions. The cross-sectional area of ​​a convex vessel region is larger than that of a normal vessel region. Therefore, in this embodiment, the convex vessel region can be determined based on the cross-sectional area change curve in the abnormal vessel region, and the centerline of the convex vessel region can then be corrected.

[0051] S305, Obtain the three-dimensional convexity measurement of the convex vascular region; Specifically, the three-dimensional convexity metric is a dimensionless geometric feature used to quantify the degree of convexity of a three-dimensional object's shape. In this embodiment, the three-dimensional convexity metric is defined as the ratio of the actual volume of a convex blood vessel region to the volume of its smallest convex portion.

[0052] Please see Figure 7 , Figure 7This is an example schematic diagram of a convex vascular region provided in an embodiment of this application, such as... Figure 7 As shown, the convex vascular region in aneurysms is a spherical protrusion with a clear boundary between it and the vessel wall. Therefore, the convex vascular region in aneurysms consists of the smallest protruding part of the aneurysm and the vessel portion connected to it. The volume of the convex vascular region is larger than the volume of the protruding part, and its three-dimensional convexity metric is greater than 1. In contrast, the convex vascular region in tortuous adhesion types is its smallest protruding part. The convex vascular region may or may not contain concave portions; therefore, its three-dimensional convexity metric is less than or equal to 1.

[0053] S306, If the three-dimensional convexity measure is less than or equal to the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be the tortuous segment adhesion type. Specifically, the convexity threshold is 1. If the three-dimensional convexity metric is less than or equal to 1, the abnormality type of the abnormal blood vessel mask region is determined to be the tortuous segment adhesion type. It should be noted that in this embodiment, the abnormal blood vessel sample regions with straight segment adhesion labels, tortuous segment adhesion labels, and aneurysm labels are pre-input into the deep learning model to train the deep learning model to learn the features of various abnormality types. After determining that the abnormal blood vessel type is tortuous segment adhesion, the adhesion regions in the abnormal blood vessel mask region are marked according to the morphological characteristics of the tortuous segment adhesion type so as to distinguish the adhesion regions during subsequent centerline correction.

[0054] Straight segment adhesions refer to adhesions in the straight segment of a blood vessel, exhibiting a closed loop characteristic after bifurcation. Aneurysms are pathological protrusions on the surface of the blood vessel wall, spherically shaped, with a clear boundary and complete lumen structure. They are connected to the vessel wall through local attachment and do not structurally fuse with other vascular segments, exhibiting morphological characteristics independent of the vessel itself. Tortuous segment adhesions refer to adhesions where the walls of two different spatial segments of a blood vessel are too close to each other, causing structural fusion at the bifurcation point, forming an abnormal protrusion.

[0055] S307, If the abnormality type is the curved segment adhesion type, then based on each voxel point in the third abnormal center line in the abnormal blood vessel mask region, generate a third inline ball sequence in the abnormal blood vessel mask region. Specifically, when the abnormality type is the tortuous segment adhesion type, the abnormal center line in the abnormal blood vessel mask region is the third abnormal center line. Based on each voxel point in the third abnormal center line, a third inscribed sphere sequence is generated in the abnormal blood vessel mask region. Each sphere in the third inscribed sphere sequence is the largest inscribed sphere in the blood vessel, and the center of each sphere is a voxel point in the third abnormal center line.

[0056] S308, determine the fourth inner ball sequence corresponding to the single-peak segment in the cross-sectional area change curve in the third inner ball sequence; Specifically, the single-peak segment in the cross-sectional area change curve provided in step S303 is determined. This single-peak segment corresponds to the convex vessel region in the abnormal vessel mask region. Due to vascular adhesion, the maximum inscribed ball diameter of the convex vessel region is greater than or equal to the maximum inscribed ball diameter of the adjacent vessel segment. The fourth inscribed ball sequence corresponding to the single-peak segment is determined in the third inscribed ball sequence, which is the maximum inscribed ball sequence in the convex vessel region. The single-peak segment includes both the ascending and descending segments in the cross-sectional area change curve.

[0057] S309, In ​​the third inside ball receiving sequence, the upstream sequence of the fourth inside ball receiving sequence is determined as the fifth inside ball receiving sequence, and the downstream sequence of the fourth inside ball receiving sequence is determined as the sixth inside ball receiving sequence; S310, the sphere in the fifth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is determined as the first sphere, and the sphere in the sixth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is determined as the second sphere; Please see Figure 8 , Figure 8 This is an example schematic diagram of a tortuous segment adhesion type of blood vessel provided in an embodiment of this application. For example... Figure 8 As shown, in the third inline sphere sequence, the upstream sequence of the fourth inline sphere sequence is designated as the fifth inline sphere sequence, and the downstream sequence of the fourth inline sphere sequence is designated as the sixth inline sphere sequence. Both the fifth and sixth inline sphere sequences are inline sphere sequences of adjacent vascular segments in convex vascular regions, where the diameter of each sphere is the same as that of a sphere in a normal vascular region. The sphere adjacent to the fourth inline sphere in the fifth inline sphere sequence is designated as the first sphere, and the sphere adjacent to the fourth inline sphere sequence in the sixth inline sphere sequence is designated as the second sphere.

[0058] S311, control the first sphere to roll along the blood vessel wall opposite to the adhesion area in the abnormal blood vessel mask area until the first sphere overlaps with the second sphere, and correct the third abnormal center line of the abnormal blood vessel mask area based on the first sphere center line during the rolling process.

[0059] Specifically, such as Figure 8 As shown, the red dashed line represents the third abnormal center line, and the blue dashed line represents the corrected target center line. In the abnormal vessel mask region of the curved segment adhesion type, the third abnormal center line is more biased towards the vessel wall of the adhesion region compared to the normal center line. The first sphere is controlled to roll along the vessel wall opposite to the adhesion region in the abnormal vessel mask region until the first sphere overlaps with the second sphere. The first line formed by the first sphere during the rolling process is determined as the expected normal center line for the convex vessel region. Based on this normal center line, the third abnormal center line is corrected to obtain the corrected target center line.

[0060] In one feasible implementation, if the single-peak morphology of the cross-sectional area change curve exhibits a sub-single-peak morphology with a large slope change, then the abnormal vessel masking region is determined to be a vessel abnormality crossing type where both tortuous segment adhesion and aneurysm types coexist, and an aneurysm appears in the convex vessel region corresponding to the tortuous segment adhesion type. For this type of vessel abnormality crossing, this application embodiment performs two corrections on the abnormal vessel masking region: first, the centerline of the tortuous segment adhesion type is corrected to obtain a first target centerline; then, the centerline of the aneurysm type is corrected on the first target centerline to obtain a second target centerline.

[0061] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological characteristics of the mask region. When the abnormality type is a tortuous adhesion, the largest inscribed sphere of the normal vessel segment within the mask region is controlled to roll along the vessel wall in the opposite direction to the adhesion region. The expected normal centerline at the abnormal convexity is determined based on the line connecting the centers of the spheres during the rolling process, thus achieving centerline correction. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. It specifically corrects the abnormal centerline caused by vessel abnormalities, improving the efficiency of centerline correction and the accuracy of centerline extraction.

[0062] Please see Figure 9 , Figure 9 This is a schematic flowchart of a method for correcting the centerline of blood vessels provided in an embodiment of this application. Figure 9 As shown, the method in this application embodiment may include the following steps S401-S411.

[0063] S401, Obtain a vascular reconstruction view of the target blood vessel, and determine the abnormal vascular segments of the target blood vessel based on the vascular reconstruction view; Please refer to step S301 for the specific process, which will not be repeated here.

[0064] S402, map the abnormal blood vessel segment to the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region; Please refer to step S302 for the specific process, which will not be repeated here.

[0065] S403, If the abnormal blood vessel mask region does not include loop features, then obtain the cross-sectional area change curve of the blood vessel in the abnormal blood vessel mask region. Please refer to step S303 for the specific process, which will not be repeated here.

[0066] S404. If the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then determine the convex blood vessel region corresponding to the single-peak cross-sectional area change curve segment in the abnormal blood vessel mask region. Please refer to step S304 for the specific process, which will not be repeated here.

[0067] S405, obtain the three-dimensional convexity measurement of the convex vascular region; Please refer to step S305 for the specific process, which will not be repeated here.

[0068] S406, If the three-dimensional convexity measure is greater than the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be an aneurysm type; Specifically, the convexity threshold is 1. If the three-dimensional convexity metric is greater than 1, the abnormality type of the abnormal blood vessel mask region is determined to be an aneurysm. It should be noted that in this embodiment, the abnormal blood vessel sample regions with labels for straight segment adhesion, curved segment adhesion, and aneurysm are pre-input into a deep learning model to train the deep learning model to learn the features of various abnormality types. After determining that the abnormal blood vessel type is an aneurysm, the aneurysms in the abnormal blood vessel mask region are marked according to the morphological characteristics of the aneurysm type so as to distinguish aneurysms during subsequent centerline correction.

[0069] Straight segment adhesions refer to adhesions in the straight segment of a blood vessel, exhibiting a closed loop characteristic after bifurcation. Aneurysms are pathological protrusions on the surface of the blood vessel wall, spherically shaped, with a clear boundary and complete lumen structure. They are connected to the vessel wall through local attachment and do not structurally fuse with other vascular segments, exhibiting morphological characteristics independent of the vessel itself. Tortuous segment adhesions refer to adhesions where the walls of two different spatial segments of a blood vessel are too close to each other, causing structural fusion at the bifurcation point, forming an abnormal protrusion.

[0070] S407, If the abnormality type is aneurysm, then based on each voxel point in the fourth abnormal center line in the abnormal vessel mask region, generate the seventh inline ball sequence in the abnormal vessel mask region. Specifically, when the abnormality type is aneurysm, the abnormal center line in the abnormal vessel mask region is the fourth abnormal center line. Based on each voxel point in the fourth abnormal center line, a seventh inline sphere sequence is generated in the abnormal vessel mask region. Each sphere in the seventh inline sphere sequence is the largest inline sphere in the vessel, and the center of each sphere is a voxel point in the fourth abnormal center line.

[0071] S408, determine the eighth inner ball sequence corresponding to the single-peak segment in the cross-sectional area change curve in the seventh inner ball sequence; Specifically, the single-peak segment in the cross-sectional area change curve provided in step S403 is determined. The single-peak segment corresponds to the convex vessel region in the abnormal vessel mask region. Due to the aneurysm protrusion, the maximum inscribed ball diameter of the convex vessel region is greater than or equal to the maximum inscribed ball diameter of the adjacent vessel segment. The eighth inscribed ball sequence corresponding to the single-peak segment is determined in the seventh inscribed ball sequence, which is the maximum inscribed ball sequence in the convex vessel region. The single-peak segment includes the rising and falling segments in the cross-sectional area change curve.

[0072] S409, In ​​the seventh inside ball receiving sequence, the upstream sequence of the eighth inside ball receiving sequence is determined as the ninth inside ball receiving sequence, and the downstream sequence of the eighth inside ball receiving sequence is determined as the tenth inside ball receiving sequence. S410, the sphere in the ninth inner ball sequence that is adjacent to the eighth inner ball sequence is identified as the third sphere, and the sphere in the tenth inner ball sequence that is adjacent to the eighth inner ball sequence is identified as the fourth sphere; Please see Figure 10 , Figure 10 This is a schematic diagram illustrating an example of an aneurysm vessel provided in an embodiment of this application. Figure 10 As shown, in the seventh inline sphere sequence, the upstream sequence of the eighth inline sphere sequence is designated as the ninth inline sphere sequence, and the downstream sequence of the eighth inline sphere sequence is designated as the tenth inline sphere sequence. Both the ninth and tenth inline sphere sequences are inline sphere sequences of adjacent vascular segments in convex vascular regions, and the diameter of each sphere is the same as that of a sphere in a normal vascular region. The sphere adjacent to the eighth inline sphere in the ninth inline sphere sequence is the third sphere, and the sphere adjacent to the eighth inline sphere sequence in the tenth inline sphere sequence is the fourth sphere.

[0073] S411, control the third sphere to roll along the vessel wall opposite to the aneurysm in the abnormal vessel mask region until the third sphere overlaps with the fourth sphere, and correct the abnormal center line of the abnormal vessel mask region based on the line connecting the centers of the second spheres during the rolling process of the third sphere.

[0074] Specifically, such as Figure 10 As shown, the red dashed line represents the fourth abnormal center line, and the blue dashed line represents the corrected target center line. In the abnormal vessel mask region corresponding to the aneurysm, the fourth abnormal center line, compared to the normal center line, is more biased towards the vessel wall on the side where the aneurysm is located. The third sphere is controlled to roll along the vessel wall on the opposite side of the aneurysm within the abnormal vessel mask region until the third sphere overlaps with the fourth sphere. The second line connecting the third spheres during the rolling process is determined as the expected normal center line for the convex vessel region. Based on this normal center line, the fourth abnormal center line is corrected to obtain the corrected target center line.

[0075] In one feasible implementation, if the single-peak morphology of the cross-sectional area change curve exhibits a sub-single-peak morphology with a large slope change, then the abnormal vessel masking region is determined to be a vessel abnormality crossing type where both tortuous segment adhesion and aneurysm types coexist, and an aneurysm appears in the convex vessel region corresponding to the tortuous segment adhesion type. For this type of vessel abnormality crossing, this application embodiment performs two corrections on the abnormal vessel masking region: first, the centerline of the tortuous segment adhesion type is corrected to obtain a first target centerline; then, the centerline of the aneurysm type is corrected on the first target centerline to obtain a second target centerline.

[0076] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological characteristics of the mask region. When the abnormality is an aneurysm, the largest inscribed sphere of the normal vessel segment within the mask region is controlled to roll along the vessel wall in the opposite direction to the abnormal convexity. The desired normal centerline at the abnormal convexity is determined based on the line connecting the centers of the spheres during the rolling process, thus achieving centerline correction. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. It specifically corrects the abnormal centerline caused by vessel abnormalities, improving the efficiency of centerline correction and the accuracy of centerline extraction.

[0077] based on Figure 1 The following is a scene illustration, which will be combined with... Figure 11 This application provides a detailed description of the vascular centerline correction device provided in its embodiments. It should be noted that... Figure 11 The vascular centerline correction device in the present application is used to perform the present application. Figures 2-10 The methods shown in the embodiments are for illustrative purposes only, illustrating the parts relevant to the embodiments of this application. For specific technical details not disclosed, please refer to this application. Figures 2-10 The example shown.

[0078] Please see Figure 11 , Figure 11 This is a schematic diagram of the structure of a vascular centerline correction device provided in an embodiment of this application. Figure 11 As shown, the vascular centerline correction device 1 of this application embodiment may include: an abnormal area determination unit 11, an abnormal type determination unit 12, and a correction unit 13.

[0079] The abnormal region determination unit 11 is used to obtain a vascular reconstruction view of the target blood vessel and obtain the abnormal blood vessel mask region of the target blood vessel based on the vascular reconstruction view. Anomaly type determination unit 12 is used to determine the anomaly type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region; The correction unit 13 is used to correct the abnormal center line of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormal type. The abnormal center line is the center line determined by extracting the center line of the abnormal blood vessel mask region.

[0080] Optionally, the abnormal region determination unit 11 is specifically used to determine the abnormal vascular segment of the target vascular vessel based on the vascular reconstruction view; The abnormal blood vessel segment is mapped onto the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region.

[0081] Optionally, the abnormality type determination unit 12 is specifically used to determine the abnormality type of the abnormal blood vessel mask region as a straight segment adhesion type if the morphological characteristics of the abnormal blood vessel mask region are the characteristics of a closed loop after bifurcation.

[0082] Optionally, the anomaly type determination unit 12 is specifically used to obtain the cross-sectional area change curve of the blood vessels in the abnormal blood vessel mask area if the abnormal blood vessel mask area does not include loop features. If the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then the convex blood vessel region corresponding to the single-peak cross-sectional area change curve segment is determined in the abnormal blood vessel mask region. Obtain a three-dimensional convexity measurement of the convex vascular region; If the three-dimensional convexity measure is greater than the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be an aneurysm. If the three-dimensional convexity metric is less than or equal to the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be the tortuous segment adhesion type.

[0083] Optionally, the correction unit 13 is specifically used to generate a first inscribed ball sequence in the abnormal blood vessel mask region based on each voxel point in the first abnormal center line in the abnormal blood vessel mask region if the abnormality type is a straight segment adhesion type. Obtain the first diameter variation curve of the first inscribed ball sequence; The first diameter change curve is matched with the second diameter change curve to identify the abnormal curve segment in the first diameter change curve that does not match the second diameter change curve. The second diameter change curve is the diameter change curve of the second inline sphere sequence. The second inline sphere sequence is the inline sphere sequence in the normal blood vessel mask region adjacent to the abnormal blood vessel mask region in the target blood vessel. Obtain the second abnormal centerline corresponding to the abnormal curve segment in the first abnormal centerline; The second abnormal center line is removed from the first abnormal center line to obtain the corrected target center line.

[0084] Optionally, the correction unit 13 is specifically used to generate a third inscribed ball sequence in the abnormal blood vessel mask region based on each voxel point in the third abnormal center line in the abnormal blood vessel mask region if the abnormality type is the curved segment adhesion type. In the third inscribed ball sequence, determine the fourth inscribed ball sequence corresponding to the single-peak segment in the cross-sectional area change curve. The single-peak segment includes the rising and falling segments in the cross-sectional area change curve. In the third inside ball receiving sequence, the upstream sequence of the fourth inside ball receiving sequence is determined as the fifth inside ball receiving sequence, and the downstream sequence of the fourth inside ball receiving sequence is determined as the sixth inside ball receiving sequence. The sphere in the fifth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the first sphere, and the sphere in the sixth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the second sphere. Control the first sphere to roll along the vessel wall opposite to the adhesion area in the abnormal vessel mask region until the first sphere overlaps with the second sphere; The third abnormal center line of the abnormal blood vessel mask region is corrected based on the line connecting the centers of the first sphere during the rolling process.

[0085] Optionally, the correction unit 13 is specifically used to generate a seventh inline ball sequence in the abnormal blood vessel mask region based on each voxel point in the fourth abnormal center line in the abnormal blood vessel mask region if the abnormality type is an aneurysm type. In the seventh inbound ball sequence, determine the eighth inbound ball sequence corresponding to the single-peak segment in the cross-sectional area change curve. The single-peak segment includes the rising and falling segments in the cross-sectional area change curve. In the seventh inside ball receiving sequence, the upstream sequence of the eighth inside ball receiving sequence is determined as the ninth inside ball receiving sequence, and the downstream sequence of the eighth inside ball receiving sequence is determined as the tenth inside ball receiving sequence. The sphere in the ninth inner ball sequence that is adjacent to the eighth inner ball sequence is designated as the third sphere, and the sphere in the tenth inner ball sequence that is adjacent to the eighth inner ball sequence is designated as the fourth sphere. Control the third sphere to roll along the vessel wall opposite to the aneurysm in the abnormal vessel mask region until the third sphere overlaps with the fourth sphere; The abnormal center line of the abnormal blood vessel mask region is corrected based on the line connecting the centers of the second sphere during the rolling process of the third sphere.

[0086] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological features of the mask region, and then the abnormal centerline of the mask region is corrected using a corresponding correction strategy. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. This targeted correction of abnormal centerlines caused by vessel abnormalities improves the efficiency of centerline correction and the accuracy of centerline extraction.

[0087] Please see Figure 12 , Figure 12 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0088] For example, such as Figure 12 As shown, the computer device 1200 includes a processor 1201 and a memory 1202, wherein the processor 1201 and the memory 1202 are electrically connected.

[0089] Processor 1201 is the control center of computer device 1200 and may include one or more processing cores. Processor 1201 connects to various parts of the computer device using various interfaces and lines. By running or calling computer programs stored in memory 1202 and calling data stored in memory 1202, it executes various functions of the computer device and processes data, thereby providing overall control of computer device 1200. Optionally, processor 1201 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 1201 may integrate one or more of the following: CPU, Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user page, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into processor 1201 and may be implemented separately through a communication chip.

[0090] The memory 1202 can be used to store software programs and modules. The processor 1201 executes various functional applications and data processing by running the computer programs and modules stored in the memory 1202. The memory 1202 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, computer programs required for at least one function, etc.; the data storage area may store data created based on the use of the computer device 1200, etc.

[0091] Furthermore, memory 1202 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, memory 1202 may also include a memory controller to provide processor 1201 with access to memory 1202.

[0092] In this embodiment, the processor 1201 in the computer device 1200 loads the instructions corresponding to the processes of one or more computer programs into the memory 1202 according to the following steps, and the processor 1201 runs the computer programs stored in the memory 1202 to realize various functions, as follows: Obtain the initial centerline and vessel segmentation mask of the target vessel, and determine the vessel reconstruction view of the target vessel based on the initial centerline and vessel segmentation mask; Obtain the abnormal vascular mask region of the target vascular vessel based on the vascular reconstruction view; The abnormality type of the abnormal blood vessel mask region is determined based on its morphological characteristics. The abnormal centerline of the abnormal blood vessel mask region is corrected based on the correction strategy corresponding to the abnormal type. The abnormal centerline is the centerline determined by extracting the centerline of the abnormal blood vessel mask region.

[0093] Optionally, when the processor 1201 executes the operation of obtaining the abnormal vessel mask region of the target vessel based on the vessel reconstruction view, it specifically performs the following operations: Identify abnormal vascular segments of the target vessel based on the reconstructed vascular view; The abnormal blood vessel segment is mapped onto the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region.

[0094] Optionally, when the processor 1201 determines the anomaly type of the abnormal blood vessel mask region based on its morphological features, it specifically performs the following operations: If the morphological characteristics of the abnormal blood vessel mask region are those of a closed loop after bifurcation, then the abnormality type of the abnormal blood vessel mask region is determined to be the straight segment adhesion type.

[0095] Optionally, when the processor 1201 determines the anomaly type of the abnormal blood vessel mask region based on its morphological features, it specifically performs the following operations: If the abnormal blood vessel mask region does not include loop features, then obtain the cross-sectional area change curve of the blood vessel in the abnormal blood vessel mask region. If the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then the convex blood vessel region corresponding to the single-peak cross-sectional area change curve segment is determined in the abnormal blood vessel mask region. Obtain a three-dimensional convexity measurement of the convex vascular region; If the three-dimensional convexity measure is greater than the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be an aneurysm. If the three-dimensional convexity metric is less than or equal to the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be the tortuous segment adhesion type.

[0096] Optionally, when the processor 1201 performs the following operations to correct the abnormal centerline of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormality type: If the abnormality type is a straight segment adhesion type, then based on each voxel point in the first abnormal center line in the abnormal blood vessel mask region, a first inscribed ball sequence is generated in the abnormal blood vessel mask region. Obtain the first diameter variation curve of the first inscribed ball sequence; The first diameter change curve is matched with the second diameter change curve to identify the abnormal curve segment in the first diameter change curve that does not match the second diameter change curve. The second diameter change curve is the diameter change curve of the second inline sphere sequence. The second inline sphere sequence is the inline sphere sequence in the normal blood vessel mask region adjacent to the abnormal blood vessel mask region in the target blood vessel. Obtain the second abnormal centerline corresponding to the abnormal curve segment in the first abnormal centerline; The second abnormal center line is removed from the first abnormal center line to obtain the corrected target center line.

[0097] Optionally, when the processor 1201 performs the following operations to correct the abnormal centerline of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormality type: If the abnormality type is the curved segment adhesion type, then based on each voxel point in the third abnormal center line in the abnormal blood vessel mask region, a third inline ball sequence is generated in the abnormal blood vessel mask region. In the third inscribed ball sequence, determine the fourth inscribed ball sequence corresponding to the single-peak segment in the cross-sectional area change curve. The single-peak segment includes the rising and falling segments in the cross-sectional area change curve. In the third inside ball receiving sequence, the upstream sequence of the fourth inside ball receiving sequence is determined as the fifth inside ball receiving sequence, and the downstream sequence of the fourth inside ball receiving sequence is determined as the sixth inside ball receiving sequence. The sphere in the fifth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the first sphere, and the sphere in the sixth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the second sphere. Control the first sphere to roll along the vessel wall opposite to the adhesion area in the abnormal vessel mask region until the first sphere overlaps with the second sphere; The third abnormal center line of the abnormal blood vessel mask region is corrected based on the line connecting the centers of the first sphere during the rolling process.

[0098] Optionally, when the processor 1201 performs the following operations to correct the abnormal centerline of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormality type: If the abnormality type is an aneurysm, then a seventh inline ball sequence is generated in the abnormal vessel mask region based on each voxel point in the fourth abnormal center line of the abnormal vessel mask region. In the seventh inbound ball sequence, determine the eighth inbound ball sequence corresponding to the single-peak segment in the cross-sectional area change curve. The single-peak segment includes the rising and falling segments in the cross-sectional area change curve. In the seventh inside ball receiving sequence, the upstream sequence of the eighth inside ball receiving sequence is determined as the ninth inside ball receiving sequence, and the downstream sequence of the eighth inside ball receiving sequence is determined as the tenth inside ball receiving sequence. The sphere in the ninth inner ball sequence that is adjacent to the eighth inner ball sequence is designated as the third sphere, and the sphere in the tenth inner ball sequence that is adjacent to the eighth inner ball sequence is designated as the fourth sphere. Control the third sphere to roll along the vessel wall opposite to the aneurysm in the abnormal vessel mask region until the third sphere overlaps with the fourth sphere; The abnormal center line of the abnormal blood vessel mask region is corrected based on the line connecting the centers of the second sphere during the rolling process of the third sphere.

[0099] In this embodiment, an abnormal vessel mask region is obtained from the reconstructed view of the target vessel. The abnormality type is determined based on the morphological features of the mask region, and then the abnormal centerline of the mask region is corrected using a corresponding correction strategy. This method, after initial centerline extraction, adds a vessel abnormality determination process, assigning different correction strategies to different abnormality types. This targeted correction of abnormal centerlines caused by vessel abnormalities improves the efficiency of centerline correction and the accuracy of centerline extraction.

[0100] It should be understood that the device provided in this application embodiment is used to perform the above-described method for correcting the vascular centerline, and therefore can achieve the same effect as the above-described implementation method.

[0101] When using integrated units, the device may include a processing module and a storage module. When applied to a computer device, the processing module can be used to control and manage the operations of the computer device. The storage module can be used to support the computer device in executing relevant program code, etc.

[0102] The processing module may be a processor or a controller, which can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0103] In addition, the device provided in this application embodiment may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute a vascular centerline correction method provided in the above embodiment.

[0104] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the vascular centerline correction method provided in the above embodiments.

[0105] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the vascular centerline correction method provided in the above embodiment.

[0106] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0107] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0108] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0109] 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 correcting the centerline of blood vessels, characterized in that, The method includes: Obtain a reconstructed vascular view of the target blood vessel, and obtain the abnormal vascular mask region of the target blood vessel based on the reconstructed vascular view; The abnormality type of the abnormal blood vessel mask region is determined based on its morphological characteristics. The abnormal centerline of the abnormal blood vessel mask region is corrected based on the correction strategy corresponding to the abnormal type. The abnormal centerline is the centerline determined by centerline extraction of the abnormal blood vessel mask region.

2. The method according to claim 1, characterized in that, The step of obtaining the abnormal vascular mask region of the target vascular vessel based on the vascular reconstruction view includes: Based on the reconstructed vascular view, the abnormal vascular segment of the target vessel is identified; The abnormal blood vessel segment is mapped onto the blood vessel segmentation mask of the target blood vessel to obtain the abnormal blood vessel mask region.

3. The method according to claim 1, characterized in that, Determining the abnormality type of the abnormal blood vessel mask region based on its morphological features includes: If the morphological characteristics of the abnormal blood vessel mask region are those of a closed loop after bifurcation, then the abnormality type of the abnormal blood vessel mask region is determined to be the straight segment adhesion type.

4. The method according to claim 3, characterized in that, The method further includes: If the abnormal blood vessel mask region does not include loop features, then obtain the cross-sectional area change curve of the blood vessel in the abnormal blood vessel mask region; If the cross-sectional area change curve shows a single-peak shape that rises first and then falls, then the convex blood vessel region corresponding to the cross-sectional area change curve segment of the single-peak shape is determined in the abnormal blood vessel mask region. Obtain the three-dimensional convexity measurement of the convex vascular region; If the three-dimensional convexity measure is greater than the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be an aneurysm. If the three-dimensional convexity metric is less than or equal to the convexity threshold, then the abnormality type of the abnormal blood vessel mask region is determined to be the curved segment adhesion type.

5. The method according to claim 3, characterized in that, The correction strategy based on the anomaly type for correcting the abnormal centerline of the abnormal blood vessel mask region includes: If the abnormality type is the straight segment adhesion type, then based on each voxel point in the first abnormal center line in the abnormal blood vessel mask region, a first inscribed ball sequence is generated in the abnormal blood vessel mask region. Obtain the first diameter variation curve of the first inscribed ball sequence; The first diameter change curve is matched with the second diameter change curve to identify the abnormal curve segment in the first diameter change curve that does not match the second diameter change curve. The second diameter change curve is the diameter change curve of the second inscribed ball sequence. The second inscribed ball sequence is the inscribed ball sequence in the normal blood vessel mask region adjacent to the abnormal blood vessel mask region in the target blood vessel. Obtain the second abnormal center line corresponding to the abnormal curve segment in the first abnormal center line; The second abnormal center line is removed from the first abnormal center line to obtain the corrected target center line.

6. The method according to claim 4, characterized in that, The correction strategy based on the anomaly type for correcting the abnormal centerline of the abnormal blood vessel mask region includes: If the abnormality type is the curved segment adhesion type, then a third inline ball sequence is generated in the abnormal blood vessel mask region based on each voxel point in the third abnormal center line of the abnormal blood vessel mask region. In the third inscribed ball sequence, determine the fourth inscribed ball sequence corresponding to the single-peak segment in the cross-sectional area change curve, wherein the single-peak segment includes the rising line segment and the falling line segment in the cross-sectional area change curve; In the third inside ball sequence, the upstream sequence of the fourth inside ball sequence is determined as the fifth inside ball sequence, and the downstream sequence of the fourth inside ball sequence is determined as the sixth inside ball sequence; The sphere in the fifth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the first sphere, and the sphere in the sixth inscribed sphere sequence that is adjacent to the fourth inscribed sphere sequence is designated as the second sphere. Control the first sphere to roll along the blood vessel wall opposite to the adhesion area in the abnormal blood vessel mask region until the first sphere overlaps with the second sphere; The third abnormal center line of the abnormal blood vessel mask region is corrected based on the first center line of the first sphere during its rolling process.

7. The method according to claim 6, characterized in that, The method further includes: If the abnormality type is the aneurysm type, then a seventh inline ball sequence is generated in the abnormal blood vessel mask region based on each voxel point in the fourth abnormal center line of the abnormal blood vessel mask region. In the seventh inscribed ball sequence, determine the eighth inscribed ball sequence corresponding to the single-peak segment in the cross-sectional area change curve, wherein the single-peak segment includes the rising line segment and the falling line segment in the cross-sectional area change curve; In the seventh inside ball receiving sequence, the upstream sequence of the eighth inside ball receiving sequence is determined as the ninth inside ball receiving sequence, and the downstream sequence of the eighth inside ball receiving sequence is determined as the tenth inside ball receiving sequence. The sphere in the ninth inscribed sphere sequence that is adjacent to the eighth inscribed sphere sequence is designated as the third sphere, and the sphere in the tenth inscribed sphere sequence that is adjacent to the eighth inscribed sphere sequence is designated as the fourth sphere. The third sphere is controlled to roll along the vessel wall opposite to the aneurysm in the abnormal vessel mask region until the third sphere overlaps with the fourth sphere; The abnormal center line of the abnormal blood vessel mask region is corrected based on the line connecting the centers of the second sphere during the rolling process of the third sphere.

8. A vascular centerline correction device, characterized in that, The device includes: An abnormal region determination unit is used to obtain a vascular reconstruction view of a target blood vessel and obtain an abnormal blood vessel mask region of the target blood vessel based on the vascular reconstruction view. An anomaly type determination unit is used to determine the anomaly type of the abnormal blood vessel mask region based on the morphological characteristics of the abnormal blood vessel mask region. The correction unit is used to correct the abnormal center line of the abnormal blood vessel mask region based on the correction strategy corresponding to the abnormality type. The abnormal center line is the center line determined by center line extraction of the abnormal blood vessel mask region.

9. A computer device, characterized in that, The computer device includes: Memory, used to store executable program code; A processor for calling and running the executable program code from the memory, causing the computer device to perform the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computing program product includes: Computer program code, when executed, implements the method as described in any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program code that, when executed, implements the method as described in any one of claims 1 to 7.

Images