Method, device and storage medium for completing the centerline of a blood vessel segment

Abstract

The present application discloses a method, device and storage medium for completing the centerline of a blood vessel segment. The method comprises: obtaining an image containing a blood vessel to be completed, and extracting the blood vessel centerline from the image; dividing the blood vessel centerline based on blood vessel information to obtain a plurality of blood vessel segment centerlines; constructing a graph structure of each of the blood vessel segment centerlines; judging whether the plurality of blood vessel segment centerlines are missing based on the blood vessel information; and inputting the graph structure of the blood vessel segment centerline with missing information and its corresponding blood vessel information into a completion model for completion to obtain the completed blood vessel segment centerline. By using the scheme of the present application, the spatial geometric features and topological structure of the blood vessel can be extracted, and the quality of the completed blood vessel segment centerline can be improved.

Description

Technical Field

[0001] The present application generally relates to the technical field of blood vessel segment centerline completion, and more specifically, to a method, device, and computer-readable storage medium for completing a blood vessel segment centerline. Background Art

[0002] The vascular centerline is an important basis for the identification of vascular diseases and minimally invasive interventional treatment. For example, it can be used as an important basis for the identification of ischemic cerebrovascular disease in stroke. Among them, when identifying ischemic cerebrovascular disease, it is often accompanied by problems such as vascular occlusion and stenosis, which will cause the medical image to be unable to show the complete blood vessels.

[0003] At present, there are methods to complete the centerline of the vascular segment through images or point clouds, among which the image method can only extract the local features of the vascular centerline, and the point cloud method can only analyze discrete points. Therefore, the existing vascular centerline completion method cannot extract the spatial geometric features and topological structure of the blood vessel (such as the connection relationship of the nodes), which makes it impossible to accurately identify whether the blood vessel is missing, making the effect of the completed vascular segment centerline poor, affecting the recognition results of vascular diseases.

[0004] In view of this, there is an urgent need to provide a solution for completing the centerline of the blood vessel segment, so as to extract the spatial geometric characteristics and topological structure of the blood vessel and improve the quality of the completed centerline of the blood vessel segment. Summary of the invention

[0005] In order to at least solve one or more of the technical problems mentioned above, the present application proposes a solution for completing the centerline of a blood vessel segment in multiple aspects.

[0006] In a first aspect, the present application provides a method for completing a blood vessel segment centerline, comprising: acquiring an image containing a blood vessel to be completed, and extracting the blood vessel centerline from the image; dividing the blood vessel centerline based on blood vessel information to obtain a plurality of blood vessel segment centerlines; constructing a graph structure of each of the blood vessel segment centerlines; determining whether the plurality of blood vessel segment centerlines are missing based on the blood vessel information; and inputting the graph structure of the missing blood vessel segment centerline and the corresponding blood vessel information into a completion model for completion to obtain a completed blood vessel segment centerline.

[0007] In one embodiment, the blood vessel information at least includes blood vessel key point information, blood vessel length information, blood vessel topology information and blood vessel segment identification information.

[0008] In another embodiment, dividing the blood vessel centerline based on blood vessel information to obtain multiple blood vessel segment centerlines includes: identifying blood vessel key point information in the blood vessel centerline; and dividing the blood vessel centerline based on the identification result to obtain multiple blood vessel segment centerlines.

[0009] In another embodiment, constructing the graph structure of each of the vascular segment centerlines includes: extracting from each of the vascular segment centerlines the points constituting the vascular centerline and connection information between the points; and constructing the graph structure of each of the vascular segment centerlines based on the connection information between the points of the vascular centerline.

[0010] In another embodiment, constructing the graph structure of each of the vascular segment centerlines based on the points of the vascular centerline and the connection information between the points includes: adding nodes containing location information of the points to the corresponding graph structure based on the points of the vascular centerline; and adding connecting edges to the corresponding graph structure based on the connection information between the points to construct the graph structure of each of the vascular segment centerlines.

[0011] In yet another embodiment, determining whether the centerlines of the multiple blood vessel segments are missing based on the blood vessel information includes: determining whether the centerlines of the multiple blood vessel segments are disconnected or occluded based on one or more of the blood vessel key point information, the blood vessel length information, the blood vessel topology information, and the blood vessel segment identification information, so as to determine whether the centerlines of the multiple blood vessel segments are missing.

[0012] In yet another embodiment, inputting the graph structure of the missing vascular segment centerline and its corresponding vascular information into a completion model for completion to obtain the completed vascular segment centerline includes: inputting the graph structure of the missing vascular segment centerline and its corresponding vascular segment identification information into the completion model for completion and outputting the completion coordinates; and obtaining the completed vascular segment centerline according to the completion coordinates.

[0013] In another embodiment, the method further includes: obtaining a coordinate offset relative to the original coordinates of the blood vessel centerline; and restoring the completed blood vessel segment centerline to the original position of the blood vessel centerline based on the coordinate offset and the completed coordinates.

[0014] In a second aspect, the present application provides a device for completing the centerline of a blood vessel segment, comprising: a processor; and a memory storing program instructions for completing the centerline of a blood vessel segment, wherein when the program instructions are executed by the processor, the device implements multiple embodiments of the aforementioned first aspect.

[0015] In a third aspect, the present application provides a computer-readable storage medium having stored thereon computer-readable instructions for completing a centerline of a blood vessel segment. When the computer-readable instructions are executed by one or more processors, multiple embodiments of the first aspect are implemented.

[0016] Through the solution for completing the centerline of a blood vessel segment as provided above, the embodiment of the present application divides the blood vessel centerline into multiple blood vessel segment centerlines, constructs a graph structure of each blood vessel segment centerline, and inputs the graph structure of the missing blood vessel segment centerline and its corresponding blood vessel information into the completion model for completion, so as to obtain the completed blood vessel segment centerline. Based on this, the present application can extract the specific spatial geometric features and topological structure of the blood vessel through the graph ("Graph") method, so as to accurately identify the missing blood vessel segment centerline, so as to more accurately complete the blood vessel segment centerline and improve the quality of the completed blood vessel segment centerline. The blood vessel segment centerline completed by the solution of the present application can be further used to accurately identify, for example, vascular diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0018] Figure 1 is an exemplary flowchart showing a method for completing a centerline of a blood vessel segment according to an embodiment of the present application;

[0019] Figure 2 is an exemplary flowchart showing an overall process for completing a centerline of a blood vessel segment according to an embodiment of the present application;

[0020] Figure 3 is an exemplary schematic diagram showing a graph structure for constructing a centerline of a blood vessel segment according to an embodiment of the present application;

[0021] Figure 4 is an exemplary schematic diagram showing the centerline of a blood vessel segment before and after completion according to an embodiment of the present application; and

[0022] Figure 5 is an exemplary structural block diagram showing a device for completing a centerline of a blood vessel segment according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. It should be understood that the embodiments described in this specification are only partial embodiments provided by this application to facilitate a clear understanding of the solutions and to comply with legal requirements, and are not all embodiments that can implement this application. Based on the embodiments disclosed in this specification, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0024] Figure 1 FIG. 1 is an exemplary flowchart of a method 100 for completing a centerline of a blood vessel segment according to an embodiment of the present application. Figure 1 As shown in , at step 101, an image of a blood vessel to be completed is acquired, and the centerline of the blood vessel is extracted from the image of the blood vessel. In one embodiment, the aforementioned image may include but is not limited to CT angiography (Computed Tomography Angiography, "CTA") or magnetic resonance angiography (Magnetic Resonance Angiography, "MRA"). In an implementation scenario, based on the acquired image, a segmentation model (such as a U-net network model) may be used to first segment the blood vessel from the image, and then the Mesh information of the blood vessel is extracted by, for example, graphics processing software vtk, and then based on the Mesh information of the blood vessel, the blood vessel centerline is extracted using blood vessel modeling software (such as vtmk). It can be understood that the Mesh information of the aforementioned blood vessel includes data structure information connecting points on the surface of the blood vessel.

[0025] According to the above-extracted vascular centerline, at step 102, the vascular centerline is divided based on the vascular information to obtain a plurality of vascular segment centerlines. In one embodiment, the vascular information may include but is not limited to vascular key point information, vascular length information, vascular topology information, and vascular segment identification information. When dividing the vascular centerline, the vascular key point information in the vascular centerline is first identified, and then the vascular centerline is divided based on the identification result to obtain a plurality of vascular segment centerlines.

[0026] It should be understood that there are different bifurcation points or blood vessels in different positions (such as intracranial, head-to-chest diameter), and the corresponding blood vessel key point information is different. For example, taking the blood vessels of the head-to-chest diameter as an example, there may be 10 blood vessel key points on the blood vessels, such as the basilar artery ("BA"), the left common carotid artery ("L-CCA"), the right common carotid artery ("R-CCA"), the left external carotid artery ("L-ECA"), the right external carotid artery ("R-ECA"), the left internal carotid artery ("L-ICA"), the right internal carotid artery ("R-ICA"), the left vertebral artery ("L-VA"), the right vertebral artery ("R-VA") and INA. In the implementation scenario, the vascular centerline can be divided into multiple vascular segment centerlines in one or more of the following ways: "BA end point", "BA starting point", "L-CCA starting point", "L-ECA and L-ICA dividing point (extracranial)", "L-VA starting point (extracranial)", "R-CCA and INA dividing point", "R-ECA and R-ICA dividing point (extracranial)", "R-VA starting point (extracranial)", "R-ICA end point" and "L-ICA end point".

[0027] After obtaining multiple vascular segment centerlines, at step 103, a graph structure of each vascular segment centerline is constructed. In one embodiment, the connection information between the points constituting the vascular centerline is extracted from each vascular segment centerline, so as to construct the graph structure of each vascular segment centerline based on the connection information between the points of the vascular centerline. More specifically, the graph structure of each vascular segment centerline can be constructed by adding nodes containing position information of the points to the corresponding graph structure based on the points of the vascular centerline, and adding connecting edges to the corresponding graph structure based on the connection information between the points. That is, nodes containing position information of the points of the vascular segment centerline are added to the corresponding graph structure, and connecting edges are used to represent the connection relationship between the points to form the graph structure of each vascular segment centerline (for example Figure 3 shown).

[0028] Next, at step 104, it is determined whether the center lines of multiple vascular segments are missing based on the vascular information. In one embodiment, it is determined whether the center lines of multiple vascular segments are disconnected or occluded based on one or more of the vascular key point information, vascular length information, vascular topology information, and vascular segment identification information, so as to determine whether the center lines of multiple vascular segments are missing. That is, the missing of multiple vascular segment centers includes two situations: disconnection or occlusion of multiple vascular segment center lines, wherein disconnection means that the center line of the vascular segment is missing more (e.g., one segment is missing), and occlusion means that the center line of the vascular segment is missing less (e.g., a few points are missing). Taking the vascular length information as an example, when determining whether the center lines of multiple vascular segments are disconnected or occluded, it can be determined based on the length threshold of each vascular segment center line, and when it is lower than or exceeds the corresponding length threshold, it is determined that the corresponding vascular segment center line is disconnected or occluded.

[0029] For example, in an exemplary scenario, since the average length of the R-CCA vascular segment is 132 mm and the variance of the length is 30, the length threshold of the R-CCA vascular segment can be set between 42 mm and 222 mm. In this scenario, when the length of the centerline of the R-CCA vascular segment is less than 42 mm or exceeds 222 mm, it is determined that the centerline of the R-CCA vascular segment is disconnected or occluded. In addition, it can also be judged in combination with vascular key point information, vascular topology information or vascular segment identification information. Among them, the aforementioned vascular topology information includes the connection relationship between points on the centerline of the vascular segment. When the connection relationship between some points is missing, it can be considered that the centerline of the vascular segment is disconnected or occluded. The aforementioned vascular segment identification information is to identify the vascular segment to which the centerline of the vascular segment that is more likely to be disconnected or occluded belongs, so as to facilitate judgment, and the vascular segment identification information can be identified by a one-hot variable, for example, 1000000 is used to identify the left internal carotid artery ("L-ICA"), and 0100000 is used to identify the right internal carotid artery ("R-ICA").

[0030] Further, at step 105, the graph structure of the missing vascular segment centerline and the corresponding vascular information are input into the completion model for completion to obtain the completed vascular segment centerline. In one embodiment, the graph structure of the missing vascular segment centerline and the corresponding vascular segment identification information can be input into the completion model for completion and the completion coordinates can be output, and then the completed vascular segment centerline is obtained according to the completed coordinates. In an implementation scenario, the aforementioned completion model can be, for example, a trained graph convolution neural network (Graph Convolution Neural Networks, "GCNNs") model, which can generate a representation of the node vi by learning a function f and aggregating its own features Xi and the features Xj of its neighbors. In an embodiment of the present application, the coordinate information of the completed node, that is, the completed coordinates, is obtained by the completion model, thereby forming the completed vascular segment centerline, and the completed vascular segment centerline is an undirected graph consisting of multiple nodes and multiple edges.

[0031] In one embodiment, before the graph structure of the missing vascular segment centerline is input into the completion model, it can be preprocessed. Specifically, the coordinates of the points of the missing vascular segment centerline in the graph structure of the missing vascular segment centerline are normalized, that is, the coordinates of the points of the missing vascular segment centerline are unified to the range of [-1,1] to obtain the preprocessed coordinate information. In an implementation scenario, the preprocessed coordinate information can be obtained by the following formula:

[0032]

[0033] Among them, Pos new Represents the preprocessed coordinate information, Pos origin The original coordinates of the points representing the missing centerline of the vessel segment, Mean(Pos origin ) and Std(Pos origin ) respectively represent the original coordinate mean and standard deviation of the point representing the missing blood vessel segment centerline, that is, the coordinate offset relative to the original coordinate. new The blood vessel identification information is input into the completion model for completion, and the completion coordinates and the center line of the completed blood vessel segment can be obtained.

[0034] In the embodiment of the present application, the coordinate offset relative to the original coordinates of the blood vessel centerline is obtained, so as to restore the completed blood vessel segment centerline to the original position of the blood vessel centerline according to the coordinate offset and the completed coordinates. In one embodiment, the completed blood vessel segment centerline can be restored to the original position of the blood vessel centerline by the following formula:

[0035] Pos comp =os out *td(Pos origin )+Mean(Pos origin )(2)

[0036] Among them, Pos comp Indicates the coordinate information corresponding to the original position of the center line of the blood vessel segment after completion, Pos out Indicates the completed coordinates, Mean(Pos origim ) and Std(Pos origin ) represents the coordinate offset.

[0037] In combination with the above description, it can be known that the embodiment of the present application extracts the blood vessel centerline from the image, and divides the blood vessel centerline into multiple blood vessel segment centerlines based on the blood vessel information. Then, by constructing a graph structure of each blood vessel segment centerline, and inputting the graph structure of the missing blood vessel segment centerline and its corresponding blood vessel information into the completion model for completion, so as to obtain the completed blood vessel segment centerline. That is, the embodiment of the present application completes the blood vessel segment centerline through the Graph method. Based on this, the specific spatial geometric features and topological structure of the blood vessel can be extracted, so as to accurately identify the missing blood vessel segment centerline, so as to more accurately complete the blood vessel segment centerline and improve the quality of the completed blood vessel segment centerline. When the blood vessel segment centerline completed by the present application solution is subsequently used to identify vascular diseases, more accurate recognition results can be obtained.

[0038] Figure 2 FIG. 1 is an exemplary flowchart showing the overall process of completing the centerline of a blood vessel segment according to an embodiment of the present application. Figure 2 As shown in , at step 201, an image containing a blood vessel to be completed is obtained, and the blood vessel centerline is extracted from the image. The aforementioned image may be, for example, CTA or MRA. As can be seen from the foregoing, when extracting the blood vessel centerline, the blood vessel may first be segmented by, for example, a U-net network model, and then, for example, a graphics processing software vtk may be used to extract the Mesh information of the blood vessel (for example, data structure information including the connection between points on the surface of the blood vessel) from the segmented blood vessel. Further, based on the Mesh information of the aforementioned blood vessel, the blood vessel centerline may be extracted using, for example, vtmk. Based on the extracted blood vessel centerline, at step 202, the blood vessel centerline is divided into a plurality of blood vessel segment centerlines. In one embodiment, the blood vessel key point information in the blood vessel centerline may be identified so that the blood vessel centerline may be divided into a plurality of blood vessel segment centerlines based on the identified key point information.

[0039] After obtaining multiple vascular segment centerlines, at step 203, a graph structure of each vascular segment centerline is constructed. Specifically, a node containing the position information of the point of the vascular segment centerline is added to the corresponding graph structure, and the connection relationship between the points is represented by connecting edges to form a graph structure of each vascular segment centerline. The figure further shows that at step 204, it is determined whether multiple vascular segment centerlines are missing. That is, it is determined whether multiple vascular segment centerlines are disconnected or occluded. In the implementation scenario, it can be determined whether multiple vascular segment centerlines are disconnected or occluded based on one or more information such as vascular key point information, vascular length information, vascular topology information, and vascular segment identification information. As an example, a length threshold of the blood vessel can be set to determine whether the corresponding vascular segment centerline is disconnected or occluded according to the length threshold of each vascular segment centerline.

[0040] When it is determined that the centerline of the current vascular segment is not disconnected or occluded (for example, when the length of the vascular segment is within the length threshold range), there is no need to complete the centerline of the current vascular segment, and return to the aforementioned step 204 to continue to determine whether the centerline of the next vascular segment is missing. When it is determined that the centerline of the current vascular segment is disconnected or occluded (for example, when the length of the vascular segment is lower than or higher than the length threshold range), at step 205, the graph structure of the missing vascular segment centerline and its corresponding vascular information are input into the completion model for completion. As mentioned above, the completion model can be, for example, a trained GCNNs model. In an implementation scenario, before the graph structure of the missing vascular segment centerline is input into the completion model, it can be normalized based on the above formula (1) to obtain the preprocessed coordinate information. Then, the preprocessed coordinate information and vascular identification information are input into the completion model for completion, so that at step 206, the completed coordinates and the completed vascular segment centerline can be obtained.

[0041] Furthermore, at step 207, the coordinate offset relative to the original coordinates of the blood vessel centerline can also be obtained, so that at step 208, the completed blood vessel segment centerline can be restored to the original position of the blood vessel centerline according to the coordinate offset and the completed coordinates. Specifically, the coordinate information corresponding to the restored original position of the blood vessel segment centerline can be obtained by the above formula (2), so that the restored original position of the blood vessel segment centerline can be realized based on the coordinate information.

[0042] Figure 3 FIG. 1 is an exemplary schematic diagram showing a graph structure for constructing a centerline of a blood vessel segment according to an embodiment of the present application. Figure 3The left side of the figure shows a blood vessel centerline 301 extracted from a blood vessel. In the implementation scenario, the blood vessel centerline is divided based on the blood vessel information to obtain multiple blood vessel segment centerlines, for example, a blood vessel segment centerline (for example, a curve shown in a circle in the figure) 302 divided from the blood vessel centerline 301 is shown in the figure. Next, the points constituting the blood vessel segment centerline 302 are extracted from the blood vessel centerline 301 according to the blood vessel information (for example, as shown by arrow ① in the figure) and the connection information between the points constituting the blood vessel segment centerline 302 is extracted from the blood vessel centerline 301 (for example, as shown by arrow ② in the figure). After extracting the points constituting the blood vessel segment centerline and the connection information between the points, a node containing the position information of the points of the blood vessel segment centerline 302 is added to the graph structure (for example, as shown by arrow ③ in the figure), and the connection information between the points is represented by connecting edges (for example, as shown by arrow ④ in the figure), thereby forming a graph structure 303 of the blood vessel segment centerline 302. In this scenario, when it is determined that the centerline of the corresponding vascular segment is disconnected or occluded, the completed centerline of the vascular segment can be obtained by inputting the graph structure of the centerline of the vascular segment and the corresponding vascular identification information into a completion model (such as a GCNNs model).

[0043] Figure 4 is an exemplary schematic diagram showing the centerline of a blood vessel segment before and after completion according to an embodiment of the present application. Figure 4 FIG. 1 (a) shows that there is a disconnected centerline of a blood vessel segment. By inputting the graph structure of the disconnected centerline of the blood vessel segment and the blood vessel identification information into a completion model (such as a GCNNs model), the completion coordinates and the like can be obtained. Figure 4 In addition, based on the completed coordinates and the coordinate offset, the coordinate information corresponding to the restored original position of the center line of the blood vessel segment after the completion is obtained according to the above formula (2), so that the restored original position of the center line of the blood vessel segment after the completion is realized based on the coordinate information.

[0044] Figure 5 is an exemplary structural block diagram showing a device 500 for completing a centerline of a blood vessel segment according to an embodiment of the present application.

[0045] like Figure 5As shown in , the device of the present application may include a central processing unit or central processing unit ("CPU") 511, which may be a general-purpose CPU, a dedicated CPU, or other information processing and program execution unit. Further, the device 500 may also include a large-capacity memory 512 and a read-only memory ("ROM") 513, wherein the large-capacity memory 512 may be configured to store various types of data, including various images of blood vessels to be completed, blood vessel centerlines, graph structures of blood vessel segment centerlines, algorithm data, intermediate results, and various programs required to run the device 500. The ROM 513 may be configured to store data and instructions required for power-on self-test of the device 500, initialization of various functional modules in the system, drivers for basic input / output of the system, and booting the operating system.

[0046] Optionally, the device 500 may also include other hardware platforms or components, such as the tensor processing unit ("TPU") 514, graphics processing unit ("GPU") 515, field programmable gate array ("FPGA") 516, and machine learning unit ("MLU") 517 shown. It is understood that although a variety of hardware platforms or components are shown in the device 500, they are merely exemplary and not restrictive, and those skilled in the art may add or remove corresponding hardware according to actual needs. For example, the device 500 may only include a CPU, related storage devices, and interface devices to implement the method for completing the centerline of a blood vessel segment of the present application.

[0047] In some embodiments, in order to facilitate the transmission and interaction of data with an external network, the device 500 of the present application further includes a communication interface 518, so that it can be connected to a local area network / wireless local area network ("LAN / WLAN") 505 through the communication interface 518, and then connected to a local server 506 or to the Internet ("Internet") 507 through the LAN / WLAN. Alternatively or additionally, the device 500 of the present application can also be directly connected to the Internet or a cellular network through the communication interface 518 based on wireless communication technology, such as wireless communication technology based on the third generation ("3G"), the fourth generation ("4G") or the fifth generation ("5G") of the third generation ("3G"), the fourth generation ("4G") or the fifth generation ("5G") of the third generation ("3G") of the fourth generation ("4G") of the fourth generation ("4G") of the fifth generation ("5G") of the third generation ("3G") of the fourth generation ("4G") of the fourth generation ("4G") of the fifth generation ("5G") of the third generation ("3G") of the fourth generation ("4G") of the fourth generation ("4G") of the fourth generation ("5 ...

[0048] The peripheral devices of the device 500 may include a display device 502, an input device 503, and a data transmission interface 504. In one embodiment, the display device 502 may include, for example, one or more speakers and / or one or more visual displays, which are configured to perform voice prompts and / or image video display for the completion of the vascular centerline of the present application. The input device 503 may include other input buttons or controls such as a keyboard, a mouse, a microphone, a gesture capture camera, etc., which are configured to receive input of audio data and / or user instructions. The data transmission interface 504 may include, for example, a serial interface, a parallel interface or a universal serial bus interface ("USB"), a small computer system interface ("SCSI"), a serial ATA, a FireWire ("FireWire"), a PCI Express, and a high-definition multimedia interface ("HDMI"), etc., which are configured for data transmission and interaction with other devices or systems. According to the solution of the present application, the data transmission interface 504 can receive an image containing a blood vessel to be completed collected from a medical image acquisition device, and transmit to the device 500 an image of the blood vessel to be completed, a blood vessel centerline, a graph structure of a blood vessel segment centerline, or various other types of data or results.

[0049] The CPU 511, mass storage 512, ROM 513, TPU 514, GPU 515, FPGA 516, MLU 517 and communication interface 518 of the device 500 of the present application can be interconnected through a bus 519, and data interaction with peripheral devices can be achieved through the bus. In one embodiment, through the bus 519, the CPU 511 can control other hardware components in the device 500 and its peripheral devices.

[0050] Combination of the above Figure 5 The device for completing the centerline of a blood vessel segment that can be used to implement the present application is described. It should be understood that the device structure or architecture here is only exemplary, and the implementation method and implementation entity of the present application are not limited thereto, but can be changed without departing from the spirit of the present application.

[0051] According to the above description in combination with the accompanying drawings, those skilled in the art can also understand that the embodiments of the present application can also be implemented by a software program. Therefore, the present application also provides a computer-readable storage medium. The computer-readable storage medium can be used to implement the present application in combination with the accompanying drawings. Figure 1-Figure 2 A method for completing a centerline of a vessel segment is described.

[0052] It should be noted that although the operations of the method of the present application are described in a specific order in the accompanying drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted in the flow chart can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0053] It should be understood that when the terms "first", "second", "third" and "fourth" are used in the claims, the specification and the drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0054] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this application specification and claims, unless the context clearly indicates otherwise, the singular forms of "a", "an" and "the" are intended to include plural forms. It should also be further understood that the term "and / or" used in this application specification and claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0055] Although the implementation methods of the present application are as above, the contents described are only examples adopted to facilitate the understanding of the present application, and are not intended to limit the scope and application scenarios of the present application. Any technician in the technical field described in the present application can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present application, but the scope of patent protection of the present application shall still be subject to the scope defined in the attached claims.

Claims

1. A method for completing a centerline of a blood vessel segment, comprising: Acquire an image containing a blood vessel to be completed, and extract a blood vessel centerline from the image; Dividing the blood vessel centerline based on the blood vessel information to obtain a plurality of blood vessel segment centerlines; constructing a graph structure of the centerline of each of the blood vessel segments; Determining whether the center lines of the plurality of blood vessel segments are missing based on the blood vessel information; and Inputting the graph structure of the missing vascular segment centerline and its corresponding vascular information into a completion model for completion to obtain a completed vascular segment centerline, wherein the completion model is a trained graph convolutional neural network model, and the vascular information at least includes vascular key point information, vascular length information, vascular topological structure information, and vascular segment identification information; The graph structure of the missing vascular segment centerline and its corresponding vascular information are input into the completion model for completion, so as to obtain the completed vascular segment centerline including: Inputting the graph structure of the missing vascular segment centerline and its corresponding vascular segment identification information into the completion model for completion and outputting the completion coordinates; as well as Obtaining a completed centerline of the blood vessel segment according to the completed coordinates; Obtaining a coordinate offset relative to the original coordinates of the centerline of the blood vessel; and Restoring the completed blood vessel segment centerline to the original position of the blood vessel centerline according to the coordinate offset and the completed coordinates; Among them, the centerline of the completed blood vessel segment is restored to the original position of the blood vessel centerline by the following formula: Pos comp =Pos out *Std(Pos origin )+Mean(Pos origin ) Among them, Pos comp Indicates the coordinate information corresponding to the original position of the center line of the blood vessel segment after completion, Pos out Indicates the completed coordinates, Mean(Pos origin ) and Std(Pos origin ) represents the coordinate offset.

2. The method according to claim 1, wherein dividing the blood vessel centerline based on the blood vessel information to obtain a plurality of blood vessel segment centerlines comprises: Identifying blood vessel key point information in the blood vessel centerline; and The blood vessel centerline is divided based on the recognition result to obtain a plurality of blood vessel segment centerlines.

3. The method according to claim 1, wherein constructing a graph structure of the centerline of each of the blood vessel segments comprises: Extracting from each of the blood vessel segment center lines the points constituting the blood vessel center line and the connection information between the points; and A graph structure of the centerline of each blood vessel segment is constructed according to the points of the blood vessel centerline and connection information between the points.

4. The method according to claim 3, wherein constructing a graph structure of each of the blood vessel segment centerlines according to the points of the blood vessel centerline and the connection information between the points comprises: According to the point of the blood vessel centerline, adding a node containing the position information of the point to the corresponding graph structure; as well as According to the connection information between the points, connecting edges are added to the corresponding graph structure to construct the graph structure of the center line of each blood vessel segment.

5. The method according to claim 1, wherein determining whether the center lines of the plurality of blood vessel segments are missing based on the blood vessel information comprises: Based on one or more of the vascular key point information, the vascular length information, the vascular topology information and the vascular segment identification information, it is determined whether the center lines of the multiple vascular segments are disconnected or occluded, so as to determine whether the center lines of the multiple vascular segments are missing.

6. A device for completing the centerline of a blood vessel segment, comprising: processor; A memory storing program instructions for completing the centerline of a blood vessel segment, wherein when the program instructions are executed by the processor, the device implements the method according to any one of claims 1 to 5.

7. A computer-readable storage medium storing computer-readable instructions for completing a centerline of a blood vessel segment, wherein the computer-readable instructions, when executed by one or more processors, implement the method according to any one of claims 1 to 5.

Images