Methods, devices, storage media, and electronic equipment for determining blood vessel endpoints

Abstract

This application discloses a method, apparatus, storage medium, and electronic device for determining blood vessel endpoints. The method involves identifying a target blood vessel from a three-dimensional medical image, determining a first adjacent blood vessel and a second adjacent blood vessel adjacent to both ends of the target blood vessel, determining a first interface between the target blood vessel and the first adjacent blood vessel, and determining a second interface between the target blood vessel and the second adjacent blood vessel. The target blood vessel endpoint is then determined based on these first and second interfaces. By obtaining the interfaces between the target blood vessel and its adjacent blood vessels at both ends, this application can more accurately locate the actual endpoint of the blood vessel, thereby improving the accuracy of determining the endpoints at both ends of the blood vessel.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a method, apparatus, storage medium, and electronic device for determining blood vessel endpoints. Background Technology

[0002] In clinical diagnosis and treatment, the vascular centerline, as a core feature line describing the topology of blood vessels, plays an important supporting role in key aspects such as vascular lesion diagnosis and interventional surgery planning. The accuracy of extracting the centerline at both ends of the blood vessel directly depends on the precise positioning of the two ends of the blood vessel. Therefore, the accuracy of determining the two ends of the blood vessel is a key factor affecting the effectiveness of vascular-related clinical applications.

[0003] In existing technologies, the determination of the endpoints of the vascular centerline often adopts a manual interactive method. This method is a traditional endpoint positioning method that relies on the operator's experience and interactive operation. Specifically, in medical images (such as CT angiography images and magnetic resonance angiography images), the operator directly specifies the starting point (such as the location of the root of the vessel) and the ending point (such as the location of the terminal of the vessel) of the centerline on the vascular image by clicking the mouse or manually marking.

[0004] However, this type of positioning method, which relies on manual interaction, is greatly affected by differences in operator experience and subjective judgment bias. It is easy to lead to poor consistency in endpoint positioning results among different operators or among the same operator at different number of operations, making it difficult to meet the clinical demand for accurate vascular endpoint positioning. Therefore, it is urgent to improve the existing methods for determining the endpoints of both ends of a blood vessel to improve the accuracy of the determination. Summary of the Invention

[0005] This application provides a method, apparatus, storage medium, and electronic device for determining the endpoints of blood vessels, which can improve the accuracy of determining the endpoints at both ends of blood vessels.

[0006] In a first aspect, embodiments of this application provide a method for determining the endpoint of a blood vessel, including: The target blood vessel to be located is determined from the three-dimensional medical image, and the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel are determined. Determine the first interface between the target blood vessel and the first adjacent blood vessel, and determine the second interface between the target blood vessel and the second adjacent blood vessel; The target vessel endpoint is determined based on the first interface and the second interface.

[0007] In some embodiments, determining the target blood vessel from the three-dimensional medical image includes: The three-dimensional medical image is segmented by a multi-class blood vessel segmentation model to obtain blood vessel masks of multiple blood vessel feature types. The target blood vessel mask corresponding to the target blood vessel is determined from the blood vessel masks of the multiple blood vessel feature types, and the target blood vessel is determined in the three-dimensional medical image based on the target blood vessel mask.

[0008] In some embodiments, determining the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel includes: From the plurality of vessel masks, determine the candidate vessel mask that is adjacent to both ends of the target vessel mask; Based on the candidate vessel mask, determine the first adjacent vessel mask corresponding to the first adjacent vessel and the second adjacent vessel mask corresponding to the second adjacent vessel. The first adjacent blood vessel is determined based on the first adjacent blood vessel mask, and the second adjacent blood vessel is determined based on the second adjacent blood vessel mask.

[0009] In some embodiments, determining the first interface between the target blood vessel and the first adjacent blood vessel, and determining the second interface between the target blood vessel and the second adjacent blood vessel, includes: The first interface is determined based on the target vessel mask and the first adjacent vessel mask, and the second interface is determined based on the target vessel mask and the second adjacent vessel mask.

[0010] In some embodiments, determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, includes: For each voxel corresponding to the target blood vessel mask, neighborhood detection is performed to obtain multiple first reference neighborhood voxels; A first target voxel belonging to the first adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels, and a second target voxel belonging to the second adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels; The first interface is determined based on the first target voxel, and the second interface is determined based on the second target voxel.

[0011] In some embodiments, determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, includes: Voxel dilation is performed on the first adjacent blood vessel mask to obtain the first adjacent blood vessel dilation mask, and voxel dilation is performed on the second adjacent blood vessel mask to obtain the second adjacent blood vessel dilation mask. A third target voxel belonging to the target blood vessel mask is determined from the first adjacent blood vessel expansion mask, and a fourth target voxel belonging to the target blood vessel mask is determined from the second adjacent blood vessel expansion mask; The first interface is determined based on the third target voxel, and the second interface is determined based on the fourth target voxel.

[0012] In some embodiments, after determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, the method further includes: Determine the first blood flow direction of the blood vessel corresponding to the first interface, take the first blood flow direction as the first normal direction of the first interface, and calibrate the first interface based on the first normal direction to obtain the calibrated first interface. In addition, the second blood flow direction of the blood vessel corresponding to the second interface is determined, the second blood flow direction is taken as the second normal direction of the second interface, and the second interface is calibrated based on the second normal direction to obtain the calibrated second interface.

[0013] In some implementations, determining the target vessel endpoint based on the first interface and the second interface includes: Determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; The first three-dimensional centroid of the voxel corresponding to the first interface is determined based on the first three-dimensional coordinates, and the second three-dimensional centroid of the voxel corresponding to the second interface is determined based on the second three-dimensional coordinates. The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

[0014] Secondly, embodiments of this application also provide a blood vessel endpoint determination device, comprising: The first determining module is used to determine the target blood vessel at the endpoint to be located from the three-dimensional medical image, and to determine the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel. The second determining module is used to determine the first interface between the target blood vessel and the first adjacent blood vessel, and to determine the second interface between the target blood vessel and the second adjacent blood vessel. The third determining module is used to determine the target vessel endpoint of the target vessel based on the first interface and the second interface.

[0015] Thirdly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when run on a computer, causes the computer to perform the vascular endpoint determination method as provided in any embodiment of this application.

[0016] Fourthly, embodiments of this application also provide an electronic device, including a processor and a memory, the memory having a computer program, the processor executing the vascular endpoint determination method as provided in any embodiment of this application by calling the computer program.

[0017] Fifthly, embodiments of this application also provide a computer program product containing instructions that, when the computer program product is run on a computer or processor, cause the computer or processor to execute the vascular endpoint determination method provided in any embodiment of this application.

[0018] The technical solution provided in this application determines the target blood vessel at the endpoint to be located from a three-dimensional medical image, and identifies a first adjacent blood vessel and a second adjacent blood vessel adjacent to both ends of the target blood vessel. It also determines a first interface between the target blood vessel and the first adjacent blood vessel, and a second interface between the target blood vessel and the second adjacent blood vessel. Based on the first and second interfaces, the endpoint of the target blood vessel is determined. By obtaining the interfaces between the target blood vessel at the endpoint to be located and the adjacent blood vessels at both ends of the target blood vessel, this application can more accurately locate the actual endpoint of the blood vessel, thereby improving the accuracy of determining the endpoints at both ends of the blood vessel. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an exemplary system architecture diagram of a method for determining blood vessel endpoints provided in an embodiment of this application.

[0021] Figure 2 This is a flowchart illustrating a method for determining blood vessel endpoints provided in an embodiment of this application.

[0022] Figure 3 This is a schematic diagram of the interface between blood vessels provided in an embodiment of this application.

[0023] Figure 4 This is a schematic diagram of a blood vessel endpoint provided in an embodiment of this application.

[0024] Figure 5This is a schematic diagram of the structure of the blood vessel endpoint determination device provided in the embodiments of this application.

[0025] Figure 6 This is a schematic diagram of a first structure of an electronic device provided in an embodiment of this application.

[0026] Figure 7 This is a schematic diagram of a second structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0028] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0029] Accurate centerline is fundamental for vascular stenosis analysis, surface reconstruction visualization, and estimation of vessel length and diameter. If centerline points are missing at both ends of the vessel, the missing portions cannot be fully analyzed; if the centerline points at both ends of the vessel are offset, the offset portions cannot be accurately used for quantitative analysis of stenosis and estimation of length and diameter, resulting in a tortuous reconstructed vessel. Therefore, accurately obtaining the endpoints of both ends of the vessel is a prerequisite for calculating the centerline.

[0030] This application provides a method for determining blood vessel endpoints that improves the accuracy of identifying both ends of a blood vessel. The execution entity of this method can be the blood vessel endpoint determination device provided in this application, or an electronic device integrating the device. The device can be implemented in hardware or software. The electronic device can be any device equipped with a processor and possessing processing capabilities, such as mobile electronic devices with processors like smartphones, tablets, PDAs, and laptops, or fixed electronic devices with processors like desktop computers, televisions, and servers.

[0031] Please see Figure 1 , Figure 1 This is an exemplary system architecture diagram of a method for determining blood vessel endpoints provided in an embodiment of this application.

[0032] like Figure 1 As shown, the system architecture may include electronic device 10, network 20, and server 30. Network 20 serves as the medium for providing a communication link between electronic device 10 and server 30. Network 20 may include various types of wired or wireless communication links, such as wired communication links including fiber optic cables, twisted-pair cables, or coaxial cables, and wireless communication links including Bluetooth communication links, Wireless-Fidelity (Wi-Fi) communication links, or microwave communication links, etc.

[0033] Electronic device 10 can interact with server 30 via network 20 to receive messages from server 30 or send messages to server 30, or electronic device 10 can interact with server 30 via network 20 to receive messages or data sent to server 30 by other users. Electronic device 10 can be hardware or software. When electronic device 10 is hardware, it can be various electronic devices, including but not limited to smartwatches, smartphones, tablets, laptops, and desktop computers. When electronic device 10 is software, it can be installed in the electronic devices listed above, and it can be implemented as multiple software programs or software modules (e.g., to provide distributed services), or it can be implemented as a single software program or software module, without specific limitations.

[0034] Server 30 can be a business server providing various services. It should be noted that server 30 can be either hardware or software. When server 30 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When server 30 is software, it can be implemented as multiple software programs or software modules (e.g., used to provide distributed services), or as a single software program or software module; no specific limitations are made here.

[0035] In this embodiment of the application, the electronic device 10 can determine the target blood vessel of the endpoint to be located from a three-dimensional medical image, and determine the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel, determine the first interface between the target blood vessel and the first adjacent blood vessel, and determine the second interface between the target blood vessel and the second adjacent blood vessel, and determine the target blood vessel endpoint of the target blood vessel based on the first interface and the second interface.

[0036] It should be understood that Figure 1 The number of electronic devices, networks, and servers shown is merely illustrative; any number of electronic devices, networks, and servers can be used as needed. Of course, the system architecture provided in this application may not include servers; that is, servers are optional in the system architecture provided in this application.

[0037] Please see Figure 2 , Figure 2 This is a flowchart illustrating a method for determining blood vessel endpoints according to an embodiment of this application. The specific flow of the method for determining blood vessel endpoints provided in this embodiment of the application can be as follows: S110. Determine the target blood vessel at the endpoint to be located from the three-dimensional medical image, and determine the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel.

[0038] Three-dimensional medical imaging refers to digital image data of blood vessels and surrounding tissues inside the human body, stored in the form of a three-dimensional voxel array, acquired through medical imaging technology. For example, this three-dimensional medical imaging may include CT angiography (CTA) images, magnetic resonance angiography (MRA) images, and other images.

[0039] In this context, the target vessel refers to a specific vessel selected from among multiple vessels for which its two ends need to be located. This target vessel is usually determined based on clinical needs or analytical objectives. For example, if the clinical need is to assess coronary artery stenosis, the target vessel is the coronary artery; if the study is to examine the terminal branches of cerebral blood vessels, the target vessel is a specific cerebral blood vessel branch.

[0040] Adjacent vessels refer to vessels that are directly spatially adjacent to one end of the target vessel and have anatomical connectivity in the three-dimensional anatomical structure presented in three-dimensional medical images. They are key reference vessels that help define the location of the corresponding endpoint of the target vessel.

[0041] It should be noted that direct spatial adjacency refers to a spatial position in which the adjacent blood vessel and the target blood vessel are directly in contact in the image without any other blood vessels or tissues separating them at one end (such as the root or terminal end). This can be visually identified through the continuity of grayscale and the course of the vessel. Anatomical functional connection refers to a physiological anatomical relationship with the target blood vessel (such as the continuity of blood flow, the derivation relationship between the main trunk and branches), rather than simply spatial proximity. For example, blood from the target blood vessel can directly flow into or out of adjacent blood vessels, or the two are different branches / segments of the same vascular tree. This can be understood as the target blood vessel having one directly adjacent blood vessel at each end, i.e., the first adjacent blood vessel and the second adjacent blood vessel.

[0042] In this step, the target blood vessel whose two endpoints need to be located is identified from the 3D medical image, and the first and second adjacent blood vessels adjacent to the two ends of the target blood vessel are determined. Specifically, the first adjacent blood vessel can be designated as the adjacent blood vessel proximal to the target blood vessel, and the second adjacent blood vessel can be designated as the adjacent blood vessel distal to the target blood vessel. That is, the first adjacent blood vessel is the blood vessel adjacent to the starting end of the target blood vessel, and the second adjacent blood vessel is the blood vessel adjacent to the ending end of the target blood vessel. It should be noted that in medical anatomy, proximal and distal are relative concepts describing the positional relationship of blood vessels, determined based on the origin of blood flow to the vessel or its distance from the heart.

[0043] Specifically, the target blood vessel, the first adjacent blood vessel, and the second adjacent blood vessel can be identified based on anatomical features and relevant image processing techniques. For example, 3D medical images can be segmented to output segmentation masks for different blood vessels, where each mask corresponds to a unique blood vessel type, representing different blood vessel branches or segments. Then, based on clinical needs or anatomical features, the target blood vessel, the first adjacent blood vessel, and the second adjacent blood vessel can be determined from the multi-class segmentation masks.

[0044] S120. Determine the first interface between the target blood vessel and the first adjacent blood vessel, and determine the second interface between the target blood vessel and the second adjacent blood vessel.

[0045] In this context, the interface refers to the voxel region in the three-dimensional space of a three-dimensional medical image where the target blood vessel is in direct contact with its adjacent blood vessels and possesses anatomical structural continuity. It can be understood that the first interface is the voxel region connecting the target blood vessel to its first adjacent blood vessel, and the second interface is the voxel region connecting the target blood vessel to its second adjacent blood vessel.

[0046] Voxel, short for volume pixel, is similar to the smallest unit in two-dimensional space—the pixel. It is the smallest unit in three-dimensional space segmentation and is widely used in fields such as 3D imaging, scientific data, and medical imaging. Voxelization of a 3D model essentially involves discretizing the 3D space containing the model into cubic or cuboid meshes, and determining whether each mesh is on the model (surface voxelization) or inside the model (solid voxelization).

[0047] In this step, after determining the target blood vessel, the first adjacent blood vessel, and the second adjacent blood vessel in the three-dimensional medical image, the first interface between the target blood vessel and the first adjacent blood vessel, and the second interface between the target blood vessel and the second adjacent blood vessel are further determined.

[0048] In some implementations, the interface between the target blood vessel and its adjacent vessels at both ends can be obtained using a deep learning-based interface recognition model. This interface extraction model is a deep learning-based intelligent model whose core function is to receive three-dimensional medical images, blood vessel segmentation masks, and target blood vessel mask values ​​as input. By learning manually annotated interface features, it automatically identifies and outputs the interfaces (presented as a three-dimensional array) connecting the two ends of the target blood vessel to its adjacent vessels from the input data. This can replace manual annotation and provide accurate spatial location data for subsequent tasks such as target blood vessel endpoint localization.

[0049] The training process of this interface recognition model is as follows: Input data: a set of original images of the whole person (3D array), a mask after blood vessel segmentation (3D array), and the mask value (integer value) of the target blood vessel at the interface to be extracted.

[0050] Among them, the original set of images of the whole person refers to the unprocessed raw data of three-dimensional medical images, such as the voxel matrix of CTA and MRA images.

[0051] The mask after blood vessel segmentation refers to the binarized / multi-valued matrix that marks the spatial locations of all blood vessels in the image after blood vessel segmentation processing.

[0052] The mask value of the target blood vessel at the interface to be extracted refers to the identifier value used to explicitly specify the specific blood vessel that the model needs to analyze.

[0053] Input labels: The interfaces at both ends of the target blood vessel to be extracted (3D array, manually labeled).

[0054] In this process, the annotator will identify the extent of the interface between the two ends of the target blood vessel on a voxel-by-voxel basis in the 3D medical image. Voxels belonging to the interface will be marked with a specific value, and voxels not belonging to the interface will be marked with another specific value, ultimately forming a label in the form of a 3D array. For example, voxels belonging to the interface will be marked as 1, and voxels not belonging to the interface will be marked as 0.

[0055] Specifically, to obtain the interface recognition model, the following steps are taken: First, obtain the sample 3D medical image, the sample blood vessel segmentation mask, and the sample target blood vessel mask value. Then, obtain the interface labels at both ends of the target blood vessel corresponding to the sample 3D medical image, sample blood vessel segmentation mask, and sample target blood vessel mask value. Simultaneously, determine the basic image processing model. Input the sample 3D medical image, sample blood vessel segmentation mask, and sample target blood vessel mask value into the basic image processing model. This basic image processing model outputs the interface labels at both ends of the reference target blood vessel for the sample 3D medical image, sample blood vessel segmentation mask, and sample target blood vessel mask value. Then, construct a loss function using the parameters corresponding to the interface labels at both ends of the reference target blood vessel and the sample target blood vessel. Determine the model loss value for the loss function corresponding to the interface labels at both ends of the reference target blood vessel and the sample target blood vessel. Then, adjust the model parameters of the basic image processing model using the model loss value. After multiple rounds of model training, the basic image processing model is trained until it is complete, resulting in the interface recognition model. The basic image processing model can employ V-Net specifically for 3D medical images, or its improved model V-Net++, etc.

[0056] S130. Determine the target vessel endpoint of the target vessel based on the first interface and the second interface.

[0057] In this step, after determining the first and second interfaces corresponding to the target vessel and its adjacent vessels, the endpoints of the target vessel are determined using these interfaces. Specifically, the endpoints of the target vessel can be located using the centroids of the first and second interfaces. This application obtains the interfaces of adjacent vessels based on vascular anatomy, enabling more precise localization of the actual endpoints and providing strong clinical interpretability.

[0058] In practice, this application is not limited by the execution order of the described steps. Without causing conflicts, some steps may be performed in other orders or simultaneously.

[0059] As can be seen from the above, the method for determining the endpoint of a blood vessel provided in this application determines the target blood vessel from a three-dimensional medical image, identifies the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel, determines the first interface between the target blood vessel and the first adjacent blood vessel, and determines the second interface between the target blood vessel and the second adjacent blood vessel. Based on the first and second interfaces, the endpoint of the target blood vessel is determined. This application, by obtaining the interface between the target blood vessel and the adjacent blood vessels at both ends of the target blood vessel, can more accurately locate the actual endpoint of the blood vessel, thereby improving the accuracy of determining the endpoints at both ends of the blood vessel. Based on the anatomical features of the blood vessel structure, this application can fully automatically and accurately locate the endpoints at both ends of the blood vessel.

[0060] In some implementations, performing the step "identifying the target blood vessel from the three-dimensional medical image" may include the following steps: (11) The three-dimensional medical image is segmented by a multi-class blood vessel segmentation model to obtain blood vessel masks of multiple blood vessel feature types; The multi-class blood vessel segmentation model refers to an algorithmic tool used to segment blood vessel types, capable of simultaneously distinguishing multiple different blood vessel types. This multi-class blood vessel segmentation model can be based on a deep learning semantic segmentation network, such as U-Net, V-Net, or U-Net++.

[0061] In this context, vascular feature type refers to the criteria used to classify blood vessels into different categories during vascular type segmentation. For example, vascular features can include dimensions such as anatomical branching features, anatomical segment features, and functional / pathological features.

[0062] It should be noted that anatomical branching features refer to the classification of blood vessels according to their branching within the human anatomy. This is the most commonly used feature type, such as the aorta, renal artery, middle cerebral artery, and left anterior descending coronary artery. Each branch is an independent vascular feature type. Anatomical segment features refer to the division of the same vascular branch into different anatomical segments. This is suitable for scenarios requiring detailed analysis of local vascular structures, such as further dividing the middle cerebral artery into the proximal, middle, and distal segments. Each segment is an independent vascular feature type. Functional / pathological features refer to the classification based on the physiological function or pathological state of the blood vessel. This is suitable for clinical diagnostic and treatment-related segmentation scenarios, such as normal arteries, stenotic arteries, thrombotic vessels, and collateral circulation vessels. Each function or pathological state corresponds to an independent vascular feature type.

[0063] In this step, a multi-class blood vessel segmentation model is used to perform blood vessel type segmentation processing on the three-dimensional medical image (also known as blood vessel segmentation processing), and outputs segmentation masks for different blood vessels. Each mask corresponds to a unique blood vessel type, such as representing different blood vessel branches or segments.

[0064] For example, the blood vessel type can be represented as a mask:

[0065] (12) Determine the target blood vessel mask corresponding to the target blood vessel from the blood vessel masks of the plurality of blood vessel feature types, and determine the target blood vessel in the three-dimensional medical image based on the target blood vessel mask.

[0066] In this step, the target vessel for the endpoint to be located can be determined from the multi-class segmentation mask based on clinical needs or anatomical features, and denoted as the target vessel mask. Based on the target blood vessel mask, the target blood vessel can be identified in the three-dimensional medical image.

[0067] In some implementations, the step of "determining the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel" may include the following steps: (21) Determine candidate blood vessel masks that are adjacent to both ends of the target blood vessel mask from the plurality of blood vessel masks; Among them, the candidate vessel mask refers to the vessel mask that has a direct three-dimensional spatial adjacency with a certain end (proximal or distal end) of the target vessel mask among the multiple vessel masks output by multi-class vessel segmentation, and has the possibility of vascular anatomical connection. It is the set of candidate objects for subsequent determination of the first adjacent vessel and the second adjacent vessel.

[0068] In this step, the spatial adjacency relationship of the target blood vessel mask is analyzed to determine its connection with... The mask of directly connected adjacent blood vessels, i.e., the candidate blood vessel mask, is denoted as . and .in, , These are the adjacent vessels proximal and distal to the target vessel, respectively.

[0069] (22) Determine the first adjacent blood vessel mask corresponding to the first adjacent blood vessel and the second adjacent blood vessel mask corresponding to the second adjacent blood vessel based on the candidate blood vessel mask; In this step, after the candidate vessel mask is determined, the first adjacent vessel mask of the first adjacent vessel (proximal end of the target vessel) and the second adjacent vessel mask of the second adjacent vessel (distal end of the target vessel) can be determined respectively.

[0070] (23) Determine the first adjacent blood vessel based on the first adjacent blood vessel mask, and determine the second adjacent blood vessel based on the second adjacent blood vessel mask.

[0071] In this step, after determining the first adjacent blood vessel mask and the second adjacent blood vessel mask, the first adjacent blood vessel in the three-dimensional medical image can be determined based on the first adjacent blood vessel mask, and the second adjacent blood vessel in the three-dimensional medical image can be determined based on the second adjacent blood vessel mask.

[0072] In some implementations, when determining the first interface between the target blood vessel and the first adjacent blood vessel, and when determining the second interface between the target blood vessel and the second adjacent blood vessel, the first interface can be determined based on the target blood vessel mask and the first adjacent blood vessel mask, and the second interface can be determined based on the target blood vessel mask and the second adjacent blood vessel mask.

[0073] In this embodiment, the first interface can be found through the voxel association relationship between the target blood vessel mask and the first adjacent blood vessel mask, and the second interface can be found through the voxel association relationship between the target blood vessel mask and the second adjacent blood vessel mask.

[0074] Specifically, in some embodiments, the steps of "determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask" may include the following steps: (31) Perform neighborhood detection on each voxel corresponding to the target blood vessel mask to obtain multiple first reference neighborhood voxels; Neighborhood detection refers to the technical operation of identifying, filtering, or statistically analyzing the attributes of all voxels within a preset neighborhood for a specific voxel in a 3D image (in this application, the voxel of the target blood vessel mask).

[0075] The first reference neighborhood voxel refers to the neighborhood voxels within the neighborhood range of each voxel after performing neighborhood detection on each voxel of the target blood vessel mask.

[0076] In this step, neighborhood detection is performed on each voxel corresponding to the target blood vessel mask to obtain multiple first reference neighborhood voxels.

[0077] For example, each voxel of the target blood vessel mask can be traversed, and a 26-neighborhood operator (containing the neighborhood range of the central voxel and 26 adjacent voxels in three-dimensional space) can be used to perform neighborhood detection on each voxel to obtain the plurality of first reference neighborhood voxels.

[0078] (32) Determine a first target voxel belonging to the first adjacent blood vessel mask from the plurality of first reference neighborhood voxels, and determine a second target voxel belonging to the second adjacent blood vessel mask from the plurality of first reference neighborhood voxels; In this step, voxels belonging to the first adjacent blood vessel mask are determined from multiple first reference neighborhood voxels as first target voxels, and voxels belonging to the second adjacent blood vessel mask are determined from multiple first reference neighborhood voxels as second target voxels.

[0079] (33) Determine the first interface based on the first target voxel, and determine the second interface based on the second target voxel.

[0080] In this step, after determining the first target voxel and the second target voxel, the first target voxel is the voxel constituting the first interface, and the second target voxel is the voxel constituting the second interface. That is, the first target voxel is used as a candidate point for the first interface, and the second target voxel is used as a candidate point for the second interface. All candidate points for the first interface constitute the first interface, and all candidate points for the second interface constitute the second interface.

[0081] Specifically, in some embodiments, a method for determining the interface is also provided. When performing the step of "determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask", the following steps may be included: (41) Perform voxel dilation on the first adjacent blood vessel mask to obtain the first adjacent blood vessel dilation mask, and perform voxel dilation on the second adjacent blood vessel mask to obtain the second adjacent blood vessel dilation mask. Among them, voxel dilation is one of the most basic operations in 3D morphological image processing. For binary masks, its core is to expand the foreground voxels to the surrounding neighborhood.

[0082] The first adjacent vessel dilation mask is a mask obtained by voxel dilation of the first adjacent vessel mask. The second adjacent vessel dilation mask is a mask obtained by voxel dilation of the second adjacent vessel mask.

[0083] In this step, voxel dilation operations are performed on the first adjacent blood vessel mask and the second adjacent blood vessel mask respectively to obtain the first adjacent blood vessel dilation mask and the second adjacent blood vessel dilation mask.

[0084] (42) Determine a third target voxel belonging to the target blood vessel mask from the first adjacent blood vessel expansion mask, and determine a fourth target voxel belonging to the target blood vessel mask from the second adjacent blood vessel expansion mask; In this step, each voxel of the target vessel mask can be traversed. If there is a voxel that is shared with the first adjacent vessel expansion mask, then this shared voxel is the third target voxel. Similarly, each voxel of the target vessel mask can be traversed. If there is a voxel that is shared with the second adjacent vessel expansion mask, then this shared voxel is the fourth target voxel.

[0085] (43) Determine the first interface based on the third target voxel, and determine the second interface based on the fourth target voxel.

[0086] In this step, after determining the third and fourth target voxels, the third target voxel becomes the voxel constituting the first interface, and the fourth target voxel becomes the voxel constituting the second interface. That is, the third target voxel is used as a candidate point for the third interface of the first interface, and the fourth target voxel is used as a candidate point for the fourth interface of the second interface. All candidate points for the third interface constitute the first interface, and all candidate points for the fourth interface constitute the second interface.

[0087] For example, such as the mask of the first adjacent blood vessel Perform a voxel dilation operation (morphological dilation) on one voxel to obtain the dilation mask of the first adjacent blood vessel. and the mask for the second adjacent blood vessel. Perform a voxel dilation operation on one voxel to obtain the dilation mask for the second adjacent blood vessel. ,calculate and , The overlapping area is the interface between the target blood vessel and its adjacent blood vessels at both ends.

[0088] The formula for calculating morphological dilation is as follows:

[0089]

[0090] The calculation formula for interface extraction is as follows:

[0091]

[0092] in, This is the first interface. This is the second interface.

[0093] In this embodiment, after performing voxel dilation on the adjacent blood vessel masks of the target blood vessel mask, each voxel of the target blood vessel mask is traversed to determine the common voxels between the target blood vessel mask and the adjacent blood vessel masks after the dilation operation. The interface between the target blood vessel and its adjacent blood vessels is determined by the common voxels, which can avoid the determination of incomplete interfaces and thus more accurately determine the location of the interface and avoid possible omission of boundary points.

[0094] Specifically, this application also provides an implementation method for calibrating the interface between a target blood vessel and its adjacent blood vessels at both ends. After performing the steps of "determining the first interface based on the target blood vessel mask and the first adjacent blood vessel mask, and determining the second interface based on the target blood vessel mask and the second adjacent blood vessel mask", the following steps may be included: (51) Determine the first blood flow direction of the blood vessel corresponding to the first interface, take the first blood flow direction as the first normal direction of the first interface, and calibrate the first interface based on the first normal direction to obtain the calibrated first interface. In this step, for the first interface, the first normal direction is determined by the first blood flow direction, and then the shape of the first interface is calibrated using the first normal direction to obtain a more accurate calibrated first interface.

[0095] And (61) determine the second blood flow direction of the blood vessel corresponding to the second interface, take the second blood flow direction as the second normal direction of the second interface, and calibrate the second interface based on the second normal direction to obtain the calibrated second interface.

[0096] In this step, for the second interface, the second normal direction is determined by the second blood flow direction, and then the morphology of the second interface is calibrated using the second normal direction to obtain a more accurate calibrated second interface.

[0097] Please refer to Figure 3 , Figure 3 Figure 3 This is a schematic diagram of the interface between blood vessels provided in an embodiment of this application. Specifically, Figure 3 The diagram illustrates the interface before and after correction between blood vessel 1 and blood vessel 2, as well as the direction of blood flow.

[0098] In this embodiment, based on the position of the interface, N units of pixels are expanded to both sides to obtain a partially truncated blood vessel mask. The blood flow direction is calculated, which is the normal direction of the three-dimensional blood vessel lumen cross section. The interface is calibrated by this normal direction so that the normal direction of the interface is consistent with the blood flow direction at the current position. In other words, the interface is made to fit the actual connection surface through which blood flow passes in the vascular anatomy structure more closely, further improving the accuracy of subsequent endpoint positioning.

[0099] In some implementations, the step of "determining the target vessel endpoint based on the first interface and the second interface" may include the following steps: (71) Determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; In this step, the three-dimensional coordinates of the first interface and the second interface in the three-dimensional medical image are extracted to obtain the first three-dimensional coordinates corresponding to the first interface and the second three-dimensional coordinates corresponding to the second interface.

[0100] (72) Determine the first three-dimensional centroid of the voxel corresponding to the first interface based on the first three-dimensional coordinates, and determine the second three-dimensional centroid of the voxel corresponding to the second interface based on the second three-dimensional coordinates; In this step, the three-dimensional centroid of the voxel corresponding to the first interface is calculated based on the first three-dimensional coordinates corresponding to the first interface as the first three-dimensional centroid, and the three-dimensional centroid of the voxel corresponding to the second interface is calculated based on the second three-dimensional coordinates corresponding to the second interface as the second three-dimensional centroid.

[0101] The formula for determining the geometric centroid is as follows:

[0102]

[0103] in, This refers to the three-dimensional coordinates of the interface voxels. This refers to the first three-dimensional centroid. It refers to the second three-dimensional centroid.

[0104] (73) The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

[0105] In this step, the first three-dimensional centroid corresponding to the first interface and the second three-dimensional centroid corresponding to the second interface are the two endpoints on both sides of the target blood vessel.

[0106] Please refer to Figure 4 , Figure 4 This is a schematic diagram of the blood vessel endpoint provided in an embodiment of this application. Figure 4The diagram illustrates the interface corresponding to the blood vessel and the three-dimensional centroid selected on the interface as the endpoint of the blood vessel.

[0107] In one embodiment, a blood vessel endpoint determination device is also provided. See also... Figure 5 , Figure 5 This is a schematic diagram of the structure of a blood vessel endpoint determination device provided in an embodiment of this application. The blood vessel endpoint determination device 200 is applied to an electronic device and includes a first determination module 201, a second determination module 202, and a third determination module 203, as follows: The first determining module 201 is used to determine the target blood vessel of the endpoint to be located from the three-dimensional medical image, and to determine the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel. The second determining module 202 is used to determine the first interface between the target blood vessel and the first adjacent blood vessel, and to determine the second interface between the target blood vessel and the second adjacent blood vessel. The third determining module 203 is used to determine the target vessel endpoint of the target vessel based on the first interface and the second interface.

[0108] In some implementations, the first determining module 201 is configured to: The three-dimensional medical image is segmented by a multi-class blood vessel segmentation model to obtain blood vessel masks of multiple blood vessel feature types. The target blood vessel mask corresponding to the target blood vessel is determined from the blood vessel masks of the multiple blood vessel feature types, and the target blood vessel is determined in the three-dimensional medical image based on the target blood vessel mask.

[0109] In some implementations, the first determining module 201 is configured to: From the plurality of vessel masks, determine the candidate vessel mask that is adjacent to both ends of the target vessel mask; Based on the candidate vessel mask, determine the first adjacent vessel mask corresponding to the first adjacent vessel and the second adjacent vessel mask corresponding to the second adjacent vessel. The first adjacent blood vessel is determined based on the first adjacent blood vessel mask, and the second adjacent blood vessel is determined based on the second adjacent blood vessel mask.

[0110] In some implementations, the second determining module 202 is configured to: The first interface is determined based on the target vessel mask and the first adjacent vessel mask, and the second interface is determined based on the target vessel mask and the second adjacent vessel mask.

[0111] In some implementations, the second determining module 202 is configured to: For each voxel corresponding to the target blood vessel mask, neighborhood detection is performed to obtain multiple first reference neighborhood voxels; A first target voxel belonging to the first adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels, and a second target voxel belonging to the second adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels; The first interface is determined based on the first target voxel, and the second interface is determined based on the second target voxel.

[0112] In some implementations, the second determining module 202 is configured to: Voxel dilation is performed on the first adjacent blood vessel mask to obtain the first adjacent blood vessel dilation mask, and voxel dilation is performed on the second adjacent blood vessel mask to obtain the second adjacent blood vessel dilation mask. A third target voxel belonging to the target blood vessel mask is determined from the first adjacent blood vessel expansion mask, and a fourth target voxel belonging to the target blood vessel mask is determined from the second adjacent blood vessel expansion mask; The first interface is determined based on the third target voxel, and the second interface is determined based on the fourth target voxel.

[0113] In some embodiments, the second determining module 202 is further configured to: Determine the first blood flow direction of the blood vessel corresponding to the first interface, take the first blood flow direction as the first normal direction of the first interface, and calibrate the first interface based on the first normal direction to obtain the calibrated first interface. In addition, the second blood flow direction of the blood vessel corresponding to the second interface is determined, the second blood flow direction is taken as the second normal direction of the second interface, and the second interface is calibrated based on the second normal direction to obtain the calibrated second interface.

[0114] In some embodiments, the third determining module 203 is configured to: Determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; The first three-dimensional centroid of the voxel corresponding to the first interface is determined based on the first three-dimensional coordinates, and the second three-dimensional centroid of the voxel corresponding to the second interface is determined based on the second three-dimensional coordinates. The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

[0115] It should be noted that the blood vessel endpoint determination device provided in this application embodiment belongs to the same concept as the blood vessel endpoint determination method in the above embodiment. The blood vessel endpoint determination device can implement any of the methods provided in the blood vessel endpoint determination method embodiment. For details of its implementation process, please refer to the blood vessel endpoint determination method embodiment, which will not be repeated here.

[0116] Furthermore, to better implement the blood vessel endpoint determination method in the embodiments of this application, this application also provides an electronic device based on the blood vessel endpoint determination method. The electronic device may be a smartwatch, smartphone, tablet computer, laptop computer, or desktop computer, etc. Please refer to [link / reference]. Figure 6 , Figure 6 This is a schematic diagram of a first structure of an electronic device provided in an embodiment of this application. The electronic device 300 includes a processor 301 and a memory 302. The processor 301 and the memory 302 are electrically connected.

[0117] The processor 301 is the control center of the electronic device 300. It connects various parts of the electronic device via various interfaces and lines, and executes various functions and processes data by running or calling computer programs stored in the memory 302 and accessing data stored in the memory 302, thereby providing overall monitoring of the electronic device. The processor 301 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0118] The memory 302 can be used to store computer programs and data. The computer programs stored in the memory 302 contain instructions that can be executed in the processor. The computer programs can be composed of various functional modules. The processor 401 executes various functional applications and data processing by calling the computer programs stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device 300 (such as audio data, video data, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0119] In this embodiment, the processor 301 in the electronic device 300 loads the instructions corresponding to the processes of one or more computer programs into the memory 302 according to the following steps, and the processor 401 runs the computer programs stored in the memory 302 to realize various functions: The target blood vessel to be located is determined from the three-dimensional medical image, and the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel are determined. Determine the first interface between the target blood vessel and the first adjacent blood vessel, and determine the second interface between the target blood vessel and the second adjacent blood vessel; The target vessel endpoint is determined based on the first interface and the second interface.

[0120] In some implementations, please refer to Figure 7 , Figure 7 This is a second structural schematic diagram of the electronic device provided in an embodiment of this application. The electronic device 300 further includes: a radio frequency circuit 303, a display screen 304, a control circuit 305, an input unit 306, an audio circuit 307, a sensor 308, and a power supply 309. The processor 301 is electrically connected to the radio frequency circuit 303, the display screen 304, the control circuit 305, the input unit 306, the audio circuit 307, the sensor 308, and the power supply 309.

[0121] The radio frequency circuit 303 is used to transmit and receive radio frequency signals to communicate with network devices or other electronic devices via wireless communication.

[0122] The display screen 304 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of electronic devices, which can be composed of images, text, icons, videos, and any combination thereof.

[0123] The control circuit 305 is electrically connected to the display screen 304 and is used to control the display screen 304 to display information.

[0124] The input unit 306 can be used to receive input numeric or character information or user characteristic information (such as fingerprints), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. The input unit 306 may include a fingerprint recognition module.

[0125] The audio circuit 307 provides an audio interface between the user and the electronic device via a speaker and a microphone. The audio circuit 307 includes a microphone, which is electrically connected to the processor 301. The microphone is used to receive voice information input by the user.

[0126] Sensor 308 is used to collect information about the external environment. Sensor 308 may include one or more sensors such as an ambient light sensor, an accelerometer, and a gyroscope.

[0127] The power supply 309 is used to supply power to the various components of the electronic device 300. In some embodiments, the power supply 309 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system.

[0128] Although not shown in the figure, electronic device 300 may also include a camera, Bluetooth module, etc., which will not be described in detail here.

[0129] In this embodiment, the processor 301 in the electronic device 300 loads the instructions corresponding to the processes of one or more computer programs into the memory 302 according to the following steps, and the processor 301 runs the computer programs stored in the memory 302 to realize various functions: The target blood vessel to be located is determined from the three-dimensional medical image, and the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel are determined. Determine the first interface between the target blood vessel and the first adjacent blood vessel, and determine the second interface between the target blood vessel and the second adjacent blood vessel; The target vessel endpoint is determined based on the first interface and the second interface.

[0130] In some implementations, when processor 301 performs the step of determining the target blood vessel from the three-dimensional medical image, it may perform the following: The three-dimensional medical image is segmented by a multi-class blood vessel segmentation model to obtain blood vessel masks of multiple blood vessel feature types. The target blood vessel mask corresponding to the target blood vessel is determined from the blood vessel masks of the multiple blood vessel feature types, and the target blood vessel is determined in the three-dimensional medical image based on the target blood vessel mask.

[0131] In some implementations, when processor 301 executes the step of determining the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel, it may perform the following: From the plurality of vessel masks, determine the candidate vessel mask that is adjacent to both ends of the target vessel mask; Based on the candidate vessel mask, determine the first adjacent vessel mask corresponding to the first adjacent vessel and the second adjacent vessel mask corresponding to the second adjacent vessel. The first adjacent blood vessel is determined based on the first adjacent blood vessel mask, and the second adjacent blood vessel is determined based on the second adjacent blood vessel mask.

[0132] In some implementations, when processor 301 executes the steps of determining the first interface between the target blood vessel and the first adjacent blood vessel, and determining the second interface between the target blood vessel and the second adjacent blood vessel, it may perform the following: The first interface is determined based on the target vessel mask and the first adjacent vessel mask, and the second interface is determined based on the target vessel mask and the second adjacent vessel mask.

[0133] In some embodiments, when processor 301 performs the steps of determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, it may perform the following: For each voxel corresponding to the target blood vessel mask, neighborhood detection is performed to obtain multiple first reference neighborhood voxels; A first target voxel belonging to the first adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels, and a second target voxel belonging to the second adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels; The first interface is determined based on the first target voxel, and the second interface is determined based on the second target voxel.

[0134] In some embodiments, when processor 301 performs the steps of determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, it may perform the following: Voxel dilation is performed on the first adjacent blood vessel mask to obtain the first adjacent blood vessel dilation mask, and voxel dilation is performed on the second adjacent blood vessel mask to obtain the second adjacent blood vessel dilation mask. A third target voxel belonging to the target blood vessel mask is determined from the first adjacent blood vessel expansion mask, and a fourth target voxel belonging to the target blood vessel mask is determined from the second adjacent blood vessel expansion mask; The first interface is determined based on the third target voxel, and the second interface is determined based on the fourth target voxel.

[0135] In some embodiments, after processor 301 performs the steps of determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, it may further perform the following: Determine the first blood flow direction of the blood vessel corresponding to the first interface, take the first blood flow direction as the first normal direction of the first interface, and calibrate the first interface based on the first normal direction to obtain the calibrated first interface. In addition, the second blood flow direction of the blood vessel corresponding to the second interface is determined, the second blood flow direction is taken as the second normal direction of the second interface, and the second interface is calibrated based on the second normal direction to obtain the calibrated second interface.

[0136] In some implementations, when processor 301 executes the step of determining the target vessel endpoint based on the first interface and the second interface, it may perform the following: Determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; The first three-dimensional centroid of the voxel corresponding to the first interface is determined based on the first three-dimensional coordinates, and the second three-dimensional centroid of the voxel corresponding to the second interface is determined based on the second three-dimensional coordinates. The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

[0137] This application also provides a computer-readable storage medium storing a computer program. When the computer program is run on a computer, the computer executes the blood vessel endpoint determination method described in any of the above embodiments.

[0138] It should be noted that those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, which may include, but is not limited to, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0139] This application also provides a computer program product containing instructions that, when run on a computer or processor, cause the computer or processor to execute the vascular endpoint determination method described in any of the above embodiments.

[0140] Furthermore, the terms "first," "second," and "third," etc., used in this application are used to distinguish different objects, not to describe a specific order. Additionally, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the listed steps or modules, but some embodiments may also include steps or modules not listed, or some embodiments may include other steps or modules inherent to these processes, methods, products, or devices.

[0141] The above provides a detailed description of the method, apparatus, storage medium, and electronic device for determining blood vessel endpoints provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for determining the endpoints of a blood vessel, characterized in that, include: The target blood vessel to be located is determined from the three-dimensional medical image, and the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel are determined. Adjacent blood vessels refer to blood vessels that are directly spatially adjacent to one end of the target blood vessel and have anatomical connection in the three-dimensional anatomical structure presented in three-dimensional medical imaging. A first interface is determined between the target blood vessel and the first adjacent blood vessel, and a second interface is determined between the target blood vessel and the second adjacent blood vessel; the first interface is the voxel region where the target blood vessel connects with the first adjacent blood vessel, and the second interface is the voxel region where the target blood vessel connects with the second adjacent blood vessel. Determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; The first three-dimensional centroid of the voxel corresponding to the first interface is determined based on the first three-dimensional coordinates, and the second three-dimensional centroid of the voxel corresponding to the second interface is determined based on the second three-dimensional coordinates. The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

2. The method according to claim 1, characterized in that, The process of identifying the target blood vessel from three-dimensional medical images includes: The three-dimensional medical image is segmented by a multi-class blood vessel segmentation model to obtain blood vessel masks of multiple blood vessel feature types. The target blood vessel mask corresponding to the target blood vessel is determined from the blood vessel masks of the multiple blood vessel feature types, and the target blood vessel is determined in the three-dimensional medical image based on the target blood vessel mask.

3. The method according to claim 2, characterized in that, The determination of the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel includes: From the plurality of vessel masks, determine the candidate vessel mask that is adjacent to both ends of the target vessel mask; Based on the candidate vessel mask, determine the first adjacent vessel mask corresponding to the first adjacent vessel and the second adjacent vessel mask corresponding to the second adjacent vessel. The first adjacent blood vessel is determined based on the first adjacent blood vessel mask, and the second adjacent blood vessel is determined based on the second adjacent blood vessel mask.

4. The method according to claim 3, characterized in that, Determining the first interface between the target blood vessel and the first adjacent blood vessel, and determining the second interface between the target blood vessel and the second adjacent blood vessel, includes: The first interface is determined based on the target vessel mask and the first adjacent vessel mask, and the second interface is determined based on the target vessel mask and the second adjacent vessel mask.

5. The method according to claim 4, characterized in that, The step of determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, includes: For each voxel corresponding to the target blood vessel mask, neighborhood detection is performed to obtain multiple first reference neighborhood voxels; A first target voxel belonging to the first adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels, and a second target voxel belonging to the second adjacent blood vessel mask is determined from the plurality of first reference neighborhood voxels; The first interface is determined based on the first target voxel, and the second interface is determined based on the second target voxel.

6. The method according to claim 4, characterized in that, The step of determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, includes: Voxel dilation is performed on the first adjacent blood vessel mask to obtain the first adjacent blood vessel dilation mask, and voxel dilation is performed on the second adjacent blood vessel mask to obtain the second adjacent blood vessel dilation mask. A third target voxel belonging to the target blood vessel mask is determined from the first adjacent blood vessel expansion mask, and a fourth target voxel belonging to the target blood vessel mask is determined from the second adjacent blood vessel expansion mask; The first interface is determined based on the third target voxel, and the second interface is determined based on the fourth target voxel.

7. The method according to claim 4, characterized in that, After determining the first interface based on the target vessel mask and the first adjacent vessel mask, and determining the second interface based on the target vessel mask and the second adjacent vessel mask, the method further includes: Determine the first blood flow direction of the blood vessel corresponding to the first interface, take the first blood flow direction as the first normal direction of the first interface, and calibrate the first interface based on the first normal direction to obtain the calibrated first interface. In addition, the second blood flow direction of the blood vessel corresponding to the second interface is determined, the second blood flow direction is taken as the second normal direction of the second interface, and the second interface is calibrated based on the second normal direction to obtain the calibrated second interface.

8. A device for determining the endpoint of a blood vessel, characterized in that, include: The first determining module is used to determine the target blood vessel at the endpoint to be located from the three-dimensional medical image, and to determine the first adjacent blood vessel and the second adjacent blood vessel adjacent to both ends of the target blood vessel; the adjacent blood vessel refers to a blood vessel that has a direct spatial adjacent relationship with one end of the target blood vessel and has anatomical connection in the three-dimensional anatomical structure presented in the three-dimensional medical image. The second determining module is used to determine the first interface between the target blood vessel and the first adjacent blood vessel, and to determine the second interface between the target blood vessel and the second adjacent blood vessel; the first interface is the voxel region connecting the target blood vessel and the first adjacent blood vessel, and the second interface is the voxel region connecting the target blood vessel and the second adjacent blood vessel. The third determining module is used to determine the first three-dimensional coordinates of the first interface and the second three-dimensional coordinates of the second interface; determine the first three-dimensional centroid of the voxel corresponding to the first interface based on the first three-dimensional coordinates, and determine the second three-dimensional centroid of the voxel corresponding to the second interface based on the second three-dimensional coordinates; The first three-dimensional centroid and the second three-dimensional centroid are used as the target vessel endpoints of the target vessel.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run on a computer, it causes the computer to perform the vascular endpoint determination method as described in any one of claims 1 to 7.

10. An electronic device comprising a processor and a memory, the memory storing a computer program, characterized in that, The processor invokes the computer program to execute the vascular endpoint determination method as described in any one of claims 1 to 7.

Images