Method and device for segmenting aneurysm and storage medium

Abstract

The invention discloses a method and equipment for segmenting an aneurysm and a storage medium. The method comprises the following steps: generating a blood vessel map structure according to a blood vessel segmentation mask, wherein the blood vessel map structure at least comprises leaf nodes containing radius attributes; determining an initial path node in the aneurysm according to the positioning point; screening target paths from the initial path node to the leaf node, and sorting the target paths according to an average radius to form a path pair; restoring the path passing through the aneurysm based on the path pair to obtain a restored path; and segmenting the aneurysm based on the reduction path to obtain an aneurysm segmentation result. By means of the scheme, stable segmentation of the large aneurysm can be guaranteed, the peripheral blood vessel condition of the aneurysm can be stably obtained, and the accuracy and reliability of aneurysm segmentation are improved.

Description

Technical Field

[0001] The present application generally relates to the field of medical image processing technology. More specifically, the present application relates to a method, apparatus, and computer-readable storage medium for segmenting an aneurysm. Background Art

[0002] In medicine, accurate segmentation of aneurysms is crucial for the diagnosis, treatment planning, and follow-up of neurosurgical diseases. Multimodal vascular imaging, such as digital subtraction angiography (DSA), CT angiography (CTA), and magnetic resonance angiography (MRA), provides important insights for aneurysm detection. However, accurately segmenting aneurysms from these images remains a challenging task.

[0003] In existing technologies, large intracranial aneurysms are characterized by diverse shapes and complex necks. Both manual segmentation and deep learning model segmentation face significant technical challenges. Specifically, manual segmentation is time-consuming and labor-intensive, relying on physician experience. Deep learning models, on the other hand, are prone to incomplete aneurysm segmentation and can mistakenly identify the parent artery as an aneurysm, leading to inaccurate segmentation results and compromising clinical diagnosis and treatment planning.

[0004] In view of this, the present application provides a solution for segmenting aneurysms, so as to ensure the stable segmentation of large aneurysms, stably obtain the surrounding blood vessels of the aneurysms, and improve the accuracy and reliability of aneurysm segmentation. Summary of the Invention

[0005] In order to at least solve one or more of the above-mentioned technical problems, the present application proposes a solution for segmenting an aneurysm in multiple aspects.

[0006] In a first aspect, the present application provides a method for segmenting an aneurysm, comprising: generating a vascular graph structure based on a vascular segmentation mask, wherein the vascular graph structure includes at least a leaf node containing a radius attribute; determining an initial path node inside the aneurysm based on a positioning point; screening a target path from the initial path node to the leaf node, and sorting the target path according to an average radius to form a path pair; restoring a path passing through the aneurysm based on the path pair to obtain a restored path; and segmenting the aneurysm based on the restored path to obtain an aneurysm segmentation result.

[0007] In some embodiments, the target path from the initial path node to the leaf node is filtered by the following operations: traversing the path set from the initial path node to the leaf node; calculating the length or average radius of each path in the path set; and filtering the target path from the initial path node to the leaf node according to the length or average radius of each path.

[0008] In some embodiments, screening the target path from the initial path node to the leaf node based on the length or average radius of each path includes: screening out paths whose length is less than a length threshold or whose average radius is less than a radius threshold from the path set to obtain the target path from the initial path node to the leaf node.

[0009] In some embodiments, the vascular graph structure further includes edge nodes including radius attributes, and restoring a path through the aneurysm based on the path pair, obtaining the restored path includes: performing edge node removal and interpolation operations on the path pair to obtain new path points; generating a sphere with all points on the path pair after adding the new path points as the center and according to the corresponding radius; and performing a union operation on the spheres to restore the path through the aneurysm to obtain the restored path.

[0010] In some embodiments, the path pairs are subjected to edge node removal by the following operations: taking the first point on each path in the path pair as the starting point; removing the edge nodes on each path in the path pair whose distance from the corresponding starting point is less than the radius of the corresponding starting point, so as to remove the edge nodes of the path pair.

[0011] In some embodiments, the path pair is interpolated by adding a target number of points containing a target radius between the first points on each path in the path pair after removing the edge nodes, so as to interpolate the path pair.

[0012] In some embodiments, the operation of segmenting the aneurysm based on the restored path to obtain the aneurysm segmentation result includes: performing a union operation on the restored path to obtain a connected domain; and segmenting the aneurysm based on the distance from the positioning point to the connected domain to obtain the aneurysm segmentation result.

[0013] In some embodiments, wherein the aneurysm is segmented based on the distance from the positioning point to the connected domain, the operation of obtaining the aneurysm segmentation result includes: in response to the distance from the target positioning point to the target connected domain being the smallest, determining the target connected domain as the aneurysm segmentation result.

[0014] In a second aspect, the present application provides a device comprising a processor and a memory, wherein the memory stores computer instructions for segmenting an aneurysm, and when the computer instructions are executed by the processor, one or more embodiments of the aforementioned first aspect are implemented.

[0015] In a third aspect, the present application provides a computer-readable storage medium having stored thereon computer program instructions for segmenting an aneurysm, wherein when the computer program instructions are executed by one or more processors, one or more embodiments of the aforementioned first aspect are implemented.

[0016] Through the above-provided solution for optimizing the neck surface of an aneurysm, the embodiment of the present application generates a vascular graph structure containing at least a leaf node with a radius attribute according to a vascular segmentation mask, determines the initial path node inside the aneurysm according to the positioning point, and determines the target path from the initial path node to the leaf node. The path pair based on the target path restores the path through the aneurysm to segment the aneurysm. Based on this, the vascular topology and size characteristics can be accurately presented to ensure that the segmentation logic fits the anatomical characteristics, focusing on the main path set that is closely related to the aneurysm morphology, and avoiding the misinclusion of the parent artery. Through path restoration, the connection relationship between the aneurysm and the surrounding blood vessels can be reconstructed, the spatial morphological contour of the aneurysm can be fitted, and the complete segmentation of the intracranial large aneurysm can be achieved, effectively solving the problems of incomplete tumor segmentation and mis-segmentation of the parent artery in the prior art, thereby stably obtaining the blood vessel conditions around the aneurysm and providing reliable image segmentation results for clinical diagnosis and treatment planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood by reading the detailed description below with reference to the accompanying drawings. 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 flow chart illustrating a method for segmenting an aneurysm according to an embodiment of the present application;

[0019] Figure 2 is an exemplary schematic diagram showing the structure of a blood vessel map according to an embodiment of the present application;

[0020] Figure 3 is a flowchart illustrating an exemplary process of determining a target path according to an embodiment of the present application;

[0021] Figure 4 is a flowchart illustrating an exemplary method of restoring a path through an aneurysm and segmenting the aneurysm based on a path pair according to an embodiment of the present application;

[0022] Figure 5 is a block diagram showing an exemplary structure of a device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0024] It should be understood that the terms "include" and "comprising" used in the description and claims of this application indicate the presence of 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 collections thereof.

[0025] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this specification and claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" as used in this specification and claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0026] As used in this specification and claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0027] The specific implementation of the present application will be described in detail below with reference to the accompanying drawings.

[0028] Figure 1 FIG. 1 is an exemplary flow chart showing a method 100 for segmenting an aneurysm according to an embodiment of the present application. Figure 1 As shown in , the method 100 includes: step S101: generating a vascular graph structure based on a vascular segmentation mask, wherein the vascular graph structure includes at least leaf nodes including a radius attribute; step S102: determining an initial path node inside the aneurysm based on a positioning point; step S103: screening a target path from the initial path node to the leaf node, and sorting the target path according to the average radius to form a path pair; step S104: restoring a path through the aneurysm based on the path pair to obtain a restored path; and step S105: segmenting the aneurysm based on the restored path to obtain an aneurysm segmentation result.

[0029] First, in step S101, a vessel map structure is generated based on a vessel segmentation mask. In some implementation scenarios, a vessel segmentation mask can be obtained by using threshold segmentation, deep learning, or other methods to extract a binary image of vessels from medical images (e.g., DSA, CTA), marking the vessel area as 1 and the background as 0.

[0030] In other implementation scenarios, a segmentation tool (such as the sknw tool, a Python library that focuses on building networks from skeleton images and performing analysis) can be used to generate a vascular graph structure based on the vascular segmentation mask. The vascular graph structure includes leaf nodes (nodes with only one adjacent node) and edge nodes (the remaining nodes), and each node has a radius attribute. The radius is the shortest distance from the node to the surface of the blood vessel, which is determined by calculating the Euclidean distance of the node to the edge of the blood vessel mask in three-dimensional space. As an example, if the vascular graph structure G contains m leaf nodes, the leaf nodes are denoted by n1~nm, and the edges are represented by (nj,nk), where 1<=j, k<=m. There are L edge nodes on the edge (nj,nk), denoted by e1~eL, and their radii are r1~rL.

[0031] In other implementation scenarios, the length of the edge (nj,nk) can be determined. If the length is less than 10 mm or less than (rj + rk), a new point nm+1 is added, positioned at the center of nj and nk. The adjacent points of nj and nk are set to the adjacent points of nm+1 to generate a vascular graph. The generated vascular graph quantifies vessel thickness using radius attributes and locates vessel ends at leaf nodes, providing a precise topological and geometric foundation for subsequent path analysis, avoiding misjudgment of aneurysm boundaries due to fuzzy vascular structures.

[0032] Then, at step S102, the initial path node inside the aneurysm is determined based on the positioning point. In some implementation scenarios, the user can mark the positioning point [x0, y0, z0] on the medical image through an interactive method (such as mouse click), or generate the positioning point through an automatic detection algorithm (such as key point detection based on aneurysm features). Based on the generated positioning point, the Euclidean distance from all leaf nodes n1~nm in the vascular structure graph G to the positioning point is calculated, and the node np corresponding to the minimum distance is found as the initial path node inside the aneurysm. Therefore, by determining the initial node based on the positioning point, it is ensured that the segmentation starting point is located in the core area of ​​the aneurysm, avoiding incomplete segmentation due to the offset of the starting point, and improving the fit of the segmentation result with clinical needs.

[0033] Based on the initial path node within the aneurysm determined above, at step S103, target paths from the initial path node to the leaf node are screened, and the target paths are sorted by average radius to form path pairs. In some embodiments, the target paths from the initial path node to the leaf node can be screened by traversing the set of paths from the initial path node to the leaf node, calculating the length or average radius of each path in the path set, and screening the target paths from the initial path node to the leaf node based on the length or average radius of each path.

[0034] In some implementation scenarios, one can first start from the initial path node np and traverse the vascular structure graph using, for example, a breadth-first search or a depth-first search to obtain a set W of paths from the initial node np to the leaf nodes. Next, the total length and average radius (the arithmetic mean of the radii of the edge nodes on the path) of each path are calculated. The total length is the sum of the lengths of each edge, and the edge length is calculated by the node coordinate difference; the aforementioned average radius is the arithmetic mean of the radii of the edge nodes on the path. Furthermore, in some embodiments, paths whose length is less than a length threshold or whose average radius is less than a radius threshold are filtered out from the path set to obtain the target path from the initial path node to the leaf node.

[0035] In some implementation scenarios, the aforementioned length threshold may be, for example, 80 mm, and the aforementioned radius threshold may be, for example, 1 mm. For example, paths with a length of less than 80 mm or an average radius of less than 1 mm will be screened out from the path set to obtain the target path W'. After obtaining the target path W', the paths in the target path W' are sorted in descending order by the average radius. For example, the paths in W' are recorded as w1~wf, and the average radius is recorded as r1~rf, where rw1>rw2>...>rwf. Adjacent paths (wi, wi+1) are sequentially formed into path pairs. Based on this, through dual screening of length and average radius, small branches and noise paths are excluded, and the main paths related to the aneurysm body are focused on to avoid missegmentation of the parent artery. By sorting the paired paths by radius, it is ensured that the path pairs can characterize the main inflow / outflow channels of the aneurysm, providing a reliable path set for restoring the aneurysm morphology.

[0036] Further, at step S104, the path through the aneurysm is restored based on the path pair to obtain a restored path. In some embodiments, edge node removal and interpolation operations are performed on the path pair to obtain new path points, with all points on the path pair after the new path points are added as the center, and a sphere is generated according to the corresponding radius, and then a union operation is performed on the sphere to restore the path through the aneurysm to obtain a restored path. In some embodiments, the path pair can be subjected to the following operation to remove edge nodes: with the first point on each path in the path pair as the starting point, the edge nodes on each path in the path pair whose distance from the corresponding starting point is less than the radius of the corresponding starting point are removed to perform edge node removal on the path pair. In other embodiments, the path pair can be interpolated by the following operation: a target number of points containing a target radius are added between the first point on each path in the path pair after the edge nodes are removed to perform an interpolation operation on the path pair.

[0037] Specifically, for a path pair (wi, wi+1), taking wi's starting node np1 (radius rp) as an example, remove the edge nodes on wi whose distance from the starting node np1 is less than rp (i.e., the proximal node located within the aneurysm). Similarly, the starting node of wi+1 is processed to obtain the truncated path starting points eyi and eyi+1. Next, linear interpolation is performed between eyi and eyi+1 using the Euclidean distance D(eyi, eyi+1). The radius of each new point is set to (ri+ri+1) / 2, where ri and ri+1 are the radii of eyi and eyi+1. This yields the path pair after the addition of the new path point. The total number of points in the path pair after the addition of the new path point is Li+Li+1-yi-yi+1+D. By generating spheres with corresponding radii centered around all points in the path pair after the addition of the new path point, and then performing a union operation on the spheres, the path through the aneurysm is restored to obtain the restored path. Among them, redundant nodes inside the aneurysm are removed by edge node removal, adjacent paths are smoothly connected by interpolation operation, and the three-dimensional sphere union accurately restores the spatial connection form between the aneurysm and the parent artery, avoiding segmentation gaps caused by path discontinuity and improving the integrity of the aneurysm contour.

[0038] Finally, at step S105, the aneurysm is segmented based on the restored path to obtain an aneurysm segmentation result. In some embodiments, a connected domain can be obtained by performing a union operation on the restored path, and the aneurysm can be segmented based on the distance from the positioning point to the connected domain to obtain an aneurysm segmentation result. Specifically, in some embodiments, in response to the distance from the target positioning point to the target connected domain being the smallest, the target connected domain is determined as the aneurysm segmentation result. As an example, a logical NOT operation (1-vessel_seg) is performed on the restored path vessel_seg, and a logical AND operation is performed with the original vessel segmentation mask to obtain n connected domains, each of which corresponds to an area in the vascular network that is not covered by the restored path. Then, by calculating the shortest distance from the positioning point to each connected domain (taking the minimum Euclidean distance from all points in the connected domain to the positioning point), the connected domain j with the smallest distance is found and determined as the aneurysm segmentation result. Based on this, the connected domain is screened based on the distance between the positioning points to ensure that the segmentation result is the area closest to the aneurysm core, avoiding misjudging the distal blood vessels as aneurysms; morphological processing eliminates segmentation noise, making the segmentation results more in line with clinical diagnostic requirements and providing an accurate basis for aneurysm volume calculation, aneurysm neck analysis, etc.

[0039] In combination with the above description, it can be seen that the embodiment of the present application generates a vascular graph structure containing leaf nodes with radius attributes based on the vascular segmentation mask, which can accurately characterize the topological structure and size characteristics of the blood vessels, and provide an accurate geometric basis for subsequent path analysis; based on the positioning point, the initial path node inside the aneurysm is determined, and the vascular path can be explored from the core area of ​​the aneurysm to ensure the accuracy of the segmentation starting point; the target path from the initial path node to the leaf node is screened and sorted by the average radius to form a path pair, which can effectively filter out irrelevant small branches and noise paths, focus on the aneurysm-related paths carrying the main blood flow, and avoid missegmentation of the tumor-bearing artery.

[0040] Furthermore, the path through the aneurysm is restored based on the path pair. Through edge node removal and interpolation operations, the morphological contour of the aneurysm can be accurately fitted, and its connection relationship with the surrounding blood vessels can be restored. Finally, the aneurysm is segmented based on the restored path. Through connected domain analysis and positioning point distance judgment, the aneurysm area can be accurately separated, achieving complete segmentation of large intracranial aneurysms with diverse shapes and complex necks. This effectively solves the problems of incomplete aneurysm segmentation and the mistaken inclusion of the parent artery in the segmentation results in the existing technology, thereby stably obtaining the blood vessel conditions around the aneurysm and providing reliable imaging basis for clinical diagnosis and treatment. In some implementation scenarios, the aneurysm segmentation results can also be eroded and expanded to remove noise, smooth the contours, and retain the main structure.

[0041] Figure 2 FIG. 1 is an exemplary schematic diagram showing the structure of a blood vessel map according to an embodiment of the present application. Figure 2 As shown in the figure, the vascular graph structure may include leaf nodes (e.g., indicated by arrow A in the figure) and edge nodes (e.g., indicated by arrow B in the figure). As previously described, a vascular segmentation mask can be skeletonized using tools such as sknw to generate the vascular graph structure. Based on the generated vascular graph structure and the anchor points, target paths from initial path nodes to leaf nodes are determined, and path pairs are identified. Paths through the aneurysm are then restored based on the path pairs. The aneurysm is segmented by restoring the restored path, resulting in an aneurysm segmentation result.

[0042] Figure 3 FIG. 1 is an exemplary flow chart illustrating a method for determining a target path according to an embodiment of the present application. Figure 3 As shown in FIG, in step S301, a positioning point is obtained. In some implementations, the positioning point can be obtained by marking the positioning point on a medical image or automatically detecting key points of the aneurysm. Next, in step S302, the Euclidean distance from the positioning point to all leaf nodes in the vascular structure graph G is calculated. In step S303, the node with the minimum distance is determined as the initial path node inside the aneurysm.

[0043] Based on the initial path node obtained above, in step S304, the path set W from the initial path node to the leaf node is traversed. In step S305, the length or average radius of each path in the path set is calculated. In step S306, paths with a length less than 80 mm or an average radius less than 1 mm are removed from the path set to obtain a target path W'. Based on this target path W', the paths in the target path W' are sorted in descending order of average radius, and adjacent paths (wi, wi+1) are sequentially formed into path pairs to restore the path through the aneurysm for segmentation.

[0044] Figure 4 FIG. 1 is an exemplary flow chart showing how to restore a path through an aneurysm and segment the aneurysm based on a path pair according to an embodiment of the present application. Figure 4 As shown in , at step S401, edge node removal and interpolation operations are performed on the path pair to obtain new path points. Specifically, with the first point on each path in the path pair as the starting point, the edge nodes on each path in the path pair whose distance from the corresponding starting point is less than the radius of the corresponding starting point are removed to perform edge node removal on the path pair. A target number of points containing a target radius are added between the first points on each path in the path pair after the edge nodes are removed to perform an interpolation operation on the path pair. The target number is D new points generated by linear interpolation according to the Euclidean distance D(eyi,eyi+1), and the target radius is (ri+ri+1) / 2.

[0045] In step S402, spheres are generated with corresponding radii, centered around all points on the path pair after the addition of the new path point. In step S403, a union operation is performed on the spheres to restore the path through the aneurysm and obtain a restored path. Based on the restored path, a union operation is performed on the restored paths in step S404 to obtain connected components. Furthermore, in step S405, the shortest distance from the anchor point to each connected component is calculated. In step S406, the connected component j with the shortest distance is determined as the aneurysm segmentation result.

[0046] Figure 5 is an exemplary structural block diagram of a device 500 according to an embodiment of the present application. It is understood that the device 500 may include the apparatus of the embodiment of the present application, and the device implementing the solution of the present application may be a single device (such as a computing device) or a multifunctional device including various peripheral devices.

[0047] like Figure 5As shown in , the device of the present application may also include a central processing unit ("CPU") 511, which may be a general-purpose CPU, a dedicated CPU, or other execution unit for information processing and program execution. Furthermore, the device 500 may also include a mass storage 512 and a read-only memory ("ROM") 513, wherein the mass storage 512 may be configured to store various types of data, including various data related to the vascular map structure, initial path nodes, target paths, path pairs, restored paths, aneurysm segmentation results, 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, basic input / output drivers for the system, and booting the operating system.

[0048] Optionally, the device 500 may further include other hardware platforms or components, such as a tensor processing unit ("TPU") 514, a graphics processing unit ("GPU") 515, a field programmable gate array ("FPGA") 516, and a machine learning unit ("MLU") 517. It will be appreciated that although various hardware platforms or components are shown in the device 500, these are merely exemplary and non-limiting, and those skilled in the art may add or remove corresponding hardware as needed. For example, the device 500 may include only a CPU, related storage devices, and interface devices to implement the method for segmenting an aneurysm of the present application.

[0049] In some embodiments, to facilitate data transmission and interaction with external networks, the device 500 of the present application further includes a communication interface 518, which can be connected to a local area network / wireless local area network ("LAN / WLAN") 505 via the communication interface 518, and further connected to a local server 506 or the Internet ("Internet") 507 via 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 via the communication interface 518 based on wireless communication technology, such as third generation ("3G"), fourth generation ("4G"), or fifth generation ("5G") wireless communication technology. In some application scenarios, the device 500 of the present application can also access a server 508 and a database 509 on an external network as needed to obtain various known algorithms, data, and modules, and can remotely store various data, such as various data or instructions for presenting, for example, vascular map structures, initial path nodes, target paths, path pairs, restored paths, aneurysm segmentation results, etc.

[0050] 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 provide voice prompts and / or image video displays for the segmentation of aneurysms 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 aneurysm images collected from CT or MRI equipment, and transmit data or results including aneurysm images or various other types to the device 500.

[0051] 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 via a bus 519 and can interact with peripheral devices via the bus. In one embodiment, the CPU 511 can control other hardware components in the device 500 and its peripheral devices via the bus 519.

[0052] Combination of the above Figure 5 The devices that can be used to implement the present application are described. It should be understood that the device structures or architectures herein are merely exemplary, and the implementation methods and implementation entities of the present application are not limited thereto, but can be modified without departing from the spirit of the present application.

[0053] 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 software programs. Therefore, the present application also provides a computer-readable storage medium, which stores computer-readable instructions for segmenting aneurysms. When the computer-readable instructions are executed by one or more processors, they can be used to implement the present application in combination with the accompanying drawings. Figure 1 A method for segmenting an aneurysm is described.

[0054] It should be noted that although the operations of the present method are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in that particular order, or that all of the operations shown must be performed to achieve the desired results. Rather, the steps depicted in the flowcharts may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into a single step, and / or a single step may be broken down into multiple steps.

[0055] It should be understood that when the terms "first," "second," "third," and "fourth," etc., are used in the claims, specification, and drawings of this application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the specification and claims of this application indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0056] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this specification and claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" as used in this specification and claims refers to any and all possible combinations of one or more of the associated listed items, including and including these combinations.

[0057] Although the implementation methods of this application are as described above, the contents are only examples adopted to facilitate understanding of this application and are not intended to limit the scope and application scenarios of this application. Any technician in the technical field described in this application can make any modifications and changes in the form and details of implementation without departing from the spirit and scope disclosed in this application, but the scope of patent protection of this application shall still be based on the scope defined by the attached claims.

[0058] In addition, the collection and acquisition of various data in this application complies with relevant laws and regulations and is authorized by the data provider. Any organization or individual that needs to obtain external data must obtain authorization in accordance with the law and ensure data security. They must not illegally collect, use, process, or transmit unauthorized or unprotected data, nor illegally buy, sell, provide, or disclose unauthorized or unprotected data.

Claims

1. A method for segmenting an aneurysm, comprising: generating a vessel graph structure according to the vessel segmentation mask, wherein the vessel graph structure at least includes a leaf node including a radius attribute; Determine the initial path node inside the aneurysm based on the positioning point; Filtering target paths from the initial path node to the leaf node, and sorting the target paths according to average radius to form path pairs; restoring a path through the aneurysm based on the path pair to obtain a restored path; The aneurysm is segmented based on the restored path to obtain an aneurysm segmentation result.

2. The method according to claim 1, wherein the target path from the initial path node to the leaf node is screened by the following operations: Traversing the path set from the initial path node to the leaf node; Calculating the length or average radius of each path in the path set; The target path from the initial path node to the leaf node is filtered according to the length or average radius of each path.

3. The method according to claim 2, wherein screening the target path from the initial path node to the leaf node according to the length or average radius of each path comprises: Paths whose lengths are less than a length threshold or whose average radii are less than a radius threshold are screened out from the path set to obtain a target path from the initial path node to the leaf node.

4. The method according to claim 1 , wherein the vascular graph structure further comprises edge nodes including a radius attribute, and restoring a path through the aneurysm based on the path pair, wherein obtaining the restored path comprises: Performing edge node removal and interpolation operations on the path pair to obtain new path points; Generate a sphere with the corresponding radius, centering on all points on the path pair after adding the new path point; A union operation is performed on the spheres to restore the path through the aneurysm to obtain a restored path.

5. The method according to claim 4, wherein the edge nodes of the path pair are removed by the following operation: Taking the first point on each path in the path pair as the starting point; The edge nodes on each path in the path pair whose distance from the corresponding starting point is less than the radius of the corresponding starting point are removed, so as to perform edge node removal on the path pair.

6. The method according to claim 4, wherein the path pair is interpolated by: A target number of points including a target radius are added between the first points on each path in the path pair after removing the edge nodes, so as to perform an interpolation operation on the path pair.

7. The method according to claim 1, wherein segmenting the aneurysm based on the restored path and obtaining an aneurysm segmentation result comprises: Performing a union operation on the restored paths to obtain a connected domain; The aneurysm is segmented based on the distance from the positioning point to the connected domain to obtain an aneurysm segmentation result.

8. The method according to claim 7, wherein segmenting the aneurysm based on the distance from the positioning point to the connected domain, and obtaining the aneurysm segmentation result comprises: In response to the distance between the target positioning point and the target connected domain being the smallest, the target connected domain is determined as the aneurysm segmentation result.

9. A device comprising: processor; as well as A memory having computer instructions for segmenting an aneurysm stored thereon, which, when executed by a processor, enable the method according to any one of claims 1 to 8 to be implemented.

10. A computer-readable storage medium having stored thereon computer program instructions for segmenting an aneurysm, wherein when the computer program instructions are executed by one or more processors, the method according to any one of claims 1 to 8 is implemented.

Images