Hepatic artery model processing method, device and server

Abstract

The present application provides a method, device, and server for processing a liver artery model, which relates to medical image processing technology. The method includes: obtaining a liver artery model reconstruction request; the liver artery model reconstruction request includes a liver artery data identifier; determining the liver artery data corresponding to the liver artery data identifier based on the liver artery model reconstruction request, the liver artery data including a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline. Determine a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline. Determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model; the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. The method of the present application can obtain a target hepatic artery vascular model corresponding to a higher-level hepatic artery, so as to facilitate determination of hepatic artery vascular information and simulation calculations.

Description

Technical Field

[0001] The present application relates to medical image processing technology, and in particular to a method, device and server for processing a liver artery model. Background Art

[0002] Currently, among many medical examination methods, computed tomography (CT) and magnetic resonance imaging (MRI) are both important means of imaging examination. Therefore, the liver can be examined based on CT data and MRI data.

[0003] In the existing technology, when examining the liver based on CT data and MRI data, a hepatic artery model can be reconstructed. Due to the complex structure of the liver tissue itself, the different shapes and sizes of the livers of different people, the variation of blood vessels, and the similar density of the liver and adjacent heart tissue, these problems lead to poor quality of the actual CT image of the liver, which in turn leads to a lower grade of the reconstructed hepatic artery. While magnetic resonance imaging has good resolution for soft tissue, the MR scan layer thickness is large, and the grayscale continuity of the microvascular structure is poor. It can only view the approximate range of grade 3-4 of the hepatic artery, making it difficult to use for accurate data collection. Therefore, the hepatic artery can generally only be divided into grade 2-3 hepatic artery vessels based on digital reconstruction, and cannot be further divided into higher-level vessels. It is necessary to determine a higher-grade hepatic artery model to improve the accuracy of the hepatic artery model.

[0004] Therefore, there is an urgent need for a method to improve the accuracy of the hepatic artery model. Summary of the Invention

[0005] The present application provides a method, device and server for processing a hepatic artery model to solve the technical problem of low accuracy of the hepatic artery model.

[0006] In a first aspect, the present application provides a method for processing a hepatic artery model, comprising:

[0007] Obtaining a liver vascular model reconstruction request; wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and determining, based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein a grade of a hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than a grade of a hepatic artery corresponding to the first hepatic artery model, and a grade of a hepatic artery corresponding to the first centerline is higher than a grade of a hepatic artery corresponding to the first hepatic artery model;

[0008] determining a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model;

[0009] Based on the first hepatic artery model and the second hepatic artery model, a target hepatic artery vascular model is determined to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery.

[0010] Furthermore, determining a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline includes:

[0011] Obtain a liver parenchymal model;

[0012] determining a plurality of voxel points within the liver parenchymal model;

[0013] determining, based on the first hepatic artery model, a plurality of starting seed points located at the distal end of the first hepatic artery model;

[0014] Based on the multiple starting seed points, the preset centerline condition information, and the preset vascular tree condition information, a target voxel point among the multiple voxel points is determined, and the target voxel point is added to a seed point queue; wherein the preset centerline condition information represents conditional rule information about the first centerline, and the preset vascular tree condition information represents conditional rule information about the simulated hepatic artery vascular tree;

[0015] A second hepatic artery model is determined based on the plurality of seed point queues.

[0016] Furthermore, determining a target voxel point among the multiple voxel points based on the multiple starting seed points, the preset centerline condition information, and the preset vascular tree condition information, and adding the target voxel point to a seed point queue includes:

[0017] Based on the plurality of starting seed points, determining a neighborhood of a preset range of each starting seed point; wherein each of the preset ranges includes at least one voxel point;

[0018] If it is determined that a first target voxel point that satisfies both the preset centerline condition information and the preset blood vessel tree condition information exists in the neighborhood, adding the first target voxel point to a seed point queue;

[0019] If it is determined that the first target voxel point does not exist in the neighborhood, determining whether a second target voxel point that meets the preset centerline condition information exists in the neighborhood;

[0020] If it is determined that the second target voxel point exists, adding the second target voxel point to the seed point queue;

[0021] If it is determined that the second target voxel point does not exist, determining whether there is a third target voxel point that meets the preset blood vessel tree condition information in the neighborhood;

[0022] If it is determined that the third target voxel point exists, adding the third target voxel point to the seed point queue;

[0023] The newly added target voxel point in the seed point queue is updated as the starting seed point, and the process of determining a neighborhood of a preset range of each starting seed point based on the multiple starting seed points, determining a target voxel point within the neighborhood, and updating the target voxel point as the starting seed point is repeated until there is no newly added starting seed point.

[0024] Furthermore, the preset centerline condition information includes: the voxel point is located on the first centerline or on a line connecting the first centerline and a starting seed point located at the distal end of the first hepatic artery model;

[0025] The preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on a line connecting the simulated hepatic artery vascular tree and a starting seed point located at the end of the first hepatic artery model.

[0026] Furthermore, determining the second hepatic artery model based on the plurality of seed point queues includes:

[0027] Acquire a second center line corresponding to each seed point queue in the plurality of seed point queues;

[0028] The second hepatic artery model is generated based on the second centerline corresponding to each seed point queue.

[0029] Furthermore, the grade of at least a portion of the hepatic arteries in the simulated hepatic artery tree is the same as the grade of the hepatic artery corresponding to the first centerline.

[0030] Furthermore, the first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery, and the hepatic arteries corresponding to both the first centerline and the simulated hepatic artery vascular tree include a third-order hepatic artery and / or a fourth-order hepatic artery.

[0031] Furthermore, the method further comprises:

[0032] Based on the simulated hepatic artery vascular tree, a third hepatic artery model is obtained.

[0033] Furthermore, the grade of the hepatic artery corresponding to the third hepatic artery model is higher than the grade of the hepatic artery corresponding to the second hepatic artery model.

[0034] Furthermore, the third hepatic artery model includes a fifth-order hepatic artery.

[0035] Furthermore, determining a target hepatic artery model based on the first hepatic artery model and the second hepatic artery model includes:

[0036] The target hepatic artery vessel model is obtained based on the first hepatic artery model, the second hepatic artery model, and the third hepatic artery model.

[0037] Furthermore, the method further comprises:

[0038] Acquire first image data and second image data;

[0039] A first hepatic artery model and a simulated hepatic artery vascular tree are generated based on the first image data, and a first centerline is generated based on the second image data.

[0040] In a second aspect, the present application provides a device for processing a hepatic artery model, comprising:

[0041] a first acquisition unit configured to acquire a liver vascular model reconstruction request, wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and determine, based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein a grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than a grade of the hepatic artery corresponding to the first hepatic artery model, and a grade of the hepatic artery corresponding to the first centerline is higher than a grade of the hepatic artery corresponding to the first hepatic artery model;

[0042] a first determining unit, configured to determine a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model;

[0043] a second determining unit, configured to determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein a grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to a grade of the hepatic artery corresponding to the second hepatic artery.

[0044] Furthermore, the first determining unit includes:

[0045] A first acquisition module is used to acquire a liver parenchymal model;

[0046] A first determining module is used to determine a plurality of voxel points in the liver parenchyma model;

[0047] a second determining module, configured to determine, based on the first hepatic artery model, a plurality of starting seed points located at the distal end of the first hepatic artery model;

[0048] a third determining module, configured to determine a target voxel point among the plurality of voxel points based on the plurality of starting seed points, preset centerline condition information, and preset vascular tree condition information, and add the target voxel point to a seed point queue; wherein the preset centerline condition information represents conditional rule information regarding the first centerline, and the preset vascular tree condition information represents conditional rule information regarding the simulated hepatic artery vascular tree;

[0049] The fourth determination module is configured to determine a second hepatic artery model based on the plurality of seed point queues.

[0050] Furthermore, the third determining module includes:

[0051] A first determining submodule is configured to determine a neighborhood of a preset range of each of the starting seed points based on the plurality of starting seed points; wherein each of the preset ranges includes at least one voxel point;

[0052] a first adding submodule, configured to add the first target voxel point to a seed point queue if it is determined that a first target voxel point that satisfies both the preset centerline condition information and the preset blood vessel tree condition information exists in the neighborhood;

[0053] a second determining submodule, configured to determine whether a second target voxel point that satisfies the preset centerline condition information exists in the neighborhood if it is determined that the first target voxel point does not exist in the neighborhood;

[0054] a second adding submodule, configured to add the second target voxel point to a seed point queue if it is determined that the second target voxel point exists;

[0055] a third determining submodule, configured to determine whether there is a third target voxel point that satisfies the preset vascular tree condition information in the neighborhood if it is determined that the second target voxel point does not exist;

[0056] a third adding submodule, configured to add the third target voxel point to a seed point queue if it is determined that the third target voxel point exists;

[0057] The fourth determination submodule is used to update the newly added target voxel point in the seed point queue as the starting seed point, and repeatedly determine the neighborhood of the preset range of each starting seed point based on multiple starting seed points, and determine the target voxel point within the neighborhood, and update the target voxel point to the starting seed point until there is no newly added starting seed point.

[0058] Furthermore, the preset centerline condition information includes: the voxel point is located on the first centerline or on a line connecting the first centerline and a starting seed point located at the distal end of the first hepatic artery model;

[0059] The preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on a line connecting the simulated hepatic artery vascular tree and a starting seed point located at the end of the first hepatic artery model.

[0060] Furthermore, the fourth determining module includes:

[0061] A first acquisition submodule is configured to acquire a second center line corresponding to each seed point queue in the plurality of seed point queues;

[0062] A generating submodule is configured to generate the second hepatic artery model based on the second center line corresponding to each seed point queue.

[0063] Furthermore, the grade of at least a portion of the hepatic arteries in the simulated hepatic artery tree is the same as the grade of the hepatic artery corresponding to the first centerline.

[0064] Furthermore, the first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery, and the hepatic arteries corresponding to both the first centerline and the simulated hepatic artery vascular tree include a third-order hepatic artery and / or a fourth-order hepatic artery.

[0065] Furthermore, the device further comprises:

[0066] The second acquisition unit is configured to acquire a third hepatic artery model based on the simulated hepatic artery vascular tree.

[0067] Furthermore, the grade of the hepatic artery corresponding to the third hepatic artery model is higher than the grade of the hepatic artery corresponding to the second hepatic artery model.

[0068] Furthermore, the third hepatic artery model includes a fifth-order hepatic artery.

[0069] Furthermore, the second determining unit is specifically configured to:

[0070] The target hepatic artery vessel model is obtained based on the first hepatic artery model, the second hepatic artery model, and the third hepatic artery model.

[0071] Furthermore, the device further comprises:

[0072] a third acquiring unit, configured to acquire the first image data and the second image data;

[0073] A generating unit is configured to generate a first hepatic artery model and a simulated hepatic artery vascular tree according to the first image data, and to generate a first centerline according to the second image data.

[0074] In a third aspect, the present application provides a server comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0075] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0076] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which implements the method described in the first aspect when executed by a processor.

[0077] The present application provides a method, device, and server for processing a hepatic artery model. The method comprises obtaining a hepatic artery model reconstruction request, wherein the hepatic artery model reconstruction request includes a hepatic artery data identifier. Based on the hepatic artery model reconstruction request, the method comprises determining hepatic artery data corresponding to the hepatic artery data identifier. The hepatic artery data comprises a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline. The hepatic artery grade corresponding to the simulated hepatic artery vascular tree is higher than the grade corresponding to the first hepatic artery model, and the hepatic artery grade corresponding to the first centerline is higher than the grade corresponding to the first hepatic artery model. Based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, a second hepatic artery model is determined. The grade corresponding to the second hepatic artery model is higher than the grade corresponding to the first hepatic artery model. Based on the first and second hepatic artery models, a target hepatic artery model is determined to determine hepatic artery vascular information based on the target hepatic artery model. The grade corresponding to the hepatic artery of the target hepatic artery vascular model is greater than or equal to the grade corresponding to the second hepatic artery. In this solution, based on the liver vascular data identifier obtained in the liver vascular model reconstruction request, the liver vascular data corresponding to the liver vascular data identifier is first determined. The liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline. Then, based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, a second hepatic artery model is determined, in which the hepatic artery is of a higher grade. Finally, the first and second hepatic artery models are integrated to determine a target hepatic artery vascular model. The hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. Therefore, based on the obtained first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, the hepatic artery vascular system, including the higher-grade hepatic artery, can be accurately reconstructed. This results in a target hepatic artery vascular model corresponding to the higher-grade hepatic artery, facilitating the determination of hepatic artery vascular information and simulation calculations, thereby resolving the technical issue of low accuracy in the hepatic artery vascular model. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0079] Figure 1 A schematic flow chart of a method for processing a hepatic artery model provided in an embodiment of the present application;

[0080] Figure 2 A schematic diagram of a liver artery model reconstructed in the prior art;

[0081] Figure 3A schematic diagram of a target hepatic artery model reconstructed in an embodiment of the present application;

[0082] Figure 4 A schematic flow chart of another method for processing a hepatic artery model provided in an embodiment of the present application;

[0083] Figure 5 1 is a flow chart of another method for processing a hepatic artery model provided in this embodiment;

[0084] Figure 6 This is a process diagram of region growth provided by this embodiment;

[0085] Figure 7 This is another process diagram of region growing provided by this embodiment;

[0086] Figure 8 This is another process diagram of region growth provided by this embodiment;

[0087] Figure 9 A schematic structural diagram of a liver artery model processing device provided in an embodiment of the present application;

[0088] Figure 10 A schematic structural diagram of another device for processing a liver artery model provided in an embodiment of the present application;

[0089] Figure 11 A schematic structural diagram of another device for processing a hepatic artery model provided in an embodiment of the present application;

[0090] Figure 12 A schematic diagram of the structure of a server provided in an embodiment of the present application.

[0091] The above drawings illustrate specific embodiments of the present disclosure, which will be described in more detail below. These drawings and textual descriptions are not intended to limit the scope of the present disclosure in any way, but rather to illustrate the concepts of the present disclosure to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0092] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present disclosure.

[0093] Currently, among many medical examination methods, computed tomography (CT) and magnetic resonance imaging (MRI) are both important means of imaging examination. Therefore, the liver can be examined based on CT data and MRI data.

[0094] In one example, when examining the liver based on CT and MRI data, a low-grade hepatic artery model can be reconstructed. However, due to the complex structure of liver tissue, the varying shapes and sizes of livers, vascular variations, and the similar density of the liver and adjacent cardiac tissue, these issues result in poor CT image quality. In particular, the hepatic artery is thinner than the portal vein, and CT sampling often loses much of the arterial detail. While the portal vein can typically be reconstructed at levels 3-4, only the hepatic artery can be reconstructed at levels 1-2, resulting in a low-grade reconstructed hepatic artery. While MRI offers good resolution for soft tissue, its thicker slices and poor grayscale continuity of microvascular structures limit the approximate range of levels 3-4, making it difficult to capture precise data. It is primarily used to annotate soft tissue and lesion areas. Therefore, digital reconstruction of the hepatic artery is generally limited to level 2-3, and cannot be segmented to higher levels. Therefore, a higher-grade hepatic artery model is needed to improve its accuracy.

[0095] The present application provides a method, device, and server for processing a liver artery model, aiming to solve the above technical problems of the prior art.

[0096] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0097] Figure 1 A schematic diagram of a process for processing a liver artery model provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0098] Step 101: Obtain a liver vascular model reconstruction request; wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and, based on the liver vascular model reconstruction request, determine liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0099] Exemplarily, the execution entity of this embodiment may be a server. First, a liver vascular model reconstruction request needs to be obtained. Specifically, a user clicks a liver vascular model reconstruction button on a display screen corresponding to the server, triggering the generation of a liver vascular model reconstruction request. The server then obtains the liver vascular model reconstruction request. Alternatively, the server receives a liver vascular model reconstruction request sent by another server. The liver vascular model reconstruction request includes a liver vascular data identifier.

[0100] In this step, the server determines liver vascular data corresponding to the liver vascular data identifier based on the liver vascular data identifier in the liver vascular model reconstruction request. The liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline. The grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. The grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. The grade of at least a portion of the hepatic artery in the simulated hepatic artery vascular tree is the same as the grade of the hepatic artery corresponding to the first centerline. The first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery. The hepatic arteries corresponding to both the first centerline and the simulated hepatic artery vascular tree include a third-order hepatic artery and / or a fourth-order hepatic artery. The fact that the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model means that the grade of at least some of the branch vessels included in the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of all the branch vessels included in the first hepatic artery model. For example, the hepatic artery corresponding to the simulated hepatic artery vascular tree includes level 3 branch vessels and level 4 branch vessels, and the hepatic artery corresponding to the first hepatic artery model includes level 1 branch vessels and level 2 branch vessels. Typically, in some embodiments, the hepatic artery corresponding to the simulated hepatic artery vascular tree may also include level 1 branch vessels, level 2 branch vessels, level 3 branch vessels, and level 4 branch vessels. It should be noted that, in the embodiments of the present application, the grade of one hepatic artery being higher than the grade of another hepatic artery means that the grade of at least some of the branch vessels of the hepatic artery is higher than the grade of all the branch vessels of the other hepatic artery.

[0101] The server also needs to pre-acquire first image data and second image data. The first image data is different from the second image data. The first image data is computed tomography (CT) data, and the second image data is magnetic resonance imaging (MRI) data. The server can generate a first hepatic artery model and a portal vein model based on the first image data, obtain a simulated hepatic artery vascular tree based on the portal vein model, and generate a first centerline based on the second image data. The first centerline is the vascular centerline of the hepatic artery.

[0102] Figure 2 is a schematic diagram of a hepatic artery model (including first-order branches and second-order branches) reconstructed in the prior art. Figure 3 Schematic diagram of the target hepatic artery model (including level 1 branches, level 2 branches, level 3 branches, level 4 branches and level 5 branches) reconstructed in the embodiment of the present application; specifically, Figure 2 and Figure 3 As shown, the present application defines the hepatic artery classification as the left hepatic artery (LHA), middle hepatic artery (MHA), and right hepatic artery (RHA) as first-level branch vessels, the right anterior hepatic artery, right posterior hepatic artery and other branches as second-level branch vessels, and the third, fourth, and fifth-level branch vessels are analogous. In layman's terms, the vessels between adjacent bifurcation nodes are vessels of the same level, and the grade of the vessels after the bifurcation node is higher than the grade of the vessels before the bifurcation node. In some embodiments, Figure 3 The target hepatic artery vessel model in may include the tertiary hepatic artery. In other embodiments, Figure 3 The target hepatic artery vascular model in may include the tertiary hepatic artery and the quaternary hepatic artery, or may further include the tertiary hepatic artery, the quaternary hepatic artery, and the quintile hepatic artery.

[0103] Step 102: Determine a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0104] Illustratively, since the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, the server can use a region growing algorithm to calculate a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, thereby obtaining a second hepatic artery model with a higher grade. There is no limitation on this calculation method. At this time, the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0105] For example, the region growing algorithm is an image segmentation technology. Its basic idea is to merge pixels with similar criteria to form regions based on certain judgment criteria. The main step is to find a seed pixel as the starting point of growth for each area to be segmented (that is, to find a pixel as a reference to determine whether other pixels are related to the reference pixel). Then, based on certain judgment criteria, similar pixels around the seed pixel are judged, and pixels with higher similarity are merged, so that they germinate and grow like seeds.

[0106] It should be noted that the first hepatic artery model and the second hepatic artery model are both three-dimensional reconstructed digital models, and the simulated vascular tree and the first centerline are both two-dimensional lines.

[0107] Step 103: Determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery.

[0108] Exemplarily, the server integrates the first hepatic artery model and the second hepatic artery model to determine a target hepatic artery vascular model, and then determines hepatic artery vascular information based on the target hepatic artery vascular model, making it convenient for users to obtain surgical plan reports, perform surgical navigation, perform surgical simulation, and other operations based on the hepatic artery vascular information. At this time, the grade of the hepatic artery corresponding to the target hepatic artery vascular model obtained is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery.

[0109] In an embodiment of the present application, a liver vascular model reconstruction request is obtained; the liver vascular model reconstruction request includes a liver vascular data identifier; and based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier is determined, the liver vascular data including a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. A second hepatic artery model is determined based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. Based on the first hepatic artery model and the second hepatic artery model, a target hepatic artery vascular model is determined, thereby determining hepatic artery vascular information based on the target hepatic artery vascular model, wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. In this solution, based on the liver vascular data identifier obtained in the liver vascular model reconstruction request, the liver vascular data corresponding to the liver vascular data identifier is first determined. The liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline. Then, based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, a second hepatic artery model is determined, in which the hepatic artery is of a higher grade. Finally, the first and second hepatic artery models are integrated to determine a target hepatic artery vascular model. The hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. Therefore, based on the obtained first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, the hepatic artery vascular system, including the higher-grade hepatic artery, can be accurately reconstructed. This results in a target hepatic artery vascular model corresponding to the higher-grade hepatic artery, facilitating the determination of hepatic artery vascular information and simulation calculations, thereby resolving the technical issue of low accuracy in the hepatic artery vascular model.

[0110] Figure 4 A flow chart of another method for processing a liver artery model provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the method includes:

[0111] Step 201: Acquire first image data and second image data.

[0112] Exemplarily, the server may pre-acquire first image data and second image data, where the first image data is different from the second image data, wherein the first image data is computed tomography (CT) data and the second image data is magnetic resonance imaging (MRI) data.

[0113] Step 202: Generate a first hepatic artery model and a simulated hepatic artery vascular tree based on the first image data, and generate a first centerline based on the second image data.

[0114] For example, the server can generate a first hepatic artery model and a portal vein model based on the first image data, and obtain a simulated hepatic artery vascular tree based on the portal vein model, wherein the grade of the hepatic artery corresponding to the portal vein model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, i.e., the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. Furthermore, the server can generate a first centerline based on the second image data.

[0115] For example, in this embodiment, the first image data is computed tomography (CT) data, i.e., the first hepatic artery model is generated based on the CT data. In this case, the first hepatic artery model includes the first-order hepatic artery and / or the second-order hepatic artery; the second image data may be magnetic resonance imaging (MRI) data, i.e., the first centerline is generated based on the MRI data. In this case, the hepatic artery corresponding to the first centerline includes the third-order hepatic artery and / or the fourth-order hepatic artery. In a specific embodiment, while the hepatic artery corresponding to the first hepatic artery model generally includes the first-order hepatic artery and the second-order hepatic artery, the hepatic artery corresponding to the first centerline may include the third-order hepatic artery and the fourth-order hepatic artery. Of course, in other embodiments, different CT data may have different accuracy and completeness. Therefore, the hepatic artery corresponding to the first hepatic artery model may also include only the first-order hepatic artery. In this case, the hepatic artery corresponding to the first centerline may include the second-order hepatic artery, the third-order hepatic artery, and the fourth-order hepatic artery.

[0116] Furthermore, a simulated hepatic artery vascular tree is obtained based on the portal vein model, and the grade of the portal vein corresponding to the portal vein model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model. The portal vein model is generated based on the first imaging data, that is, the portal vein model is generated based on the CT data. Specifically, since the portal vein is relatively thick, the CT data can capture more details, and the portal vein can usually be reconstructed to a higher grade based on the CT data, such as grade 3, grade 4, or even grade 5. Moreover, according to the triad anatomical structure of the portal vein, the hepatic artery and the portal vein are in a coexisting state. Therefore, the portal vein model can be reconstructed based on the first imaging data (i.e., CT data), and the grade of the portal vein corresponding to the portal vein model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model; further, a simulated hepatic artery vascular tree can be obtained by simulation and reconstruction based on the portal vein model, and the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the simulated hepatic artery vascular tree is higher than or equal to the grade of the portal vein corresponding to the portal vein model. For example, the hepatic artery corresponding to the first hepatic artery model reconstructed based on the first image data includes the first-level hepatic artery and the second-level hepatic artery, and the portal vein corresponding to the portal vein model reconstructed based on the first image data may include the first-level portal vein, the second-level portal vein, the third-level portal vein and the fourth-level portal vein. At this time, based on the accompanying state of the portal vein and the hepatic artery and the portal vein model, a simulated hepatic artery vascular tree can be simulated and reconstructed. The simulated hepatic artery vascular tree may include the first-level hepatic artery, the second-level hepatic artery, the third-level hepatic artery and the fourth-level hepatic artery, or, in other embodiments, the simulated hepatic artery vascular tree may only include the third-level hepatic artery and the fourth-level hepatic artery. This embodiment does not impose specific restrictions on this.

[0117] As described above, in this embodiment, since CT data cannot capture the details of the tertiary hepatic artery and the fourth hepatic artery, the tertiary hepatic artery and the fourth hepatic artery cannot be reconstructed. MRI data can view the tertiary hepatic artery and the fourth hepatic artery, but due to poor grayscale continuity, the tertiary hepatic artery and the fourth hepatic artery cannot be reconstructed. Therefore, in this embodiment, the centerline corresponding to the tertiary hepatic artery and the fourth hepatic artery is obtained by using MRI data, and the portal vein model is reconstructed by using CT data, and a simulated hepatic artery vascular tree corresponding to the tertiary hepatic artery and the fourth hepatic artery is obtained based on the portal vein model. Based on the centerline corresponding to the tertiary hepatic artery and the fourth hepatic artery and the simulated hepatic artery vascular tree, the tertiary hepatic artery model and the fourth hepatic artery model are reconstructed by using a region growing algorithm, thereby reconstructing a hepatic artery vascular model.

[0118] Step 203: Obtain a liver vascular model reconstruction request; the liver vascular model reconstruction request includes a liver vascular data identifier; and, based on the liver vascular model reconstruction request, determine liver vascular data corresponding to the liver vascular data identifier, the liver vascular data including a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0119] In one example, the grade of the hepatic artery in at least a portion of the simulated hepatic artery tree is the same as the grade of the hepatic artery corresponding to the first centerline.

[0120] In one example, the first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery, and the hepatic arteries corresponding to the first centerline and the simulated hepatic artery vascular tree both include a third-order hepatic artery and / or a fourth-order hepatic artery.

[0121] For example, this step can be referred to Figure 1 Step 101 in the above description will not be repeated.

[0122] Step 204: Acquire a liver parenchyma model.

[0123] Exemplarily, the server can obtain a liver parenchymal model, wherein the liver parenchymal model refers to the substantial part of the liver, and vascular models such as the first hepatic artery model and the second hepatic artery model are interspersed and embedded in the liver parenchymal model. The liver parenchymal model and the vascular model together constitute the liver model. The liver parenchymal model can be reconstructed based on CT data. In this embodiment, the liver parenchymal model is a three-dimensional digital model.

[0124] Step 205: Determine a plurality of voxel points in the liver parenchyma model.

[0125] For example, the server may determine a plurality of voxel points within the liver parenchymal model according to a preset accuracy, wherein the determined plurality of voxel points are evenly distributed within the liver parenchymal model, i.e., the space within the liver parenchymal model is evenly divided into the plurality of voxel points. The preset accuracy may mean that a unit length includes a preset number of voxel points.

[0126] Step 206 : Based on the first hepatic artery model, determine a plurality of starting seed points located at the distal end of the first hepatic artery model.

[0127] For example, taking the first hepatic artery model including first-order branch vessels and second-order branch vessels as an example, the starting seed point is a voxel point on the end of each second-order branch vessel, that is, there are multiple starting seed points.

[0128] Step 207: Based on multiple starting seed points, preset centerline condition information, and preset vascular tree condition information, determine a target voxel point among the multiple voxel points, and add the target voxel point to a seed point queue; wherein the preset centerline condition information represents conditional rule information about the first centerline, and the preset vascular tree condition information represents conditional rule information about the simulated hepatic artery vascular tree.

[0129] In one example, step 207 includes: determining a preset range neighborhood of each starting seed point based on multiple starting seed points, wherein each preset range includes at least one voxel point; if it is determined that a first target voxel point that satisfies both preset centerline condition information and preset vascular tree condition information exists in the neighborhood, adding the first target voxel point to a seed point queue; if it is determined that the first target voxel point does not exist in the neighborhood, determining whether a second target voxel point that satisfies the preset centerline condition information exists in the neighborhood; if it is determined that the second target voxel point exists, adding the second target voxel point to a seed point queue; if it is determined that the second target voxel point does not exist, determining whether a third target voxel point that satisfies the preset vascular tree condition information exists in the neighborhood; if it is determined that the third target voxel point exists, adding the third target voxel point to a seed point queue; updating the newly added target voxel point in the seed point queue as the starting seed point, and repeatedly performing the steps of determining the preset range neighborhood of each starting seed point based on the multiple starting seed points, determining the target voxel point in the neighborhood, and updating the target voxel point as the starting seed point until no newly added starting seed point exists.

[0130] In one example, the preset centerline condition information includes: the voxel point is located on the first centerline or on the line connecting the first centerline and the starting seed point located at the end of the first hepatic artery model; the preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on the line connecting the simulated hepatic artery vascular tree and the starting seed point located at the end of the first hepatic artery model.

[0131] For example, the server may predetermine centerline condition information A based on the first centerline. Specifically, in this embodiment, the first centerline is used as a criterion for seed point directional growth. That is, centerline condition information A refers to whether the voxel point is located on the first centerline. Alternatively, in some embodiments, the first centerline may be a certain distance away from the first hepatic artery model. Therefore, centerline condition information A may also include whether the voxel point is located on the line connecting the starting seed point on the first hepatic artery model and the first centerline.

[0132] The server may also pre-determine vascular tree condition information B based on the simulated hepatic artery vascular tree. Specifically, in this embodiment, the simulated hepatic artery vascular tree is used as another criterion for directional growth of the seed point. That is, the vascular tree condition information B refers to the voxel point being located on the simulated hepatic artery vascular tree. Similarly, in some embodiments, the simulated hepatic artery vascular tree may be a certain distance away from the first hepatic artery model. Therefore, the vascular tree condition information B may also include the voxel point being located on the line between the starting seed point on the first hepatic artery model and the simulated hepatic artery vascular tree.

[0133] Figure 5 This is a flow chart of another method for processing a liver artery model provided in this embodiment. Figure 5 As shown, in this step, the server determines a preset neighborhood of each starting seed point based on multiple starting seed points, centerline condition information A, and vascular tree condition information B. It then determines whether each voxel point within the neighborhood satisfies both the preset centerline condition information and the preset vascular tree condition information. If both are satisfied, the voxel point that satisfies both is determined to be a first target voxel point and is added to a seed point queue, thereby obtaining a newly added first target voxel point. If the first target voxel point is determined not to exist within the neighborhood, the server determines whether a second target voxel point that satisfies the centerline condition information exists within the neighborhood. If the second target voxel point exists, the server adds the second target voxel point to the seed point queue, thereby obtaining a newly added second target voxel point. If the second target voxel point is determined not to exist, the server determines whether a third target voxel point that satisfies the vascular tree condition information exists within the neighborhood. If the third target voxel point exists, the server adds the third target voxel point to the seed point queue, thereby obtaining a newly added third target voxel point. Finally, the newly added target voxel point in the seed point queue is updated as the starting seed point, and each newly added target voxel point in the seed point queue is used as the starting seed point, and the above steps of "determining the neighborhood of the preset range of each starting seed point based on multiple starting seed points, centerline condition information A and vascular tree condition information B, and determining the target voxel point within the neighborhood, and updating the target voxel point as the starting seed point" are repeated until all target voxel points in the seed point queue are traversed and there is no newly added starting seed point, the region growing process is terminated.

[0134] For example, in the process of the server determining whether multiple voxel points are seed points, the first step is to determine the neighborhood of the preset range of the starting seed point based on the starting seed point; usually, in three-dimensional space, 26 neighborhoods of the starting seed point can be determined, that is, a 3*3*3 voxel point area is obtained around the voxel point where the starting seed point is located as the neighborhood of the starting seed point; more specifically, the 26 voxel points in the three-dimensional space adjacent to the starting seed point are considered to be its neighborhood, including the voxel points directly connected to it and diagonally connected to it. For example, assuming that the coordinates of the starting seed voxel point are (i, j, k), then there are 27 points in total (i±1, j±1, k±1). Excluding the original center point (i, j, k), there are a total of 26 voxel points around it. These 26 voxel points are the 26 neighborhoods of the starting seed point.

[0135] In the second step, region growing is performed on the initial seed point by combining the centerline condition information A and the vascular tree condition information B. That is, the first centerline obtained based on the MRI data and the simulated hepatic artery vascular tree obtained based on the portal vein model reconstructed from the CT data are used as discrimination criteria. In the neighborhood of the starting seed point, the first target voxel point located on both the first centerline and the simulated hepatic artery vascular tree is preferentially selected as the seed point. When the first target voxel point exists, all other voxel points in the neighborhood of the starting seed point are eliminated. When the first target voxel point does not exist, the second target voxel point located on the first centerline but not on the simulated hepatic artery vascular tree is selected as the seed point. When the second target voxel point exists, all other voxel points in the neighborhood of the starting seed point are eliminated. When the second target voxel point does not exist, the third target voxel point located on the simulated hepatic artery vascular tree is selected as the seed point. When the third target voxel point exists, all other voxel points in the neighborhood of the starting seed point are eliminated. In this way, the optimal growth path can be selected in the neighborhood of the starting seed point, thereby improving the accuracy of the reconstruction of the target hepatic artery vascular model.

[0136] More specifically, Figure 6 、 Figure 7 and Figure 8 are all process diagrams of region growth in this embodiment. Figure 6 、 Figure 7 and Figure 8 The specific process of region growing in this embodiment is shown in FIG. Figure 6 、 Figure 7 and Figure 8In the figure, M is the first hepatic artery model, O is the starting seed point, P is the neighborhood, T is the simulated hepatic artery vascular tree, L is the first centerline, H is the liver parenchyma model, and the area shown by the grid is at least part of the area where the liver parenchyma model is located. Each grid represents a voxel position, and the voxel point of each voxel position is located at the center of the grid (shown as a dot). This embodiment uses the 26 neighborhoods of the starting seed point as an example. For ease of explanation, the thick black box is used to represent the area where the neighborhood is located. Figure 6 、 Figure 7 and Figure 8 Only the 8 neighbors on the cross section where the seed point finally grows are shown, that is, in three-dimensional space, the 9 neighbors above and below perpendicular to the paper surface are not shown.

[0137] More specifically, Figure 6 As shown in FIG, based on the first hepatic artery model M, a starting seed point O is determined at its distal end, and based on the starting seed point O, a neighborhood P of the starting seed point is determined around it. The simulated hepatic artery vascular tree T and the first centerline L are also located in the liver parenchyma model H according to their actual positions. Specifically, first, it is confirmed in the neighborhood P whether there is a first target voxel point that satisfies both the centerline condition information A and the vascular tree condition information B (i.e., is located on both the simulated hepatic artery vascular tree T and the first centerline L), as shown in FIG. Figure 6 As shown in FIG, there is no first target voxel point in the neighborhood P that satisfies both the centerline condition information A and the vascular tree condition information B. Therefore, it is further confirmed in the neighborhood P whether there is a second target voxel point that satisfies the centerline condition information A (i.e., located on the first centerline L), as shown in FIG. Figure 6 As shown, there are two second target voxels N1 and N2 in the neighborhood P that satisfy the centerline condition information A. Therefore, these second target voxels N1 and N2 are used as seed points and added to the seed point queue. At this point, there is no need to further confirm whether there is a third target voxel point that satisfies the vascular tree condition information B in the neighborhood P. It should be noted that when there are two target voxels in the neighborhood P that satisfy both the centerline condition information A and the vascular tree condition information B, or both the centerline condition information A and the vascular tree condition information B, it indicates that the vascular growth path has bifurcated at that location.

[0138] Furthermore, after the second target voxel points N1 and N2 are added to the seed point queue, the second target voxel points N1 and N2 need to be used as new starting seed points O for region growing. Figure 7 This is a schematic diagram of region growing using seed point N1 as the starting seed point. Figure 8 Schematic diagram of region growing with seed point N2 as the starting seed point. Figure 7In the process, first, it is confirmed in the neighborhood P whether there is a first target voxel point that satisfies both the centerline condition information A and the vascular tree condition information B (i.e., it is located on the simulated hepatic artery vascular tree T and the first centerline L at the same time), such as Figure 7 As shown in , there is no first target voxel point in the neighborhood P that satisfies both the centerline condition information A and the vascular tree condition information B. Therefore, it is further confirmed in the neighborhood P whether there is a second target voxel point that satisfies the centerline condition information A (i.e., located on the first centerline L), as shown in Figure 7 As shown in , there is no second target voxel point that meets the centerline condition information A in the neighborhood P. Therefore, it is necessary to further confirm whether there is a third target voxel point that meets the vascular tree condition information B in the neighborhood P, such as Figure 7 As shown, there is a third target voxel point N3 in the neighborhood P that meets the vascular tree condition information B. The third target voxel point N3 is used as a seed point and added to the seed point queue. Figure 8 In the process, first, it is confirmed in the neighborhood P whether there is a first target voxel point that satisfies both the centerline condition information A and the vascular tree condition information B (i.e., it is located on the simulated hepatic artery vascular tree T and the first centerline L at the same time), such as Figure 8 As shown in , there is no first target voxel point in the neighborhood P that satisfies both the centerline condition information A and the vascular tree condition information B. Therefore, it is further confirmed in the neighborhood P whether there is a second target voxel point that satisfies the centerline condition information A (i.e., located on the first centerline L), as shown in Figure 8 As shown in , there is no second target voxel point that meets the centerline condition information A in the neighborhood P. Therefore, it is necessary to further confirm whether there is a third target voxel point that meets the vascular tree condition information B in the neighborhood P, such as Figure 8 As shown, there is a third target voxel point N4 in the neighborhood P that meets the blood vessel tree condition information B. The third target voxel point N4 is used as a seed point and added to the seed point queue.

[0139] Alternatively, in other embodiments, there is no first target voxel point in the neighborhood P that satisfies both the centerline condition information A and the vascular tree condition information B; therefore, it is necessary to further confirm whether there is a second target voxel point in the neighborhood P that satisfies the centerline condition information A (i.e., is located on the first centerline L). If there is a second target voxel point in the neighborhood P that satisfies the centerline condition information A, then the second target voxel point is used as a seed point and added to the seed point queue. Furthermore, it is no longer necessary to further confirm whether there is a third target voxel point in the neighborhood P that satisfies the vascular tree condition information B. Furthermore, it should be noted that if there is no first target voxel point in the neighborhood P that satisfies both the centerline condition information A and the vascular tree condition information B, no second target voxel point that satisfies the centerline condition information A, and no third target voxel point that satisfies the vascular tree condition information B, then all voxels in the neighborhood P can be eliminated.

[0140] Alternatively, in some embodiments, when no seed point is found in the neighborhood P of the starting seed point, the neighborhood P can be expanded, and the presence of a seed point can be further confirmed in the expanded neighborhood P, and a threshold can be set for the neighborhood P. When the neighborhood P is expanded to exceed the threshold, step 207 is automatically terminated; for example, when no seed point is found in the 26 neighborhood (3*3*3), the neighborhood can be expanded to the 124 neighborhood (5*5*5) to search for a seed point. When the threshold is set to 125, if no seed point is found in the 124 neighborhood, the neighborhood is further expanded to the 342 neighborhood (7*7*7). At this time, since the 342 neighborhood exceeds the threshold 125, step 207 is automatically terminated, and all voxel points in the neighborhood are eliminated. At this time, the region growing process ends at the starting seed point.

[0141] Step 208: Determine a second hepatic artery model based on the multiple seed point queues.

[0142] In one example, step 208 includes: acquiring a second centerline corresponding to each seed point queue in the plurality of seed point queues; and generating a second hepatic artery model based on the second centerline corresponding to each seed point queue.

[0143] Exemplarily, obtaining the second centerline based on the seed point queue may be to connect the seed points in the seed point queue in sequence to generate the second centerline, and the second centerline is the blood vessel growth path obtained by the region growing algorithm. Generating the second hepatic artery model based on the second centerline may refer to expanding based on the second centerline as a reference to obtain the corresponding blood vessel model. Wherein, expanding based on the second centerline as a reference includes: determining the expansion size (diameter) according to the level of the blood vessel, and then expanding based on the determined expansion size to obtain the blood vessel model. It should be noted that the blood vessels are usually vertebral, so the expansion sizes of blood vessels of the same level at different positions may be different; or, in some other embodiments, the blood flow rate may be calculated based on information such as the patient's height and / or weight, and the expansion size may be determined based on the blood flow rate. There is no limitation on this expansion method. In this embodiment, the expansion size (diameter) may be 0.5mm-5mm.

[0144] Step 209: Acquire a third hepatic artery model based on the simulated hepatic artery vascular tree.

[0145] In one example, the grade of the hepatic artery corresponding to the third hepatic artery model is higher than the grade of the hepatic artery corresponding to the second hepatic artery model.

[0146] In one example, the third hepatic artery model includes the fifth hepatic artery.

[0147] For example, because MRI data cannot obtain relevant information about the fifth hepatic artery, that is, the first centerline corresponding to the fifth hepatic artery cannot be obtained, the above scheme still cannot reconstruct a liver vascular model that includes the fifth hepatic artery. This step, based on the second hepatic artery model obtained using the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, further obtains a third hepatic artery model based on the simulated hepatic artery vascular tree. Specifically, the third hepatic artery model is reconstructed based on the portal vein model. The hepatic artery corresponding to the third hepatic artery model includes the fifth hepatic artery. Since the maximum grade of the hepatic artery corresponding to the second hepatic artery model is grade 4, the grade of the hepatic artery corresponding to the third hepatic artery model is higher than that of the hepatic artery corresponding to the second hepatic artery model. Therefore, a target hepatic artery vascular model with a higher grade can be obtained based on the third hepatic artery model.

[0148] It should be noted that the steps for obtaining the third hepatic artery model based on the simulated hepatic artery vascular tree are similar to the steps for obtaining the second hepatic artery model based on the second centerline. For details, please refer to the above steps 204 to 208, which will not be repeated here.

[0149] Step 210: Determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery.

[0150] In one example, step 210 includes obtaining a target hepatic artery vessel model based on the first hepatic artery model, the second hepatic artery model, and the third hepatic artery model.

[0151] Exemplarily, step 210 includes two implementations. In a first implementation, the server can directly integrate the first hepatic artery model and the second hepatic artery model to determine a target hepatic artery vascular model, and then determine the hepatic artery vascular information based on the target hepatic artery vascular model. In this case, the grade of the hepatic artery corresponding to the target hepatic artery vascular model is equal to the grade of the hepatic artery corresponding to the second hepatic artery model. Alternatively, in a second implementation, the server integrates the first, second, and third hepatic artery models to determine a target hepatic artery vascular model, and then determine the hepatic artery vascular information based on the target hepatic artery vascular model. In this case, the grade of the hepatic artery corresponding to the target hepatic artery vascular model is equal to the grade of the hepatic artery corresponding to the third hepatic artery model.

[0152] In an embodiment of the present application, first image data and second image data are acquired. A first hepatic artery model and a simulated hepatic artery vascular tree are generated based on the first image data, and a first centerline is generated based on the second image data. A liver vascular model reconstruction request is acquired; wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier is determined, the liver vascular data including the first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree being higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline being higher than the grade of the hepatic artery corresponding to the first hepatic artery model. A liver parenchymal model is acquired. Multiple voxel points within the liver parenchymal model are determined. Based on the first hepatic artery model, multiple starting seed points located at the end of the first hepatic artery model are determined. Based on multiple starting seed points, preset centerline condition information, and preset vascular tree condition information, a target voxel point is determined from the multiple voxel points and added to a seed point queue. The preset centerline condition information represents conditional rule information regarding the first centerline, and the preset vascular tree condition information represents conditional rule information regarding the simulated hepatic artery vascular tree. A second hepatic artery model is determined based on the multiple seed point queues. A third hepatic artery model is obtained based on the simulated hepatic artery vascular tree. A target hepatic artery vascular model is determined based on the first and second hepatic artery models, and hepatic artery vascular information is determined based on the target hepatic artery vascular model. The grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. In this solution, first, based on the liver vascular data identifier obtained in the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier is determined. The liver vascular data includes the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline. Then, based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline, a second hepatic artery model is determined, wherein the hepatic artery has a higher grade. Finally, the first hepatic artery model and the second hepatic artery model are integrated to determine a target hepatic artery vascular model. The grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery. Therefore, by combining the lower-grade first hepatic artery model reconstructed based on the first image data, the simulated hepatic artery vascular tree corresponding to the higher-grade hepatic artery obtained from the portal vein model reconstructed based on the first image data, and the first centerline corresponding to the higher-grade hepatic artery obtained from the second image data, a region growing algorithm is used to accurately reconstruct the hepatic artery vascular system, including the higher-grade hepatic artery (e.g., grade 3 or higher, and some embodiments can reconstruct grade 5 or higher), to facilitate determination of hepatic artery vascular information and simulation calculations, thereby resolving the technical issue of low accuracy of the hepatic artery vascular model.

[0153] Figure 9 This is a structural diagram of a liver artery model processing device provided in an embodiment of the present application, as shown in FIG. Figure 9 As shown, the device includes:

[0154] The first acquisition unit 31 is configured to acquire a liver vascular model reconstruction request, wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and determine, based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein the grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than the grade of the hepatic artery corresponding to the first hepatic artery model, and the grade of the hepatic artery corresponding to the first centerline is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0155] The first determining unit 32 is configured to determine a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model.

[0156] The second determining unit 33 is configured to determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery.

[0157] The device of this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principles are the same and will not be repeated here.

[0158] Figure 10 This is a structural diagram of another liver artery model processing device provided in an embodiment of the present application. Figure 9 Based on the embodiment shown, Figure 10 As shown, the first determining unit 32 includes:

[0159] The first acquisition module 321 is used to acquire a liver parenchymal model.

[0160] The first determining module 322 is configured to determine a plurality of voxel points within the liver parenchyma model.

[0161] The second determining module 323 is configured to determine a plurality of starting seed points located at the distal end of the first hepatic artery model based on the first hepatic artery model.

[0162] The third determination module 324 is used to determine a target voxel point among multiple voxel points based on multiple starting seed points, preset centerline condition information, and preset vascular tree condition information, and add the target voxel point to a seed point queue; wherein the preset centerline condition information represents conditional rule information about the first centerline, and the preset vascular tree condition information represents conditional rule information about the simulated hepatic artery vascular tree.

[0163] The fourth determination module 325 is configured to determine a second hepatic artery model based on multiple seed point queues.

[0164] In one example, the third determining module 324 includes:

[0165] The first determining submodule 3241 is configured to determine a neighborhood of a preset range of each starting seed point based on the plurality of starting seed points, wherein each preset range includes at least one voxel point;

[0166] The first adding submodule 3242 is configured to add the first target voxel point to the seed point queue if it is determined that there is a first target voxel point in the neighborhood that satisfies both the preset centerline condition information and the preset blood vessel tree condition information.

[0167] The second determining submodule 3243 is configured to determine whether there is a second target voxel point that meets the preset centerline condition information in the neighborhood if it is determined that the first target voxel point does not exist in the neighborhood.

[0168] The second adding submodule 3244 is configured to add the second target voxel point to the seed point queue if it is determined that the second target voxel point exists.

[0169] The third determining submodule 3245 is configured to determine whether there is a third target voxel point that meets preset vascular tree condition information in the neighborhood if it is determined that the second target voxel point does not exist.

[0170] The third adding submodule 3246 is configured to add the third target voxel point to the seed point queue if it is determined that the third target voxel point exists.

[0171] The fourth determination submodule 3247 is used to update the newly added target voxel point in the seed point queue to the starting seed point, and repeatedly determine the neighborhood of the preset range of each starting seed point based on multiple starting seed points, and determine the target voxel point within the neighborhood, and update the target voxel point to the starting seed point until there is no newly added starting seed point.

[0172] In one example, the preset centerline condition information includes: the voxel point is located on the first centerline or on a line connecting the first centerline and a starting seed point located at the distal end of the first hepatic artery model.

[0173] The preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on a line connecting the simulated hepatic artery vascular tree and a starting seed point located at the end of the first hepatic artery model.

[0174] In one example, the fourth determining module 325 includes:

[0175] The first acquisition submodule 3251 is configured to acquire a second center line corresponding to each seed point queue in the plurality of seed point queues.

[0176] The generating submodule 3252 is configured to generate a second hepatic artery model based on the second centerline corresponding to each seed point array.

[0177] In one example, the grade of the hepatic artery in at least a portion of the simulated hepatic artery tree is the same as the grade of the hepatic artery corresponding to the first centerline.

[0178] In one example, the first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery, and the hepatic arteries corresponding to the first centerline and the simulated hepatic artery vascular tree both include a third-order hepatic artery and / or a fourth-order hepatic artery.

[0179] Figure 11 This is a structural diagram of another liver artery model processing device provided in an embodiment of the present application. Figure 10 Based on the embodiment shown, Figure 11 As shown, the device also includes:

[0180] The second acquisition unit 41 is configured to acquire a third hepatic artery model based on the simulated hepatic artery vascular tree.

[0181] In one example, the grade of the hepatic artery corresponding to the third hepatic artery model is higher than the grade of the hepatic artery corresponding to the second hepatic artery model.

[0182] In one example, the third hepatic artery model includes the fifth hepatic artery.

[0183] In one example, the second determining unit 33 is specifically configured to:

[0184] A target hepatic artery vascular model is obtained based on the first hepatic artery model, the second hepatic artery model, and the third hepatic artery model.

[0185] In one example, the apparatus further includes:

[0186] The third acquiring unit 42 is configured to acquire the first image data and the second image data.

[0187] The generating unit 43 is configured to generate a first hepatic artery model and a simulated hepatic artery vascular tree according to the first image data, and to generate a first centerline according to the second image data.

[0188] The device of this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principles are the same and will not be repeated here.

[0189] Figure 12 A schematic diagram of the structure of a server provided in an embodiment of the present application is shown in FIG. Figure 12 As shown, the server includes: a memory 51 and a processor 52.

[0190] The memory 51 stores computer programs that can be executed on the processor 52 .

[0191] The processor 52 is configured to execute the method provided in the above embodiments.

[0192] The server further includes a receiver 53 and a transmitter 54. The receiver 53 is used to receive instructions and data sent by an external device, and the transmitter 54 is used to send instructions and data to the external device.

[0193] An embodiment of the present application also provides a non-temporary computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of the server, enables the server to execute the method provided in the above embodiment.

[0194] An embodiment of the present application also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, at least one processor of the server can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the server executes the solution provided by any of the above embodiments.

[0195] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0196] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for processing a liver artery model, characterized in that: include: Obtaining a liver vascular model reconstruction request; wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and determining, based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein a grade of a hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than a grade of a hepatic artery corresponding to the first hepatic artery model, and a grade of a hepatic artery corresponding to the first centerline is higher than a grade of a hepatic artery corresponding to the first hepatic artery model; determining a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model; Determining a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein the grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to the grade of the hepatic artery corresponding to the second hepatic artery; The method further comprises: Acquire first image data and second image data; A first hepatic artery model and a simulated hepatic artery vascular tree are generated based on the first image data, and a first centerline is generated based on the second image data.

2. The method according to claim 1, characterized in that Determining a second hepatic artery model according to the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline includes: Obtain a liver parenchymal model; determining a plurality of voxel points within the liver parenchymal model; determining, based on the first hepatic artery model, a plurality of starting seed points located at the distal end of the first hepatic artery model; Based on the multiple starting seed points, the preset centerline condition information, and the preset vascular tree condition information, a target voxel point among the multiple voxel points is determined, and the target voxel point is added to a seed point queue; wherein the preset centerline condition information represents conditional rule information about the first centerline, and the preset vascular tree condition information represents conditional rule information about the simulated hepatic artery vascular tree; A second hepatic artery model is determined based on the plurality of seed point queues.

3. The method according to claim 2, characterized in that The step of determining a target voxel point from among the plurality of voxel points based on the plurality of starting seed points, the preset centerline condition information, and the preset vascular tree condition information, and adding the target voxel point to a seed point queue, comprises: Based on the plurality of starting seed points, determining a neighborhood of a preset range of each starting seed point; wherein each of the preset ranges includes at least one voxel point; If it is determined that a first target voxel point that satisfies both the preset centerline condition information and the preset blood vessel tree condition information exists in the neighborhood, adding the first target voxel point to a seed point queue; If it is determined that the first target voxel point does not exist in the neighborhood, determining whether a second target voxel point that meets the preset centerline condition information exists in the neighborhood; If it is determined that the second target voxel point exists, adding the second target voxel point to the seed point queue; If it is determined that the second target voxel point does not exist, determining whether there is a third target voxel point that meets the preset blood vessel tree condition information in the neighborhood; If it is determined that the third target voxel point exists, adding the third target voxel point to the seed point queue; The newly added target voxel point in the seed point queue is updated as the starting seed point, and the process of determining a neighborhood of a preset range of each starting seed point based on the multiple starting seed points, determining a target voxel point within the neighborhood, and updating the target voxel point as the starting seed point is repeated until there is no newly added starting seed point.

4. The method according to claim 2, characterized in that The preset centerline condition information includes: the voxel point is located on the first centerline or on the line connecting the first centerline and the starting seed point located at the distal end of the first hepatic artery model; The preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on a line connecting the simulated hepatic artery vascular tree and a starting seed point located at the end of the first hepatic artery model.

5. The method according to claim 2, characterized in that Determining a second hepatic artery model based on the plurality of seed point queues includes: Acquire a second center line corresponding to each seed point queue in the plurality of seed point queues; The second hepatic artery model is generated based on the second centerline corresponding to each seed point queue.

6. The method according to any one of claims 1 to 5, characterized in that The grade of at least a portion of the hepatic arteries in the simulated hepatic artery tree is the same as the grade of the hepatic artery corresponding to the first center line.

7. The method according to claim 6, characterized in that The first hepatic artery model includes a first-order hepatic artery and / or a second-order hepatic artery, and the hepatic arteries corresponding to the first center line and the simulated hepatic artery vascular tree both include a third-order hepatic artery and / or a fourth-order hepatic artery.

8. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Based on the simulated hepatic artery vascular tree, a third hepatic artery model is obtained.

9. The method according to claim 8, characterized in that The grade of the hepatic artery corresponding to the third hepatic artery model is higher than the grade of the hepatic artery corresponding to the second hepatic artery model.

10. The method according to claim 9, characterized in that The third hepatic artery model includes the fifth order hepatic artery.

11. The method according to claim 10, characterized in that The step of determining a target hepatic artery model based on the first hepatic artery model and the second hepatic artery model includes: The target hepatic artery vessel model is obtained based on the first hepatic artery model, the second hepatic artery model, and the third hepatic artery model.

12. A liver artery model processing device, characterized in that: include: a first acquisition unit configured to acquire a liver vascular model reconstruction request, wherein the liver vascular model reconstruction request includes a liver vascular data identifier; and determine, based on the liver vascular model reconstruction request, liver vascular data corresponding to the liver vascular data identifier, wherein the liver vascular data includes a first hepatic artery model, a simulated hepatic artery vascular tree, and a first centerline, wherein a grade of the hepatic artery corresponding to the simulated hepatic artery vascular tree is higher than a grade of the hepatic artery corresponding to the first hepatic artery model, and a grade of the hepatic artery corresponding to the first centerline is higher than a grade of the hepatic artery corresponding to the first hepatic artery model; a first determining unit, configured to determine a second hepatic artery model based on the first hepatic artery model, the simulated hepatic artery vascular tree, and the first centerline; wherein the grade of the hepatic artery corresponding to the second hepatic artery model is higher than the grade of the hepatic artery corresponding to the first hepatic artery model; a second determining unit, configured to determine a target hepatic artery vascular model based on the first hepatic artery model and the second hepatic artery model, so as to determine hepatic artery vascular information according to the target hepatic artery vascular model; wherein a grade of the hepatic artery corresponding to the target hepatic artery vascular model is greater than or equal to a grade of the hepatic artery corresponding to the second hepatic artery; a third acquiring unit, configured to acquire the first image data and the second image data; A generating unit is configured to generate a first hepatic artery model and a simulated hepatic artery vascular tree according to the first image data, and to generate a first centerline according to the second image data.

13. The device according to claim 12, characterized in that The first determining unit includes: A first acquisition module is used to acquire a liver parenchymal model; A first determining module is used to determine a plurality of voxel points in the liver parenchyma model; a second determining module, configured to determine, based on the first hepatic artery model, a plurality of starting seed points located at the distal end of the first hepatic artery model; a third determining module, configured to determine a target voxel point among the plurality of voxel points based on the plurality of starting seed points, preset centerline condition information, and preset vascular tree condition information, and add the target voxel point to a seed point queue; wherein the preset centerline condition information represents conditional rule information regarding the first centerline, and the preset vascular tree condition information represents conditional rule information regarding the simulated hepatic artery vascular tree; The fourth determination module is configured to determine a second hepatic artery model based on the plurality of seed point queues.

14. The device according to claim 13, characterized in that The third determining module includes: A first determining submodule is configured to determine a neighborhood of a preset range of each of the starting seed points based on the plurality of starting seed points; wherein each of the preset ranges includes at least one voxel point; a first adding submodule, configured to add the first target voxel point to a seed point queue if it is determined that a first target voxel point that satisfies both the preset centerline condition information and the preset blood vessel tree condition information exists in the neighborhood; a second determining submodule, configured to determine whether a second target voxel point that satisfies the preset centerline condition information exists in the neighborhood if it is determined that the first target voxel point does not exist in the neighborhood; a second adding submodule, configured to add the second target voxel point to a seed point queue if it is determined that the second target voxel point exists; a third determining submodule, configured to determine whether there is a third target voxel point that satisfies the preset vascular tree condition information in the neighborhood if it is determined that the second target voxel point does not exist; a third adding submodule, configured to add the third target voxel point to a seed point queue if it is determined that the third target voxel point exists; The fourth determination submodule is used to update the newly added target voxel point in the seed point queue as the starting seed point, and repeatedly determine the neighborhood of the preset range of each starting seed point based on multiple starting seed points, and determine the target voxel point within the neighborhood, and update the target voxel point to the starting seed point until there is no newly added starting seed point.

15. The device according to claim 13, characterized in that The preset centerline condition information includes: the voxel point is located on the first centerline or on the line connecting the first centerline and the starting seed point located at the distal end of the first hepatic artery model; The preset vascular tree condition information includes: the voxel point is located on the simulated hepatic artery vascular tree or on a line connecting the simulated hepatic artery vascular tree and a starting seed point located at the end of the first hepatic artery model.

16. The device according to claim 13, characterized in that The fourth determining module includes: A first acquisition submodule is configured to acquire a second center line corresponding to each seed point queue in the plurality of seed point queues; A generating submodule is configured to generate the second hepatic artery model based on the second center line corresponding to each seed point queue.

17. A server, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the method according to any one of claims 1 to 11 is implemented.

18. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 11 when executed by a processor.

19. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 11 when being executed by a processor.

Images