Bone tissue blood vessel network reconstruction method and system

Abstract

The invention relates to the field of image processing, in particular to a reconstruction method and system of a bone tissue blood vessel network, and provides a brand-new and full-automatic method and system for performing three-dimensional reconstruction, measurement and analysis of different-scale blood vessels on a bone tissue CT structure image, which are used for reflecting the blood vessel distribution of bone tissues, providing quantitative information of the blood vessel density and improving the blood vessel density. And bone tissue perfusion metabolism, bone lesion and bone repair evaluation are assisted. According to the system, individual differences can be restored, anisotropic characteristics of bone tissue material attributes are introduced, mechanical simulation is completed, and the similarity between mechanical simulation and real physiological conditions is increased. The invention also provides a blood vessel density estimation means, a blood vessel network generation and visualization means, and a full-automatic standardized finite element mechanical simulation process and equipment.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a method and system for reconstructing a bone tissue vascular network. Background Art

[0002] Bone tissue is one of the main structural tissues of the human body, with a complex hierarchical structure, in which cancellous bone and compact bone are the two main tissue types that make up the inner and outer layers of bones. Cancellous bone is composed of trabeculae, and there are many pores between the trabeculae, which are filled with bone marrow, which contains abundant blood vessels. Compact bone is denser and harder than cancellous bone, and is mainly composed of canaliculi, platelets, bone chambers and connecting channels between canaliculi, and the tissues are tightly arranged. Among them, canaliculi are tubular structures that contain blood vessels, nerve fibers and lymphatic vessels. Compact bone provides strength and rigidity to bones, enabling them to withstand pressure and resist stress. Cancellous bone mainly reflects the elasticity of bone tissue, while compact bone mainly reflects the rigidity of bone tissue.

[0003] CT (computed tomography) imaging of bone tissue is a commonly used non-invasive three-dimensional medical imaging method that can provide high-resolution bone tissue images. Femoral head CT can provide high-resolution bone images, display bone tissue and surrounding soft tissue structures, evaluate bone tissue morphology and bone quality, and assist doctors in diagnosing and evaluating bone abnormalities such as fractures, osteoporosis, bone cysts, bone tumors, and osteonecrosis. Among them, osteonecrosis refers to the necrosis of bone tissue due to insufficient blood supply. CT imaging can show the blood supply and the scope of the necrotic area.

[0004] Bone tissue vascular imaging. Bone tissue CTA (CT angiography) and MRA (magnetic resonance angiography) can provide high-resolution, non-invasive vascular imaging. The former uses CT imaging combined with intravenous injection of contrast agents to obtain bone tissue vascular images, while the latter uses magnetic resonance imaging technology combined with intravenous injection of contrast agents to display bone tissue and surrounding blood vessels to evaluate bone tissue blood supply and vascular patency, thrombosis, and aneurysms. Bone tissue DSA (digital subtraction angiography) is an interventional vascular imaging method that uses X-rays to obtain images of bone tissue and its blood vessels through vascular cannulation and intravascular injection of contrast agents. Microangiography (microangiography) injects dye into the blood vessels and uses a microscope to observe the flow of dye in the capillaries, thereby imaging the capillary network. Vascular imaging of bone tissue can indirectly evaluate vascular distribution and density.

[0005] However, none of the above technologies can image both structure and function at the same time. The imaged blood vessels are usually the blood supply / drainage vessels of bone tissue, which are large arteries / veins. For the imaging of capillaries in bone tissue for scientific research and laboratory environments, the imaging tissue needs to be removed from the body and the sample needs to be prepared, which has clinical limitations. Existing vascular imaging technologies may cause allergic reactions when contrast agents are injected, and interventional imaging may cause vascular damage and complications. Imaging technologies that rely on X-rays will expose subjects to radiation and have certain risks. Summary of the invention

[0006] Based on the above objectives, the purpose of the present invention is to overcome the shortcomings of the prior art and propose a new, fully automatic method and system for three-dimensional reconstruction, measurement and analysis of blood vessels of different scales in bone tissue CT structural images, so as to reflect the vascular distribution of bone tissue, provide quantitative information on vascular density, and assist in the evaluation of bone tissue perfusion metabolism, bone lesions and bone repair.

[0007] To this end, the first technical solution of the present application discloses a method for reconstructing a bone tissue vascular network, comprising:

[0008] S1. Preprocess the target image to obtain a finite element mesh unit;

[0009] S2. construct the coordinate correspondence between the voxel and each vertex of the finite element mesh unit;

[0010] S3. respectively calculating the bone tissue density of each vertex of the voxel and the finite element mesh unit, and obtaining the relative blood vessel density of each vertex as the numerical input for generating the vascular network;

[0011] S4. Arrange the relative blood vessel density of each vertex of the finite element mesh unit in descending order, and set the first N surface vertices as the starting point of the blood vessel network according to the order, where N∈(0,1); traverse the length of the edge where the first N surface vertices are located, and take the shortest edge as the blood vessel extension direction. The other vertex involved in the shortest edge is the blood vessel growth sub-node, which serves as the input vertex for continued extension; repeat this process until the distance between the growth terminal sub-node and the corresponding starting point is greater than the radius of the circumscribed circle of the starting point along the center line cross section, and the blood vessel network growth ends;

[0012] S5. Output the bone tissue vascular network, including vertex index numbers, the number of vascular extension child nodes, and the child node index numbers.

[0013] Furthermore, the coordinate correspondence between the voxels and the vertices of the finite element grid unit constructed in S2 is specifically as follows: a three-dimensional orthogonal coordinate system and a transformation matrix are constructed between the target image voxel space and the finite element grid unit space described in S1, thereby obtaining the coordinate correspondence between the voxels and the vertices of the grid unit in the target image voxel space and the finite element grid space.

[0014] Furthermore, the bone tissue density of the voxel in S3 is calculated based on the linear relationship between the bone tissue CT value and the apparent density; the bone tissue density of the finite element grid unit is obtained according to the transformation matrix of the spatial coordinate system according to claim 2; the relative blood vessel density of each vertex is calculated by simplifying the bone tissue into porous material, and simplifying the density and porosity of the porous material into a linear relationship.

[0015] Furthermore, S4 also includes inputting a custom threshold D. When the difference between the shortest side and the second shortest side of the vertex or growing sub-node is less than D, the blood vessel will bifurcate at this point. At this time, two growing sub-nodes will be formed. The S4 cycle is repeated for both sub-nodes, but the points that have already appeared in the vascular network will no longer appear in subsequent cycles.

[0016] And, the bone tissue vascular network reconstructed according to the above method.

[0017] The second technical solution of the present application provides a bone tissue vascular network reconstruction system, comprising:

[0018] Original data: target image used for reconstruction;

[0019] Preprocessing module: used to convert the target image into finite element mesh units;

[0020] Spatial registration module: used to construct the coordinate correspondence between voxels and vertices of finite element mesh units;

[0021] Parameter calculation module: used to calculate the bone tissue density of each vertex of the voxel and finite element mesh unit, obtain the relative blood vessel density of each vertex, and generate the numerical input for the generation of the vascular network;

[0022] Vascular network generation module: used to process numerical input, reconstruct and generate vascular networks;

[0023] Vascular network output module: used to output the reconstructed vascular network, including vertex index number, number of vascular extension sub-nodes, and sub-node index number.

[0024] Furthermore, the vascular network generation module includes a custom vascular bifurcation threshold D, which is used to determine whether a blood vessel has a bifurcation.

[0025] The third technical solution of the present application discloses a computer device, comprising:

[0026] Memory; and

[0027] The processor is configured to execute the above-mentioned method for reconstructing the bone tissue vascular network based on the instructions stored in the memory.

[0028] The fourth technical solution of the present application discloses a computer scale storage medium storing computer instructions, which can implement the above-mentioned bone tissue vascular network reconstruction method when executed by a processor.

[0029] Beneficial effects: This application designs a reconstruction method for bone tissue vascular network, which can restore individual differences, introduce the anisotropic characteristics of bone tissue material properties and input them into mechanical simulation, thereby increasing the similarity between mechanical simulation and real physiological conditions. It also provides a means of estimating vascular density, a means of generating and visualizing vascular networks, and a fully automatic standardized finite element mechanical simulation process and equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a schematic diagram of the reconstruction of the vascular network of bone tissue;

[0031] Figure 2 To construct the finite element mesh unit diagram;

[0032] Figure 3 Generate a schematic diagram for the bone tissue vascular network;

[0033] Figure 4 Generate a schematic diagram of module iterations for the bone tissue vascular network;

[0034] Figure 5 Reconstruction system for the vascular network of bone tissue. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0036] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to those of ordinary skill in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be considered as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so that once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0037] Unless otherwise specified, the meanings of the technical terms in this specification are the same as those generally understood by those skilled in the art. However, if there is any conflict, the definitions in this specification shall prevail.

[0038] The term "mask" in this application refers to blocking (all or part of) the target image to be processed to control the area or process of image processing.

[0039] The term "porosity (φ)" refers to the ratio of the volume of pores (voids) in a material to the total volume.

[0040] Taking a specific bone tissue CT scan as an example, a method for reconstructing a bone tissue vascular network is specifically disclosed in combination with the embodiments of the present application. The method flow chart is as follows: Figure 1 As shown, specifically including:

[0041] S1. Preprocess the target image to obtain a finite element mesh unit.

[0042] In this embodiment, the target image is a bone tissue scan image, such as a CT image or a MRI image, etc. This embodiment takes a CT image as an example.

[0043] The preprocessing is to process the original CT DICOM image, segment the bone tissue and obtain the bone tissue mask, and generate a three-dimensional bone tissue structure model (such as Figure 2 Left), the structural model is discretized to obtain the finite element mesh unit (such as Figure 2 Right), denoted as a mesh T = (V, ε, F, U) with vertices V, edges ε, faces F, and units U. Vertices contain coordinates and index numbers, edges contain vertex index numbers and lengths, faces contain surface vertex index numbers, and units contain vertex index numbers. The resolution of the finite element mesh should be equal to or higher than the original image resolution.

[0044] It is understandable that the segmentation, reconstruction and finite element mesh discretization processing described in this embodiment are existing technologies or general technologies in this industry. This application only proposes some preferred implementations and does not serve as a limitation to this application.

[0045] S2. Construct the coordinate correspondence between voxels and vertices of finite element mesh units: by constructing the three-dimensional orthogonal coordinate system and transformation matrix of the target image voxel space and the finite element mesh unit space described in S1, the coordinate correspondence between the voxels and vertices of the mesh units in the target image voxel space and the finite element mesh space is obtained.

[0046] S3. Calculate the bone tissue density of each voxel and each vertex of the finite element mesh unit respectively, and obtain the relative blood vessel density of each vertex as the numerical input for generating the vascular network.

[0047] In this embodiment, the bone tissue density of the voxel is calculated based on the linear relationship between the bone tissue CT value and the apparent density; the bone tissue density of the finite element grid unit is obtained according to the transformation matrix of the spatial coordinate system according to claim 2; the relative blood vessel density of each vertex is calculated by simplifying the bone tissue into a porous material and simplifying the density and porosity of the porous material into a linear relationship, specifically:

[0048] The bone tissue CT image is imported into the parameter calculation module, and the bone tissue density ρ of each image voxel is calculated according to the image voxel CT value Voxel , and map the voxel attributes to the finite element mesh unit vertices to obtain the bone tissue density ρ at each vertex Vertex Considering bone tissue as a porous material, the relationship between material density and porosity is simplified to be linear (Formula 1), and the relative vascular density rρ at each vertex is calculated. Vessel The vertex is denoted as V = (index, rρ Vessel ), is the numerical input for generating the vascular network;

[0049] Where, ρ = (1-φ)*ρ Theory (1);

[0050] φ Bone +φ Vessel =1 (2);

[0051]

[0052] S4. Arrange the relative vascular density of each vertex of the finite element mesh unit in descending order, and set the first N surface vertices as the starting point of the vascular network according to the order, where N∈(0,1); traverse the length of the edges where the first N surface vertices are located, and take the shortest side as the vascular extension direction. The other vertex involved in the shortest side is the vascular growth sub-node, which serves as the input vertex for continued extension; repeat this process until the distance between the growth terminal sub-node and the corresponding starting point is greater than the radius of the circumscribed circle of the starting point along the centerline cross section, and the vascular network growth is terminated.

[0053] In a specific embodiment, the vertices included in the surface F correspond to the relative blood vessel density rρ Vessel Arrange in descending order, and input the custom parameter vascular network starting point percentage N∈(0,1) at the same time, calculate the number of vascular starting point vertices and set the first N surface vertices as the vascular network starting point according to the sorting.

[0054] By traversing the connection relationship between the starting point and the neighboring vertices of the finite element mesh unit and calculating the edge length, the shortest path algorithm is used to assign weights to the extension direction of the vascular mesh. The relative vascular density of the neighboring vertices in each possible extension direction is combined to determine whether it is bifurcated. The radius of the circumscribed circle of the cross section passing through the starting point of the blood vessel along the center line of the bone tissue structure model is taken as the end point of each vascular network growth (such as Figure 3 : Left: 2D and 3D examples of single blood vessel growth; Middle: Local enlarged illustration of generated vascular network; Right: Vascular network generation using the femoral head as an example).

[0055] like Figure 4 A schematic diagram of the iterations of vascular network generation is shown in FIG. Figure 4 As shown on the left, with a 0 As the starting point, 00 , l 01 …l 0m0 for a 0 m 0 possible extension directions, namely m 0 edges, where l 01 is the shortest side, so it is recorded as a 0 The extension direction of the edge corresponds to another vertex a 1 That is, the blood vessel derived subnode, which serves as the input vertex of the next cycle, such as Figure 4 As shown on the right, with a 1 As the starting point, 10 , l 11 …l 1m1 for a 1 m 1 possible extension directions, namely m 1 edges, where l 11 is the shortest side, so it is recorded as a 1 The extension direction of the edge corresponds to another vertex a 2 That is, the blood vessel derived subnode, which is used as the input vertex of the next cycle. Thus, we get Figure 4 The extension direction of the middle blood vessel is a 0 —>a 1 —>…—>a n When the distance between the last child node and the corresponding starting point is greater than the radius of the circumscribed circle of the cross section along the center line of the starting point, the growth of the vascular network ends, and the following is obtained: Figure 3 The vascular network generation diagram shown in the figure, in this case the last child node is the growth terminal child node.

[0056] It should be noted that, considering the problem of vascular bifurcation in the actual blood vessel growth process, a custom vascular bifurcation threshold D is added in this step. The threshold D is determined by the density of the capillary network (i.e., the microscopic imaging energy number: the number of capillary branches per cubic millimeter). During the vascular iteration process, when the difference between the shortest side and the second shortest side of the vertex is less than the input custom threshold D, it is recorded as a blood vessel bifurcation at this vertex. 1 Vertex, l 11 is the shortest side, l 10 is the second shortest side, if l 10-l 11 <D, that is, a 1 is a forking vertex. At this time, two growing child nodes will be formed. The S4 loop is repeated for both child nodes, but the points that have already appeared in the blood vessel network will not appear in subsequent loops.

[0057] Beneficial effects of the above steps: After increasing the construction of the blood vessel network, three-dimensional simulation of the blood perfusion effect of arterioles and capillary networks can be achieved. At present, the simulation of the blood vessel network is mainly based on imaging to reconstruct blood vessels, but it is relatively difficult to achieve small-scale blood vessel imaging and reconstruction. If the blood vessel structure tree model is used, the radii of sub-vessels and mother vessels are determined based on flow assumptions. This method is mainly applied to the blood flow simulation in the blood vessel wall and lumen, and is not suitable for simulating the supply and mass exchange process of the small-scale fluid cavity (capillary network) to the surrounding tissues. In the simulation of the capillary network, for example, in the mathematical modeling of the glomerulus to simulate the filtration of the kidney (glomerular dialysis), the capillary is abstracted as a one-dimensional network, lacking a three-dimensional structure, and the simulation is only for the overall function of the organ.

[0058] S5. Output the bone tissue blood vessel network, including vertex index numbers, the number of blood vessel extension child nodes, and child node index numbers.

[0059] The second embodiment of the present application provides a bone tissue blood vessel network reconstruction system, as Figure 5 shown, including:

[0060] Original data: the target picture for reconstruction;

[0061] Preprocessing module: used to convert the target picture into finite element mesh cells;

[0062] Spatial registration module: used to construct the coordinate correspondence between voxels and each vertex of the finite element mesh cells;

[0063] Parameter calculation module: used to calculate the bone tissue density of each vertex of the voxel and the finite element mesh cells, obtain the relative blood vessel density of each vertex, and generate the numerical input for blood vessel network generation;

[0064] Blood vessel network generation module: used to process the numerical input, reconstruct and generate the blood vessel network;

[0065] Blood vessel network output module: used to output the reconstructed blood vessel network, including vertex index numbers, the number of blood vessel extension child nodes, and child node index numbers.

[0066] On the other hand, the present application also discloses a computer device, including:

[0067] A memory; and

[0068] The processor is configured to execute the above-mentioned method for reconstructing the bone tissue vascular network based on the instructions stored in the memory.

[0069] Finally, the present application also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the above-mentioned method for reconstructing a bone tissue vascular network can be implemented.

[0070] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for reconstructing a bone tissue vascular network, characterized in that: include: S1. Preprocess the target image to obtain a finite element mesh unit; S2. construct the coordinate correspondence between the voxel and each vertex of the finite element mesh unit; S3. respectively calculating the bone tissue density of each vertex of the voxel and the finite element mesh unit, and obtaining the relative blood vessel density of each vertex as the numerical input for generating the vascular network; S4. Arrange the relative blood vessel density of each vertex of the finite element mesh unit in descending order, and set the first N surface vertices as the starting point of the blood vessel network according to the order, where N∈(0,1); traverse the length of the edge where the first N surface vertices are located, and take the shortest edge as the blood vessel extension direction. The other vertex involved in the shortest edge is the blood vessel growth sub-node, which serves as the input vertex for continued extension; repeat this process until the distance between the growth terminal sub-node and the corresponding starting point is greater than the radius of the circumscribed circle of the starting point along the center line cross section, and the blood vessel network growth ends; S5. Output the bone tissue vascular network, including vertex index numbers, the number of vascular extension child nodes, and the child node index numbers.

2. The reconstruction method according to claim 1, characterized in that: The coordinate correspondence between the voxels and the vertices of the finite element grid unit constructed in S2 is specifically as follows: a three-dimensional orthogonal coordinate system and a transformation matrix are constructed between the target image voxel space and the finite element grid unit space in S1, thereby obtaining the coordinate correspondence between the voxels and the vertices of the grid unit in the target image voxel space and the finite element grid space.

3. The reconstruction method according to claim 1, characterized in that: The bone tissue density of the voxel in S3 is calculated based on the linear relationship between the bone tissue CT value and the apparent density; the bone tissue density of the finite element grid unit is obtained according to the transformation matrix of the spatial coordinate system according to claim 2; the relative blood vessel density of each vertex is calculated by simplifying the bone tissue into porous material and simplifying the density and porosity of the porous material into a linear relationship.

4. The reconstruction method according to claim 1, characterized in that: S4 also includes inputting a custom threshold D. When the difference between the shortest side and the second shortest side of a vertex or a growing sub-node is less than D, the blood vessel will bifurcate at this point. At this time, two growing sub-nodes will be formed. The S4 cycle will be repeated for both sub-nodes, but points that have already appeared in the vascular network will no longer appear in subsequent cycles.

5. A bone tissue vascular network reconstructed according to any one of the reconstruction methods of claims 1-4.

6. A bone tissue vascular network reconstruction system, characterized in that: include: Original data: target image used for reconstruction; Preprocessing module: used to convert the target image into finite element mesh units; Spatial registration module: used to construct the coordinate correspondence between voxels and vertices of finite element mesh units; Parameter calculation module: used to calculate the bone tissue density of each vertex of the voxel and finite element mesh unit, obtain the relative blood vessel density of each vertex, and generate the numerical input for the generation of the vascular network; Vascular network generation module: used to process numerical input, reconstruct and generate vascular networks; Vascular network output module: used to output the reconstructed vascular network, including vertex index number, number of vascular extension sub-nodes, and sub-node index number.

7. The system according to claim 6, characterized in that: The vascular network generation module includes a custom vascular bifurcation threshold D, which is used to determine whether a blood vessel has bifurcation.

8. A computer device, characterized in that: include: Memory; as well as A processor is configured to execute the bone tissue vascular network reconstruction method according to any one of claims 1 to 4 based on the instructions stored in the memory.

9. A computer scale storage medium, characterized in that: It stores computer instructions, which, when executed by a processor, can implement the bone tissue vascular network reconstruction method described in any one of claims 1-4.

Images