A method for acquiring parameters of a lumbar endplate and related device

Abstract

This application discloses a method and related apparatus for obtaining parameters of the lumbar vertebral endplate, which is used to improve parameter accuracy. The method includes: obtaining the centroid coordinates and the midline vector parallel to the left-right direction of the human body for each vertebra based on the acquired initial image of the lumbar spine; performing interpolation fitting based on the centroid coordinates of all vertebrae to obtain the spinal center curve, and calculating the tangent vector of the spinal center curve at the centroid of each vertebra; for each vertebra, determining the constant term by multiplying the inverse vector of the cross product of the midline vector and the tangent vector with the centroid coordinates, and determining the cutting plane based on the cross product and the constant term; cutting the pre-constructed target triangular mesh covering the surface of the vertebrae according to the cutting plane to obtain the target mesh, wherein both the target triangular mesh and the target mesh are meshes including multiple triangular facets, and the coverage of the target mesh includes the surfaces of the upper and lower endplates of the vertebrae; and analyzing the target mesh to obtain the parameters of the endplate.

Description

Technical Field

[0001] This application relates to the field of image processing, and more particularly to a method and related apparatus for obtaining parameters of the lumbar endplate. Background Technology

[0002] The lumbar spine is a vital part of the human body, playing an indispensable role in daily life. Lumbar spine diseases, such as herniated discs, vertebral compression fractures, and scoliosis, can significantly impact a patient's quality of life and work capacity. Therefore, understanding the real-time condition of the lumbar spine, particularly the parameters of the lumbar endplates (such as the depth and angle of endplate depression), is crucial for medical professionals' research and analysis. In current methods, after acquiring images of the lumbar spine, medical professionals manually annotate key points based on practical experience and then calculate the relevant parameters of the lumbar endplates based on these key points.

[0003] However, in existing solutions, the parameters of the lumbar endplates are obtained based on manual annotation, which is highly subjective and results in low accuracy of the obtained parameters, failing to meet higher requirements. Summary of the Invention

[0004] To address the technical problem of low parameter accuracy in the prior art, this application provides a method and related apparatus for obtaining parameters of the lumbar endplate, which improves parameter accuracy.

[0005] The first aspect of this application provides a method for obtaining parameters of the lumbar vertebral endplate, including: Based on the initial image of the lumbar spine, the centroid coordinates and midline vector of each vertebra are obtained. The midline vector is parallel to the left and right direction of the human body. Interpolation fitting is performed based on the centroid coordinates of all cones to obtain the spinal center curve, and the tangent vector of the spinal center curve at the centroid of each cone is calculated. For each cone, the product of the inverse of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. The pre-constructed target triangular mesh covering the surface of the cone is cut according to the cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes that include multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper and lower end plates of the cone. For each cone, the parameters of the final plate are obtained by analyzing the target mesh.

[0006] Optionally, the parameters of the final plate can be obtained by analyzing the target mesh, including: For each cone, determine the upper end plate patch covering the upper end plate surface and the lower end plate patch covering the lower end plate surface from the target mesh; The least squares method is used to perform plane fitting on the points on the upper and lower end plate patches respectively to obtain the corresponding upper and lower end plate fitting planes. The parameters of the end plate are then calculated based on the upper and lower end plate fitting planes.

[0007] Optionally, for each cone, an upper endplate patch covering the upper endplate surface and a lower endplate patch covering the lower endplate surface are determined from the target mesh, including: For each cone, the triangular facets in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is greater than a first preset threshold are determined as the upper end plate facets covering the upper end plate surface; the triangular facets in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is less than a second preset threshold are determined as the lower end plate facets covering the lower end plate surface.

[0008] Optionally, the least squares method is used to perform plane fitting on the points on the upper and lower endplate patches respectively, to obtain the corresponding upper and lower endplate fitting planes, including: Based on the preset relationship between the distance from a point to the centroid of the cone and the point weight, the corresponding weights of the upper end plate points are obtained according to the points on the upper end plate surface, and the corresponding weights of the lower end plate points are obtained according to the points on the lower end plate surface. Based on the least squares method, the corresponding upper fitting plane coefficients are calculated according to the weights of the upper end plate points and the coordinates of the points on the upper end plate surface, thus obtaining the upper end plate fitting plane. Based on the least squares method, the corresponding lower fitting plane coefficients are calculated according to the weights of the lower end plate points and the coordinates of the points on the lower end plate surface, thus obtaining the lower end plate fitting plane. or, From the points on the upper end plate, select the upper fitting points that are farthest from the centroid of the cone by a predetermined percentage. From the points on the lower end plate, select the lower fitting points that are farthest from the centroid of the cone by a predetermined percentage. Based on the least squares method, calculate the corresponding upper fitting plane coefficients based on the upper fitting points to obtain the upper end plate fitting plane. Based on the least squares method, calculate the corresponding lower fitting plane coefficients based on the lower fitting points to obtain the lower end plate fitting plane.

[0009] Optional parameters include upper concave depth, lower concave depth, upper concave angle, and lower concave angle. The parameters of the final plate are calculated based on the upper and lower final plate fitting planes, including: Based on the preset concave depth formula, the distance between the upper concave point of the upper end plate and the fitting plane of the upper end plate is determined as the upper concave depth, and based on the preset concave depth formula, the distance between the lower concave point of the lower end plate and the fitting plane of the lower end plate is determined as the lower concave depth. The two points closest to the upper concave point among the boundary points of the upper endplate fitting plane are designated as the first point and the second point. The angle between the first and second connecting lines is designated as the upper concave angle. The first connecting line is the straight line between the first point and the upper concave point, and the second connecting line is the straight line between the second point and the upper concave point. The two points closest to the lower concave point among the boundary points of the lower endplate fitting plane are designated as the third point and the fourth point. The angle between the third and fourth connecting lines is designated as the lower concave angle. The third connecting line is the straight line between the third point and the lower concave point, and the fourth connecting line is the straight line between the fourth point and the lower concave point.

[0010] Optionally, after calculating the parameters of the final plate based on the upper and lower final plate fitting planes, the method further includes: For each cone, the sum of the areas of all the upper endplates is determined as the surface area of ​​the upper endplate, and the sum of the areas of all the lower endplates is determined as the surface area of ​​the lower endplate. For each cone, the maximum projection distance of a point on the upper end plate in the front-back direction is determined as the front-back diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the front-back direction is determined as the front-back diameter of the lower end plate, wherein the front-back direction is perpendicular to the tangent vector and the midline vector, respectively. For each cone, the maximum projection distance of a point on the upper end plate in the left-right direction is determined as the left-right diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the left-right direction is determined as the left-right diameter of the lower end plate. The cosine of the angle between the normal vector of the upper endplate fitting plane of the current cone and the normal vector of the lower endplate fitting plane of the previous cone is determined as the endplate angle between the two endplates, and the Euclidean distance between the center point of the upper endplate fitting plane of the current cone and the center point of the lower endplate fitting plane of the previous cone is determined as the distance between the two cones.

[0011] Optionally, the centroid coordinates and midline vector of each vertebra are obtained from the acquired initial image of the lumbar spine, including: The initial image of the lumbar spine is acquired, and the initial image is resampled and the voxel spacing is normalized to obtain the target image. Based on a preset deep learning segmentation algorithm, the target image is processed into three-dimensional cone segmentation to obtain three-dimensional segmentation masks of multiple cones. For each cone, based on the stereolithography algorithm, the three-dimensional segmentation mask of the cone is processed to obtain the corresponding initial triangular mesh covering the surface of the cone, and the initial triangular mesh is processed by morphological closing operation to obtain the corresponding target triangular mesh. For each cone, the centroid coordinates of the corresponding cone are calculated based on the coordinates of the vertices of all triangular faces of the target triangular mesh, using a preset centroid coordinate algorithm. For each cone, the mean vector is calculated based on the coordinates of all vertices of the target triangular mesh corresponding to the cone. The covariance matrix is ​​then calculated based on the coordinates of the centered vertex, the mean vector, and the number of all vertices. The coordinates of the centered vertex are the coordinates of all vertices after centering based on the mean vector. The covariance matrix is ​​then subjected to eigenvalue decomposition to obtain multiple eigenvalues ​​and multiple eigenvectors corresponding to the eigenvalues. The eigenvector corresponding to the second largest eigenvalue is determined as the central axis vector of the cone.

[0012] A second aspect of this application provides a device for acquiring parameters of the lumbar endplate, comprising: The acquisition unit is used to obtain the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine. The midline vector is parallel to the left and right direction of the human body. The calculation unit is used to perform interpolation fitting based on the centroid coordinates of all cones to obtain the spinal center curve and calculate the tangent vector of the spinal center curve at the centroid of each cone. The unit is defined as follows: for each cone, the product of the inverse vector of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. The cutting unit is used to cut the pre-constructed target triangular mesh covering the surface of the cone according to the cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes that include multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper and lower end plates of the cone. The element determination is also used to analyze the target mesh for each cone to obtain the parameters of the final plate.

[0013] A third aspect of this application provides a device for acquiring parameters of the lumbar endplate, comprising: Central processing unit, memory, and input / output interfaces; The memory can be either temporary or permanent storage. The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the aforementioned method.

[0014] A fourth aspect of this application provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the aforementioned method.

[0015] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: First, the centroid coordinates and midline vector of each vertebra are obtained from the initial image of the lumbar spine. Then, interpolation fitting is performed based on the centroid coordinates of all vertebrae to obtain the spinal center curve, and the tangent vector of the spinal center curve at the centroid of each vertebra is calculated. For each vertebra, the product of the inverse vector of the cross product of the midline vector and the tangent vector and the centroid coordinate is determined as a constant term. The cutting plane is determined based on the cross product and the constant term. Then, the pre-constructed target triangular mesh covering the vertebrae surface is cut according to the cutting plane to obtain the target mesh. Finally, for each vertebra, the target mesh is analyzed to obtain the parameters of the endplate. The method of this application is based on automatic execution by the device to obtain the parameters of the lumbar vertebrae endplate. The spinal center curve is used for cutting to obtain the target mesh. Since the spinal condition of each patient is different, a corresponding relatively objective spinal center curve is constructed for each patient's spine. The endplate parameters obtained based on the spinal center curve are more accurate and have higher precision, which can meet higher requirements.

[0016] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of an embodiment of a method for obtaining parameters of the lumbar endplate disclosed in this application; Figure 2 This is a schematic diagram of another embodiment of the method for obtaining parameters of the lumbar endplate disclosed in this application; Figure 3 This is a schematic diagram of an embodiment of the parameter acquisition device for the lumbar endplate disclosed in this application; Figure 4 This is a schematic diagram of another embodiment of the parameter acquisition device for the lumbar endplate disclosed in this application.

[0019] The realization of the objectives, functional features and advantages of the embodiments of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] To conduct real-time research and analysis of the lumbar spine, it is necessary to understand the parameters of the lumbar endplates. In existing methods, after acquiring images of the lumbar spine, medical personnel manually annotate key points based on practical experience and calculate the relevant parameters of the lumbar endplates based on these key points. However, the current method, relying on manual annotation to obtain lumbar endplate parameters, is highly subjective, resulting in low accuracy and failing to meet higher requirements. To address these technical problems, this application provides a method and related apparatus for obtaining lumbar endplate parameters. The method is based on automatic device execution to obtain lumbar endplate parameters. It utilizes the spinal center curve for cutting operations to obtain a target mesh. Since each patient's spinal condition is different, a corresponding, more objective spinal center curve is constructed for each patient's spine. The endplate parameters obtained based on the spinal center curve are more accurate and precise, meeting higher requirements.

[0021] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the embodiments of this application, and should not be construed as limiting the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of the embodiments of this application.

[0022] In the description of the embodiments of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "circumferential", "radial", etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "first," "second," "third," "fourth," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. In the description of embodiments of this application, "a plurality of" means two or more, unless otherwise expressly specified.

[0024] In the embodiments of this application, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0025] In the embodiments of this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature being directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0026] The following describes a method for obtaining parameters of the lumbar endplate according to this application. Please refer to [link / reference]. Figure 1 An embodiment of the method for obtaining parameters of the lumbar endplate according to this application includes: 101. Based on the initial image of the lumbar vertebrae, obtain the centroid coordinates and midline vector of each vertebra; Based on the acquired initial images of the lumbar spine, the centroid coordinates and midline vector of each vertebra are obtained, where the midline vector is parallel to the left-right direction of the body. The initial images are computed tomography (CT) images, and their format can be DICOM or other formats, which are not limited here. The lumbar spine consists of multiple vertebrae, each with its own centroid and corresponding coordinates.

[0027] 102. Based on the centroid coordinates of all cones, perform interpolation fitting to obtain the spinal center curve, and calculate the tangent vector of the spinal center curve at the centroid of each cone. Interpolation fitting is performed based on the centroid coordinates of all cones to obtain the spinal center curve, and the tangent vector of the spinal center curve at the centroid of each cone is calculated. Specifically, sample interpolation fitting can be performed based on a preset deep learning segmentation algorithm to obtain the spinal center curve passing through the coordinates of all centroids. The spinal center curve has a tangent at the centroid of each cone, and the corresponding tangent vector can be calculated based on the cumulative chord length parameter.

[0028] 103. For each cone, the product of the inverse of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. For each cone, the product of the inverse of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term. The cutting plane is then determined based on the cross product and the constant term. The cross product is the normal vector of the cutting plane, and the cutting plane is determined by the coefficient of the normal vector and the constant term.

[0029] 104. Cut the pre-constructed target triangular mesh covering the cone surface according to the cutting plane to obtain the target mesh; The pre-constructed target triangular mesh covering the surface of the cone is cut using a cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes comprising multiple triangular facets. The target mesh covers the surfaces of the upper and lower end plates of the cone. The target triangular mesh can be constructed using a 3D algorithm based on a photopolymerization stereo modeling format, and the target mesh is obtained after cutting with the cutting plane.

[0030] 105. For each cone, analyze the target mesh to obtain the parameters of the final plate.

[0031] For each cone, the parameters of the final plate are obtained by analyzing the target mesh. The target mesh can be processed first to distinguish between the upper and lower final plate sections, and then each section can be analyzed separately to obtain its respective final plate parameters.

[0032] In this embodiment, the centroid coordinates and midline vector of each vertebra are first obtained from the initial image of the lumbar spine. Then, interpolation fitting is performed based on the centroid coordinates of all vertebrae to obtain the spinal center curve, and the tangent vector of the spinal center curve at the centroid of each vertebra is calculated. For each vertebra, the product of the inverse vector of the cross product of the midline vector and the tangent vector and the centroid coordinate is determined as a constant term. The cutting plane is determined based on the cross product and the constant term. Then, the pre-constructed target triangular mesh covering the vertebrae surface is cut according to the cutting plane to obtain the target mesh. Finally, for each vertebra, the target mesh is analyzed to obtain the parameters of the endplate. The method of this application is based on automatic execution by the device to obtain the parameters of the lumbar vertebrae endplate. It uses the spinal center curve to perform the cutting operation to obtain the target mesh. Since the spinal condition of each patient is different, a corresponding relatively objective spinal center curve is constructed for each patient's spine. The parameters of the endplate obtained based on the spinal center curve are more accurate and have higher precision, which can meet higher requirements.

[0033] Please see Figure 2 Another embodiment of the method for obtaining parameters of the lumbar endplate according to this application includes: 201. Based on the initial image of the lumbar vertebrae, obtain the centroid coordinates and midline vector of each vertebra; Based on the acquired initial images of the lumbar spine, the centroid coordinates and midline vector of each vertebra are obtained, where the midline vector is parallel to the left-right direction of the body. The initial images are computed tomography (CT) images, and their format can be DICOM or other formats, which are not limited here. The lumbar spine consists of multiple vertebrae, each with its own centroid and corresponding coordinates.

[0034] In one implementation, an initial image of the lumbar spine is first acquired, and then resampling and voxel spacing normalization are performed on the initial image to obtain the target image. Specifically, the initial CT image is preprocessed, such as through resampling and voxel spacing normalization, to obtain the corresponding target image, which is the processed CT image.

[0035] Next, based on a preset deep learning segmentation algorithm, the target image is subjected to 3D cone segmentation processing to obtain 3D segmentation masks for multiple cones. Preferably, the deep learning segmentation algorithm can be the totalsegmentator algorithm, or other algorithms, which are not limited here.

[0036] Then, for each cone, based on a stereolithography algorithm, the 3D segmentation mask of the cone is processed to obtain the corresponding initial triangular mesh covering the cone surface. Morphological closing operations are then performed on the initial triangular mesh to obtain the corresponding target triangular mesh. Specifically, both the initial and target triangular meshes are composed of multiple triangular facets, which can be understood as multiple triangular facets closely surrounding the cone. Morphological closing operations can fill local holes and smooth boundaries.

[0037] Then, for each cone, based on a preset centroid coordinate algorithm, the centroid coordinates of the corresponding cone are calculated according to the coordinates of the vertices of all triangular faces of the target triangular mesh. Specifically, the centroid coordinate algorithm can be: ; in, Let be the coordinates of the centroid of the i-th cone. , and All of these are the coordinates of the j-th vertex on the target triangular mesh of the i-th cone. Let be the total number of vertices on the target triangular mesh of the i-th cone.

[0038] Finally, for each cone, the mean vector is calculated based on the coordinates of all vertices of the target triangular mesh corresponding to the cone. The covariance matrix is ​​then calculated using the centered vertex coordinates, the mean vector, and the number of vertices. The centered vertex coordinates are the coordinates of all vertices after centering based on the mean vector. The covariance matrix is ​​then subjected to eigenvalue decomposition to obtain multiple eigenvalues ​​and their corresponding eigenvectors. The eigenvector corresponding to the second largest eigenvalue is determined as the central axis vector of the cone. This step uses principal component analysis to determine the central axis vector. Specifically, after obtaining the mean vector, the formulas for calculating the covariance matrix and eigenvalues ​​are as follows: ; ; in, Let covariance matrix be the variance matrix. Let be the total number of vertices on the target triangular mesh of the i-th cone. A vertex on the triangular facet. It is the mean vector. The eigenvector matrix, Let be the eigenvalue diagonal matrix. After obtaining the eigenvector matrix and the eigenvalue diagonal matrix, the eigenvector corresponding to the second largest eigenvalue is determined as the median vector.

[0039] 202. Based on the centroid coordinates of all cones, perform interpolation fitting to obtain the spinal center curve, and calculate the tangent vector of the spinal center curve at the centroid of each cone. Interpolation fitting is performed based on the centroid coordinates of all cones to obtain the spinal center curve, and the tangent vector of the spinal center curve at each cone centroid is calculated. Preferably, cubic spline difference fitting can be performed to construct the spinal center curve, and the cumulative chord length parameter of each centroid can be calculated. The tangent vector is then calculated based on the cumulative chord length parameter. In one embodiment, the formula is: ; in, The tangent vector, For the cumulative chord length parameter, Let be the first derivative of the curve in the x-direction. Let be the first derivative of the curve in the y-direction. Let be the first derivative of the curve in the z-direction.

[0040] 203. For each cone, the product of the inverse of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. For each cone, the product of the inverse of the cross product of the median vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. The cross product is the normal vector of the cutting plane. In one embodiment, the cutting plane is: ; The relevant formula is: ; ; in, The result of the cross product. Using the centroid coordinates, For constant terms, and These are the midline vector and the tangent vector.

[0041] 204. Cut the pre-constructed target triangular mesh covering the cone surface according to the cutting plane to obtain the target mesh; The pre-constructed target triangular mesh covering the surface of the cone is cut using a cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes consisting of multiple triangular facets. The target mesh covers the surfaces of the upper and lower endplates of the cone. The cutting plane cuts the target triangular mesh into the target mesh, which "wraps" around the cone body, upper endplate, and lower endplate, while other irrelevant tissues such as spikes are ignored. Irrelevant tissues and the target mesh can be distinguished based on a pre-defined model. For example, the model can be trained using mesh instances covering the surfaces of the cone body, upper endplate, and lower endplate, allowing for differentiation.

[0042] 205. For each cone, determine the upper endplate patch covering the upper endplate surface and the lower endplate patch covering the lower endplate surface from the target mesh. For each cone, an upper endplate patch covering the upper endplate surface and a lower endplate patch covering the lower endplate surface are determined from the target mesh. There are many implementation methods, such as principal component analysis or the methods described below, and no specific method is limited here. Specifically, in one implementation, for each cone, triangular patches in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is greater than a first preset threshold are determined as upper endplate patches covering the upper endplate surface; triangular patches in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is less than a second preset threshold are determined as lower endplate patches covering the lower endplate surface. Simply put, triangular patches whose cosine value of the angle between the normal vector and the tangent vector is greater than the first preset threshold are identified as upper endplate patches, and the process for determining lower endplate patches is similar. Preferably, the first preset threshold can be 0.84, and the second preset threshold is -0.84. It is understandable that the normal vector of the triangular facet can also be compared with the vertical downward vector, but the specifics are not limited here.

[0043] 206. Perform plane fitting processing on the points on the upper end plate and the lower end plate respectively using the least squares method to obtain the corresponding upper end plate fitting plane and lower end plate fitting plane, and calculate the parameters of the end plate based on the upper end plate fitting plane and lower end plate fitting plane. The points on the upper and lower final plate surfaces are subjected to least squares plane fitting, respectively, to obtain the corresponding upper and lower final plate fitting planes. The parameters of the final plate are then calculated based on these fitting planes. The step of "applying least squares plane fitting to the points on the upper and lower final plate surfaces to obtain the corresponding upper and lower final plate fitting planes" can be implemented in several ways, and no specific implementation is limited here. Two implementation methods are described below.

[0044] In one implementation, based on a preset relationship between the distance from a point to the centroid of the cone and the point weights, the corresponding weights of the upper endplate points are obtained from the points on the upper endplate surface, and the corresponding weights of the lower endplate points are obtained from the points on the lower endplate surface. Using the least squares method, the corresponding upper fitting plane coefficients are calculated based on the upper endplate point weights and the coordinates of the points on the upper endplate surface, thus obtaining the upper endplate fitting plane. Similarly, using the least squares method, the corresponding lower fitting plane coefficients are calculated based on the lower endplate point weights and the coordinates of the points on the lower endplate surface, thus obtaining the lower endplate fitting plane. It is understood that least squares calculations can also be performed without setting weights.

[0045] In another implementation, upper fitting points with a predetermined percentage of distance from the centroid of the cone are selected from the points on the upper end plate, and lower fitting points with a predetermined percentage of distance from the centroid of the cone are selected from the points on the lower end plate. Based on the least squares method, the corresponding upper fitting plane coefficients are calculated based on the upper fitting points to obtain the upper end plate fitting plane. Similarly, based on the least squares method, the corresponding lower fitting plane coefficients are calculated based on the lower fitting points to obtain the lower end plate fitting plane. The predetermined percentage can be 80% or other values, and can be set according to actual needs; no specific limitation is made here.

[0046] The step "calculate the parameters of the end plate based on the fitting plane of the upper end plate and the fitting plane of the lower end plate" involves parameters including the upper concave depth, the lower concave depth, the upper concave angle, and the lower concave angle.

[0047] Specifically, based on a preset indentation depth formula, the distance between the upper indentation point of the upper end plate and the fitting plane of the upper end plate is determined as the upper indentation depth, and based on the preset indentation depth formula, the distance between the lower indentation point of the lower end plate and the fitting plane of the lower end plate is determined as the lower indentation depth. In one embodiment, the indentation point is: ; The formula for the depth of the depression is: ; in, The depth of the depression , and The coefficients are the fitting plane coefficients of the final plate.

[0048] Specifically, the two points closest to the upper concave point among the boundary points of the upper endplate fitting plane are designated as the first point and the second point. The angle between the first and second connecting lines is designated as the upper concave angle. The first connecting line is the straight line between the first point and the upper concave point, and the second connecting line is the straight line between the second point and the upper concave point. The two points closest to the lower concave point among the boundary points of the lower endplate fitting plane are designated as the third point and the fourth point. The angle between the third and fourth connecting lines is designated as the lower concave angle. The third connecting line is the straight line between the third point and the lower concave point, and the fourth connecting line is the straight line between the fourth point and the lower concave point.

[0049] 207. Calculate the surface area, front-to-back diameter, and left-to-right diameter of the upper and lower end plates respectively, and calculate the included angle and distance between two adjacent end plates.

[0050] Calculate the surface area, front-to-back diameter, and left-to-right diameter of the upper and lower end plates respectively, and calculate the included angle and distance between two adjacent end plates.

[0051] Specifically, for each cone, the sum of the areas of all the upper endplate surfaces is determined as the surface area of ​​the upper endplate, and the sum of the areas of all the lower endplate surfaces is determined as the surface area of ​​the lower endplate.

[0052] For each cone, the maximum projection distance of a point on the upper end plate in the front-back direction is determined as the front-back diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the front-back direction is determined as the front-back diameter of the lower end plate, wherein the front-back direction is perpendicular to the tangent vector and the midline vector, respectively.

[0053] For each cone, the maximum projection distance of a point on the upper end plate in the left-right direction is determined as the left-right diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the left-right direction is determined as the left-right diameter of the lower end plate.

[0054] The cosine of the angle between the normal vector of the upper endplate fitting plane of the current cone and the normal vector of the lower endplate fitting plane of the previous cone is determined as the endplate angle between the two endplates. The Euclidean distance between the center point of the upper endplate fitting plane of the current cone and the center point of the lower endplate fitting plane of the previous cone is determined as the distance between the two cones. It is understood that other corresponding points can be chosen; the center point is not required. Finally, all relevant parameters are output in structured JSON format.

[0055] Furthermore, the method in this embodiment can process initial images from multiple patients in batches, ultimately returning a structured result dictionary. It supports both one-time batch processing and periodic scheduled processing, achieving automation. It can also unify the local coordinates of each cone, for example, using the principal axis direction of one cone as a reference, unifying the principal axis directions of other cones with that cone.

[0056] In this embodiment, the method is based on automatic device execution to obtain parameters of the lumbar endplates. The spinal center curve is used for cutting to obtain the target mesh. Since each patient's spinal condition is different, a corresponding, more objective spinal center curve is constructed for each patient's spine. The endplate parameters obtained based on this spinal center curve are more accurate and precise, meeting higher requirements. Furthermore, the method can automatically calculate the relevant endplate parameters and set reasonable thresholds to classify the upper and lower endplates, improving the reliability of identification and ensuring high reproducibility.

[0057] The above describes a method for obtaining parameters of the lumbar endplate according to an embodiment of this application. The following describes a device for obtaining parameters of the lumbar endplate according to an embodiment of this application. Please refer to... Figure 3 One embodiment of the parameter acquisition device for the lumbar endplate in this application includes: The acquisition unit 301 is used to obtain the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine. The midline vector is parallel to the left and right direction of the human body. The calculation unit 302 is used to perform interpolation fitting based on the centroid coordinates of all cones to obtain the spinal center curve, and to calculate the tangent vector of the spinal center curve at the centroid of each cone. The determining unit 303 is used to determine the constant term by multiplying the inverse vector of the cross product of the midline vector and the tangent vector with the centroid coordinates for each cone, and to determine the cutting plane based on the cross product result and the constant term. Cutting unit 304 is used to cut the pre-constructed target triangular mesh covering the surface of the cone according to the cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes including multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper end plate and the lower end plate of the cone. The element 303 is also used to analyze the target mesh for each cone to obtain the parameters of the final plate.

[0058] In this embodiment, the acquisition unit 301 first obtains the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine. Then, the calculation unit 302 performs interpolation fitting based on the centroid coordinates of all vertebrae to obtain the spinal center curve and calculates the tangent vectors of the spinal center curve at the centroid of each vertebra. The determination unit 303 then, for each vertebra, determines the constant term by multiplying the inverse vector of the cross product of the midline vector and the tangent vector with the centroid coordinates, and determines the cutting plane based on the cross product and the constant term. Next, the cutting unit 304 cuts the pre-constructed target triangular mesh covering the vertebrae surface according to the cutting plane to obtain the target mesh. Finally, for each vertebra, the target mesh is analyzed to obtain the parameters of the endplate. The method of this application is based on automatic device execution to obtain the parameters of the lumbar vertebrae endplate. It utilizes the spinal center curve for cutting to obtain the target mesh. Since each patient's spinal condition is different, a corresponding, relatively objective spinal center curve is constructed for each patient's spine. The endplate parameters obtained based on the spinal center curve are more accurate and precise, meeting higher requirements. The device can also be integrated into medical imaging workstations or CT / MRI terminals as a real-time measurement module, or as a PACS plugin or cloud service to provide remote measurement capabilities. It is also compatible with mobile applications and scientific data acquisition platforms, making it easy to promote and apply in clinical, scientific research and commercial scenarios.

[0059] The following is a detailed description of a parameter acquisition device for lumbar endplates according to an embodiment of this application. Another embodiment of the parameter acquisition device for lumbar endplates according to an embodiment of this application includes: The acquisition unit is used to obtain the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine. The midline vector is parallel to the left and right direction of the human body. The calculation unit is used to perform interpolation fitting based on the centroid coordinates of all cones to obtain the spinal center curve and calculate the tangent vector of the spinal center curve at the centroid of each cone. The unit is defined as follows: for each cone, the product of the inverse vector of the cross product of the midline vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. The cutting unit is used to cut the pre-constructed target triangular mesh covering the surface of the cone according to the cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes that include multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper and lower end plates of the cone. The element determination is also used to analyze the target mesh for each cone to obtain the parameters of the final plate.

[0060] Determine the unit, specifically for: For each cone, determine the upper end plate patch covering the upper end plate surface and the lower end plate patch covering the lower end plate surface from the target mesh; The least squares method is used to perform plane fitting on the points on the upper and lower end plate patches respectively to obtain the corresponding upper and lower end plate fitting planes. The parameters of the end plate are then calculated based on the upper and lower end plate fitting planes.

[0061] Determine the unit, specifically for: For each cone, the triangular facets in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is greater than a first preset threshold are determined as the upper end plate facets covering the upper end plate surface; the triangular facets in the target mesh whose cosine value of the angle between the normal vector and the tangent vector is less than a second preset threshold are determined as the lower end plate facets covering the lower end plate surface.

[0062] Determine the unit, specifically for: Based on the preset relationship between the distance from a point to the centroid of the cone and the point weight, the corresponding weights of the upper end plate points are obtained according to the points on the upper end plate surface, and the corresponding weights of the lower end plate points are obtained according to the points on the lower end plate surface. Based on the least squares method, the corresponding upper fitting plane coefficients are calculated according to the weights of the upper end plate points and the coordinates of the points on the upper end plate surface, thus obtaining the upper end plate fitting plane. Based on the least squares method, the corresponding lower fitting plane coefficients are calculated according to the weights of the lower end plate points and the coordinates of the points on the lower end plate surface, thus obtaining the lower end plate fitting plane. or, From the points on the upper end plate, select the upper fitting points that are farthest from the centroid of the cone by a predetermined percentage. From the points on the lower end plate, select the lower fitting points that are farthest from the centroid of the cone by a predetermined percentage. Based on the least squares method, calculate the corresponding upper fitting plane coefficients based on the upper fitting points to obtain the upper end plate fitting plane. Based on the least squares method, calculate the corresponding lower fitting plane coefficients based on the lower fitting points to obtain the lower end plate fitting plane.

[0063] The parameters include the upper recess depth, lower recess depth, upper recess angle, and lower recess angle, which determine the element and are specifically used for: Based on the preset concave depth formula, the distance between the upper concave point of the upper end plate and the fitting plane of the upper end plate is determined as the upper concave depth, and based on the preset concave depth formula, the distance between the lower concave point of the lower end plate and the fitting plane of the lower end plate is determined as the lower concave depth. The two points closest to the upper concave point among the boundary points of the upper endplate fitting plane are designated as the first point and the second point. The angle between the first and second connecting lines is designated as the upper concave angle. The first connecting line is the straight line between the first point and the upper concave point, and the second connecting line is the straight line between the second point and the upper concave point. The two points closest to the lower concave point among the boundary points of the lower endplate fitting plane are designated as the third point and the fourth point. The angle between the third and fourth connecting lines is designated as the lower concave angle. The third connecting line is the straight line between the third point and the lower concave point, and the fourth connecting line is the straight line between the fourth point and the lower concave point.

[0064] The device also includes a processing unit for: For each cone, the sum of the areas of all the upper endplates is determined as the surface area of ​​the upper endplate, and the sum of the areas of all the lower endplates is determined as the surface area of ​​the lower endplate. For each cone, the maximum projection distance of a point on the upper end plate in the front-back direction is determined as the front-back diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the front-back direction is determined as the front-back diameter of the lower end plate, wherein the front-back direction is perpendicular to the tangent vector and the midline vector, respectively. For each cone, the maximum projection distance of a point on the upper end plate in the left-right direction is determined as the left-right diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the left-right direction is determined as the left-right diameter of the lower end plate. The cosine of the angle between the normal vector of the upper endplate fitting plane of the current cone and the normal vector of the lower endplate fitting plane of the previous cone is determined as the endplate angle between the two endplates, and the Euclidean distance between the center point of the upper endplate fitting plane of the current cone and the center point of the lower endplate fitting plane of the previous cone is determined as the distance between the two cones.

[0065] The acquisition unit is specifically used for: The initial image of the lumbar spine is acquired, and the initial image is resampled and the voxel spacing is normalized to obtain the target image. Based on a preset deep learning segmentation algorithm, the target image is processed into three-dimensional cone segmentation to obtain three-dimensional segmentation masks of multiple cones. For each cone, based on the stereolithography algorithm, the three-dimensional segmentation mask of the cone is processed to obtain the corresponding initial triangular mesh covering the surface of the cone, and the initial triangular mesh is processed by morphological closing operation to obtain the corresponding target triangular mesh. For each cone, the centroid coordinates of the corresponding cone are calculated based on the coordinates of the vertices of all triangular faces of the target triangular mesh, using a preset centroid coordinate algorithm. For each cone, the mean vector is calculated based on the coordinates of all vertices of the target triangular mesh corresponding to the cone. The covariance matrix is ​​then calculated based on the coordinates of the centered vertex, the mean vector, and the number of all vertices. The coordinates of the centered vertex are the coordinates of all vertices after centering based on the mean vector. The covariance matrix is ​​then subjected to eigenvalue decomposition to obtain multiple eigenvalues ​​and multiple eigenvectors corresponding to the eigenvalues. The eigenvector corresponding to the second largest eigenvalue is determined as the central axis vector of the cone.

[0066] The functions and processes performed by each unit in the parameter acquisition device for the lumbar endplate in this embodiment are the same as those described above. Figures 1 to 2 The function and process of the parameter acquisition device for the mid-lumbar endplate are similar, and will not be described in detail here.

[0067] Figure 4 This is a schematic diagram of a parameter acquisition device for lumbar endplates provided in an embodiment of this application. The parameter acquisition device 400 for lumbar endplates may include one or more central processing units (CPUs) 401 and a memory 405, in which one or more application programs or data are stored.

[0068] The memory 405 can be volatile or persistent storage. The program stored in the memory 405 can include one or more modules, each module including a series of instruction operations on the lumbar endplate parameter acquisition device 400. Furthermore, the central processing unit 401 can be configured to communicate with the memory 405 and execute the series of instruction operations in the memory 405 on the lumbar endplate parameter acquisition device 400.

[0069] The parameter acquisition device 400 for the lumbar endplate may also include one or more power supplies 402, one or more wired or wireless network interfaces 403, one or more input / output interfaces 404, and / or one or more operating systems, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0070] The central processing unit 401 can perform the aforementioned... Figures 1 to 2 The specific operations performed by the parameter acquisition device for the lumbar endplate in the illustrated embodiment will not be described in detail here.

[0071] This application also provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the methods described in the foregoing embodiments.

[0072] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0073] It should be noted that although the steps in the flowcharts of the various embodiments are drawn sequentially according to the arrows, unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the various embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.

[0074] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the embodiments of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0075] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0076] The above are merely preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structural transformations made using the description and drawings of the present application under the inventive concept of the present application, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present application.

Claims

1. A method for obtaining parameters of the lumbar vertebral endplate, characterized in that, include: Based on the initial image of the lumbar spine, the centroid coordinates and midline vector of each vertebra are obtained, and the midline vector is parallel to the left-right direction of the human body; Interpolation fitting is performed based on the centroid coordinates of all the cones to obtain the spinal center curve, and the tangent vector of the spinal center curve at the centroid of each cone is calculated. For each cone, the product of the inverse vector of the cross product of the central axis vector and the tangent vector and the centroid coordinates is determined as a constant term, and the cutting plane is determined based on the cross product and the constant term. The pre-constructed target triangular mesh covering the surface of the cone is cut according to the cutting plane to obtain the target mesh. Both the target triangular mesh and the target mesh are meshes including multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper and lower end plates of the cone. For each cone, the parameters of the final plate are obtained by analyzing the target mesh.

2. The method for obtaining parameters of the lumbar endplate according to claim 1, characterized in that, The parameters of the final plate obtained by analyzing the target mesh include: For each cone, an upper endplate patch covering the upper endplate surface and a lower endplate patch covering the lower endplate surface are determined from the target mesh; The points on the upper end plate and the points on the lower end plate are respectively subjected to plane fitting using the least squares method to obtain the corresponding upper end plate fitting plane and lower end plate fitting plane, and the parameters of the end plate are calculated based on the upper end plate fitting plane and the lower end plate fitting plane.

3. The method for obtaining parameters of the lumbar endplate according to claim 2, characterized in that, For each of the cones, determining the upper endplate patch covering the upper endplate surface and the lower endplate patch covering the lower endplate surface from the target mesh includes: For each cone, the triangular facet in the target mesh whose cosine of the angle between the normal vector and the tangent vector is greater than a first preset threshold is determined as the upper end plate facet covering the upper end plate surface; the triangular facet in the target mesh whose cosine of the angle between the normal vector and the tangent vector is less than a second preset threshold is determined as the lower end plate facet covering the lower end plate surface.

4. The method for obtaining parameters of the lumbar endplate according to claim 2, characterized in that, The step of performing least squares plane fitting on the points on the upper endplate and the points on the lower endplate respectively to obtain the corresponding upper endplate fitting plane and lower endplate fitting plane includes: Based on the preset relationship between the distance from a point to the centroid of the cone and the point weight, the corresponding upper end plate point weights are obtained according to the points on the upper end plate surface, and the corresponding lower end plate point weights are obtained according to the points on the lower end plate surface. Based on the least squares method, the corresponding upper fitting plane coefficients are calculated according to the upper end plate point weights and the coordinates of the points on the upper end plate surface, thereby obtaining the upper end plate fitting plane. Similarly, based on the least squares method, the corresponding lower fitting plane coefficients are calculated according to the lower end plate point weights and the coordinates of the points on the lower end plate surface, thereby obtaining the lower end plate fitting plane. or, From the points on the upper endplate, select the upper fitting points that are farthest from the centroid of the cone by a predetermined percentage; from the points on the lower endplate, select the lower fitting points that are farthest from the centroid of the cone by a predetermined percentage. Based on the least squares method, calculate the corresponding upper fitting plane coefficients based on the upper fitting points to obtain the upper endplate fitting plane. Based on the least squares method, calculate the corresponding lower fitting plane coefficients based on the lower fitting points to obtain the lower endplate fitting plane.

5. The method for obtaining parameters of the lumbar endplate according to claim 2, characterized in that, The parameters include the upper concave depth, the lower concave depth, the upper concave angle, and the lower concave angle. The calculation of the endplate parameters based on the upper endplate fitting plane and the lower endplate fitting plane includes: Based on the preset concave depth formula, the distance between the upper concave point of the upper end plate and the fitting plane of the upper end plate is determined as the upper concave depth, and based on the preset concave depth formula, the distance between the lower concave point of the lower end plate and the fitting plane of the lower end plate is determined as the lower concave depth. The two points closest to the upper concave point among the boundary points of the upper endplate fitting plane are designated as the first point and the second point. The angle between the first line and the second line is designated as the upper concave angle. The first line is the straight line between the first point and the upper concave point, and the second line is the straight line between the second point and the upper concave point. The two points closest to the lower concave point among the boundary points of the lower endplate fitting plane are designated as the third point and the fourth point. The angle between the third line and the fourth line is designated as the lower concave angle. The third line is the straight line between the third point and the lower concave point, and the fourth line is the straight line between the fourth point and the lower concave point.

6. The method for obtaining parameters of the lumbar endplate according to claim 2, characterized in that, After calculating the parameters of the final plate based on the upper final plate fitting plane and the lower final plate fitting plane, the method further includes: For each cone, the sum of the areas of all the upper endplates is determined as the surface area of ​​the upper endplate, and the sum of the areas of all the lower endplates is determined as the surface area of ​​the lower endplate. For each cone, the maximum projection distance of a point on the upper end plate in the front-back direction is determined as the front-back diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the front-back direction is determined as the front-back diameter of the lower end plate, wherein the front-back direction is perpendicular to the tangent vector and the central axis vector, respectively. For each cone, the maximum projection distance of a point on the upper end plate in the left-right direction is determined as the left-right diameter of the upper end plate, and the maximum projection distance of a point on the lower end plate in the left-right direction is determined as the left-right diameter of the lower end plate. The cosine of the angle between the normal vector of the upper endplate fitting plane of the current cone and the normal vector of the lower endplate fitting plane of the previous cone is determined as the endplate angle between the two endplates, and the Euclidean distance between the center point of the upper endplate fitting plane of the current cone and the center point of the lower endplate fitting plane of the previous cone is determined as the distance between the two cones.

7. The method for obtaining parameters of the lumbar endplate according to claim 1, characterized in that, The step of obtaining the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine includes: An initial image of the lumbar spine is acquired, and the initial image is resampled and voxel spacing normalized to obtain the target image. Based on a preset deep learning segmentation algorithm, the target image is subjected to three-dimensional cone segmentation processing to obtain three-dimensional segmentation masks of multiple cones; For each cone, based on a stereolithography algorithm, the three-dimensional segmentation mask of the cone is processed to obtain an initial triangular mesh covering the surface of the cone, and the initial triangular mesh is subjected to morphological closing operation to obtain the corresponding target triangular mesh. For each cone, the centroid coordinates of the cone are calculated based on the coordinates of the vertices of all the triangular faces of the target triangular mesh, according to a preset centroid coordinate algorithm. For each cone, the mean vector is calculated based on the coordinates of all the vertices of the triangular facets of the target triangular mesh corresponding to the cone. The covariance matrix is ​​then calculated based on the centered vertex coordinates, the mean vector, and the number of vertices. The centered vertex coordinates are the coordinates of all vertices after centering based on the mean vector. The covariance matrix is ​​then subjected to eigenvalue decomposition to obtain multiple eigenvalues ​​and multiple eigenvectors corresponding to the eigenvalues. The eigenvector corresponding to the second largest eigenvalue is determined as the central axis vector of the cone.

8. A device for acquiring parameters of the lumbar vertebral endplate, characterized in that, include: The acquisition unit is used to obtain the centroid coordinates and midline vector of each vertebra based on the acquired initial image of the lumbar spine, wherein the midline vector is parallel to the left-right direction of the human body; The calculation unit is used to perform interpolation fitting based on the centroid coordinates of all the cones to obtain the spinal center curve, and to calculate the tangent vector of the spinal center curve at the centroid of each of the cones. The determining unit is configured to, for each of the cones, determine the product of the inverse vector of the cross product of the central axis vector and the tangent vector and the centroid coordinates as a constant term, and determine the cutting plane based on the cross product and the constant term; A cutting unit is used to cut a pre-constructed target triangular mesh covering the surface of the cone according to the cutting plane to obtain a target mesh. Both the target triangular mesh and the target mesh are meshes including multiple triangular facets. The coverage area of ​​the target mesh includes the surfaces of the upper and lower end plates of the cone. The determining unit is further configured to analyze the target mesh for each cone to obtain the parameters of the final plate.

9. A device for acquiring parameters of the lumbar vertebral endplate, characterized in that, include: Central processing unit, memory, and input / output interfaces; The memory is either a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instructions in the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Includes instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images