Vertebral collapse end plate area identification method

By automatically identifying the collapsed vertebral endplate area, the problem of doctors' difficulty in quickly identifying it is solved, achieving efficient and safe treatment of vertebral collapse.

CN121921238APending Publication Date: 2026-04-24JIANGSU PROVINCE HOSPITAL (THE FIRST AFFILIATED HOSPITAL OF NANJING MEDICAL UNIVERSITY)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411946283.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-12-26
Filing Date
2024-12-27
Publication Date
2026-04-24

Smart Images

  • Figure CN121921238A_ABST
    Figure CN121921238A_ABST
Patent Text Reader

Abstract

The invention discloses a centrum collapse end plate area identification method, which comprises the following steps: S1, acquiring a vertebral section image of a single vertebral section, and performing centrum hole center segmentation along the left-right direction of a human body to obtain a centrum image; s2, recognizing the front and rear edge end face directions of the vertebral body according to the vertebral body image obtained in the S1; s3, respectively defining a front edge plane and a rear edge plane according to the front edge end face direction and the rear edge end face direction of the cone and a point on the front edge plane and the rear edge plane obtained in the S2, and respectively translating a set distance into the cone along the corresponding normal directions by taking the defined front edge plane and the defined rear edge plane as references to obtain a front edge interception plane and a rear edge interception plane; and S4, respectively obtaining areas formed by points on the front end plate and the rear end plate of the cone obtained in the step S3, wherein the distance between the points and the corresponding front edge interception plane and the rear edge interception plane is greater than a set collapse distance threshold value, and the areas are corresponding collapse areas of the front end plate and the rear end plate. According to the invention, the corresponding front and rear end plate collapse areas on the front and rear end plates of the vertebral body can be automatically identified, the operation efficiency can be effectively improved, the treatment effect is optimized, and the pain of a patient is relieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method for identifying the endplate region of a collapsed vertebral body. Background Technology

[0002] Vertebral body collapse usually refers to a reduction in vertebral height caused by osteoporosis, compression fracture, or other pathological factors. Clinically, it is treated with spinal vertebral body expansion surgery. The main method is to use a spreader to open up the collapsed endplate area of ​​the vertebral body. Through the expansion effect of the spreader, the fractured vertebral body is repositioned and supported, partially restoring vertebral height and stability, thereby relieving the patient's pain and increasing the patient's mobility.

[0003] The conventional method involves doctors visually locating the collapsed area using X-ray fluoroscopy and manually inserting a diffuser. This method is inefficient and heavily reliant on the doctor's clinical experience, making it difficult to quickly and accurately identify the collapsed vertebral body during surgery. Therefore, the trend is towards replacing the traditional method with an automated diffuser insertion procedure. The automated identification of the collapsed vertebral endplate is of great significance for improving surgical efficiency, optimizing treatment outcomes, and reducing patient suffering. Summary of the Invention

[0004] Purpose of the invention: To address the above-mentioned problems, this invention proposes a method for identifying vertebral body collapse endplate regions, which can automatically identify vertebral body collapse endplates, effectively improving surgical efficiency, optimizing treatment outcomes, and reducing patient suffering.

[0005] Technical solution:

[0006] This invention provides a method for identifying the collapsed endplate region of the vertebral body, including:

[0007] S1. Obtain the image of a single vertebral segment, and segment it along the left-right direction of the human body, passing through the center of the vertebral foramen to obtain the vertebral body image;

[0008] S2. Identify the orientation of the anterior and posterior edges of the vertebral body in the vertebral body image obtained in S1;

[0009] S3. Based on the directions of the anterior and posterior edge faces of the vertebral body obtained in S2 and a point on them, define the anterior and posterior edge planes respectively. Using the defined anterior and posterior edge planes as references, translate them into the vertebral body by a set distance along their corresponding normal directions to obtain the anterior and posterior edge cutting planes.

[0010] S4. Obtain the regions formed by the points on the anterior and posterior endplates of the vertebral body obtained in S3 whose distance between them and the corresponding anterior and posterior edge cutting planes is greater than the set collapse distance threshold. These regions are the corresponding anterior and posterior endplate collapse regions.

[0011] Specifically, S2 includes:

[0012] S21. Calculate the centroid of the vertebral body in the vertebral body image obtained in S1, and construct a vertebral body interception plane through the centroid along the left-right direction of the human body.

[0013] S22. Set the angle step size so that the vertebral body in the vertebral body image obtained in S1 rotates around the axis along the front-back direction of the human body and passes through the centroid of the vertebral body in the vertebral body image. After each rotation, obtain the point farthest from the centroid of the vertebral body image among the intersection points of the vertebral body cutting plane and the vertebral body, and obtain the point set.

[0014] S23. Based on the point set obtained in S22, obtain the rotation angle corresponding to the intersection point closest to the centroid of the vertebra, thereby obtaining the corresponding vertebral pose;

[0015] S24. Maintain the vertebral body pose obtained in S23. Using the vertebral body cutting plane constructed in S21 as a reference, define several cutting planes along the anterior-posterior direction of the human body at set intervals, and calculate the cutting intersection points of these cutting planes and the vertebral body respectively.

[0016] S25. Fit a straight line to the set of points formed by the intersection points obtained in S24, and calculate the product between the normal direction of the fitted straight line and the vertebral pose obtained in S23. This product is the direction of the anterior and posterior edge faces of the vertebral body.

[0017] More specifically, in S22, the angle range of the vertebral body rotating around the axis can be taken as ±45° based on the vertebral body image obtained in S1.

[0018] More specifically, in S22, the angle step size is 0.3°-0.8°.

[0019] More specifically, in S24, based on the vertebral body cutting plane constructed in S21, several cutting planes are defined at set intervals within a range of ±10mm along the anterior-posterior direction of the human body.

[0020] The set interval is 0.5-1.5mm.

[0021] Specifically, in S3, the set distance for translation is 1.2-2.8mm.

[0022] Specifically, in S4, the set collapse distance threshold is 3mm.

[0023] Beneficial Effects: This invention constructs corresponding anterior and posterior cutting planes by identifying the anterior and posterior endface directions of the vertebral body in vertebral imaging. This allows for the segmentation of the anterior and posterior endplates of the vertebral body. By setting a collapse distance threshold, the corresponding anterior and posterior endplate collapse areas can be automatically identified, helping doctors assess the severity of vertebral collapse and guiding targeted surgical treatment. It also aids in postoperative monitoring of vertebral endplate recovery and evaluation of surgical outcomes. This invention allows for personalized diagnosis and treatment based on the specific vertebral collapse condition of different patients, improving the targetedness and effectiveness of treatment. Through automated image processing technology, it reduces the workload of doctors during surgery, improves surgical efficiency and treatment effectiveness, and resolves the uncertainty and potential surgical risks associated with visual assessment of collapse areas in traditional surgery, thus enhancing treatment safety. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in this invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the vertebral body collapse endplate region identification method of the present invention;

[0026] Figure 2 Example image of a single vertebral segment, where, Figure 2 (a) Draw an image of the surface of the vertebral body. Figure 2 (b) Draw an image of the vertebral body. Figure 2 (c) is an example diagram of the established vertebral coordinate system;

[0027] Figure 3 Example image of the segmented vertebral body; where 1 is the boundary outline of the endplate, and 2 is the endplate;

[0028] Figure 4 A schematic diagram for identifying the anterior and posterior end faces of the vertebral body, wherein, Figure 4 (a) is an example diagram of the established cone coordinate system. Figure 4 (b) An example diagram showing the intersection of the vertebral body after rotation and the vertebral body interception plane. Figure 4 (c) Example diagram to obtain the rotation angle corresponding to the intersection point with the smallest distance from the center of the vertebral foramen. Figure 4 (d) is an example diagram showing the orientation of the anterior end face of the identified vertebral body;

[0029] Figure 5 This is a schematic diagram of the anterior and posterior endplates of the vertebral body, in which... Figure 5(a) is an example diagram of the front and rear edge intercept planes. Figure 5 (b) is an example diagram of the anterior and posterior endplates of the vertebral body;

[0030] Figure 6 A schematic diagram to identify the collapse areas of the front and rear endplates, wherein, Figure 6 (a) is Figure 5 Example diagram of the anterior and posterior endplates of the vertebral body. Figure 6 (b) A schematic diagram for identifying the collapse areas of the front and rear end plates. Figure 6 (c) is an example diagram of the identified front and rear endplate collapse areas. Detailed Implementation

[0031] To make the objectives, technical solutions and advantages of the present invention clearer, the present application will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0032] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of the present invention should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0033] The method for identifying the collapsed endplate region of the vertebral body in this invention is as follows: Figure 1 As shown, the steps include:

[0034] S1. Obtain the image of a single vertebral segment, and segment it along the left-right direction of the human body, passing through the center of the vertebral foramen to obtain the vertebral body image;

[0035] In this invention, after obtaining the image of a single vertebral segment, based on human anatomy, the vertebral body image can be obtained by segmenting along the left-right direction of the human body through the center of the vertebral foramen. In fact, the vertebral segment is segmented by passing through the two pedicle isthmuses of the vertebral segment, thereby obtaining the vertebral body image.

[0036] In a specific embodiment of the present invention, a three-dimensional image containing the thoracic or lumbar vertebrae is acquired, and the three-dimensional images of each vertebral segment are obtained by segmentation. For any segment of the human body, the collapsed lamina region can be identified by the algorithm, which can comprehensively assess the lamina collapse of the entire thoracic or lumbar vertebrae. This covers the main part of the spine and has broad clinical application value.

[0037] Reference Figure 2 The three-dimensional image of a vertebral segment contains all the voxel information and triangular facet information of a single vertebral segment, such as Figure 2 (a) Draw an image of the vertebral body to represent the surface information of the segmented vertebral segments; Figure 2 (b) Draw a body image of the vertebral body, representing... Figure 2 (a) Voxel information of the segmented vertebrae.

[0038] In this invention, after obtaining the image of a single vertebral segment, a vertebral segment coordinate system is established. Specifically, referring to... Figure 2 (c) Based on human anatomy, the vertebral segment coordinate system is a coordinate system established with the center of the vertebral foramen of the vertebral segment as the origin, the axis of the vertebral foramen as the Y-axis, the left-right direction of the human body as the X-axis, and the front-back direction of the human body as the Z-axis.

[0039] In this invention, since a segmental coordinate system has already been established, the XY plane of the segmental coordinate system is directly used to segment the aforementioned segmental images through the center of the vertebral foramen. That is, along the left-right direction of the human body, the segmental images are obtained by segmenting the vertebral body through the center of the vertebral foramen. Figure 3 As shown.

[0040] S2. Identify the anterior and posterior end faces of the vertebral body in the vertebral body image obtained from S1;

[0041] In this invention, the direction of the leading and trailing edge end faces is defined as the planar direction corresponding to the boundary contours of the leading and trailing end plates, such as... Figure 3 As shown, 1 represents the boundary outline of the final plate, and 2 represents the final plate.

[0042] For the specific procedure of identifying the anterior and posterior end faces of the vertebral body, please refer to [link / reference needed]. Figure 4 Taking the identification of the anterior end face direction of the vertebral body as an example, it includes:

[0043] S21. Calculate the centroid of the vertebral body in the vertebral body image obtained in S1, and construct a vertebral body interception plane through the centroid along the left-right direction of the human body.

[0044] like Figure 4 As shown in (a), based on the vertebral body image obtained by S1, a vertebral body coordinate system is constructed. The directions of each coordinate axis of the vertebral body coordinate system are consistent with the aforementioned vertebral segment coordinate system. Its origin is taken as the centroid P1 of the vertebral body in the vertebral body image. Then, the vertebral body interception plane S0 is defined, which coincides with the XY plane of the vertebral body coordinate system.

[0045] S22. Set the angle step size so that the vertebral body in the vertebral body image obtained in S1 rotates around the axis along the front-back direction of the human body and passes through the centroid of the vertebral body in the vertebral body image. After each rotation, obtain the point with the largest distance from the centroid of the vertebral body image among the intersection points of the vertebral body cutting plane and the vertebral body, and then obtain the point set.

[0046] like Figure 4 As shown in (b), an angle step size is set so that the vertebral body rotates around the Z-axis of the vertebral coordinate system, which is the axis along the anterior-posterior direction of the human body and passes through the centroid of the vertebral body image. After each rotation, the point farthest from the centroid of the vertebral body among the intersection points of the vertebral body cutting plane and the vertebral body is obtained, that is, the intersection point Q with the largest X-direction component among the intersection points of the vertebral body cutting plane and the vertebral body. nIn this embodiment, the range of angles in which the vertebral body rotates around the aforementioned axis can be taken within ±45° of the vertebral body image obtained in S1. That is, with the vertebral body image obtained in S1 as the reference, the range of rotation of the vertebral body around the Z-axis of the vertebral coordinate system can be within 45° of its counterclockwise or clockwise rotation.

[0047] In a specific embodiment of the present invention, the angle step size ranges from 0.3° to 0.8°, and is preferably set to 0.5°.

[0048] S23. Based on the point set obtained in S22, obtain the rotation angle corresponding to the intersection point closest to the centroid of the vertebra, thereby obtaining the corresponding vertebral pose;

[0049] Reference Figure 4 (c) The intersection point Q of the point set obtained by S22 that is closest to the centroid of the cone. m The corresponding rotation angle β m Thus, the corresponding vertebral pose M is obtained. m At this point, the anterior end face of the vertebral body is approximately perpendicular to the X-axis.

[0050] S24. Maintain the vertebral body pose obtained in S23. Using the vertebral body cutting plane constructed in S21 as a reference, define several cutting planes along the anterior-posterior direction of the human body at set intervals, and calculate the cutting intersection points of these cutting planes and the vertebral body respectively.

[0051] In this invention, the vertebral body pose obtained in S23 is maintained, and the vertebral body cutting plane constructed in S21 is used as a reference. A series of cutting planes S are defined along the positive and negative directions of the Z-axis of the vertebral body coordinate system in the anterior-posterior direction of the human body. v The intersection point Q between the cutting plane and the vertebral body was calculated. v ,like Figure 4 As shown in (d).

[0052] In this invention, the range of the cutting plane defined in the positive and negative directions of the Z-axis of the vertebral coordinate system can be defined within ±10mm of the Z-axis of the vertebral coordinate system, with the vertebral cutting plane constructed by S21 as the reference. That is, with the vertebral cutting plane constructed by S21 as the reference, several cutting planes are defined at set intervals within ±10mm along the anterior-posterior direction of the human body.

[0053] In a specific embodiment of the present invention, the interval range is set to 0.5mm-1.5mm, preferably 1mm.

[0054] S25. Fit a straight line to the set of points formed by the intersection points obtained in S24, and calculate the product between the normal direction of the fitted straight line and the vertebral pose obtained in S23. This product is the direction of the anterior end face of the vertebral body.

[0055] Specifically, such as Figure 4As shown in (d), the formula for calculating the product between the normal direction of the fitted straight line and the vertebral pose obtained from S23 is as follows:

[0056]

[0057] in, To identify the orientation of the anterior end face of the vertebral body, To fit the straight line l h The direction of the normal.

[0058] The identification of the posterior end face direction of the vertebral body is consistent with the identification of the anterior end face direction of the vertebral body mentioned above.

[0059] S3. Based on the directions of the anterior and posterior end faces of the vertebral body obtained in S2 and a point on them, define the anterior and posterior edge planes respectively. Using the defined anterior and posterior edge planes as references, translate them into the vertebral body by a set distance along their corresponding normal directions to obtain the anterior and posterior edge intercepting planes. An example is shown below. Figure 5 As shown in (a), the anterior and posterior endplates of the vertebral body can be obtained by segmenting the corresponding anterior and posterior margin cutting planes, as illustrated in the example. Figure 5 As shown in (b).

[0060] In this invention, the anterior and posterior edge planes are shifted inwards by a predetermined distance to remove uneven and rugged portions of the endplate edge. This allows for a more accurate assessment of the actual condition of the endplate and avoids diagnostic errors caused by irregular edges and unstable data. The shifting distance ranges from 1.2mm to 2.8mm, with 2mm being the preferred setting.

[0061] S4. Obtain the regions formed by the points on the anterior and posterior endplates of the vertebral body obtained in S3 whose distance between the corresponding anterior and posterior edge cutting planes is greater than the set collapse distance threshold. These regions are the corresponding anterior and posterior endplate collapse regions.

[0062] In this invention, the most severely collapsed area of ​​the endplate is the area on the endplate that is furthest from its corresponding cutting plane. Taking the identification of the anterior endplate collapse area as an example, the distance d from each point on the anterior endplate of the vertebral body obtained by S3 to the anterior cutting plane is calculated. i This leads to the distance set, from which the collapse distance threshold d is set. h The collapse distance threshold d h Let d be a value from the aforementioned set of distances; each value on the anterior endplate of the vertebral body that satisfies distance d i >d h The set of points constitutes the anterior endplate collapse region ψ, as shown in 6(b) and 6(c). Similarly, the posterior endplate collapse region can be obtained.

[0063] In this invention, on the collapsed areas of the front and rear endplates, the point corresponding to the maximum value of the plane intercepted at the corresponding front and rear edges is the farthest point Gs, such as... Figure 6 As shown in (a) and 6(b).

[0064] In this embodiment, the set collapse distance threshold d h 3mm is acceptable.

[0065] This invention identifies the anterior and posterior endface directions of the vertebral body in vertebral imaging, constructing corresponding anterior and posterior cutting planes. This allows for the segmentation of the anterior and posterior endplates of the vertebral body. By setting a collapse distance threshold, the corresponding anterior and posterior endplate collapse areas can be automatically identified, helping doctors assess the severity of vertebral collapse and guiding targeted surgical treatment. It also aids in postoperative monitoring of vertebral endplate recovery and evaluation of surgical outcomes. This invention allows for personalized diagnosis and treatment based on the individual patient's vertebral collapse condition, improving the targetedness and effectiveness of treatment. Through automated image processing technology, it reduces the workload of doctors during surgery, improving surgical efficiency and treatment outcomes. It also eliminates the uncertainty and potential surgical risks associated with visual assessment of collapse areas in traditional surgery, enhancing treatment safety.

[0066] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of the invention as described above, which are not provided in the details for the sake of brevity.

[0067] The embodiments of this invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this invention should be included within the protection scope of this invention.

Claims

1. A method for identifying the collapsed endplate region of a vertebral body, characterized in that, include: S1. Obtain the image of a single vertebral segment, and segment it along the left-right direction of the human body, passing through the center of the vertebral foramen to obtain the vertebral body image; S2. Identify the orientation of the anterior and posterior edges of the vertebral body in the vertebral body image obtained in S1; S3. Based on the directions of the anterior and posterior edge faces of the vertebral body obtained in S2 and a point on them, define the anterior and posterior edge planes respectively. Using the defined anterior and posterior edge planes as references, translate them into the vertebral body by a set distance along their corresponding normal directions to obtain the anterior and posterior edge cutting planes. S4. Obtain the regions formed by the points on the anterior and posterior endplates of the vertebral body obtained in S3 whose distance between them and the corresponding anterior and posterior edge cutting planes is greater than the set collapse distance threshold. These regions are the corresponding anterior and posterior endplate collapse regions.

2. The method for identifying the collapsed endplate region of the vertebral body according to claim 1, characterized in that, S2 includes: S21. Calculate the centroid of the vertebral body in the vertebral body image obtained in S1, and construct a vertebral body interception plane through the centroid along the left-right direction of the human body. S22. Set the angle step size so that the vertebral body in the vertebral body image obtained in S1 rotates around the axis along the front-back direction of the human body and passes through the centroid of the vertebral body in the vertebral body image. After each rotation, obtain the point farthest from the centroid of the vertebral body image among the intersection points of the vertebral body cutting plane and the vertebral body, and obtain the point set. S23. Based on the point set obtained in S22, obtain the rotation angle corresponding to the intersection point closest to the centroid of the vertebra, thereby obtaining the corresponding vertebral pose; S24. Maintain the vertebral body pose obtained in S23. Using the vertebral body cutting plane constructed in S21 as a reference, define several cutting planes along the anterior-posterior direction of the human body at set intervals, and calculate the cutting intersection points of these cutting planes and the vertebral body respectively. S25. Fit a straight line to the set of points formed by the intersection points obtained in S24, and calculate the product between the normal direction of the fitted straight line and the vertebral pose obtained in S23. This product is the direction of the anterior and posterior edge faces of the vertebral body.

3. The method for identifying the collapsed endplate region of the vertebral body according to claim 2, characterized in that, In S22, the angle range of the vertebral body rotating around the axis can be taken as ±45° based on the vertebral body image obtained in S1.

4. The method for identifying the collapsed vertebral endplate region according to claim 2, characterized in that, In step S22, the angle step size is 0.3°-0.8°.

5. The method for identifying the collapsed endplate region of the vertebral body according to claim 2, characterized in that, In S24, based on the vertebral body cutting plane constructed in S21, several cutting planes are defined at set intervals within a range of ±10mm along the anterior-posterior direction of the human body.

6. The method for identifying the collapsed endplate region of the vertebral body according to claim 5, characterized in that, The set interval is 0.5-1.5mm.

7. The method for identifying the collapsed endplate region of the vertebral body according to claim 1, characterized in that, In S3, the set distance for translation is 1.2-2.8mm.

8. The method for identifying the collapsed endplate region of the vertebral body according to claim 1, characterized in that, In step S4, the set collapse distance threshold is 3mm.