Oligamentum flavum segmentation method and resection path planning method

Abstract

The invention discloses a ligamentum flavum ossificum segmentation method and a resection path planning method.The ligamentum flavum ossificum segmentation method comprises the steps that a three-dimensional image of a target area of a patient is taken, and a first bone substance area and a second bone substance area in the three-dimensional image are obtained through segmentation; and setting a Hu value threshold according to voxel characteristics of the ligamentum flavum ossiforme, and searching and extracting voxel points which are located between the first bone region and the second bone region and have Hu values greater than the Hu value threshold to obtain a ligamentum flavum ossiforme region. According to the method, the ossified ligamentum flavum can be accurately recognized and extracted from numerous tissue areas of a human body, and the accuracy and success rate of subsequent excision of the ossified ligamentum flavum are effectively ensured. According to the method, a safer and more effective ligamentum flavum excision path can be planned, operation guidance is provided for doctors, the success rate of ligamentum flavum excision is increased, and then the operation risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an ossified ligamentum flavum segmentation method and a resection path planning method. Background Art

[0002] The yellow ligament is a tissue structure in the spinal canal that maintains the stability of the spine. The yellow ligament is elastic and is the main ligament structure involved in the formation of the posterior wall of the spinal canal. Ossification of the yellow ligament refers to a disease in which the yellow ligament ossifies and abnormally proliferates, resulting in a decrease in the volume of the spinal canal. The hypertrophic yellow ligament compresses the surrounding muscles, fascia and other soft tissues, causing patients to experience low back pain. Stenosis in the spinal canal can also cause radiating pain and numbness in the lower limbs, causing great pain to patients.

[0003] Surgical removal of the ossified yellow ligament is the only effective treatment. Currently, CT scans are used to collect patient images before surgery, and an endoscope is inserted into the patient's body for observation during surgery. The imaging quality of the endoscope itself is average, and the yellow ligament is usually adhered to the surrounding dura mater and is not easy to separate. Correctly identifying and removing the ossified yellow ligament under endoscopy depends heavily on the doctor's clinical experience, which also affects the success rate of ossified yellow ligament resection. Therefore, how to accurately identify and extract the ossified yellow ligament from the many tissue areas of the human body and help plan a more effective surgical path is of great significance in helping more doctors guide surgical planning and improve the success rate of ossified yellow ligament resection in the future. Summary of the invention

[0004] Purpose of the invention: In view of the above-mentioned deficiencies, the present invention provides an ossified yellow ligament segmentation method and a resection path planning method, which can accurately identify and extract the ossified yellow ligament from many tissue areas of the human body, provide surgical guidance for doctors, and thus reduce surgical risks.

[0005] Technical solution:

[0006] The present invention provides a method for segmenting an ossified ligamentum flavum, comprising:

[0007] Acquire a three-dimensional image of a target area of ​​a patient, and segment the image to obtain a first bone area and a second bone area;

[0008] A Hu value threshold is set according to the voxel characteristics of the ossified yellow ligament, and voxel points with a Hu value greater than the Hu value threshold located between the first bone region and the second bone region are searched and extracted to obtain the ossified yellow ligament region.

[0009] Specifically, the search extracts voxel points between the first bone region and the second bone region whose Hu values ​​are greater than a Hu value threshold, specifically:

[0010] The voxel points in the first bone region are traversed, and a three-dimensional search space of a set size is established with any voxel point as the center. The voxel points in the three-dimensional search space whose Hu values ​​are greater than the set Hu value threshold and are not in the first bone region and the second bone region are extracted, and the voxel points are put into a seed point set. After the voxel points in the first bone region are traversed, the seed point set obtained is the ossified yellow ligament point set.

[0011] More specifically, the seed point set is traversed, and if the voxel value of any one of the 8 neighboring points of any voxel point is greater than the Hu value threshold and is not within the first bone region and the second bone region, the neighboring point is put into the seed point set, and after traversing the voxel points of the seed point set, the target point set is obtained, which is the final ossified yellow ligament point set.

[0012] More specifically, the three-dimensional search space of the set size is:

[0013] In the human anatomical coordinate system, the height of the three-dimensional search space is one-third of the calculated distance between the lowest point of the vertebra in the second bone region and the highest point of the first bone region, its width is one-fifth of the width of the first bone region, and its length is one-half of the depth of the vertebra in the second bone region.

[0014] Specifically, the ossified yellow ligament area is obtained as follows:

[0015] Calculate the boundary of the minimum circumscribed rectangle of the seed point set composed of the extracted points, and determine the candidate area of ​​the ossified yellow ligament based on it;

[0016] Based on the Sato classification method, the candidate regions of the ossified yellow ligament are classified into the original classification template that is most similar to it. The vertebral regions in the candidate regions of the ossified yellow ligament are excluded accordingly, and finally the ossified yellow ligament regions are obtained.

[0017] More specifically, the Sato classification method is used to classify the candidate region of the ossified yellow ligament into the original classification template that is most similar to it, as follows:

[0018] (1) For any original typing template image, scale and rotation transformation are performed on it to obtain multiple candidate typing template images;

[0019] a. Performing a scale transformation on each original typing template image to generate a series of scaled original typing template images;

[0020] b. Performing rotation transformation on each original typing template image to generate a series of rotated original typing template images.

[0021] (2) For each candidate classification template image obtained after scale and rotation transformation, a matching degree test is performed between it and the candidate region of the ossified yellow ligament, and the maximum matching coefficient is calculated. The candidate classification template is retained, that is, the candidate region of the ossified yellow ligament is classified into the corresponding candidate classification template.

[0022] Specifically, the segmentation obtains a first bone region and a second bone region in the target region of the patient, as follows:

[0023] The spinous process, lamina, and transverse process below the vertebral body in the vertebral segment are defined as the first bone region, and the vertebral body region is defined as the second bone region;

[0024] The three-dimensional image of the target area of ​​the patient is segmented using a pre-trained segmentation model to obtain a first bone area and a second bone area;

[0025] The segmentation model is obtained by training a plurality of three-dimensional images of the first bone region and the second bone region which are pre-marked as training samples.

[0026] The present invention also provides a method for planning an ossified yellow ligament resection path, comprising the steps of:

[0027] (1) obtaining a preoperative three-dimensional image of the patient's target area, and segmenting the ossified ligamentum flavum area according to the aforementioned ossified ligamentum flavum segmentation method;

[0028] (2) marking the necessary areas and constraint areas of the resection path in the preoperative three-dimensional image, using this as prior knowledge and adopting an ant colony algorithm to plan the resection path;

[0029] (3) obtaining a three-dimensional image of the target area of ​​the patient during the operation, and registering the surface point cloud of the vertebral segment therein with the surface point cloud of the vertebral segment in the preoperative three-dimensional image;

[0030] (4) Based on the registration in step (3), the ossified yellow ligament resection path marked in the preoperative three-dimensional image is transformed into the three-dimensional image acquired during the operation.

[0031] Beneficial effects: The present invention adopts an automated image segmentation method with high processing efficiency, and can accurately identify and extract the ossified yellow ligament from many tissue areas of the human body. By combining the voxel characteristics of the ossified yellow ligament and the shape prior characteristics of clinical image data of different types, the accuracy of positioning, identifying and segmenting the ossified yellow ligament is improved, which helps to formulate a more accurate surgical plan, thereby improving the success rate of the operation. The present invention can plan a safer and more effective ossified yellow ligament resection path, provide surgical guidance for doctors, improve the success rate of ossified yellow ligament resection, and thus reduce surgical risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are merely embodiments of the present invention, and ordinary technicians in this field can obtain other drawings based on these drawings without creative work.

[0033] Figure 1 is a flow chart of the ossified yellow ligament segmentation method of the present invention;

[0034] Figure 2 This is an example of a sagittal image of a CT image of a patient's affected area;

[0035] Figure 3 An example diagram of segmentation of a first bone region and a second bone region of a cross-sectional image of a CT image of an affected part of a patient;

[0036] Figure 4 An example of the segmentation effect of the ossified yellow ligament. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0038] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the common meanings understood by persons having ordinary skills in the field to which the present invention belongs.

[0039] The present invention provides a method for segmenting ossified yellow ligament. Figure 1 As shown, including:

[0040] S1, obtaining a three-dimensional image of a target area of ​​a patient, and segmenting the three-dimensional image to obtain a first bone area and a second bone area;

[0041] In the present invention, the yellow ligament of the human body is connected from the lower edge and inner surface of the upper vertebral lamina to the upper edge and outer edge of the lower vertebral lamina, and participates in forming the posterior wall and posterolateral wall of the cone canal. The yellow ligament undergoes heterotopic ossification under specific conditions or due to metabolic disorders, which can be seen in Figure 2 The yellow ligament exists in the cervical, thoracic and lumbar vertebrae. Theoretically, ossification of the yellow ligament can occur in any part of the spine.

[0042] In the present invention, the three-dimensional image of the target area of ​​the patient is exemplified as a CT cross-sectional image of a certain vertebral segment of a patient with ossification of the yellow ligament, such as Figure 3 As shown, the ossified yellow ligament ( Figure 3The spinous process, lamina, transverse process, articular process, etc. are located below the vertebral body in the CT cross-sectional image. They have similar voxel values ​​and all have bone characteristics. However, since the boundary of the ossified yellow ligament is not clear, it is difficult to directly segment and remove it from the plexus. Therefore, it is necessary to first segment the spinous process, lamina, transverse process, and articular process located below the vertebral body, and then screen out the ossified yellow ligament area. In the present invention, the definition Figure 3 The spinous process, lamina, and transverse process below the vertebral body are the first bone region, and the vertebral body region is the second bone region. Figure 3 shown.

[0043] In the present invention, since the anatomical features of the ossified yellow ligaments are different and do not have typical shape features, and the voxels in the ossified yellow ligament lesion area account for a very small proportion of the background of the entire image, and belong to a small target sample area, the accuracy of directly segmenting the ossified yellow ligaments by training samples of the ossified yellow ligaments is very low. The first bone region and the second bone region have typical shape features, so the segmentation model of these two regions is trained to segment the first bone region and the second bone region, and then the ossified yellow ligament can be segmented by the prior spatial information of the segmented first bone region and the second bone region.

[0044] In the present invention, the first bone region and the second bone region in the CT cross-sectional image of the patient with ossification of the yellow ligament can be segmented by a pre-trained segmentation model, and the segmentation model is obtained by pre-marking a number of three-dimensional images of the first bone region (the vertebral body does not include the ossified yellow ligament region) and the second bone region as training samples, and the specific segmentation model can be a pre-trained 3d_unet deep learning network model. In the present invention, the segmentation model can also use a pre-trained DeepLab V2 model or a Medical SAM 2 model.

[0045] S2, based on the first bone region and the second bone region obtained in S1, searching and obtaining an ossified yellow ligament region located between the first bone region and the second bone region according to the voxel characteristics of the ossified yellow ligament;

[0046] The voxel value of the ossified yellow ligament is very large, generally similar to the voxel value of the bone. Based on this, a Hu value threshold can be set. In this embodiment, the Hu value threshold is set between 100 and 800. Figure 1 , the ossified yellow ligament is shown in the CT cross-sectional image as an area above the first bone area. Therefore, the ossified yellow ligament area can be obtained by searching and extracting voxel points whose voxel values ​​are greater than the set Hu value threshold in the set space between the first bone area and the second bone area.

[0047] Specifically, the voxel points whose voxel values ​​are greater than the set Hu value threshold in the set space between the first bone region and the second bone region are searched and extracted, as follows:

[0048] Traverse the voxel points in the first bone region, and for any voxel point p i , establish the voxel point p i A three-dimensional search space of size w*h*l is taken as the center. Other voxel points are searched in the three-dimensional search space. If a voxel point does not belong to the first bone region or the second bone region, and the voxel value is greater than the set Hu value threshold, the voxel point is extracted and put into the seed point set. After traversing all the voxel points in the first bone region, the obtained seed point set is the rough ossified yellow ligament point set.

[0049] In the present invention, the size of the three-dimensional search space is determined as follows: in the human anatomical coordinate system, its height h is one-third of the calculated distance between the lowest point of the vertebra in the second bone region and the highest point of the first bone region (i.e., the height along the front-to-back direction of the human body), its width w is one-fifth of the width of the first bone region (i.e., the width along the left-right direction of the human body), and its length l is one-half of the depth of the vertebra in the second bone region (i.e., the length along the head-to-foot direction of the human body).

[0050] In the present invention, after the above-mentioned search and extraction process is completed and the corresponding voxel points are obtained, all voxel points constitute a seed point set. The seed point set is traversed. If a neighboring point of the 8 neighboring points of any voxel point satisfies the voxel value greater than the set Hu value threshold and is not in the first bone region and the second bone region, the neighboring point is also put into the seed point set. After traversing all the voxel points in the seed point set, no voxel points will be added to the seed point set. At this time, the target point set is obtained, that is, the more accurate ossified yellow ligament point set of the above steps is obtained.

[0051] In the present invention, after obtaining a more accurate set of ossified yellow ligament points, it is necessary to locate the candidate area of ​​the ossified yellow ligament, as follows: Calculate the boundary of the minimum circumscribed rectangle of the target point set obtained above, and its boundary is determined by obtaining the leftmost, rightmost, frontmost, backmost, topmost and bottommost points of the area corresponding to the target point set. Define the coordinates of the upper left corner of the minimum circumscribed rectangle in the image coordinate system as (x min ,y min， z min ), the coordinates of the lower right corner of the minimum bounding rectangle in the image coordinate system are (x max ,y max， z max ), then the width of the minimum enclosing rectangle is W = x max -x min , height is H = y max -y min, length D = z max -z min , thereby determining the candidate area of ​​ossified yellow ligament.

[0052] The candidate region of the ossified yellow ligament determined by the above method may include a small area of ​​vertebrae, so it is necessary to remove the small area of ​​vertebrae included therein. The present invention classifies the candidate region of the ossified yellow ligament into the original classification template most similar to it based on the Sato classification method, and then excludes the small area of ​​vertebrae included therein, thereby accurately locating and segmenting the final true ossified yellow ligament region.

[0053] Sato classification is divided into lateral type, extended type, enlarged type, fusion type and nodular type according to the different forms of ossification of the yellow ligament and the degree of compression on the spinal cord. These classifications reflect the morphological differences of most ossified yellow ligaments. Therefore, it is necessary to collect clinical images of ossified yellow ligaments corresponding to the above classifications as original classification templates (one classification in the present invention obtains one clinical image, and five classifications obtain a total of five clinical images. More clinical images corresponding to different classifications can also be collected according to actual needs). The ossified yellow ligament candidate region of the present invention has its specific scale and rotation, and the scale and rotation of each collected original classification template are also different. The ossified yellow ligament candidate region cannot be directly classified. Therefore, it is necessary to continuously transform the scale and rotation of each original classification template until the matching degree with the candidate region of the ossified yellow ligament is the highest, so that the ossified yellow ligament candidate region can be classified into the corresponding original classification template.

[0054] The new position coordinates of the voxel points in each original classification template image after each rotation and scale transformation are:

[0055]

[0056] Where: (x, y, z) is the position coordinate of the original voxel point in the image coordinate system, (x new ,y new ,z new ) is the position coordinate of the transformed voxel point in the image coordinate system;

[0057]

[0058] in:

[0059]

[0060] Among them, S is the scale transformation matrix, that is, the scale transformation step of the original voxel point, s x is the scale in the x direction, s y is the scale in the y direction, s zis the scaling factor in the z direction; R x is the rotation θ around the x-axis x The transformation matrix, R y is the rotation θ around the y axis y The transformation matrix, R z is the rotation θ around the z axis z The transformation matrix, R x , R y , R z Together they constitute the rotation transformation step size of the original voxel point; θ x The value range is 0-360°, θ y The value range is 0-360°, θ z The value range is 0-360°.

[0061] In the present invention, in order to ensure accuracy, a finer scale transformation step and rotation transformation step can also be set, that is, a scale transformation step is set, which can be applied to the x, y, and z directions respectively, and each application is a scale transformation; similarly, a rotation transformation step is set, which can be applied to rotations around the x, y, and z axes respectively, and each application is a rotation transformation.

[0062] Therefore, after calculating the new position coordinates of the voxel points of each transformed original typing template image, it may not be possible to directly obtain the voxel value. At this time, according to the voxel values ​​of the integer position points around it, the voxel value is obtained using the linear interpolation method, that is, the voxel value of the new voxel point of each original typing template image after transformation is obtained, and then the typing template image data after rotation and scale transformation can be calculated and used as the candidate typing template image. According to the scale transformation step size and rotation transformation step size set above, different candidate typing template images can be obtained.

[0063] Based on this, in the present invention, the candidate regions of the ossified yellow ligament are classified into the corresponding original typing templates, as follows:

[0064] (1) For any original typing template image, scale and rotation transformation are performed on it to obtain multiple candidate typing template images;

[0065] a. Performing a scale transformation on each original typing template image to generate a series of scaled original typing template images;

[0066] b. Performing rotation transformation on each original typing template image to generate a series of rotated original typing template images.

[0067] (2) For each candidate classification template image obtained after scale and rotation transformation, the candidate region of the ossified yellow ligament is traversed, and the matching degree between the candidate region and the ossified yellow ligament is detected. The largest matching coefficient is calculated and counted, and the candidate classification template is retained. Then, the candidate region of the ossified yellow ligament can be classified into the corresponding candidate classification template.

[0068] In the present invention, the matching coefficient adopts the normalized mutual correlation coefficient, which is as follows:

[0069]

[0070] Where:

[0071]

[0072] Where m is the voxel width of the candidate classification template image, n is the voxel height of the candidate classification template image, and d is the voxel depth of the candidate classification template image. m*n*d represents the window size of the candidate classification template image, f is the candidate region of the ossified yellow ligament, and t is the candidate classification template image. x+i, y+j, z+k are the positions of the center point of the candidate classification template image in the x, y, and z directions of the coordinate system when it slides on the image corresponding to the candidate region of the ossified yellow ligament. The value range of δ is [-1,1]. When δ is equal to -1, it means that it is completely irrelevant, and when δ is equal to 1, it means that it is completely relevant.

[0073] In the present invention, each original typing template image is scaled and rotated based on the set scale transformation step size and rotation transformation step size to obtain multiple candidate typing template images. Each candidate typing template image is matched with the candidate area of ​​the ossified yellow ligament to obtain a corresponding matching coefficient. All candidate typing template images are traversed to obtain the maximum matching coefficient, and a matching threshold is set. If the maximum matching coefficient obtained above is greater than the matching threshold, the maximum matching coefficient is valid, and the corresponding original typing template is classified into the corresponding candidate typing template. Based on the corresponding candidate typing template, a small amount of vertebral area contained in the candidate area of ​​the ossified yellow ligament is removed, and then the real ossified yellow ligament is accurately located and segmented, such as Figure 4 shown.

[0074] The present invention first segments the first bone region and the second bone region located below the vertebral body in the three-dimensional image, and then searches and extracts the region near the first bone region according to the voxel characteristics of the ossified yellow ligament, so as to obtain an approximate ossified yellow ligament region, and then classifies the ossified yellow ligament candidate region into the corresponding Sato classification according to the shape prior characteristics of the ossified yellow ligament Sato classification, thereby removing a small amount of vertebral regions in the aforementioned approximate ossified yellow ligament region, thereby accurately locating, identifying, and segmenting the ossified yellow ligament from numerous tissue regions of the human body, and effectively ensuring the accuracy and success rate of subsequent resection of the ossified yellow ligament.

[0075] The present invention also provides a method for planning a path for ossified yellow ligament resection based on the aforementioned ossified yellow ligament segmentation method, comprising the steps of:

[0076] (1) obtaining a preoperative three-dimensional image of the patient's target area and obtaining the ossified ligamentum flavum area according to the aforementioned ossified ligamentum flavum segmentation method;

[0077] (2) Mark the necessary areas and constrained areas in the resection path in the preoperative 3D images, use this as prior knowledge, and use the ant colony algorithm to plan the resection path;

[0078] The constraint area of ​​the set resection path is the area that cannot be passed through during the resection process, such as the spinal cord, nerve roots and other related areas.

[0079] In this embodiment, the necessary area of ​​the resection path is the area that must be passed during the resection process, such as the inferior articular process and lamina of the upper vertebra, the superior articular process and lamina of the lower vertebra, and the ossified yellow ligament area.

[0080] (3) obtaining a three-dimensional image of the patient's target area during surgery and registering the vertebrae therein with the vertebrae in the preoperative image;

[0081] Because the ossified yellow ligament resection path is planned in the preoperative three-dimensional image, it is necessary to transform the ossified yellow ligament resection path planned in the preoperative three-dimensional image to the three-dimensional image acquired during the operation. First, it is necessary to establish a transformation relationship between the preoperative three-dimensional image and the three-dimensional image acquired during the operation. Based on this, the vertebral segments in the preoperative three-dimensional image and the vertebral segments in the intraoperative three-dimensional image are aligned, and the two are unified into the same coordinate system. In this way, the ossified yellow ligament resection path planned in the preoperative three-dimensional image can be transformed to the three-dimensional image acquired during the operation for navigation.

[0082] In the present invention, the three-dimensional image of the target area of ​​the patient collected during the operation is a CBCT image.

[0083] Specifically, the registration of the vertebral surface point cloud in the three-dimensional image collected during the operation with the vertebral surface point cloud in the aforementioned preoperative image is as follows:

[0084] (31) using the aforementioned segmentation method for the first bone region and the second bone region to respectively segment and obtain surface point clouds of the vertebrae in the preoperative image and the three-dimensional image collected during the operation;

[0085] In the present invention, after obtaining the vertebrae in the preoperative image and the three-dimensional image collected during the operation by using the aforementioned segmentation method of the first bone region and the second bone region, the coordinates of the junction of the foreground and the background can be directly calculated as the surface point cloud position of the vertebrae.

[0086] (32) using the ICP algorithm to register the surface point clouds of the two groups of vertebrae obtained in step (31) to obtain an initial registration result T0;

[0087] (33) using a bounding box generation algorithm to generate bounding boxes of vertebral segments in the preoperative image respectively, and calculating the bounding boxes of vertebral segments in the three-dimensional image collected during the operation according to the initial registration result T0 obtained in step (32);

[0088] (34) The vertebral bounding boxes in the preoperative image obtained in step (33) and the three-dimensional image collected during the operation are input into a voxel-based gradient descent optimization algorithm for optimization to obtain the final registration result T1.

[0089] In the present invention, since the rough registration in step (32) has been performed, the preoperative image and the three-dimensional image collected during the operation have been roughly aligned, and the initial value of the optimization variable is set to T0. The optimization variable T0 is already near the global optimal solution, which ensures that no local optimal solution will be obtained in the subsequent optimization process. In addition, the initial registration result T0 obtained by the rough registration of the present invention can reduce the number of optimization iterations, thereby reducing the time consumption of the optimization algorithm.

[0090] In the present invention, in the voxel-based gradient descent optimization algorithm, the mutual correlation coefficient between the two images can be used to determine whether the two images are properly registered.

[0091] (4) According to the registration in step (3), the resection path of the ossified yellow ligament planned in the preoperative three-dimensional image is transformed into the three-dimensional image acquired during the operation.

[0092] In the present invention, in order to further facilitate practical application operations, after the ossified yellow ligament resection path planned in the preoperative three-dimensional image is transformed into the three-dimensional image collected during the operation, the ossified yellow ligament resection path planned in the preoperative three-dimensional image can also be transformed into the two-dimensional image displayed by the endoscope during the operation. This process can obtain the transformation relationship between the three-dimensional image collected during the operation and the two-dimensional image of the endoscope through camera calibration, and accordingly transform the ossified yellow ligament resection path planned in the preoperative three-dimensional image into the two-dimensional image of the endoscope.

[0093] In the present invention, camera calibration can be achieved by any means in the prior art, which will not be described in detail here.

[0094] The present invention extracts the ossified yellow ligament area based on the aforementioned ossified yellow ligament segmentation method. On this basis, by marking the necessary areas and constraint areas of the resection path as prior knowledge, an ant colony algorithm is used to plan the resection path. After that, the ossified yellow ligament resection path planned in the preoperative three-dimensional image is transformed into the three-dimensional image collected during the operation through registration. A safer and more effective ossified yellow ligament resection path can be planned, providing surgical guidance for doctors, improving the success rate of ossified yellow ligament resection, and thereby reducing surgical risks.

[0095] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0096] The embodiments of the present invention are intended to cover all such substitutions, modifications and variations that fall 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 the present invention should be included in the protection scope of the present invention.

Claims

1. A method for segmenting an ossified ligamentum flavum, characterized in that: include: Acquire a three-dimensional image of a target area of ​​a patient, and segment the image to obtain a first bone area and a second bone area; A Hu value threshold is set according to the voxel characteristics of the ossified yellow ligament, and voxel points with a Hu value greater than the Hu value threshold located between the first bone region and the second bone region are searched and extracted to obtain the ossified yellow ligament region.

2. The method for segmenting the ossified yellow ligament according to claim 1, characterized in that: The search extracts voxel points between the first bone region and the second bone region whose Hu values ​​are greater than a Hu value threshold, specifically: The voxel points in the first bone region are traversed, and a three-dimensional search space of a set size is established with any voxel point as the center. The voxel points in the three-dimensional search space whose Hu values ​​are greater than the set Hu value threshold and are not in the first bone region and the second bone region are extracted, and the voxel points are put into a seed point set. After the voxel points in the first bone region are traversed, the seed point set obtained is the ossified yellow ligament point set.

3. The method for segmenting the ossified yellow ligament according to claim 2, characterized in that: The seed point set is traversed. If the voxel value of any neighboring point of the 8 neighboring points of any voxel point is greater than the Hu value threshold and is not in the first bone area and the second bone area, the neighboring point is added to the seed point set. After traversing the voxel points of the seed point set, the target point set is obtained, which is the final ossified yellow ligament point set.

4. The method for segmenting the ossified yellow ligament according to claim 2, characterized in that: The three-dimensional search space of the set size is specifically: In the human anatomical coordinate system, the height of the three-dimensional search space is one-third of the calculated distance between the lowest point of the vertebra in the second bone region and the highest point of the first bone region, its width is one-fifth of the width of the first bone region, and its length is one-half of the depth of the vertebra in the second bone region.

5. The method for segmenting the ossified ligamentum flavum according to claim 1, characterized in that: The ossified yellow ligament area is obtained as follows: Calculate the boundary of the minimum circumscribed rectangle of the seed point set composed of the extracted points, and determine the candidate area of ​​the ossified yellow ligament based on it; Based on the Sato classification method, the candidate regions of the ossified yellow ligament are classified into the original classification template that is most similar to it. The vertebral regions in the candidate regions of the ossified yellow ligament are excluded accordingly, and finally the ossified yellow ligament regions are obtained.

6. The method for segmenting the ossified ligamentum flavum according to claim 5, characterized in that: The Sato classification method is used to classify the candidate regions of the ossified yellow ligament into the original classification template that is most similar to it, as follows: (1) For any original typing template image, scale and rotation transformation are performed on it to obtain multiple candidate typing template images; a. Performing a scale transformation on each original typing template image to generate a series of scaled original typing template images; b. Performing rotation transformation on each original typing template image to generate a series of rotated original typing template images. (2) For each candidate classification template image obtained after scale and rotation transformation, a matching degree test is performed between it and the candidate region of the ossified yellow ligament, and the maximum matching coefficient is calculated. The candidate classification template is retained, that is, the candidate region of the ossified yellow ligament is classified into the corresponding candidate classification template.

7. The method for segmenting the ossified ligamentum flavum according to claim 1, characterized in that: The segmentation obtains a first bone region and a second bone region in the target region of the patient, as follows: The spinous process, lamina, and transverse process below the vertebral body in the vertebral segment are defined as the first bone region, and the vertebral body region is defined as the second bone region; The three-dimensional image of the target area of ​​the patient is segmented using a pre-trained segmentation model to obtain a first bone area and a second bone area; The segmentation model is obtained by training a plurality of three-dimensional images of the first bone region and the second bone region which are pre-marked as training samples.

8. A method for planning a path for resection of an ossified yellow ligament, characterized in that: Includes steps: (1) obtaining a preoperative three-dimensional image of a target area of ​​a patient, and segmenting the ossified ligamentum flavum area according to the ossified ligamentum flavum segmentation method described in any one of claims 1 to 7; (2) marking the necessary areas and constraint areas of the resection path in the preoperative three-dimensional image, using this as prior knowledge and adopting an ant colony algorithm to plan the resection path; (3) obtaining a three-dimensional image of the target area of ​​the patient during the operation, and registering the surface point cloud of the vertebral segment therein with the surface point cloud of the vertebral segment in the preoperative three-dimensional image; (4) Based on the registration in step (3), the ossified yellow ligament resection path marked in the preoperative three-dimensional image is transformed into the three-dimensional image acquired during the operation.

Images