Method and device for determining a thread of a medical screw, computer device and storage medium

Abstract

The application relates to a nail path determination method and device of a medical screw, a computer device and a storage medium. The method comprises the following steps: acquiring a candidate nail path of a medical screw in a target model of a target part; the target model is obtained according to a medical image of the target part; determining the distance between the candidate nail path and the surface of the target model and the image parameters of the candidate nail path in the medical image; the image parameters are used for characterizing the bone density of the target part; and determining the target nail path of the medical screw from the candidate nail path according to the distance and the image parameters. By using the method, the nail path with good safety and stability can be automatically determined, and the problem of time and labor consumption caused by manual marking is solved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer device, and storage medium for determining the path of a medical screw. Background Technology

[0002] With the development of medical technology, pedicle screw fixation has emerged. This technique involves inserting medical screws through the pedicle into the vertebral body of the spine, providing strong and reliable fixation for the spine. It has a wide range of applications in the treatment of spinal diseases.

[0003] Pedicle screw fixation usually requires preoperative planning. Doctors manually mark the screw paths based on the patient's CT (Computed Tomography) images. During the operation, the screws are placed according to the pre-planned paths. The manual marking process is time-consuming and labor-intensive.

[0004] Therefore, current methods for determining the path of medical screws are time-consuming and labor-intensive. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for determining the path of medical screws in a time-saving and labor-saving manner to address the above-mentioned technical problems.

[0006] Firstly, this application provides a method for determining the path of a medical screw. The method includes:

[0007] Candidate screw tracks of medical screws are obtained in a target model at a target site; the target model is obtained based on medical images of the target site.

[0008] The distance between the candidate pin path and the surface of the target model is determined, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0009] Based on the distance and the image parameters, the target path of the medical screw is determined from the candidate paths.

[0010] In one embodiment, obtaining candidate screw tracks of the medical screw in the target model at the target site includes:

[0011] Determine the initial path of the medical screw in the target model;

[0012] Based on the directed distance map of the target location, the initial spike track is iteratively updated to obtain the updated spike track;

[0013] The initial nail path and the updated nail path are used as candidate nail paths for the medical screw in the target model.

[0014] In one embodiment, determining the initial path of the medical screw in the target model includes:

[0015] Based on the surface of the target model, the target point cloud of the target region is obtained;

[0016] Determine the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target region.

[0017] Based on the registration relationship, the standard pin track corresponding to the standard model is subjected to coordinate transformation to obtain the initial pin track.

[0018] In one embodiment, the step of iteratively updating the initial spike track based on the directed distance map of the target location to obtain the updated spike track includes:

[0019] Determine the gradient value of the initial spike in the directed distance map;

[0020] The initial spike is updated based on the gradient value to obtain the updated spike. The updated spike is then used as the initial spike, and the process returns to the step of determining the gradient value of the initial spike in the directed distance map, until the number of candidate spikes meets the preset number.

[0021] In one embodiment, updating the initial spike line based on the gradient value to obtain the updated spike line includes:

[0022] Based on the gradient value, the initial set of sampling points of the spike path is updated to obtain the updated set of sampling points;

[0023] Determine the fitted straight line corresponding to the updated set of sampling points;

[0024] The entry and exit points of the updated nail track are obtained based on the intersection of the fitted straight line and the target model.

[0025] Connect the nail entry point and the nail exit point to obtain the updated nail path.

[0026] In one embodiment, determining the distance between the candidate spike and the surface of the target model includes:

[0027] At least one target sampling point is determined from the candidate spike path;

[0028] Determine the minimum distance between each target sampling point and the target point cloud of the target location;

[0029] The minimum value of at least one of the minimum distances is determined as the distance between the candidate spike and the target model.

[0030] In one embodiment, determining the target kerf path of the medical screw from the candidate kerf paths based on the distance and the image parameters includes:

[0031] The suitability of each candidate spike is determined based on the distance and the image parameters.

[0032] The maximum applicability is determined from the applicability, and the candidate nail path corresponding to the maximum applicability is determined as the target nail path.

[0033] Secondly, this application also provides a device for determining the path of a medical screw. The device includes:

[0034] The candidate screw path module is used to acquire candidate screw paths of medical screws in a target model at the target site; the target model is obtained based on medical images of the target site.

[0035] A parameter determination module is used to determine the distance between the candidate pin path and the surface of the target model, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0036] The pin path determination module is used to determine the target pin path of the medical screw from the candidate pin paths based on the distance and the image parameters.

[0037] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0038] Candidate screw tracks of medical screws are obtained in a target model at a target site; the target model is obtained based on medical images of the target site.

[0039] The distance between the candidate pin path and the surface of the target model is determined, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0040] Based on the distance and the image parameters, the target path of the medical screw is determined from the candidate paths.

[0041] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0042] Candidate screw tracks of medical screws are obtained in a target model at a target site; the target model is obtained based on medical images of the target site.

[0043] The distance between the candidate pin path and the surface of the target model is determined, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0044] Based on the distance and the image parameters, the target path of the medical screw is determined from the candidate paths.

[0045] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0046] Candidate screw tracks of medical screws are obtained in a target model at a target site; the target model is obtained based on medical images of the target site.

[0047] The distance between the candidate pin path and the surface of the target model is determined, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0048] Based on the distance and the image parameters, the target path of the medical screw is determined from the candidate paths.

[0049] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for determining the path of a medical screw acquires candidate paths of the medical screw in a target model at the target site, determines the distance between the candidate paths and the surface of the target model, and the image parameters of the candidate paths in medical images. Based on the distance and image parameters, the target path of the medical screw is determined from the candidate paths. Since the greater the distance from the path to the bone surface, the safer the screw placement, and the greater the bone density on the path, the stronger the stability of the screw implantation, the path with good safety and stability can be automatically determined based on the distance from the path to the bone surface and the image parameters reflecting the bone density on the path, solving the problem of time-consuming and labor-intensive manual annotation. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating a method for determining the path of a medical screw in one embodiment;

[0051] Figure 2This is a schematic diagram illustrating the segmentation of a target area from a medical image in one embodiment;

[0052] Figure 3 This is a schematic diagram illustrating the registration of the target model and the standard model in one embodiment;

[0053] Figure 4 This is a schematic diagram of the initial nail path in one embodiment;

[0054] Figure 5 This is a schematic diagram of a directed distance map in one embodiment;

[0055] Figure 6 This is a schematic diagram of the target spike track in one embodiment;

[0056] Figure 7 This is a flowchart illustrating a method for determining the path of a medical screw in another embodiment;

[0057] Figure 8 This is a flowchart illustrating a method for determining the path of a medical screw in another embodiment;

[0058] Figure 9 This is a structural block diagram of a device for determining the path of a medical screw in one embodiment;

[0059] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0061] In one embodiment, such as Figure 1 As shown, a method for determining the path of a medical screw is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0062] Step S110: Obtain candidate screw paths of the medical screw in the target model of the target site; the target model is obtained based on the medical image of the target site.

[0063] The target site can be any part of the human body where the medical screw is to be implanted, such as the lumbar spine.

[0064] The target model can be a three-dimensional model of the target part.

[0065] Among them, medical images can be, but are not limited to, CT images or MR (Magnetic Resonance) images.

[0066] In practice, medical images of the target area can be acquired, the medical images can be segmented to obtain a target model of the target area, and candidate screw paths for medical screws can be determined in the target model.

[0067] In practical applications, a standard model of the target area and its pin placement rules can be pre-obtained. The pin placement rules include the insertion and exit points of the medical screw on the standard model. Medical images containing the target area can also be acquired, and the images can be segmented using deep learning methods to obtain a target model of the target area. After registering the target model with the standard model, the pin placement rules of the standard model are applied to the target model to obtain the insertion and exit points of the medical screw on the target model. The line connecting the insertion and exit points is used as the initial pin path of the target model. Based on the directed distance map corresponding to the medical image, the initial pin path is iteratively updated to obtain multiple updated pin paths. The initial pin path and the multiple updated pin paths are used as candidate pin paths for the medical screw in the target model.

[0068] Step S120: Determine the distance between the candidate pin path and the surface of the target model, as well as the imaging parameters of the candidate pin path in medical imaging; the imaging parameters are used to characterize the bone density of the target site.

[0069] Among them, the image parameters can be the average value of the image values ​​corresponding to the candidate spikes.

[0070] In practice, sampling points can be determined on the candidate pin path, and the minimum distance from each sampling point to the surface of the target model can be determined. The minimum value is then used as the distance between the candidate pin path and the surface of the target model. Alternatively, the image values ​​of each sampling point on the candidate pin path in the medical image can be obtained, and the average value of the image values ​​can be used as the image parameter of the candidate pin path in the medical image.

[0071] It should be noted that when the medical image is a CT image, the image parameters are used to characterize the degree of X-ray absorption by the tissue at the target site along the path of the candidate pin; when the medical image is an MR image, the image parameters are used to characterize the relaxation time of the tissue at the target site along the path of the candidate pin; when the medical image is another type of image, the image parameters can be selected according to the parameters of the actual image type to characterize the bone mineral density at the target site along the path of the candidate pin.

[0072] For example, for each candidate pin path, we can avoid the vicinity of the entry and exit points and extract several sampling points in the middle of the candidate pin path. We can calculate the minimum distance from each sampling point to the surface of the target model and take the minimum value of the minimum distance as the distance between the candidate pin path and the surface of the target model. We can also obtain the CT value of each sampling point on the candidate pin path in the CT image and take the average value of the CT value as the image parameter of the candidate pin path.

[0073] Step S130: Based on distance and image parameters, determine the target path of the medical screw from the candidate paths.

[0074] The target spike track can be the final determined spike track.

[0075] In practice, the applicability of each candidate screw path can be determined based on preset rules, distance and image parameters. The applicability can reflect the safety and stability of the candidate screw path. The target screw path of the medical screw can be determined from the candidate screw paths based on the applicability.

[0076] In practical applications, the formula for calculating the applicability v can be:

[0077] v = D * log(E),

[0078] Where D represents the distance between the candidate screw path and the target model surface, and E represents the image parameters of the candidate screw path in medical imaging, i.e., the average image value of each sampling point on the candidate screw path. The suitability of each candidate screw path is calculated, and the candidate screw path with the highest suitability is determined as the target screw path for the medical screw.

[0079] The above-mentioned method for determining the path of a medical screw involves acquiring candidate paths of the medical screw in a target model at the target site, determining the distance between the candidate paths and the surface of the target model, and the image parameters of the candidate paths in medical imaging. Based on the distance and image parameters, the target path of the medical screw is determined from the candidate paths. Since the greater the distance from the path to the bone surface, the safer the screw placement, and the greater the bone density on the path, the stronger the stability of the screw implantation, the method can automatically determine a path with good safety and stability based on the distance from the path to the bone surface and the image parameters reflecting the bone density on the path, thus solving the problem of time-consuming and labor-intensive manual annotation.

[0080] In one embodiment, step S110 may specifically include: determining the initial path of the medical screw in the target model; iteratively updating the initial path based on the directed distance map of the target area to obtain the updated path; and using the initial path and the updated path as candidate paths of the medical screw in the target model.

[0081] The initial spike path can be a candidate spike path that has been initially determined.

[0082] In practice, a standard model of the target area and its pin placement rules can be pre-obtained. The medical image containing the target area is segmented to obtain a target model of the target area. After registering the target model and the standard model, the pin placement rules of the standard model are applied to the target model to obtain the initial pin path of the medical screw in the target model. The directed distance map corresponding to the medical image is determined. The initial pin path is updated according to the gradient value of the initial pin path in the directed distance map to obtain the updated pin path. The number of initial pin paths and updated pin paths is counted. If the number does not meet the preset number, the updated pin path is used as the initial pin path. The process returns to the step of updating the initial pin path according to the gradient value of the initial pin path in the directed distance map to obtain the updated pin path. This process continues until the number of initial pin paths and updated pin paths meets the preset number. At this point, the initial pin path and all updated pin paths are used as candidate pin paths of the medical screw in the target model.

[0083] Figure 2 This provides a schematic diagram of segmenting a target area from a medical image. Based on... Figure 2 Based on deep learning methods, CT images of the human lumbar spine can be segmented to obtain a three-dimensional model of the third lumbar vertebra, which can then be used as the target model.

[0084] Figure 3 A schematic diagram illustrating the registration of the target model and the standard model is provided. According to... Figure 3 You can select a standard model of the third lumbar vertebra from the template library and obtain the screw insertion and exit points of the standard model. Register the standard model with the target model, and perform coordinate transformation on the screw insertion and exit points of the standard model according to the registration relationship between the two to obtain the screw insertion and exit points of the target model.

[0085] Figure 4 A schematic diagram of an initial nail path is provided. According to... Figure 4 The standard model can be set with two nail tracks, left and right, corresponding to two nail entry points and two nail exit points. Based on the two nail tracks of the standard model, the following can be determined: Figure 4 The white line segments indicate the two spikes on the target model.

[0086] In this embodiment, the initial path of the medical screw in the target model is determined; the initial path is iteratively updated according to the directed distance map of the target area to obtain the updated path; the initial path and the updated path are used as candidate paths of the medical screw in the target model, which can automatically determine multiple candidate paths of the target model and improve the efficiency of candidate path determination.

[0087] In one embodiment, the step of determining the initial nail path of the medical screw in the target model may specifically include: obtaining the target point cloud of the target part based on the surface of the target model; determining the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target part; and performing coordinate transformation processing on the standard nail path corresponding to the standard model according to the registration relationship to obtain the initial nail path.

[0088] The target point cloud can be the point cloud on the surface of the target model. The standard point cloud can be the point cloud on the surface of the standard model.

[0089] The registration relationship can be the coordinate transformation relationship between the target point cloud and the standard point cloud.

[0090] Among them, the standard screw track can be the screw track of the standard model for medical screw implantation.

[0091] In practice, a mesh can be constructed on the surface of the target model, and the intersection of the mesh can be used as the target point cloud of the target part. Alternatively, a mesh can be constructed on the surface of the standard model, and the intersection of the mesh can be used as the standard point cloud of the target part. The target point cloud and the standard point cloud are registered to obtain the registration relationship. Based on the registration relationship, the standard nail track of the medical screw implantation standard model is transformed to obtain the initial nail track of the medical screw in the target model.

[0092] In practical applications, the standard model and the target model can be constructed separately, such as Figure 3 The grid shown is used to obtain standard point clouds and target point clouds based on the grid intersections. The number of standard and target point clouds can be reduced by methods such as grid aggregation. The CPD (Coherent Point Drift) point cloud registration method is used to determine the coordinate transformation relationship between the standard and target point clouds. Based on the coordinate transformation relationship, the infeed and outfeed points of the standard model are transformed to obtain the infeed and outfeed points of the target model. The infeed and outfeed points of the target model are connected to obtain the initial nail path of the medical screw in the target model.

[0093] In this embodiment, the target point cloud of the target part is obtained based on the surface of the target model; the registration relationship between the target point cloud and the standard point cloud is determined; the standard point cloud is obtained based on the standard model of the target part; according to the registration relationship, the standard pin track corresponding to the standard model is subjected to coordinate transformation processing to obtain the initial pin track. By performing coordinate transformation on the standard pin track of the standard model, the initial pin track of the target model can be obtained quickly, further improving the efficiency of candidate pin track determination.

[0094] In one embodiment, the step of iteratively updating the initial spike track based on the directed distance map of the target location to obtain the updated spike track may specifically include: determining the gradient value of the initial spike track in the directed distance map; updating the initial spike track based on the gradient value to obtain the updated spike track; using the updated spike track as the initial spike track; and returning to the step of determining the gradient value of the initial spike track in the directed distance map until the number of candidate spike tracks meets the preset number.

[0095] Among them, the directed distance map can be a graph composed of the shortest directed distances from each point on the medical image to the edge of the target model.

[0096] In the specific implementation, a directed distance map corresponding to the medical image can be determined based on the target model. Several sampling points are determined on the initial pin track, and the gradient value of each sampling point in the directed distance map is determined. The initial pin track is updated according to the gradient value to obtain the updated pin track. The number of initial pin tracks and updated pin tracks is counted to obtain the number of candidate pin tracks. If the number of candidate pin tracks meets the preset number, the determination of candidate pin tracks is completed. Otherwise, if the number of candidate pin tracks is less than the preset number, the updated pin track is used as the initial pin track, and the process returns to the steps of determining several sampling points on the initial pin track, determining the gradient value of each sampling point in the directed distance map, and updating the initial pin track according to the gradient value to obtain the updated pin track, until the number of candidate pin tracks meets the preset number.

[0097] In this embodiment, the gradient value of the initial spike in the directed distance map is determined; the initial spike is updated according to the gradient value to obtain the updated spike; the updated spike is used as the initial spike, and the process returns to the step of determining the gradient value of the initial spike in the directed distance map until the number of candidate spikes meets the preset number. The spikes can be iteratively updated according to the gradient value of the spikes in the directed distance map, which speeds up the spike determination and reduces the spike determination time.

[0098] In one embodiment, the step of updating the initial spike track based on the gradient value to obtain the updated spike track may specifically include: updating the sampling point set of the initial spike track based on the gradient value to obtain the updated sampling point set; determining the fitted line corresponding to the updated sampling point set; obtaining the entry point and exit point of the updated spike track based on the intersection point between the fitted line and the target model; and connecting the entry point and exit point to obtain the updated spike track.

[0099] In the specific implementation, several sampling points can be determined on the initial spike path to obtain the sampling point set of the initial spike path. The gradient value of each sampling point in the sampling point set on the directed distance map is determined. Each sampling point is updated according to the gradient value to obtain the updated sampling points. All updated sampling points form the updated sampling point set. The fitted line corresponding to the updated sampling point set is determined. The intersection point between the fitted line and the target model is determined as the entry point and exit point. The line connecting the entry point and exit point is used as the updated spike path.

[0100] Figure 5 A schematic diagram of a directed distance map is provided. According to... Figure 5 First, the spatial resolution of the CT image and the target model can be normalized. Then, the shortest directed distance from each pixel in the CT image to the edge of the target model can be calculated to obtain a directed distance map (SDM). In the directed distance map, pixels outside the target model can be marked as negative values, pixels inside the target model as positive values, and pixels at the edge of the target model as zero. Alternatively, pixels outside the target model can be marked as positive values, pixels inside the target model as negative values, and pixels at the edge of the target model as zero. For the initial pin path, avoiding the vicinity of the pin entry and exit points, several sampling points are extracted in the middle of the initial pin path, and a sampling point update formula is set.

[0101] p U =p+s*f,

[0102] Where p represents the initial sampling point on the spike track, f represents the gradient value of p in the directed distance map, and s represents the update step size. U Let p represent the updated sampling point. For each sampling point p on the initial spike path, determine its gradient value f in the directed distance map, and obtain the corresponding updated sampling point p according to the sampling point update formula. U Find a straight line that can fit all the updated sampled points, and determine the two intersection points of this line with the target model as the entry point and exit point, respectively. The line connecting the entry point and exit point is taken as the updated nail path.

[0103] In this embodiment, the initial sampling point set of the spike path is updated based on the gradient value to obtain the updated sampling point set; the fitted line corresponding to the updated sampling point set is determined; the entry and exit points of the updated spike path are obtained based on the intersection of the fitted line and the target model; the entry and exit points are connected to obtain the updated spike path. The spike path can be iteratively updated based on the gradient value of the spike path in the directed distance map, which speeds up the spike path determination and reduces the spike path determination time.

[0104] In one embodiment, step S120 may specifically include: determining at least one target sampling point from the candidate spike path; determining the minimum distance between each target sampling point and the target point cloud of the target location; and determining the minimum value of the at least one minimum distance as the distance between the candidate spike path and the target model.

[0105] The target sampling point can be a sampling point on the candidate spike track.

[0106] In practice, a mesh can be constructed on the surface of the target model, and the target point cloud of the target part can be obtained based on the intersection of the mesh. At least one target sampling point can be determined on the candidate spike path, and the minimum distance from each target sampling point to the target point cloud can be determined. The minimum value is selected from the minimum distances as the distance between the candidate spike path and the target model.

[0107] For example, if 90 target sampling points are identified on the candidate spike path, and the minimum distances to the target point cloud are 0.5, 0.8, 0.9, 0.1, ..., 0.4 respectively, and the minimum value among the minimum distances is selected, we get 0.1. Then the distance between the candidate spike path and the target model is 0.1.

[0108] In this embodiment, by determining at least one target sampling point from the candidate pin path; determining the minimum distance between each target sampling point and the target point cloud of the target part; and determining the minimum value among the at least one minimum distance as the distance between the candidate pin path and the target model, the shortest distance between the candidate pin path and the edge of the target model can be determined, which facilitates the selection of pin paths far from the edge of the target part and ensures the safety of pin placement.

[0109] In one embodiment, step S130 may specifically include: determining the applicability of each candidate spike path based on distance and image parameters; the applicability is used to characterize the safety and stability of the candidate spike path; determining the maximum applicability from the applicability, and determining the candidate spike path corresponding to the maximum applicability as the target spike path.

[0110] In practice, the mapping relationship between applicability and distance and image parameters can be preset. After determining the distance between the candidate pin path and the target model surface, as well as the image parameters of the candidate pin path in the medical image, the applicability of the candidate pin path can be determined according to the mapping relationship. In this way, the applicability of each candidate pin path is obtained, and the candidate pin path with the highest applicability is determined as the target pin path.

[0111] In practical applications, the formula for calculating the applicability v can be:

[0112] v = D * log(E),

[0113] Where D represents the distance between the candidate screw path and the target model surface, and E represents the image parameters of the candidate screw path in medical imaging, i.e., the average image value of each sampling point on the candidate screw path. The suitability of each candidate screw path is calculated, and the candidate screw path with the highest suitability is determined as the target screw path for the medical screw.

[0114] Figure 6 A schematic diagram of the target spike track is provided. According to... Figure 6 Corresponding to the two spikes on the left and right sides set in the standard model, it can be determined that... Figure 6 The two initial spike tracks shown by the white line segments, and as shown in the image. Figure 6 The two target spikes are shown by the black line segment in the middle.

[0115] In this embodiment, the suitability of each candidate screw path is determined based on distance and imaging parameters; the maximum suitability is determined from the suitability, and the candidate screw path corresponding to the maximum suitability is determined as the target screw path. This can identify the target screw path that is far from the bone surface and passes through a bone density with high density, thus ensuring the safety and stability of medical screw implantation.

[0116] To facilitate a deeper understanding of the embodiments of this application by those skilled in the art, a specific example will be used for illustration below.

[0117] The automatic screw path positioning process for preoperative planning in lumbar pedicle screw surgery mainly includes the following steps:

[0118] Step 1: Based on 3D organ segmentation technology, a deep learning-based 3D vertebral segmentation model is used to quickly and accurately extract the target vertebra (target location), for example, the third lumbar vertebra, from 3D lumbar spine CT images (medical images), resulting in... Figure 2 The lumbar spine segmentation results shown (target model).

[0119] Step 2: Select a standard vertebral body (standard model) corresponding to the third lumbar vertebra from the template library, along with the screw insertion points and endpoints (entry and exit points of the standard screw track) as the baseline template. Apply the screw placement rules of the baseline template to the target model to predict the screw insertion points and endpoints (entry and exit points of the initial screw track) of the left and right screw tracks when placing screws on the target model. Figure 3 As shown.

[0120] The specific process for predicting the initial screw insertion point and screw endpoint of the target model based on the baseline template is as follows:

[0121] Step 2.1: Obtain the point clouds of the target cone in the target model and the standard model, and obtain the target point cloud and the standard point cloud respectively. Specifically, the target point cloud can be obtained based on the mask of the target cone in the lumbar spine segmentation result, and the standard point cloud can be obtained in the same way.

[0122] Step 2.2: Perform mesh reconstruction based on the obtained point cloud, and further perform mesh remesh operation on the mesh reconstruction result. For example, perform mesh reconstruction by mesh aggregation to reduce the size of the point cloud and obtain a smaller point cloud, for example, by an order of magnitude compared to before mesh reconstruction.

[0123] Step 2.3: Use the CPD point cloud registration method to learn the registration rules (registration relationship) from the standard point cloud of the reference template to the target point cloud of the target model;

[0124] Step 2.4: Obtain the point cloud of the screw in the reference template (the entry and exit points of the standard screw track). Using the registration rules learned in Step 2.3, convert the screw insertion point and end point of the standard screw track in the reference template into the insertion point and end point of the left and right screw initial tracks of the target model.

[0125] Step 3, automatic navigation of the nail track. For example... Figure 4 and Figure 6 As shown, the white line segment represents the nail path obtained from the initial insertion point and endpoint of the left and right screw paths of the target model, while the black line segment represents the optimal nail path found by automatic navigation.

[0126] Step 3.1: First, the spatial resolution of the target model and its mask is normalized. Then, a directed distance map corresponding to the 3D lumbar spine CT image is calculated based on the mask. Since the mask is 3D, the directed distance map is also 3D. Here, "distance map" refers to calculating the distance of each pixel in the 3D lumbar spine CT image from the mask boundary. "Directed" indicates whether a point in space is inside or outside the mask, using positive and negative signs respectively, and the sign setting is not unique. In this example, if a point in space is inside the mask, its distance from the mask boundary is positive; if it is outside the mask, its distance from the mask boundary is negative. The calculation result of the directed distance map is as follows: Figure 5 As shown;

[0127] Step 3.2: Calculate the gradient at each pixel in the directed distance map, where the gradient is also three-dimensional;

[0128] Step 3.3: Update the pixels of the initial nail path in the target model using the obtained gradient. The specific operation is as follows:

[0129] The iterative optimization of the left and right spikes uses the same method. Taking the left spike as an example, we first define the optimization iteration step i = 0, which corresponds to the initial spike of the target model.

[0130] Step 3.3.1: For the spike path in step i, sequentially sample m points along the spike path (the line connecting the entry and exit points), for example, m = 100, to obtain the sequential sampling point p. i , i = 1, ..., 100;

[0131] Step 3.3.2: Extract the middle 90% of points from m sequential sampling points to construct a measurement point (target sampling point) set, that is, from the 5th point to the 95th point starting from the starting point. Use the middle 90 points of 100 points to construct the measurement point set, calculate the distance D of the measurement point set from the target point cloud, and the average value E (image parameter) of the CT value traversed by each point in the measurement point set.

[0132] Specifically, for distance D, the minimum distance from each point in the metric point set to the target point cloud can be calculated to obtain the minimum distance set. Then, the minimum distance in the minimum distance set can be selected as the distance D from the metric point set to the target cone point cloud.

[0133] The evaluation value (suitability) of the current pin track is set as v = D * log(E). The evaluation value of each pin track is calculated, and the pin track with the highest evaluation value is selected as the approximately optimal pin track. The physical meaning of the evaluation value is that the larger the distance D and the larger the average CT value E along the path, the larger the evaluation value. This indicates that the found approximately optimal pin track is located inside the vertebral body, far from the cortical bone, and passes through areas with high bone density. The logarithmic function is used for E because bone tissue CT values ​​are high, and the calculation of the evaluation value mainly focuses on distance; therefore, the logarithmic function is used to reduce the influence of CT values.

[0134] Step 3.3.3, for each sequential sampling point p i The corresponding gradient f can be found in the gradient plot of the directed distance map;

[0135] Step 3.3.4, update each sequential sampling point, p i U =p i +s*f, where s is the update step size, which can be predefined as s = 0.5;

[0136] Step 3.3.5: Update all m sequential sampling points on the spike path to obtain m new sampling points (updated sampling points). Then, a fitted straight line L can be found. c Fit the m new sampling points;

[0137] Step 3.3.6, Fit the straight line L c It can intersect with the target model to obtain the screw implantation point and the endpoint respectively. The line connecting the screw implantation point and the endpoint is used to construct a new nail path (updated nail path), which is the nail path in step i+1.

[0138] Repeat steps 3.3.1-3.3.6 above up to n times, finding the spike with the highest evaluation value. This spike is taken as the approximate optimal spike. For example, n=200. In the experiment, the highest evaluation value was obtained near step 100. Figure 6 The black spike track in the image is taken as the approximate optimal spike track.

[0139] The aforementioned automatic screw track positioning method for preoperative planning of lumbar pedicle screw surgery can achieve fully automatic screw track navigation through point cloud registration-based initial screw track positioning and near-optimal screw track automatic navigation, without manual intervention. Compared with traditional methods, it does not need to traverse all candidate trajectories to find the near-optimal screw track, and is faster.

[0140] Moreover, the optimized positioning of the screw path takes into account both the fact that the farther away from the outer edge of the vertebral body is, the better (that is, the farther away the same size screw is implanted along the trajectory, the farther away from the bone cortex, which means that the screw implantation is safer), and also ensures that the CT value of the trajectory is as high as possible (that is, the higher the CT value, the higher the bone density, and the stronger the stability after screw implantation), and also ensures the safety and stability of medical screw implantation.

[0141] In one embodiment, such as Figure 7 As shown, a method for determining the path of a medical screw is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0142] Step S201: Obtain the target point cloud of the target part based on the surface of the target model;

[0143] Step S202: Determine the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target part.

[0144] Step S203: Based on the registration relationship, perform coordinate transformation on the standard pin track corresponding to the standard model to obtain the initial pin track;

[0145] Step S204: Determine the gradient value of the initial spike track in the directed distance map;

[0146] Step S205: Update the initial nail track according to the gradient value to obtain the updated nail track;

[0147] Step S206: Use the updated pin track as the initial pin track, and return to step S204 until the number of candidate pin tracks meets the preset number; the candidate pin tracks include the initial pin track and the updated pin track.

[0148] Step S207: Determine the distance between each candidate pin track and the surface of the target model, as well as the imaging parameters of each candidate pin track in medical imaging; the imaging parameters are used to characterize the bone density of the target site.

[0149] Step S208: Determine the suitability of each candidate spike track based on distance and image parameters; suitability is used to characterize the safety and stability of the candidate spike track.

[0150] Step S209: Determine the maximum applicability from the applicability and determine the candidate pin path corresponding to the maximum applicability as the target pin path.

[0151] In the specific implementation, the target point cloud of the target part can be obtained based on the surface mesh of the target model. The target point cloud is then registered with the standard point cloud of the standard model to obtain the registration relationship. Based on the registration relationship, the standard pin track of the standard model is subjected to coordinate transformation to obtain the initial pin track. The gradient value of the initial pin track in the directed distance map of the target part is determined. The initial pin track is iteratively updated based on the gradient value to obtain candidate pin tracks. The distance between each candidate pin track and the surface of the target model, as well as the average value of each candidate pin track in the medical image, are determined. The applicability of each candidate pin track is determined based on the distance and the average value. The candidate pin track with the maximum applicability is determined as the target pin track.

[0152] Since the terminal processing procedure has been described in detail in the foregoing embodiments, it will not be repeated here.

[0153] The above-mentioned method for determining the path of medical screws is effective because the greater the distance from the path to the bone surface, the safer the screw placement; the greater the bone density on the path, the stronger the stability of the screw implantation. Based on the distance from the path to the bone surface and the imaging parameters reflecting the bone density on the path, a path with good safety and stability can be automatically determined, solving the problem of time-consuming and labor-intensive manual marking.

[0154] In one embodiment, such as Figure 8 As shown, a method for determining the path of a medical screw is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0155] Step S301: Obtain the target point cloud of the target part based on the surface of the target model;

[0156] Step S302: Determine the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target part.

[0157] Step S303: Based on the registration relationship, perform coordinate transformation on the standard pin track corresponding to the standard model to obtain the initial pin track;

[0158] Step S304: Determine the distance between the initial pin track and the surface of the target model, as well as the imaging parameters of the initial pin track in medical imaging; the imaging parameters are used to characterize the bone density of the target site.

[0159] Step S305: Determine the suitability of the initial spike track based on distance and image parameters; suitability is used to characterize the safety and stability of the initial spike track.

[0160] Step S306: Determine the gradient value of the initial spike track in the directed distance map;

[0161] Step S307: Update the initial nail track according to the gradient value to obtain the updated nail track;

[0162] Step S308: Use the updated pin track as the initial pin track, and return to step S304 until the number of candidate pin tracks meets the preset number; the candidate pin tracks include the initial pin track and the updated pin track.

[0163] Step S309: Determine the maximum applicability from the applicability and determine the candidate pin path corresponding to the maximum applicability as the target pin path.

[0164] In practice, the target point cloud of the target area can be obtained from the surface mesh of the target model. The target point cloud is then registered with the standard point cloud of the standard model to obtain the registration relationship. Based on the registration relationship, the standard pin track of the standard model is subjected to coordinate transformation to obtain the initial pin track. The distance between the initial pin track and the surface of the target model, as well as the average value of the initial pin track in the medical image, are determined. The applicability of the initial pin track is determined based on the distance and the average value. The gradient value of the initial pin track in the directed distance map of the target area can also be determined. The initial pin track is updated based on the gradient value to obtain the updated pin track. The applicability of the updated pin track is determined until a preset number of candidate pin tracks are obtained. The candidate pin track with the highest applicability is determined as the target pin track.

[0165] Since the terminal processing procedure has been described in detail in the foregoing embodiments, it will not be repeated here.

[0166] The above-mentioned method for determining the path of medical screws is effective because the greater the distance from the path to the bone surface, the safer the screw placement; the greater the bone density on the path, the stronger the stability of the screw implantation. Based on the distance from the path to the bone surface and the imaging parameters reflecting the bone density on the path, a path with good safety and stability can be automatically determined, solving the problem of time-consuming and labor-intensive manual marking.

[0167] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by 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 embodiments described above 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 of other steps.

[0168] Based on the same inventive concept, this application also provides a device for determining the path of a medical screw to implement the above-described method for determining the path of a medical screw. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the device for determining the path of a medical screw provided below can be found in the limitations of the method for determining the path of a medical screw described above, and will not be repeated here.

[0169] In one embodiment, such as Figure 9 As shown, a device for determining the path of a medical screw is provided, comprising: a candidate path module 410, a parameter determination module 420, and a path determination module 430, wherein:

[0170] The candidate screw path module 410 is used to acquire candidate screw paths of medical screws in a target model at the target site; the target model is obtained based on medical images of the target site.

[0171] The parameter determination module 420 is used to determine the distance between the candidate pin path and the surface of the target model, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site.

[0172] The pin path determination module 430 is used to determine the target pin path of the medical screw from the candidate pin paths based on the distance and the image parameters.

[0173] In one embodiment, the candidate pin track module 410 further includes:

[0174] An initial path module is used to determine the initial path of the medical screw in the target model;

[0175] The spike update module is used to iteratively update the initial spike based on the directed distance map of the target location to obtain the updated spike.

[0176] The candidate pin path module is used to select the initial pin path and the updated pin path as candidate pin paths for the medical screw in the target model.

[0177] In one embodiment, the aforementioned initial pin track module is further configured to: obtain a target point cloud of the target region based on the surface of the target model; determine the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target region; and perform coordinate transformation processing on the standard pin track corresponding to the standard model based on the registration relationship to obtain the initial pin track.

[0178] In one embodiment, the spike update module is further configured to determine the gradient value of the initial spike in the directed distance map; update the initial spike according to the gradient value to obtain the updated spike; use the updated spike as the initial spike; and return to the step of determining the gradient value of the initial spike in the directed distance map until the number of candidate spikes meets a preset number.

[0179] In one embodiment, the above-mentioned spike update module is further configured to update the sampling point set of the initial spike according to the gradient value to obtain an updated sampling point set; determine the fitted line corresponding to the updated sampling point set; obtain the entry point and exit point of the updated spike according to the intersection point between the fitted line and the target model; and connect the entry point and the exit point to obtain the updated spike.

[0180] In one embodiment, the parameter determination module 420 is further configured to determine at least one target sampling point from the candidate spike path; determine the minimum distance between each of the target sampling points and the target point cloud of the target location; and determine the minimum value of at least one of the minimum distances as the distance between the candidate spike path and the target model.

[0181] In one embodiment, the spike determination module 430 is further configured to determine the applicability of each candidate spike based on the distance and the image parameters; the applicability is used to characterize the safety and stability of the candidate spike; the maximum applicability is determined from the applicability, and the candidate spike corresponding to the maximum applicability is determined as the target spike.

[0182] Each module in the aforementioned medical screw path determination device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0183] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for determining the path of a medical screw. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0184] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0185] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0186] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0187] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0188] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0189] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0190] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0191] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining the path of a medical screw, characterized in that, The method includes: Obtain candidate screw tracks of a medical screw in a target model of a target site; the target model is obtained based on medical images of the target site. Obtaining candidate screw tracks of a medical screw in a target model of a target site includes: determining the gradient value of an initial screw track in a directed distance map of the target site; updating the initial screw track according to the gradient value to obtain an updated screw track; using the updated screw track as the initial screw track; returning to the step of determining the gradient value of the initial screw track in the directed distance map of the target site; until the number of candidate screw tracks meets a preset number; and using the initial screw track and the updated screw track as the candidate screw tracks of the medical screw in the target model. The distance between the candidate pin path and the surface of the target model is determined, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site. Based on the distance and the image parameters, the target path of the medical screw is determined from the candidate paths.

2. The method according to claim 1, characterized in that, The process of obtaining candidate screw tracks in a target model at the target site for medical screws includes: Determine the initial path of the medical screw in the target model; Based on the directed distance map of the target location, the initial spike track is iteratively updated to obtain the updated spike track.

3. The method according to claim 2, characterized in that, Determining the initial nail path of the medical screw in the target model includes: Based on the surface of the target model, the target point cloud of the target region is obtained; Determine the registration relationship between the target point cloud and the standard point cloud; the standard point cloud is obtained based on the standard model of the target region. Based on the registration relationship, the standard pin track corresponding to the standard model is subjected to coordinate transformation to obtain the initial pin track.

4. The method according to claim 1, characterized in that, The step of updating the initial spike track based on the gradient value to obtain the updated spike track includes: Based on the gradient value, the initial set of sampling points of the spike path is updated to obtain the updated set of sampling points; Determine the fitted straight line corresponding to the updated set of sampling points; The entry and exit points of the updated nail track are obtained based on the intersection of the fitted straight line and the target model. Connect the nail entry point and the nail exit point to obtain the updated nail path.

5. The method according to claim 3, characterized in that, Determining the distance between the candidate spike and the surface of the target model includes: At least one target sampling point is determined from the candidate spike path; Determine the minimum distance between each target sampling point and the target point cloud of the target location; The minimum value of at least one of the minimum distances is determined as the distance between the candidate spike and the target model.

6. The method according to claim 1, characterized in that, The step of determining the target path of the medical screw from the candidate paths based on the distance and the image parameters includes: The suitability of each candidate spike is determined based on the distance and the image parameters. The maximum applicability is determined from the applicability, and the candidate nail path corresponding to the maximum applicability is determined as the target nail path.

7. A device for determining the path of a medical screw, characterized in that, The device includes: The candidate screw path module is used to acquire candidate screw paths of medical screws in a target model at the target site; the target model is obtained based on medical images of the target site. The candidate pin path module is further configured to determine the gradient value of the initial pin path in the directed distance map of the target location, update the initial pin path according to the gradient value to obtain the updated pin path, use the updated pin path as the initial pin path, return to the step of determining the gradient value of the initial pin path in the directed distance map of the target location, until the number of candidate pin paths meets the preset number, and use the initial pin path and the updated pin path as the candidate pin paths of the medical screw in the target model; A parameter determination module is used to determine the distance between the candidate pin path and the surface of the target model, as well as the image parameters of the candidate pin path in the medical image; the image parameters are used to characterize the bone density of the target site. The pin path determination module is used to determine the target pin path of the medical screw from the candidate pin paths based on the distance and the image parameters.

8. The apparatus according to claim 7, characterized in that, The candidate screw path module is further configured to determine the initial screw path of the medical screw in the target model, and to perform iterative update processing on the initial screw path according to the directed distance map of the target location to obtain the updated screw path.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images