A method for planning a surgical screw implantation path for pelvic fracture reduction

Abstract

The application discloses a kind of medical instrument technical field, a kind of planning method for pelvic fracture reduction surgical screw implantation path, comprising the following steps: the implantation number of retaining screw is calculated;The implantation position of retaining screw is calculated;The implantation direction of retaining screw is calculated.The input data of the application is the three-dimensional model of each broken bone of fracture pelvis and each fracture section point cloud that is reconstructed by preoperative CT of patient and spliced reduction completion, and the final task is to automatically generate the scheme of retaining screw implantation path in operation for the fracture condition of patient, including how many retaining screws are implanted, where retaining screw is implanted, in which direction screw is implanted and how deep is implanted.The application compared with existing conventional statistical model method, can guarantee the safety and optimal implantation effect of implantation path planning on personalized case;The retaining screw implantation planning path obtained by the application has stronger stability performance;Compared with traditional manual planning method, the planning time of single screw by the application is only about 5s, and the efficiency is improved by about 20 times.

Description

Technical Field

[0001] This invention relates to screw implantation methods in the field of medical device technology, and in particular to a method for planning screw implantation paths for pelvic fracture reduction surgery that provides stronger path stability. Background Technology

[0002] Screw placement is a crucial step in pelvic reduction surgery. Its function is to anchor the reduced pelvic fragments together, maintaining proper positioning until the fracture surface heals. Modern pelvic reduction surgery often employs closed reduction, resulting in minimal surgical trauma and minimal exposure of subcutaneous tissue. The clinical error rate for screw placement is approximately 18%. To improve the success rate, surgeons typically plan the screw placement path preoperatively or intraoperatively, manually designing the screw placement plan before execution.

[0003] Traditional manual planning relies heavily on the doctor's subjective experience, is time-consuming and labor-intensive, has low repeatability, and its inefficiency during surgery leads to increased exposure time for both doctors and patients under strong radiation, increasing surgical risks. Previous automated screw implantation planning algorithms were mostly based on statistical models or deep learning for probabilistic planning, which cannot guarantee the optimal implantation path in individual cases and may also penetrate the bone, damaging important soft tissues such as nerves and blood vessels around the lesion. These automated methods are currently mostly used for implantation planning in other hard tissues of the human body. For example, in oral implantation, Chinese patent CN115670661A discloses a method for planning the implantation path in zygomatic and pterygoid implant surgery. It uses the alveolar ridge of the dental region as the starting point and the zygomatic and pterygoid bone regions as the ending point, taking the possible paths connecting the starting and ending points. However, this method belongs to the field of deep learning and still requires doctor intervention while suffering from the aforementioned problems. There is still a research gap in the field of screw implantation planning for pelvic repositioning surgery. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by proposing a method for planning screw implantation paths in pelvic fracture reduction surgery. This method not only ensures the safety and optimal implantation effect of implantation path planning in individualized cases, but also provides stronger stability of the resulting fixation screw implantation path. Compared with traditional manual planning methods, the planning time for a single screw is significantly shorter.

[0005] The present invention includes the following steps: Step S1, calculating the number of fixation screws implanted; Step S2, calculating the implantation position of the fixation screws; Step S3, calculating the implantation direction of the fixation screws.

[0006] Furthermore, the above-mentioned step S1 of the present invention determines the number of fixation screws implanted based on the following three principles: First, "small-scale" fracture surfaces are not considered; second, at least one fixation screw should be implanted in each "qualified" fracture surface; third, fragments of "large-scale" fracture surfaces should be connected with at least two screws.

[0007] Furthermore, in this invention, the above step S1 includes the following steps: Step S1.1, performing principal component analysis on the point cloud of each fracture surface and combining it with threshold classification to obtain an effective point cloud; Step S1.2, each point cloud obtained in the end needs to be implanted with a fixation screw, that is, the number of effective point clouds is the same as the number of implanted screws.

[0008] Furthermore, in this invention, step S2 includes the following steps: Step S2.1, for each effective point cloud, continuously intercept the point cloud within a certain width along the first principal major axis direction of principal component analysis with a fixed step size, take its centroid, and obtain the coordinates corresponding to multiple centroids; Step S2.2, calculate the distance from the above coordinates to the boundary of the point cloud, and take the one with the largest distance as the fixation screw implantation point on the point cloud.

[0009] Furthermore, in this invention, step S3 includes the following steps: Step S3.1, using the least squares method to calculate the normal vector of each effective point cloud, which is used as the reference direction for implanting each fixation screw; Step S3.2, using a rotation matrix to generate a dense candidate direction in a conical scattering pattern; Step S3.3, introducing executability and safety constraints to eliminate illegal directions and leave effective paths; Step 4, the direction that allows the maximum implantation depth and is safest among the connected bone fragments is the optimal implantation direction.

[0010] Compared with existing technologies, this invention has the following beneficial effects: First, compared with existing conventional statistical model methods, this invention can ensure the safety and optimal implantation effect of implantation path planning in personalized cases. An experiment with 30 fixation screws based on clinical data shows that the clinical compliance rate of this method is 100%, and the clinical compliance rate of implantation planning results is 86.7%. Second, the fixation screw implantation planning path obtained by this invention has stronger stability. The optimal implantation position obtained has a larger boundary margin compared with directly calculating the centroid of the point cloud; the optimal implantation direction obtained has a greater implantation depth compared with principal component analysis or plane normal vectors. Third, compared with traditional manual planning methods, this invention takes only about 5 seconds to plan a single screw, improving efficiency by about 20 times. Attached Figure Description

[0011] Figure 1 This is a flowchart of the present invention;

[0012] Figure 2This is a schematic diagram of the three-dimensional model of each bone fragment in the fractured pelvis and the point cloud of each fracture section in Embodiment 1 of the present invention;

[0013] Figure 3 This is a projection width diagram of the point cloud along the first major axis in Embodiment 1 of the present invention;

[0014] Figure 4 This is the maximum inner circle diagram of the effective fracture cross-section point cloud in Embodiment 1 of the present invention;

[0015] Figure 5 This is a fitted plane diagram of the effective point cloud predicted using the least squares method in Embodiment 1 of the present invention;

[0016] Figure 6 This is the final path planning diagram of Embodiment 1 of the present invention;

[0017] Figure 7 This is a path planning diagram of Embodiment 2 of the present invention;

[0018] Figure 8 This is a path planning diagram of Embodiment 3 of the present invention;

[0019] Among them, 1. Fractured pelvis in Example 1, 2. Point cloud of fracture surface in Example 1, 3. Effective point cloud, 4. Maximum inner circle of point cloud, 5. Center of maximum inner circle, 6. First sliding window, 7. Second sliding window, 8. Centroid of local point cloud, 9. First principal major axis direction, 10. Effective point cloud of fitted plane, 11. Normal vector of fitted plane, 12. First principal major axis direction of fitted plane, 13. Fitted plane, 14. First planned path in Example 1, 15. Second planned path in Example 1, 16. Third planned path in Example 1, 17. Fourth planned path in Example 1, 18. Fifth planned path in Example 1, 19. First planned path in Example 2, 20. Second planned path in Example 2, 21. Third planned path in Example 2, 22. Fourth planned path in Example 2, 23. First planned path in Example 3, 24. Second planned path in Example 3, 25. Third planned path in Example 3. Detailed Implementation

[0020] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. These embodiments are based on the technical solutions of the present invention and provide detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the following embodiments.

[0021] Example 1

[0022] The input data for this invention consists of a three-dimensional model (STL format) of each bone fragment in the fractured pelvis, reconstructed and stitched together from the patient's preoperative CT scan, and point clouds of each fracture section. (See attached image.) Figure 2The ultimate goal is to automatically generate a surgical plan for screw placement based on the patient's fracture condition, including the number of screws, placement location, direction, and depth. This method decouples the problem into three steps: calculating the number of screws, calculating the placement location, and calculating the placement depth. The flowchart is shown below. Figure 1 .

[0023] The first step is to calculate the number of fixation screws required. The number of fixation screws is highly correlated with the complexity of the fracture. This method determines the number of fixation screws based on the following three principles: 1) "Small-scale" fracture surfaces are not considered; 2) At least one fixation screw should be implanted in each "qualified" fracture surface; 3) Fragments in "large-scale" fracture surfaces should be connected with at least two screws. First, principal component analysis is performed on the point cloud of each fracture section to obtain the direction of the first principal major axis. Then, the point cloud is projected along the direction of the first principal major axis to obtain the projection width of the point cloud in the direction of the first principal major axis. See [link to relevant documentation]. Figure 3 Finally, the width is compared with two preset thresholds (denoted as threshold 1 and threshold 2, where threshold 2 > threshold 1). If the width is greater than threshold 1, the point cloud is judged as "small-scale" and discarded; if the width is greater than threshold 1 and less than threshold 2, the point cloud is recorded as a valid point cloud; if the width is greater than threshold 2, the point cloud is considered a "large-scale" point cloud, and it is cut into two valid point clouds along a plane perpendicular to the first principal major axis of the projection center. Each valid point cloud should ultimately have one retaining screw implanted, meaning the number of retaining screws should be consistent with the number of valid point clouds obtained.

[0024] The second step is to calculate the implantation location of the fixation screws. As known from the first step, each fixation screw corresponds to an effective point cloud. To achieve optimal screw stability, the ideal implantation location should be the center of the largest inner circle of the effective fracture section point cloud, see [link to relevant documentation]. Figure 4 However, the point cloud on the fracture surface is highly irregular, making analytical solutions impossible. This method first calculates the first principal major axis direction for each effective point cloud cluster, sets a fixed step size and sliding window width, and then uses the sliding window to extract a local point cloud within the sliding window width along the first principal major axis direction at the specified step size, obtaining the local centroid. Next, the distance from this centroid to the point cloud boundary is calculated. Finally, the local centroid with the largest boundary distance is selected as the optimal implantation location.

[0025] Finally, the implantation direction of the fixation screw is calculated. In determining the screw implantation direction, this method plans the screw path sequentially from smallest to largest effective point cloud size, considering: 1) Safety constraints: the planned screw path must not penetrate the bone wall and damage the surrounding soft tissues of the pelvis; 2) Feasibility constraints: the newly planned screw path must not interfere with other already planned paths; 3) Stability principle: the planned screw path maximizes the implantation depth within connected bone fragments to ensure optimal fixation. The calculation process is as follows: First, see... Figure 5 The least squares method was used to predict the fitting plane of the effective point cloud, obtaining the normal vector. Then, starting from the implantation position of this effective point cloud, four rays were drawn along the positive and negative directions of the normal vector and the positive and negative directions of the first principal major axis obtained from principal component analysis. Without penetrating the bone wall, the total length of the four rays was obtained. The sum of the ray lengths related to the normal vector direction was compared with the sum of the ray lengths related to the first principal major axis direction, and the direction with the larger value was taken as the reference direction. Next, using this reference direction as the parent direction, a dense array of candidate directions surrounding the reference direction was generated using a rotation matrix. Finally, the maximum implantation depth under safety and executability constraints was calculated along these candidate directions, and the path with the largest implantation depth was selected as the optimal implantation path. The results are shown below. Figure 6 .

[0026] Example 2

[0027] The patient had three pelvic fractures. Using the method described in this invention, it was calculated that three fixation screws should be implanted. The time taken for each screw was 2.95, 3.65, and 3.71 seconds, respectively, all meeting clinical surgical criteria. Figure 7 As shown.

[0028] The patient suffered a large fracture. Using the method described in this invention, two fixation screws were implanted, with implantation times of 3.24 seconds and 3.29 seconds respectively, meeting clinical surgical standards. Figure 8 As shown.

[0029] The above embodiments are merely illustrative of the design principles and uses of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for planning screw implantation paths in pelvic fracture reduction surgery, characterized in that... Includes the following steps: Step S1: Calculate the number of retention screws implanted; Step S2: Calculate the implantation position of the retention screw; Step S3: Calculate the implantation direction of the retention screw; In step S1, the number of fixation screws implanted is determined based on the following three principles: First, "small-scale" fracture surfaces are not considered; second, at least one fixation screw should be implanted in each "qualified" fracture surface; third, fragments of "large-scale" fracture surfaces should be connected with at least two screws. Step S1 includes the following steps: Step S1.1: Perform principal component analysis on the point cloud of each fracture surface, and combine it with threshold classification to obtain the effective point cloud; In step S1.2, each point cloud obtained in the end needs to be implanted with a fixing screw, that is, the number of effective point clouds is the same as the number of implanted screws; Step S2 includes the following steps: Step S2.1: For each effective point cloud cluster, continuously extract the point cloud within a certain width along the first principal major axis of principal component analysis with a fixed step size, take its centroid, and obtain the coordinates corresponding to multiple centroids; Step S2.2: Calculate the distance from the above coordinates to the boundary of the point cloud, and take the largest distance as the fixation screw implantation point on the point cloud; Step S3 includes the following steps: Step S3.1: Calculate the normal vector of each effective point cloud using the least squares method, and use it as the reference direction for implanting each fixation screw. Step S3.2: Use a rotation matrix to generate a dense array of candidate directions in a cone-shaped scattering pattern; Step S3.3 introduces executability and security constraints, eliminates illegal directions, and leaves valid paths; Step S3.4: The optimal implantation direction is the one that allows for the greatest and safest implantation depth among the connected bone fragments.

Images