Methods and systems for positioning a blocking pin for a distal tibial fracture intramedullary nail

Abstract

The application discloses a tibial fracture intramedullary nail blocking nail positioning guide method and system, and belongs to the technical field of medical image processing and spatial data calculation. The method first extracts a target point cloud subset meeting preset conditions of bone density and cortical bone thickness based on preoperative CT image data, and generates initial path parameters of the blocking nail in combination with three-dimensional data of the intramedullary nail; further, real-time two-dimensional X-ray image data in operation is acquired, a bone block transformation matrix and an intramedullary nail transformation matrix are calculated by using an image registration algorithm, and dynamic path parameters, dynamic intramedullary nail data and dynamic cortical bone inner wall point cloud data are obtained; then, spatial relationship parameters are calculated according to the dynamic data, when the spatial relationship parameters do not meet preset safety threshold conditions, iterative optimization is performed with the dynamic target point cloud subset as a search range, and final path parameters are output. The application realizes the combination of preoperative bone quality evaluation and intraoperative dynamic real-time tracking, and improves the accuracy, effectiveness and safety of blocking nail positioning.

Description

Technical Field

[0001] This invention relates to the field of medical image processing and spatial data computing technology, and in particular to a method and system for positioning and guiding intramedullary nail blocking screws in distal tibial fractures. Background Technology

[0002] Distal tibial fractures require high levels of distal stability control due to the wider medullary canal and proximity of the fracture to the ankle joint when fixed with intramedullary nails. The blocking screw technique can improve fixation stability and fracture reduction maintenance by altering the spatial position of the intramedullary nail within the medullary canal; however, the accuracy of the blocking screw's positioning directly impacts its auxiliary fixation effect. Therefore, utilizing computer-aided technology to achieve precise positioning and guidance of the blocking screw is crucial for improving the quality of intramedullary nail fixation for distal tibial fractures and ensuring standardized surgical procedures.

[0003] Currently, with the development of computer-aided technology, the positioning of intramedullary nail blocking screws for distal tibial fractures is increasingly being combined with digital image processing and data analysis methods for assisted planning. Existing techniques typically involve acquiring preoperative CT images of the patient and using medical image processing software to perform three-dimensional reconstruction of the tibial bone structure, fracture area, and medullary canal morphology, creating an individualized bone model for the patient. Based on this, computer analysis is used to spatially analyze the medullary canal axis, fracture end position, intramedullary nail implantation path, and preset blocking screw position in the reconstructed model, obtaining relevant anatomical parameters and spatial relationship data. Some methods further incorporate three-dimensional visualization technology to simulate the matching relationship between the intramedullary nail and the medullary canal, and use the data analysis results to help determine the positioning area and insertion direction of the blocking screw. Furthermore, based on technologies such as digital image registration, three-dimensional coordinate measurement, and computer-aided design, virtual planning of the blocking screw positioning process can also be achieved, providing data support for intraoperative positioning operations.

[0004] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0005] In surgical scenarios where fracture reduction status changes dynamically, preoperative planning models based on static CT data struggle to capture the real-time evolution of fracture end positions caused by intraoperative traction, positioning adjustments, and intramedullary nail insertion. This leads to deviations between the planned path and the actual anatomical state. The pre-set blocking nail position is prone to relative displacement due to intraoperative micro-movements at the fracture ends, causing the blocking nail to transform from "effective blocking" into "ineffective interference," or even generating unexpected levering effects leading to "reverse deformity." Furthermore, for complex cases such as severe osteoporosis or comminuted fractures, existing methods overemphasize geometric matching of anatomical morphology, neglecting the quantitative assessment of bone density distribution and cortical bone thickness—crucial for blocking nail holding force—increasing the risk of internal fixation failure. Therefore, the limitations of current technology ultimately result in a lack of real-time feedback mechanisms and mechanical stability guarantees in the blocking nail positioning and guidance process, making it difficult to achieve precise and safe surgical guidance in complex and variable clinical environments. Summary of the Invention

[0006] To address the technical problems of existing technologies, which lack a real-time sensing and feedback mechanism for dynamic changes in fracture reduction and differences in bone quality, resulting in insufficient precision in guiding the blocking screw, reduced effective blocking effect, and inadequate positioning safety in complex fracture cases, this invention provides a method and system for guiding the positioning of intramedullary nails for distal tibial fractures. The technical solution is as follows:

[0007] On the one hand, a method for guiding the positioning of the blocking screw in intramedullary nailing for distal tibial fractures is provided. This method includes: acquiring preoperative CT image data of the distal tibia; reconstructing three-dimensional point cloud data of the tibia based on the preoperative CT image data; and extracting point cloud data of the cortical bone wall; calculating bone density based on the preoperative CT image data and cortical bone thickness based on the cortical bone wall point cloud data; extracting three-dimensional point cloud data where bone density and cortical bone thickness meet preset conditions to form a target point cloud subset; acquiring preset three-dimensional data of the intramedullary nail; calculating and generating initial path parameters for the blocking screw within the coordinate range of the target point cloud subset, combining the three-dimensional data of the intramedullary nail, wherein the initial path parameters include at least the coordinates of the insertion point and the path direction; acquiring real-time two-dimensional X-ray image data during fracture reduction and intramedullary nail insertion; and using an image registration algorithm to calculate the bone block transformation matrix and intramedullary nail transformation matrix relative to the real-time two-dimensional X-ray image data of the tibial three-dimensional point cloud data and the intramedullary nail three-dimensional data, respectively. The algorithm involves several steps: First, it transforms the initial path parameters, cortical bone wall point cloud data, and target point cloud subset using the bone block transformation matrix to obtain dynamic path parameters, dynamic cortical bone wall point cloud data, and a dynamic target point cloud subset. Second, it transforms the intramedullary nail 3D data using the intramedullary nail transformation matrix to obtain dynamic intramedullary nail data. Third, it calculates the minimum gap between the blocking nail path and the intramedullary nail, the safe distance between the blocking nail path and the cortical bone wall, and the angle between the blocking nail path and the intramedullary nail based on the dynamic path parameters, dynamic intramedullary nail data, and dynamic cortical bone wall point cloud data, thus obtaining spatial relationship parameters. Fourth, it iteratively optimizes the dynamic path parameters by using the coordinate set of the dynamic target point cloud subset as the search range and employing a dual constraint of the spatial relationship parameters being contained within the dynamic target point cloud subset and satisfying the preset safety threshold condition, outputting the final path parameters.

[0008] On the other hand, an intramedullary nail blocking screw positioning and guidance system for distal tibial fractures is provided, including: a tibial target area extraction module, an initial blocking planning module, a screw pose registration module, a screw dynamic mapping module, a blocking interference analysis module, and a blocking guidance optimization module; the tibial target area extraction module is used to acquire preoperative CT image data of the distal tibia, reconstruct 3D point cloud data of the tibia based on the preoperative CT image data, and extract point cloud data of the cortical bone inner wall; bone density is calculated based on the preoperative CT image data, and cortical bone thickness is calculated based on the cortical bone inner wall point cloud data. Three-dimensional point cloud data with bone density and cortical bone thickness meeting preset conditions are extracted to form a target point cloud subset. An initial blocking planning module acquires preset intramedullary nail three-dimensional data and, within the coordinate range of the target point cloud subset, calculates and generates initial path parameters for the blocking nail based on the intramedullary nail three-dimensional data. These initial path parameters include at least the nail insertion point coordinates and path direction. A bone nail pose registration module acquires real-time two-dimensional X-ray image data during fracture reduction and intramedullary nail insertion. Using an image registration algorithm, it calculates the tibial three-dimensional point cloud data and the intramedullary nail three-dimensional data respectively. Based on the bone block transformation matrix and intramedullary nail transformation matrix of the relative real-time two-dimensional X-ray image data; the bone nail dynamic mapping module is used to perform coordinate transformation on the initial path parameters, cortical bone inner wall point cloud data and target point cloud subset based on the bone block transformation matrix to obtain dynamic path parameters, dynamic cortical bone inner wall point cloud data and dynamic target point cloud subset; the intramedullary nail transformation matrix is ​​used to perform coordinate transformation on the three-dimensional data of the intramedullary nail to obtain dynamic intramedullary nail data; the obstruction interference analysis module is used to calculate the minimum gap between the obstruction nail path and the intramedullary nail, the safe distance between the obstruction nail path and the cortical bone inner wall, and the angle between the obstruction nail path and the intramedullary nail based on the dynamic path parameters, dynamic intramedullary nail data and dynamic cortical bone inner wall point cloud data to obtain spatial relationship parameters; the obstruction guidance optimization module is used to perform iterative optimization on the dynamic path parameters when the spatial relationship parameters do not meet the preset safety threshold conditions, using the coordinate set of the dynamic target point cloud subset as the search range, and with the dual constraints of being contained within the dynamic target point cloud subset and the spatial relationship parameters meeting the preset safety threshold conditions, and output the final path parameters.

[0009] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0010] Compared to existing technologies that primarily rely on preoperative static CT planning, intraoperative two-dimensional fluoroscopy, and manual experience for blocking nail positioning, this invention achieves dynamic perception and adaptive optimization of the blocking nail positioning and guidance process by integrating preoperative three-dimensional bone structure information, bone density, cortical bone thickness, and other bone quality parameters, as well as real-time intraoperative image feedback. Existing technologies struggle to reflect changes in bone block position during fracture reduction, traction adjustment, and intramedullary nail insertion, easily leading to deviations between the preset path and the actual anatomical state, affecting the effective blocking effect of the blocking nail; simultaneously, they ignore local bone quality differences, easily resulting in insufficient fixation stability in complex fractures or osteoporosis. This invention improves the bone compatibility of the blocking nail implantation site by constructing a three-dimensional point cloud model of the tibia and selecting target areas that meet the requirements for bone density and cortical bone thickness. It also achieves real-time correlation between preoperative three-dimensional planning and the actual intraoperative state by acquiring the dynamic transformation relationship between the bone block and the intramedullary nail through real-time two-dimensional X-ray image registration, reducing positioning deviations caused by changes in fracture reduction. Furthermore, by calculating the spatial relationship between the blocking nail path, the intramedullary nail, and the cortical bone, and combining iterative optimization algorithms to optimize paths that do not meet safety thresholds, the invention achieves automatic adjustment and safety constraint control of the blocking nail path. Therefore, this invention can improve the accuracy of blocking nail positioning and implantation stability, reduce the risks of intramedullary nail interference, cortical bone damage, blocking failure, and internal fixation failure, and enhance the safety and reliability of intramedullary fixation surgery for complex distal tibial fractures. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart of a method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures provided in an embodiment of this application;

[0013] Figure 2 A logic diagram of the intramedullary nail blocking screw positioning and guiding method for distal tibial fractures provided in the embodiments of this application;

[0014] Figure 3 This is a schematic diagram of the structure of the intramedullary nail blocking screw positioning and guiding system for distal tibial fractures provided in the embodiments of this application. Detailed Implementation

[0015] Embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present disclosure are shown in the drawings, it should be understood that embodiments of the present disclosure may be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure.

[0016] It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure. In the description of the embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "this embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects.

[0017] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0018] Distal tibial fractures are a common and challenging type of lower limb fracture in clinical practice. Due to the thin soft tissue coverage and complex anatomical structure of the distal tibia, and the often significant instability of the distal fracture fragments, traditional conservative treatment or simple external fixation methods are prone to problems such as loss of reduction and malunion. Currently, intramedullary nailing has become an important treatment method for distal tibial fractures due to its advantages of minimal trauma, good soft tissue protection, and allowing for early functional exercises. However, due to the enlarged medullary canal of the distal tibia, insufficient distal fracture control, and the tendency for fracture fragments to rotate or shift, intramedullary nailing alone often fails to achieve ideal distal stability. Therefore, it is usually necessary to use Poller nails to limit the displacement of the intramedullary nail within the medullary canal and improve fracture reduction stability.

[0019] Current methods for positioning blocking screws primarily rely on preoperative CT imaging assessment, intraoperative two-dimensional X-ray fluoroscopy, and the surgeon's experience to determine their location. During preoperative planning, the approximate implantation area of ​​the blocking screw is typically determined based on the patient's static bone structure information. However, during actual surgery, the fracture reduction status is affected by traction forces, patient positioning adjustments, micromovements at the fracture ends, and the mechanical forces during intramedullary nail insertion. This causes dynamic changes in the fracture fragments and intramedullary nail relative to the preoperative planned state, resulting in deviations between the pre-set blocking screw path and the actual intraoperative anatomical position. When the blocking screw position fails to adjust synchronously with the fracture reduction status, it may fail to provide effective mechanical resistance, or even cause abnormal prying forces due to deviations in the blocking direction, leading to problems such as lost fracture reduction, abnormal force alignment, or reverse deformity.

[0020] Furthermore, existing methods for positioning detaining screws typically focus more on the overall skeletal morphology and spatial geometry, while lacking effective assessment of crucial factors affecting detaining screw fixation effectiveness, such as local bone density distribution, cortical bone thickness, and bone load-bearing capacity. For patients with osteoporosis, comminuted fractures, or complex distal bone structures, significant differences in local bone conditions mean that path planning based solely on anatomical morphology may result in insufficient bone support in the detaining screw implantation area, reducing the screw's holding power and increasing the risk of internal fixation loosening, failure, or even a second surgery.

[0021] Meanwhile, existing technologies lack a real-time feedback mechanism for changes in fracture status, the actual position of the intramedullary nail, and the spatial relationship between the insertion path of the blocking nail during intraoperative positioning. This typically requires the surgeon to repeatedly adjust the guide position based on fluoroscopic images, increasing surgical complexity and radiation exposure time, and making it difficult to guarantee positioning accuracy in complex cases. Especially in cases of distal tibial fractures with significant dynamic changes in reduction status and severe bone structural damage, traditional methods relying on static planning and manual judgment struggle to simultaneously address anatomical reduction requirements, blocking effect, and local bone biomechanical stability.

[0022] Against this backdrop, addressing the shortcomings of existing technologies such as the lack of intraoperative dynamic sensing mechanisms in the positioning process of intramedullary nails, the inability to adjust in real time according to changes in fracture reduction, and insufficient consideration of local bone quality differences leading to inadequate positioning accuracy and fixation stability, this invention provides a method for guiding the positioning of intramedullary nails for distal tibial fractures, such as... Figure 1 The diagram shown is a flowchart of a method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures provided in an embodiment of this application; see reference. Figure 1 It should be understood that this invention, by innovatively integrating preoperative quantitative assessment of bone quality with real-time dynamic tracking technology during surgery, effectively addresses the limitations of existing technologies in dealing with dynamic changes in fracture reduction and complex bone conditions. Specifically, firstly, preoperative CT data is used to accurately screen target areas with high bone density and appropriate cortical bone thickness, providing a stable mechanical basis for the blocking nail. Secondly, by registering real-time two-dimensional X-ray images with a three-dimensional model during surgery, the dynamic spatial position of the fracture ends and intramedullary nail is captured, enabling real-time updates and corrections to the planned path. Finally, by combining spatial relationship parameter analysis and a path optimization algorithm under dual constraints, the blocking nail path is ensured to be in the optimal bone region while meeting safety thresholds. This technical solution not only effectively eliminates the risk of blocking failure or reverse deformity caused by deviations between static planning and actual intraoperative anatomical state, but also significantly improves the accuracy, safety, and stability of the surgery and internal fixation, possessing significant clinical value for improving the surgical treatment of complex distal tibial fractures.

[0023] Furthermore, it should be added that, Figure 2This is a logic diagram of the intramedullary nail blocking screw positioning and guiding method for distal tibial fractures provided in the embodiments of this application; based on Figure 2 It should also be understood that the various steps of this invention are interconnected and work together to form a closed-loop positioning mechanism consisting of preoperative analysis, path planning, intraoperative feedback, dynamic updates, safety evaluation, and optimization adjustments. Preoperative CT data reconstruction and analysis of bone density and cortical bone thickness are used to screen a subset of target point clouds, providing a basis for blocking nail positioning that combines anatomical adaptability and mechanical stability. Initial path parameters are generated by combining intramedullary nail 3D data to achieve individualized planning. Furthermore, intraoperative X-ray image registration is used to obtain real-time pose changes of bone blocks and intramedullary nails, and the blocking nail path, bone structure, and intramedullary nail model are updated synchronously through a transformation matrix, adapting the positioning scheme to dynamic changes during surgery. Based on this, the safety of the current path is evaluated through spatial relationship parameters, and schemes that do not meet the conditions are optimized and adjusted using a path optimization algorithm. Through data transmission and constraint association, each step achieves continuous control from target area screening and path planning to dynamic correction and intelligent optimization. Compared with existing methods that rely on experience and static planning, this improves the accuracy, safety, and stability of blocking nail positioning, and reduces the risks of positioning deviation, instrument interference, and fixation failure.

[0024] Specifically, as a further option, refer to Figure 1 , Figure 2 The first step of the intramedullary nail blocking screw positioning and guidance method for distal tibial fractures is as follows: acquiring preoperative CT image data of the distal tibia, reconstructing three-dimensional point cloud data of the tibia based on the preoperative CT image data, and extracting point cloud data of the cortical bone inner wall; calculating bone density based on the preoperative CT image data, and calculating cortical bone thickness based on the cortical bone inner wall point cloud data, extracting three-dimensional point cloud data that meet preset conditions for bone density and cortical bone thickness to form a target point cloud subset; and achieving precise selection of the blocking screw implantation area by fusing bone structure information and bone quality parameters, avoiding the problem of insufficient bone quality caused by positioning based solely on geometric morphology, providing a target area with good bone biomechanical support for subsequent path planning, and improving the reliability of blocking screw fixation.

[0025] In this embodiment of the invention, the first step is specifically implemented through the following sub-steps:

[0026] Step S101: Obtain preoperative CT image data of the distal tibia, reconstruct three-dimensional point cloud data of the tibia based on the preoperative CT image data, and extract point cloud data of the inner wall of the cortical bone.

[0027] Specifically, the distal tibia of the patient was first scanned using a multi-slice spiral CT scanner to obtain raw DICOM data with a slice thickness of 0.5 mm to 1.0 mm. The raw DICOM data was then processed using 3D reconstruction software (such as Mimics or 3D Slicer). A threshold segmentation algorithm was used to extract bone tissue, with the Huntsfield unit (HU) threshold set to a range of 226–3000 (corresponding to cortical and cancellous bone). Noise interference was removed using a region growing method.

[0028] Subsequently, a triangular mesh surface model of the tibia was constructed using the moving cube algorithm, and the model was then smoothed and simplified. Based on the triangular mesh model, three-dimensional point cloud data of the tibia was generated using a mesh vertex extraction method.

[0029] To extract point cloud data of the cortical bone inner wall, morphological operations or edge detection algorithms are used to distinguish the boundary between the cortical bone and the medullary cavity. In a specific embodiment, a simulated cavity wall extraction algorithm can be used to calculate the distance from each point in the point cloud to the center line of the medullary cavity. The cortical bone inner wall is defined by a preset distance threshold (such as the bone wall layer closest to the center line), thereby obtaining accurate cortical bone inner wall point cloud data, denoted as set. ,in These are the three-dimensional coordinate points on the inner wall.

[0030] In some preferred embodiments, to further refine preoperative individualized planning, anatomical morphological features of the tibia can be further extracted based on preoperative CT image data and reconstructed tibial 3D point cloud data. Specifically, morphological parameters, including but not limited to bone length, minimum transverse diameter and anteroposterior diameter of the bone, and maximum transverse diameter and anteroposterior diameter of the bone, can be extracted by calculating the spatial extrema and slice contours of the point cloud data.

[0031] The extraction of the aforementioned morphological parameters provides a quantitative basis for the precise preoperative assessment of the degree of anatomical narrowing of the medullary canal and the curvature of the bone shaft. This morphological data can be directly used to assist in determining the 'pre-set three-dimensional data of the intramedullary nail' in subsequent steps (such as assisting the surgeon in accurately selecting the length specification and main nail diameter of the intramedullary nail). By combining this anatomical morphological data (used for intramedullary nail matching) with the aforementioned bone mineral density and cortical bone thickness data (used for assessing the holding force of the blocking nail), the geometric fit and biomechanical stability of the fixation of distal tibial fractures can be comprehensively considered.

[0032] Step S102: Calculate bone mineral density based on preoperative CT image data. This step specifically includes the following sub-steps:

[0033] First, the DICOM file header information is parsed to obtain the Rescale Slope and Rescale Intercept parameters, and the stored pixel values ​​of the original CT image are converted into standard Huntsfield unit (HU) grayscale values.

[0034] Secondly, a spatial mapping relationship is established. The coordinate space (world coordinate system) of the tibial three-dimensional point cloud data is mapped to the voxel coordinate system of the preoperative CT image. For any coordinate point in the cortical bone inner wall point cloud data... By combining inverse projection with trilinear interpolation, the precise grayscale value corresponding to this coordinate in the CT volume data space is calculated. .

[0035] Finally, bone mineral density conversion is performed. Based on the physical definition of CT values, the Huntsfield unit grayscale values ​​are converted to apparent bone mineral density. The conversion formula is: ;in, coordinate point Bone mineral density at the site (unit: ), The density of pure water (usually taken as 1.0) Therefore, each point in the point cloud data They were all assigned corresponding bone density attribute values.

[0036] Step S103: Calculate cortical bone thickness based on point cloud data of the cortical bone inner wall. This step specifically includes the following sub-steps:

[0037] First, a model of the cortical bone outer surface is constructed. Based on the tibial 3D model obtained in step S101, a high-precision triangular mesh model of the cortical bone outer surface is generated using curvature filtering or edge extraction algorithms. .

[0038] Secondly, calculate the normal vector. This is done for the point cloud dataset of the cortical bone inner wall. Each data point in Select its k-nearest neighbor points (e.g., k=15), construct the covariance matrix, and use principal component analysis (PCA) to calculate the normal vector of that point. By adjusting the direction of the normal vector, it is made to point towards the outer side of the cortical bone (i.e., away from the center of the medullary cavity).

[0039] Next, ray projection and intersection. Using data points... Starting from the normal vector, along A ray is emitted in the direction The ray-triangle intersection algorithm (such as the Möller-Trumbore algorithm) is used to calculate the ray. Cortical bone outer surface model The set of intersection points of all triangular faces in the triangle.

[0040] Finally, the cortical bone thickness is determined. From the set of intersection points, those located in the positive direction of the ray and at a distance from the origin are selected. The nearest intersection .calculate and Euclidean distance between This distance That is, data points The thickness of the cortical bone at that location.

[0041] Step S104: Extract three-dimensional point cloud data that meet the preset conditions for bone density and cortical bone thickness to form a subset of the target point cloud.

[0042] To ensure the stability of the depressor implantation, it is necessary to select the bone wall region with optimal mechanical properties. Specific selection criteria are as follows:

[0043] Preset bone mineral density threshold and cortical bone thickness threshold In a preferred embodiment, the bone mineral density threshold Set to 0.8 (Corresponding to a HU value of approximately 800), cortical bone thickness threshold Set to 3.0mm (this value can be adjusted according to the diameter of the intramedullary nail, and is usually required to be no less than 80% of the screw diameter).

[0044] Traverse the point cloud data of the cortical bone inner wall and extract all data that simultaneously satisfy the following conditions. and The data points are defined as a subset of the target point cloud. This subset excludes areas of severe osteoporosis (low bone density) or areas of thin cortical bone (insufficient support), providing a feasible solution space with sufficient holding force and safety margin for subsequent blocking pin path planning.

[0045] By mapping the Huntsfield unit grayscale values ​​in preoperative CT images to the three-dimensional point cloud coordinates of the tibia, a correlation between bone density parameters and spatial location was established. Simultaneously, cortical bone thickness was calculated based on the spatial distance between the point clouds of the inner and outer surfaces of the cortical bone, enabling the quantitative expression of bone quality parameters in three-dimensional space. Compared to methods that rely solely on bone morphology for localization, this approach considers both bone tissue density and cortical bone structural strength, providing a more accurate mechanical evaluation basis for selecting the blocking screw implantation area, improving the reliability of the target localization area, and reducing the risk of decreased fixation stability due to insufficient bone mass.

[0046] The second step in the intramedullary nail blocking pin positioning and guidance method for distal tibial fractures is as follows: acquire the preset three-dimensional data of the intramedullary nail, and calculate and generate the initial path parameters of the blocking pin within the coordinate range of the target point cloud subset, combined with the three-dimensional data of the intramedullary nail. The initial path parameters include at least the coordinates of the entry point and the path direction. This step achieves individualized initial planning of the blocking pin path by combining the target bone region with the spatial relationship of the intramedullary nail, so that the blocking pin can effectively block the intramedullary nail while meeting the bone conditions.

[0047] This embodiment provides a second step in the intramedullary nail blocking screw positioning and guiding method for distal tibial fractures, specifically including the following sub-steps:

[0048] S201: Extract the central axis of the three-dimensional data of the intramedullary nail.

[0049] Obtain the pre-defined 3D model data of the intramedullary nail, which is typically a triangular mesh model in STL (Standard Tessellation Language) format. To accurately extract the central axis, the Random Sample Consensus Algorithm (RANSAC) is used to fit a cylinder to the main trunk of the intramedullary nail model, obtaining the central axis vector of the cylinder.

[0050] If the intramedullary nail has an anatomically curved design, a slicing method is further employed to improve accuracy: vertical cross-sections are cut along the long axis of the intramedullary nail at predetermined intervals (e.g., 1.0 mm), and the geometric center (centroid) of each cross-section is calculated, resulting in a series of discrete center points. Subsequently, these discrete center points are fitted using cubic B-spline curves or the least squares method to generate a smooth, continuous curve as the central axis of the intramedullary nail. The central axis Includes its spatial equation and direction vector .

[0051] S202: Based on bone density and cortical bone thickness, perform connectivity analysis on the target point cloud subset, screen out the largest connected region with bone density greater than the preset neighborhood mean, and mark it as the high holding power core region.

[0052] The target point cloud subset generated based on the aforementioned steps (Including bone density) and cortical bone thickness (Attributes) to perform connectivity analysis to locate the optimal nail insertion area.

[0053] First, a subset of the target point cloud is indexed using a KD-Tree data structure to accelerate neighborhood search. The search radius is then set. (e.g., 3.0mm), for each point in the point set Search for all neighboring points within its spherical neighborhood. .

[0054] Secondly, calculate the average bone mineral density of all neighboring points within the neighborhood. Determine the current point Bone density Is it greater than the neighborhood mean? If satisfied Then the point Marked as local high-density feature points.

[0055] Finally, connected component clustering is performed. The density-based noise applied spatial clustering algorithm (DBSCAN) is used to cluster all locally high-density feature points. A neighborhood radius is set. (e.g., 2.0mm) and a minimum number of points (MinPts) (e.g., 15). Points that meet the density requirements are grouped into the same connected region. The number of points contained within each connected region is calculated, and the connected region with the largest number of points is selected and marked as the "high holding power core region". .

[0056] S203: Calculate the three-dimensional geometric center of the high holding force core area, and project this three-dimensional geometric center onto the outer surface of the cortical bone. Set the surface coordinates obtained from the projection as the entry point coordinates of the initial path parameters. The specific calculation is as follows:

[0057] First, calculate the high-holding-force core area. The three-dimensional geometric center of all points inside That is, the average value of the coordinates of each point.

[0058] Secondly, obtain a mesh model of the cortical bone outer surface of the tibia. Using spatial nearest neighbor search algorithms (such as KD-Tree search), a model of the cortical bone outer surface was constructed. Find the distance from the geometric center The nearest surface vertex.

[0059] Finally, the three-dimensional coordinates of the surface vertex are set as the entry point coordinates of the initial path parameters. .

[0060] S204: Construct a spatial normal plane passing through the coordinates of the entry point and perpendicular to the central axis; determine the intersection point of the central axis and the spatial normal plane; generate candidate path lines passing through the coordinates of the entry point within the spatial normal plane, where the shortest distance between the candidate path lines and the intersection point is equal to a preset offset distance; use the direction vector of the candidate path lines as the path direction of the initial path parameters. The specific implementation is as follows:

[0061] First, construct the spatial normal plane. Using the coordinates of the nail entry point A point on the plane, with the central axis of the intramedullary nail as the reference point. Direction vector Given the normal vector of the plane, construct a spatial normal plane passing through the entry point.

[0062] Secondly, determine the spatial reference point. Solve for the central axis of the intramedullary nail. With the spatial normal plane The intersection point is denoted as .

[0063] Next, calculate the preset offset distance. This distance is given by the formula. The calculation shows that, among which The radius of the intramedullary nail is [missing information]. The radius of the blocking pin, This is a preset safety gap (usually 0~1mm).

[0064] Finally, candidate path lines are generated and their directions are determined. In the spatial normal plane... Inside, using the coordinates of the nail entry point Starting from a point, generate one or more candidate straight paths that meet the following condition: the candidate straight paths intersect at a point. The shortest distance is equal to the preset offset distance .

[0065] In a specific embodiment, if the nail entry point to the intersection The distance is greater than the preset offset distance Then there exist two candidate straight paths that satisfy the conditions (i.e., with...). Center of the circle (Two tangents to a circle with radius ). At this point, based on the direction of the insertion of the blocking screw (e.g., from the outside to the inside), select the candidate straight line whose tangential direction points inward to the tibia and facilitates surgical operation.

[0066] The unit direction vector of the candidate path line is used to determine the path direction of the initial path parameters. .

[0067] It's important to understand that by selecting a "high-holding-force core area," the entry point is locked in the region with the highest bone density and best connectivity, solving the problem that relying solely on geometric planning can easily lead to loosening or cutting of the blocking screw. Simultaneously, geometric constraints (tangential distance conditions) ensure that the path direction accurately guides the blocking screw to contact the intramedullary nail, achieving a unity of "fixed-point strong holding" and "precise blocking and repositioning."

[0068] Furthermore, existing technologies typically require the entry point to be directly aligned with the intramedullary nail axis for effective blocking. This invention, by introducing generation logic for a "preset offset distance" and "candidate path straight line," ensures effective contact of the blocking nail even when there is a certain distance deviation between the high-density area of ​​the cortical bone (entry point) and the intramedullary nail axis. This is achieved by adjusting the implantation angle (generating a tangential path). This significantly improves the algorithm's adaptability and robustness when facing irregular fracture surfaces or eccentric medullary cavities.

[0069] Simultaneously, the calculation of the initial path parameters integrates three-dimensional anatomical features and the intramedullary nail model, generating an initial solution containing precise nail entry point coordinates and the optimal direction vector. This high-precision initial solution serves as the starting point for subsequent intraoperative real-time registration and dynamic optimization, significantly reducing iterative calculation time, improving the response speed of the surgical navigation system, and ensuring reliable guidance suggestions in real time in complex dynamic surgical environments.

[0070] The third step in the intramedullary nail blocking pin positioning and guidance method for distal tibial fractures is to acquire real-time two-dimensional X-ray image data during fracture reduction and intramedullary nail insertion. Using an image registration algorithm, the transformation matrices of the bone blocks and the intramedullary nail relative to the real-time two-dimensional X-ray image data are calculated, respectively. This step establishes a correlation between the preoperative three-dimensional model and the actual intraoperative state, enabling real-time perception of changes in fracture reduction and intramedullary nail position, overcoming the problem that traditional methods rely on preoperative static planning and cannot adapt to intraoperative changes.

[0071] This embodiment provides the third step of the intramedullary nail blocking screw positioning and guiding method for distal tibial fractures, which specifically includes the following sub-steps:

[0072] S301: Obtain pre-labeled fracture line feature parameters, and segment the tibial three-dimensional point cloud data into proximal bone block point cloud data and distal bone block point cloud data based on the fracture line feature parameters.

[0073] Specifically, during the preoperative planning phase, a surface mesh model of the tibia was constructed based on high-resolution CT data, and fracture line feature parameters were manually or automatically labeled. These feature parameters include, but are not limited to, the spatial plane equation of the fracture surface (ax+by+cz+d=0) or the closed three-dimensional contour curve of the fracture line.

[0074] Based on the aforementioned feature parameters, a spatial geometric segmentation algorithm (such as a half-space cutting algorithm or ray intersection method) is used to process the complete tibial 3D point cloud data. The system logically divides the point cloud data into two independent rigid body parts according to the fracture line: proximal bone block point cloud data, defined as a set of point clouds containing the tibial plateau and the bone shaft above it. Distal bone fragment point cloud data: defined as a set of point clouds containing the medial malleolus, distal metaphysis, and the underlying diaphysis. .

[0075] After the segmentation is completed, they are respectively and Establish an independent local coordinate system so that its motion can be tracked independently in subsequent steps.

[0076] S302: Perform edge detection on real-time two-dimensional X-ray image data to extract the actual proximal bone block contour, the actual distal bone block contour, and the actual intramedullary nail contour, respectively.

[0077] Real-time two-dimensional X-ray digital images (typically in DICOM format) are acquired by connecting to an intraoperative C-arm X-ray machine. To accurately extract the target contour against complex intraoperative backgrounds, the following image processing workflow is implemented:

[0078] First, a dynamic region of interest (ROI) is defined. To improve processing speed and suppress noise, the 3D bounding box of the proximal bone block, distal bone block, and intramedullary nail is projected onto the current 2D image plane using the transformation matrix calculated in the previous time step (or the initial preoperative pose). Based on the projected bounding box, a preset margin is added outward (e.g., 20 pixels or 10 mm corresponding to the actual physical size), generating three independent dynamic ROI regions. This dynamic update mechanism ensures that the target is completely contained within the detection window even when the bone moves rapidly.

[0079] Next, image enhancement and edge extraction are performed. Image preprocessing: Contrast-limited adaptive histogram equalization (CLAHE) is applied to the image pixels within each ROI to enhance the contrast of soft tissue and bone; subsequently, a Gaussian filter kernel (kernel size, for example...) is used. or Standard deviation Smoothing and denoising are performed. Edge detection: The Canny edge detection operator is applied. A high-low threshold ratio is set (e.g., high:low = 2:1) for hysteresis thresholding to extract edge pixels.

[0080] For the actual proximal and distal bone block contours, within their respective ROIs, edge curves with continuous gradient magnitudes and lengths greater than a preset length threshold (e.g., 50 pixels) are extracted, and scattered noise is removed to obtain the contour pixel set. and .

[0081] For the actual intramedullary nail contour, due to the high-density metallic material, the nail exhibits extremely high brightness and clear boundaries under X-rays. The Otsu's maximum inter-class variance method is used to automatically calculate the segmentation threshold, extract the bright areas, and calculate their gradient extrema as precise edges, thus obtaining the contour pixel set. .

[0082] S303: Using a 2D-3D projection matching algorithm, calculate the spatial pose deviation between the proximal bone block point cloud data, the distal bone block point cloud data, and the intramedullary nail 3D data under virtual projection and the corresponding actual contours. Independently output the proximal bone block transformation matrix and the distal bone block transformation matrix, which together constitute the bone block transformation matrix, and output the intramedullary nail transformation matrix.

[0083] This step iteratively optimizes the pose of the 3D model in the 2D image plane. The specific implementation method is as follows:

[0084] First, a virtual projection and similarity metric function are constructed. Using calibrated C-arm X-ray imaging parameters (including source-to-detector distance SID, detector pixel size, X-ray source focal coordinates, etc.), a digitally reconstructed radiographic (DRR) model is built. For the target to be registered (proximal, distal, or intramedullary nail), its 3D point cloud or mesh model is projected onto the virtual imaging plane using a ray projection algorithm based on the current assumed transformation matrix T (containing rotation matrix R and translation vector t), generating a simulated contour (i.e., a digitally synthesized image). Spatial pose deviation is calculated using a distance-transform-based similarity metric. The actual contour pixel set extracted in step S302 (…) is then processed. , or Construct a 2D distance transformation map, where the value of each pixel represents the Euclidean distance from that point to the nearest actual contour edge. Calculate the sum of the distance values ​​of all sampled points on the virtual projected contour in the distance transformation map, using this as the objective function to measure spatial pose deviation. ,in For the coordinates of the virtual projection point, This is the distance transformation value.

[0085] Secondly, multi-resolution pyramid optimization is performed. To avoid getting trapped in local optima and improve convergence speed, an image pyramid (e.g., 3 layers) is constructed. Coarse registration is performed on the lowest resolution (top layer) image, using the pose result of the previous frame as an initial guess. The Levenberg-Marquardt (LM) nonlinear least squares optimization algorithm is used to iteratively update the transformation parameter T, minimizing the objective function E(T). The optimization result of the current layer is passed up as the initial value for the next layer (higher resolution) until the optimum is reached on the original resolution image.

[0086] Finally, the transformation matrix is ​​output. When the descent gradient of the objective function is less than a preset threshold (e.g., ...), ... Convergence is determined when the number of iterations reaches the upper limit and the reprojection error (RMS) is less than a preset accuracy threshold (e.g., less than 0.5 pixels). The proximal bone block transformation matrix corresponding to the proximal bone block is output. And the distal bone block transformation matrix corresponding to the distal bone block. The set of both is defined as the bone block transformation matrix. The intramedullary nail transformation matrix corresponding to the intramedullary nail is output independently. The above matrices are all... The homogeneous transformation matrix precisely describes the translation components of each target relative to the X-ray imaging coordinate system (or the robot's reference coordinate system). and rotational components .

[0087] This embodiment segments the proximal and distal bone blocks of the tibia using three-dimensional point clouds based on fracture line features. A 2D-3D projection matching algorithm is then used to calculate the spatial pose changes of the bone blocks and intramedullary nail relative to intraoperative X-ray images, enabling independent dynamic acquisition of the fracture area and the implanted instrument's status. Compared to traditional overall registration methods, this approach can distinguish the motion changes of different bone blocks and establish dynamic transformation relationships between bone blocks and intramedullary nails, improving the accuracy of intraoperative status perception and providing an accurate data foundation for subsequent dynamic adjustment of the blocking nail path.

[0088] The fourth step in the intramedullary nail blocking pin positioning and guidance method for distal tibial fractures is as follows: Based on the bone block transformation matrix, coordinate transformation is performed on the initial path parameters, cortical bone inner wall point cloud data, and target point cloud subset to obtain dynamic path parameters, dynamic cortical bone inner wall point cloud data, and dynamic target point cloud subset; based on the intramedullary nail transformation matrix, coordinate transformation is performed on the intramedullary nail three-dimensional data to obtain dynamic intramedullary nail data; this step dynamically updates the spatial information on which the blocking pin positioning is based, enabling the path planning to adapt to changes in intraoperative anatomical state, realizing the transformation from static planning to dynamic adjustment, and reducing positioning errors caused by position changes.

[0089] This embodiment details the fourth step of the intramedullary nail blocking screw positioning and guiding method for distal tibial fractures, specifically including the following sub-steps:

[0090] S401: Based on the bone block transformation matrix, coordinate transformation is performed on the initial path parameters, the cortical bone inner wall point cloud data, and the target point cloud subset to obtain the dynamic path parameters, dynamic cortical bone inner wall point cloud data, and dynamic target point cloud subset; based on the intramedullary nail transformation matrix, coordinate transformation is performed on the intramedullary nail three-dimensional data to obtain the dynamic intramedullary nail data.

[0091] First, obtain the data describing the real-time pose of the bone and implant output in step S303. Specifically, this includes: a proximal bone block transformation matrix: describing the rigid body transformation of the proximal bone block from the preoperative coordinate system to the intraoperative coordinate system; a distal bone block transformation matrix: describing the rigid body transformation of the distal bone block from the preoperative coordinate system to the intraoperative coordinate system; and an intramedullary nail transformation matrix: describing the rigid body transformation of the intramedullary nail from the preoperative coordinate system to the intraoperative coordinate system.

[0092] All of the above matrices are The homogeneous transformation matrix of is given by the standard mathematical form as follows: Among them, the upper left corner The submatrix R is the rotation matrix, and the column vector t on the right is the translation vector.

[0093] S402: Calculate the shortest spatial distance between the initial path parameter's entry point coordinates and the proximal bone block point cloud data and the distal bone block point cloud data, respectively, and determine the bone block point cloud data with the smaller shortest spatial distance as the target bone block to which the entry point coordinates belong.

[0094] To determine which bone segment the blocking pin should follow, the system uses a nearest neighbor search algorithm.

[0095] First, using the KD-Tree (K-Dimensional Tree) spatial index data structure, the preoperative proximal bone fragment point cloud data were analyzed. and remote bone block point cloud data Build a topological index to accelerate spatial queries.

[0096] Secondly, extract the coordinates of the entry point from the initial path parameters. Calculate separately Euclidean distance to the two point cloud sets: ; .

[0097] Finally, the magnitudes of the two distances are compared and combined with a preset distance difference threshold. (For example Perform robustness assessment: if The insertion point was determined to be a proximal bone fragment; if The insertion point is determined to be a distal bone fragment; if If the distance between two points is too close (which usually occurs when the insertion point is exactly at the edge of the fracture line), then the determination is made according to the pre-defined anatomical rules. Since the blocking screw for distal tibial fractures is mainly used to stabilize the distal bone fragment, the target bone fragment is determined by default to be the distal bone fragment.

[0098] S403: When the target bone block is a proximal bone block point cloud data, extract the translation and rotation matrices of the proximal bone block transformation matrix, update the pin entry point coordinates using the translation and rotation matrices, and independently update the path direction using the rotation matrix.

[0099] When the target bone block is identified as a proximal bone block, the proximal bone block transformation matrix is ​​extracted. rotation matrix in Translation vector .

[0100] Use rigid body transformation formulas to determine the initial nail entry point coordinates Mapped to dynamic nail entry point coordinates : .

[0101] Due to path direction This is a unit direction vector, whose characteristic is that it changes only with the rotation of the rigid body, and not with translation. It is used using a rotation matrix. Perform a linear transformation on it: To ensure the accuracy of subsequent angle calculations, the transformed vector... Normalization is performed: .

[0102] S404: When the target bone block is the point cloud data of the distal bone block, extract the translation matrix and rotation matrix of the transformation matrix of the distal bone block, update the coordinates of the nail entry point using the translation matrix and rotation matrix, and update the path direction independently using the rotation matrix.

[0103] When the target bone block is identified as a distal bone block, the system extracts the distal bone block transformation matrix. rotation matrix in Translation vector Perform coordinate transformation according to the exact same mathematical logic as step S403: ; And perform normalization processing in the same way.

[0104] S405: Recombine the updated entry point coordinates with the updated path direction to obtain dynamic path parameters.

[0105] The calculated dynamic nail entry point coordinates With dynamic path direction The combination constitutes the dynamic path parameters of the blocking pin at this moment, which is used to drive the real-time display of the surgical navigation system or the motion planning of the robotic arm.

[0106] Simultaneously, perform batch data transformation to update the global view: utilizing a defined target bone block transformation matrix (e.g., the distal bone block transformation matrix). The preoperatively extracted cortical bone inner wall point cloud data and target point cloud subset are traversed and processed. That is, each point in the point cloud set is processed... , apply formula This allows for the acquisition of dynamic cortical bone wall point cloud data and a dynamic target point cloud subset. The intramedullary nail transformation matrix is ​​then utilized. The rigid body transformation described above is performed on the vertex data of the three-dimensional model of the intramedullary nail to obtain dynamic intramedullary nail data.

[0107] At this point, all key geometric elements of the preoperative planning have been synchronized in real time to the current surgical space coordinate system.

[0108] By calculating the spatial distance between the initial nail insertion point and different bone segments, the target bone segment to which the nail insertion point belongs is determined. The coordinates and path direction of the nail insertion point are then updated according to the transformation matrix of the corresponding bone segment, enabling dynamic adjustment of the blocking nail path parameters as the fracture reduction status changes. This method avoids path update errors caused by treating the entire bone structure as a rigid whole, allowing for precise correction of the blocking nail positioning to the target bone segment that has actually shifted. This improves the consistency between the dynamic path parameters and the actual anatomical state during surgery, enhancing the adaptability of the positioning guidance.

[0109] The fifth step in the intramedullary nail blocking screw positioning and guidance method for distal tibial fractures is as follows: Based on dynamic path parameters, dynamic intramedullary nail data, and dynamic cortical bone inner wall point cloud data, calculate the minimum gap between the blocking screw path and the intramedullary nail, the safe distance between the blocking screw path and the cortical bone inner wall, and the angle between the blocking screw path and the intramedullary nail to obtain spatial relationship parameters; specifically including the following sub-steps:

[0110] S501: Parse the dynamic path parameters to obtain the coordinates of the dynamic nail entry point and the dynamic path direction, and generate a cylindrical virtual bounding box for the blocking nail along the dynamic path direction, starting from the coordinates of the dynamic nail entry point.

[0111] First, analyze the dynamic path parameters output in step S405 and extract the coordinates of the dynamic nail entry point. and dynamic path direction unit vector Secondly, based on the type and specifications of the blocking staples used in the actual surgery, the geometric dimensions of the virtual bounding box are determined. Specifically, this includes: radius parameter: set as the nominal radius of the blocking staples. (For example, a commonly used 4.0mm diameter corresponds to a 2.0mm radius). Length parameter: Set as the effective working length of the stop pin. This length is typically determined based on preoperative measurements or pre-defined general values ​​(e.g., 30mm to 50mm), ensuring the cylinder completely penetrates the cortical bone and extends into the medullary canal to interact with the intramedullary nail. Finally, with Centered on the base of the cylinder, with Using the central axis direction, construct a radius of [missing information]. Height is A cylindrical mesh model is used as a virtual bounding box for the blocking pins.

[0112] S502: Calculate the shortest spatial distance between the three-dimensional outer surface of the virtual bounding box of the blocking nail and the three-dimensional outer surface of the dynamic intramedullary nail data, and use it as the minimum gap.

[0113] To accurately assess the contact or proximity between the blocking nail and the intramedullary nail, a surface distance calculation method based on discrete sampling is used:

[0114] First, the cylindrical surface and bottom surface of the virtual bounding box of the blocking pin are discretized into a mesh. Then, a uniformly distributed set of surface sampling points is generated according to a preset sampling density (e.g., one point every 10 degrees along the circumference and one point every 1 mm along the axis). .

[0115] Secondly, the KD-Tree (K-Dimensional Tree) spatial indexing algorithm is used to preprocess the vertices or surfaces of the dynamic intramedullary nail data (usually an STL triangular mesh model) to construct a fast query structure.

[0116] Then, traverse the set of sampling points. Each point in KD-Tree was used to search for its nearest neighbor on the surface of dynamic intramedullary nail data. And calculate the Euclidean distance between the two points. .

[0117] Finally, take the minimum value among all calculated distances, that is... This is used as the minimum gap between the blocking nail path and the intramedullary nail. This parameter reflects the potential support or interference of the blocking nail with the intramedullary nail.

[0118] S503: Calculate the shortest spatial distance between the three-dimensional outer surface of the virtual bounding box of the blocking nail and each point in the dynamic cortical bone inner wall point cloud data, and use it as the safety distance.

[0119] To assess whether the blocking screw will laterally breach the cortical bone boundary as it travels through the medullary canal, a minimum distance algorithm from the point cloud to the mesh was used:

[0120] First, the dynamic cortical bone wall point cloud data obtained in step S405 is processed using KD-Tree. Build a spatial index.

[0121] Secondly, extract the core working segment (i.e., the outer edge of the nail insertion and exit sides minus the preset thickness) of the virtual bounding box of the blocking nail generated in step S502 located inside the medullary cavity, and generate a set of sampling points on the surface of the intramedullary segment. .

[0122] Then, iterate through... Each point in Dot cloud on the dynamic cortical bone inner wall Search for its nearest neighbor bone wall point And calculate the Euclidean distance. .

[0123] Finally, take the minimum value among all calculated distances, that is... This distance is used as a safety distance between the blocking pin path and the inner wall of the cortical bone. If this distance is less than 0 or less than a preset safety threshold (e.g., 1 mm), it indicates an extremely high risk of unexpected perforation of the cortical bone.

[0124] S504: Extract the central axis vector of the virtual bounding box of the blocking nail and the central axis vector of the dynamic intramedullary nail data, calculate the intersection angle of the two central axis vectors in three-dimensional space, and use it as the included angle.

[0125] To evaluate the mechanical rationality of the insertion direction of the blocking pin, the spatial angle between the two axes is calculated:

[0126] First, the central axis vector of the virtual bounding box of the blocking pin is the dynamic path direction. .

[0127] Secondly, obtain the central axis vector of the dynamic intramedullary nail data. This vector can be read directly from the intramedullary nail model transformed in step S405, or obtained by fitting the cylindrical surface of the intramedullary nail using the RANSAC algorithm.

[0128] Finally, the angle between the two unit vectors is calculated using the dot product formula. . ; .

[0129] To ensure the stability of numerical calculations and prevent errors caused by floating-point arithmetic errors leading to the dot product result slightly exceeding the domain [-1.0, 1.0], the dot product result is clamped before performing the inverse cosine operation: ; Calculated angle (Unit: degrees) is the angle between the blocking nail path and the intramedullary nail.

[0130] S505: Combine the calculated minimum gap, safety distance, and included angle to form spatial relationship parameters.

[0131] The above three key geometric indicators, namely minimum gap Safe distance and included angle Combined into a multidimensional feature vector or structure, denoted as spatial relation parameters. . This parameter set will be output to the navigation system's interactive interface to display the current path status in real time, or as input constraints for subsequent steps (such as path optimization and adjustment).

[0132] By calculating the gap, distance, and angle between the blocking nail path and the intramedullary nail and cortical bone, the safety of the path and the blocking effect can be quantitatively evaluated, which can identify the risk of spatial interference and bone damage in advance and improve the safety of implantation.

[0133] The sixth step in the intramedullary nail blocking screw positioning and guidance method for distal tibial fractures is as follows: When the spatial relationship parameters do not meet the preset safety threshold condition, the coordinate set of the dynamic target point cloud subset is used as the search range, and the path optimization algorithm is used to iteratively optimize the dynamic path parameters by using the subset of dynamic target point clouds containing the points and whose spatial relationship parameters meet the preset safety threshold condition as a double constraint, and outputting the final path parameters. Specifically, this includes the following sub-steps:

[0134] S601: When the spatial relationship parameters do not meet the preset safety threshold conditions, the coordinate set of the dynamic target point cloud subset is used as the search range, and the path optimization algorithm is run to iteratively optimize the dynamic path parameters with the constraints of the spatial relationship parameters being contained in the dynamic target point cloud subset and meeting the preset safety threshold conditions, and the final path parameters are output.

[0135] First, read the current spatial relationship parameters calculated in step S505. The system has a preset set of safety threshold conditions. If the current... If any of the indicators fails to meet the standard, the system will immediately trigger the path optimization mode.

[0136] This optimization mode follows the "double constraint" principle: the first constraint is a geometric position constraint: the new entry point must be located in the dynamic target point cloud subset updated in step S405. Within the coordinate range, that is, on the cortical bone surface located in the high holding force core area. The second constraint is a safety performance constraint: the generated path must ensure that its corresponding spatial relationship parameters satisfy all safety threshold conditions defined later in this embodiment.

[0137] S602: Within the coordinate set of the dynamic target point cloud subset, a three-dimensional mesh is generated according to the preset spatial sampling step size. Each node of the three-dimensional mesh is used as the coordinate of the candidate entry point, and multiple candidate path directions are generated at each candidate entry point coordinate, which are combined to form multiple candidate path parameters.

[0138] To find the optimal solution within a limited time, a discretized sampling strategy is used to construct a candidate path set: First, a dynamic target point cloud subset is calculated. The axis-aligned bounding box (AABB) is then used. Next, within the space defined by the AABB, sampling is performed according to a preset spatial step size. (For example Construct a uniform 3D mesh. Utilize a KD-Tree to accelerate the search and eliminate points that do not belong to the mesh. The nodes (i.e., those locations without bone tissue support) and the remaining valid mesh nodes constitute the set of candidate nail insertion point coordinates. .

[0139] Then, for each candidate entry point The principal component analysis (PCA) algorithm is used to estimate the normal vector of the local surface where the point is located. Based on this, a set of candidate path directions is generated: calculation point Connect the two points closest to the central axis of the dynamic intramedullary nail to obtain the reference vector pointing towards the intramedullary nail. .by To guide the direction, in the following With axis as the axis and half-cone angle as the ... is as the half-con (For example Within the conical range of ), according to a preset angular resolution (e.g., per interval) Spatial sampling is performed. Simultaneously, samples with excessively large angles to the bone surface normal (e.g., greater than...) are discarded. (i.e., the direction in which the nail is inserted at an excessively tilted angle).

[0140] Finally, the coordinates of the candidate entry points are combined with the filtered candidate direction vectors to form a set of candidate path parameters. .

[0141] S603: Calculate the spatial relationship parameters corresponding to each candidate path parameter one by one, eliminate candidate path parameters whose spatial relationship parameters do not meet the preset safety threshold conditions, and retain the set of valid path parameters.

[0142] Using the methods in steps S501 to S504, virtual construction and geometric analysis are performed on each candidate path parameter, and its corresponding... (Minimum gap) (Safe distance) and (Angle), and filter according to the following preset safety threshold conditions:

[0143] Minimum clearance constraint (contact effectiveness): must satisfy .in, This is the spatial clearance distance threshold (usually a negative value or 0, such as -1.0 mm, indicating that a slight volumetric penetration between the blocking nail model and the intramedullary nail model is allowed to simulate physical contact). This is the maximum effective blocking gap threshold (e.g., 2.0 mm). This condition ensures that the blocking pin is tightly attached to the intramedullary nail to exert the "Bourdon" blocking effect and prevents the blocking from failing due to excessive distance between them.

[0144] Safety distance constraints (cortical bone integrity): must meet .in, This is the safe margin threshold for cortical bone (e.g., 1.0 mm). This condition ensures that the blocking screw will not penetrate the inner wall of the cortical bone, thus preventing the fracture fragments from splitting.

[0145] Angle constraint (effectiveness of mechanical support): must meet the following requirements. .in, For the range of mechanical blocking angles (e.g.) This condition restricts the implantation direction to be as perpendicular as possible to the intramedullary nail axis (i.e., the path direction is approximately orthogonal to the intramedullary nail axis vector) in order to maximize the ability to resist lateral and angular displacement of the intramedullary nail relative to the inner wall of the cortical bone.

[0146] The system eliminates candidate paths that do not meet any of the above conditions and retains all those that meet the conditions as the set of valid path parameters.

[0147] S604: Extract the bone density and cortical bone thickness corresponding to the coordinates of each candidate nail entry point in the effective path parameter set, calculate the weighted score of bone density and cortical bone thickness, and take the effective path parameter with the highest weighted score as the final path parameter.

[0148] Among the effective pathways that meet safety requirements, further optimization is performed based on biomechanical performance:

[0149] For each valid path parameter, the system uses spatial mapping technology to query the corresponding voxel attributes of the nail insertion point coordinates in the preoperative CT data and obtain the bone mineral density at that point. (Usually based on HU value conversion) and the corresponding cortical bone thickness (Determined by calculating the distance from the point to the inner wall of the contralateral cortical bone).

[0150] The weighted summation method is used to calculate the comprehensive score. : ;in, and The maximum value is used as a reference (for normalization, to ensure the score falls between 0 and 1). and For example, preset weighting coefficients (Focusing on bone mineral density) (Focusing on cortical bone thickness).

[0151] The system calculates the scores of all valid paths and selects the scores. Mark the largest path parameter as the final path parameter. .

[0152] S605: Based on the coordinate data of the final path parameters, a virtual entity image of the blocking nail is simultaneously rendered and superimposed on the three-dimensional anatomical image and real-time two-dimensional X-ray image data output by the display device.

[0153] To achieve intuitive intraoperative navigation, the system performs augmented reality (AR) rendering: first, using the final path parameters... First, construct a three-dimensional virtual solid model (cylindrical model) of the blocking nail. Then, in the three-dimensional anatomical view of the display device, using a virtual reality rendering engine (such as OpenGL or VTK), overlay and blend this virtual solid model with a semi-transparent highlight color (such as fluorescent green) onto the real-time reconstructed three-dimensional bone model, and display its relative position to the intramedullary nail and bone wall.

[0154] Meanwhile, in the real-time two-dimensional X-ray image (fluoroscopic anterior and lateral views) window of the display device, digital reconstructed radiography (DRR) projection technology is used to calculate the projection contour of the virtual entity model on the imaging plane by combining the current imaging parameters of the C-arm. This contour is then superimposed on the real-time X-ray video stream with high-contrast lines (such as red solid lines) to achieve "virtual and real on the same screen" display, which helps doctors to intuitively confirm the nail insertion point and direction under two-dimensional fluoroscopy.

[0155] S606: Transform the final path parameters to the base coordinate system of the external surgical robot, generate and output control guidance commands for driving the end effector of the external surgical robot to the corresponding spatial pose.

[0156] To achieve surgical automation or robot-assisted operation, the system performs coordinate transformation and command generation:

[0157] First, retrieve the hand-eye calibration matrix that was pre-calibrated. This matrix precisely characterizes the rigid body transformation relationship (including rotation and translation) between the intraoperative C-arm / navigation camera coordinate system (usually defined as the world coordinate system) and the surgical robot base coordinate system.

[0158] Secondly, the coordinates of the nail entry point in the final path parameters and path direction vector Transformation from image coordinate system to robot base coordinate system: ; .

[0159] Finally, using the robot's inverse kinematics algorithm, based on and The required rotation angles for each joint of the robot or the target pose of the end effector are calculated. A model containing Cartesian coordinates (x, y, z) and a pose description (such as rotation quaternions) is generated. Control guidance commands (or Euler angles) are sent in real time to the controller of the external surgical robot via a standard communication interface (such as TCP / IP protocol or ROS message packets) to drive the robot's end effector (such as the drill sleeve held by the robotic arm) to move precisely and lock into the implantation pose of the stop screw.

[0160] By combining dynamic target point cloud range, safety threshold constraints, and bone quality scores for path optimization, automatic optimization of the blocking screw path is achieved. At the same time, through virtual image overlay and robot-guided output, the optimized path is visualized and precisely executed, improving positioning accuracy, surgical efficiency, and fixation reliability.

[0161] like Figure 3 The diagram shown is a structural schematic of an intramedullary nail blocking screw positioning and guiding system for distal tibial fractures provided in an embodiment of this application. (Refer to...) Figure 3 The system includes: a tibial target area extraction module, an initial blocking planning module, a bone screw pose registration module, a bone screw dynamic mapping module, a blocking interference analysis module, and a blocking guidance optimization module.

[0162] Specifically, the tibial target area extraction module is used to acquire preoperative CT image data of the distal tibia, reconstruct three-dimensional point cloud data of the tibia based on the preoperative CT image data, and extract point cloud data of the inner wall of the cortical bone; calculate bone density based on the preoperative CT image data, calculate cortical bone thickness based on the point cloud data of the inner wall of the cortical bone, and extract three-dimensional point cloud data that meet the preset conditions for bone density and cortical bone thickness to form a subset of target point clouds.

[0163] The initial blocking planning module is used to acquire preset three-dimensional data of intramedullary nails. Within the coordinate range of the target point cloud subset, it calculates and generates the initial path parameters of the blocking nails in combination with the three-dimensional data of the intramedullary nails. The initial path parameters include at least the coordinates of the nail entry point and the path direction.

[0164] The bone nail pose registration module is used to acquire real-time two-dimensional X-ray image data during fracture reduction and intramedullary nail insertion. Using the image registration algorithm, the transformation matrix of the bone block and the transformation matrix of the intramedullary nail relative to the real-time two-dimensional X-ray image data are calculated respectively.

[0165] The bone nail dynamic mapping module is used to perform coordinate transformation on the initial path parameters, the cortical bone inner wall point cloud data, and the target point cloud subset based on the bone block transformation matrix, thereby obtaining the dynamic path parameters, dynamic cortical bone inner wall point cloud data, and dynamic target point cloud subset; and to perform coordinate transformation on the intramedullary nail three-dimensional data based on the intramedullary nail transformation matrix to obtain the dynamic intramedullary nail data.

[0166] The blocking interference analysis module is used to calculate the minimum gap between the blocking nail path and the intramedullary nail, the safe distance between the blocking nail path and the inner wall of the cortical bone, and the angle between the blocking nail path and the intramedullary nail based on dynamic path parameters, dynamic intramedullary nail data, and dynamic cortical bone inner wall point cloud data, thereby obtaining spatial relationship parameters.

[0167] The obstacle-guided optimization module is used to perform iterative optimization of the dynamic path parameters by using the coordinate set of a subset of dynamic target point clouds as the search range when the spatial relationship parameters do not meet the preset safety threshold conditions. It also uses the dual constraints of the subset of dynamic target point clouds containing the spatial relationship parameters that meet the preset safety threshold conditions as the running path optimization algorithm to output the final path parameters.

[0168] In summary, this invention achieves a transformation from traditional static planning to dynamic intelligent positioning by constructing a blocking screw positioning and guidance mechanism that integrates bone structural features, bone quality parameters, and intraoperative dynamic feedback. Compared to existing technologies, this invention can adjust the blocking screw positioning path in real time based on changes in fracture reduction status and the actual position of the intramedullary nail. It also optimizes the implantation plan by combining bone density, cortical bone thickness, and spatial safety constraints, thereby improving the accuracy of blocking screw positioning while enhancing the safety and fixation stability of the implantation process. This invention effectively reduces the risks of positioning deviations, fixation failures, and related complications caused by intraoperative anatomical changes, insufficient bone quality, and unreasonable path planning. It is particularly suitable for precise positioning and guidance of intramedullary nail blocking screws in cases of complex distal tibial fractures, osteoporosis, and comminuted fractures, and has high clinical application value.

[0169] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the above functions can be divided into different functional modules to complete all or part of the functions described above.

[0170] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0171] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units, located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0172] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0173] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the solution, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0174] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures, characterized in that, Includes the following steps: Preoperative CT image data of the distal tibia is acquired, and three-dimensional point cloud data of the tibia is reconstructed based on the preoperative CT image data. Point cloud data of the inner wall of the cortical bone is extracted. Bone density is calculated based on the preoperative CT image data, and cortical bone thickness is calculated based on the point cloud data of the inner wall of the cortical bone. Three-dimensional point cloud data that meet the preset conditions of bone density and cortical bone thickness are extracted to form a target point cloud subset. Acquire preset three-dimensional data of intramedullary nails, and within the coordinate range of a subset of target point clouds, calculate and generate initial path parameters for the blocking nail based on the three-dimensional data of the intramedullary nails. The initial path parameters include at least the coordinates of the nail entry point and the path direction. Real-time two-dimensional X-ray image data were acquired during fracture reduction and intramedullary nail insertion. Using an image registration algorithm, the transformation matrices of the tibial three-dimensional point cloud data and the intramedullary nail three-dimensional data relative to the real-time two-dimensional X-ray image data were calculated, and the transformation matrices of the intramedullary nail were calculated, respectively. Based on the bone block transformation matrix, coordinate transformation is performed on the initial path parameters, the cortical bone inner wall point cloud data, and the target point cloud subset to obtain the dynamic path parameters, dynamic cortical bone inner wall point cloud data, and dynamic target point cloud subset; based on the intramedullary nail transformation matrix, coordinate transformation is performed on the intramedullary nail three-dimensional data to obtain the dynamic intramedullary nail data. Based on the dynamic path parameters, dynamic intramedullary nail data, and dynamic cortical bone inner wall point cloud data, the minimum gap between the blocking nail path and the intramedullary nail, the safe distance between the blocking nail path and the cortical bone inner wall, and the angle between the blocking nail path and the intramedullary nail are calculated to obtain the spatial relationship parameters. When the spatial relationship parameters do not meet the preset safety threshold conditions, the coordinate set of the dynamic target point cloud subset is used as the search range, and the path optimization algorithm is used as a double constraint that the spatial relationship parameters are contained in the dynamic target point cloud subset and meet the preset safety threshold conditions. The algorithm iteratively optimizes the dynamic path parameters and outputs the final path parameters.

2. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 1, characterized in that, The calculation of bone density based on preoperative CT image data and the calculation of cortical bone thickness based on cortical bone inner wall point cloud data specifically include: The gray values ​​of each voxel in the preoperative CT image data were determined as Huntsfield unit gray values, and the Huntsfield unit gray values ​​were spatially mapped to the corresponding three-dimensional coordinate points of the tibial three-dimensional point cloud data. The mapped gray values ​​were then converted into bone density at the corresponding coordinate points. Generate the cortical bone outer surface point cloud data corresponding to the tibial 3D point cloud data. Generate a spatial ray along the normal vector direction of each data point in the cortical bone inner wall point cloud data. Calculate the spatial intersection coordinates of the spatial ray and the cortical bone outer surface point cloud data respectively, and take the Euclidean distance between the data point and the corresponding spatial intersection coordinates as the cortical bone thickness of each data point.

3. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 2, characterized in that, The initial path parameters for generating the blocking nail by combining the three-dimensional data of the intramedullary nail specifically include: Extract the central axis of the three-dimensional data of the intramedullary nail; Based on bone density and cortical bone thickness, connectivity analysis is performed on the target point cloud subset to screen out the largest connected region with bone density greater than the preset neighborhood mean, and it is marked as the high holding power core region. Calculate the three-dimensional geometric center of the high holding force core area, and project the three-dimensional geometric center onto the outer surface of the cortical bone. Set the surface coordinates obtained from the projection as the entry point coordinates of the initial path parameters. Construct a spatial normal plane that passes through the coordinates of the entry point and is perpendicular to the central axis, and determine the intersection point of the central axis and the spatial normal plane; generate candidate path lines that pass through the coordinates of the entry point in the spatial normal plane, and the shortest distance between the candidate path lines and the intersection point is equal to a preset offset distance; use the direction vector of the candidate path lines as the path direction of the initial path parameters.

4. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 1, characterized in that, The calculation of the bone block transformation matrix and intramedullary nail transformation matrix relative to the real-time two-dimensional X-ray image data of the tibial three-dimensional point cloud data and intramedullary nail three-dimensional data specifically includes: Obtain pre-labeled fracture line feature parameters, and segment the tibial three-dimensional point cloud data into proximal bone block point cloud data and distal bone block point cloud data based on the fracture line feature parameters; Edge detection was performed on real-time two-dimensional X-ray image data to extract the actual proximal bone block contour, the actual distal bone block contour, and the actual intramedullary nail contour, respectively. Using a 2D-3D projection matching algorithm, the spatial pose deviations between the proximal bone block point cloud data, the distal bone block point cloud data, and the intramedullary nail 3D data under virtual projection and their corresponding actual contours are calculated. The proximal bone block transformation matrix and the distal bone block transformation matrix, which contain translation and rotation components, are output independently to form the bone block transformation matrix, and the intramedullary nail transformation matrix is ​​also output.

5. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 4, characterized in that, The specific steps for obtaining the dynamic path parameters are as follows: Calculate the shortest spatial distance between the initial path parameters, the pin insertion point coordinates, and the proximal and distal bone block point cloud data respectively. The bone block point cloud data with the smaller shortest spatial distance is identified as the target bone block to which the pin insertion point coordinates belong. When the target bone block is a proximal bone block point cloud data, the translation and rotation matrices of the proximal bone block transformation matrix are extracted. The coordinates of the nail insertion point are updated using the translation and rotation matrices, and the path direction is updated independently using the rotation matrix. When the target bone block is the point cloud data of the distal bone block, the translation matrix and rotation matrix of the transformation matrix of the distal bone block are extracted, the coordinates of the nail entry point are updated using the translation matrix and rotation matrix, and the path direction is updated independently using the rotation matrix. The updated entry point coordinates and the updated path direction are recombined to obtain dynamic path parameters.

6. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 1, characterized in that, The calculation of the minimum gap between the blocking nail path and the intramedullary nail, the safe distance between the blocking nail path and the inner wall of the cortical bone, and the angle between the blocking nail path and the intramedullary nail specifically includes: The dynamic path parameters are parsed to obtain the coordinates of the dynamic nail entry point and the dynamic path direction. A cylindrical virtual bounding box for the blocking nail is generated starting from the coordinates of the dynamic nail entry point and along the dynamic path direction. Calculate the shortest spatial distance between the three-dimensional outer surface of the virtual bounding box of the blocking nail and the three-dimensional outer surface of the dynamic intramedullary nail data, and use it as the minimum gap; Calculate the shortest spatial distance between the three-dimensional outer surface of the virtual bounding box of the blocking nail and each point in the dynamic cortical bone inner wall point cloud data, and use it as the safety distance; Extract the central axis vector of the virtual bounding box of the blocking nail and the central axis vector of the dynamic intramedullary nail data, calculate the intersection angle of the two central axis vectors in three-dimensional space, and use it as the included angle; The calculated minimum gap, safety distance, and included angle are combined to form spatial relationship parameters.

7. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 1, characterized in that, The path optimization algorithm iteratively optimizes the dynamic path parameters and outputs the final path parameters, specifically including: Within the coordinate set of the dynamic target point cloud subset, a three-dimensional mesh is generated according to the preset spatial sampling step size. Each node of the three-dimensional mesh is used as the coordinate of the candidate entry point, and multiple candidate path directions are generated at each candidate entry point coordinate, which are combined to form multiple candidate path parameters. Calculate the spatial relationship parameters corresponding to each candidate path parameter one by one, eliminate candidate path parameters whose spatial relationship parameters do not meet the preset safety threshold conditions, and retain the set of valid path parameters; Extract the bone density and cortical bone thickness corresponding to the coordinates of each candidate nail entry point in the effective path parameter set, calculate the weighted score of bone density and cortical bone thickness, and take the effective path parameter with the highest weighted score as the final path parameter.

8. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 7, characterized in that, The preset security threshold conditions specifically include: The minimum gap is greater than or equal to a preset space avoidance distance threshold, and the minimum gap is less than or equal to a preset maximum effective blocking gap threshold. The safe distance is greater than or equal to the preset cortical bone safe margin threshold; The included angle is within a preset mechanical blocking angle range to limit the lateral and angular offset of the dynamic intramedullary nail data relative to the dynamic cortical bone inner wall point cloud data.

9. The method for positioning and guiding intramedullary nail blocking screws in distal tibial fractures as described in claim 1, characterized in that, Following the output of the final path parameters, the following is also included: Based on the coordinate data of the final path parameters, a virtual entity image of the blocking nail is simultaneously rendered and superimposed on the three-dimensional anatomical image and real-time two-dimensional X-ray image data output by the display device. The final path parameters are transformed into the base coordinate system of the external surgical robot, and control guidance commands are generated and output to drive the end effector of the external surgical robot to the corresponding spatial pose.

10. A positioning and guiding system for an intramedullary nail blocking screw in distal tibial fractures, employing the positioning and guiding method for an intramedullary nail blocking screw in distal tibial fractures as described in any one of claims 1-9, characterized in that... include: Tibial target area extraction module, initial blocking planning module, bone nail pose registration module, bone nail dynamic mapping module, blocking interference analysis module, and blocking guidance optimization module; The tibial target area extraction module is used to acquire preoperative CT image data of the distal tibia, reconstruct three-dimensional point cloud data of the tibia based on the preoperative CT image data, and extract point cloud data of the inner wall of the cortical bone; calculate bone density based on the preoperative CT image data, calculate cortical bone thickness based on the inner wall point cloud data of the cortical bone, and extract three-dimensional point cloud data that meet preset conditions for bone density and cortical bone thickness to form a subset of target point clouds. The initial blocking planning module is used to acquire preset three-dimensional data of intramedullary nails, and within the coordinate range of the target point cloud subset, calculate and generate the initial path parameters of the blocking nails in combination with the three-dimensional data of the intramedullary nails. The initial path parameters include at least the coordinates of the nail entry point and the path direction. The bone nail pose registration module is used to acquire real-time two-dimensional X-ray image data during fracture reduction and intramedullary nail insertion. Using the image registration algorithm, the transformation matrix of the bone block and the transformation matrix of the intramedullary nail relative to the real-time two-dimensional X-ray image data of the tibial three-dimensional point cloud data and the intramedullary nail three-dimensional data are calculated respectively. The bone nail dynamic mapping module is used to perform coordinate transformation on the initial path parameters, the cortical bone inner wall point cloud data and the target point cloud subset based on the bone block transformation matrix, so as to obtain the dynamic path parameters, the dynamic cortical bone inner wall point cloud data and the dynamic target point cloud subset; and to perform coordinate transformation on the intramedullary nail three-dimensional data based on the intramedullary nail transformation matrix to obtain the dynamic intramedullary nail data. The blocking interference analysis module is used to calculate the minimum gap between the blocking nail path and the intramedullary nail, the safe distance between the blocking nail path and the inner wall of the cortical bone, and the angle between the blocking nail path and the intramedullary nail based on the dynamic path parameters, dynamic intramedullary nail data, and dynamic cortical bone inner wall point cloud data, so as to obtain spatial relationship parameters. The obstruction-guided optimization module is used to perform iterative optimization of the dynamic path parameters by using the coordinate set of a subset of dynamic target point clouds as the search range and the path optimization algorithm as a dual constraint that the spatial relationship parameters are contained within the subset of dynamic target point clouds and meet the preset safety threshold conditions, and outputs the final path parameters.

Images