A high tibial osteotomy osteotomy surface planning method, system and storage medium

Abstract

The application relates to the technical field of medical devices and discloses a high-position tibial osteotomy osteotomy surface planning method and system and a storage medium. The method first reconstructs a three-dimensional model based on medical images, constructs a local anatomical coordinate system and an initial osteotomy plane; then constructs an osteotomy parameter geometric constraint model, processes parameter interaction by using a linear constraint equation, realizes linkage update of an osteotomy position, depth and lateral hinge retention amount and safety threshold boundary control; further establishes a virtual distraction model, performs a lower limb mechanical force line path pre-performance, and performs instrument interference detection based on a signed distance field; and finally generates a navigation protocol. The application effectively avoids the risk of lateral hinge fracture through a parameter linkage mechanism, realizes accurate prediction of force line correction effect and active avoidance of surgical risks by combining virtual kinematics simulation, and improves the accuracy and safety of preoperative planning.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a method, system and storage medium for planning the osteotomy surface of a high tibial osteotomy. Background Technology

[0002] High tibial osteotomy (HTO) is an important knee-preserving surgery for treating medial compartment osteoarthritis of the knee and correcting varus deformity of the lower extremities. The key to the success of this surgery lies in the precise planning of the osteotomy plane, that is, by determining the position, angle, and depth of the osteotomy plane, the mechanical force line of the lower extremity is precisely adjusted to improve the joint load distribution. Therefore, the accuracy and reliability of the preoperative planning directly determine the postoperative knee joint stability and corrective effect.

[0003] Currently, conventional preoperative planning mainly relies on manual measurements from two-dimensional X-rays or CT images. This method lacks the visualization support of three-dimensional space, cannot intuitively present the spatial relationship between the osteotomy surface and the proximal tibia and distal femur, and is difficult to fully adapt to individual differences in bone morphology (such as the height difference of the tibial plateau and bone thickness) of patients. As a result, the planning scheme is highly general but not specific, and is prone to spatial planning errors.

[0004] With the development of computer-aided technology, some existing navigation systems (such as the knee joint high tibial osteotomy navigation system mentioned in existing public technologies) have introduced three-dimensional reconstruction and osteotomy line planning functions, digitizing the surgical process. However, they still have limitations in the core osteotomy surface parameter planning logic. Specifically, existing technologies usually treat the horizontal position, vertical height, osteotomy depth, and lateral hinge retention as independent parameters, without establishing a linkage mechanism based on anatomical geometry at the system level. In actual planning, when the surgeon adjusts the position or angle of the osteotomy surface, the system often cannot automatically update the osteotomy depth, which can easily lead to logical conflicts between parameters. For example, excessive osteotomy depth may damage the stability of the lateral hinge, or insufficient hinge retention may lead to breakage risks. This isolation of parameter adjustment not only forces surgeons to engage in inefficient trial and error but also increases their reliance on surgical experience.

[0005] Furthermore, existing planning techniques largely focus on static geometric segmentation, lacking real-time dynamic simulation of force line changes after osteotomy and instrument compatibility verification. Surgeons often struggle to correlate lower limb force line changes in real time during the planning phase, typically requiring postoperative or intraoperative verification of the corrective effect, leading to a disconnect between planning goals and actual needs. Simultaneously, the lack of pre-interference detection of intraoperative oscillating saw cutting paths and plate fixation positions may cause conflicts between the planned scheme and the patient's anatomical structures (such as the posterior cortex or soft tissues), increasing the risk and time cost of intraoperative adjustments. Therefore, there is an urgent need for a high tibial osteotomy planning technique capable of multi-parameter coordinated operation, dynamic force line simulation, and active safety detection. Summary of the Invention

[0006] The purpose of this invention is to provide a method, system, and storage medium for planning the osteotomy surface of a high tibial osteotomy, which solves the problems in the prior art where the adjustment of osteotomy parameters in the preoperative planning of a high tibial osteotomy lacks a linkage mechanism, resulting in low planning efficiency and easy logical conflicts, inability to preview the force line correction effect in real time, and lack of active safety detection for the lateral hinge and surgical instruments.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a method for planning the osteotomy surface of a high tibial osteotomy, comprising the following steps:

[0008] Acquire medical imaging data of the patient's knee joint, perform threshold segmentation and surface reconstruction, and generate a three-dimensional mesh model of the tibia and femur;

[0009] Key anatomical landmarks were identified on the three-dimensional mesh model, and a local anatomical coordinate system of the tibia was constructed based on the key anatomical landmarks.

[0010] Based on preset statistical rules and the key anatomical landmarks, an initial geometric model of the osteotomy plane is generated in the local anatomical coordinate system of the tibia.

[0011] Construct a geometric constraint model for osteotomy parameters, and process interactive changes of osteotomy parameters based on the geometric constraint model. When a single parameter changes, automatically calculate and update the values ​​of other related parameters.

[0012] A virtual osteotomy and expansion model is established based on the updated osteotomy parameters. The calculation and simulation of the mechanical force line path of the lower limb are performed, and the safety of the osteotomy plan is tested.

[0013] The final confirmed osteotomy planning parameter set is locked, and the osteotomy planning parameter set is subjected to inverse coordinate space mapping to generate a standard navigation protocol file and output it.

[0014] Preferably, the initial geometric model for generating the osteotomy plane specifically includes: defining the osteotomy plane using a point-normal equation, wherein the point-normal equation includes a positioning point located on the osteotomy plane and a unit normal vector perpendicular to the osteotomy plane; determining the target reference point of the lateral hinge area, obtaining the coordinate value of the lateral point of the proximal tibia in the Z-axis direction, subtracting a preset safety distance from the coordinate value to obtain the target height of the lateral hinge; constructing a virtual ray emanating from the origin along the negative X-axis in the local anatomical coordinate system of the tibia with a height equal to the target height of the lateral hinge, and determining the initial positioning point of the osteotomy plane by identifying the spatial intersection point of the virtual ray with the surface of the tibial three-dimensional mesh model.

[0015] Preferably, the construction of the osteotomy parameter geometric constraint model specifically includes: performing a Boolean intersection operation between the plane and the mesh to obtain the intersection profile of the current osteotomy plane and the tibial 3D mesh model; reorganizing the intersection profile into a closed ordered polyline, which is defined as the osteotomy profile line; orthogonally projecting the X-axis of the tibial local anatomical coordinate system onto the current osteotomy plane to obtain the osteotomy direction unit vector; identifying the two intersection points of the osteotomy profile line and a straight line passing through the positioning point and parallel to the osteotomy direction unit vector; marking the intersection point located on the inner side as the osteotomy entry point and the intersection point located on the outer side as the osteotomy exit point; calculating the straight-line distance between the two based on the spatial coordinate difference between the osteotomy entry point and the osteotomy exit point, which is defined as the total width of the tibial cross section under the current osteotomy plane height and angle.

[0016] Preferably, the automatic calculation and updating of other related parameters specifically includes: establishing a linear constraint equation between osteotomy depth and lateral hinge retention, wherein the linear constraint equation defines the lateral hinge retention as the total width of the tibial section minus the osteotomy depth; establishing an active-passive response mechanism for parameter adjustment, wherein when the system detects that the user has modified the osteotomy plane position as an active variable, the system locks the lateral hinge retention as a fixed value and automatically adjusts the osteotomy depth according to the depth update formula; wherein the depth update formula is: the updated recommended value of osteotomy depth is equal to the total width of the tibial section recalculated after the osteotomy plane position is updated minus the currently locked fixed value of the lateral hinge retention.

[0017] In one specific embodiment, the interactive change of osteotomy parameters further includes performing boundary condition control based on a safety threshold: real-time monitoring of changes in the values ​​of the driven variables; if the osteotomy depth input by the user results in the calculated lateral hinge retention amount being less than a preset minimum safety threshold for the lateral hinge, then boundary restriction processing is performed; the system resets the current input value to the maximum allowable value according to the boundary restriction formula, which is: the maximum allowable osteotomy depth under the current osteotomy plane state is equal to the total width of the tibial section minus the minimum safety threshold for the lateral hinge; wherein, the minimum safety threshold for the lateral hinge is defined as the minimum straight-line distance between the osteotomy exit point and the edge of the lateral cortex of the tibia, used to characterize the minimum bone retention width required to prevent lateral hinge breakage.

[0018] Preferably, the establishment of the virtual osteotomy and dislocation model specifically includes: performing mesh cutting operations on the tibial three-dimensional mesh model using the point-normal equation of the osteotomy plane, marking the set of vertices with positive directed distances as the proximal tibial sub-mesh, and marking the set of vertices with negative directed distances as the distal tibial sub-mesh; constructing a virtual dislocation kinematic model based on a rotation matrix, selecting the osteotomy exit point as the virtual hinge rotation center, and selecting the Y-axis of the tibial local anatomical coordinate system as the rotation axis vector; constructing a homogeneous transformation matrix according to the user-defined target correction angle, applying the homogeneous transformation matrix to the distal tibial sub-mesh and the ankle joint center point, and calculating the new spatial coordinates of the distal geometry and its associated points.

[0019] Preferably, the calculation and pre-simulation of the lower limb mechanical force line path specifically includes: obtaining the femoral head center and connecting it to the updated ankle joint center to construct a new lower limb force line vector; fitting the articular surface vertex set of the proximal tibial sub-mesh to obtain the tibial plateau plane, and determining the spatial intersection point of the lower limb force line vector and the tibial plateau plane; establishing a normalized coordinate system, defining the medial edge point of the tibial plateau as the 0% reference point, defining the lateral edge point of the tibial plateau as the 100% reference point, and calculating the percentage of the lower limb force line penetration point on the tibial plateau based on the ratio of the projected distance to the total distance.

[0020] Preferably, the safety detection of the osteotomy scheme specifically includes: constructing a virtual geometric model of the surgical instruments, the virtual geometric model including a virtual oscillating saw model generated based on the osteotomy path and a virtual plate model registered by the iterative nearest point algorithm; performing spatial interference detection based on the symbolic distance field algorithm, constructing a discretized symbolic distance field for the solidified model of the posterior cortical boundary of the tibia and the osteotomy gap; sampling the distance values ​​of the surface points of the virtual geometric model in the symbolic distance field, traversing all sampling points, and if there is any sampling point with a negative distance value, it is determined that the virtual geometric model has caused penetrating interference with the anatomical structure.

[0021] A second aspect of the present invention provides a high tibial osteotomy surface planning system, comprising:

[0022] The bone model reconstruction module is used to receive and parse medical imaging data of the patient's knee joint, perform threshold segmentation and surface reconstruction, and generate a three-dimensional mesh model of the tibia and femur.

[0023] The feature point association module is used to identify key anatomical landmarks on the three-dimensional mesh model and construct a local anatomical coordinate system of the tibia based on the key anatomical landmarks.

[0024] The osteotomy plane generation module is used to generate an initial geometric model of the osteotomy plane in the local anatomical coordinate system of the tibia according to preset statistical rules and the key anatomical landmarks.

[0025] The multi-parameter linkage module is used to construct a geometric constraint model for osteotomy parameters, and to process interactive changes in osteotomy parameters based on the geometric constraint model. When a single parameter changes, the values ​​of other related parameters are automatically calculated and updated.

[0026] The pre-analysis and verification module is used to build a virtual osteotomy and expansion model based on the updated osteotomy parameters, perform calculation and pre-analysis of the mechanical force line path of the lower limb, and perform safety checks on the osteotomy plan.

[0027] The result output module is used to lock the final confirmed osteotomy planning parameter set, perform coordinate space inverse mapping on the osteotomy planning parameter set, generate a standard navigation protocol file, and output it.

[0028] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect above.

[0029] In summary, the present invention has at least one of the following beneficial technical effects:

[0030] 1. This invention solves the problems of low efficiency and logical conflicts in preoperative planning by constructing a geometric constraint model of osteotomy parameters and a multi-parameter linkage mechanism. It establishes a dynamic relationship between the position of the osteotomy plane, the osteotomy depth and the amount of lateral hinge retention using linear constraint equations. When the user adjusts any active variable, the system automatically calculates and updates the associated parameters. This changes the trial-and-error mode of traditional planning that requires repeated manual measurement and independent adjustment of various parameters. While ensuring the rigor of the geometric logic of the parameters, it improves planning efficiency and parameter synergy.

[0031] 2. This invention accurately calculates the mechanical force line path of the lower limb after correction by simulating the rotational movement of the distal tibia. It introduces minimum safety threshold control of the lateral hinge and instrument interference detection based on the symbolic distance field, enabling doctors to know the force line target status in real time during the planning stage and to detect the risk of lateral hinge breakage or penetrating interference of the virtual oscillating saw on the posterior cortex in advance, thereby ensuring the dual safety of the osteotomy plan in terms of mechanics and anatomy.

[0032] 3. This invention identifies key anatomical landmarks on a three-dimensional mesh model and generates an initial geometric model that accurately conforms to the patient's specific proximal tibial morphology. By planning within a unified anatomical coordinate system, spatial errors in osteotomy surface parameters are effectively controlled, ensuring that the generated standard navigation protocol file can accurately guide subsequent surgical procedures. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the system architecture of the present invention;

[0034] Figure 2 This is a schematic diagram of the method flow of the present invention.

[0035] The module includes: 101 Bone Model Reconstruction Module; 102 Feature Point Association Module; 103 Osteotomy Surface Generation Module; 104 Multi-Parameter Linkage Module; 105 Pre-Show Verification Module; and 106 Result Output Module. Detailed Implementation

[0036] The following is in conjunction with the appendix Figure 1 - Appendix Figure 2 The present invention will be further described in detail below.

[0037] See attached document Figure 1 The present invention provides a high tibial osteotomy osteotomy surface planning system, which mainly includes a hardware operating environment and multiple software functional modules running on it.

[0038] The system's hardware operating environment includes a processor, memory, and a display. The memory stores computer program instructions. The processor is configured to execute the computer program instructions in the memory to perform osteotomy surface planning calculations. The display presents a three-dimensional visual interactive interface, showing the bone model, osteotomy surfaces, and related parameter information. In some embodiments, the system is also communicatively connected to an optical locator for acquiring three-dimensional positional coordinate data within the surgical space.

[0039] The high tibial osteotomy osteotomy surface planning system includes the following logical architecture: bone model reconstruction module 101, feature point association module 102, osteotomy surface generation module 103, multi-parameter linkage module 104, pre-run verification module 105, and result output module 106.

[0040] The bone model reconstruction module 101 receives and parses medical imaging data of the patient's knee joint. This medical imaging data is typically DICOM-compliant CT tomographic scan data. The bone model reconstruction module 101 performs threshold segmentation on the image data, extracts bone tissue pixel information, and uses a surface reconstruction algorithm to generate a high-precision three-dimensional mesh model of the tibia and femur. The bone model reconstruction module 101 renders the generated three-dimensional mesh model to the display and responds to user view operation commands, enabling the model to be rotated, scaled, and translated.

[0041] The feature point association module 102 is used to identify and mark key anatomical landmarks on the three-dimensional mesh model generated by the bone model reconstruction module 101. These anatomical landmarks include, but are not limited to, the proximal medial tibial point, the proximal lateral tibial point, the intercondylar spine of the tibia, and the distal medial and lateral epicondyles of the femur. Based on these anatomical landmarks, the feature point association module 102 constructs a local coordinate system for the tibia and establishes a mapping relationship between the subsequently generated osteotomy surface spatial vector and this local coordinate system, thereby unifying the planned spatial reference.

[0042] The osteotomy surface generation module 103 is used to automatically generate an osteotomy surface model containing initial geometric parameters in three-dimensional space based on the osteotomy surgery type selected by the user (such as open wedge high tibial osteotomy) and the coordinates of anatomical feature points provided by the feature point association module 102. This module is responsible for calculating the initial position and normal vector of the osteotomy surface, providing a starting scheme that conforms to anatomical statistical laws for subsequent parameter adjustments.

[0043] The multi-parameter linkage module 104 is used to handle parameter interaction and calculation during the osteotomy surface planning process. This module receives user-inputted osteotomy parameter values ​​or drag commands to the osteotomy surface model in the 3D view. The osteotomy parameters include osteotomy depth, distance from the articular surface, lateral hinge retention, and osteotomy plane angle. The multi-parameter linkage module 104 has pre-set parameter geometric constraint logic. When a single parameter changes, the module automatically calculates and updates the values ​​of other related parameters based on this geometric constraint logic to maintain the integrity and rationality of the osteotomy geometry.

[0044] The pre-simulation verification module 105 is used to simulate and test the safety of the current osteotomy plan. Based on the osteotomy surface parameters, the module 105 establishes a virtual osteotomy and expansion model, calculating the mechanical force line path of the lower limb after expansion and its penetration position on the tibial plateau. This module also monitors whether the amount of lateral hinge retention is below a preset safety threshold. Furthermore, the pre-simulation verification module 105 is also used to load a virtual surgical instrument model and detect whether there is spatial interference between the osteotomy surface path and the virtual oscillating saw or virtual plate model.

[0045] The result output module 106 is used to lock the final confirmed set of osteotomy surface parameters. The result output module 106 encapsulates the planning scheme, which includes spatial coordinates, angle parameters, and force line pre-simulation data, into a navigation data file. This module stores the navigation data file on a medium or transmits it to the surgical navigation execution unit via a communication interface.

[0046] See attached document Figure 2 This invention provides a method for planning the osteotomy surface of a high tibial osteotomy. The method is executed by the aforementioned planning system and includes the following steps:

[0047] S10: Acquire medical imaging data of the patient's knee joint and reconstruct a three-dimensional mesh model of the tibia and femur based on the medical imaging data.

[0048] S20, identify key anatomical feature points on the three-dimensional mesh model, and establish a spatial coordinate system for osteotomy planning based on the anatomical feature points;

[0049] S30, Generate an osteotomy surface model containing initial geometric parameters based on the osteotomy surgery type and the coordinate positions of the anatomical feature points;

[0050] S40, receive adjustment instructions for the osteotomy surface model, and perform linkage calculation and synchronous update of multiple parameters such as osteotomy depth, distance from joint surface and lateral hinge retention amount according to preset geometric constraint logic;

[0051] S50 performs virtual dislocation and lower limb force line pre-simulation based on the current osteotomy surface parameters, and verifies the numerical safety of the lateral hinge retention amount and the spatial adaptability of the virtual surgical instruments;

[0052] S60, confirm the final osteotomy surface parameter set, and generate and output a planning data file containing the parameter set.

[0053] The steps described above will be explained in detail below with reference to specific embodiments.

[0054] In step S10, the bone model reconstruction module 101 establishes a data mapping relationship between medical imaging data and three-dimensional geometric space. This step constructs a three-dimensional mesh model by parsing, resampling, and segmenting the original two-dimensional tomographic scan data, providing a digital benchmark for subsequent osteotomy planning. Specifically, step S10 can be further divided into the following sub-steps:

[0055] S101, the bone model reconstruction module 101 reads the DICOM format sequence file of the patient's knee joint. This module parses the DICOM file header information and extracts the pixel spacing and slice thickness parameters. Since the resolution of the original CT scan data is usually lower between slices than within a slice, the bone model reconstruction module 101 selects the minimum value between the original pixel spacing and slice thickness as the target voxel size, performs linear interpolation resampling on the voxel data, and generates an isotropic voxel dataset to correct the discontinuity of the original data in the inter-slice direction. Based on this, the module performs grayscale thresholding based on the radiometric density characteristics of bone tissue. The system sets a preset Huntsfield unit threshold range, marking voxels with grayscale values ​​within this range as bone tissue voxels, and marking voxels outside the range as soft tissue or background noise, thereby generating binarized three-dimensional bone mask data. The specific algorithm implementation for thresholding is a well-known technique in the field of digital image processing and will not be elaborated here.

[0056] S102, the bone model reconstruction module 101 constructs and optimizes a three-dimensional surface mesh model based on the binarized three-dimensional bone mask data. This module uses a moving cube algorithm to traverse the voxel space, calculates the intersections of isosurfaces and voxels, and generates an initial mesh model containing triangular facets. To address the high-frequency noise and vertex redundancy issues in the initial mesh model, this module further performs Laplacian smoothing and mesh simplification operations. Laplacian smoothing removes high-frequency noise by iteratively adjusting the positions of mesh vertices; its position update follows the Laplacian smoothing formula:

[0057] ;

[0058] in, The target vertex coordinate vector after smoothing; The target vertex coordinate vector before smoothing; These are the smoothing factor coefficients; The number of neighboring vertices directly connected to the target vertex; For the first The coordinate vectors of the neighboring vertices; Let be the sum of the displacement vectors of the target vertex relative to all its neighboring vertices.

[0059] The system smooths high-frequency geometric noise on the bone surface without altering the overall anatomical morphology, as calculated by the above formula. Subsequently, the bone model reconstruction module 101 uses a quadratic error metric algorithm to simplify the smoothed model's mesh. By iteratively performing edge folding operations, the number of triangular facets is reduced while maintaining the model's geometric feature error below a preset value, generating a 3D mesh model of the tibia and femur for interactive rendering. This model preserves the geometric features of key anatomical structures such as the tibial plateau, tibial tuberosity, and femoral condyle, serving as the data foundation for subsequent feature point recognition.

[0060] After the construction and optimization of the three-dimensional mesh model are completed in step S10, the feature point association module 102 performs the identification and extraction of key anatomical feature points based on the geometric model in step S20. This step transforms the anatomical morphological features into computer-recognizable spatial coordinate data, providing a mathematical basis for the geometric definition of the osteotomy plane. Specifically, the definition and extraction process of key anatomical feature points in step S20 is detailed below:

[0061] S201, the feature point association module 102 performs initial pose correction based on principal component analysis (PCA) on the tibial 3D mesh model to determine the search area for feature points. This module calculates the covariance matrix of all vertices of the mesh model and decomposes it into three orthogonal principal eigenvectors, corresponding to the vertical major axis, horizontal transverse axis, and anterior-posterior axis of the tibia, respectively. The system uses a rotation matrix to transform the mesh model into a temporary coordinate system defined by these principal eigenvectors. In this coordinate system, the feature point association module 102 calculates the axial bounding box length of the tibial model in the Z-axis direction and extracts a set of vertices within a preset length proportion (e.g., 15% of the length at the tip) in the positive Z-axis direction (i.e., the proximal direction), defining this as the proximal vertex set. This step, through geometric spatial constraints, eliminates interference from the morphology of the tibial shaft and distal bones on the identification of proximal feature points.

[0062] S202, Feature point association module 102 in the near vertex set The search is performed to determine the medial proximal tibial point, the lateral proximal tibial point, and the intercondylar spine of the tibia.

[0063] For the medial and lateral points of the proximal tibia, their geometric definitions are the extreme points of the medial and lateral edges of the tibial plateau in the coronal plane, respectively. The feature point association module 102 uses an extreme value search algorithm to determine these two points, and the calculation logic follows the coordinate extreme value formula:

[0064] ;

[0065] in, The coordinates of the target feature point to be solved; The set of proximal apexes of the tibia; Let the coordinate vectors of any vertex in the set be ; The unit vector is the search direction; To make the objective function The independent variable that takes the maximum value When calculating the medial proximal tibia point, Set as a unit vector pointing in the direction of the medial aspect of the tibia; when calculating the lateral proximal point of the tibia, Set as a unit vector pointing in the direction of the lateral aspect of the tibia.

[0066] The intercondylar spine of the tibia is geometrically defined as the highest point of the central bulge of the tibial plateau. The feature point association module 102 sets a region of interest (ROI) in the middle section of the line connecting the medial and lateral proximal points of the tibia (e.g., within 40% to 60% of the line length) and searches for the vertex with the largest Z-axis coordinate value within this region as the intercondylar spine of the tibia.

[0067] Similarly, the feature point association module 102 identifies the medial epicondyle and lateral epicondyle of the femur in the distal region of the femoral 3D mesh model. This module calculates the discrete Gaussian curvature of the vertices of the distal femur mesh, identifies convex regions with curvature values ​​greater than a preset threshold, and marks the geometric center of the convex region as the corresponding epicondyle.

[0068] S203, the feature point association module 102 provides a graphical user interface-based visual verification and human-computer interaction correction function. Because osteophyte formation or bone defects cause the automatic recognition algorithm to capture non-anatomically significant extreme points (e.g., bone spur tips), the feature point association module 102 projects the automatically calculated feature points onto the 3D view using spherical markers. The system responds to the user's picking command via the input device, using a ray-mesh intersection algorithm to calculate the coordinates of the nearest intersection point between the picked ray and the bone mesh surface. The system updates the spatial position data of the selected feature point in real time based on these intersection coordinates, thereby correcting recognition errors caused by pathological morphological abnormalities.

[0069] After the interactive confirmation of feature points is completed in step S203, the feature point association module 102 continues to construct the tibial local anatomical coordinate system in step S20. This step aims to establish a standardized orthogonal Cartesian coordinate system, unifying the anatomical structures of different patients under the same measurement space, and providing a unified measurement benchmark for osteotomy parameters. Specifically, the process of coordinate system construction in step S20 is detailed below:

[0070] S204, Feature point association module 102 defines the origin and vertical axis (Z-axis) of the coordinate system. The system selects the tibial intercondylar spine point identified in step S202 as the origin of the coordinate system. To determine the mechanical axis orientation of the tibia, this module obtains the geometric center point of the distal tibial articular surface. Specifically, the system extracts all bone tissue voxels within a preset height range of the distal tibial epiphysis (e.g., a region 5mm to 15mm above the distal articular surface) and calculates the spatial centroid of the voxel set as the distal geometric center point. The feature point association module 102 constructs a coordinate system originating from the distal geometric center point. The unit vector is defined as the positive Z-axis direction vector of the tibial local coordinate system. This axis represents the direction of the principal mechanical force line of the tibia and serves as a reference axis for subsequent adjustments to the osteotomy surface inclination.

[0071] S205, Feature point association module 102 constructs an auxiliary reference vector based on the medial and lateral feature points of the proximal tibia. This module constructs a vector pointing from the medial point of the proximal tibia to the lateral point of the proximal tibia, defined as the initial medial-lateral radial quantity. Due to natural variations in anatomical structure, this initial internal and external radial quantity... It is usually not strictly perpendicular to the Z-axis vector. It cannot be directly used as the axis vector of an orthogonal coordinate system and needs to be orthogonalized.

[0072] S206, the feature point association module 102 uses vector cross product operations to construct the front and rear radial axes (Y-axis) and the final inner and outer radial axes (X-axis). The system first calculates... The cross product of the axial vector and the initial inner and outer radial vectors is used to obtain the normal vector perpendicular to the plane containing these two vectors, which is defined as the Y-axis direction vector. Subsequently, the system recalculates the cross product of the Y-axis vector and the Z-axis vector to obtain the corrected X-axis direction vector. This process follows the vector orthogonalization formula:

[0073] ;

[0074] in, It is the normalized unit vector of the Y-axis (front and rear radial directions); This is a vector normalization operator used to transform an input vector into a unit vector with a magnitude of 1. It is the unit vector along the Z-axis (the direction of the mechanical axis); This is the vector cross product operator; These are the initial inner and outer radial quantities; It is the normalized X-axis (inner and outer diameter directions) unit vector.

[0075] Through the above calculations, the feature point association module 102 constructs a system based on... , , For rotational components, with the origin This is the homogeneous transformation matrix for the translation components. The system left-multiplies the bone mesh model and feature point data in the medical image coordinate space by the inverse of this transformation matrix, thereby uniformly mapping them to this local anatomical coordinate system. In this coordinate system, the XY plane is parallel to the ideal tibial cross-section, the XZ plane is parallel to the coronal plane, and the YZ plane is parallel to the sagittal plane, ensuring that the osteotomy depth and osteotomy surface height parameters in subsequent steps have clear clinical anatomical significance.

[0076] After constructing the tibial local anatomical coordinate system in step S20, the system proceeds to step S30, which involves generating the initial osteotomy plan. In this step, the osteotomy surface generation module 103 automatically generates an initial geometric model of the osteotomy plane in three-dimensional space based on the identified anatomical feature points and preset statistical rules, providing a starting state for subsequent parameter fine-tuning. Specifically, the geometric definition of the osteotomy plane and the initial parameter calculation process in step S30 are detailed below:

[0077] S301, the osteotomy plane generation module 103 constructs a mathematical model of the osteotomy plane in the local anatomical coordinate system of the tibia. To accurately represent the osteotomy path in the computer, the system uses point-normal equations to define the osteotomy plane. This definition includes a location point on the osteotomy plane (usually chosen as the osteotomy initiation point) and a normal vector perpendicular to the osteotomy plane. In the aforementioned tibial local anatomical coordinate system, the osteotomy plane... The geometric constraints follow the osteotomy plane equation:

[0078] ;

[0079] in, The unit normal vector of the osteotomy plane; Let be the spatial coordinate vector of any point on the osteotomy plane; The spatial coordinate vector of the osteotomy starting point; This is the vector dot product operator.

[0080] The normal vector The spatial orientation of the osteotomy surface is determined, which in turn determines the post-osteotomy tilt angle and the varus / valgus correction angle. Positioning points This determines the vertical position of the osteotomy surface on the proximal tibia. By adjusting... The amount and The coordinate values ​​are used to define a virtual plane passing through the proximal end of the tibia in the osteotomy surface generation module 103.

[0081] S302, the osteotomy surface generation module 103 performs the calculation of initial anatomical parameters to determine... and The initial values. To ensure the clinical safety of the initial plan, the module automatically calculates the spatial pose of the osteotomy surface based on the anatomical feature points identified in step S202 and in combination with statistical rules of orthopedic surgery (e.g., the lateral hinge point should usually maintain a safe distance of at least 15 mm from the lateral articular surface, or the anatomical safety threshold of 5-8 mm between the osteotomy tip and the subchondral bone).

[0082] In practice, the osteotomy surface generation module 103 first determines the target reference point for the lateral hinge area. This module extracts the coordinate values ​​of the lateral point of the proximal tibia in the Z-axis direction and offsets it distally by a preset safety distance (e.g., 15mm) to calculate the target height of the lateral hinge.

[0083] Next, in the local anatomical coordinate system, the module constructs a virtual ray emanating from the origin along the negative X-axis (i.e., pointing towards the medial side of the tibia), and sets the Z-axis height of this ray to the target height of the lateral hinge. The osteotomy surface generation module 103 calculates the intersection point of this virtual ray with the surface of the tibial three-dimensional mesh model, and defines this intersection point as the osteotomy starting point. (i.e., the aforementioned) (initial value).

[0084] The initial normal vector is calculated based on the anatomical assumption that the osteotomy surface should be approximately parallel to the tibial plateau articular surface, while also incorporating a preset posterior tilt angle compensation. The osteotomy surface generation module 103 generates the normal vector according to the initial vector calculation formula:

[0085] ;

[0086] in, This is the unit normal vector of the osteotomy plane in the initial state; It is the unit vector along the Z-axis (the direction of the mechanical axis); Rotation about the X-axis (inner and outer diameter directions) The rotation matrix of the angle. The initial value of the posterior tilt angle of the osteotomy surface is corresponding to the preset value; Rotation about the Y-axis (front-to-back radial direction) The rotation matrix of the angle. The corresponding preset initial value of the coronal tilt angle.

[0087] Based on the above calculations, the osteotomy plane generation module 103 generates a predetermined initial osteotomy plane and renders and overlays it onto the tibial 3D mesh model. This initial osteotomy plane satisfies basic anatomical safety constraints and serves as the reference state for subsequent multi-parameter linkage adjustments.

[0088] After generating the initial osteotomy plan in step S30, the system proceeds to step S40, the multi-parameter linkage optimization stage. To ensure the accuracy of the osteotomy operation, the multi-parameter linkage module 104 first constructs a geometric constraint model for the osteotomy parameters in step S40. This model transforms independent variables such as osteotomy depth, lateral hinge retention, osteotomy surface position, and tibial cross-sectional width into mathematical relationships, ensuring that when any parameter changes, the remaining related parameters can be automatically updated according to anatomical geometry. Specifically, the construction process of the geometric constraint model for the osteotomy parameters in step S40 is detailed below:

[0089] S401, the multi-parameter linkage module 104 obtains the intersection profile of the current osteotomy plane and the tibial 3D mesh model by performing a Boolean intersection operation between the plane and the mesh. Whenever the system detects osteotomy plane parameters (location points)... or normal vector When changes occur, this module performs a plane-mesh intersection operation. The system traverses all triangular faces of the tibial mesh, identifying those that intersect with the current osteotomy plane. The system identifies the sides of intersecting triangles and calculates the coordinates of the intersection points. It generates a series of discrete intersection segments and, by matching the endpoint coordinates of adjacent segments, reassembles all discrete intersection segments into an ordered closed polyline connected end-to-end, defined as the osteotomy outline. This contour line precisely represents the geometry of the anatomical plane where the current osteotomy surface is located.

[0090] S402, the multi-parameter linkage module 104 extracts the full width parameters of the tibial cross-section based on the osteotomy contour line. To accurately measure the bone width along the osteotomy path, the system first orthogonally projects the X-axis (internal and external diameter directions) of the tibial local anatomical coordinate system onto the current osteotomy plane, obtaining a unit vector of the osteotomy direction. Subsequently, the system calculates the osteotomy contour line. and through the positioning point And parallel to The two intersection points of the straight lines. The system marks the intersection point located on the medial side as the osteotomy entry point. Mark the intersection point on the outer side as the osteotomy exit point. The system calculates the Euclidean distance between the two intersection points and defines it as the total width of the tibial cross-section under the current osteotomy plane height and angle. Due to the irregular shape of the proximal tibial cross-section, this total width is a dynamic function of the osteotomy plane height, posterior tilt angle, and varus angle.

[0091] In step S403, the multi-parameter linkage module 104 establishes a linear constraint equation between the osteotomy depth and the amount of lateral hinge retention. In actual surgical planning, to ensure the integrity of the lateral hinge, the osteotomy operation cannot penetrate the entire width of the tibia. The system defines the osteotomy depth as the distance cut along the osteotomy direction from the osteotomy entry point, and the lateral hinge retention as the remaining bone width from the osteotomy termination point to the osteotomy exit point. Based on the cross-sectional geometry data calculated in step S402, the system constructs the following parameter coupling formula:

[0092] ;

[0093] in, Reserved amount for outer hinges; For the positioning point of the osteotomy plane and normal vector The function represents the total width of the tibial cross-section measured along the osteotomy direction; This refers to the depth of osteotomy.

[0094] This formula establishes the parameter linkage logic under the constraint of a constant total cross-sectional width. The system stores this constraint model in memory as the basis for subsequent interactive operations. Based on this model, if the user adjusts the height of the osteotomy surface (changing...) The system recalculates the new And update the associated parameters according to the preset logic; if the user increases the osteotomy depth The system then reduces the amount in real time based on the subtraction relationship. The numerical value. This mechanism encapsulates discrete geometric parameters into a holistic model with inherent consistency.

[0095] Based on the osteotomy parameter geometric constraint model established in step S403, the multi-parameter linkage module 104 further executes a dynamic linkage strategy based on the constraint model in step S40. This strategy aims to solve the real-time update problem in a multi-parameter coupled environment, ensuring that adjustments to a single parameter do not cause the overall surgical plan to deviate from the anatomically safe range. Specifically, the process of dynamic linkage and boundary control in step S40 is detailed below:

[0096] S404, the multi-parameter linkage module 104 establishes an active and passive response mechanism for parameter adjustment. The system identifies the target parameter being modified by listening to input events from user interface controls, defining it as the active variable, and defining other parameters affected by this active variable as passive variables. To solve the unique solution problem of the underdetermined equation system, the system presets a "hinge priority" constraint logic. When the active variable is the osteotomy plane position (i.e., the user-adjusted...),... or When this is done, the system defaults to locking the outer hinge retention amount. For a fixed value, the osteotomy depth will be... Set as a floating variable. This logic ensures that the width of the lateral bone bridge is always the first priority constraint when adjusting the osteotomy height or tilt angle.

[0097] S405, the multi-parameter linkage module 104 performs real-time reverse calculation of the associated parameters. When the system detects that the user has modified the osteotomy depth... When (as an active variable), the system directly calls the osteotomy parameter coupling formula established in step S403, substitutes the input depth value into the equation, and calculates the corresponding lateral hinge retention amount. And update the display on the user interface.

[0098] When the system detects that the user has modified the osteotomy plane position (as an active variable), the system first recalculates the total width of the tibial section at the new position according to the method in step S402. Subsequently, based on the aforementioned "hinge priority" logic, the system maintains the lateral hinge retention amount unchanged and automatically adjusts the osteotomy depth according to the following depth update formula:

[0099] ;

[0100] in, The updated recommended value for osteotomy depth. The total width of the tibia section measured along the osteotomy direction is recalculated after updating the osteotomy plane position; This is a fixed value for the amount of outer hinges currently locked by the system.

[0101] Through this reverse calculation, the system enables the osteotomy depth value to be automatically updated as the width of the tibial cross-section changes when the osteotomy surface moves in three-dimensional space, thus maintaining a constant hinge width.

[0102] S406, the multi-parameter linkage module 104 performs boundary condition control based on a safety threshold. The system presets a minimum safety threshold for the outer hinge. (e.g., 5mm). During user-interactive parameter adjustment, the system monitors the changes in the value of the driven variable in real time. If the osteotomy depth input by the user causes the calculated hinge retention amount to be less than the minimum safe threshold, the system triggers boundary clamping logic. The system resets the data in the input buffer to the maximum allowed value, which follows the boundary limit formula:

[0103] ;

[0104] in, This represents the maximum permissible osteotomy depth under the current osteotomy plane condition. For the positioning point of the osteotomy plane and normal vector The function represents the total width of the tibial cross-section measured along the osteotomy direction; This is the preset minimum safety threshold for the outer hinge.

[0105] Simultaneously, the system renders the affected hinge area in a warning color in the visualization view and prohibits saving operations until the parameters are adjusted back to within the safety constraints. This step solidifies the anatomical safety rules into logical control constraints at the software level, preventing surgical planning risks caused by human error.

[0106] After the multi-dimensional linkage optimization of the osteotomy parameters is completed in step S40, the pre-operative simulation and verification module 105 performs a high-precision pre-operative simulation in step S50. To intuitively evaluate the corrective effect of the osteotomy plan on the patient's lower limb biomechanical alignment, the system constructs a virtual surgical environment based on the principles of rigid body kinematics, simulating the osteotomy spreading process and the dynamic shift of the lower limb biomechanical alignment in real time. Specifically, the process of virtual osteotomy and lower limb biomechanical alignment simulation in step S50 is detailed below:

[0107] S501, the pre-performance verification module 105 performs mesh segmentation and topology reconstruction based on the osteotomy plane. The system reads the final osteotomy plane parameters determined in step S40 and performs mesh cutting operations on the tibial 3D mesh model using the point-normal equation of the plane. The system traverses all vertices of the mesh and calculates the directed distance from each vertex to the osteotomy plane. The system marks the set of vertices with positive directed distances as the proximal tibial sub-mesh (defined as the stationary point set) and the set of vertices with negative directed distances as the distal tibial sub-mesh (defined as the moving point set). To simulate a realistic open wedge osteotomy, the system identifies the ankle joint center point... It is marked as part of the motion point set, and in subsequent calculations, it is subjected to the same rigid body transformation matrix as the distal tibia submesh.

[0108] S502, the pre-performance verification module 105 constructs a virtual opening kinematic model based on a rotation matrix. The system selects the osteotomy exit point determined in step S402 as the virtual hinge rotation center. The Y-axis (anteroposterior diameter direction) of the tibial local anatomical coordinate system was selected as the rotation axis vector. The system corrects the angle according to the user-defined target. (i.e., the spread angle), construct a homogeneous transformation matrix to calculate the new spatial coordinates of the distal tibia and ankle joint center. The rigid body motion of the distal geometry and its associated points follows the kinematic transformation formula:

[0109] ;

[0110] in, The transformed target point coordinate vector (including the far-end mesh vertices and the updated ankle joint center) ); The initial point coordinate vector before the transformation; The coordinate vector of the virtual hinge rotation center (osteotomy exit point); This is a translation matrix used to translate the origin of the coordinate system to the center of the hinge. for The inverse matrix; The vector about the rotation axis is constructed based on the Rodriguez formula. Rotation Rotation matrix of angle.

[0111] Using this algorithm, the system can generate the tibial morphology after expansion in real time without destroying the mesh topology, and dynamically present the opening process of the osteotomy gap on the screen.

[0112] S503, the pre-performance verification module 105 calculates the lower limb force line trajectory and its percentage of penetration points on the tibial plateau in real time. The system obtains the femoral head center. (The coordinates of this point have been obtained and fixed through full-length image registration in the model construction of step S10), and are connected to the updated ankle joint center in step S502. Construct new lower limb force line vectors .

[0113] Subsequently, the system uses the least squares method to fit the articular surface vertex set of the proximal tibial sub-mesh to obtain the tibial plateau plane, and calculates the intersection point of the lower limb force line vector with this plane. To quantify the position of the force line, a normalized coordinate system was established, defining the medial edge point of the tibial plateau. Define the lateral edge of the tibial plateau as the 0% reference point. This is a 100% reference point. The system derives quantitative indicators based on the force line percentage calculation formula:

[0114] ;

[0115] in, This represents the percentage of the lower limb force line intersection point located on the tibial plateau. The coordinates of the intersection point of the lower limb force line vector and the fitted plane of the tibial plateau; represents the coordinates of the feature point on the medial edge of the tibial plateau; · represents the vector dot product operator; For the reason point to The unit direction vector; The coordinates of the feature points on the outer edge of the tibial plateau; This is the Euclidean norm (vector magnitude) operator.

[0116] The system will calculate the results The numerical values ​​are displayed in real-time on one side of the 3D view. This is shown when the doctor adjusts the osteotomy angle. The system updates in real time. Location and The numerical values ​​enable closed-loop simulation, helping doctors to accurately control the degree of force line correction.

[0117] While virtual osteotomy and dynamic force line pre-simulation are performed in steps S501 to S503, the pre-simulation verification module 105 executes steps S504 to S506 in parallel, namely, dual verification of hinge safety and instrument compatibility. This process aims to use computer graphics algorithms to quantitatively evaluate the physical feasibility of the surgical plan in three-dimensional space and detect potential anatomical risks and instrument implantation conflicts. Details are as follows:

[0118] In step S504, the pre-performance verification module 105 performs real-time three-dimensional volumetric monitoring of the lateral hinge bone thickness. Although step S40 has calculated the two-dimensional hinge retention width based on the cross-sectional profile, considering the surface irregularity of the lateral tibial condyle in three-dimensional space, the system needs to further verify whether the bone volume around the hinge axis is sufficient. The system constructs a cylindrical region of interest (VOI) centered on the osteotomy exit point determined in step S402, along the Z-axis (mechanical axis) of the tibial local anatomical coordinate system. The system calculates the minimum Euclidean distance from the bone mesh surface points within this VOI to the hinge axis originating from the osteotomy exit point, using this as a three-dimensional hinge thickness safety indicator. If this minimum distance is less than a preset fracture threshold (e.g., 3 mm), the system issues an alarm signal.

[0119] S505, the pre-performance verification module 105 constructs a virtual geometric model of the surgical instruments. The system retrieves a virtual oscillating saw model and a virtual steel plate model that match the current plan from a pre-set instrument database. The virtual oscillating saw model is constructed as a scanning volume generated based on the osteotomy path, with a preset thickness (e.g., 1.2 mm) and width. Its spatial pose is determined by the osteotomy plane parameters defined in step S301. and The virtual plate model is spatially registered with the proximal medial surface of the tibia using an iterative nearest-point algorithm to minimize the distance error between the inner surface of the plate and the bone surface, thereby determining the virtual implantation position of the plate.

[0120] S506, the pre-verification module 105 performs spatial interference detection based on the symbolic distance field algorithm. In order to accurately determine whether the osteotomy surface (i.e., the virtual oscillating saw trajectory) intrudes into the posterior cortical protection zone of the tibia, or whether the locking screw of the virtual plate accidentally penetrates the osteotomy gap, the system uses the distance field gradient operator for collision analysis.

[0121] The system first constructs a discretized symbolic distance field for the physical model of the posterior cortical boundary of the tibia and the osteotomy gap. For any point in space , The absolute value of represents the distance from the point to the nearest object surface, and the sign indicates whether the point is inside the object (negative value) or outside (positive value). Subsequently, the system samples the scanning body of the virtual oscillating saw and the trajectory of the steel plate screw, calculating the function value of the sampled point set in the target distance field. The system evaluates the conflict state based on the interference determination formula:

[0122] ;

[0123] in, It is the minimum interference distance coefficient; The operator for finding the minimum value; Discrete sampling points on the surface of virtual surgical instruments (including virtual oscillating saw scanning bodies or plate screws); For set membership symbols; This refers to the set of surface sampling points of the virtual surgical instrument. For the target object (including the solid model of the posterior cortical boundary of the tibia or the osteotomy gap), at the sampling point The sign distance field function value at that location.

[0124] System basis Numerical execution logic determination:

[0125] like If a penetrating interference is detected (e.g., a screw penetrates the osteotomy suture, or a saw blade cuts through the posterior cortex), the system marks the interference area with a highlighted color block in the 3D view and locks the export function of the osteotomy plan.

[0126] like ( If the system determines a critical danger state by setting a preset safety buffer distance (e.g., 2mm), it will output a warning.

[0127] like The device was deemed to be well-suited for use.

[0128] Through the aforementioned dual verification mechanism, the system preemptively eliminates the risk of hinge breakage and instrument installation conflicts in the virtual environment, ensuring that the output osteotomy parameters comply with clinical anatomical limitations and instrument operation specifications.

[0129] After completing the virtual osteotomy pre-simulation and safety dual verification in step S50, confirming that the current surgical plan meets all anatomical constraints and instrument compatibility requirements, the system proceeds to step S60, the result output and data encapsulation stage. In this step, the result output module 106 converts the geometric parameters, optimized through multiple iterations, into a standardized command protocol that the surgical navigation system can directly read. Specifically, the process of locking the final planning parameters and converting the navigation data protocol in step S60 is detailed below:

[0130] S601, the result output module 106 performs the locking and structured encapsulation of the final planning parameters. The system first locks all parameter values ​​in the current interactive interface to prevent unexpected changes in the data during the output stage. Subsequently, the system extracts the key geometric features confirmed in steps S40 and S50 to construct a set of osteotomy planning parameters containing complete surgical information. This set contains not only geometric location information, but also dynamic correction targets and safety constraint boundaries, and its mathematical definition follows the formula for the set of planning parameters:

[0131] ;

[0132] in, This is the final set of osteotomy planning parameters; The three-dimensional coordinates of the final confirmed osteotomy starting point in the local coordinate system; This is the final confirmed unit normal vector for the osteotomy plane; This is the final confirmed osteotomy depth value; To determine the final osteotomy angle; The coordinates of the rotation center of the outer hinge; The percentage of the target lower limb alignment position confirmed in the pre-operative simulation.

[0133] S602, the result output module 106 performs the inverse coordinate space mapping of the navigation data. Since the aforementioned steps S20 to S50 are all calculations performed in the tibial local anatomical coordinate system, while surgical navigation systems are typically guided based on the original medical image (DICOM) coordinate system, the system needs to... The geometric data is mapped back to the original coordinate space. The system obtains the homogeneous transformation matrix from the original coordinate system to the local coordinate system generated when constructing the local anatomical coordinate system of the tibia in step S20, and performs an inverse operation on it to obtain the inverse mapping matrix. The system calculates according to the output coordinate transformation formula:

[0134] ;

[0135] in, The coordinates of the position points are transformed to the original image coordinate system; This is the inverse mapping matrix from the local anatomical coordinate system of the tibia to the original image coordinate system; Input position points in the local coordinate system (taken from...) gather); This is the direction vector transformed to the original image coordinate system; for A submatrix containing rotation components; The input direction vector in the local coordinate system (taken from...) gather).

[0136] This step ensures that the output spatial data is accurately registered with the patient's original CT data, enabling the navigation system to correctly project the planned path onto the patient's physical anatomical structures.

[0137] S603, the result output module 106 performs the generation and integrity verification of the standard navigation protocol file. The system uses Extensible Markup Language (XML) as the data carrier and, according to the preset navigation device communication protocol, formats and writes the converted geometric data and patient identification information. The system generates a planning description file containing a header, data body, and check bits.

[0138] In its implementation, the system converts floating-point coordinate and vector data into strings of specified precision and writes them into the corresponding tag nodes of the XML file. For example, the `<link>` node stores the converted osteotomy point and normal vector, and the `<link>` node stores the boundary coordinates of the hinge protection zone. To prevent data corruption during transmission, the system performs a hash operation on all generated data content, generating a unique digital digest and writing it to the end of the file. The specific encoding implementation for XML file generation and hash verification can be accomplished by those skilled in the art using existing software engineering methods; these are well-known technologies in the field of computer data processing and will not be elaborated upon here.

[0139] Finally, the packaged and verified planning file is transmitted to the operating room navigation equipment via a network interface, completing the data interaction from virtual planning to physical execution. This process ensures that the high-precision geometric information of the preoperative planning can be completely transmitted to the intraoperative support system, providing the surgeon with precise osteotomy guidance.

[0140] The present invention also provides a storage medium storing a computer program, which is executed by a processor to perform the method described above.

[0141] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

Claims

1. A method for planning the osteotomy surface of a high tibial osteotomy, characterized in that, Includes the following steps: Acquire medical imaging data of the patient's knee joint, perform threshold segmentation and surface reconstruction, and generate a three-dimensional mesh model of the tibia and femur; Key anatomical landmarks were identified on the three-dimensional mesh model, and a local anatomical coordinate system of the tibia was constructed based on the key anatomical landmarks. Based on preset statistical rules and the key anatomical landmarks, an initial geometric model of the osteotomy plane is generated in the local anatomical coordinate system of the tibia. Construct a geometric constraint model for osteotomy parameters, and process interactive changes of osteotomy parameters based on the geometric constraint model. When a single parameter changes, automatically calculate and update the values ​​of other related parameters. A virtual osteotomy and expansion model is established based on the updated osteotomy parameters, the calculation and pre-simulation of the mechanical force line path of the lower limb are performed, and the safety of the osteotomy scheme is tested. The final confirmed osteotomy planning parameter set is locked, and the osteotomy planning parameter set is subjected to inverse coordinate space mapping to generate a standard navigation protocol file and output it.

2. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 1, characterized in that, The initial geometric model for generating the osteotomy plane specifically includes: The osteotomy plane is defined using a point-normal equation, which includes a location point on the osteotomy plane and a unit normal vector perpendicular to the osteotomy plane. Determine the target reference point for the outer hinge area, obtain the coordinate value of the lateral point of the proximal tibia in the Z-axis direction, and subtract a preset safety distance from the coordinate value to obtain the target height of the outer hinge; In the local anatomical coordinate system of the tibia, a virtual ray is constructed emanating from the origin along the negative X-axis with a height equal to the height of the outer hinge target. The spatial intersection of the virtual ray with the surface of the tibial three-dimensional mesh model is identified, and the spatial intersection is defined as the initial positioning point of the osteotomy plane.

3. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 1, characterized in that, The construction of the osteotomy parameter geometric constraint model specifically includes: Perform a Boolean intersection operation between the plane and the mesh to obtain the intersection profile of the current osteotomy plane and the tibial 3D mesh model. Reorganize the intersection profile into an ordered closed polyline with the beginning and end connected, and define it as the osteotomy profile line. The X-axis of the local anatomical coordinate system of the tibia is orthogonally projected onto the current osteotomy plane to obtain the unit vector of the osteotomy direction, and the two intersection points of the osteotomy outline and the straight line passing through the positioning point and parallel to the unit vector of the osteotomy direction are identified. The intersection point located on the inner side is marked as the osteotomy entry point, and the intersection point located on the outer side is marked as the osteotomy exit point. The straight-line distance between the osteotomy entry point and the osteotomy exit point is determined based on the spatial coordinate difference between them, and is defined as the total width of the tibial section under the current osteotomy plane height and angle.

4. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 3, characterized in that, The automatic calculation and updating of other related parameters specifically includes: A linear constraint equation is established between the osteotomy depth and the amount of lateral hinge retention, wherein the linear constraint equation defines the amount of lateral hinge retention as the total width of the tibial section minus the osteotomy depth; An active and passive response mechanism for parameter adjustment is established. When the system detects that the user has modified the position of the osteotomy plane as an active variable, the system locks the amount of the outer hinge retention to a fixed value and automatically adjusts the osteotomy depth according to the depth update formula. The depth update formula is as follows: the updated osteotomy depth recommendation value is equal to the total width of the tibia section obtained by recalculation after the osteotomy plane position is updated, minus the fixed value of the lateral hinge retention currently locked by the system.

5. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 4, characterized in that, The interactive modification of osteotomy parameters also includes performing boundary condition control based on safety thresholds: The system monitors the changes in the values ​​of the driven variables in real time. If the osteotomy depth input by the user causes the calculated amount of the lateral hinge to be less than the preset minimum safety threshold for the lateral hinge, the system triggers the boundary clamping logic. The system resets the data in the input buffer to the maximum allowed value according to the boundary limit formula, which is: the maximum allowed osteotomy depth under the current osteotomy plane state is equal to the total width of the tibial section minus the minimum safe threshold of the lateral hinge; The minimum safety threshold of the lateral hinge is defined as the minimum straight-line distance between the osteotomy exit point and the edge of the lateral cortical bone of the tibia, which is used to characterize the minimum bone preservation width required to prevent the lateral hinge from breaking.

6. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 3, characterized in that, The establishment of the virtual osteotomy and dislocation model specifically includes: The point-normal equation of the osteotomy plane is used to perform mesh cutting operation on the tibial three-dimensional mesh model. The set of vertices with positive directed distance is marked as the proximal tibial sub-mesh, and the set of vertices with negative directed distance is marked as the distal tibial sub-mesh. A virtual opening kinematic model based on a rotation matrix is ​​constructed, the osteotomy exit point is selected as the virtual hinge rotation center, and the Y-axis of the local anatomical coordinate system of the tibia is selected as the rotation axis vector; A homogeneous transformation matrix is ​​constructed based on the user-defined target correction angle. The homogeneous transformation matrix is ​​then applied to the distal tibial subgrid and the ankle joint center point to calculate the new spatial coordinates of the distal geometry and its associated points.

7. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 6, characterized in that, The calculation and simulation of the mechanical force path of the lower limbs specifically includes: Obtain the femoral head center and connect it to the updated ankle joint center to construct a new lower limb force line vector; The tibial plateau plane is obtained by fitting the articular surface vertex set of the proximal tibial sub-mesh, and the spatial intersection point of the lower limb force line vector and the tibial plateau plane is determined. A normalized coordinate system is established, the inner edge point of the tibial plateau is defined as the 0% reference point, and the outer edge point of the tibial plateau is defined as the 100% reference point. The percentage of the lower limb force line penetration point on the tibial plateau is obtained according to the force line percentage calculation formula.

8. The method for planning the osteotomy surface of a high tibial osteotomy according to claim 1, characterized in that, The safety testing of the osteotomy procedure specifically includes: A virtual geometric model of surgical instruments is constructed, the virtual geometric model including a virtual oscillating saw model generated based on the osteotomy path and a virtual steel plate model registered by the iterative nearest point algorithm; Spatial interference detection is performed based on the symbolic distance field algorithm, and a discretized symbolic distance field is constructed for the solid model of the posterior cortical boundary of the tibia and the osteotomy gap. The distance values ​​of the surface points of the virtual geometric model in the symbolic distance field are sampled. All sampling points are traversed and retrieved. If the distance value of any sampling point is negative, it is determined that the virtual geometric model has undergone penetrating interference with the anatomical structure.

9. A high tibial osteotomy surface planning system, characterized in that, A method for planning a high tibial osteotomy surface according to any one of claims 1-8 includes: The bone model reconstruction module is used to receive and parse medical imaging data of the patient's knee joint, perform threshold segmentation and surface reconstruction, and generate a three-dimensional mesh model of the tibia and femur. The feature point association module is used to identify key anatomical landmarks on the three-dimensional mesh model and construct a local anatomical coordinate system of the tibia based on the key anatomical landmarks. The osteotomy plane generation module is used to generate an initial geometric model of the osteotomy plane in the local anatomical coordinate system of the tibia according to preset statistical rules and the key anatomical landmarks. The multi-parameter linkage module is used to construct a geometric constraint model for osteotomy parameters, and to process interactive changes in osteotomy parameters based on the geometric constraint model. When a single parameter changes, the values ​​of other related parameters are automatically calculated and updated. The pre-analysis and verification module is used to build a virtual osteotomy and expansion model based on the updated osteotomy parameters, perform calculation and pre-analysis of the mechanical force line path of the lower limb, and perform safety checks on the osteotomy plan. The result output module is used to lock the final confirmed osteotomy planning parameter set, perform coordinate space inverse mapping on the osteotomy planning parameter set, generate a standard navigation protocol file, and output it.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-8.

Images