Biplanar image-based preoperative planning method and system for total knee arthroplasty

Abstract

Provided are a biplanar image-based preoperative planning method and system for total knee arthroplasty. Based on a static biplanar lower-limb image of a knee joint of a target object in a standing position and a dynamic biplanar image of the knee joint during functional motion, as well as an imported three-dimensional model, functional parameters, morphological parameters, and a three-dimensional medial and lateral joint gap of the knee joint are calculated, and then iterative calculation is continuously carried out using a two-dimensional model and a three-dimensional dynamic model to obtain a dynamically optimized placement position, so that preoperative TKA prosthesis model selection for balanced joint space across full flexion angles can be achieved; meanwhile, TKA prosthesis position planning for correcting the mechanical axis of the knee joint under three-dimensional weight-bearing conditions is considered, thereby accurately restoring the real dynamic motion of a patient, and improving the prosthesis-patient matching degree.

Description

A Method and System for Preoperative Planning of Total Knee Arthroplasty Based on Dual-Plane Imaging Technical Field

[0001] This invention belongs to the field of intelligent medical technology, and in particular relates to a method and system for preoperative planning of total knee arthroplasty based on dual-plane imaging. Background Technology

[0002] Currently, TKA (total knee arthroplasty) surgical planning mainly utilizes two-dimensional X-ray images or three-dimensional supine CT images. However, due to limitations in imaging technology, it is impossible to obtain three-dimensional anatomical parameters of the joint in a weight-bearing position, resulting in discrepancies between the actual plan and the patient's actual condition. Surgeons still need to make adjustments during the operation, reducing the effectiveness of preoperative planning. Preoperative planning that does not consider the patient's dynamic mobility may ultimately lead to unsatisfactory postoperative functional performance.

[0003] The invention patent with publication number CN112842529A mainly utilizes anteroposterior and lateral X-ray images of the knee joint to plan the size and position of the prosthesis. It uses a neural network to identify key points and axes of the knee joint, and plans the prosthesis type and position based on these key points and axes. However, two-dimensional X-ray images cannot reflect the true three-dimensional morphology of the skeleton, and the TKA planning scheme based on two-dimensional images cannot effectively plan the three-dimensional placement of the prosthesis.

[0004] The invention patent with publication number CN113842211A mainly utilizes a three-dimensional CT model for total knee arthroplasty planning. Based on a three-dimensional skeletal model of the knee joint, key skeletal parameters are determined, thereby determining the type and model of the prosthesis, as well as its placement position and angle. However, due to significant differences between the force lines measured in the patient's standing position and the force lines and angle parameters obtained from traditional supine CT imaging, ideal correction of the patient's knee joint cannot be achieved. Although this method considers the principle of joint space balance in knee arthroplasty, it simulates extension and flexion of the knee joint, which does not match the dynamic movement of a real patient, and it also does not consider joint space balance within the full flexion range.

[0005] Therefore, preoperative planning for TKA needs further optimization.

[0006] Summary of the Invention

[0007] To address the aforementioned issues, the present invention aims to provide a preoperative planning method and system for total knee arthroplasty based on dual-plane imaging. This system enables preoperative TKA prosthesis selection with joint space balance at full flexion angles, as well as TKA prosthesis position planning that simultaneously considers knee joint alignment correction under three-dimensional weight-bearing positions. This accurately reproduces the patient's true dynamic movements and improves the compatibility between the prosthesis and the patient.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows: a preoperative planning method for total knee arthroplasty based on dual-plane imaging, comprising the following steps: acquiring static dual-plane lower limb images of the knee joint in a standing position and dynamic dual-plane images of the knee joint during functional movement; acquiring a three-dimensional model of the knee joint of the target object, and performing two-dimensional-three-dimensional spatial registration and alignment with the static dual-plane lower limb images and dynamic dual-plane images of the knee joint in a standing position, respectively, to obtain a three-dimensional registration image of the knee joint in a standing position and a three-dimensional registration image of the knee joint during functional movement; calculating knee joint functional parameters, morphological parameters, and medial and lateral joint three-dimensional gaps based on the knee joint standing three-dimensional registration image, and determining the tibial prosthesis and tibial osteotomy plane based on the functional and morphological parameters; determining the femoral prosthesis based on the functional and morphological parameters, importing the determined three-dimensional model of the femoral prosthesis, and optimizing the femoral articular surface and the femoral prosthesis articular surface using a three-dimensional point cloud registration algorithm. The static preliminary placement position of the femoral prosthesis is determined, and the static preliminary relative positional relationship between the femoral prosthesis and the femur is obtained as P1. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides of the knee joint, a static optimized placement position in the standing position is calculated based on a preset optimization algorithm, yielding the static optimized relative positional relationship between the femoral prosthesis and the femur as P2. This static optimized relative positional relationship P2 is imported into three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and the three-dimensional joint gap between the medial and lateral sides after femoral prosthesis placement is simulated and calculated during functional movement. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides of the knee joint throughout the full flexion range, the position of the femoral prosthesis is iteratively calculated and adjusted based on the optimization algorithm until a dynamic optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional joint gap balance throughout the full flexion range. The dynamic optimized relative positional relationship between the femoral prosthesis and the femur as well as the femoral osteotomy plane are then output.

[0009] Preferably, before acquiring static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movement, the preoperative planning method for total knee arthroplasty further includes correcting the biplane imaging system to obtain the relative spatial position relationship and imaging parameters of the biplane system, including the following steps: placing a correction box containing multiple steel balls in the imaging space for imaging; solving the spatial positions of the X-ray source and the flat panel detector in the biplane imaging system by using the acquired steel ball images and the known relative spatial positions of the steel balls; calibrating the coordinate system of the biplane imaging system based on the spatial positions of the X-ray source and the flat panel detector.

[0010] Preferably, the method for obtaining a three-dimensional model of the knee joint of the target object includes: obtaining CT images / magnetic resonance images of the knee joint from multiple angles and the corresponding angle information, and performing three-dimensional modeling based on the CT images / magnetic resonance images from multiple angles and the corresponding angle information to obtain a three-dimensional model of the knee joint.

[0011] Preferably, acquiring a three-dimensional model of the knee joint of the target object and performing two-dimensional-three-dimensional spatial registration and alignment with the standing static biplane lower limb image and dynamic biplane image of the knee joint further includes: importing the acquired standing static biplane lower limb image and the dynamic biplane image of the knee joint during functional movement into a preset registration program, importing the three-dimensional model of the knee joint of the target object, and determining the position of the femur and tibia in three-dimensional space by matching the bone contour with the biplane dynamic perspective image in each frame in the reconstructed virtual projection environment.

[0012] Preferably, calculating the functional parameters, morphological parameters, and medial-lateral joint three-dimensional gap of the knee joint based on the three-dimensional registration image of the knee joint in a standing position further includes: obtaining the three-dimensional spatial position of key anatomical feature points based on the three-dimensional registration image of the knee joint in a standing position, wherein the key anatomical feature points include at least the center of the hip joint, the center of the knee joint, the center of the ankle joint, the medial and lateral femoral condyles, and the medial and lateral tibial plateau centers; calculating the functional parameters and morphological parameters of the knee joint, wherein the functional parameters include at least the joint force line, and the morphological parameters include at least the anteroposterior diameter of the femur, the size of the medial and lateral femoral condyles, the length and width of the tibial plateau after internal and external rotation of the posterior femoral condyle, and the posterior tilt angle of the tibial plateau; determining the tibial osteotomy plane based on the functional and morphological parameters of the knee joint, selecting the medial and lateral femoral condyle articular surfaces, and calculating the shortest distance from the medial and lateral femoral condyles to the medial and lateral tibial plateau planes in three-dimensional space as the medial-lateral joint three-dimensional gap of the knee joint.

[0013] Preferably, the calculation of the functional parameters, morphological parameters and medial and lateral joint three-dimensional gap of the knee joint based on the knee joint standing three-dimensional registration image further includes: when the knee joint functional parameters do not meet the preset conditions, after calculating the joint force line, rotating the femur to perform varus / valgus pre-correction and determine the simulated force line, and then determining the tibial osteotomy plane based on the simulated force line.

[0014] Preferably, the static preliminary placement of the femoral prosthesis based on the femoral articular surface and the femoral prosthesis articular surface using a three-dimensional point cloud registration optimization algorithm further includes: setting the points on the femoral articular surface as point cloud P. t Set the point cloud of the femoral prosthesis joint surface as P s The goal is to find an optimal set of rotation transformations R. * Translation and shift transformation T * The goal is to minimize the relative distance between PC2 and point cloud PC1 after rotational and / or translational transformations, where PC2 refers to the point cloud on the surface of the femoral prosthesis and PC1 refers to the point cloud on the surface of the femoral joint. The problem description is... The point cloud is preprocessed and filtered, the rotation and translation transformations are solved, the weights of corresponding points are adjusted, unreasonable corresponding points are identified, the loss value is calculated, the loss value is minimized, the current optimal change is solved to update the rotation and translation transformations, and the iteration continues until convergence is achieved to obtain the static preliminary placement position of the femoral prosthesis and the static preliminary relative positional relationship P1 between the femoral prosthesis and the femur.

[0015] Preferably, the static optimized placement position in the standing position calculated based on the preset optimization algorithm further includes: calculating the medial and lateral gaps between the femoral prosthesis and the tibial osteotomy surface in the initial static placement position after simulated force line correction, and adjusting the femoral prosthesis position to balance the absolute value difference of the medial and lateral gaps; wherein, the loss function is the minimum absolute difference of the medial and lateral gaps: loss1=||d l |-|d m ||, where |d m | and |d l | represents the absolute distance from the medial and lateral condyles of the femoral prosthesis to the tibial osteotomy plane, respectively; the optimization object is the 6-DOF value of the femoral prosthesis relative to the femur [α,β,λ,x,y,z], where α represents the flexion angle of the femoral prosthesis, β represents the varus / valgus angle of the femoral prosthesis, λ represents the internal / external rotation angle of the femoral prosthesis, x represents the anterior / posterior movement of the femoral prosthesis, y represents the lateral / hyperaxial movement of the femoral prosthesis, and z represents the proximal / distal movement of the femoral prosthesis.

[0016] Preferably, iteratively calculating and adjusting the femoral prosthesis position until a dynamically optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional joint clearance balance across the full flexion range further includes: adjusting the placement position of the femoral prosthesis relative to the femur within a preset range, calculating the three-dimensional joint clearance at different flexion angles under different femoral prosthesis positions, until three-dimensional joint clearance balance is achieved across the full flexion range, wherein the loss function is defined as: n represents different buckling positions, ||d l |-|d m || i These represent the absolute differences between the medial and lateral interarticular spaces at each flexion position; the individualized variation curves of the three-dimensional interarticular spaces between the medial and lateral joints during knee flexion, with the loss function set as loss3 = |f(d m )-g(d m )|+|f(d l )-g(d l |f(d)-g(d)| represents the difference between the three-dimensional joint clearance and the target joint clearance after the prosthesis is placed.

[0017] Based on the same concept, this invention also provides a preoperative planning system for total knee arthroplasty based on dual-plane imaging, comprising: an image acquisition module for acquiring static dual-plane lower limb images of the knee joint in a standing position and dynamic dual-plane images of the knee joint during functional movement; a model registration module for acquiring a three-dimensional model of the knee joint of the target object and performing two-dimensional-three-dimensional spatial registration and alignment with the static dual-plane lower limb images and dynamic dual-plane images of the knee joint in a standing position, respectively, to obtain a three-dimensional registration image of the knee joint in a standing position and a three-dimensional registration image of the knee joint during functional movement; a calculation module for calculating knee joint functional parameters, morphological parameters, and medial and lateral joint three-dimensional gaps based on the knee joint standing three-dimensional registration image, and determining the tibial prosthesis and tibial osteotomy plane based on the functional and morphological parameters; and a static position determination module for determining the femoral prosthesis based on the functional and morphological parameters, importing the determined three-dimensional model of the femoral prosthesis, and determining the position based on the femoral articular surface and the femoral prosthesis articular surface through three-dimensional... The point cloud registration optimization algorithm locates the initial static placement position of the femoral prosthesis, obtaining the initial static relative positional relationship P1 between the femoral prosthesis and the femur. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides of the knee, a preset optimization algorithm calculates the static optimized placement position in the standing position, obtaining the static optimized relative positional relationship P2 between the femoral prosthesis and the femur. A dynamic position determination module imports the static optimized relative positional relationship P2 between the femoral prosthesis and the femur into three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and simulates and calculates the three-dimensional joint gap between the medial and lateral sides after femoral prosthesis placement during functional movement. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides across the full flexion range, the algorithm iteratively calculates and adjusts the position of the femoral prosthesis until a dynamic optimized placement position is found that achieves three-dimensional joint gap balance across the full flexion range. The module outputs the dynamic optimized relative positional relationship P3 between the femoral prosthesis and the femur, as well as the femoral osteotomy plane.

[0018] Because the present invention adopts the above technical solution, it has the following advantages and positive effects compared with the prior art:

[0019] 1. The technical solution of this invention solves the problem of inaccurate prosthesis size and position planning caused by the inaccuracy of image magnification and differences in shooting angle in traditional preoperative planning methods that only use single-plane weight-bearing images for TKA preoperative planning. It also solves the defect of traditional two-dimensional planning that cannot obtain the true three-dimensional relative position of the prosthesis and bone. Compared with traditional preoperative planning, the technical solution of this invention is based on static dual-plane lower limb images in a standing position and dynamic dual-plane images of the knee joint during functional movement. It can not only realize TKA three-dimensional planning based on the joint force line in the weight-bearing position, but also realize preoperative three-dimensional planning of joint space balance at the full flexion angle of the knee joint, reduce the surgeon's intraoperative operation time, and help patients achieve better postoperative joint function.

[0020] 2. The technical solution of this invention, in conjunction with an X-ray dual-plane imaging system, performs standing weight-bearing imaging and functional motion imaging of the knee joint. Combined with the skeletal model reconstructed from the patient's CT scan, the three-dimensional spatial position of the knee joint in weight-bearing position and dynamic functional motion is obtained using two-dimensional-three-dimensional registration technology. This makes the preoperative planning more accurate, ensuring that the femoral prosthesis and tibial prosthesis can achieve a better spatial matching relationship, and the patient can obtain better joint function after surgery. Attached Figure Description

[0021] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, wherein:

[0022] Figure 1 is a schematic diagram of the preoperative planning method for total knee arthroplasty based on dual-plane imaging according to the present invention.

[0023] Figure 2 is a schematic diagram of the medial and lateral spaces between the femur and tibia in this invention;

[0024] Figure 3 is a schematic diagram of the femur and tibia at multiple angles according to the present invention.

[0025] Explanation of reference numerals in the attached figures:

[0026] 1- Lateral joint space; 2- Medial joint space; 3- Tibial osteotomy surface. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise ratios, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0028] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0029] Figure 1 shows a schematic diagram of the preoperative planning method for total knee arthroplasty based on dual-plane imaging according to the present invention.

[0030] This embodiment discloses a preoperative planning method for total knee arthroplasty based on dual-plane imaging, including the following steps:

[0031] Acquire static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movements of the target object.

[0032] A three-dimensional model of the knee joint of the target object is obtained, and two-dimensional-three-dimensional spatial registration and alignment are performed with the static biplane lower limb image and dynamic biplane image of the knee joint in a standing position, respectively, to obtain a three-dimensional registration image of the knee joint in a standing position and a three-dimensional registration image of the knee joint in functional movement.

[0033] Based on the three-dimensional registration images of the knee joint in a standing position, calculate the functional parameters, morphological parameters, and three-dimensional gaps between the medial and lateral joints of the knee joint. Based on the functional and morphological parameters, determine the tibial prosthesis and the tibial osteotomy plane.

[0034] Based on the aforementioned functional and morphological parameters, the femoral prosthesis is determined. The three-dimensional model of the determined femoral prosthesis is imported. Based on the femoral articular surface and the femoral prosthesis articular surface, the static preliminary placement position of the femoral prosthesis is located using a three-dimensional point cloud registration optimization algorithm, obtaining the static preliminary relative positional relationship P1 between the femoral prosthesis and the femur. With the three-dimensional gap balance of the medial and lateral joints of the knee as the optimization target, the static optimized placement position in the standing position is calculated based on a preset optimization algorithm, obtaining the static optimized relative positional relationship P2 between the femoral prosthesis and the femur.

[0035] The static optimized relative position relationship P2 between the femoral prosthesis and the femur is imported into the three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and the three-dimensional gap between the medial and lateral joints after the femoral prosthesis is placed is simulated and calculated during functional movement. With the goal of balancing the three-dimensional gap between the medial and lateral joints in the full range of flexion, the position of the femoral prosthesis is continuously calculated and adjusted based on the optimization algorithm until a dynamic optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional gap balance in the full range of flexion. The dynamic optimized relative position relationship P3 between the femoral prosthesis and the femur and the femoral osteotomy plane are output.

[0036] The technical solution of this embodiment is based on the static biplane lower limb images in a standing position and the dynamic biplane images of the knee joint during functional movement for preoperative planning. It can not only realize TKA three-dimensional planning based on the joint force line in the weight-bearing position, but also realize the preoperative three-dimensional planning of joint space balance at the full flexion angle of the knee joint, reduce the doctor's operation time during operation, and help patients obtain better postoperative joint function.

[0037] Preferably, before acquiring static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movement, the preoperative planning method for total knee arthroplasty further includes correcting the biplane imaging system to obtain the relative spatial position relationship and imaging parameters of the biplane system, including the following steps: placing a correction box containing multiple steel balls in the imaging space for imaging; solving the spatial positions of the X-ray source and the flat panel detector in the biplane imaging system by using the acquired steel ball images and the known relative spatial positions of the steel balls; calibrating the coordinate system of the biplane imaging system based on the spatial positions of the X-ray source and the flat panel detector.

[0038] First, static spatial correction is performed on the dual-plane imaging system. The dual-plane imaging system consists of two pairs of X-ray generators and flat-panel detectors. A correction box containing multiple steel balls is placed in the imaging space for imaging. By using the acquired images of the steel balls and their known relative spatial positions, the spatial positions of the X-ray source and the flat-panel detectors in the dual-plane system are determined. The dual-plane skeletal images acquired after spatial correction can reconstruct realistic three-dimensional anatomical feature points, providing a basis for subsequent prosthesis placement planning.

[0039] Preferably, the method for obtaining a three-dimensional model of the knee joint of the target object includes: obtaining CT images or magnetic resonance images of the knee joint from multiple angles and the corresponding angle information, and performing three-dimensional modeling based on the CT images or magnetic resonance images from multiple angles and the corresponding angle information to obtain a three-dimensional model of the knee joint.

[0040] This embodiment can acquire a three-dimensional model of the patient's knee joint based on currently available, relatively accurate multi-angle CT / MRI images, thereby improving modeling accuracy and consequently enhancing the accuracy of subsequent prosthesis placement determination. Alternatively, a three-dimensional model can be directly created based on static or dynamic images from the dual-plane X-ray system of this embodiment, thereby increasing the utilization rate of the equipment and reducing the cost per treatment for the patient.

[0041] Preferably, acquiring a three-dimensional model of the knee joint of the target object and performing two-dimensional-three-dimensional spatial registration and alignment with the standing static biplane lower limb image and dynamic biplane image of the knee joint further includes: importing the acquired standing static biplane lower limb image and the dynamic biplane image of the knee joint during functional movement into a preset registration program, importing the three-dimensional model of the knee joint of the target object, and determining the position of the femur and tibia in three-dimensional space by matching the bone contour with the biplane dynamic perspective image in each frame in the reconstructed virtual projection environment.

[0042] This embodiment aligns and registers the dual-plane static / dynamic images with the 3D model based on the obtained 3D model. The static images can be easily implemented by those skilled in the art using this embodiment. The dynamic images can be aligned and registered frame by frame or by selecting a predetermined number of frames to determine the positions of the femur and tibia in 3D space. However, this embodiment does not impose any limitations on the alignment and registration method, as long as it achieves the registration of the image and model.

[0043] Preferably, calculating the functional parameters, morphological parameters, and medial-lateral joint three-dimensional gap of the knee joint based on the three-dimensional registration image of the knee joint in a standing position further includes: obtaining the three-dimensional spatial position of key anatomical feature points based on the three-dimensional registration image of the knee joint in a standing position, wherein the key anatomical feature points include at least the center of the hip joint, the center of the knee joint, the center of the ankle joint, the medial and lateral femoral condyles, and the medial and lateral tibial plateau centers; calculating the functional parameters and morphological parameters of the knee joint, wherein the functional parameters include at least the joint force line, and the morphological parameters include at least the anteroposterior diameter of the femur, the size of the medial and lateral femoral condyles, the length and width of the tibial plateau after internal and external rotation of the posterior femoral condyle, and the posterior tilt angle of the tibial plateau; determining the tibial osteotomy plane based on the functional and morphological parameters of the knee joint, selecting the medial and lateral femoral condyle articular surfaces, and calculating the shortest distance from the medial and lateral femoral condyles to the medial and lateral tibial plateau planes in space as the medial and lateral joint three-dimensional gap of the knee joint.

[0044] In this embodiment, the three-dimensional joint gap refers to the distance between the femoral articular surface and the tibial articular surface, which is equivalent to the distance between surfaces.

[0045] Preferably, the calculation of the knee joint's functional parameters, morphological parameters, and medial-lateral joint three-dimensional gap based on the knee joint standing three-dimensional registration image further includes: if the knee joint functional parameters do not meet preset conditions, after calculating the joint force line, rotating the femur to perform varus / valgus pre-correction and determine the simulated force line, and then determining the tibial osteotomy plane based on the simulated force line. The preset conditions refer to a pre-set threshold describing the degree of varus / valgus deformity.

[0046] After calculating the joint alignment, the degree of varus / valgus deformity in the patient can be obtained. Simulated varus / valgus correction is then performed by pre-rotating the femur, and a simulated alignment is obtained. Subsequent calculations are based on this simulated alignment, allowing the patient to achieve the expected varus / valgus correction effect post-operatively, thereby achieving better joint function. The goal of this embodiment, involving femoral rotation, is to correct varus / valgus deformities in patients.

[0047] Preferably, the static preliminary placement of the femoral prosthesis based on the femoral articular surface and the femoral prosthesis articular surface using a three-dimensional point cloud registration optimization algorithm further includes: setting the points on the femoral articular surface as point cloud P. t Set the point cloud of the femoral prosthesis joint surface as P s The goal is to find an optimal set of rotation transformations R. * Translation and shift transformation T * The goal is to minimize the relative distance between PC2 and point cloud PC1 after rotational and / or translational transformations, where PC2 refers to the point cloud on the surface of the femoral prosthesis and PC1 refers to the point cloud on the surface of the femoral joint. The problem description is... The point cloud is preprocessed and filtered, the rotation and translation transformations are solved, the weights of corresponding points are adjusted, unreasonable corresponding points are identified, the loss value is calculated, the loss value is minimized, the current optimal change is solved to update the rotation and translation transformations, and the iteration continues until convergence is achieved to obtain the static preliminary placement position of the femoral prosthesis and the static preliminary relative positional relationship P1 between the femoral prosthesis and the femur.

[0048] This embodiment uses a point cloud registration optimization algorithm to initially determine the static preliminary placement position of the femoral prosthesis, thereby making the static preliminary placement position closer to the target position, and only minor adjustments are needed subsequently.

[0049] Preferably, the static optimized placement position in the standing position calculated based on the preset optimization algorithm further includes: calculating the medial and lateral gaps between the femoral prosthesis and the tibial osteotomy surface in the initial static placement position after simulated force line correction, and adjusting the femoral prosthesis position to balance the absolute value difference of the medial and lateral gaps; wherein, the loss function is the minimum absolute difference of the medial and lateral gaps: loss1=||d l |-|d m ||, where |d m | and |d l | represents the absolute distance from the medial and lateral condyles of the femoral prosthesis to the tibial osteotomy plane, respectively; the optimization object is the 6-DOF value of the femoral prosthesis relative to the femur [α,β,λ,x,y,z], where α represents the flexion angle of the femoral prosthesis, β represents the varus / valgus angle of the femoral prosthesis, λ represents the internal / external rotation angle of the femoral prosthesis, x represents the anterior / posterior movement of the femoral prosthesis, y represents the lateral / hyperaxial movement of the femoral prosthesis, and z represents the proximal / distal movement of the femoral prosthesis.

[0050] This embodiment adjusts the final femoral prosthesis position based on multiple different degrees of freedom, resulting in more precise positional adjustments and ensuring better postoperative joint function for the patient. In this embodiment, the medial-lateral gap is the distance from the lowest point of the femoral surface to the tibial osteotomy surface. See Figure 2 for a schematic diagram of the medial-lateral gap between the femur and tibia.

[0051] Preferably, iteratively calculating and adjusting the femoral prosthesis position until a dynamically optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional joint clearance balance across the full flexion range further includes: adjusting the placement position of the femoral prosthesis relative to the femur within a preset range, calculating the three-dimensional joint clearance at different flexion angles under different femoral prosthesis positions, until three-dimensional joint clearance balance is achieved across the full flexion range, wherein the loss function is defined as: n represents different buckling positions, ||d l |-|d m || i These represent the absolute differences between the medial and lateral interarticular spaces at each flexion position; the individualized variation curves of the three-dimensional interarticular spaces between the medial and lateral joints during knee flexion, with the loss function set as loss3 = |f(d m )-g(d m )|+|f(d l )-g(d l )|,|f(d)-g(d)| represents the difference between the three-dimensional joint clearance and the target joint clearance after the prosthesis is placed.

[0052] This embodiment employs an optimization algorithm to iteratively calculate the optimal placement position and optimizes the calculation based on multiple positional relationships in a dynamic model, thereby ensuring better joint function for the patient during postoperative movement. Compared to existing technologies that rely on two-dimensional or three-dimensional static preoperative planning, this embodiment clearly achieves superior pre-simulation results, thus providing a better guarantee for the patient's later experience. See Figure 3, which schematically illustrates multiple angles of the femur and tibia.

[0053] Based on the same concept, this invention also provides a preoperative planning system for total knee arthroplasty based on dual-plane imaging, comprising: an image acquisition module for acquiring static dual-plane lower limb images of the knee joint in a standing position and dynamic dual-plane images of the knee joint during functional movement; a model registration module for acquiring a three-dimensional model of the knee joint of the target object and performing two-dimensional-three-dimensional spatial registration and alignment with the static dual-plane lower limb images and dynamic dual-plane images of the knee joint in a standing position, respectively, to obtain a three-dimensional registration image of the knee joint in a standing position and a three-dimensional registration image of the knee joint during functional movement; a calculation module for calculating knee joint functional parameters, morphological parameters, and medial and lateral joint three-dimensional gaps based on the knee joint standing three-dimensional registration image, and determining the tibial prosthesis and tibial osteotomy plane based on the functional and morphological parameters; and a static position determination module for determining the femoral prosthesis based on the functional and morphological parameters, importing the determined three-dimensional model of the femoral prosthesis, and determining the position based on the femoral articular surface and the femoral prosthesis articular surface through three-dimensional... The point cloud registration optimization algorithm locates the initial static placement position of the femoral prosthesis, obtaining the initial static relative positional relationship P1 between the femoral prosthesis and the femur. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides of the knee, a preset optimization algorithm calculates the static optimized placement position in the standing position, obtaining the static optimized relative positional relationship P2 between the femoral prosthesis and the femur. A dynamic position determination module imports the static optimized relative positional relationship P2 between the femoral prosthesis and the femur into three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and simulates and calculates the three-dimensional joint gap between the medial and lateral sides after femoral prosthesis placement during functional movement. With the goal of balancing the three-dimensional joint gap between the medial and lateral sides across the full flexion range, the algorithm iteratively calculates and adjusts the position of the femoral prosthesis until a dynamic optimized placement position is found that achieves three-dimensional joint gap balance across the full flexion range. The module outputs the dynamic optimized relative positional relationship P3 between the femoral prosthesis and the femur, as well as the femoral osteotomy plane.

[0054] The technical solution of this embodiment is based on static biplane lower limb images in a standing position and dynamic biplane images of the knee joint during functional movements for preoperative planning. It can not only achieve three-dimensional planning of TKA based on the joint force line in the weight-bearing position, but also achieve preoperative three-dimensional planning of joint space balance at the full flexion angle of the knee joint, reducing the surgeon's intraoperative operation time and helping patients achieve better postoperative joint function. It solves the problem of inaccurate prosthesis size and position planning caused by inaccurate image magnification and differences in shooting angles in traditional preoperative planning methods that only use single-plane weight-bearing images for TKA preoperative planning. It also overcomes the deficiency of traditional two-dimensional planning in obtaining the true three-dimensional relative position of the prosthesis and bone.

[0055] In this embodiment, some methods not described in detail are implemented using any existing method that can achieve the function or effect, and the present invention does not require any restrictions.

[0056] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0057] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they shall still fall within the protection scope of the present invention.

Claims

1. A preoperative planning method for total knee arthroplasty based on dual-plane imaging, characterized in that, Includes the following steps: Acquire static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movements of the target object. A three-dimensional model of the knee joint of the target object is obtained, and two-dimensional-three-dimensional spatial registration and alignment are performed with the static biplane lower limb image and dynamic biplane image of the knee joint in a standing position, respectively, to obtain a three-dimensional registration image of the knee joint in a standing position and a three-dimensional registration image of the knee joint in functional movement. Based on the three-dimensional registration images of the knee joint in a standing position, calculate the functional parameters, morphological parameters, and three-dimensional gaps between the medial and lateral joints of the knee joint. Based on the functional and morphological parameters, determine the tibial prosthesis and the tibial osteotomy plane. Based on the aforementioned functional and morphological parameters, the femoral prosthesis is determined. The three-dimensional model of the determined femoral prosthesis is imported. Based on the femoral articular surface and the femoral prosthesis articular surface, the static preliminary placement position of the femoral prosthesis is located using a three-dimensional point cloud registration optimization algorithm, obtaining the static preliminary relative positional relationship P1 between the femoral prosthesis and the femur. With the three-dimensional gap balance of the medial and lateral joints of the knee as the optimization target, the static optimized placement position in the standing position is calculated based on a preset optimization algorithm, obtaining the static optimized relative positional relationship P2 between the femoral prosthesis and the femur. The static optimized relative position relationship P2 between the femoral prosthesis and the femur is imported into the three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and the three-dimensional gap between the medial and lateral joints after the femoral prosthesis is placed is simulated and calculated during functional movement. With the goal of balancing the three-dimensional gap between the medial and lateral joints in the full range of flexion, the position of the femoral prosthesis is continuously calculated and adjusted based on the optimization algorithm until a dynamic optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional gap balance in the full range of flexion. The dynamic optimized relative position relationship P3 between the femoral prosthesis and the femur and the femoral osteotomy plane are output.

2. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 1, characterized in that, Before acquiring static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movement, the preoperative planning method for total knee arthroplasty also includes correcting the biplane imaging system to obtain the relative spatial position relationship and imaging parameters of the biplane system, including the following steps: A correction box containing multiple steel balls is placed in the imaging space for imaging. The spatial positions of the X-ray source and the flat panel detector in the dual-plane imaging system are determined by the captured images of the steel balls and the known relative spatial positions of the steel balls. The coordinate system of the dual-plane imaging system is calibrated based on the spatial position of the X-ray source and the flat panel detector.

3. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 1, characterized in that, Methods for obtaining a 3D model of the knee joint of a target object include: Acquire CT or MRI images of the knee joint from multiple angles and the corresponding angle information, and perform 3D modeling based on the CT or MRI images from multiple angles and the corresponding angle information to obtain a 3D model of the knee joint.

4. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 1, characterized in that, Obtaining a 3D model of the knee joint of the target object, and performing 2D-3D spatial registration and alignment with the standing static biplane lower limb image and dynamic biplane image of the knee joint, respectively, further includes: Import the acquired static biplane lower limb images in a standing position and the dynamic biplane knee joint images during functional movements into a preset registration program. Import the three-dimensional model of the knee joint of the target object. By matching the bone contour with the biplane dynamic perspective images in each frame in the reconstructed virtual projection environment, determine the position of the femur and tibia in three-dimensional space.

5. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 1, characterized in that, The calculation of the functional parameters, morphological parameters, and medial-lateral joint three-dimensional gap of the knee joint based on the aforementioned standing three-dimensional registered images of the knee joint further includes: The three-dimensional spatial positions of key anatomical feature points are obtained based on the three-dimensional registration image of the knee joint in standing position. The key anatomical feature points include at least the center of the hip joint, the center of the knee joint, the center of the ankle joint, the medial and lateral femoral condyles, and the medial and lateral centers of the tibial plateau. Calculate the functional and morphological parameters of the knee joint. The functional parameters include at least the joint alignment, and the morphological parameters include at least the anteroposterior diameter of the femur, the size of the medial and lateral femoral condyles, the length and width of the tibial plateau due to internal and external rotation of the posterior femoral condyle, and the posterior tilt angle of the tibial plateau. Based on the functional and morphological parameters of the knee joint, the tibial osteotomy plane is determined, the medial and lateral femoral condyles are selected, and the shortest distance from the medial and lateral femoral condyles to the medial and lateral tibial plateau planes in three-dimensional space is calculated as the three-dimensional joint gap between the medial and lateral knee joints.

6. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 5, characterized in that, The calculation of the functional parameters, morphological parameters, and medial-lateral joint three-dimensional gap of the knee joint based on the aforementioned standing three-dimensional registered images of the knee joint further includes: If the knee joint function parameters do not meet the preset conditions, after calculating the joint force line, the femur is rotated to perform varus / valgus pre-correction and determine the simulated force line, and then the tibial osteotomy plane is determined based on the simulated force line.

7. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 5 or 6, characterized in that, Based on the femoral articular surface and the femoral prosthesis articular surface, a three-dimensional point cloud registration optimization algorithm is used to locate the static preliminary placement position of the femoral prosthesis, which further includes: Let the points on the surface of the femoral joint be a point cloud P. t Set the point cloud of the femoral prosthesis joint surface as P s The goal is to find an optimal set of rotation transformations R. * Translation and shift transformation T * The goal is to minimize the relative distance between PC2 and point cloud PC1 after rotational and / or translational transformations, where PC2 refers to the point cloud on the surface of the femoral prosthesis and PC1 refers to the point cloud on the surface of the femoral joint. The problem description is... The point cloud is preprocessed and filtered, the rotation and translation transformations are solved, the weights of corresponding points are adjusted, unreasonable corresponding points are identified, the loss value is calculated, the loss value is minimized, the current optimal change is solved to update the rotation and translation transformations, and the iteration continues until convergence is achieved to obtain the static preliminary placement position of the femoral prosthesis and the static preliminary relative positional relationship P1 between the femoral prosthesis and the femur.

8. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 6, characterized in that, The static optimized placement position for standing positions, calculated based on a preset optimization algorithm, further includes: Calculate the medial and lateral gaps between the femoral prosthesis and the tibial osteotomy surface in the initial static placement position after simulated force line correction, and adjust the position of the femoral prosthesis to balance the absolute value difference of the medial and lateral gaps. Among them, the loss function is the one where the absolute difference between the inner and outer gaps is minimized: loss1 = ||d l |-|d m ||, where |d m | and |d l | These represent the absolute distances from the medial and lateral condyles of the femoral prosthesis to the tibial osteotomy plane, respectively; The optimization target is the 6-DOF values ​​[α,β,λ,x,y,z] of the femoral prosthesis relative to the femur, where α represents the flexion angle of the femoral prosthesis, β represents the varus / valgus angle of the femoral prosthesis, λ represents the internal / external rotation angle of the femoral prosthesis, x represents the anterior / posterior movement of the femoral prosthesis, y represents the lateral / proximal movement of the femoral prosthesis, and z represents the proximal / distal movement of the femoral prosthesis.

9. The preoperative planning method for total knee arthroplasty based on dual-plane imaging according to claim 1, characterized in that, The femoral prosthesis position is continuously calculated and adjusted until a dynamically optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional joint space balance throughout the full range of flexion. This further includes: The placement of the femoral prosthesis relative to the femur is adjusted within a preset range, and the three-dimensional joint gap at different flexion angles under different femoral prosthesis positions is calculated until the three-dimensional joint gap is balanced across the full flexion range. The loss function is defined as follows: n represents different buckling positions, ||d l |-|d m || i These represent the absolute difference between the medial and lateral intercostal spaces at each buckling position; Personalized three-dimensional space variation curves of the medial and lateral joints during knee flexion, with the loss function set as loss3=|f(d m )-g(d m )|+|f(d l )-g(d l |f(d)-g(d)| represents the difference between the three-dimensional joint clearance and the target joint clearance after the prosthesis is placed.

10. A preoperative planning system for total knee arthroplasty based on dual-plane imaging, characterized in that, include: The image acquisition module is used to acquire static biplane lower limb images of the knee joint in a standing position and dynamic biplane images of the knee joint during functional movements of the target object. The model registration module is used to acquire a 3D model of the knee joint of the target object, and to perform 2D-3D spatial registration and alignment with the static biplane lower limb image and dynamic biplane image of the knee joint in a standing position, respectively, to obtain a 3D registered image of the knee joint in a standing position and a 3D registered image of the knee joint in functional movement. Quasi-image; The calculation module is used to calculate the knee joint functional parameters, morphological parameters and medial and lateral joint three-dimensional gaps based on the knee joint standing three-dimensional registration image, and to determine the tibial prosthesis and tibial osteotomy plane based on the functional parameters and morphological parameters. The static position determination module is used to determine the femoral prosthesis based on the functional parameters and morphological parameters, import the three-dimensional model of the determined femoral prosthesis, and locate the static preliminary placement position of the femoral prosthesis based on the femoral articular surface and the femoral prosthesis articular surface through a three-dimensional point cloud registration optimization algorithm to obtain the static preliminary placement relative position relationship P1 between the femoral prosthesis and the femur; with the three-dimensional gap balance of the medial and lateral joints of the knee as the optimization target, the static optimized placement position in the standing position is calculated based on a preset optimization algorithm to obtain the static optimized placement relative position relationship P2 between the femoral prosthesis and the femur; The dynamic position determination module is used to import the static optimized relative position relationship P2 between the femoral prosthesis and the femur into the three-dimensional registration images of the knee joint at multiple flexion angles during functional movement, and to simulate and calculate the three-dimensional gap between the medial and lateral joints after the femoral prosthesis is placed during functional movement. With the goal of balancing the three-dimensional gap between the medial and lateral joints in the full flexion range, the module iteratively calculates and adjusts the position of the femoral prosthesis based on the optimization algorithm until a dynamic optimized placement position is found that allows the femoral prosthesis to achieve three-dimensional gap balance in the full flexion range. The module outputs the dynamic optimized relative position relationship P3 between the femoral prosthesis and the femur and the femoral osteotomy plane.

Images