Acetabular prosthesis placement guidance system and method based on feature point registration and augmented reality

Abstract

The present invention discloses an acetabular prosthesis placement guidance system and method based on feature point registration and augmented reality, the method comprising: obtaining hip joint image information and three-dimensional reconstruction of hip joint model; solving the corresponding acetabular cup model; importing the corresponding acetabular cup and simulating the placement of the acetabular cup; solving and visualizing the placement direction of the acetabular prosthesis; using the virtual three-dimensional hip joint model as a floating image, using the actual hip joint surgical site as a reference image, solving the rotation and translation matrix based on the quaternion method of selecting feature point matching to perform virtual and real scene registration; using augmented reality technology to render the registered virtual planning path to the actual scene in real time; using the Zhang-style calibration method to realize the hand-eye calibration of the robot arm, obtaining the transformation matrix from the camera coordinate system to the robot arm coordinate system, and guiding the robot arm to grab the acetabular rasp to reach the spatial position of the preoperative planning path. The present invention reduces the error caused by subjective judgment alone, improves the accuracy and safety of prosthesis placement, and shortens the prosthesis placement time.

Description

Technical Field

[0001] The present invention relates to the technical field of computer graphics and image processing, and in particular to an acetabular prosthesis placement guidance system and method based on feature point registration and augmented reality. Background Art

[0002] Studies have found that incorrect implantation position during surgery is an important cause of various complications such as joint instability and dislocation after hip replacement, and accurate placement of the acetabular prosthesis is the most difficult part of prosthesis placement. Therefore, many experts at home and abroad believe that accurate placement of the acetabular prosthesis is the key to the success of the operation.

[0003] Due to the particularity and complexity of the hip joint structure and the influence of multiple potential factors, such as changes in the patient's position during surgery, the surgeon's visual errors, deformation or loss of reference markers, etc., there is no unified standard for the precise placement of acetabular prostheses, resulting in inaccurate positioning, high surgical difficulty and risk.

[0004] At present, domestic doctors mostly use CT images to understand the patient's lesion information before hip replacement surgery. However, CT images are two-dimensional images. Doctors can only rely on experience to estimate the size and shape of the lesion from the sequence of two-dimensional images and imagine the relationship between the lesion and the surrounding tissues, which brings certain difficulties to the formulation of surgical plans. In hip replacement surgery, the placement of the acetabular prosthesis is mainly based on clinical experience and naked eye observation, and there is a problem of inaccurate prosthesis implantation. Summary of the invention

[0005] The purpose of the present invention is to overcome the problem of inaccurate implantation of acetabular prosthesis in the prior art, and proposes an acetabular prosthesis placement guidance system and method based on feature point alignment and augmented reality. The present invention reduces the errors caused by subjective judgment alone, improves the accuracy and safety of acetabular prosthesis placement, and shortens the prosthesis placement time.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The present invention provides an acetabular prosthesis placement guidance system based on feature point registration and augmented reality, comprising: a hip joint image analysis module, a hip joint model three-dimensional reconstruction module, an acetabular cup model solving module, an acetabular cup placement simulation module, a prosthesis placement path planning module, a depth camera information acquisition module, a rotation and translation matrix solving module, a virtual and real scene registration module, and a robotic arm guidance module;

[0008] The hip joint image analysis module is used to parse the DICOM format file of the hip joint part to obtain the hip joint image information;

[0009] The hip joint model three-dimensional reconstruction module is used to obtain a virtual three-dimensional hip joint model based on a marching cube algorithm;

[0010] The acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting;

[0011] The simulated acetabular cup placement module is used to import the corresponding acetabular cup in STL file format, and use the Lewinnek safety zone as a reference standard for acetabular prosthesis placement to simulate the acetabular cup placement;

[0012] The prosthesis placement path planning module is used to determine the path direction of the acetabular grinding and implanting acetabular prosthesis and visualize it according to the acetabular cup edge information and interactive point picking technology;

[0013] The depth camera information acquisition module is used to acquire the actual hip joint surgery site target model video scene, perform format analysis on the video image, and acquire the color information and depth information of the video image;

[0014] The rotation and translation matrix solving module is used to use the virtual three-dimensional hip joint model as a floating image, the actual hip joint surgery site target model in the video scene as a reference image, and solve the rotation matrix and translation matrix based on the quaternion method of selecting feature point matching;

[0015] The virtual-real scene registration module is used to apply the solved rotation and translation matrix to the virtual scene to achieve virtual-real scene registration, and at the same time use augmented reality technology to render the registered virtual planning path to the actual scene in real time;

[0016] The robotic arm guidance module is used to implement hand-eye calibration of the robotic arm based on Zhang's calibration method, obtain the transformation matrix from the camera coordinate system to the robotic arm coordinate system, thereby guiding the robotic arm to grasp the acetabular rasp to the spatial position of the preoperative planned path, and place the acetabular rasp and acetabular prosthesis in combination with augmented reality effects.

[0017] As a preferred technical solution, the acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting, specifically including:

[0018] The virtual three-dimensional hip joint model was transformed into a three-dimensional posture so that the projection area of ​​the acetabular cross section on the computer screen was maximized. Interactive point picking technology was used to select multiple feature points on the acetabular cross section to fit the size of the acetabular cup. The corresponding outer diameter of the acetabular cup was solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

[0019] As a preferred technical solution, the prosthesis placement path planning module is used to determine and visualize the path direction of the acetabular grinding and implanting acetabular prosthesis based on the acetabular cup edge information and interactive point picking technology, specifically including:

[0020] According to the edge information of the acetabular cup, the interactive point picking technology is used to select feature points to construct the acetabular cup plane. The acetabular grinding and rasp points are selected according to the center area of ​​the plane. The normal vector of the acetabular grinding and rasp points based on the plane is solved according to the three-dimensional spatial geometric relationship. The marching cube algorithm is used to visualize the normal vector as the path direction of the acetabular grinding and rasp and the implanted acetabular prosthesis.

[0021] As a preferred technical solution, the rotation and translation matrix solving module is used to use the virtual three-dimensional hip joint model as a floating image, the actual hip joint surgery site target model in the video scene as a reference image, and solve the rotation matrix and translation matrix based on the quaternion method of selecting feature point matching, specifically including:

[0022] Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i ,i=0,1,2,…,n};

[0023] Set the objective function to:

[0024] Among them, R represents the rotation matrix and T represents the translation matrix;

[0025] Find the centroid of the target point set U and the reference point set X respectively, expressed as:

[0026]

[0027] The covariance matrix K is constructed based on the target point set U and the reference point set X, which is expressed as:

[0028]

[0029] Based on the element K in the covariance matrix K ij Construct a symmetric matrix N, expressed as:

[0030]

[0031] Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the symmetric matrix N are calculated, and the unit eigenvector q=[q0 q1 q2 q3] corresponding to the maximum eigenvalue is obtained. The rotation matrix R is further solved, which is expressed as:

[0032]

[0033] Solve the translation matrix T based on the rotation matrix R, expressed as:

[0034] T=μ X -Rμ U

[0035] Among them, μ U represents the centroid of the target point set U, μ X represents the centroid of the reference point set X.

[0036] The present invention also provides an acetabular prosthesis placement guidance method based on feature point registration and augmented reality, comprising the following steps:

[0037] Parse the DICOM format file of the hip joint to obtain the hip joint image information;

[0038] 3D reconstruction of virtual 3D hip joint model based on marching cube algorithm;

[0039] The corresponding acetabular cup model is solved based on interactive point picking technology and least square fitting method;

[0040] The acetabular cup model was imported in STL file format, and the acetabular cup placement was simulated using the Lewinnek safety zone as the reference standard for acetabular prosthesis placement;

[0041] Determine and visualize the path direction of acetabular grinding and implantation of acetabular prosthesis based on acetabular cup edge information and interactive point picking technology;

[0042] Use a depth camera to obtain a video image of the actual hip joint surgical site target model, and analyze the color information and depth information of the image;

[0043] The virtual 3D hip joint model is used as a floating image, the actual hip joint surgical site target model is used as a reference image, and the rotation matrix and translation matrix are solved based on the quaternion method of selecting feature point matching;

[0044] The spatial transformation relationship between the virtual scene and the actual scene is obtained according to the rotation matrix and the translation matrix to achieve the registration of the virtual and real scenes;

[0045] Use augmented reality technology to render the registered virtual planning path into the actual scene in real time;

[0046] The Zhang-style calibration method is used to realize the hand-eye calibration of the robotic arm, obtain the transformation matrix from the camera coordinate system to the robotic arm coordinate system, transform the pose information corresponding to the virtual planned path in the depth camera coordinate system to the robotic arm coordinate system, guide the robotic arm to grasp the acetabular reamer to the virtual planned path, and combine the augmented reality effect to place the acetabular reamer and acetabular prosthesis.

[0047] As a preferred technical solution, the method of solving the corresponding acetabular cup model based on interactive point picking and least squares fitting specifically includes:

[0048] The virtual three-dimensional hip joint model was transformed into a three-dimensional posture so that the projection area of ​​the acetabular cross section on the computer screen was maximized. Interactive point picking technology was used to select multiple feature points on the acetabular cross section to fit the size of the acetabular cup. The corresponding outer diameter of the acetabular cup was solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

[0049] As a preferred technical solution, the method of determining the path direction of the acetabular grinding and implanting the acetabular prosthesis according to the acetabular cup edge information and the interactive point picking technology and visualizing it specifically includes:

[0050] According to the edge information of the acetabular cup, the interactive point picking technology is used to select feature points to construct the acetabular cup plane. The acetabular grinding and rasp points are selected according to the center area of ​​the plane. The normal vector of the acetabular grinding and rasp points based on the plane is solved according to the three-dimensional spatial geometric relationship. The marching cube algorithm is used to visualize the normal vector as the path direction of the acetabular grinding and rasp and the implanted acetabular prosthesis.

[0051] As a preferred technical solution, the virtual three-dimensional hip joint model is used as a floating image, the actual hip joint surgery site target model is used as a reference image, and the rotation matrix and translation matrix are solved based on the quaternion method of selecting feature point matching, which specifically includes:

[0052] Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i ,i=0,1,2,…,n};

[0053] Set the objective function to:

[0054] Among them, R represents the rotation matrix and T represents the translation matrix;

[0055] Find the centroid of the target point set U and the reference point set X respectively, expressed as:

[0056]

[0057] The covariance matrix K is constructed based on the target point set U and the reference point set X, which is expressed as:

[0058]

[0059] Based on the element K in the covariance matrix K ij Construct a symmetric matrix N, expressed as:

[0060]

[0061] Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the symmetric matrix N are calculated, and the unit eigenvector q=[q0 q1 q2 q3] corresponding to the maximum eigenvalue is obtained. The rotation matrix R is further solved, which is expressed as:

[0062]

[0063] Solve the translation matrix T based on the rotation matrix R, expressed as:

[0064] T=μ X -Rμ U

[0065] Among them, μ U represents the centroid of the target point set U, μ X represents the centroid of the reference point set X.

[0066] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0067] (1) The present invention utilizes interactive point picking technology and least squares fitting to solve the acetabular cup model, uses the Lewinnek safety zone as a reference standard to simulate the acetabular cup placement, solves the acetabular grinding and acetabular prosthesis placement path based on the three-dimensional spatial geometric relationship and visualizes it, so as to achieve the purpose of planning the prosthesis placement path.

[0068] (2) The present invention does not rely on artificial markings, but achieves matching between virtual scenes and actual scenes based on feature points, and uses augmented reality technology to render the registered virtual planning path into the actual scene in real time, thereby achieving the purpose of accurate placement of the prosthesis.

[0069] (3) The present invention is combined with a robotic arm guidance module, and the robotic arm can grab the acetabular rasp to reach the planned reference path, thereby reducing the error caused by subjective judgment alone, improving the accuracy and safety of prosthesis placement, and shortening the prosthesis placement time. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a schematic diagram of the framework structure of the acetabular prosthesis placement guidance system based on feature point registration and augmented reality of the present invention;

[0071] Figure 2 It is a flow chart of the acetabular prosthesis placement guidance method based on feature point registration and augmented reality of the present invention;

[0072] Figure 3 This is a schematic diagram of the effect of the virtual planning path of the present invention;

[0073] Figure 4 (a) is a schematic diagram of the effect of the virtual acetabular cup and prosthesis placement path aligned with the present invention;

[0074] Figure 4 (b) is a schematic diagram showing the effect of only displaying the virtual prosthesis placement path of the present invention;

[0075] Figure 5 It is a schematic diagram of the guiding effect of the robot arm of the present invention. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0077] Example 1

[0078] like Figure 1 As shown, this embodiment provides an acetabular prosthesis placement guidance system based on feature point registration and augmented reality, including: a hip joint image analysis module, a hip joint model three-dimensional reconstruction module, an acetabular cup model solving module, an acetabular cup placement simulation module, a prosthesis placement path planning module, a depth camera information acquisition module, a rotation and translation matrix solving module, a virtual and real scene registration module, and a robotic arm guidance module;

[0079] In this embodiment, the hip joint image parsing module is used to parse the DICOM format file of the hip joint part to obtain the hip joint image information;

[0080] In this embodiment, the hip joint model three-dimensional reconstruction module is used to obtain a virtual three-dimensional hip joint model based on a marching cube algorithm;

[0081] In this embodiment, the acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting;

[0082] In this embodiment, the acetabular cup placement simulation module is used to import the corresponding acetabular cup in STL (StereoLithography) file format, and use the Lewinnek safety zone as a reference standard for acetabular prosthesis placement to perform planning simulation of acetabular cup placement;

[0083] In this embodiment, the prosthesis placement path planning module is used to determine the acetabular grinding and implanting path direction of the acetabular prosthesis according to the acetabular cup edge information and the interactive point picking technology and visualize the visualization result. Figure 3 shown.

[0084] In this embodiment, the depth camera information acquisition module is used to acquire the actual hip joint surgery site target model video scene, perform format analysis on the video image, and acquire the color information and depth information of the video image;

[0085] In this embodiment, the rotation and translation matrix solving module is used to use the virtual three-dimensional hip joint model as a floating image, the actual hip joint surgery site target model in the video scene as a reference image, and solve the rotation matrix and translation matrix based on the quaternion method of selecting feature point matching;

[0086] In this embodiment, the virtual-real scene registration module is used to apply the solved rotation and translation matrix to the virtual scene to achieve virtual-real scene registration, and at the same time, the registered virtual planning path is rendered to the actual scene in real time using augmented reality technology. Figure 4 (a) and Figure 4 As shown in (b), the registered virtual acetabular cup and prosthesis placement path are obtained, as well as the registered rendering effect of only displaying the virtual prosthesis placement path;

[0087] In this embodiment, the robot arm guidance module is used to implement the hand-eye calibration of the robot arm based on the Zhang calibration method, obtain the transformation matrix from the camera coordinate system to the robot arm coordinate system, so as to guide the robot arm to grasp the acetabular rasp to the spatial position of the preoperative planned path, and place the acetabular rasp and acetabular prosthesis in combination with the augmented reality effect. The robot arm guidance effect is as follows: Figure 5 shown.

[0088] In this embodiment, the acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting, specifically including:

[0089] The virtual three-dimensional hip joint model was transformed into a three-dimensional posture so that the projection area of ​​the acetabular cross section on the computer screen was maximized. Interactive point picking technology was used to select multiple feature points on the acetabular cross section to fit the acetabular cup model. The corresponding acetabular cup outer diameter size was solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

[0090] In this embodiment, the path direction visualization module is used to determine and visualize the path direction of the acetabular grinding and implanting of the acetabular prosthesis according to the acetabular cup edge information and the interactive point picking technology, specifically including:

[0091] According to the edge information of the acetabular cup, the interactive point picking technology is used to select feature points to construct the acetabular cup plane. The acetabular grinding and rasp points are selected according to the center area of ​​the plane. The normal vector of the acetabular grinding and rasp points based on the plane is solved according to the three-dimensional spatial geometric relationship. The marching cube algorithm is used to visualize the normal vector as the path direction of the acetabular grinding and rasp and the implanted acetabular prosthesis.

[0092] In this embodiment, the rotation and translation matrix solving module is used to use the virtual three-dimensional hip joint model as a floating image, the actual hip joint surgery site target model in the video scene as a reference image, and solve the rotation matrix and translation matrix based on the quaternion method of selecting feature point matching, specifically including:

[0093] Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i ,i=0,1,2,…,n};

[0094] Set the objective function to:

[0095] Among them, R represents the rotation matrix and T represents the translation matrix;

[0096] Find the centroid of the target point set U and the reference point set X respectively, expressed as:

[0097]

[0098] The covariance matrix K is constructed based on the target point set U and the reference point set X, which is expressed as:

[0099]

[0100] Based on the element K in the covariance matrix K ij (i, j = 1, 2, 3) constructs a symmetric matrix N, expressed as:

[0101]

[0102] Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the symmetric matrix N are calculated, and the unit eigenvector q=[q0 q1 q2 q3] corresponding to the maximum eigenvalue is obtained. The rotation matrix R is further solved, which is expressed as:

[0103]

[0104] Solve the translation matrix T based on the rotation matrix R, expressed as:

[0105] T=μ X -Rμ U

[0106] Among them, μ U represents the centroid of the target point set U, μ X represents the centroid of the reference point set X.

[0107] Example 2

[0108] like Figure 2 As shown, this embodiment provides an acetabular prosthesis placement guidance method based on feature point registration and augmented reality, comprising the following steps:

[0109] Step 1: Parse the DICOM format file of the hip joint to obtain the image information it contains, obtain the three-dimensional hip joint model based on the moving cube algorithm and display it on the computer screen.

[0110] Step 2: Perform a three-dimensional posture transformation on the virtual three-dimensional hip joint model to maximize the projection area of ​​the acetabular cross section on the computer screen. Use interactive point picking technology to select a series of feature points on the acetabular cross section to fit the acetabular cup model. After the feature points are selected, the corresponding acetabular cup outer diameter size is solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

[0111] Step 3: Import the determined acetabular cup model in STL file format and display it in the same window. Use the Lewinnek safety zone as the reference standard for acetabular prosthesis placement to plan and simulate the placement of the acetabular cup.

[0112] Step 4: Based on the edge information of the placed acetabular cup model, the interactive point picking technology is used to select feature points to construct the acetabular cup plane, and then the acetabular grinding and filing points are selected according to the center area of ​​the plane; the normal vector of the acetabular grinding and filing points based on the plane is solved using the three-dimensional spatial geometric relationship, and the marching cube algorithm is used to visualize the direction of the virtual normal vector for easy observation. The visualization effect is as follows: Figure 3 This normal vector direction is the direction of the acetabular reamer and implanted acetabular prosthesis planned before surgery, and this normal vector direction is used as the reference standard for the direction of the acetabular reamer and implanted acetabular prosthesis during surgery.

[0113] Step 5: Use a 3D depth camera to obtain a 3D video scene of the 3D printed hip joint model, parse the video image format, and obtain color information and depth information of the video image.

[0114] Step 6: The virtual 3D hip joint model reconstructed in step 1 is used as a floating image, and the 3D printed hip joint model in the video scene is used as a reference image. The rotation matrix and the translation matrix are solved by using the quaternion method based on the matching of selected feature points to realize the registration of the virtual floating image to the actual reference image, which specifically includes the following steps:

[0115] 6.1) Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i , i=0,1,2,…,n}, the two point sets contain the same number of points. In order to align the target point set U with the reference point set X, it is necessary to solve the optimal rotation matrix R and translation matrix T to satisfy the objective function: Where R is a 3×3 rotation matrix, T is a 3×1 translation matrix.

[0116] 6.2) Find the centroid of the target point set U and the reference point set X respectively.

[0117]

[0118] 6.3) Construct the covariance matrix K based on the target point set U and the reference point set X.

[0119]

[0120] 6.4) Based on the element K in the covariance matrix K ij (i,j=1,2,3) constructs a 4×4 symmetric matrix N.

[0121]

[0122] 6.5) Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the matrix N are calculated, and the unit eigenvector q = [q0 q1 q2 q3] corresponding to the maximum eigenvalue is obtained. Then the rotation matrix R can be further solved.

[0123]

[0124] 6.6) Solve the translation matrix T based on the rotation matrix R.

[0125] T=μ X -Rμ U

[0126] Step 7: Based on the calculated R, T matrix, the spatial transformation relationship between the preoperative virtual scene and the actual scene during the operation can be obtained, thereby realizing the registration of the virtual and real scenes. Then, the augmented reality fusion technology is used to render the registered virtual planning path to the actual scene during the operation in real time, such as Figure 4 As shown in (a) and FIG4(b), the registered virtual acetabular cup and prosthesis placement path, as well as the registered rendering effect showing only the virtual prosthesis placement path, are obtained;

[0127] Step 8: Implement hand-eye calibration of the robotic arm. Use Zhang’s calibration method to calculate the relative position relationship between the robotic arm and the 3D depth camera, and obtain the transformation matrix from the camera coordinate system to the robotic arm coordinate system. From step 7, we can obtain the pose information of the virtual planning path in the depth camera coordinate system, and then use the solved It can be converted to the robot arm coordinate system. Finally, according to the posture information of the virtual planning path in the robot arm coordinate system, the robot arm is guided to grab the acetabular rasp and reach the planned spatial position. At the same time, combined with the augmented reality effect, the rasp and acetabular prosthesis are placed. Figure 5 shown.

[0128] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. An acetabular prosthesis placement guidance system based on feature point registration and augmented reality, characterized in that: include: Hip joint image analysis module, hip joint model 3D reconstruction module, acetabular cup model solving module, acetabular cup placement simulation module, prosthesis placement path planning module, depth camera information acquisition module, rotation and translation matrix solving module, virtual and real scene registration module, and robotic arm guidance module; The hip joint image analysis module is used to parse the DICOM format file of the hip joint part to obtain the hip joint image information; The hip joint model three-dimensional reconstruction module is used to obtain a virtual three-dimensional hip joint model based on a marching cube algorithm; The acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting; The simulated acetabular cup placement module is used to import the corresponding acetabular cup in STL file format, and use the Lewinnek safety zone as a reference standard for acetabular prosthesis placement to simulate the acetabular cup placement; The prosthesis placement path planning module is used to determine the path direction of the acetabular grinding and implanting acetabular prosthesis and visualize it according to the acetabular cup edge information and interactive point picking technology; The depth camera information acquisition module is used to acquire the actual hip joint surgery site target model video scene, perform format analysis on the video image, and acquire the color information and depth information of the video image; The rotation and translation matrix solving module is used to use the virtual three-dimensional hip joint model as a floating image, and the actual hip joint surgery site target model in the video scene as a reference image, and solve the rotation matrix and translation matrix based on the quaternion method of selecting feature point matching, specifically including: Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i ,i=0,1,2,…,n}; Set the objective function to: Among them, R represents the rotation matrix and T represents the translation matrix; Find the centroid of the target point set U and the reference point set X respectively, expressed as: The covariance matrix K is constructed based on the target point set U and the reference point set X, which is expressed as: Based on the element K in the covariance matrix K mn Construct a symmetric matrix N, expressed as: Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the symmetric matrix N are calculated, and the unit eigenvector q=[q0q1q2q3] corresponding to the maximum eigenvalue is obtained. The rotation matrix R is further solved, which is expressed as: Solve the translation matrix T based on the rotation matrix R, expressed as: T=μ X -Rμ U Among them, μ U represents the centroid of the target point set U, μ X represents the centroid of the reference point set X; The virtual-real scene registration module is used to apply the solved rotation and translation matrix to the virtual scene to achieve virtual-real scene registration, and at the same time use augmented reality technology to render the registered virtual planning path to the actual scene in real time; The robotic arm guidance module is used to implement hand-eye calibration of the robotic arm based on Zhang's calibration method, obtain the transformation matrix from the camera coordinate system to the robotic arm coordinate system, thereby guiding the robotic arm to grasp the acetabular rasp to the spatial position of the preoperative planned path, and place the acetabular rasp and acetabular prosthesis in combination with augmented reality effects.

2. The acetabular prosthesis placement guidance system based on feature point registration and augmented reality according to claim 1, characterized in that: The acetabular cup model solving module is used to solve the corresponding acetabular cup model based on interactive point picking technology and least squares fitting, specifically including: The virtual three-dimensional hip joint model was transformed into a three-dimensional posture so that the projection area of ​​the acetabular cross section on the computer screen was maximized. Interactive point picking technology was used to select multiple feature points on the acetabular cross section to fit the size of the acetabular cup. The corresponding outer diameter of the acetabular cup was solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

3. The acetabular prosthesis placement guidance system based on feature point registration and augmented reality according to claim 1, characterized in that: The prosthesis placement path planning module is used to determine and visualize the path direction of the acetabular grinding and implanting acetabular prosthesis according to the acetabular cup edge information and the interactive point picking technology, and specifically includes: According to the edge information of the acetabular cup, the interactive point picking technology is used to select feature points to construct the acetabular cup plane. The acetabular grinding and rasp points are selected according to the center area of ​​the plane. The normal vector of the acetabular grinding and rasp points based on the plane is solved according to the three-dimensional spatial geometric relationship. The marching cube algorithm is used to visualize the normal vector as the path direction of the acetabular grinding and rasp and the implanted acetabular prosthesis.

4. A method for guiding acetabular prosthesis placement based on feature point registration and augmented reality, characterized in that: The steps include: Parse the DICOM format file of the hip joint to obtain the hip joint image information; 3D reconstruction of virtual 3D hip joint model based on marching cube algorithm; The corresponding acetabular cup model is solved based on interactive point picking technology and least square fitting method; The acetabular cup model was imported in STL file format, and the acetabular cup placement was simulated using the Lewinnek safety zone as the reference standard for acetabular prosthesis placement; Determine and visualize the path direction of acetabular grinding and implantation of acetabular prosthesis based on acetabular cup edge information and interactive point picking technology; Use a depth camera to obtain a video image of the actual hip joint surgical site target model, and analyze the color information and depth information of the image; The virtual 3D hip joint model is used as a floating image, and the actual hip joint surgical site target model is used as a reference image. The rotation matrix and translation matrix are solved based on the quaternion method of selecting feature point matching, including: Obtain the target point set U = {u i , i=0,1,2,…,n}, obtain the reference point set X={x i ,i=0,1,2,…,n}; Set the objective function to: Among them, R represents the rotation matrix and T represents the translation matrix; Find the centroid of the target point set U and the reference point set X respectively, expressed as: The covariance matrix K is constructed based on the target point set U and the reference point set X, which is expressed as: Based on the element K in the covariance matrix K mn Construct a symmetric matrix N, expressed as: Based on the Jacobi method, the eigenvalues ​​and eigenvectors corresponding to the symmetric matrix N are calculated, and the unit eigenvector q=[q0q1q2q3] corresponding to the maximum eigenvalue is obtained. The rotation matrix R is further solved, which is expressed as: Solve the translation matrix T based on the rotation matrix R, expressed as: T=μ X -Rμ U Among them, μ U represents the centroid of the target point set U, μ X represents the centroid of the reference point set X; The spatial transformation relationship between the virtual scene and the actual scene is obtained according to the rotation matrix and the translation matrix to achieve the registration of the virtual and real scenes; Use augmented reality technology to render the registered virtual planning path into the actual scene in real time; The Zhang-style calibration method is used to realize the hand-eye calibration of the robotic arm, obtain the transformation matrix from the camera coordinate system to the robotic arm coordinate system, transform the pose information corresponding to the virtual planned path in the depth camera coordinate system to the robotic arm coordinate system, guide the robotic arm to grasp the acetabular reamer to the virtual planned path, and combine the augmented reality effect to place the acetabular reamer and acetabular prosthesis.

5. The acetabular prosthesis placement guidance method based on feature point registration and augmented reality according to claim 4, characterized in that: The method of solving the corresponding acetabular cup model based on the interactive point picking technology and the least squares method fitting specifically includes: The virtual three-dimensional hip joint model was transformed into a three-dimensional posture so that the projection area of ​​the acetabular cross section on the computer screen was maximized. Interactive point picking technology was used to select multiple feature points on the acetabular cross section to fit the size of the acetabular cup. The corresponding outer diameter of the acetabular cup was solved based on the least squares method to determine the acetabular cup model required for total hip replacement.

6. The acetabular prosthesis placement guidance method based on feature point registration and augmented reality according to claim 4, characterized in that: The method of determining the path direction of the acetabular grinding and implanting the acetabular prosthesis according to the acetabular cup edge information and the interactive point picking technology and visualizing the path direction of the acetabular grinding and implanting the acetabular prosthesis specifically includes: According to the edge information of the acetabular cup, the interactive point picking technology is used to select feature points to construct the acetabular cup plane. The acetabular grinding and rasp points are selected according to the center area of ​​the plane. The normal vector of the acetabular grinding and rasp points based on the plane is solved according to the three-dimensional spatial geometric relationship. The marching cube algorithm is used to visualize the normal vector as the path direction of the acetabular grinding and rasp and the implanted acetabular prosthesis.

Images