Verification method, equipment and storage medium for TCP accuracy of TKA surgical robot

Abstract

The present invention proposes a method for verifying the TCP accuracy of a TKA surgical robot, comprising the following steps: resecting the distal portion of a femoral model; controlling a robotic arm to move a saw blade to each osteotomy plane of the femoral model, respectively, to obtain the position data of each osteotomy plane of the femoral model; obtaining the position data of each osteotomy plane of the prosthesis model; calculating a transformation matrix from the femoral model to the prosthesis model based on the position data of each osteotomy plane of the femoral model and the prosthesis model; using the transformation matrix to move the femoral model to align the femoral model with the prosthesis model, and judging the TCP accuracy based on the degree of matching between the femoral model and the prosthesis model. The present invention utilizes the matrix transformation relationship between each osteotomy plane of the femoral model and each osteotomy plane of the prosthesis model to align the femoral model and the prosthesis model in the same three-dimensional image space, thereby more accurately judging the angular impact of the saw blade TCP error on each osteotomy plane, and intuitively visualizing the osteotomy plane offset results.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a method, device and storage medium for verifying the TCP accuracy of a TKA surgical robot. Background Art

[0002] The TCP (Tool Center Point) of a surgical robot refers to the position of the end effector carried by the robotic arm under the end flange of the robotic arm itself. It is usually described using a 4x4 rigid body transformation matrix or translation vector combined with Euler angles. The accuracy of the surgical robot's TCP is a prerequisite for its ability to accurately position the actuator.

[0003] During robot-assisted total knee arthroplasty (TKA), the robot must precisely position the saw blade on five planes of the patient's distal femur. These five cutting planes must perfectly align with the five bony interfaces of the planned femoral prosthesis to ensure smooth prosthesis installation. The accuracy of the robot's saw blade's TCP directly impacts the success of this process, making it crucial to verify TCP accuracy before the actual procedure.

[0004] There are currently two common ways to verify the TCP of saw blades:

[0005] 1) A groove point is set on the saw blade to accommodate the optical probe, and the binocular positioning camera and the robotic arm are calibrated with their hand and eye to establish a coordinate conversion relationship between the two. The optical probe is then used to collect the coordinate value of the saw blade groove point in the camera coordinate system, and this value is compared with the coordinate value of the saw blade groove point in the camera coordinate system transmitted back by the robotic arm. The closer the two are (the smaller the point-to-point distance), the more accurate the robotic arm's understanding of the saw blade's position, that is, the more accurate the saw blade TCP.

[0006] 2) Perform osteotomy tests on the bone model and check the fit between the prosthesis and the cut bone model. The better the fit, the more accurate the TCP.

[0007] The disadvantage of the first method is that the distance error between the two points cannot fully reflect the influence of the saw blade TCP error on the accuracy of positioning the five osteotomy planes, and it cannot rule out the situation where the point error is small but the actual osteotomy plane angle deviates greatly.

[0008] The disadvantage of the second method is that the high-frequency vibration of the saw blade during osteotomy makes it difficult to ensure that the bone model's position will not shift due to this vibration. As a result, the final bone model cross-section does not fully reflect the true positioning accuracy of the robotic arm, making it impossible to assess the accuracy of the TCP. Furthermore, this method is subject to significant subjective influences in determining fit, and cannot provide a quantitative value.

[0009] Therefore, there is an urgent need to invent a TCP accuracy verification method that can fully reflect the impact of saw blade TCP error on the accuracy of positioning the five osteotomy planes. Summary of the Invention

[0010] The present invention proposes a method, device and storage medium for verifying the TCP accuracy of a TKA surgical robot, which solves the problem that the TCP accuracy verification method in the prior art cannot fully reflect the impact of the saw blade TCP error on the accuracy of positioning the five osteotomy planes.

[0011] The technical solution of the present invention is achieved as follows:

[0012] The first aspect of the present invention provides a method for verifying the accuracy of the TCP of a TKA surgical robot, comprising the following steps:

[0013] The distal part of the femoral model was removed;

[0014] Control the robotic arm to move the saw blade to each osteotomy plane of the femoral model and obtain the position data of each osteotomy plane of the femoral model;

[0015] Obtain the posture data of each osteotomy plane of the prosthesis model;

[0016] According to the posture data of each osteotomy plane of the femoral model and the prosthesis model, the transformation matrix from the femoral model to the prosthesis model is calculated;

[0017] The femoral model was moved using the transformation matrix to align the femoral model with the prosthesis model, and the accuracy of TCP was determined based on the degree of matching between the femoral model and the prosthesis model.

[0018] By pre-removing the distal part of the femoral model, the present invention can ensure that the saw blade at the end of the robotic arm can be normally positioned to the various osteotomy planes of the femoral model, and at the same time avoid the saw blade from contacting the femoral model and causing the position of the femoral model to shift; by utilizing the matrix conversion relationship between the various osteotomy planes of the femoral model and the various osteotomy planes of the prosthesis model, the femoral model and the prosthesis model are aligned in the same three-dimensional image space, so that the angular influence of the saw blade TCP error on the various osteotomy planes can be more accurately judged, and the osteotomy plane offset results can be intuitively visualized.

[0019] Specifically, the method for obtaining the posture data of each osteotomy plane of the femoral model includes the following steps:

[0020] Set at least three non-collinear laser target points on the osteotomy surface of the saw blade;

[0021] Control the robotic arm to move the saw blade to the distal osteotomy plane, anterior malleolus osteotomy plane, posterior malleolus osteotomy plane, anterior oblique osteotomy plane and posterior oblique osteotomy plane of the femoral model respectively;

[0022] A laser tracker is used to obtain the coordinate values ​​of multiple laser target points on each osteotomy plane;

[0023] The least square method is used to fit each osteotomy plane according to the coordinate values, and the normal vector of each osteotomy plane and the coordinate values ​​of the geometric center points of multiple laser target points on the corresponding osteotomy plane are calculated.

[0024] Specifically, the method for obtaining the posture data of each osteotomy plane of the prosthesis model is: importing the prosthesis model into the three-dimensional image space, and directly obtaining the coordinate values ​​of the normal vector and center point of each osteotomy plane of the prosthesis model in the three-dimensional image space coordinate system.

[0025] Furthermore, before calculating the transformation matrix from the femoral model to the prosthesis model, each osteotomy plane of the femoral model is first drawn in the three-dimensional image space based on the coordinate data obtained by the laser tracker.

[0026] Specifically, the method for calculating the transformation matrix from the femoral model to the prosthesis model includes the following steps:

[0027] Draw five square planes with a side length of L, using the normal vectors of each osteotomy plane of the prosthesis model as the normal vector of the plane and the center point of each osteotomy plane of the prosthesis model as the center point of the plane. This plane set is denoted as A.

[0028] Draw five square planes with a side length of M, using the normal vectors of each osteotomy plane of the femoral model as the normal vector of the plane and the geometric center points of multiple laser targets on each osteotomy plane of the femoral model as the center points of the plane. This plane set is denoted as B.

[0029] The center point of each osteotomy plane of the prosthesis model is used as the target data, and the geometric center point of multiple laser target points on each osteotomy plane of the femoral model is used as the starting data. The vtk Landmark Transform algorithm is used to calculate the transformation matrix from the starting data to the target data, which is recorded as T1.

[0030] Use the transformation matrix T1 to move the plane set B, and take the vertices of the five square planes in the moved plane set B, which are recorded as point set C;

[0031] Taking the plane set A as the target data and the point set C as the starting data, the vtk Iterative ClosestPoint Transform algorithm is used to calculate the transformation matrix from the starting data to the target data, which is recorded as T2.

[0032] Furthermore, the plane set B is first moved using the transformation matrix T1 to roughly align the femoral model with the prosthesis model, and then the plane set B is further moved using the transformation matrix T2 to accurately align the femoral model with the prosthesis model.

[0033] Furthermore, the side length M of the square plane corresponding to the femoral model is smaller than the side length L of the square plane corresponding to the prosthesis model.

[0034] Specifically, the angle between the normal vectors of the corresponding osteotomy planes of the femoral model and the prosthesis model is calculated based on the dot product of the normal vectors. The smaller the angle, the higher the TCP accuracy.

[0035] A second aspect of the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and wherein the processor implements the steps of the above-mentioned verification method when executing the computer program.

[0036] A third aspect of the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the above-mentioned verification method are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 Schematic diagram of a flow chart of a method for verifying the TCP accuracy of a TKA surgical robot according to the present invention;

[0039] Figure 2 Schematic diagram of the distal end of the femoral model after resection in an embodiment of the present invention;

[0040] Figure 3 Schematic diagram of a plane set A corresponding to the prosthesis model in an embodiment of the present invention;

[0041] Figure 4 Schematic diagram of a plane set B corresponding to the femur model in an embodiment of the present invention;

[0042] Figure 5 Schematic diagram of alignment of the femoral model and the prosthesis model in three-dimensional image space in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0044] Reference Figure 1 In a first aspect, an embodiment of the present invention provides a method for verifying the accuracy of a TKA surgical robot TCP, comprising the following steps:

[0045] Manually remove the distal part of the femoral model (such as Figure 2 The area marked by the dotted box);

[0046] Control the robotic arm to move the saw blade to each osteotomy plane of the femoral model and obtain the position data of each osteotomy plane of the femoral model;

[0047] Obtain the posture data of each osteotomy plane of the prosthesis model;

[0048] According to the posture data of each osteotomy plane of the femoral model and the prosthesis model, the transformation matrix from the femoral model to the prosthesis model is calculated;

[0049] The femoral model was moved using the transformation matrix to align the femoral model with the prosthesis model, and the accuracy of TCP was determined based on the degree of matching between the femoral model and the prosthesis model.

[0050] By pre-removing the distal part of the femoral model, the present invention can ensure that the saw blade at the end of the robotic arm can be normally positioned to the various osteotomy planes of the femoral model, and at the same time avoid the saw blade from contacting the femoral model and causing the position of the femoral model to shift; by utilizing the matrix conversion relationship between the various osteotomy planes of the femoral model and the various osteotomy planes of the prosthesis model, the femoral model and the prosthesis model are aligned in the same three-dimensional image space, so that the angular influence of the saw blade TCP error on the various osteotomy planes can be more accurately judged, and the osteotomy plane offset results can be intuitively visualized.

[0051] During the specific implementation process, before manually cutting the femoral model, it is necessary to start the robotic system according to the instructions in the manual, and complete the standard pre-osteotomy preparation processes such as robotic arm alignment, instrument calibration, bone optical tracking array fixation, and image alignment in sequence.

[0052] Specifically, the method for obtaining the posture data of each osteotomy plane of the femoral model includes the following steps:

[0053] At least three non-collinear laser target points are set on the osteotomy surface of the saw blade (three laser target points are selected in this embodiment);

[0054] Control the robotic arm to move the saw blade to the distal osteotomy plane of the femoral model;

[0055] A laser tracker was used to obtain the coordinates of the three laser target points on the distal osteotomy plane;

[0056] The distal osteotomy plane is fitted using the least squares method according to the coordinate values, and the normal vector of the distal osteotomy plane is calculated. And the coordinate value p1 of the geometric center point of the three laser target points (obtained by calculating the weighted average of the coordinate values ​​of the three laser target points).

[0057] Control the robotic arm to move the saw blade to the anterior malleolus osteotomy plane of the femoral model;

[0058] A laser tracker was used to obtain the coordinates of the three laser target points on the anterior ankle osteotomy plane;

[0059] The anterior malleolus osteotomy plane was fitted using the least squares method according to the coordinate values, and the normal vector of the anterior malleolus osteotomy plane was calculated. And the coordinate value p2 of the geometric center point of the three laser target points (obtained by calculating the weighted average of the coordinate values ​​of the three laser target points).

[0060] Control the robotic arm to move the saw blade to the posterior malleolus osteotomy plane of the femoral model;

[0061] A laser tracker was used to obtain the coordinates of the three laser target points on the posterior malleolus osteotomy plane;

[0062] The posterior malleolus osteotomy plane was fitted using the least squares method according to the coordinate values, and the normal vector of the posterior malleolus osteotomy plane was calculated. And the coordinate value p3 of the geometric center point of the three laser target points (obtained by calculating the weighted average of the coordinate values ​​of the three laser target points).

[0063] Control the robotic arm to move the saw blade to the anterior oblique osteotomy plane of the femoral model;

[0064] A laser tracker was used to obtain the coordinates of the three laser target points on the anterior oblique osteotomy plane;

[0065] The anterior oblique osteotomy plane was fitted using the least squares method according to the coordinate values, and the normal vector of the anterior oblique osteotomy plane was calculated. And the coordinate value p4 of the geometric center point of the three laser target points (obtained by calculating the weighted average of the coordinate values ​​of the three laser target points).

[0066] Control the robotic arm to move the saw blade to the posterior oblique osteotomy plane of the femoral model;

[0067] A laser tracker was used to obtain the coordinates of the three laser target points on the posterior oblique osteotomy plane;

[0068] The posterior oblique osteotomy plane was fitted using the least squares method according to the coordinate values, and the normal vector of the posterior oblique osteotomy plane was calculated. And the coordinate value p5 of the geometric center point of the three laser target points (obtained by calculating the weighted average of the coordinate values ​​of the three laser target points).

[0069] Specifically, the method for obtaining the posture data of each osteotomy plane of the prosthesis model is as follows: the prosthesis model is imported into the three-dimensional image space (common three-dimensional mapping software can be used), and the normal vectors of the distal osteotomy plane, the anterior condyle osteotomy plane, the posterior condyle osteotomy plane, the anterior oblique osteotomy plane, and the posterior oblique osteotomy plane of the prosthesis model can be directly obtained in the three-dimensional image space coordinate system. The coordinate values ​​of the center point of each osteotomy plane (the location of the center point is selected by the user and only needs to fall roughly in the center of the osteotomy plane) are p a 、p b 、p c 、p d 、p e .

[0070] Furthermore, before calculating the transformation matrix from the femoral model to the prosthesis model, the various osteotomy planes of the femoral model are drawn in the three-dimensional image space based on the coordinate data obtained by the laser tracker, so as to facilitate the subsequent matrix transformation in the three-dimensional image space.

[0071] Specifically, the method for calculating the transformation matrix from the femoral model to the prosthesis model includes the following steps:

[0072] The normal vectors of each osteotomy plane of the prosthesis model are is the normal vector of the plane, and the center point p of each osteotomy plane of the prosthesis model is a 、p b 、p c 、p d 、p e As the center point of the plane, draw 5 square planes with side length L (side length L is specified by the user, no special requirements), and record the plane set as A, such as Figure 3 As shown;

[0073] The normal vectors of each osteotomy plane of the femoral model are is the normal vector of the plane, and the geometric center points p1, p2, p3, p4, and p5 of multiple laser targets on each osteotomy plane of the femoral model are used as the center points of the plane. Five square planes with a side length of M (the side length M is specified by the user) are drawn, and the plane set is denoted as B, as follows: Figure 4 As shown;

[0074] Taking the central points of each osteotomy plane of the prosthesis model as the target data and the geometric central points of multiple laser target points on each osteotomy plane of the femoral model as the starting data, the vtk Landmark Transform algorithm provided by the VTK library is used to calculate the transformation matrix from the starting data to the target data, denoted as T1;

[0075] The plane set B is moved using the transformation matrix T1. The plane set B will be moved to a position close to the plane set A. The vertices of 5 square planes in the moved plane set B are taken, denoted as the point set C;

[0076] Taking the plane set A as the target data and the point set C as the starting data, the vtk Iterative Closest Point Transform algorithm (i.e., the icp algorithm) provided by the VTK library is used to calculate the transformation matrix from the starting data to the target data, denoted as T2. The principle of this algorithm is to first find the points in the target data that are closest to each starting point, then perform landmark operations, and then move the starting point set, iterating back and forth until the matching degree between the starting data and the target data is higher than the preset threshold. This algorithm is prior art in this field.

[0077] Further, first use the transformation matrix T1 to move the plane set B to roughly align the femoral model and the prosthesis model, and then use the transformation matrix T2 to further move the plane set B to precisely align the femoral model and the prosthesis model (here it means aligning each osteotomy plane of the femoral model with the corresponding osteotomy plane of the prosthesis model), as Figure 5 shown;

[0078] Further, the side length M of the square plane corresponding to the femoral model is smaller than the side length L of the square plane corresponding to the prosthesis model. Because the 5 square planes (with side length M) of the plane set B will be matched to the 5 planes (with side length L) of the plane set A using the icp algorithm later, the icp algorithm needs to find the points on the planes of the plane set A that are closest to the corner points of the planes of the plane set B. If the plane of the plane set A is smaller than the plane of the plane set B (L < M), the plane size of the plane set B is larger than the plane size of the plane set A, which will cause the corner point of a certain plane in the plane set B to easily find the wrong corresponding plane when looking for the closest point on the plane of the plane set A (for example, the corner point of the distal osteotomy plane in the plane set B finds the plane point on the anterior oblique osteotomy plane in the plane set A), resulting in a matching error.

[0079] Specifically, the angles between the normal vectors of the femoral model and the prosthesis model corresponding to the osteotomy planes are calculated based on the dot product results of the normal vectors (the angle between the two vectors can be obtained by taking the inverse cosine value of the absolute value of the dot product result of the vectors). They are θ1, θ2, θ3, θ4, and θ5, respectively. The smaller the angle, the higher the TCP accuracy. When the saw blade TCP is completely correct, the above angle values ​​should all be 0.

[0080] A second aspect of an embodiment of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and is characterized in that the processor implements the steps of the above-mentioned verification method when executing the computer program.

[0081] A third aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the above-mentioned verification method are implemented.

[0082] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for verifying the TCP accuracy of a TKA surgical robot, characterized in that: The following steps are involved: The distal part of the femoral model was removed; Control the robotic arm to move the saw blade to each osteotomy plane of the femoral model and obtain the position data of each osteotomy plane of the femoral model; Obtain the posture data of each osteotomy plane of the prosthesis model; According to the posture data of each osteotomy plane of the femoral model and the prosthesis model, the transformation matrix from the femoral model to the prosthesis model is calculated; The femoral model is moved using the transformation matrix to align the femoral model with the prosthesis model, and the accuracy of the TCP is determined based on the degree of matching between the femoral model and the prosthesis model. The method for calculating the transformation matrix from the femoral model to the prosthesis model includes the following steps: Draw five square planes with a side length of L, using the normal vectors of each osteotomy plane of the prosthesis model as the normal vector of the plane and the center point of each osteotomy plane of the prosthesis model as the center point of the plane. This plane set is denoted as A. Draw five square planes with a side length of M, using the normal vectors of each osteotomy plane of the femoral model as the normal vector of the plane and the geometric center points of multiple laser targets on each osteotomy plane of the femoral model as the center points of the plane. This plane set is denoted as B. The center point of each osteotomy plane of the prosthesis model is used as the target data, and the geometric center point of multiple laser target points on each osteotomy plane of the femoral model is used as the starting data. The vtk Landmark Transform algorithm is used to calculate the transformation matrix from the starting data to the target data, which is recorded as T1. Use the transformation matrix T1 to move the plane set B, and take the vertices of the five square planes in the moved plane set B, which are recorded as point set C; With plane set A as target data and point set C as starting data, the vtk Iterative Closest Point Transform algorithm is used to calculate the transformation matrix from starting data to target data, which is recorded as T2; The angle between the normal vectors of the corresponding osteotomy planes of the femoral model and the prosthesis model was calculated based on the dot product of the normal vectors. The smaller the angle, the higher the TCP accuracy.

2. The method for verifying the TCP accuracy of a TKA surgical robot according to claim 1, wherein: The method for obtaining the posture data of each osteotomy plane of the femoral model comprises the following steps: Set at least three non-collinear laser target points on the osteotomy surface of the saw blade; Control the robotic arm to move the saw blade to the distal osteotomy plane, anterior malleolus osteotomy plane, posterior malleolus osteotomy plane, anterior oblique osteotomy plane and posterior oblique osteotomy plane of the femoral model respectively; A laser tracker is used to obtain the coordinate values ​​of multiple laser target points on each osteotomy plane; The least square method is used to fit each osteotomy plane according to the coordinate values, and the normal vector of each osteotomy plane and the coordinate values ​​of the geometric center points of multiple laser target points on the corresponding osteotomy plane are calculated.

3. The method for verifying the TCP accuracy of a TKA surgical robot according to claim 2, wherein: The method for obtaining the posture data of each osteotomy plane of the prosthesis model is: importing the prosthesis model into the three-dimensional image space, and directly obtaining the coordinate values ​​of the normal vector and center point of each osteotomy plane of the prosthesis model in the three-dimensional image space coordinate system.

4. The method for verifying the TCP accuracy of a TKA surgical robot according to claim 2, wherein: Before calculating the transformation matrix from the femoral model to the prosthesis model, the various osteotomy planes of the femoral model are first drawn in the three-dimensional image space based on the coordinate data obtained by the laser tracker.

5. The method for verifying the TCP accuracy of a TKA surgical robot according to claim 1, wherein: First, the plane set B is moved using the transformation matrix T1 to roughly align the femoral model with the prosthesis model, and then the plane set B is further moved using the transformation matrix T2 to accurately align the femoral model with the prosthesis model.

6. The method for verifying the TCP accuracy of a TKA surgical robot according to claim 1, wherein: The side length M of the square plane corresponding to the femoral model is smaller than the side length L of the square plane corresponding to the prosthesis model.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the computer program, the steps of the verification method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the verification method according to any one of claims 1 to 6 are implemented.

Images