A method for evaluating the accuracy of robot-assisted nail insertion

Abstract

The present invention discloses a method for evaluating the accuracy of robot-assisted nail placement, comprising: S1, obtaining medical images of a patient through a C-arm machine and implanting an implant under intraoperative robot navigation; S2, obtaining multi-angle fluoroscopic images and the implants in the corresponding fluoroscopic images respectively; S3, extracting the implanted vertebral body parts of each implant in each fluoroscopic image; S4, obtaining the projection points of the inner and outer points of the planned access channel on the axes of each implant in each fluoroscopic image; S5, based on the projection points obtained in S4 and polar line constraint, obtaining the actual positions of the inner and outer points of the implant on the vertebral body; S6, calculating the theoretical positions of the inner and outer points of the implant on the vertebral body according to the inner and outer points of the preoperatively planned access channel, and combining with S5 to calculate the implant placement accuracy. The present invention can achieve a rapid evaluation of the accuracy of the robot-assisted nail placement system, avoiding the lack of intuitive observation in intraoperative fluoroscopy itself, with simple operation, and greatly reducing the radiation of patients.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for evaluating the accuracy of robot-assisted nail placement. Background Art

[0002] In conventional vertebroplasty surgeries, the construction of the access channel is a crucial step. With the wide application of orthopedic-assisted nail placement robots, this step is increasingly assisted by robots. The accuracy of access channel construction is of vital importance. The conventional method is to confirm through anteroposterior and lateral fluoroscopy or intraoperative three-dimensional CBCT after the Kirschner wire is inserted into the vertebra. However, intraoperative fluoroscopy itself is not intuitive to observe, and three-dimensional CBCT not only increases the operational complexity but also increases the patient's radiation. Therefore, a rapid intraoperative evaluation of the accuracy of the access channel construction for orthopedic-assisted nail placement surgery robots is of great significance. Summary of the Invention

[0003] Object of the Invention: Aiming at the above deficiencies, the present invention proposes a method for evaluating the accuracy of robot-assisted nail placement, which can achieve a rapid evaluation of the accuracy of the robot-assisted nail placement system, avoid the non-intuitive observation of intraoperative fluoroscopy itself, is simple to operate, and greatly reduces the patient's radiation.

[0004] Technical Solution:

[0005] The present invention provides a method for evaluating the accuracy of robot-assisted nail placement, including:

[0006] S1. Obtain the medical image of the patient through a C-arm machine, and intraoperatively navigate the robot to place the implant;

[0007] S2. Operate the C-arm machine to perform multi-angle fluoroscopy on the surgical area where the implant is placed to obtain corresponding fluoroscopic images, and respectively extract the implants in the fluoroscopic images corresponding to each angle;

[0008] S3. According to the projection model of the C-arm machine, respectively transform the inner and outer points of the preoperatively planned access channel on the medical image into the fluoroscopic images of each angle, and accordingly extract the parts of the implants placed in the vertebra in each fluoroscopic image obtained in S2;

[0009] S4. Respectively fit straight lines to the parts of the implants placed in the vertebra in each fluoroscopic image obtained in S3 to obtain the axes of the implants in each fluoroscopic image, and thus obtain the projection points of the inner and outer points of the planned access channel on the axes of the implants in each fluoroscopic image according to S3;

[0010] S5. Based on the epipolar constraint, obtain the actual positions of the inner and outer points of the implant in the medical image according to the projection points obtained in S4, and further obtain the actual positions of the inner and outer points of the implant on the vertebra;

[0011] S6. Calculate the theoretical positions of the inner and outer points of the implant on the vertebral body based on the inner and outer points of the planned access channel, and combine with S5 to calculate the implant placement accuracy.

[0012] Specifically, the S3 includes:

[0013] S31. Obtain the projection points of the inner and outer points of the planned access channel in each fluoroscopic image according to the projection model of the C-arm machine and the pose of the C-arm at the corresponding angle;

[0014] S32. Combine the implant in each fluoroscopic image obtained in S2 and the projection points in each fluoroscopic image obtained in S31 to extract the vertebral body part where each implant is placed in each fluoroscopic image.

[0015] More specifically, in the S32, the extraction of the vertebral body part where each implant is placed in the fluoroscopic image is specifically as follows:

[0016] According to the projection points of the inner and outer points of the planned access channel in each fluoroscopic image, construct the corresponding reference unit vectors, traverse the pixel points in the area where the implant is located in each fluoroscopic image, and judge whether the projection of the pixel point on the reference unit vector is located between the two aforementioned projection points. If so, retain the pixel point; otherwise, eliminate it. After the traversal is completed, the vertebral body part where each implant is placed in each fluoroscopic image is extracted.

[0017] Specifically, in the S5, the obtaining of the actual positions of the inner and outer points of the implant in the medical image based on the epipolar constraint is specifically as follows:

[0018] S51. According to the projection points of the inner and outer points of the planned access channel on the axis of each implant in each fluoroscopic image obtained in S4, calculate the corresponding straight lines in the fluoroscopic images of other perspectives, and use the intersection point of the straight line and the axis of the corresponding implant in the fluoroscopic image obtained in S4 as the point corresponding to the projection point in the fluoroscopic image of this perspective;

[0019] S52. According to the corresponding points in the perspective views of each perspective obtained in S51, obtain several straight lines passing through the optical center and the corresponding points, fit the intersection points of these straight lines, and obtain the actual position of the corresponding point in the medical image, that is, the actual positions of the inner and outer points of the implant in the medical image.

[0020] Specifically, in the S5, after obtaining the actual positions of the inner and outer points of the implant in the medical image, transform them under the reference of the optical array on the C-arm or under the reference of the optical array on the patient, and then obtain the actual positions of the inner and outer points of the implant on the vertebral body;

[0021] In the S6, according to the inner and outer points of the planned access channel, transform their points under the reference of the optical array on the C-arm or under the reference of the optical array on the patient, and then obtain the theoretical positions of the inner and outer points of the implant on the vertebral body.

[0022] Specifically, in S6, the implant placement accuracy includes the implant angle error and the implant distance error. Among them, the implant angle error is the included angle between the line connecting the theoretical positions of the inner and outer points of the implant on the vertebral body and the line connecting the actual positions of the inner and outer points of the implant on the vertebral body. The implant distance error is the average value of the distances from the theoretical positions of the inner and outer points of the implant on the vertebral body to the line connecting the actual positions of the inner and outer points of the implant on the vertebral body.

[0023] Specifically, in S2, after performing threshold segmentation on the fluoroscopic image, connected component detection is performed to obtain the implant in the fluoroscopic image.

[0024] More specifically, the connected component is defined as a region where image pixels are connected. The specific feature of pixel connection is four-neighborhood or eight-neighborhood connection.

[0025] More specifically, after the connected component detection, the number of pixel points in each connected component is calculated, and the connected components with the number of pixel points less than the set threshold are deleted, that is, the implant in the fluoroscopic image is obtained.

[0026] Beneficial effects: The present invention realizes the rapid evaluation of the intraoperative accuracy of the robotic nail-making system through fluoroscopic images such as the anteroposterior and lateral views, and has the following remarkable benefits:

[0027] 1) It avoids repeated fluoroscopy confirmation through multiple C-arms, reducing the intraoperative iatrogenic radiation.

[0028] 2) The present invention can achieve sub-millimeter quantitative evaluation, improving the accuracy and safety of the surgery, and greatly improving the safety and success rate of surgical procedures with high precision requirements (such as cervical spine surgery, etc.). BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is a flowchart of the robotic-assisted nail placement accuracy evaluation method of the present invention;

[0031] Figure 2 It is an example diagram of a robotic navigation and positioning system;

[0032] Figure 3 It is an example diagram of multi-angle perspective projection;

[0033] Figure 4An example diagram of the fluoroscopic image after the robot has completed screw placement. Among them, Figure 4 (a) is the anteroposterior view after the robot has completed screw placement; Figure 4 (b) is the lateral view after the robot has completed screw placement;

[0034] Figure 5 An example diagram of implant extraction. Among them, Figure 5 (a) is an example diagram of the image obtained by performing threshold segmentation on Figure 4 ; Figure 5 (b) is an example diagram of pixel - to - pixel connectivity judgment; Figure 5 (c) is an example diagram of the result of implant connectivity detection; Figure 5 (d) is an example diagram of the result of implant extraction;

[0035] Figure 6 An example diagram of segmenting the implant placed in the vertebral body. Among them, Figure 6 (a) is an example diagram of transforming the points inside and outside the planned channel in the image to the image; Figure 6 (b) is an example diagram of the result of segmenting the implant inserted into the vertebral body;

[0036] Figure 7 An example diagram for evaluating the accuracy of implant placement. Detailed implementation manners

[0037] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the following further elaborates on this application in detail with reference to specific embodiments and the accompanying drawings.

[0038] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the ordinary meaning understood by those of ordinary skill in the field to which the present invention pertains.

[0039] The method for evaluating the accuracy of robot - assisted screw placement of the present invention is as Figure 1 shown and includes:

[0040] S1. Obtain the medical image of the patient through a C - arm machine, and the intraoperative robot navigation is used to place the implant;

[0041] As Figure 2 shown, the robot navigation positioning system mainly correlates the medical image of the patient, the intraoperative patient, and the C - arm machine. Thus, the access channel information in the medical image of the patient can be converted under the reference of the intraoperative patient, and the intraoperative patient is associated with the execution robotic arm system in real - time through the optical array. Thereby, the access channel information planned in the medical image of the patient can be converted to the execution component system to complete the screw placement operation.

[0042] In the present invention, the implanted implant is generally a Kirschner wire. Of course, it may also be other implants that need to be implanted into the human body.

[0043] S2. Operate the C-arm machine to perform multi-angle fluoroscopy on the surgical area where the implant is placed to obtain corresponding fluoroscopic images, and extract the implants in the fluoroscopic images corresponding to each angle respectively.

[0044] After the implant is placed by intraoperative robotic navigation, operate the C-arm machine to perform multi-angle fluoroscopy on the surgical area where the implant is placed. As Figure 4 shown, when performing fluoroscopy, the optical array on the C-arm and the optical array on the patient need to be seen simultaneously. That is, essentially, the fluoroscopy processes at multiple angles constitute a two-dimensional fluoroscopy model. As Figure 3 shown, the two-dimensional fluoroscopy of the C-arm machine itself is a process of sampling three-dimensional images into two-dimensional images, which can be expressed by a projection model. The calibration has been completed before the C-arm machine leaves the factory, that is, the transformation relationship of the projection model relative to the optical array on the C-arm is known.

[0045] In the present invention, extracting the implants in the fluoroscopic images corresponding to each angle specifically means: performing threshold segmentation on the fluoroscopic images and then performing connected component detection to obtain the implants in the fluoroscopic images.

[0046] In the present invention, after threshold segmentation of the fluoroscopic images, the binary images show connected regions, which also include some impurity noise regions that do not belong to the implants. As Figure 5 shown at B in (a), therefore, connected component detection is still required. After connected component detection, several connected components can be obtained in the binary image. As Figure 5 shown in (c).

[0047] In the present invention, a connected component is defined as a region where image pixels are connected. The specific feature of pixel connection is four-neighborhood or eight-neighborhood connection. In this embodiment, four-neighborhood connection is adopted. As Figure 5 shown for pixel p1 in (b), its neighborhood is regions 1, 2, 3, and 4. Then p1 and p2 are adjacent, p1 and p3 are not adjacent, and p1 and p4 are not adjacent.

[0048] The detected regions through connected component detection are φ1 - φ5. Also, since the area of the region image of the implant is relatively large, for each connected component, calculate the number of pixel points therein, and delete the connected components with the number of pixel points less than the set threshold, then the region corresponding to the implant can be obtained. As Figure 5 shown in (d), that is, the implants in the fluoroscopic images are obtained.

[0049] In the present invention, the set threshold can be taken as 50.

[0050] S3. Transform the internal and external points of the preoperatively planned access channel to the fluoroscopic images at each angle according to the projection model of the C-arm machine, and extract the vertebral body parts of each implant in each fluoroscopic image obtained in S2.

[0051] In the present invention, the implant has been extracted in S2. Taking the Kirschner wire as an example, generally, the Kirschner wire is inserted into the vertebral body about 2-3 cm. For the convenience of actual fluoroscopy, the external Kirschner wire is bent and clamped on the sterile cloth, which will cause a certain bending of the Kirschner wire. However, the part of the tip of the Kirschner wire with a length of about 2 cm will not be bent because it is inserted into the vertebral body. Therefore, the present invention needs to segment the Kirschner wire inserted into the vertebral body part.

[0052] Specifically, it includes:

[0053] S31. According to the projection model of the C-arm machine and the pose of the C-arm at the corresponding angle, the projection points of the internal and external points of the access channel planned on the medical image in each fluoroscopic image can be obtained.

[0054] In the present invention, the operator, such as a doctor, can plan the internal and external points of the access channel in the medical image, which can be carried out before and during the operation.

[0055] Taking the first fluoroscopic projection as an example, referring to Figure 6 (a), it is as follows:

[0056]

[0057] Among them, are respectively the projection points of the internal and external points of the planned access channel in the fluoroscopic image obtained by the first fluoroscopic projection; F2 represents the mapping relationship between the points in the medical image and the points on the fluoroscopic image, which can be obtained through the projection model, specifically according to the transformation relationship of the aforementioned projection model relative to the optical array on the C-arm and the pose of the optical array on the C-arm obtained in real time.

[0058] Specifically, according to the img relative transformation relationship between C carm and C transform the internal and external points of the planned access channel to the reference of the optical array on the C-arm, that is:

[0059]

[0060] Among them, are respectively the internal and external points of the access channel transformed to the reference of the optical array on the C-arm;

[0061] After that, according to the transformation relationship of the aforementioned projection model relative to the optical array on the C-arm and the pose of the optical array on the C-arm obtained in real time, the mapping relationship F2 between the points in the medical image and the points on the fluoroscopic image can be obtained.

[0062] S32. Extract the implanted vertebral body parts of each implant in each perspective image obtained by combining S2 and the projection points in each perspective image obtained by S31;

[0063] Since the implants are placed according to the planned internal and external points, mapping the internal and external points in the medical image to the positions on the perspective image will appear near the area where the implants are located. Then, extract the areas where the implants are located in each perspective image by combining the implants in each perspective image obtained by S2 and the projection points in each perspective image obtained by S31, as follows:

[0064] According to the projection points of the internal and external points of the planned access channel in each perspective image, construct the corresponding reference unit vectors. Traverse the pixel points in the area where the implants are located in each perspective image, and judge whether the projection of the pixel point on the reference unit vector is between the two aforementioned projection points. If so, retain the pixel point; otherwise, eliminate it. After the traversal is completed, the implanted vertebral body parts of each implant in each perspective image can be extracted, as shown in Figure 6 (b).

[0065] Furthermore, in this embodiment, the constructed reference unit vector is:

[0066]

[0067] Then, for each regional pixel point Construct a vector:

[0068]

[0069] It can be known that the pixel length on the perspective image of the projection points of the internal and external points of the planned access channel in each perspective image is Then, judge whether the projection of a certain pixel point in the area where the implant is located in the perspective image on the reference unit vector is between the two aforementioned projection points, as follows:

[0070] Calculate the vector in The length of the projection If d i ≥0 and d i ≤d pixel , then retain the pixel point; otherwise, eliminate it.

[0071] S4. Fit a straight line to the implanted vertebral body parts of each implant in each perspective image obtained in S3 to obtain the axis of each implant in each perspective image. Then, calculate the projection points of the internal and external points of the planned access channel on the axes of each implant in each perspective image according to the transformation in S3;

[0072] In this embodiment, the fluoroscopic images after the robot navigates and implants the implant are as follows Figure 4 (a) and 4(b). Then, combining S1 and S3, if the corresponding implant axis is extracted on the fluoroscopic image, the implant axis can be mapped under the reference of the optical array on the C-arm or under the reference of the optical array on the patient, so as to calculate the accuracy deviation between the current implant axis and the planned access channel.

[0073] In S3, the inner and outer points of the planned access channel have been respectively transformed into the fluoroscopic images at various angles. The projection points of the inner and outer points of the planned access channel in each fluoroscopic image are and Then, the vertical points corresponding to the axes of the corresponding implants can be respectively obtained by fitting through That is, the vertical points of the axes of the corresponding implants are and These are the projection points of the inner and outer points of the planned access channel on the axes of each implant in each fluoroscopic image.

[0074] S5. Based on the projection points of the inner and outer points of the planned access channel on the axes of each implant in each fluoroscopic image obtained in S4, and based on the epipolar constraint, obtain the actual positions of the inner and outer points of the implant in the medical image, and transform them to the same reference as the theoretical positions of the inner and outer points of the implant on the vertebral body, so as to obtain the actual positions of the inner and outer points of the implant on the vertebral body;

[0075] In the present invention, the epipolar constraint means that a point on the fluoroscopic image of one view corresponds to a straight line on the fluoroscopic image of other views, which includes:

[0076] S51. According to the projection points of the inner and outer points of the planned access channel on the axes of each implant in each fluoroscopic image obtained in S4, calculate the corresponding straight lines in the fluoroscopic images of other views, and use the intersection point of the straight line and the axis of the corresponding implant in the fluoroscopic image obtained in S4 as the point corresponding to the projection point in the fluoroscopic image of this view;

[0077] In this embodiment, as shown in Figure 3 , there are a total of three views. Then, the point on view 1 corresponds to the straight line on view 2. The intersection point of the straight lines and is the point corresponding to on the fluoroscopic image of view 2. Similarly, it can be obtained that

[0078] S52. Obtain several lines passing through the optical center and the corresponding points from the corresponding points in the perspective views of each perspective obtained in S51, and fit the intersection points of these lines, thereby obtaining the actual position of the corresponding point in the medical image, that is, the actual position of the inner and outer points of the implant in the medical image;

[0079] The points obtained according to S4 and the points obtained according to S51 and Then the actual position of the corresponding point of the implant on the vertebral body can be obtained, specifically as follows:

[0080] Referring to Figure 3 , is the pose of the optical array on the C-arm when the C-arm performs fluoroscopy for the i-th time. According to the transformation relationship of the aforementioned projection model with respect to the optical array on the C-arm, the following mapping relationship can be established:

[0081]

[0082] Among them, is a certain point on the fluoroscopic image obtained from the i-th fluoroscopy, which is obtained by the i-th perspective projection of the point p1 in space, is the line passing through the optical center and the point ; F1 represents the mapping relationship between the point on the fluoroscopic image and the space line ;

[0083] Then, by obtaining multiple lines through the points on the fluoroscopic images at multiple angles and fitting the intersection points of these multiple lines, the actual position of the point in space can be obtained, as follows:

[0084]

[0085] Among them, p1 carm is the actual position of the point in space, and E represents the fitting of the intersection points of all lines.

[0086] S6. Calculate the theoretical positions of the inner and outer points of the implant on the vertebral body according to the inner and outer points of the planned access channel, and combine the actual positions of the inner and outer points of the implant on the vertebral body obtained in S5 to calculate the implant placement accuracy, that is, the robot-assisted nail placement accuracy;

[0087] In the present invention, by transforming the inner and outer points of the planned access channel to the reference of the optical array on the C-arm or the reference of the optical array on the patient, the theoretical positions of the inner and outer points of the implant on the vertebral body can be obtained.

[0088] Continue to refer to Figure 2 , C img is the image coordinate system corresponding to the medical image, C pat and Ccarm are the poses of the optical arrays on the upper part of the patient and on the C-arm of the C-arm machine respectively, where C pat and C carm The relative transformation relationship between them can be obtained by real-time recognition and tracking of the optical array. C img and C carm The relative transformation relationship between them is generally determined and remains unchanged during the factory calibration of the C-arm machine, that is, the image obtained by the C-arm machine and the C-arm are a rigid body and are the internal and external points on the planned access channel. Then, the internal and external points on the planned access channel and can be transformed to the coordinate system corresponding to the optical array on the patient, that is, the internal and external points on the planned access channel are transformed to the intraoperative image, as follows:

[0089]

[0090] Among them, and are the theoretical positions of the internal and external points of the implant on the vertebral body.

[0091] In the present invention, the theoretical positions of the internal and external points of the implant on the vertebral body can be obtained in any step.

[0092] In the present invention, by transforming the actual positions of the internal and external points of the implant in the medical image obtained in S52 to the same reference as the theoretical positions of the internal and external points of the implant on the vertebral body, the actual positions of the internal and external points of the implant on the vertebral body can be obtained.

[0093] In the present invention, referring to Figure 7 , the calculated implant placement accuracy includes the implant angle error and the implant distance error. Among them, the implant angle error is the angle between the line connecting the theoretical positions of the internal and external points of the implant on the vertebral body, that is, its axis A bat and the line connecting the actual positions of the internal and external points of the implant on the vertebral body, that is, its axis B bat angle(A bat , B bat ), and the implant distance error is the average value (d1 + d2) / 2 of the distances d1 and d2 from the theoretical positions of the internal and external points of the implant on the vertebral body to the line connecting the actual positions of the internal and external points of the implant on the vertebral body, that is, its axis B bat .

[0094] In the present invention, the Kirschner wire is a commonly used tool in clinical nail-making surgery. The algorithm for calculating and positioning the axis in the Kirschner wire image in the present invention has important reference significance for the recognition and positioning of the Kirschner wire axis in medical images.

[0095] In the present invention, a C-arm machine is used to perform multi-angle fluoroscopy on the surgical area where an implant is placed during the operation to obtain corresponding fluoroscopic images. The implant can be recognized on the multi-angle fluoroscopic images, and then based on the epipolar constraint, the implant image information on each fluoroscopic image is correlated. Finally, the actual position of the in-point of the implant on the vertebral body is accurately obtained, thereby enabling a rapid evaluation of the accuracy of the robot-assisted nail insertion system, avoiding the unintuitive observation during intraoperative fluoroscopy, with simple operation and greatly reducing the radiation exposure of the patient.

[0096] The present invention can also achieve a rapid evaluation and recording of the intraoperative accuracy of the efficient robot-assisted nail making surgery based on the above accuracy evaluation method. Through a large number of long-term quantitative accuracy evaluation records, a large amount of real intraoperative accuracy data is provided for subsequent research on the problem of the drift of the robot surgery accuracy over time, which is of great significance for the optimization and upgrade of the subsequent robot navigation system.

[0097] Those of ordinary skill in the art should understand that: the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples; under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of the embodiments of the present invention as described above, which are not provided in detail for the sake of brevity.

[0098] The embodiments of the present invention are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent substitution, improvement, etc. made within the spirit and principle of the embodiments of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for evaluating the accuracy of robot-assisted nail placement, characterized in that, Including: S1. Obtain the medical image of the patient through the C-arm machine, and place the implant under intraoperative robotic navigation; S2. Operate the C-arm machine to perform multi-angle fluoroscopy on the surgical area where the implant is placed to obtain corresponding fluoroscopic images, and extract the implants in the fluoroscopic images corresponding to each angle respectively; S3. According to the projection model of the C-arm machine, transform the internal and external points of the access channel planned on the medical image into the fluoroscopic images of each angle respectively, and accordingly extract the implanted vertebral body parts of each implant in each fluoroscopic image obtained in S2; S4. Fit straight lines to the implanted vertebral body parts of each implant in each fluoroscopic image obtained in S3 respectively to obtain the axes of each implant in each fluoroscopic image, and thus obtain the projection points of the internal and external points of the planned access channel on the axes of each implant in each fluoroscopic image according to S3; S5. Based on the projection points obtained in S4 and the epipolar constraint, obtain the actual positions of the internal and external points of the implant in the medical image, and further obtain the actual positions of the internal and external points of the implant on the vertebral body, specifically: S51. According to the projection points of the internal and external points of the planned access channel on the axes of each implant in each fluoroscopic image obtained in S4, calculate the corresponding straight lines in the fluoroscopic images of other perspectives, and use the intersection point of the straight line and the axis of the corresponding implant in the fluoroscopic image obtained in S4 as the point corresponding to the projection point in the fluoroscopic image of this perspective; S52. According to the corresponding points in the perspective views of each perspective obtained in S51, obtain several straight lines passing through the optical center and the corresponding points, fit the intersection points of these straight lines, and obtain the actual position of the corresponding point in the medical image, that is, the actual positions of the internal and external points of the implant in the medical image; S6. Calculate the theoretical positions of the internal and external points of the implant on the vertebral body according to the internal and external points of the planned access channel, and combine with S5 to calculate the implant placement accuracy.

2. The robot-assisted nail placement accuracy evaluation method according to claim 1, wherein The said S3 includes: S31. According to the projection model of the C-arm machine and the pose of the C-arm at the corresponding angle, obtain the projection points of the internal and external points of the planned access channel in each fluoroscopic image; S32. Combine the implants in each fluoroscopic image obtained in S2 and the projection points in each fluoroscopic image obtained in S31 to extract the implanted vertebral body parts of each implant in each fluoroscopic image.

3. The robot-assisted nail placement accuracy evaluation method according to claim 2, wherein The said S32 is specifically: According to the projection points of the internal and external points of the planned access channel in each fluoroscopic image, construct corresponding reference unit vectors, traverse the pixel points in the area where the implant is located in each fluoroscopic image, and judge whether the projection of the pixel point on the reference unit vector is located between the projection points of the internal and external points of the planned access channel obtained above in each fluoroscopic image. If so, retain the pixel point, otherwise eliminate it. After the traversal is completed, the implanted vertebral body parts of each implant in each fluoroscopic image are extracted.

4. The robot-assisted nail placement accuracy evaluation method according to claim 1, wherein In the said S5, after obtaining the actual positions of the internal and external points of the implant in the medical image, transform them under the reference of the optical array on the C-arm or under the reference of the optical array on the patient, and then obtain the actual positions of the internal and external points of the implant on the vertebral body; In the said S6, according to the internal and external points of the planned access channel, transform the points under the reference of the optical array on the C-arm or under the reference of the optical array on the patient, and then obtain the theoretical positions of the internal and external points of the implant on the vertebral body.

5. The robot-assisted nail placement accuracy evaluation method according to claim 1, wherein In S6, the implant placement accuracy includes the implant angle error and the implant distance error. Among them, the implant angle error is the included angle between the line connecting the theoretical positions of the inner and outer points of the implant on the vertebral body and the line connecting the actual positions of the inner and outer points of the implant on the vertebral body. The implant distance error is the average value of the distances from the theoretical positions of the inner and outer points of the implant on the vertebral body to the line connecting the actual positions of the inner and outer points of the implant on the vertebral body.

6. The robot-assisted nail placement accuracy evaluation method according to claim 1, wherein In S2, after threshold segmentation of the fluoroscopic image, connected component detection is performed to obtain the implant in the fluoroscopic image.

7. The robot-assisted nail placement accuracy evaluation method according to claim 6, wherein, The connected component is defined as a region where image pixels are connected. The specific feature of pixel connection is four-neighborhood or eight-neighborhood connection.

8. The robot-assisted nail placement accuracy evaluation method according to claim 6, wherein After the connected component detection, the number of pixel points in each connected component is calculated, and the connected components with the number of pixel points less than the set threshold are deleted, thus obtaining the implant in the fluoroscopic image.

Images