A spatial model precision verification device, verification method and registration system precision verification method

Abstract

This invention discloses a spatial model accuracy verification device, a verification method, and a registration system accuracy verification method. The spatial model accuracy verification method includes the following steps: placing the accuracy verification device in the imaging space of an imaging device; acquiring a frontal and lateral 2D perspective image of the accuracy verification device and the positions of its marker points; reconstructing the marker points based on the spatial model and the frontal and lateral 2D perspective image to obtain reconstructed marker points, and acquiring the positions of the reconstructed marker points; calculating the accuracy of the spatial model based on the marker point positions and the reconstructed marker point positions. This invention introduces an accuracy verification device, utilizes the spatial model and the accuracy verification device to reconstruct the theoretical points of the verification device's marker points, and calculates the accuracy of the imaging spatial model by comparing the deviations between the marker points and the theoretical points, providing a theoretical data analysis basis for the correction and improvement of the spatial model.

Description

Technical Field

[0001] This invention relates to the field of accuracy verification technology, and in particular to a spatial model accuracy verification device, verification method, and registration system accuracy verification method. Background Technology

[0002] Currently, surgical robots are widely used in orthopedic surgery, and registration methods based on 2D medical images are increasingly being applied. For example, there are methods for registering preoperative 3D images with intraoperative 2D images. These methods calculate the pose transformation relationship between 3D and 2D fluoroscopic images. In image-guided orthopedic surgery, planning information from preoperative 3D images can be integrated into intraoperative 2D fluoroscopic images. Existing algorithms typically use preoperative 3D images to generate 2D DRR (Digital Reconstructed Radiographs) images, and then register them based on the similarity between the 2D DRR images and intraoperative 2D fluoroscopic images, as shown in CN112233155B. Generating 2D DRR from 3D images requires constructing an imaging space model of the intraoperative imaging device. This space model includes precise calibration of the imaging device's projection parameters, enabling accurate registration in orthopedic robotic surgery.

[0003] However, in actual use, human operation and normal mechanical aging can affect the imaging system of the imaging device, such as changes in the optical center position or deformation of the flat panel detector, which can reduce the accuracy of the spatial model and thus affect the registration accuracy. Therefore, a method is needed to verify the accuracy of the spatial model. Summary of the Invention

[0004] Purpose of the invention: To address the above-mentioned shortcomings, this invention proposes a spatial model accuracy verification device, a verification method, and a registration system accuracy verification method for imaging devices, used to verify the accuracy of the imaging spatial model and further characterize the registration accuracy of the registration system.

[0005] Technical solution: A spatial model accuracy verification device, comprising: a model and a first tracer for tracing the model, the model having at least one marker point.

[0006] A method for verifying the accuracy of a spatial model, comprising the following steps:

[0007] Place the aforementioned accuracy verification device in the imaging space of the imaging device, and acquire the frontal and lateral 2D perspective images of the accuracy verification device and the positions of the marker points on it.

[0008] The reconstructed marker points and their locations are obtained by reconstructing the marker points based on the spatial model and the frontal and lateral 2D perspective images;

[0009] The accuracy of the spatial model is calculated based on the location of the marker points and the location of the reconstructed marker points.

[0010] The process of reconstructing the landmark points based on the spatial model and the frontal and lateral 2D perspective images includes:

[0011] Acquire the image plane pose during imaging of frontal and lateral 2D perspective images;

[0012] Based on the image plane pose presentation spatial model and the frontal and lateral 2D perspective image, the optical center point in the spatial model and the image point of the marker point in the frontal and lateral 2D perspective image are obtained;

[0013] The marker point is reconstructed based on the optical center and the image point.

[0014] Reconstructing the marker point based on the optical center and the image point includes:

[0015] Construct frontal and lateral projection lines based on the optical center and the image point, respectively;

[0016] The reconstructed marker points are determined based on the intersection of the aforementioned frontal and lateral projection lines;

[0017] The position of the marker point is reconstructed based on the position of the optical center point and the position of the image point of the marker point.

[0018] Image plane pose during imaging includes:

[0019] A second tracer for tracing the image plane is fixed on the imaging device;

[0020] The pose of the second tracer during imaging is obtained through an optical tracking system;

[0021] The image plane pose is obtained based on the pose of the second tracer and the positional relationship between the image plane and the second tracer.

[0022] The optical center position in the spatial model is obtained by: obtaining the optical center position based on the positional relationship between the optical center and the image plane in the spatial model and the pose of the image plane.

[0023] Obtaining the image point position of the marker includes:

[0024] Based on the optical center and the marker point, a search line is constructed, and the intersection point of the search line and the image plane is calculated by combining the image plane pose.

[0025] Acquire the image within a defined pixel range around the intersection points on the image plane;

[0026] Identify the landmark image points in the image and obtain the image point positions.

[0027] The acquisition of the marker point location includes:

[0028] The pose of the first tracer is obtained through an optical tracking system, and the position of the marker is obtained based on the pose of the first tracer and the positional relationship between the marker and the first tracer.

[0029] The accuracy of the calculated spatial model includes:

[0030] Calculate the distance between each marker point and its corresponding reconstructed marker point, and calculate the accuracy of the spatial model based on the distance.

[0031] The at least one marker point includes a first marker point and a second marker point; the calculation of the accuracy of the spatial model includes:

[0032] A first channel is constructed using the first and second marker points, and a second channel is constructed using the reconstructed marker points corresponding to the first and second marker points.

[0033] The pose deviations of the first channel and the second channel are compared, and the accuracy of the spatial model is calculated based on the pose deviations.

[0034] The at least one marker point further includes a third marker point located between the first marker point and the second marker point and collinear with the first marker point and the second marker point, and the calculation of the accuracy of the spatial model includes:

[0035] Calculate the distance from the reconstructed marker of the third marker point to the second channel, and calculate the accuracy of the spatial model based on the distance.

[0036] A method for verifying the registration accuracy of a registration system includes:

[0037] Establish the relationship between the accuracy of the spatial model and the accuracy of the registration system;

[0038] The registration accuracy of the registration system is calculated based on the aforementioned spatial model accuracy verification method and the relationship described.

[0039] Beneficial effects: By introducing a precision verification device, this invention reconstructs the theoretical points of the verification device's marker points using the spatial model and the 2D perspective image of the precision verification device. By comparing the deviation between the marker points and the theoretical points, the accuracy of the imaging spatial model of the imaging device can be quickly and accurately analyzed. The operation is simple and the computational load is low, providing a theoretical data analysis basis for the correction and improvement of the spatial model. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the accuracy verification device;

[0041] Figure 2 This is a schematic diagram of the spatial model;

[0042] Figure 3 Here is a flowchart of the spatial model accuracy verification method;

[0043] Figure 4 A schematic diagram illustrating the correct orientation of a spatial model for accuracy verification.

[0044] Figure 5 This is a schematic diagram for verifying the accuracy of a spatial model.

[0045] 11. Base, 12. Bone model, 13. First tracer, 14. Landmark point, P1. Simulation in point, P2. Simulation out point;

[0046] 2. Spatial model, 21. Image plane, 22. Optical center, 23. Center of image plane, 211. Orthogonal 2D perspective image, 212. Lateral 2D perspective image, 221. Orthogonal optical center, 222. Lateral optical center;

[0047] 31. Second tracer;

[0048] L1. First exploration line, L2. Second exploration line, L1'. Third exploration line, L2'. Fourth exploration line; L3. First projected line, L3'. Second projected line, L4. Third projected line, L4'. Fourth projected line;

[0049] P3. First intersection point search, P4. Second intersection point search, P3'. Third intersection point search, P4'. Fourth intersection point search, P5. Entering point positive image point, P6. Exiting point positive image point, P5'. Entering point lateral image point, P6'. Exiting point lateral image point, P1'. Reconstructing the entering point, P2'. Reconstructing the exit point.

[0050] In this invention, the point represents the position of the point, which can be represented by coordinates. Detailed Implementation

[0051] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0052] This invention provides an accuracy verification device, such as Figure 1As shown, the device includes a base 11, a bone model 12 mounted on the base 11, a first tracer 13, and marker points 14 disposed on the bone model 12. The bone model 12 is modeled after a real vertebra, but is not limited to a vertebra. The model is not limited to a bone model and can also be other tissue models. The first tracer 13 is used to track the bone model 12 and has at least three coplanar but non-collinear reflective spheres mounted on it. In this embodiment, there are four spheres. The pose of the first tracer 13 can be identified by an optical tracking system, and a coordinate system for the verification device can be established. There is at least one marker point 14. In this embodiment, there are two marker points. The line connecting the two marker points can simulate a bone screw channel. One marker point serves as the simulated entry point P1 of the bone screw channel, and the other marker point serves as the simulated exit point P2 of the bone screw channel. In this invention, both the simulated entry point P1 and the simulated exit point P2 are small steel balls with a diameter of 2 mm.

[0053] The spatial model 2 of the present invention is as follows Figure 2 The image includes: optical center 22, image plane 21, and image plane center 23, whose positions are fixed; the image plane is represented by a direction vector. and Indicates the direction vector and Corresponding to the x-axis and y-axis in the image coordinate system; the spatial model of the X-ray imaging device has been constructed and stored in the system in the early stage, and the construction method is the existing technology, see literature CN112168357A for details.

[0054] Reference Figure 3 The present invention provides a method for verifying the accuracy of a spatial model system, comprising:

[0055] (1) Place the aforementioned accuracy verification device in the imaging space of the imaging device and obtain the position of the marker point on it:

[0056] The markers include the simulated in point P1 and the simulated out point P2;

[0057] The accuracy verification device is placed in the imaging space of the imaging device. The pose of the first tracer 13 on the accuracy verification device is obtained through the optical tracking system, and the coordinate system of the verification device is established. The relative positional relationship between the simulated entry point P1 and the simulated exit point P2 and the first tracer 13 is a known quantity, which is obtained through coordinate measuring machine measurement. The coordinates of the simulated entry point P1 and the simulated exit point P2 in the coordinate system of the verification device are obtained according to the pose of the first tracer 13 and the known relative positional relationship. That is, the positions of the simulated entry point P1 and the simulated exit point P2 are also the positions of the marker points on the accuracy verification device.

[0058] In this invention, the imaging device is an X-ray imaging device, more specifically a C-arm machine, but not limited to a C-arm machine, it can also be a G-arm machine, a CT machine or an O-arm machine, etc.

[0059] (2) Image the precision verification device in the imaging space to obtain a frontal and lateral 2D perspective image of the precision verification device:

[0060] The C-arm camera captures images of the precision verification device located in its imaging space, obtaining frontal and lateral 2D perspective images of the precision verification device. The frontal and lateral 2D perspective images include frontal 2D perspective image 211 and lateral 2D perspective image 212.

[0061] (3) Reconstruct the landmark points using the spatial model and the aforementioned 2D perspective images, and obtain the reconstructed landmark point positions. The reconstructed landmark points include the reconstruction entry point P1' and the reconstruction exit point P2'. The specific steps include:

[0062] (31) Obtain the image plane pose during imaging:

[0063] like Figure 4 As shown, a second tracer 31 is installed on the flat panel detector of the C-arm camera before shooting. The second tracer 31 is used to trace the image plane of the C-arm camera. When the C-arm camera acquires the frontal and lateral 2D perspective images of the verification device, it acquires the pose of the second tracer 31 through the optical tracking system and establishes the C-arm coordinate system. Since there is a rigid positional relationship T1 between the C-arm tracer and the image plane of the C-arm camera, the image plane pose when acquiring the frontal and lateral 2D perspective images of the verification device is obtained based on the rigid positional relationship T1 and the pose of the second tracer 31, and an image coordinate system is established. The image plane pose includes the frontal image plane pose and the lateral image plane pose, and the image coordinate system includes the frontal image coordinate system and the lateral image coordinate system.

[0064] (32) Based on the spatial model of image plane pose and the frontal and lateral 2D perspective images:

[0065] Because of the image plane direction vector and Corresponding to the x and y axes in the image coordinate system, the frontal and lateral 2D perspective images can be presented in space based on the image plane pose. Specifically, a frontal 2D perspective image 211 is presented based on the frontal image plane pose, and a lateral 2D perspective image 212 is presented based on the lateral image plane pose, especially presenting the frontal and lateral image points of the marker points. Furthermore, the spatial model includes the image plane 21 and the positional relationship between the optical center 22 and the image plane 21. Therefore, the spatial model can be presented based on the image plane pose, especially presenting the frontal optical center 221 and the lateral optical center 222, such as... Figures 4-5 As shown.

[0066] (33) Obtain the optical center point:

[0067] Based on the image plane pose and the positional relationship between the image plane and the optical center in the spatial model, the coordinates of the optical center in the corresponding image coordinate system are obtained, and then transformed to the verification device coordinate system to obtain the coordinates of the optical center in the verification device coordinate system, i.e., the optical center position. The optical center position includes the positive optical center position and the lateral optical center position.

[0068] Specifically, transforming the coordinates from the image coordinate system to the verification device coordinate system involves:

[0069] The positional relationship T2 between the second tracer 31 and the first tracer 13 can be obtained through the optical tracking system. Based on this positional relationship T2 and the rigid positional relationship T1 between the C-arm tracer and the image plane of the C-arm machine, the pose transformation relationship between the image plane of the C-arm machine and the first tracer 13 can be obtained, that is, the transformation relationship T3 between the image coordinate system and the verification device coordinate system. The coordinates in the image coordinate system are transformed to the verification device coordinate system using this transformation relationship T3.

[0070] In this embodiment, the verification device coordinate system C0 corresponding to the pose of the first tracer 13 is used as the reference coordinate system. However, the present invention is not limited to this. The present invention can also use the C-arm coordinate system corresponding to the pose of the second tracer 31 as the reference coordinate system, or the world coordinate system corresponding to the optical tracking system as the reference coordinate system, or the image coordinate system corresponding to the pose of the frontal or side view image as the reference coordinate system. It is only necessary to transform according to the pose transformation relationship between the C-arm coordinate system, the verification device coordinate system, the world coordinate system, and the image coordinate system.

[0071] (34) Obtain the image point position of the marker:

[0072] The first exploration line L1 is constructed using the positive optical center 221 and the simulated entry point P1;

[0073] The coordinates of the first exploration line L1 and the first exploration intersection point P3 of the positive optical center 221, the simulated entry point P1, and the positive image plane pose are calculated in the corresponding image coordinate system.

[0074] Taking the first exploration intersection point P3 as the center, extract the image within a 30*30 pixel range around it. After a series of digital image processing, the extracted image is processed to obtain the coordinates of the positive image point P5 in the positive image coordinate system. This coordinate is then transformed according to the transformation relationship T3 between the image coordinate system and the verification device coordinate system to obtain the coordinates of the positive image point P5 in the verification device coordinate system, i.e., the position of the positive image point P5.

[0075] Using the orthogonal optical center 221 and the simulated exit point P2, the lateral optical center 222 and the simulated entry point P1, and the lateral optical center 222 and the simulated entry point P2 respectively, construct the second exploration line L2, the third exploration line L1', and the fourth exploration line L2', and obtain the corresponding second, third, and fourth exploration intersection points P4, P3', and P4' respectively. Repeat the above steps to obtain the position of the exit point orthogonal image point P6, the entry point lateral image point P5', and the exit point lateral image point P6'.

[0076] The image processing involves identifying and segmenting a steel ball within a defined pixel range. The steel ball is imaged as a circle; that is, the circle is identified and extracted from the image within the defined pixel range, and the coordinates of its center in the image coordinate system are calculated. The defined pixel range can be 30*30 pixels. This range selection allows for quick searching of corresponding image points while avoiding the search for redundant image points.

[0077] (35) Based on the image point, the corresponding optical center, and the marker point, the reconstructed marker point is obtained, and the position of the reconstructed marker point is acquired:

[0078] The first projection line L3 is constructed using the positive optical center 221 and the positive image point P5 at the entrance point;

[0079] Construct the second projection line L3′ using the lateral optical center 222 and the lateral image point P5′ at the entrance point;

[0080] The first projection line L3 and the second projection line L3′ intersect to obtain the reconstruction entry point P1′. The coordinates of the intersection point of the first projection line L3 and the second projection line L3′ are calculated using the optical center coordinates and image point coordinates. These coordinates are the coordinates of the reconstruction entry point P1′, which is also the position of the reconstruction entry point P1′.

[0081] By repeating the above steps using the optical center and the image point, the reconstructed image point P2′ and its location can be obtained.

[0082] like Figures 3-5 As shown, the frontal 2D perspective image 211 includes a frontal entrance image point P5 and a frontal exit image point P6, and the lateral 2D perspective image 212 includes an entrance lateral image point P5′ and an exit lateral image point P6′. The line connecting the frontal optical center 221 and the entrance frontal image point P5 forms the first projection line L3, the line connecting the lateral optical center 222 and the lateral image point P5′ forms the second projection line L3′, the line connecting the frontal optical center 221 and the exit frontal image point P6 forms the third projection line L4, and the line connecting the lateral optical center 222 and the exit frontal image point P6 forms the fourth projection line L4′. The intersection of the third projection line L4 and the fourth projection line L4′ is the reconstructed exit point P2′.

[0083] (4) Calculate the accuracy of the spatial model based on the location of the marker points obtained in step (1) (specifically, the location of the simulated entry point P1 and the simulated exit point P2) and the location of the reconstructed marker points obtained in step (3) (specifically, the location of the reconstructed entry point P1′ and the reconstructed exit point P2′).

[0084] The accuracy of the computational space model includes:

[0085] Compare the positional deviations of the simulated in point P1, simulated out point P2, and reconstructed in point P1′, reconstructed out point P2′:

[0086] Calculate the distance S1 between the simulated in-point P1 and the reconstructed in-point P1′, and use this distance as the in-point deviation;

[0087] Calculate the distance S2 between the simulated output point P2 and the reconstructed output point P2′, and use this distance S2 as the error of the output point;

[0088] The accuracy k of the spatial model is calculated using the formula: k = max(S1, S2).

[0089] The accuracy of the computational space model also includes:

[0090] A first channel is constructed based on the simulated inlet point P1 and the simulated outlet point P2 to simulate a bone nail channel or other surgical channel, and the pose of the first channel is obtained using the coordinates of these two points; a second channel is constructed based on the reconstructed inlet point P1′ and the reconstructed outlet point P2′, and the pose of the second channel is obtained using the coordinates of these two points; the pose deviations of the first channel and the second channel are compared, and the accuracy of the spatial model is calculated based on the pose deviations. Specifically, the standard deviations of each component of the poses of the first channel and the second channel can be calculated, and the standard deviations are used as the accuracy of the spatial model.

[0091] Two marker points are used to verify the accuracy of the spatial model. Based on the principle of two points defining a line, surgical channels, such as bone nail channels, can be simulated, which is more in line with the needs of actual applications. It can not only verify the accuracy of the spatial model, but also reflect the ability of the spatial model to reconstruct the simulated channel to a certain extent, including the ability of the spatial model to reconstruct the entry point, exit point, and pose of the simulated channel.

[0092] The spatial model accuracy verification method described in this invention can use at least one marker point to verify the accuracy of the spatial model. That is, the distance between any point of simulation entry point P1 or simulation exit point P2 and the corresponding reconstruction point can be used as the accuracy of the spatial model. Using a single point to verify the accuracy of the spatial model results in faster calculation and verification speed.

[0093] Furthermore, the present invention can also set three non-collinear marker points on the bone model 12. Correspondingly, the accuracy verification of the spatial model in step (3) can be performed based on the position of the three channel simulation points and the pose of the three simulation channels. The three non-collinear points can construct three straight lines in space, thus simulating three channels.

[0094] Furthermore, the present invention can also set three collinear marker points on the bone model 12, with the third marker point located between the simulation entry point P1 and the simulation exit point P2; the method of obtaining the third marker point and its position, the corresponding reconstruction point and its position is the same as in steps (1) and (2); correspondingly, the calculation of the accuracy of the spatial model in step (3) also includes: calculating the distance S3 from the reconstruction point of the third marker point to the second channel; calculating the accuracy of the spatial model based on the distance S3. This accuracy verification method can verify the accuracy of the spatial model from multiple dimensions, and can also reflect the ability of the spatial model to reconstruct the simulation channel from multiple dimensions.

[0095] This invention also provides a method for verifying the registration accuracy of a registration system. Since preoperative 3D images and intraoperative 2D images require the generation of 2D DRR (Digital Reconstructed Radiography) images from preoperative 3D CT images, registration is performed based on the similarity between the DRR images and the intraoperative 2D fluoroscopic images. The accuracy of the C-arm spatial model directly affects the accuracy of the generated 2D DRR images, thus influencing the registration accuracy of the registration system. A relationship is established between the spatial model accuracy k and the registration system accuracy K, where K = f(k). The registration system accuracy error is evaluated based on the accuracy error of the spatial model; the worse the accuracy of the spatial model, the greater the registration system error.

[0096] More specifically, the relationship between the spatial model accuracy k and the registration system accuracy K is K=ak+b, where a and b are relationship coefficients, which can be obtained through fitting.

[0097] The verification principle of this invention is as follows: Based on the imaging principle of an imaging device, in a spatial model, both the orthogonal and lateral projection rays of an object point pass through the object point and form an image point on the image plane. The spatial model contains optical center information, and the 2D perspective image contains image point information. Therefore, the projection line can be reconstructed using the optical center in the spatial model and the image point in the orthogonal and lateral 2D perspective image. The object point can be reconstructed using the intersection of the orthogonal and lateral projection rays. The intersection point is the reconstructed object point. The smaller the positional deviation between the reconstructed object point and the actual object point, the higher the accuracy of the spatial model. The larger the positional deviation between the reconstructed object point and the actual object point, the worse the accuracy of the spatial model.

[0098] This invention introduces an accuracy verification device, which uses the spatial model and the 2D perspective image of the accuracy verification device to reconstruct the theoretical points of the marker points of the verification device. By comparing the deviation between the marker points and the theoretical points, the accuracy of the imaging spatial model of the imaging device can be analyzed quickly and accurately. The operation is simple and the computation is low. It provides a theoretical data analysis basis for the correction and improvement of the spatial model, and can reflect the ability of the spatial model to reconstruct the simulation channel. It can also further characterize the registration accuracy of the registration system.

[0099] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations (such as quantity, shape, position, etc.) can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for verifying the accuracy of a spatial model, the method comprising: The method comprises the steps of: placing a precision verification device in the imaging space of an imaging device, the precision verification device comprising a model and a first tracer for tracing the model, the model having at least one marker point thereon; acquiring a front-lateral 2D perspective image of the precision verification device and the marker point thereon; acquiring an image plane pose during imaging, based on the image plane pose, a spatial model and a front-lateral 2D perspective image, and according to the position relationship between the image plane and the optical center in the spatial model, acquiring a front-lateral optical center point; constructing a search straight line according to the optical center point and the marker point, and calculating the intersection point of the search straight line and the image plane based on the image plane pose, acquiring an image within a set pixel range around the intersection point on the image plane, identifying the image point of the marker point in the image and acquiring the image point; constructing a front-lateral projection straight line according to the optical center point and the image point, determining a reconstructed marker point based on the intersection of the front-lateral projection straight line, and calculating the point of the reconstructed marker point based on the optical center point and the image point of the marker point; calculating the precision of the spatial model based on the point of the marker point and the point of the reconstructed marker point.

2. The spatial model accuracy verification method of claim 1, wherein, The step of acquiring the image plane pose during imaging comprises: fixing a second tracer for tracing the image plane on the imaging device; acquiring the pose of the second tracer during imaging through an optical tracking system; acquiring the image plane pose based on the pose of the second tracer and the position relationship between the image plane and the second tracer.

3. The spatial model accuracy verification method of claim 1, wherein, The step of acquiring the optical center point in the spatial model comprises:

4. The spatial model accuracy verification method of claim 1, wherein, acquiring the pose of the first tracer through an optical tracking system, and acquiring the point of the marker based on the pose of the first tracer and the position relationship between the marker and the first tracer. The step of calculating the precision of the spatial model comprises:

5. The spatial model accuracy verification method of claim 1, wherein, calculating the distance between each marker point and its corresponding reconstructed marker point, and calculating the precision of the spatial model based on the distance. The at least one marker point comprises a first marker point and a second marker point, and the step of calculating the precision of the spatial model comprises:

6. The spatial model accuracy verification method of claim 1, wherein, constructing a first channel with the first marker point and the second marker point, and constructing a second channel with the reconstructed marker points corresponding to the first marker point and the second marker point; comparing the pose deviation between the first channel and the second channel, and calculating the precision of the spatial model based on the pose deviation. The at least one marker point further comprises a third marker point located between the first marker point and the second marker point and collinear with the first marker point and the second marker point, and the step of calculating the precision of the spatial model comprises:

7. The spatial model accuracy verification method of claim 6, wherein, calculating the distance from the reconstructed marker point of the third marker point to the second channel, and calculating the precision of the spatial model based on the distance. The method comprises:

8. A method of verifying registration accuracy of a registration system, characterized by, establishing a relationship between the precision of the spatial model and the precision of the registration system; calculating the registration precision of the registration system based on the spatial model precision verification method according to any one of claims 1-7 and the relationship. ​

Images