Tooth implantation robot positioning method applying oral cavity scanning

Abstract

According to the tooth implantation robot positioning method applying oral cavity scanning, a mechanical arm and an end effector installed on the mechanical arm are included, and a dental drill point used for preparing an implantation cavity and an equipment probe used for obtaining the three-dimensional characteristics of the oral cavity are installed on the end effector. The equipment probe is electrically connected with a computer which is used for acquiring image data acquired by the equipment probe and calculating a spatial pose; comprising the following steps: acquiring oral cavity information of a patient, pulling an end effector to a position where implantation is needed, executing oral cavity three-dimensional feature acquisition, and performing coordinate comparison to complete confirmation of a pose relation between an equipment probe and the oral cavity of the patient; compared with the prior art, the problems that an external marker is high in cost and prone to external interference, and positioning is inaccurate due to uncontrollable deformation along with time are solved through biological oral cavity three-dimensional feature positioning.

Description

Technical Field

[0001] The present invention relates to the technical field of implant robots, and in particular to a tooth implant robot positioning method using oral cavity scanning. Background Art

[0002] In recent years, modern engineering technologies such as robots and artificial intelligence have gradually been widely used in the medical field. Implant doctors can complete operations with the assistance of implant robots, which can greatly improve surgical accuracy, reduce surgical errors, maximize the goal of minimally invasive surgery, reduce surgical risks, shorten operation time, and achieve standardized operations of implant surgery.

[0003] When a robot assists an implant surgeon in performing oral implant surgery, one of the important steps is to determine the relative position of the drill under the robot flange. Currently, the drill positioning of the oral implant robot mainly uses an infrared binocular camera to emit infrared light. After being reflected by the reflective balls on the sign board, the infrared binocular camera can accurately identify a collection of multiple reflective balls and calculate the position of the oral implant robot drill through a relative coordinate transformation algorithm, thereby determining the position relationship between the oral implant robot drill and the patient's affected tooth. However, the above method for determining the position relationship between the oral implant robot drill and the patient's affected tooth has the following problems:

[0004] 1. If the infrared light emitted by the infrared emitter is blocked, the infrared calibration system will not be able to detect the location of the marker. In the complex environment of oral implantology, it is difficult for doctors and patients to ensure that the infrared light path will not be blocked, so this positioning method is not stable.

[0005] 2. The reflective balls in the marking system are consumables and are expensive, which greatly increases the cost of treatment.

[0006] 3. The above-mentioned positioning system uses a fixed marker to assist in positioning during the positioning process. This marker is divided into two parts: a fixed module and a marker module. Before the operation, the patient wears the fixed module of the marker on his teeth to take a CT scan. During the operation, the patient wears the same fixed module to connect the marker module for positioning. However, such fixed modules are standard parts and do not match the shape of the patient's teeth. They must be filled with fillings such as gutta-percha / silicone rubber / EVA to fix them to the patient's teeth. After these fillings are made, they will produce uncontrollable deformation over time, resulting in inaccurate position of the connected marker, which will cause errors in the optical positioning and navigation system. The errors in the robot positioning and navigation system can easily lead to inaccurate positioning of the end effector (dental drill) due to inaccurate coordinates of the target position, thereby increasing the risk of implant surgery.

[0007] Chinese Publication No. CN113693723A discloses a cross-modal navigation and positioning system and method for oral and pharyngeal surgery, designed for real-time navigation and positioning of a surgical robot during surgery. The cross-modal navigation and positioning system includes a self-made mouth opener for supporting the oral cavity; self-identifying visual markers for assisting positioning; a visual positioning device for detecting and locating the self-identifying visual markers on the self-made mouth opener and the surgical robot; and a control host for preoperative registration and fusion of multi-source scan data, visual registration of a three-dimensional model of the self-identifying visual markers, and calibration between various coordinate systems. The visual positioning device then performs real-time detection and positioning of the self-identifying visual markers during surgery.

[0008] The positioning system disclosed above navigates and positions the surgical robot through self-identifying visual markers and visual positioning devices, and the self-identifying visual markers are installed on a homemade mouth opener. However, since the patient needs to keep his mouth open for a long time during the operation, the patient's swallowing, discomfort caused by keeping the mouth open for a long time, and discomfort caused by installing a homemade mouth opener will cause the patient's mouth to move, thereby causing the relative position of the homemade mouth opener in the patient's mouth to change, affecting the navigation and positioning effects of the surgical robot. Summary of the Invention

[0009] The present invention aims to overcome the above-mentioned defects in the prior art and provide a dental implant robot positioning method using oral scanning, which has accurate positioning, controllable costs, and does not affect the patient's comfort.

[0010] To achieve the above-mentioned object of the invention, the present invention adopts the following technical solution: a dental implant robot positioning method using oral scanning, comprising a robotic arm, an end effector mounted on the robotic arm, and a CT machine, wherein the end effector is mounted with a dental drill for preparing an implant cavity and a device probe for obtaining three-dimensional features of the oral cavity, and the device probe is electrically connected to a computer for collecting image data and calculating spatial posture; the method comprises the following steps:

[0011] Step S1: Obtain the patient's oral information, obtain the oral 3D feature A of the desired implant area of ​​the patient through the device probe, and obtain the oral 3D data B through the CT machine, and align the oral 3D feature A and the oral 3D data B to establish a unified coordinate system N xyz , taking the center point of the implant area required by the patient as the coordinate system N xyz The base point, the lingual side is the positive direction of the X axis, and the maxillofacial normal is the positive direction of the Z axis of the coordinate system;

[0012] Step S2: Pull the end effector to the desired implant area, align the device probe with the patient's desired implant area, and set the local coordinate system of the end effector to S xyz , the global coordinate system of the robot is R xyz , and use the homogeneous transformation matrix Indicates that the end effector is in the global coordinate system R of the robot arm xyz The posture relationship in

[0013] Step S3: Execute oral 3D feature acquisition, obtain the patient's real-time oral 3D feature C, and establish the coordinate system M xyz , taking the center point of the implant area required by the patient as the coordinate system M xyz The base point, the side of the tongue is the positive direction of the X axis, the normal line of the maxillofacial face is the positive direction of the Z axis of the coordinate system, and the real-time oral 3D feature C is registered with the oral 3D feature A and unified in the global coordinate system R of the robot arm. xyz middle;

[0014] Step S4: Establish a coordinate system M for the real-time oral 3D feature C obtained in step S3 xyz , taking the center point of the implant area required by the patient as the coordinate system M xyz The base point, the lingual side is the positive direction of the X axis, the maxillofacial normal is the positive direction of the coordinate system Z axis, in the set local coordinate system S xyz Under the action of xyz With the local coordinate system S xyz The relative position of the oral cavity is obtained by registering the acquired oral cavity 3D feature A with the real-time oral cavity 3D feature C to obtain the coordinate system M xyz With coordinate system N xyz The relative pose of the coordinate system N is obtained by the following calculation formula xyz With the robot global coordinate system R xyz The relative pose of is as follows:

[0015]

[0016] Step S5: completing the confirmation of the posture relationship between the device probe and the patient's oral cavity. Under the condition that the relative posture of the dental drill and the device probe is known, the posture relationship between the dental drill and the patient's oral cavity is confirmed.

[0017] As a preferred embodiment of the present invention, the oral three-dimensional feature A of step S1 is the global three-dimensional feature of the patient, and the oral three-dimensional feature A is composed of the patient's teeth and mucosa. The oral three-dimensional data B is the global three-dimensional data of the patient, and the oral three-dimensional data B is composed of the patient's teeth, mucosa, and alveolar bone taken by a CT machine.

[0018] As a preferred solution of the present invention, the end effector in step S2 is fixedly mounted on the end of the robotic arm, and the relative position and posture of the end effector and the robotic arm can be obtained through calibration.

[0019] As a preferred solution of the present invention, step S3 obtains a real-time display image of oral features through the device probe, obtains data on the location of the patient's teeth and mucosa based on the patient's teeth and mucosa features in the real-time display image, adjusts the position of the device probe to align the device probe with the patient's desired implant area, and obtains the patient's local real-time oral three-dimensional features C through the device probe.

[0020] As a preferred solution of the present invention, step S3 includes the following sub-steps:

[0021] Step S3.1: The device probe transmits the real-time display image of the oral features captured to the computer. The computer extracts feature points from each image and matches them with the feature points in the global oral three-dimensional feature A to find similar feature point pairs. Based on the matched feature point pairs, the computer calculates the geometric transformation relationship between the multiple images captured by the device probe (6), and performs corresponding transformation on the images captured by the device probe (6) so that they are aligned in the same coordinate system to achieve preliminary matching.

[0022] Step S3.2: Extract feature points from the preliminarily matched image, perform feature matching, and verify whether the registration result meets the requirements. If not, perform preliminarily registration on the image again.

[0023] Step S3.3: Generate a set of point cloud data on the oral surface based on the matched image under the action of triangulation method, convert the point cloud data into a mesh or triangular facet, and generate a smooth surface model of the oral structure based on the mesh data using a surface reconstruction algorithm.

[0024] As a preferred solution of the present invention, the computer in step S3.1 pre-processes the collected images before comparison.

[0025] As a preferred solution of the present invention, the feature points extracted from the image in step S3.1 are the patient's teeth, mucosa, and alveolar bone.

[0026] As a preferred solution of the present invention, step S3 further includes the following sub-steps:

[0027] Step S3.4: Multi-level feature fusion segmentation grid, meshing each feature extracted from the oral 3D data A, and performing detailed registration between the real-time oral 3D feature C and each feature mesh after segmentation;

[0028] Step S3.5: Let the vertex set of the mesh reconstructed from the oral 3D data A be P c , the vertex set of the real-time oral 3D feature C grid is P i , the transformation of the oral 3D data A and the real-time oral 3D feature C grid registration is T *= [R|t] (where R∈SO(3) represents the rotation transformation, t∈R 3 Translation transformation), so that P c With P i The spatial distance between corresponding points is minimized, and the registration formula can be defined as follows:

[0029]

[0030] Step S3.6: Locally align the tooth feature points in the real-time oral 3D feature grid C with the tooth feature points reconstructed from the global oral 3D data A, so as to achieve one-to-one alignment between the tooth feature points in the real-time oral 3D feature grid C and the tooth feature points reconstructed from the global oral 3D data A, use the iterative closest point algorithm to estimate the local transformation matrix, and obtain the final refined alignment effect by fusing the alignment results of all tooth feature points.

[0031] As a preferred solution of the present invention, the local registration in step S3.6 correctly maps the tooth categories in the real-time oral 3D feature C to the reconstructed mesh of the oral 3D data A. Assume that the real-time oral 3D feature C contains M groups of tooth point clouds. The N tooth center points in the oral 3D data A are expressed as For each point cloud P i Find the center point with the smallest average distance

[0032] , where |P| is P i The number of point clouds, P ik Point cloud P i The kth point of the tooth, t is the tooth number of the center point with the smallest distance. At this time, the tooth t in the oral 3D data A and the tooth i in the real-time oral 3D feature grid C are considered to be of the same category number. Finally, M groups of optimal transformation matrices can be obtained, which are expressed as IOS grid overall transformation matrix Calculated by the following formula:

[0033]

[0034] , where W is the weight coefficient, F i is the registration fitness of the i-th group of teeth, which indicates the proportion of point pairs successfully matched in the source point cloud to the total point cloud in the source point cloud.

[0035] As a preferred solution of the present invention, the coordinate system M in step S4 xyz With the local coordinate system S xyz The relative pose is the homogeneous coordinate transformation matrix Coordinate system M xyz With coordinate system Nxyz The relative pose is the homogeneous coordinate transformation matrix is the coordinate system N xyz With the robot global coordinate system R xyz The homogeneous coordinate transformation matrix of .

[0036] Compared with existing technologies, the device probe acquires overall information about the teeth and mucosa, and the CT machine acquires information about the bones and implant plan. The three-dimensional oral features acquired by the device probe and the three-dimensional oral data acquired by the CT machine are matched to obtain a unified coordinate system. The device probe then acquires local tooth and mucosal information during the actual scan to obtain real-time three-dimensional oral features. By matching the real-time three-dimensional oral features with the three-dimensional oral features, the relative position of the patient in the unified coordinate system during the actual scan is obtained, completing the positioning.

[0037] It is suitable for large-sized oral locations of patients, such as dental implants at the patient's front teeth, which is convenient for the patient to grow to a larger size and facilitates the end effector and device probe to be inserted into the patient's mouth.

[0038] The use of biological oral three-dimensional features for positioning avoids the problems of high cost of external markers, susceptibility to external interference, and inaccurate positioning caused by uncontrollable deformation of fillings over time. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a structural schematic diagram of the present invention;

[0040] Figure 2 It is a structural diagram of the end effector;

[0041] Figure 3 is a schematic diagram of the extracted oral three-dimensional data B;

[0042] Figure 4 It is a schematic flow diagram of the present invention;

[0043] Figure 5 is a sub-step flow chart of step S3;

[0044] Figure 6 is a schematic diagram of grid separation in step S3;

[0045] Reference numerals: robotic arm 1 , end effector 2 , dental drill 21 , device probe 22 , flange 23 , patient 3 , computer 4 . DETAILED DESCRIPTION

[0046] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0047] like Figures 1-6As shown, a dental implant robot positioning method using oral scanning includes a robotic arm 1, an end effector 2 mounted on the robotic arm 1, and a CT machine, wherein the end effector 2 is mounted with a dental drill 21 for preparing an implant cavity and a device probe 22 for obtaining three-dimensional features of the oral cavity, and the device probe 22 is electrically connected to a computer 4 for collecting image data and for calculating spatial posture.

[0048] The end effector 2 is installed at the end of the robotic arm 1. Under the action of the robotic arm 1, the end effector 2 can be moved within a certain range of space. During the movement, the end effector 2 synchronously drives the movement of the dental drill 21 and the equipment probe 22. The relative position of the dental drill 21 and the equipment probe 22 can be recorded in advance, and the end effector 2 is connected to the robotic arm 1 with a flange 23.

[0049] The actual use process includes the following steps: Step S1: Obtain the patient's oral information, obtain the oral three-dimensional features A of the patient's required implant area through the device probe 22, and obtain the oral three-dimensional data B through the CT machine.

[0050] The device probe 22 captures the three-dimensional features A of the oral cavity and reconstructs a three-dimensional model. The device probe 22 can be, but is not limited to, oral scanning or optical coherence tomography (OCT). Oral scanning technologies include three mainstream approaches: laser triangulation (e.g., 3Shape R700) uses the time difference of laser reflections for 3D reconstruction; structured light technology (e.g., 3DSS STDII) utilizes phase shifts of coded gratings to acquire high-precision data; and stereo photogrammetry (e.g., 3dMD Face) uses multi-camera parallax to resolve spatial coordinates. OCT technology, based on the principle of optical interference, achieves micron-level resolution and can non-invasively examine subsurface structures of teeth.

[0051] The oral 3D feature A in step S1 is the patient's global 3D feature, which is composed of the patient's teeth and mucosa. The oral 3D data B is the patient's global 3D data, which is composed of the patient's teeth, mucosa, and alveolar bone taken by a CT machine. The cavity 3D feature A is registered with the oral 3D data B to establish a unified coordinate system N. xyz , taking the center point of the implant area required by the patient as the coordinate system N xyz The base point is the lingual side as the positive direction of the X axis, and the maxillofacial normal is the positive direction of the Z axis of the coordinate system.

[0052] Step S2: After the end effector 2 is mounted on the end of the robotic arm 1, the follow-up function of the robotic arm 1 is turned on, and the end effector 2 is pulled to the desired implantation area by holding the end of the robotic arm 1. The device probe 22 is aligned with the desired implantation area of ​​the patient, and the local coordinate system of the end effector 2 is set to S xyz , the global coordinate system of robot arm 1 is Rxyz , and use the homogeneous transformation matrix Indicates that the end effector 2 is in the global coordinate system R of the robot 1 xyz The posture relationship in .

[0053] In step S2 , the end effector 2 is fixedly mounted on the end of the robotic arm 1 , and the relative position and posture of the end effector 2 and the robotic arm 1 can be obtained through calibration.

[0054] Step S3: Execute oral 3D feature acquisition to obtain the patient's real-time oral 3D feature C, and align the real-time oral 3D feature C with the oral 3D feature A to determine whether the device probe 22 is in the global coordinate system R of the robot arm 1. xyz The current pose in .

[0055] In step S3, the device probe 22 is used to obtain a real-time display image of the oral features, and data on the positions of the patient's teeth and mucosa are obtained based on the features of the patient's teeth and mucosa in the real-time display image. The position of the device probe 22 is adjusted to align the device probe 22 with the patient's desired implant area, and the patient's local real-time oral three-dimensional features C are obtained through the device probe 22.

[0056] During the process of the device probe 22 acquiring the local real-time three-dimensional features C of the patient's oral cavity, the device probe 22 is aligned with the patient's desired implant area, and the end effector 2 is pulled slowly back and forth near the patient's desired implant area by manually pulling the robotic arm 1 until the device probe 22 scans the complete patient's desired implant area, thereby acquiring a clear real-time three-dimensional feature C of the patient's oral cavity.

[0057] After the scanning program determines that the data of the patient's desired implant area is obtained, the operator stops pulling the end effector 2 back and forth according to the prompt and waits for the current posture to be calculated.

[0058] Step S3 includes the following sub-steps:

[0059] Step S3.1: The device probe 22 transmits the real-time display image of the oral features captured to the computer 4. The computer 4 extracts feature points from each image and matches them with the feature points in the global oral three-dimensional feature A to find similar feature point pairs. Based on the matched feature point pairs, the geometric transformation relationship between the multiple images captured by the device probe (6) is calculated, and the images captured by the device probe (6) are transformed accordingly so that they are aligned in the same coordinate system to achieve preliminary matching.

[0060] The computer 4 pre-processes the collected images before comparison, including operations such as removing noise and enhancing contrast, to ensure image quality. Feature points extracted from the images are the patient's teeth and mucosa.

[0061] Step S3.1 is to align multiple images using one of the feature points in the patient's teeth and mucosa to achieve preliminary registration.

[0062] Step S3.2: Perform preliminary registration on the aligned images, align the coordinate systems of the aligned images to the same standard, extract feature points from the preliminarily matched images, perform feature matching, and verify whether the registration results meet the requirements. If not, perform preliminary registration on the images again.

[0063] Step S3.2 is to re-register the image after the preliminary registration. Based on the preliminary registration, this re-registration identifies whether the other two feature points are in a matching state when one of the feature points is aligned, thereby performing a secondary detection of the image matching state.

[0064] Step S3.3: Based on the matched images, a set of point cloud data on the oral surface is generated under the action of triangulation. The point cloud data is converted into mesh data to represent the geometric shape of the oral surface. Based on the mesh data, a smooth surface model of the oral structure is generated using a surface reconstruction algorithm. Finally, the quality of the reconstructed three-dimensional model is evaluated, including checking indicators such as surface smoothness, topological structure, and model accuracy to verify whether the constructed three-dimensional model meets the requirements.

[0065] Step S3.4: Multi-level feature fusion segmentation grid, grid division of each feature extracted from the oral 3D data A, and detailed registration of the real-time oral 3D feature C with each divided feature grid.

[0066] Each tooth extracted from the oral 3D data A is meshed in an accurate area, and each tooth is rendered in a different color to clearly mesh each tooth. The crown mesh uses the tooth mesh instance segmentation result as the alignment benchmark.

[0067] Step S3.5: Let the vertex set of the mesh reconstructed from the oral 3D data A be P c , the vertex set of the real-time oral 3D feature C grid is P i , the transformation of the oral 3D data A and the real-time oral 3D feature C grid registration is T * =[R|t] where R∈SO(3) represents the rotation transformation, t∈R 3 Translation transformation, so that P c With P i The spatial distance between corresponding points is minimized, and the registration formula can be defined as follows:

[0068]

[0069] Step S3.6: Locally align the tooth feature points in the real-time oral 3D feature grid C with the tooth feature points reconstructed from the global oral 3D data A, so as to achieve one-to-one alignment between the tooth feature points in the real-time oral 3D feature grid C and the tooth feature points reconstructed from the global oral 3D data A, use the iterative closest point algorithm to estimate the local transformation matrix, and obtain the final refined alignment effect by fusing the alignment results of all tooth feature points.

[0070] The tooth instances in the real-time oral 3D feature C grid are locally aligned with the tooth instances reconstructed from the oral 3D data A. For each pair of teeth, the iterative closest point (ICP) algorithm is used to estimate the local transformation matrix. Finally, all the alignment results are fused to obtain the final refined alignment effect.

[0071] The local registration in step S3.6 correctly maps the tooth categories in the real-time oral 3D feature C to the reconstructed mesh of the oral 3D data A. Assume that the real-time oral 3D feature C contains M groups of tooth point clouds. The N tooth center points in the oral 3D data A are expressed as For each point cloud P i Find the center point with the smallest average distance

[0072] , where |P| is P i The number of point clouds, P ik Point cloud P i The kth point of the tooth, t is the tooth number of the center point with the smallest distance. At this time, the tooth t in the oral 3D data A and the tooth i in the real-time oral 3D feature grid C are considered to be of the same category number. Finally, M groups of optimal transformation matrices can be obtained, which are expressed as IOS grid overall transformation matrix Calculated by the following formula:

[0073]

[0074] , where W is the weight coefficient, F i is the registration fitness of the i-th group of teeth, which indicates the proportion of point pairs successfully matched in the source point cloud to the total point cloud in the source point cloud.

[0075] Step S4: Establish a coordinate system M for the real-time oral 3D feature C obtained in step S3 xyz , taking the center point of the implant area required by the patient as the coordinate system M xyz The base point, the lingual side is the positive direction of the X axis, the maxillofacial normal is the positive direction of the coordinate system Z axis, in the set local coordinate system S xyz Under the action ofxyz With the local coordinate system S xyz The relative pose of the coordinate system M xyz With the local coordinate system S xyz The relative pose is the homogeneous coordinate transformation matrix The coordinate system M is obtained by fitting the acquired oral 3D feature A with the real-time oral 3D feature C. xyz With coordinate system N xyz The relative pose of the coordinate system M xyz With coordinate system N xyz The relative pose is the homogeneous coordinate transformation matrix The coordinate system N is obtained by the following calculation formula xyz With the local coordinate system S xyz The relative posture of is the coordinate system N xyz With the global coordinate system R of the robot arm 1 xyz The homogeneous coordinate transformation matrix is ​​as follows:

[0076]

[0077] Step S5: completing the confirmation of the posture relationship between the device probe 22 and the patient's oral cavity. When the relative posture of the dental drill 21 and the device probe 22 is known, the posture relationship between the dental drill 21 and the patient's oral cavity is confirmed.

[0078] The above description of the disclosed embodiments will enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein, but is to be embodied in the widest possible manner consistent with the principles and novel features disclosed herein.

[0079] Although this document frequently uses the following terms, such as robotic arm 1, end effector 2, dental drill 21, device probe 22, flange 23, patient 3, and computer 4, the use of other terms is not excluded. These terms are used solely to facilitate description and explanation of the present invention; any interpretation of these terms as additional limitations would be contrary to the spirit of the present invention.

Claims

1. A dental implant robot positioning method using oral scanning, based on a robotic arm (1), an end effector (2) mounted on the robotic arm (1), and a CT machine, wherein the end effector (2) is mounted with a dental drill (21) for preparing an implant cavity and a device probe (22) for obtaining three-dimensional features of the oral cavity, and the device probe (22) is electrically connected to a computer (4) for collecting image data and calculating spatial posture; characterized in that: The following steps are involved: Step S1: Obtain the patient's oral information, obtain the oral three-dimensional features A of the patient's desired implant area through the device probe (22), and obtain the oral three-dimensional data B through the CT machine, and align the oral three-dimensional features A and the oral three-dimensional data B to establish a unified coordinate system N xyz , taking the center point of the implant area required by the patient as the coordinate system N xyz The base point, the lingual side is the positive direction of the X axis, and the maxillofacial normal is the positive direction of the Z axis of the coordinate system; Step S2: Pull the end effector (2) to the desired implantation area, align the device probe (22) with the patient's desired implantation area, and set the local coordinate system of the end effector (2) to S xyz , the global coordinate system of the robot arm (1) is R xyz The homogeneous transformation matrix R ST is used to represent the end effector (2) in the global coordinate system R of the manipulator (1). xyz The posture relationship in Step S3: Execute oral 3D feature acquisition, obtain the patient's real-time oral 3D feature C, and establish the coordinate system M xyz , taking the center point of the implant area required by the patient as the coordinate system M xyz The base point, the side of the tongue is the positive direction of the X axis, the normal line of the maxillofacial face is the positive direction of the Z axis of the coordinate system, and the real-time oral 3D feature C is registered with the oral 3D feature A and unified in the global coordinate system R of the robot arm. xyz middle; Step S4: Establish a coordinate system M for the real-time oral 3D feature C obtained in step S3 xyz , taking the center point of the implant area required by the patient as the coordinate system M xyz The base point, the lingual side is the positive direction of the X axis, the maxillofacial normal is the positive direction of the coordinate system Z axis, in the set local coordinate system S xyz Under the action of xyz With the local coordinate system S xyz The relative position of the oral cavity is obtained by registering the acquired oral cavity 3D feature A with the real-time oral cavity 3D feature C to obtain the coordinate system M xyz With coordinate system N xyz The relative pose of the coordinate system N is obtained by the following calculation formula xyz With the robot global coordinate system R xyz The relative pose of is as follows: Step S5: Complete the confirmation of the posture relationship between the device probe (22) and the patient's oral cavity. Under the known relative posture of the dental drill (21) and the device probe (22), the posture relationship between the dental drill (21) and the patient's oral cavity is confirmed.

2. The dental implant robot positioning method using oral scanning according to claim 1, characterized in that: The oral three-dimensional feature A in step S1 is the patient's global three-dimensional feature, which is composed of the patient's teeth and mucosa. The oral three-dimensional data B is the patient's global three-dimensional data, which is composed of the patient's teeth, mucosa, and alveolar bone taken by a CT machine.

3. The dental implant robot positioning method using oral scanning according to claim 1, characterized in that: The end effector (2) in step S2 is fixedly mounted on the end of the robotic arm (1), and the relative position and posture of the end effector (2) and the robotic arm (1) can be obtained through calibration.

4. The dental implant robot positioning method using oral scanning according to claim 1, characterized in that: In step S3, a real-time display image of oral features is obtained through the device probe (22), data on the positions of the patient's teeth and mucosa are obtained based on the features of the patient's teeth and mucosa in the real-time display image, the position of the device probe (22) is adjusted to align the device probe (22) with the patient's desired implant area, and the local real-time three-dimensional features C of the patient's oral cavity are obtained through the device probe (22).

5. The dental implant robot positioning method using oral scanning according to claim 4, characterized in that: The step S3 includes the following sub-steps: Step S3.1: The device probe (22) transmits the real-time display image of the oral features captured to the computer (4). The computer (4) extracts feature points from each image and matches them with the feature points in the global oral three-dimensional feature A to find similar feature point pairs. Based on the matched feature point pairs, the computer calculates the geometric transformation relationship between the multiple images captured by the device probe (6), and performs corresponding transformation on the images captured by the device probe (6) so that they are aligned in the same coordinate system to achieve preliminary matching. Step S3.2: Extract feature points from the preliminarily matched image, perform feature matching, and verify whether the registration result meets the requirements. If not, perform preliminarily registration on the image again. Step S3.3: A set of point cloud data on the oral surface is generated based on the matched image under the action of triangulation method, the point cloud data is converted into mesh data, and a smooth surface model of the oral structure is generated based on the mesh data using a surface reconstruction algorithm.

6. The dental implant robot positioning method using oral scanning according to claim 5, characterized in that: The computer (4) in step S3.1 pre-processes the acquired images before comparison.

7. The dental implant robot positioning method using oral scanning according to claim 5, characterized in that: The feature points extracted from the image in step S3.1 are the patient's teeth, mucosa, and alveolar bone.

8. The dental implant robot positioning method using oral scanning according to claim 1, characterized in that: The step S3 further includes the following sub-steps: Step S3.4: Multi-level feature fusion segmentation grid, meshing each feature extracted from the oral 3D data A, and performing detailed registration between the real-time oral 3D feature C and each feature grid after division; Step S3.5: Let the vertex set of the mesh reconstructed by the oral 3D data A be , the vertex set of the real-time oral 3D feature C grid is , the transformation of the oral 3D data A and the real-time oral 3D feature C grid registration is (where R Î SO (3) represents the rotation transformation, translation transformation), so that and The spatial distance between corresponding points is minimized, and the registration formula can be defined as follows: ； Step S3.6: Locally align the tooth feature points in the real-time oral 3D feature grid C with the tooth feature points reconstructed from the global oral 3D data A, so as to achieve one-to-one alignment between the tooth feature points in the real-time oral 3D feature grid C and the tooth feature points reconstructed from the global oral 3D data A, use the iterative closest point algorithm to estimate the local transformation matrix, and obtain the final refined alignment effect by fusing the alignment results of all tooth feature points.

9. The dental implant robot positioning method using oral scanning according to claim 8, characterized in that: The local registration in step S3.6 correctly maps the tooth categories in the real-time oral 3D feature C to the reconstructed mesh of the oral 3D data A. Assume that the real-time oral 3D feature C contains M groups of tooth point clouds. , the N tooth center points in the oral 3D data A are expressed as , for each set of point clouds Find the center point with the smallest average distance , ,in for The number of point clouds, Point Cloud The kth point of the tooth is t, and t is the tooth number of the center point with the smallest distance. At this time, the tooth t in the oral 3D data A and the tooth i in the real-time oral 3D feature grid C are considered to be of the same category number. Finally, M groups of optimal transformation matrices can be obtained, which are expressed as , the transformation matrix of the entire IOS grid Calculated by the following formula: , where W is the weight coefficient, is the registration fitness of the i-th group of teeth, which indicates the proportion of point pairs successfully matched in the source point cloud to the total point cloud in the source point cloud.

10. The dental implant robot positioning method using oral scanning according to claim 1, characterized in that: In step S4, the coordinate system M xyz With the local coordinate system S xyz The relative pose of is the homogeneous coordinate transformation matrix S MT, the coordinate system M xyz With coordinate system N xyz The relative pose is the homogeneous coordinate transformation matrix N MT, R NT is the coordinate system N xyz With the global coordinate system R of the robot (1) xyz The homogeneous coordinate transformation matrix of .

Images