Method, mold and device for manufacturing immediate digital planting guide plate

Abstract

The invention provides a method for manufacturing an immediate digital planting guide plate, which comprises the following steps of: S1, acquiring a tooth impression of a patient, and shooting a CBCT (Cone Beam Computed Tomography) image; s2, identifying positioning anchor point information on the CBCT image, registering the positioning anchor point information with positioning anchor point coordinates of a dental cast tray, placing a simulated implant according to an edentulous position, and obtaining space coordinate position information; s3, rapid hardening gypsum is poured, and a gypsum male mold is manufactured; s4, fixing the gypsum male die on a guide ring placing device, calculating movement parameters of each shaft of the guide ring placing device, and operating the guide ring placing device to respectively move the guide pin and the gypsum male die to required positions and angles; and S5, solidifying the guide ring to the surface of the gypsum male mold through light curing, taking out the guide pin, manufacturing a diaphragm on the surface of the gypsum male mold, molding, finishing, grinding, polishing, and disinfecting for later use. The method improves the implantation precision of the implant, and is simple, rapid, economical and practical. The invention further provides a corresponding coordinate conversion mold and a guide ring placing device.

Description

Technical Field

[0001] The invention relates to the technical field of dentistry, and in particular to a method for manufacturing an immediate digital implant guide. Background Art

[0002] Artificial dental implants, also known as implant dentures, are a type of bionic tooth made of biomaterials. They are divided into two parts: the implant (i.e., the root part) that is implanted in the human jawbone to support and fix the tooth, and the crown that is responsible for chewing. According to the changes in the alveolar bone and jawbone after the patient loses his teeth, the doctor selects an implant of a certain shape and implants it into the jawbone as an artificial root. Then, the crown is installed on the implant post exposed in the mouth to obtain a shape and function similar to that of a natural tooth. With the continuous development of oral implant technology, its clinical success rate and restoration effect have been greatly improved, and it has been accepted by more and more doctors and patients. However, for some important anatomical structures such as the adjacent sites of the neural tube, maxillary sinus, and poor bone conditions in the implant area, the accurate implantation of the implant places high demands on the doctor. The traditional method of drilling holes for dental implants cannot accurately locate the holes. It is operated only by the dentist's experience, and the drilling accuracy and stability are insufficient, which poses surgical risks. In order to overcome the above problems, a simple implant guide technology is usually used. According to the information of radiological imaging, the implantation point of the implant is located on the plaster model, the implantation channel is made, and the guide is made by self-curing plastic or hot pressing by a laminating machine. This type of guide is simple to make and low in cost, and can basically guarantee the effect after restoration. However, its production only refers to the surface information of the model, and lacks consideration of the three-dimensional structure of the bone in the implant area, especially the width and height of the alveolar ridge in the edentulous area, and the precise information of the distribution of nerves and sinus cavities in the jaw, resulting in poor accuracy and safety, and it cannot be combined with CT technology.

[0003] At the same time, in the field of oral medicine, oral cone beam CT (CBCT) is widely used in the diagnosis and plastic surgery of oral and maxillofacial diseases due to its low radiation, high precision and the advantages of providing three-dimensional oral information. For example, it can be used to locate and remove impacted teeth, measure the mandibular nerve canal, and diagnose and treat jaw lesions. In clinical work, dentists can intuitively understand the patient's dental morphology through CBCT images, and use tools to measure oral and maxillofacial parameters to assist in judging the condition and formulating personalized and precise treatment and plastic surgery plans. In particular, in some oral restoration surgeries, doctors need to place a tray with a metal ball into the patient's mouth, and then identify the anchor point based on the CT image, and accurately align it with the point in the mold space coordinate system to complete the accurate calculation of oral data.

[0004] The purpose of the present invention is to provide a method and tool for making a digital implant guide by combining CBCT image positioning anchor point recognition and matching. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention proposes a method for making an immediate digital implant guide, which combines CBCT image anchor point recognition and coordinate transfer matching to improve the implantation accuracy of the implant and has the advantages of being simple, fast, economical and practical.

[0006] To achieve the above object, a method for making an instant digital implant guide of the present invention comprises the following steps:

[0007] S1: Put the prepared oral impression material into the dental tray with positioning anchor points, wear it in the patient's mouth, and ask the patient to do centric occlusion. After the impression solidifies, take the patient to take a CBCT scan. After the CBCT image scan is completed, take out the dental tray;

[0008] S2: Identify the positioning anchor point information on the CBCT image and align it with the positioning anchor point coordinates of the dental model tray. According to the position of the missing tooth, place the simulated implant simulated by the software at the implant position. Associate the spatial coordinate information of the simulated implant with the coordinate information of the positioning anchor point on the CBCT image to obtain the spatial coordinate position information.

[0009] S3: pour quick-setting plaster on the dental mold tray and put it into the mold body, apply plaster on the mold top cover, and then buckle the mold top cover on the mold body. After the plaster solidifies, demould the plaster positive mold and take it out. At this time, the mold top cover, the dental mold tray and the plaster positive mold are integrated;

[0010] S4: The mold top cover with the plaster positive mold is turned over and fixed on the guide ring placement device, the positioning anchor point information on the dental mold tray is associated with the axes of the guide ring placement device, the spatial coordinate information of the simulated implant to be expressed is calculated to be equivalent to the movement parameters of the axes of the guide ring placement device, and the guide ring placement device is operated to move the guide pin and the plaster positive mold to the required position and angle respectively;

[0011] S5: After combining the guide ring and the guide pin, apply a semi-fluid photosensitive adhesive material on the bottom cover of the guide ring, press the guide pin down to contact the surface of the gypsum positive mold, expose the adhesive to light, and take out the guide pin after the adhesive solidifies. At this time, the guide ring has been fixed on the gypsum positive mold at a predetermined angle, and a membrane is made on the surface of the gypsum positive mold. After molding, the membrane including the guide ring is trimmed, and the top and bottom covers of the guide ring are removed, and the guide ring is polished and disinfected for use.

[0012] Furthermore, in step S2, the process of identifying the positioning anchor point information on the CBCT image and aligning it with the positioning anchor point coordinates of the dental mold tray further includes the following steps:

[0013] S201: Data collection and annotation: a dental model tray with positioning anchor points is placed in the oral cavity of multiple patients to collect CBCT images as training data, and the positioning anchor points of the CBCT images are regionally annotated to construct a training set for the deep learning model;

[0014] S202: annotation optimization, extracting the center point position information of each positioning anchor point area based on morphological algorithm and geometric calculation, and automatically generating a circular area on the center coordinate as a detection mark of the positioning anchor point;

[0015] S203: Preprocessing the training data: performing data enhancement and data normalization on each two-dimensional slice of the CBCT image in the training set based on a digital image processing algorithm to increase the diversity and quantity of the training data, thereby improving the generalization performance of the deep learning model;

[0016] S204: Deep network model construction and training, based on the Pytorch deep learning platform, a multi-target detection convolutional neural network model is implemented, CBCT images and corresponding positioning anchor point detection marks are input, and the model loss function is minimized using the built-in Adam optimization algorithm of Pytorch to obtain the optimal deep network model parameters and obtain the trained CBCT image positioning anchor point detection;

[0017] S205: optimizing the detection result of the positioning anchor point, processing all the CBCT image training data based on the trained CBCT image positioning anchor point detection model and the isosurface extraction algorithm, obtaining the 3D segmentation result of the positioning anchor point of each subject, optimizing the positioning anchor point detection result based on the post-processing algorithm, saving it in the ".stl" format, further calculating the spatial information of each positioning anchor point, and reducing the false positive rate of the detection result;

[0018] S206: Based on PCA, the CBCT positioning anchor point is roughly registered to the mold space coordinate system. The three-dimensional coordinate information of the positioning anchor point in the dental mold tray is extracted based on the three-dimensional grid processing algorithm, and the mapping vector between it and the CBCT image positioning anchor point in step S205 is calculated by the principal component analysis algorithm to achieve preliminary registration of the CBCT image space coordinate system to the dental mold tray space coordinate system.

[0019] S207: Based on the ICP fine registration of the CBCT positioning anchor point to the mold space coordinate system, the roughly registered CBCT image positioning anchor point and the dental mold tray positioning anchor point are re-registered based on the iterative nearest neighbor algorithm, and the registration results and parameters are saved. Finally, the three-dimensional surface of the virtual implant is registered to the dental mold tray space.

[0020] Furthermore, in step S204, the multi-target detection convolutional neural network model built based on the Pytorch deep learning platform includes:

[0021] The feature learning network module is composed of ResNet, including 28 convolutional layers with 3×3 kernels, 3 pooling layers, 3 dropout layers and 4 batch normalization layers;

[0022] The feature fusion network module is constructed through the feature pyramid network (FPN), including 6 convolution layers with a convolution kernel of 3×3, 3 upsampling layers, 3 channel connection layers, 2 dropout layers with a dropout rate of 0.6, and 1 spatial attention mechanism module;

[0023] The pixel classifier consists of a convolutional layer with a kernel of 3×3 and a softmax function.

[0024] Furthermore, in step S204, the construction process of the multi-target detection convolutional neural network model includes: designing a loss function of the multi-target detection convolutional neural network, then inputting the CBCT image and the corresponding positioning anchor point markers, and using the Adam optimization algorithm built into Pytorch to minimize the loss function to obtain the optimal model parameters. The loss function is as follows:

[0025]

[0026] Where P L (y=G(x)|x, W) is the probability predicted by the deep network model that the pixel x belongs to the anchor point category, G(x) is the true label of the anchor point region of the pixel x, which is 0 or 1, I is the image data of each slice in the training data CBCT image, λ is the weight value of the second loss importance, N is the number of pixels in the image, Ω is the spatial coordinate domain of the image, the learning rate is set to 0.00001, the number of iterations is 500, and the number of training images in the batch is 20.

[0027] Further, in step S205, firstly, the trained multi-target detection convolutional neural network model is used to process all CBCT image training data to obtain the positioning anchor point classification probability map P of each data. L (x), set the threshold to T L , using the logical operation P L (x)>T L The probability map is converted into a binary result, and then the morphological processing algorithm is used to extract the three-dimensional connected area in the segmentation result. If the number of connected areas is greater than 8, the spatial position information of each connected area is calculated to obtain the average coordinates of all anchor points. The distance between each candidate anchor point and the center point is calculated respectively, and the 8 results with the closest distance are retained as the positioning anchor point detection results. Then, the isosurface extraction algorithm is used to extract the surface information of the three-dimensional positioning anchor point detection results, and the mesh data is generated and saved in the ".stl" format.

[0028] Further, in step S506, the process of using the PCA algorithm to calculate the transfer matrix and distance deviation between the positioning anchor point on the CBCT image and the positioning anchor point on the dental model tray includes:

[0029] First, the PCA algorithm is used to calculate the positioning anchor point surface point cloud V on the CBCT image. B ={[x1, y1, z1], ..., [x m ,y m , z m ]}The mean value v C =[x c ,y c , z c ], remove the mean value of the point cloud data and get Then the covariance matrix is ​​calculated by SVD singular value decomposition method The eigenvalues ​​and eigenvectors of are sorted from large to small, and the eigenvectors are adjusted in order from large to small to obtain the eigenvector matrix P B ;

[0030] The PCA algorithm is used again to calculate the surface point cloud V of the positioning anchor point on the dental model tray M ={[x1, y1, z1], ..., [x n ,y n , z n ]}The mean value v mc =[x mc ,y mc , z mc ], and the eigenvector matrix P M ;

[0031] Based on the above parameters, the rough registration of the positioning anchor points on the CBCT image and the positioning anchor points on the dental model tray is achieved as shown in the following formula:

[0032]

[0033] v t =v m -v m R C

[0034] V BT =V B R C +v T

[0035] Among them, R C is the transfer matrix, v t is the displacement bias term, V BT The point cloud result of roughly registering the anchor points of the CBCT image to the spatial coordinate system of the dental model tray.

[0036] The present invention also provides a coordinate conversion mold for making a digital implant guide, comprising a mold body and a mold top cover, a dental mold tray is arranged between the mold body and the mold top cover, and a plurality of positioning anchor points are arranged on the dental mold tray.

[0037] The present invention also proposes a guide ring placement device for making a digital implant guide plate. The guide ring placement device is provided with a guide pin, which is used to place the guide ring on the surface of the plaster positive mold on the dental mold tray. The guide ring placement device has linear axes in two directions and rotation axes in two directions, and the guide pin has a linear axis in one direction.

[0038] Furthermore, the guide ring comprises a guide ring body, and two ends of the guide ring body are respectively provided with a guide ring top cover and a guide ring bottom cover.

[0039] The beneficial effects of the present invention are:

[0040] 1. By introducing image prior information, the accuracy and precision of the algorithm in identifying the location of the anchor points are greatly improved, and the model's requirements for data annotation accuracy are reduced. In addition, based on the isosurface extraction algorithm and the iterative nearest neighbor algorithm, the three-dimensional coordinate information of the anchor points in the CBCT image is generated, and the three-dimensional coordinate information of the anchor points in the dental mold tray as the standard mold is automatically registered. Experiments show that the algorithm has the advantages of high efficiency, accuracy, and high robustness, and can achieve good detection and registration accuracy in noisy and complex CBCT images;

[0041] 2. The residual network model is integrated with the feature pyramid network model to extract the image features of the CBCT image, and the pixel-by-pixel classifier is used to efficiently identify the position information of the CBCT image positioning anchor points. The PCA and ICP algorithms are further used to align the CBCT image positioning anchor point detection results to the origin coordinate system of the dental mold tray to facilitate the subsequent production of the implant guide.

[0042] In summary, the method of the present invention combines CBCT image anchor point recognition and coordinate transfer matching, improves the guide ring setting accuracy of the implant plate, and thus improves the implantation accuracy of the implant, and has the advantages of being simple, fast, economical and practical. Using the method of the present invention, the guide plate production process for simple cases can be shortened from several hours to several days of traditional 3D printing guide plates to 10-15 minutes, and immediate implantation can be completed according to the case design plan. Moreover, due to the spatial matching of the anchor point and the non-tooth shape registration, it is particularly convenient and quick for edentulous cases and cases where the abutment teeth in the mouth are interfered by metal crowns. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described and explained below in conjunction with the accompanying drawings.

[0044] Figure 1 It is a flow chart of the detection of positioning anchor points on CBCT images and their registration with the positioning anchor points on the dental model tray.

[0045] Figure 2 This is the architecture diagram of the network model for positioning anchor points on CBCT images.

[0046] Figure 3 It is a schematic diagram of CBCT image positioning anchor point annotation, detection results, and three-dimensional reconstruction.

[0047] Figure 4 It is a structural schematic diagram of the coordinate transformation mold.

[0048] Figure 5 It is a structural schematic diagram of a guide ring placement device.

[0049] Figure 6 It is a structural diagram of the guide ring.

[0050] Figure numerals: 11. mold body; 12. mold top cover; 13. dental model tray; 14. positioning anchor point; 15. conversion interface; 21. base; 22. longitudinal beam; 23. support plate; 24. Y-axis electric slide; 25. X-axis electric slide; 26. A-axis rotary stepper motor; 261. working platform; 27. B-axis rotary stepper motor; 271. commutation plate; 272. guide pin; 31. guide ring body; 32. guide ring top closure; 33. guide ring bottom closure. DETAILED DESCRIPTION

[0051] The technical solution of the present invention will be more clearly and completely explained below through description of preferred embodiments of the present invention in combination with the accompanying drawings.

[0052] Example 1: A method for making an instant digital implant guide according to a preferred embodiment of the present invention comprises the following steps:

[0053] S1: Put the prepared oral impression material into the dental mold tray with the positioning anchor point, wear it in the patient's mouth, and ask the patient to do central occlusion. After the impression solidifies, take the patient to take a CBCT. After the CBCT image scanning is completed, take out the dental mold tray. The positioning anchor point uses a high-density radiation-blocking anchor point. In this embodiment, a metal anchor point is used as the positioning anchor point.

[0054] S2: The information collected at this time includes the patient's intraoral (occlusal) bone data and the positioning anchor point data on the dental mold tray. The positioning anchor point information on the CBCT image is identified and aligned with the positioning anchor point coordinates of the dental mold tray. The doctor places the simulated implant simulated by the software in the correct position according to the position of the missing teeth on the CBCT image. The spatial coordinate information of the simulated implant is associated with the coordinate information of the positioning anchor point on the CBCT image to obtain the relevant spatial coordinate position information.

[0055] like Figure 1As shown, the process of identifying the positioning anchor point information on the CBCT image and aligning it with the positioning anchor point coordinates of the dental mold tray specifically includes the following steps:

[0056] S201: Data collection and annotation. A dental mold tray with positioning anchor points is placed in the mouths of multiple patients to collect 50 CBCT images containing dental mold trays as training data. Professionals with oral medicine knowledge use medical image annotation software to perform regional annotation on the positioning anchor points of the CBCT images to build a training set for the deep learning model. The annotation rules are as follows: Each CBCT image has a total of 8 metal anchor points, which are annotated layer by layer from the mandible to the maxilla. Each metal anchor point is annotated with a different label ID number (1-8) to distinguish different metal anchor points from each other.

[0057] S202: Annotation optimization: extract the center point position information of each positioning anchor point area based on morphological algorithm and geometric calculation, and automatically generate a circular area on the center coordinate as the detection mark of the positioning anchor point. In order to improve the accuracy of annotation, calculate the central moment of the annotation result of each positioning anchor point, extract the center point position information of the positioning anchor point, and generate a circular area with a radius of 3 pixels at the center position as the detection mark of the metal anchor point, and finally save it as a binary image.

[0058] S203: Preprocessing of training data: Based on the digital image processing algorithm, data enhancement and data normalization are performed on each two-dimensional slice of the CBCT image in the training set to increase the diversity and quantity of the training data, so as to improve the generalization performance of the deep learning model. The CBCT images and corresponding detection marks are processed using image enhancement algorithms such as image flipping (up and down), angle rotation (90 degrees, 180 degrees and 270 degrees clockwise) and random shearing (the size of the sheared image is 320×320), which increases the order of magnitude of the image by 32 times and improves the diversity of the data. The data is then subjected to maximum and minimum normalization and de-meaning processing to reduce the pixel distribution differences of the data.

[0059] S204: Deep network model construction and training. A multi-target detection convolutional neural network model is implemented based on the Pytorch deep learning platform. The CBCT image and the corresponding positioning anchor point detection mark are input. The model loss function is minimized using the built-in Adam optimization algorithm of Pytorch to obtain the optimal deep network model parameters and the trained CBCT image positioning anchor point detection.

[0060] like Figure 2 As shown in the figure, the multi-target detection convolutional neural network model built on the Pytorch deep learning platform includes:

[0061] The feature learning network module is composed of ResNet, including 28 convolutional layers with 3×3 kernels, 3 pooling layers, 3 dropout layers and 4 batch normalization layers;

[0062] The feature fusion network module is constructed through the feature pyramid network (FPN), including 6 convolution layers with a convolution kernel of 3×3, 3 upsampling layers, 3 channel connection layers, 2 dropout layers with a dropout rate of 0.6, and 1 spatial attention mechanism module;

[0063] The pixel classifier consists of a convolutional layer with a kernel of 3×3 and a softmax function.

[0064] The construction process of the multi-target detection convolutional neural network model includes: designing the loss function of the multi-target detection convolutional neural network, then inputting the CBCT image and the corresponding positioning anchor mark, and using the built-in Adam optimization algorithm of Pytorch to minimize the loss function to obtain the optimal model parameters. The loss function is as follows:

[0065]

[0066] Where P L (y=G(x)|x, W) is the probability predicted by the deep network model that the pixel x belongs to the anchor point category, G(x) is the true label of the anchor point region of the pixel x, which is 0 or 1, I is the image data of each slice in the training data CBCT image, λ is the weight value of the second loss importance, N is the number of pixels in the image, Ω is the spatial coordinate domain of the image, the learning rate is set to 0.00001, the number of iterations is 500, and the number of training images in the batch is 20.

[0067] S205: Optimizing the detection results of the positioning anchor points. First, the trained multi-target detection convolutional neural network model is used to process all CBCT image training data to obtain the positioning anchor point classification probability map P for each data. L (x), set the threshold to T L , using the logical operation P L (x)>T L The probability map is converted into a binary result, and then the morphological processing algorithm is used to extract the three-dimensional connected area in the segmentation result. If the number of connected areas is greater than 8, the spatial position information of each connected area is calculated to obtain the average coordinates of all anchor points. The distance between each candidate anchor point and the center point is calculated respectively, and the 8 results with the closest distance are retained as the positioning anchor point detection results. Then, the marching cubes algorithm is used to extract the surface information of the three-dimensional positioning anchor point detection results, and the grid data is generated and saved in ".stl" format. The spatial information of each positioning anchor point is further calculated to reduce the false positive rate of the detection results.

[0068] S206: Figure 3 As shown, the positioning anchor points of the CBCT image and the positioning anchor points on the dental mold tray have a corresponding spatial position relationship, but there are directional differences and relative displacements in the coordinate systems of the two. The PCA algorithm can be used to calculate the transfer matrix and distance deviation of the two. Based on the PCA rough registration of the CBCT positioning anchor points to the mold space coordinate system, the three-dimensional coordinate information of the positioning anchor points in the dental mold tray is extracted based on the three-dimensional grid processing algorithm, and the mapping vector between the positioning anchor points of the CBCT image in step S205 is calculated by the principal component analysis algorithm, so as to achieve the preliminary registration of the CBCT image space coordinate system to the dental mold tray space coordinate system.

[0069] The process of using the PCA algorithm to calculate the transfer matrix and distance deviation between the positioning anchor points on the CBCT image and the positioning anchor points on the dental model tray includes:

[0070] First, the PCA algorithm is used to calculate the positioning anchor point surface point cloud V on the CBCT image. B ={[x1, y1, z1], ..., [x m ,y m , z m ]}The mean value v C =[x c ,y c , z c ], remove the mean value of the point cloud data and get Then the covariance matrix is ​​calculated by SVD singular value decomposition method The eigenvalues ​​and eigenvectors of are sorted from large to small, and the eigenvectors are adjusted in order from large to small to obtain the eigenvector matrix P B ;

[0071] The PCA algorithm is used again to calculate the surface point cloud V of the positioning anchor point on the dental model tray M ={[x1, y1, z1], ..., [x n ,y n , z n ]}The mean value v mc =[x mc ,y mc , z mc ], and the eigenvector matrix P M ;

[0072] Based on the above parameters, the rough registration of the positioning anchor points on the CBCT image and the positioning anchor points on the dental model tray is achieved as shown in the following formula:

[0073]

[0074] v t =v m -v m R C

[0075] V BT =V B R C +v T

[0076] Among them, R C is the transfer matrix, v t is the displacement bias term, V BT The point cloud result of roughly registering the anchor points of the CBCT image to the spatial coordinate system of the dental model tray.

[0077] S207: Based on ICP fine registration, the CBCT positioning anchor point is registered to the mold space coordinate system. Although the PCA algorithm can efficiently realize the registration of the CBCT metal anchor point to the mold space coordinate system, the algorithm is easily affected by noise, which reduces its registration accuracy. It is necessary to use a local registration method to optimize its registration accuracy. Based on the iterative nearest point algorithm, the anchor point cloud V of the roughly registered CBCT image is located. BT And dental tray positioning anchor point cloud V M The secondary registration is performed, and the number of iterations is set to 500. The algorithm is not allowed to perform size changes, flips, and other operations on the CBCT crown point cloud. The point cloud transfer matrix M is obtained. B , so based on the transfer matrix M B The anchor point results of the CBCT image after fine registration are obtained, the registration results and parameters are saved, and finally the three-dimensional surface of the virtual implant is registered to the dental model tray space.

[0078] S3: After pouring quick-setting plaster on the dental mold tray, place it in the mold body, apply plaster on the mold top cover, and then buckle the mold top cover on the mold body. After the plaster solidifies, demould the plaster positive mold and take it out. At this time, the mold top cover, the dental mold tray and the plaster positive mold are integrated. The positioning anchor point information on the dental mold tray is consistent with the positioning anchor point mapping point in the CBCT image, so that the coordinate information of the simulated implant on the CBCT image is associated with the positioning anchor point of the dental mold tray. The position of the mold top cover and the dental mold tray is relatively fixed, so the coordinate information of the positioning anchor point on the dental mold tray can be associated with the mold top cover, and finally the association between the mold top cover and the virtual implant coordinates is established.

[0079] S4: Flip the mold top cover with the plaster positive mold and fix it on the guide ring placement device. The reference planes of each axis of the guide ring placement device and the reference plane of the mold top cover are fixed parameters. Therefore, the positioning anchor point information on the mold top cover can be associated with the axes of the guide ring placement device. The spatial coordinate information of the simulated implant to be expressed is calculated to be equivalent to the movement parameters of each axis of the guide ring placement device, and the guide ring placement device is operated to move the guide pin and the plaster positive mold to the required positions and angles respectively.

[0080] S5: After combining the guide ring and the guide pin, apply semi-fluid photosensitive adhesive material on the bottom cover of the guide ring, press the guide pin down according to the calculated angle until it contacts the surface of the gypsum positive mold, expose the adhesive to light, and take out the guide pin after the adhesive solidifies. At this time, the guide ring has been fixed on the gypsum positive mold at a predetermined angle. Referring to the orthodontic retainer method, use the guide ring membrane for thermoplastic negative pressure molding, or thermoplastic wax covering, or self-hardening material accumulation, and make a membrane on the surface of the gypsum positive mold. After molding, trim the membrane including the guide ring, remove the top and bottom covers of the guide ring, grind and polish, and disinfect for use.

[0081] Embodiment 2: A coordinate conversion mold for a method of making an instant digital implant guide, such as Figure 4 As shown, it includes a mold body 11 and a mold top cover 12. The top surface of the mold body 11 is provided with a plurality of connecting columns, and the bottom surface of the mold top cover 12 is provided with a plurality of countersunk holes that cooperate with the connecting columns. A dental mold tray 13 is provided between the mold body 11 and the mold top cover 12, and 8 positioning anchor points 14 are provided on the dental mold tray 13. The top surface of the mold top cover 12 is provided with a conversion interface 15 that fixes it to the guide ring placement device.

[0082] Embodiment 3: A guide ring placement device for a method of making an instant digital implant guide, such as Figure 5 As shown, it includes a base 21, a pair of longitudinal beams 22 are arranged on the base 21, and a support plate 23 is arranged on the top of the longitudinal beam 22. A Y-axis electric slide 24 is arranged on the base 21, and an X-axis electric slide 25 is arranged on the slider of the Y-axis electric slide 24, and the sliding directions of the sliders of the Y-axis electric slide 24 and the sliders of the X-axis electric slide 25 are perpendicular to each other on the horizontal plane. An A-axis rotating stepper motor 26 is arranged on the X-axis electric slide 25, and a working platform 261 is arranged on the A-axis rotating stepper motor 26, and the working platform 261 rotates with the rotating shaft of the A-axis rotating stepper motor 26, and a mounting groove for matching the conversion interface 15 is arranged on the working platform 261. A B-axis rotating stepper motor 27 is arranged on the support plate 23, and a commutation plate 271 that rotates with its rotating shaft is arranged on the B-axis rotating stepper motor 27, and a guide pin 272 is arranged on the commutation plate 271, and the guide pin 272 can be pressed down. The positioning of the guide ring is achieved by pressing the guide pin 272 down on the surface of the plaster positive mold on the working platform 261.

[0083] like Figure 6As shown, the guide ring includes a guide ring body 31, and two ends of the guide ring body 31 are respectively provided with a guide ring top cover and a guide ring bottom cover.

[0084] The above specific implementations are only descriptions of the preferred implementations of the present invention, and do not limit the protection scope of the present invention. Without departing from the design concept and spirit of the present invention, various modifications, substitutions and improvements made by ordinary technicians in this field to the technical solution of the present invention based on the text description and drawings provided by the present invention should all fall within the protection scope of the present invention. The protection scope of the present invention is determined by the claims.

Claims

1. A method for making an instant digital implant guide, characterized in that: The following steps are involved: S1: Put the prepared oral impression material into the dental tray with positioning anchor points, wear it in the patient's mouth, and ask the patient to do centric occlusion. After the impression solidifies, take the patient to take a CBCT scan. After the CBCT image scan is completed, take out the dental tray; S2: Identify the positioning anchor point information on the CBCT image and align it with the positioning anchor point coordinates of the dental model tray. According to the position of the missing tooth, place the simulated implant simulated by the software at the implant position. Associate the spatial coordinate information of the simulated implant with the coordinate information of the positioning anchor point on the CBCT image to obtain the spatial coordinate position information. S3: pour quick-setting plaster on the dental mold tray and put it into the mold body, apply plaster on the mold top cover, and then buckle the mold top cover on the mold body. After the plaster solidifies, demould the plaster positive mold and take it out. At this time, the mold top cover, the dental mold tray and the plaster positive mold are integrated; S4: The mold top cover with the plaster positive mold is turned over and fixed on the guide ring placement device, the positioning anchor point information on the dental mold tray is associated with the axes of the guide ring placement device, the spatial coordinate information of the simulated implant to be expressed is calculated to be equivalent to the movement parameters of the axes of the guide ring placement device, and the guide ring placement device is operated to move the guide pin and the plaster positive mold to the required position and angle respectively; S5: After combining the guide ring and the guide pin, apply a semi-fluid photosensitive adhesive material on the bottom cover of the guide ring, press the guide pin down to contact the surface of the gypsum positive mold, expose the adhesive to light, and take out the guide pin after the adhesive solidifies. At this time, the guide ring has been fixed on the gypsum positive mold at a predetermined angle, and a membrane is made on the surface of the gypsum positive mold. After molding, the membrane including the guide ring is trimmed, and the top and bottom covers of the guide ring are removed, and the guide ring is polished and disinfected for use.

2. The method for making an instant digital implant guide according to claim 1, characterized in that: In step S2, the process of identifying the positioning anchor point information on the CBCT image and aligning it with the positioning anchor point coordinates of the dental model tray also includes the following steps: S201: Data collection and annotation: a dental model tray with positioning anchor points is placed in the oral cavity of multiple patients to collect CBCT images as training data, and the positioning anchor points of the CBCT images are regionally annotated to construct a training set for the deep learning model; S202: annotation optimization, extracting the center point position information of each positioning anchor point area based on morphological algorithm and geometric calculation, and automatically generating a circular area on the center coordinate as a detection mark of the positioning anchor point; S203: Preprocessing the training data: performing data enhancement and data normalization on each two-dimensional slice of the CBCT image in the training set based on a digital image processing algorithm to increase the diversity and quantity of the training data, thereby improving the generalization performance of the deep learning model; S204: Deep network model construction and training, based on the Pytorch deep learning platform, a multi-target detection convolutional neural network model is implemented, CBCT images and corresponding positioning anchor point detection marks are input, and the model loss function is minimized using the built-in Adam optimization algorithm of Pytorch to obtain the optimal deep network model parameters and obtain the trained CBCT image positioning anchor point detection; S205: optimizing the detection result of the positioning anchor point, processing all the CBCT image training data based on the trained CBCT image positioning anchor point detection model and the isosurface extraction algorithm, obtaining the 3D segmentation result of the positioning anchor point of each subject, optimizing the positioning anchor point detection result based on the post-processing algorithm, saving it in the ".stl" format, further calculating the spatial information of each positioning anchor point, and reducing the false positive rate of the detection result; S206: Based on PCA, the CBCT positioning anchor point is roughly registered to the mold space coordinate system. The three-dimensional coordinate information of the positioning anchor point in the dental mold tray is extracted based on the three-dimensional grid processing algorithm, and the mapping vector between it and the CBCT image positioning anchor point in step S205 is calculated by the principal component analysis algorithm to achieve preliminary registration of the CBCT image space coordinate system to the dental mold tray space coordinate system. S207: Based on the ICP fine registration of the CBCT positioning anchor point to the mold space coordinate system, the roughly registered CBCT image positioning anchor point and the dental mold tray positioning anchor point are re-registered based on the iterative nearest neighbor algorithm, and the registration results and parameters are saved. Finally, the three-dimensional surface of the virtual implant is registered to the dental mold tray space.

3. The method for making an instant digital implant guide according to claim 2, characterized in that: In step S204, the multi-target detection convolutional neural network model built based on the Pytorch deep learning platform includes: The feature learning network module is composed of ResNet, including 28 convolutional layers with 3×3 kernels, 3 pooling layers, 3 dropout layers and 4 batch normalization layers; The feature fusion network module is constructed through the feature pyramid network (FPN), including 6 convolution layers with a convolution kernel of 3×3, 3 upsampling layers, 3 channel connection layers, 2 dropout layers with a dropout rate of 0.6, and 1 spatial attention mechanism module; The pixel classifier consists of a convolutional layer with a kernel of 3×3 and a softmax function.

4. The method for making an instant digital implant guide according to claim 3, characterized in that: In step S204, the construction process of the multi-target detection convolutional neural network model includes: designing the loss function of the multi-target detection convolutional neural network, then inputting the CBCT image and the corresponding positioning anchor mark, and using the Adam optimization algorithm built into Pytorch to minimize the loss function to obtain the optimal model parameters. The loss function is as follows: Where P L (y=G(x)|x, W) is the probability predicted by the deep network model that the pixel x belongs to the anchor point category, G(x) is the true label of the anchor point region of the pixel x, which is 0 or 1, I is the image data of each slice in the training data CBCT image, λ is the weight value of the second loss importance, N is the number of pixels in the image, Ω is the spatial coordinate domain of the image, the learning rate is set to 0.00001, the number of iterations is 500, and the number of training images in the batch is 20.

5. The method for making an instant digital implant guide according to claim 2, characterized in that: In step S205, firstly, the trained multi-target detection convolutional neural network model is used to process all CBCT image training data to obtain the positioning anchor point classification probability map P of each data. L (x), set the threshold to T L , using logical operation P L (x)>T L The probability map is converted into a binary result, and then the morphological processing algorithm is used to extract the three-dimensional connected area in the segmentation result. If the number of connected areas is greater than 8, the spatial position information of each connected area is calculated to obtain the average coordinates of all anchor points. The distance between each candidate anchor point and the center point is calculated respectively, and the 8 results with the closest distance are retained as the positioning anchor point detection results. Then, the isosurface extraction algorithm is used to extract the surface information of the three-dimensional positioning anchor point detection results, and the mesh data is generated and saved in ".stl" format.

6. The method for making an instant digital implant guide according to claim 2, characterized in that: In step S506, the process of using the PCA algorithm to calculate the transfer matrix and distance deviation between the positioning anchor point on the CBCT image and the positioning anchor point on the dental model tray includes: First, the PCA algorithm is used to calculate the positioning anchor point surface point cloud V on the CBCT image. B ={[x1, y1, z1], ..., [x m ,y m , z m ]}The mean value v C =[x c ,y c , z c ], remove the mean value of the point cloud data and get Then the covariance matrix is ​​calculated by SVD singular value decomposition method The eigenvalues ​​and eigenvectors of are sorted from large to small, and the eigenvectors are adjusted in order from large to small to obtain the eigenvector matrix P B ; The PCA algorithm is used again to calculate the surface point cloud V of the positioning anchor point on the dental model tray M ={[x1, y1, z1], ..., [x n ,y n , z n ]}The mean value v mc =[x mc ,y mc , z mc ], and the eigenvector matrix P M ; Based on the above parameters, the rough registration of the positioning anchor points on the CBCT image and the positioning anchor points on the dental model tray is achieved as shown in the following formula: v t =v m -v m R C V ET =V B R C +v r Among them, R C is the transfer matrix, v t is the displacement bias term, V BT The point cloud result of roughly registering the anchor points of the CBCT image to the spatial coordinate system of the dental model tray.

7. A coordinate conversion mold for a method for making an instant digital implant guide according to any one of claims 1 to 6, characterized in that: The invention comprises a mould body and a mould top cover, wherein a dental mould tray is arranged between the mould body and the mould top cover, and a plurality of positioning anchor points are arranged on the dental mould tray.

8. A guide ring placement device for a method for making an instant digital implant guide according to any one of claims 1 to 6, characterized in that: The guide ring placement device is provided with a guide needle, which is used to place the guide ring on the surface of the plaster positive mold on the dental mold tray. The guide ring placement device has linear axes in two directions and rotation axes in two directions, and the guide needle has a linear axis in one direction.

9. A guide ring placement device for making a digital implant guide according to claim 8, characterized in that: The guide ring comprises a guide ring body, and two ends of the guide ring body are respectively provided with a guide ring top cover and a guide ring bottom cover.

Images