Information processing device, information processing method, and program

Abstract

This provides a mechanism that can relax the constraints related to the design of dental prosthetics. [Solution] An information processing device comprising: a control unit that, based on a patient dentition model which is a 3D model of the patient's dentition, selects one reference dentition model from a plurality of reference dentition models which are 3D models of the dentition of a plurality of other people, and generates a crown model which is a 3D model of a crown to be attached to the abutment teeth of the patient based on the selected reference dentition model.

Description

[[Technical Field]]

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program. [[Background Art]]

[0002] In recent years, the smartification of dental prosthesis production has progressed. For example, designing and producing dental prostheses using 3D (Dimension) cameras, CT (Computed Tomography) scanners, CAD / CAM (Computer-Aided Design / Computer-Aided Manufacturing) devices and the like has become widespread. Further, Patent Document 1 below discloses a technology related to a UI that enables smooth input of production instructions for dental prostheses. [[Prior Art Documents]] [[Patent Documents]]

[0003] [[Patent Document 1]] International Publication No. 2024 / 202093 [[Summary of the Invention]] [[Problem to be Solved by the Invention]]

[0004] However, the technology disclosed in the above Patent Document 1 has not been developed for a long time, and there remains room for improvement from various perspectives. For example, at present, the design of dental prostheses is often performed manually by dental technicians, and the structure that depends on the skill or experience of dental technicians has been a constraint in terms of productivity and the like.

[0005] In view of the above, the present invention has been made in consideration of the above problem, and an object of the present invention is to provide a mechanism capable of relaxing constraints related to the design of dental prostheses. [[Means for Solving the Problem]]

[0006] To solve the above problems, according to one aspect of the present invention, an information processing device is provided which includes a control unit that selects one reference dentition model from a plurality of reference dentition models, which are 3D models of the dentition of a plurality of other people, based on a patient dentition model, which is a 3D model of the patient's dentition, and generates a crown model, which is a 3D model of a crown to be attached to the abutment teeth of the patient, based on the selected reference dentition model.

[0007] The control unit may select a reference dentition model that is similar to the patient dentition model based on labels indicating the position of teeth assigned to each of the multiple vertices that constitute the patient dentition model and the reference dentition model, respectively.

[0008] The control unit may select a reference dentition model in which the teeth adjacent to the abutment tooth are similar to the patient dentition model.

[0009] The control unit may select a reference dentition model in which the opposing teeth of the abutment teeth are similar to the patient dentition model.

[0010] The labels indicating the positions of the teeth are assigned by inputting a 3D model of the dentition into a GNN (Graph Neural Network), and the control unit may select a reference dentition model similar to the patient dentition model based on the feature vectors output from the intermediate layer of the GNN when outputting the labels indicating the positions of the teeth.

[0011] The control unit may select a reference dentition model that is similar to the patient dentition model based on labels indicating the crown morphology assigned to each of the multiple vertices constituting the patient dentition model and the reference dentition model, respectively.

[0012] The control unit may select a reference dentition model that is similar to the patient dentition model based on labels indicating the surface direction assigned to each of the multiple vertices constituting the patient dentition model and the reference dentition model, respectively.

[0013] The control unit may select the reference dentition model of another person whose attribute information is similar to that of the patient.

[0014] The control unit may select a reference dentition model that is similar to the patient dentition model based on the accuracy of the reference dentition model.

[0015] The control unit may generate the crown model by processing a reference tooth model, which is a 3D model of a tooth corresponding to the abutment tooth of the patient from among the selected reference dentition models, based on the patient dentition model.

[0016] The control unit may set the position and orientation of the reference tooth model in the patient dentition model, and generate the crown model by processing the reference tooth model so that it aligns with the adjacent and opposing teeth of the abutment tooth at the set position and orientation.

[0017] The control unit may set the position of the reference tooth model in the patient dentition model so that the reference tooth model is located on the arch line formed by the abutment teeth and adjacent teeth in the patient dentition model.

[0018] The control unit may set the orientation of the reference tooth model based on the angle formed by the tangent to the arch line at the position of the reference tooth model and the boundary line separating the lingual and buccal sides of the surface direction label applied to the reference tooth model.

[0019] The control unit may set the orientation of the reference tooth model based on the distribution of labels indicating the crown morphology applied to the reference tooth model and the distribution of labels indicating the crown morphology applied to the adjacent teeth or the opposing teeth.

[0020] The control unit may readjust the position of the reference tooth model based on the interference between the reference tooth model and the opposing tooth in the set position and orientation.

[0021] The control unit may generate the crown model by changing the size of the reference tooth model or changing the shape of the reference tooth model based on interference between the reference tooth model and the adjacent teeth or the opposing tooth at the set position and posture.

[0022] The control unit may form an inner crown of the crown model by processing the reference tooth model based on interference between the reference tooth model and the abutment tooth at the set position and posture.

[0023] Furthermore, in order to solve the above problem, according to another aspect of the present invention, there is provided an information processing method executed by a computer, the method comprising: selecting one reference dental arch model from a plurality of reference dental arch models that are three-dimensional models of dental arches of a plurality of other people, based on a patient dental arch model that is a three-dimensional model of the patient's dental arch; and generating a crown model that is a three-dimensional model of a crown to be mounted on the patient's abutment tooth based on the selected reference dental arch model.

[0024] Furthermore, in order to solve the above problem, according to another aspect of the present invention, there is provided a program for causing a computer to function as: a control unit that selects one reference dental arch model from a plurality of reference dental arch models that are three-dimensional models of dental arches of a plurality of other people, based on a patient dental arch model that is a three-dimensional model of the patient's dental arch, and generates a crown model that is a three-dimensional model of a crown to be mounted on the patient's abutment tooth based on the selected reference dental arch model. Effects of the Invention

[0025] As described above, according to the present invention, there is provided a mechanism capable of relaxing constraints related to the design of a dental prosthesis. Brief Description of the Drawings

[0026] [Figure 1] It is a diagram showing an example of the configuration of system 1 according to an embodiment of the present disclosure. [Figure 2]This figure shows an example of the information processing flow performed by System 1 according to this embodiment. [Figure 3] This figure shows an example of the information processing flow performed by System 1 according to this embodiment. [Figure 4] This is a diagram illustrating the registration process in this embodiment. [Figure 5] This is a block diagram showing an example of the hardware configuration of the information processing device according to this embodiment. [Modes for carrying out the invention]

[0027] Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings. In this specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0028] <1. Embodiments> Figure 1 is a diagram showing an example of the configuration of System 1 according to one embodiment of the present disclosure. As shown in Figure 1, System 1 according to this embodiment includes a terminal device 10, an application server 20 (hereinafter also referred to as AP server 20), a database server 30 (hereinafter also referred to as DB server 30), and a machine learning server 40.

[0029] System 1 may be configured as a cloud system. For example, the AP server 20, DB server 30, and machine learning server 40 may be built on the cloud, and the terminal device 10 may function as a client.

[0030] System 1 according to this embodiment automatically designs a 3D model of a crown to be attached to the abutment tooth of a patient being treated, through the cooperation of these various devices. The abutment tooth is the tooth being treated. The crown is an example of a dental prosthesis in this embodiment.

[0031] Terminal device 10 is an information processing device operated by an end user. An example of an end user is a dental technician. Terminal device 10 may be, for example, a PC (Personal Computer), and information processing by terminal device 10 may be performed by a browser installed on terminal device 10. Terminal device 10 uploads various patient information to AP server 20 and downloads a 3D model of the crown designed by AP server 20. The dental technician can then manufacture the crown based on the downloaded 3D model.

[0032] The AP Server 20 is an information processing device that controls the overall operation of System 1. Specifically, first, the AP Server 20 selects a reference dentition model similar to the patient dentition model through a similarity search, which will be described later. The patient dentition model is a 3D model of the patient's dentition. The reference dentition model is a 3D model of another person's dentition. Then, the AP Server 20 generates a crown model, which is a 3D model of the crown to be fitted to the patient's abutment teeth, based on the selected reference dentition model. Specifically, the AP Server 20 generates the crown model by processing the reference tooth model, which is a 3D model of the tooth corresponding to the patient's abutment teeth from the selected reference dentition model, based on the patient dentition model through registration and internal crown preparation, which will be described later. With this configuration, it becomes possible to automatically generate a 3D model of a crown that harmonizes with other teeth in the patient's oral cavity, similar to the harmonious relationship between the tooth corresponding to the abutment tooth and other teeth in the reference dentition model. As a result, it becomes possible to improve the productivity of dental technicians, improve the accuracy of crown models, and improve patient satisfaction, thereby easing the constraints related to crown design.

[0033] The DB server 30 is an information processing device on which a DB 32 (see Figure 2) is constructed to store various information necessary for the operation of System 1. The DB 32 constructed by the DB server 30 may be, for example, a VectorDB that stores vectorized feature quantities.

[0034] The machine learning server 40 is an information processing device that performs machine learning-related processing. The machine learning server 40 may have a training GPU (Graphics Processing Unit) equipped with the memory necessary for training a neural network that represents 3D models of several hundred cases. The machine learning server 40 may also have an inference GPU equipped with the memory necessary to perform inference on a neural network that represents a 3D model of a single case.

[0035] The processes performed by System 1 will be described in detail below with reference to Figures 2 and 3. Figures 2 and 3 are diagrams showing an example of the information processing flow performed by System 1 according to this embodiment.

[0036] (Step S102: Data Upload) As shown in Figure 2, first, the terminal device 10 uploads the patient's dental arch model, attribute information, and information identifying the abutment teeth, adjacent teeth, and opposing teeth to the AP server 20.

[0037] A patient dentition model is a 3D model of the patient's dentition. For example, a patient dentition model can be obtained by 3D scanning the patient's oral cavity. Alternatively, a patient dentition model may be obtained by scanning a model obtained from taking an impression of the patient's oral cavity (i.e., a dental impression). Preferably, the patient dentition model is a 3D model of the entire upper and lower jaw in occlusion, or a 3D model that partially includes the teeth to be treated and their surrounding areas in occlusion.

[0038] In this embodiment, the 3D model is a collection of polygons (e.g., triangles) formed by three or more vertices and edges connecting those vertices, and includes positional information for each vertex and information indicating the two vertices connected to each edge. The 3D model may be in various data formats such as STL, PLY, or OBJ. Furthermore, a 3D model of a dental arch containing two or more teeth will be referred to as a dental arch model below. Similarly, a 3D model of a tooth will be referred to as a tooth model below.

[0039] As a post-scan preprocessing step for the patient's dental arch model, the normals of each vertex may be calculated, and the patient's dental arch model may be given the calculated results.

[0040] Attribute information refers to information that indicates the patient's attributes, such as age, gender, race, or ethnicity.

[0041] The information identifying the abutment tooth, adjacent teeth, and opposing teeth is a label indicating the position of the abutment tooth, adjacent teeth, and opposing teeth. In this embodiment, the label indicating the position of the teeth is an FDI number assigned in accordance with the FDI system. Of course, the label indicating the position of the teeth may also conform to the Universal system or the Zsigmondy-Palmer system, etc.

[0042] Adjacent teeth are one or more teeth adjacent to an abutment tooth. Opposing teeth are the teeth on the opposite side of an abutment tooth that come into direct contact when the upper and lower teeth bite together.

[0043] (Step S104: Segmentation) Next, the AP server 20 performs segmentation of the patient dentition model. Segmentation is the process of assigning a label to each vertex of the dentition model that indicates the characteristics of that vertex. The AP server 20 passes the segmented patient dentition model to the next stage of processing. Furthermore, the AP server 20 passes the feature vectors of the patient dentition model obtained during the segmentation process, which will be described later, to the next stage of processing.

[0044] Segmentation involves determining which tooth or gum each vertex of the 3D model of the dentition belongs to and assigning the determination result. The AP server 20 assigns an FDI number or a label indicating the gum to each vertex of the patient's dentition model.

[0045] Segmentation may include the process of assigning anatomical feature labels to each vertex of the patient dentition model. Anatomical feature labels are labels that indicate the anatomical features of each part of the tooth surface, i.e., the crown morphology. Examples of anatomical feature labels include the distal buccal cusp, mesiolingual triangular groove, or central groove. Anatomical feature labels may be expressed numerically. Assigning anatomical feature labels to abutment teeth may be omitted.

[0046] Segmentation may include the process of assigning region annotations to each vertex of the patient's dentition model. Region annotations are labels indicating the surface direction of the tooth within the oral cavity. Examples of region annotations include lingual, buccal, distal, mesial, and occlusal surfaces. Region annotations may be expressed numerically. Note that assigning region annotations to abutment teeth may be omitted.

[0047] The AP server 20 may have the machine learning server 40 perform the actual segmentation processing. For example, the AP server 20 sends the patient dentition model to the machine learning server 40 and obtains the segmented patient dentition model from the machine learning server 40.

[0048] The machine learning server 40 may have already built a GNN (Graph Neural Network) that outputs a segmented 3D model of the dentition when a 3D model of the dentition is input. The GNN built by the machine learning server 40 may be a DGCNN (Dynamic Graph Convolutional Neural Network).

[0049] The AP server 20 may obtain the output from the GNN's intermediate layer as a feature vector during segmentation. A single feature vector is extracted for multiple vertices and edges belonging to the same segment (i.e., those with the same FDI number). The feature vector is a vector that stores the feature quantities for each tooth, i.e., the position information of each vertex, the anatomical feature label assigned to each vertex, and the feature quantities related to region annotation.

[0050] The intermediate layer from which the feature vectors are obtained may be the layer immediately preceding the output layer. By aggregating the outputs from each of the multiple nodes constituting this final hidden layer on a segment-by-segment basis, a feature vector for each tooth can be obtained.

[0051] (Step S106: Similarity Search) Next, the AP server 20 selects one reference dentition model from several reference dentition models, which are 3D models of multiple other people's dentitions, based on the patient's dentition model. Specifically, the AP server 20 compares the patient's dentition model with the multiple reference dentition models, selects the reference dentition model that is determined to be the most similar to the patient's dentition model, and passes it to the next stage of processing. The AP server 20 also passes the reference tooth model, which is a 3D model of the tooth corresponding to the abutment tooth in the selected reference dentition model, i.e., the tooth that has been assigned the same FDI number as the abutment tooth, to the next stage of processing. The reference dentition model and the reference tooth model are referenced when the crown model is automatically generated in the next stage of processing. Note that in the reference dentition model, the teeth corresponding to the patient's abutment tooth, adjacent tooth, or opposing tooth may be simply referred to as abutment tooth, adjacent tooth, or opposing tooth below.

[0052] The DB server 30 has a DB 32 that stores multiple reference dentition models and a DB access layer 31 that provides access to the DB. The AP server 20 searches for a reference dentition model similar to the patient's dentition model from the multiple reference dentition models stored in DB 32 via the DB access layer 31.

[0053] The reference dentition model to be searched is preferably a dentition model in which there are no missing teeth (i.e., no history of treatment) in at least the abutment teeth, adjacent teeth, and opposing teeth of the patient. The reference dentition model has already undergone the segmentation described above, and the same information as the segmented patient dentition model is associated with it. Specifically, the reference dentition model includes positional information for each vertex and information indicating the two vertices to which each edge connects, and further associates the normal vector, FDI number, anatomical feature label, and region annotation for each vertex, as well as feature vectors for each FDI number.

[0054] The AP server 20 selects a reference dentition model similar to the patient dentition model based on the FDI numbers assigned to each of the multiple vertices that make up the patient dentition model and the reference dentition model, respectively. More specifically, the AP server 20 selects a reference dentition model that includes tooth models with the same FDI numbers and similar shapes as at least some of the teeth in the patient dentition model. Considering that teeth with different FDI numbers tend to have significantly different shapes, this configuration makes it possible to select a reference dentition model suitable for generating a crown model.

[0055] The AP server 20 may select a reference dentition model in which the adjacent teeth of the abutment tooth are similar to those of the patient's dentition model. For example, the AP server 20 may select a reference dentition model in which the shapes of the adjacent teeth of the abutment tooth are similar to those of the patient's dentition model. With such a configuration, it becomes possible to generate a 3D model of a crown that harmonizes with the adjacent teeth in the patient's oral cavity, similar to the harmonious relationship between the abutment tooth and the adjacent teeth in the reference dentition model.

[0056] The AP server 20 may select a reference dentition model in which the opposing teeth of the abutment teeth are similar to those of the patient's dentition model. For example, the AP server 20 may select a reference dentition model in which the shape of the opposing teeth of the abutment teeth is similar to that of the patient's dentition model. With such a configuration, it becomes possible to generate a 3D model of a crown that harmonizes with the opposing teeth in the patient's oral cavity, similar to the harmonious relationship between the abutment teeth and opposing teeth in the reference dentition model.

[0057] The AP server 20 may select a reference dentition model similar to the patient dentition model based on anatomical feature labels assigned to each of the multiple vertices that make up the patient dentition model and the reference dentition model, respectively. For example, the AP server 20 may select a reference dentition model in which the distribution of anatomical feature labels assigned to each of the vertices belonging to adjacent teeth or opposing teeth is similar to that of the patient dentition model. With such a configuration, it becomes possible to select a reference dentition model that is more suitable for generating a crown model.

[0058] The AP server 20 may select a reference dentition model similar to the patient dentition model based on the region annotations assigned to each of the multiple vertices that make up the patient dentition model and the reference dentition model, respectively. For example, the AP server 20 may select a reference dentition model in which the distribution of region annotations assigned to each of the vertices belonging to adjacent teeth or opposing teeth is similar to that of the patient dentition model. With such a configuration, it becomes possible to select a reference dentition model that is more suitable for generating a crown model.

[0059] The AP server 20 may select a reference dentition model similar to the patient dentition model based on the feature vectors output from the intermediate layer of the GNN during segmentation. For example, the AP server 20 may select a reference dentition model from the patient dentition model that has a high cosine similarity between at least one feature vector of adjacent teeth or opposing teeth obtained during the segmentation process. The AP server 20 may evaluate the similarity between the patient dentition model and the reference dentition model by summing the cosine similarities of opposing teeth and adjacent teeth or by a weighted average. In the case of a weighted average, the cosine similarity of opposing teeth may be given a greater weight than the cosine similarity of adjacent teeth. With such a configuration, it becomes possible to comprehensively numerically evaluate the various aspects described above and select a reference dentition model that is more suitable for generating a crown model.

[0060] The AP server 20 may select a reference dentition model from another person whose attribute information is similar to that of the patient. For example, the AP server 20 may limit its search to reference dentition models stored in the DB server 30 whose attribute information is similar to that of the patient. The anatomical features of teeth may be similar for each attribute, and the anatomical features of teeth may differ between different attributes. Therefore, by limiting the search to only attributes that are considered to have similar anatomical features to those of the patient, it is possible to speed up the search process and improve the accuracy of the similarity determination.

[0061] The AP server 20 may select a reference dentition model similar to the patient's dentition model based on the accuracy of the reference dentition model. For example, the AP server 20 may select a reference dentition model with higher accuracy from among several reference dentition models with higher similarity obtained through the search. The accuracy of the reference dentition model can be evaluated by the scan resolution or vertex density, etc. The accuracy of the reference dentition model differs depending on whether the scanned object is the actual oral cavity or a model obtained by impression taking, or the accuracy of the scanning equipment, etc. The accuracy of the crown model generated by system 1 may depend on the accuracy of the reference dentition model. In this respect, with this configuration, it is possible to generate a highly accurate crown model.

[0062] The AP server 20 may have the DB server 30 perform the actual similarity search processing. For example, the AP server 20 may send the feature vectors of adjacent teeth and opposing teeth to the DB server 30, and the DB server 30 may obtain a reference dentition model to which the adjacent teeth and opposing teeth with the highest similarity belong.

[0063] (Step S108: Registration) Next, as shown in Figure 3, the AP server 20 performs registration. Then, the AP server 20 passes the registered crown model to the next stage of processing. Registration is a process that generates a crown model by processing a reference tooth model as a base. The AP server 20 may output the crown model and a 4-dimensional homogeneous transformation matrix that shows the position and orientation of the crown model in the patient's dentition model as the registered crown model.

[0064] The AP server 20 generates a crown model by virtually placing a reference tooth model in place of the abutment teeth in the patient dentition model and then processing the reference tooth model. Specifically, the AP server 20 may generate a crown model by setting the position and orientation of the reference tooth model in the patient dentition model and then processing the reference tooth model so that it aligns with the adjacent and opposing teeth of the abutment tooth at the set position and orientation. With this configuration, it becomes possible to harmonize the crown attached to the abutment tooth with the adjacent and opposing teeth.

[0065] More specifically, the AP server 20 may set the position of the reference tooth model in the patient dentition model so that the reference tooth model lies on the arch line formed by the abutment teeth and adjacent teeth in the patient dentition model. For example, the AP server 20 calculates the center position of the bounding box of each tooth in the jaw on the side to which the abutment teeth belong in the patient dentition model, and sets a spline curve that smoothly connects the calculated center positions as the arch line of the patient's dentition. The bounding box may be the region enclosed by the outermost edges connecting vertices with the same FDI number assigned by segmentation, or a rectangle circumscribing that region. The AP server 20 may then set the position of the reference tooth model in the patient dentition model so that the center position of the bounding box of the reference tooth model lies on the arch line.

[0066] This point will be explained in detail with reference to Figure 4. Figure 4 is a diagram illustrating registration in this embodiment. In Figure 4, the center positions PT31-PT38 and PT41-PT48 of the bounding boxes of each tooth in the patient dentition model, and the archline AL, which is a spline curve connecting them, are shown. For example, the AP server 20 sets the position of the reference tooth model RT36 such that the center position of the reference tooth model RT36 lies on the archline AL. As shown in Figure 4, the AP server 20 may also align the center position of the reference tooth model RT36 with the center position PT36 of the abutment tooth.

[0067] In cases of malocclusion, the calculated spline curve may deviate significantly from the patient's actual tooth alignment. As a result, the position of the reference tooth model may differ considerably from its intended placement, potentially requiring manual adjustment by a dental technician. However, various methods can be considered to avoid manual adjustment by a dental technician. For example, the AP server 20 may place the reference tooth model at the positions of abutment teeth recognized by the GNN during the segmentation process within the patient's dentition model. Another example is that the AP server 20 may estimate the arch line of the patient's dentition using a higher-order spline curve, or modify the spline curve to minimize the deviation from the patient's tooth alignment using methods such as the least squares method.

[0068] The AP server 20 may set the orientation of the reference tooth model based on the angle formed by the tangent to the arch line at the position of the reference tooth model and the boundary line separating the lingual and buccal sides of the region annotation applied to the reference tooth model. Note that the orientation of the reference tooth model is synonymous with the orientation of the crown model and can be defined as a rotation angle with the center position of the reference tooth model as the center of rotation. For example, the AP server 20 may set the orientation of the reference tooth model so that the tangent to the arch line at the center position of the reference tooth model and the boundary line separating the lingual and buccal sides of the region annotation applied to the reference tooth model are parallel or approximately parallel. In this case, the AP server 20 may set the orientation of the reference tooth model so that the orientation of the lingual and buccal sides of the reference tooth model matches the orientation of the lingual and buccal sides of the patient dentition model. With such a configuration, it is possible to appropriately set the orientation of the reference tooth model.

[0069] The AP server 20 may set the orientation of the reference tooth model based on the angle formed by the tangent to the arch line at the position of the reference tooth model and the boundary line separating the occlusal side and the gingival side of the region annotation applied to the reference tooth model. For example, the AP server 20 may set the orientation of the reference tooth model so that the tangent to the arch line at the center position of the reference tooth model and the boundary line separating the occlusal side and the gingival side of the region annotation applied to the reference tooth model are parallel or approximately parallel. In this case, the AP server 20 may set the orientation of the reference tooth model so that the orientation of the occlusal side and the gingival side of the reference tooth model matches the orientation of the occlusal side and the gingival side of the patient dentition model. With such a configuration, it becomes possible to set the orientation of the reference tooth model more appropriately.

[0070] The AP server 20 may set the orientation of the reference tooth model based on the distribution of anatomical feature labels assigned to the reference tooth model and the distribution of anatomical feature labels assigned to adjacent or opposing teeth. For example, the AP server 20 may set the orientation of the reference tooth model so that the positions where identical or corresponding anatomical feature labels (e.g., labels with the same direction of concavity / convexity) are distributed smoothly between the reference tooth model and adjacent teeth. As another example, the AP server 20 may set the orientation of the reference tooth model so that the positions where corresponding anatomical feature labels (e.g., labels with opposite directions of concavity / convexity) are distributed between the reference tooth model and opposing teeth coincide or nearly coincide during occlusion. Such correspondence of anatomical feature labels is sometimes referred to as a cusp-to-ridge or cusp-to-fossa. With such a configuration, it becomes possible to set the orientation of the reference tooth model more appropriately.

[0071] The AP server 20 may readjust the position of the reference tooth model based on the interference between the reference tooth model and the opposing tooth at the set position and orientation. For example, after setting the position and orientation of the reference tooth model by the process described above, the AP server 20 determines whether or not the reference tooth model and the opposing tooth interfere. If the reference tooth model and the opposing tooth interfere, the AP server 20 shifts the position of the reference tooth model in the direction of the inner coronal side (i.e., the gingival direction of the abutment tooth) until the reference tooth model and the opposing tooth no longer interfere. The interference determination can be performed by comparing the nearest neighbor distance between vertices with a threshold, or by comparing the volume crossover amount with a threshold, etc. With such a configuration, it becomes possible to set the orientation of the reference tooth model more appropriately.

[0072] The AP server 20 may generate a crown model by changing the size or shape of the reference tooth model based on interference between the reference tooth model and adjacent or opposing teeth at the set position and orientation. For example, if the position of the reference tooth model is readjusted based on interference with opposing teeth, the reference tooth model may come into interference with adjacent teeth. In this case, the AP server 20 may remove the interference between the reference tooth model and adjacent teeth by performing a scaling process in the tangential direction at the center position of the reference tooth model on the arch line. Furthermore, as a result of the above scaling process, the reference tooth model may come into interference with opposing teeth. In this case, the AP server 20 may remove the interference between the reference tooth model and opposing teeth by changing the shape of the interference area in the reference tooth model using anatomical feature labels. With this configuration, it is possible to generate a crown model from which the interference portion has been removed from the reference tooth model.

[0073] (Step S110A: Manual endodontic preparation) Next, the terminal device 10 may download the registered crown model output from the AP server 20 and perform internal crown preparation based on manual operation by a dental technician. Internal crown preparation is the process of shaping the inside of the crown (also called the inner crown) that will be placed on the abutment tooth to match the shape of the abutment tooth. Manual internal crown preparation can be achieved, for example, by using 3DCAD (Three-Dimensional Computer Aided Design) software.

[0074] (Step S110B: Automated endodontic crown formation) Alternatively, the AP server 20 may automatically perform internal crown formation based on the registered crown model.

[0075] More specifically, the AP server 20 may form the inner crown of the crown model by processing the crown model based on the interference between the crown model and the abutment tooth at the position and orientation set in the registration. For example, the AP server 20 may generate a crown model having a shape that fits tightly with the abutment tooth without interfering with each other by removing the part of the crown model that interferes with the abutment tooth. Furthermore, the AP server 20 may generate a crown model with a cement space by removing from the crown model after the part that interferes with the abutment tooth has been removed a part corresponding to the space for filling with cement or other luting material between the abutment tooth and the crown model.

[0076] (Step S112: Adjustment) Finally, the terminal device 10 adjusts the crown model, which has been formed manually or automatically, based on manual operation by a dental technician, and outputs the adjusted crown model. Manual adjustment can be achieved, for example, by 3D CAD operation.

[0077] <2. Hardware Configuration Example> The hardware configuration of the information processing device according to this embodiment will be described below with reference to Figure 5. Figure 5 is a block diagram showing an example of the hardware configuration of the information processing device according to this embodiment. The information processing device 900 shown in Figure 5 can, for example, realize the terminal device 10, AP server 20, DB server 30, or machine learning server 40 shown in Figure 1. Information processing by the terminal device 10, AP server 20, DB server 30, or machine learning server 40 according to this embodiment is realized through the cooperation of software and the hardware described below.

[0078] As shown in Figure 5, the information processing device 900 includes a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, a RAM (Random Access Memory) 903, and a host bus 904a. The information processing device 900 also includes a bridge 904, an external bus 904b, an interface 905, an input device 906, an output device 907, a storage device 908, a communication device 909, and a GPU (Graphics Processing Unit) 910.

[0079] The CPU 901 functions as both an arithmetic processing unit and a control unit, controlling the overall operation of the information processing unit 900 according to various programs. The information processing unit 900 may have, in place of or together with, the CPU 901, an electrical circuit such as a microprocessor, a DSP (Digital Signal Processor), or an ASIC (Application Specific Integrated Circuit). The ROM 902 stores the programs and arithmetic parameters used by the CPU 901. The RAM 903 temporarily stores parameters that change as appropriate during program execution by the CPU 901. The CPU 901 can, for example, constitute a control unit that controls the overall operation of a terminal device 10, an AP server 20, a DB server 30, or a machine learning server 40.

[0080] The CPU 901, ROM 902, and RAM 903 are interconnected by a host bus 904a, which includes the CPU bus. The host bus 904a is connected to an external bus 904b, such as a PCI (Peripheral Component Interconnect / Interface) bus, via a bridge 904. The input device 906, output device 907, storage device 908, communication device 909, and GPU 910 are connected to the external bus 904b via an interface 905. It is not necessary to separate the host bus 904a, bridge 904, and external bus 904b; these functions may be implemented on a single bus.

[0081] The input device 906 is a device that receives information from the user. Examples of such devices include a mouse, keyboard, touch panel, buttons, switches, and levers. In addition, the input device 906 may include a microphone that accepts voice input or a camera that accepts gesture input. The user of the information processing device 900 can input various types of data to the information processing device 900 or instruct it to execute processing by operating the input device 906. The input device 906 accepts user input, for example, in the terminal device 10.

[0082] The output device 907 is a device that outputs information to the user. Examples of such devices include devices that output visual information such as displays and projectors, devices that output auditory information such as speakers, and devices that output tactile information such as eccentric motors. The output device 907 outputs, for example, the results obtained from various processes performed by the information processing device 900. The output device 907 displays, for example, a UI (User Interface) screen that presents various information to the user in the terminal device 10.

[0083] The storage device 908 is a device for storing data. Examples of such devices include magnetic storage devices such as HDDs, semiconductor storage devices, optical storage devices, and magneto-optical storage devices. The storage device 908 may also include a storage medium, a recording device for recording data on the storage medium, a reading device for reading data from the storage medium, and a deletion device for deleting data recorded on the storage medium. For example, the storage device 908 stores DB32 in DB server 30. Also, for example, the storage device 908 stores trained parameters of GNN in machine learning server 40.

[0084] The communication device 909 is a device that communicates with other devices by wire or wireless connection. The communication device 909 performs communication in accordance with any communication standard such as Wi-Fi (registered trademark), Bluetooth (registered trademark), LTE (Long Term Evolution), LPWA (Low Power Wide Area), or USB (Universal Serial Bus).

[0085] The GPU (Graphics Processing Unit) 910 is a processing unit. The GPU 910 may be a parallel processing unit equipped with multiple processors capable of performing parallel processing. The GPU 910 may be equipped with multiple thread processors that perform processing, a control processor that comprehensively controls the thread processors, and local memory for temporarily storing the calculation results. The GPU 910 can perform parallel processing in applications such as image processing, 3D rendering, or machine learning, in accordance with instructions from the CPU 901. For example, the GPU 910 performs training or inference of a GNN on the machine learning server 40.

[0086] The above describes an example of a hardware configuration capable of realizing the functions of the information processing device 900 according to this embodiment. Each of the above components may be realized using general-purpose materials, or it may be realized using hardware specialized for the functions of each component. Therefore, it is possible to change the hardware configuration used as appropriate depending on the level of technology at the time of implementing this embodiment.

[0087] <3. Supplement> Although preferred embodiments of the present invention have been described in detail above with reference to the attached drawings, the present invention is not limited to these examples. It is clear to any person with ordinary skill in the art to which the present invention belongs that various modifications or alterations can be conceived within the scope of the technical idea described in the claims, and these are also understood to fall within the technical scope of the present invention.

[0088] For example, in the above embodiment, a crown was given as an example of a dental prosthesis designed by System 1, but the present invention is not limited to such examples. System 1 may also design bridges, inlays, onlays, dentures, or implant superstructures.

[0089] The series of processes performed by each device described herein may be implemented using software, hardware, or a combination of software and hardware. The programs constituting the software are pre-stored in a recording medium (more specifically, a non-temporary storage medium readable by a computer) provided inside or outside each device. Each program is then loaded into RAM (Random Access Memory) when executed by a computer controlling each device described herein, and executed by a processing circuit such as a CPU (Central Processing Unit). The recording medium is, for example, a magnetic disk, an optical disk, a magneto-optical disk, or flash memory. The computer program may also be distributed via a network, for example, without using a recording medium. The computer may be an application-specific integrated circuit (ASIC), a general-purpose processor that performs functions by loading software programs, or a computer on a server used for cloud computing. The series of processes performed by each device described herein may be centrally processed by a single computer or distributed among multiple computers. Methods for executing the series of processes performed by each device described herein, which are performed by a computer, may also be provided. Furthermore, in each of the above embodiments, two or more communication means present in a single device may be physically implemented in a single medium.

[0090] Furthermore, the processes described herein using flowcharts or sequence diagrams do not necessarily have to be executed in the order shown. Some processing steps may be executed in parallel. Additional processing steps may be adopted, and some processing steps may be omitted. [Explanation of symbols]

[0091] 1 System 10 Terminal devices 20 AP servers 30 DB Servers 40 Machine Learning Servers

Claims

1. A control unit that, based on a patient dentition model which is a 3D model of the patient's dentition, selects one reference dentition model from multiple reference dentition models which are 3D models of multiple other people's dentitions, and generates a crown model which is a 3D model of a crown to be attached to the patient's abutment teeth based on the selected reference dentition model. Equipped with, The control unit selects a reference dentition model that is similar to the patient dentition model based on labels indicating the position of teeth assigned to each of the multiple vertices constituting the patient dentition model and the reference dentition model, respectively. The control unit selects the reference dentition model in which the opposing teeth of the abutment teeth are similar to the patient dentition model. Information processing device.

2. The control unit selects a reference dentition model in which the teeth adjacent to the abutment tooth are similar to the patient dentition model. The information processing apparatus according to claim 1.

3. The labels indicating the position of the teeth are assigned by inputting a 3D model of the dental arch into a GNN (Graph Neural Network). The control unit, when outputting a label indicating the position of the tooth, selects a reference dentition model similar to the patient dentition model based on the feature vector output from the intermediate layer of the GNN. The information processing apparatus according to claim 1.

4. The control unit selects a reference dentition model that is similar to the patient dentition model based on labels indicating the crown morphology assigned to each of the multiple vertices constituting the patient dentition model and the reference dentition model, respectively. The information processing apparatus according to claim 1.

5. The control unit selects a reference dentition model that is similar to the patient dentition model based on labels indicating the surface direction assigned to each of the plurality of vertices constituting the patient dentition model and the reference dentition model, respectively. The information processing apparatus according to claim 1.

6. The control unit selects the reference dentition model of the other person whose attribute information is similar to that of the patient. The information processing apparatus according to claim 1.

7. The control unit selects a reference dentition model that is similar to the patient dentition model based on the accuracy of the reference dentition model. The information processing apparatus according to claim 1.

8. The control unit generates the crown model by processing the reference tooth model, which is a 3D model of the teeth corresponding to the abutment teeth of the patient from among the selected reference dentition models, based on the patient dentition model. The information processing apparatus according to any one of claims 1 to 7.

9. The control unit sets the position and orientation of the reference tooth model in the patient dentition model, and generates the crown model by processing the reference tooth model so that it aligns with the adjacent and opposing teeth of the abutment tooth at the set position and orientation. The information processing apparatus according to claim 8.

10. The control unit sets the position of the reference tooth model in the patient dentition model so that the reference tooth model is positioned on the arch line formed by the abutment teeth and adjacent teeth in the patient dentition model. The information processing apparatus according to claim 9.

11. The control unit sets the orientation of the reference tooth model based on the angle formed by the tangent to the arch line at the position of the reference tooth model and the boundary line separating the lingual and buccal sides of the label indicating the surface direction applied to the reference tooth model. The information processing apparatus according to claim 10.

12. The control unit sets the orientation of the reference tooth model based on the distribution of labels indicating the crown morphology applied to the reference tooth model and the distribution of labels indicating the crown morphology applied to the adjacent teeth or the opposing teeth. The information processing apparatus according to claim 9.

13. The control unit readjusts the position of the reference tooth model based on the interference between the reference tooth model and the opposing tooth in the set position and orientation. The information processing apparatus according to claim 9.

14. The control unit generates the crown model by changing the size of the reference tooth model or changing the shape of the reference tooth model based on the interference between the reference tooth model and the adjacent tooth or the opposing tooth at the set position and orientation. The information processing apparatus according to claim 9.

15. The control unit forms the inner crown of the crown model by machining the reference tooth model based on the interference between the reference tooth model and the abutment tooth at the set position and orientation. The information processing apparatus according to claim 9.

16. A method of information processing performed by a computer, Based on a patient's dentition model, which is a 3D model of the patient's dentition, one reference dentition model is selected from multiple reference dentition models, which are 3D models of multiple other people's dentitions. Based on the selected reference dentition model, a crown model is generated, which is a 3D model of the crown to be fitted onto the patient's abutment teeth. Includes, The selection of the reference dentition model includes selecting a reference dentition model that is similar to the patient dentition model based on labels indicating the position of teeth assigned to each of the multiple vertices that make up each of the patient dentition model and the reference dentition model, Selecting a reference dentition model similar to the patient dentition model includes selecting a reference dentition model similar to the patient dentition model for the opposing teeth of the abutment teeth. Information processing methods.

17. Computers, A control unit that, based on a patient dentition model which is a 3D model of the patient's dentition, selects one reference dentition model from multiple reference dentition models which are 3D models of multiple other people's dentitions, and generates a crown model which is a 3D model of a crown to be attached to the patient's abutment teeth based on the selected reference dentition model. To make it function as, The control unit selects a reference dentition model that is similar to the patient dentition model based on labels indicating the position of teeth assigned to each of the multiple vertices constituting the patient dentition model and the reference dentition model, respectively. The control unit selects the reference dentition model in which the opposing teeth of the abutment teeth are similar to the patient dentition model. program.

Images