Method for automatic detection of dental landmarks

Abstract

To provide a method of identifying an indication of a digital dental object based on a geometric shape so that a suitable manufacturing method can be selected.SOLUTION: A dental recognition method using a manufacturing device (200) is provided, comprising providing a digital dental object in a coordinate system describing a shape of a dental object to be manufactured (S101), and automatically assigning the digital dental object to a predetermined class based on the shape using a self-learning algorithm (S102).SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to a dental object recognition method, a dental object manufacturing device and a computer program for recognizing a dental object. [Background technology]

[0002] In known digital workflows, dental restorations are generated in CAD software and saved in a 3D CAD file format. Depending on the CAD software used and the subsequent workflow, so-called metadata can be generated and stored together with the 3D CAD file, which contains additional information.

[0003] Metadata may help to control or optimize ablative manufacturing methods, such as assigning milling strategies to specific restoration types. In 3D printing methods where support structures are required, such as stereolithography, VAT polymerization, 3D DLP printing, or selective laser sintering of metals, certain surfaces of dental restorations should be free from support structure connections, in order to have a positive impact on accuracy and surface quality.

[0004] However, in the case of metadata, there is the disadvantage that these are already generated in advance in the CAD software and are proprietary data formats. In this case, any CAM software used must be able to import and interpret the associated proprietary metadata. Since these metadata formats are not universal, a separate interface must be programmed for each proprietary metadata format. If an object has no metadata available, the CAM software cannot determine what indicators it is during import.

[0005] However, the index is important for determining which manufacturing method should be used to manufacture the dental object. The classification can also be used in further data processing and to determine further method steps for manufacturing or material processing in a digital workflow.

[0006] US Patent No. 5,399,633 relates to a method for generating a three-dimensional digital model of a denture base for manufacturing using a light-based three-dimensional printing device. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] U.S. Patent No. 10,856,957(B2) Summary of the Invention [Problem to be solved by the invention]

[0008] A technical object of the present invention is to identify indicia of a digital dental object based on its geometric shape so that an appropriate manufacturing method can be selected.

[0009] This technical object is solved by the subject matter of the independent claims. Technically advantageous embodiments are the subject matter of the dependent claims, the description and the drawings. [Means for solving the problem]

[0010] According to a first aspect, the technical object is solved by a method for recognition of a dental object, comprising the steps of providing a digital dental object in a coordinate system describing the shape of the dental object to be manufactured and automatically assigning the digital dental object to a predefined class based on its shape by a self-learning algorithm. The classes may be associated with specific manufacturing methods that can be used to manufacture a real dental object based on the digital dental object. Depending on the class of the dental object, an adapted manufacturing method with optimal parameters can be used.

[0011] In a technically advantageous embodiment of the recognition method, several points on the surface of the digital dental object are detected, which achieves the technical advantage that, for example, a small amount of classification data can be quickly assigned to the digital dental object.

[0012] In a further technically advantageous embodiment of the recognition method, the points on the surface of the digital dental object are selected randomly, which has the technical advantage that, for example, a population of classification data can be obtained without complex calculations.

[0013] In a further technically advantageous embodiment of the recognition method, the coordinates of the detected points form an input for an artificial neural network, whereby the technical advantage is achieved that, for example, digital dental objects can be efficiently assigned to classes.

[0014] In a further technically advantageous embodiment of the recognition method, the artificial neural network is trained with a plurality of training data sets, which also achieves the technical advantage that, for example, digital dental objects can be efficiently assigned to classes.

[0015] In a further technically advantageous embodiment of the recognition method, the class of the digital dental object is output by an artificial neural network, which also achieves the technical advantage that, for example, the digital dental object can be efficiently assigned to a class.

[0016] In a further technically advantageous embodiment of the recognition method, the digital reference object is assigned to the digital dental object based on the assigned class, which has the technical advantage, for example, that the workflow can be further optimized.

[0017] In a further technically advantageous embodiment of the recognition method, the digital dental object is transformed in a coordinate system based on the assigned class and / or the reference object. The digital dental object can, for example, be aligned with the reference object or moved to the reference object. This achieves the technical advantage that, for example, the manufacturing method can be performed with an adapted orientation and / or position of the digital dental object.

[0018] In a further technically advantageous embodiment of the recognition method, a manufacturing method is assigned to the digital dental object based on the assigned class, which has the technical advantage, for example, that dental objects can be manufactured efficiently and that the manufacturing of different dental objects can be automated.

[0019] In a further technically advantageous embodiment of the recognition method, further spatial structures are added to the digital dental object on the basis of the associated manufacturing method. The spatial structures can be support structures for a building platform or holding structures for the blank. This achieves the technical advantage that, for example, the manufacturing of the dental object can be further improved.

[0020] In a further technically advantageous embodiment of the recognition method, the dental object is manufactured by a manufacturing method. This provides the technical advantage, for example, that the dental object is manufactured using a suitable manufacturing method.

[0021] In a further technically advantageous embodiment of the recognition method, the manufacturing method is an additive or subtractive manufacturing method, whereby the technical advantage is achieved, for example, of using a manufacturing method that is particularly suitable for dental purposes.

[0022] In a further technically advantageous embodiment of the recognition method, the accuracy of the assignment is checked by geometric features of the digital dental object, which has the technical advantage, for example, of improving the accuracy of the method.

[0023] According to a second aspect, the technical object is solved by an apparatus for manufacturing dental objects, adapted to perform the recognition method according to the first aspect. By means of the manufacturing apparatus the same technical advantages are achieved as by the recognition method.

[0024] According to a third aspect, the technical problem is solved by a computer program comprising instructions for causing the computer program to perform the method according to the first aspect when the computer program is executed by a computer. By means of the computer program the same technical advantages are achieved as by the recognition method.

[0025] Examples of embodiments of the invention are illustrated in the drawings and are explained in more detail below.

[0026] The drawings show: [Brief description of the drawings]

[0027] [Figure 1] FIG. 1 is a schematic diagram of a classification method. [Diagram 2] FIG. 1 illustrates a number of points from a classification dataset. [Diagram 3] FIG. 1 is a schematic diagram of an artificial neural network. [Figure 4] FIG. 1 illustrates different classes of digital dental objects. [Diagram 5] FIG. 1 illustrates a representation of recognized classes. [Figure 6] FIG. 1 is a block diagram of a method for recognition of a dental object. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] 1 shows a schematic diagram of the classification method. For example, the actual dental object 100-2 is a crown, a bridge, a veneer, an abutment, an inlay, an onlay, a splint, or a partial or complete denture. In general, the dental object 100-2 can be any object in the dental field that is additively or subtractively manufactured as part of a three-dimensional manufacturing method.

[0029] For each of these real dental objects 100-2, there may be a digital dental object 100-1 with specified three-dimensional shape and color values. This information is stored, for example, in a data set for the digital dental object 100-1.

[0030] The classification of the digital dental object 100-1 can be used to determine which type it belongs to. For example, the type of the digital dental object 100-1 can be "crown", "bridge", "veneer", "abutment", "inlay", "onlay", "sprint" or "partial denture" or "complete denture". In general, the type of the digital dental object 100-1 can be any type that can distinguish the real dental object 100-2 from each other.

[0031] For example, differentiation of the digital dental objects 100-1 may be performed based on their respective shapes. Thus, classification may be performed based on a classification data set 101 that includes a number of points on the surface of the digital dental object 100-1. If a large number of points 115 on the surface of the digital dental object 100-1 are known, it may be possible to automatically determine, for example, whether the object is an abutment or a denture.

[0032] Each point in the graph represents a single classification dataset 101 of digital dental objects 100-1 as input data. In nearest neighbor classification, a non-parametric method for estimating a probability density function is performed on the classification dataset 101. The resulting K-nearest neighbors algorithm is a classification procedure in which class assignment is performed considering its k nearest neighbors.

[0033] Classification by Linear Support Vector Machine-SVM (Support Vector Machine) for Classification Dataset 101 uses a classifier and regressor for regression analysis. Support Vector Machines divide a set of digital dental objects into classes such that the widest possible range of objects remains free around the class boundaries. Support Vector Machines are so-called large margin classifiers.

[0034] Additionally, Radial Basis Functions (RBFs) whose values ​​depend on the distance to the origin can be used. In Gaussian classification, a stochastic process is used on the classification data set 101, where each finite subset of random variables is multidimensionally normally distributed (Gaussian distributed).

[0035] Each classification data set 101 can be assigned to a class 103-1, ..., 103-n according to a classification method, such as being the class "abutment" or the class "complete prosthesis". Based on the class of the digital dental object 100-1, a manufacturing method for the real dental object 100-2 can finally be determined.

[0036] 2 shows a number of points 115 on the surface of a digital dental object 100-1 that may be present in the classification dataset 101. Each of the points 115 is defined by coordinates X, Y, and Z in a coordinate system.

[0037] This classification data set 101 is assigned to the class "Complete Prosthesis" by the classification procedure. To make the classification procedure invariant to input permutations, symmetric functions are used. Symmetric functions are functions whose variables can be exchanged without changing the function value.

[0038] A classification method can be trained using multiple classification datasets 101, where the classes of each are known. New classification datasets 101 can then be classified based on their similarity to the trained classification datasets 101.

[0039] In the classification method, depending on the input data and its characteristics, the classification algorithm independently learns an appropriate similarity measure during training to distinguish between different classes of dental objects 100-1. Thus, there is no need to define an explicit similarity measure. The classification algorithm implicitly determines, during the training phase, based on the training data, which properties of the point cloud are relevant for mapping similarity, similar to, for example, a neural network. This is then used to classify the classification dataset 101. Depending on the shape of the training dataset, the similarity measure may vary. This is an effect that commonly occurs in more complex classification algorithms and is intentionally accepted.

[0040] 3 shows a schematic diagram of an artificial neural network 109. The artificial neural network 109 is a network of artificial neurons and can be used to classify a digital dental object 100-1. The artificial neural network 109 comprises, for example, an input layer 111-IN having a number of neurons 113 corresponding to the number of points 115 on the surface of the digital dental object 100-1 in the classification dataset 101. In this case, each point from the classification dataset 101 is input to a separate neuron 113.

[0041] These are transferred to neurons 113 in the hidden layer 111-HIDDEN, which then weight each signal individually from one neuron 113 to another neuron 113, and output the result to the output layer 111-OUT. For example, the number of neurons 113 in the output layer 111-OUT corresponds to the number of classes. In this way, a mapping is created.

[0042]

number

[0043] When the neural network 109 is trained, it is fed multiple classification data sets 101, each of whose classes is known. The neural network 109 learns by modifying the weights between neurons 113 and adjusting the weights until the output classes correspond to the classes known for the classification data sets 101.

[0044] In general, a combination of convolutional and dense layers can be used. Sigmoid, Tanh or ReLU functions can be used as activation functions. Batch normalization can be performed after each layer.

[0045] For example, the present invention can be implemented by the source code shown in Tables 1 and 2 below.

[0046] [Table 1]

[0047] [Table 2]

[0048] 4 shows several digital dental objects 100-1 of different classes, such as a crown (left), an abutment (center), and another crown (right). The surfaces of these dental objects 100-1 are described by a point cloud 117 of random points.

[0049] 5 shows a display of the detected classes. The classes of the digital dental objects 100-1 may be displayed. The unique identifier may then be added to the data set of the digital dental objects 100-1. Thus, the initially unknown data set may become a properly classified data set that can be used again to train a classification algorithm.

[0050] 6 shows a block diagram of a recognition method for a dental object 100-1. The recognition method comprises a step S101 of providing a digital dental object 100-1 in a coordinate system describing the shape of the dental object 100-2 to be manufactured. For this purpose, for example, a 3D dental data set of the digital dental object 100-1 is imported. Then, a number of uniformly distributed random points on the surface of the dental object 100-1, for example 4096 random points, can be determined and stored in the classification data set 101.

[0051] Subsequently, in step S102, the digital dental object 100-1 is automatically assigned to a predefined class based on its shape by a self-learning algorithm. For this purpose, the algorithm is pre-trained with a number of known digital dental objects 100-1 and their respective dental classifications and / or indices. Each indices is a subset of a class. The self-learning algorithm can optimize parameters of a function that maps the geometric information of the digital dental object 100-1 to the respective class.

[0052] The recognition method allows the identification of the three-dimensional data set for the digital dental object 100-1 to be performed automatically and without attached metadata when imported into the CAM software. No static algorithm is used to explicitly identify specific features of the digital dental object 100-1, but a dynamic machine learning algorithm is used where features are learned implicitly.

[0053] The results depend on the 3D structure and the number of available digital dental objects 100-1 with indices used for training. Then, in a further step, validation and validity check of the classification can be performed using simple geometric features based on the dimensions of the digital dental objects 100-1 (bounding box, centroid, local curvature) etc.

[0054] A validity check is performed to check again after classification if a classification result has already been determined. For example, the volume of the dental object 100-1 can be used to check the validity. The volume can be used, for example, to check if the previous classification result "single crown" is correct. The volume can be used to check if it is actually a single crown or if the volume is too large for a single crown and rather corresponds to the volume of a bridge.

[0055] Furthermore, features such as the bounding box dimensions, the maximum bounding box size, the aspect ratio of the bounding box, the curvature of the dental object, or the location of the center of mass of the dental object can be used for validation. A combination of several features can also be used for this purpose.

[0056] In order to enable the CAM-AM software to optimally orient and position the digital dental object 100-1 in the dental technical workflow, the classification of the digital dental object 100-1 should be known. In this case, for example, rules for the orientation of the digital dental object can be defined so that the generation of optimized support structures can be performed automatically.

[0057] If the class, i.e. the indicia, of the digital dental object 100-1 is known to the CAD-CAM software, further measures can be taken on this basis in order to optimize the workflow during manufacture. These measures include, for example, optimizing the alignment of the digital dental object so that the manufactured dental object requires as few support structures as possible and so that most of the dental object's surface no longer needs to be reworked.

[0058] For this purpose, the digital reference object can for example be retrieved from a database based on the determined class. For example, if the digital dental object 100-1 is recognized as a complete denture, a digital reference object of the class "complete denture" can be searched for from the database. The digital dental object 100-1 may then be transformed to the reference object in a coordinate system. For example, the orientation, position or size of the digital dental object 100-1 is adapted to the reference object. In addition, a manufacturing method for the reference object of each class used to manufacture the digital dental object can be provided.

[0059] The method for recognizing the dental object 100-1 can be implemented in a manufacturing device 200 that can be used to manufacture the real dental object 100-2 on the basis of the digital dental object 100-1, such as a printing device for 3D printing or a milling device. For this purpose, the manufacturing device 200 comprises, for example, a computer having a digital memory and a processor for executing a computer program according to the recognition method.

[0060] All the features described and shown in connection with individual embodiments of the invention may also be provided in different combinations in the subject matter of the invention in order to simultaneously realize their advantageous effects.

[0061] All method steps may be performed by means suitable for performing the respective method step. All functions performed by the objective features may be method steps of the method.

[0062] The scope of protection of the invention is given by the claims and is not limited by the features described in the specification or shown in the drawings. [Explanation of symbols]

[0063] 100-1 Digital Dental Object 100-2 Manufactured dental objects 101 Classification Dataset Class 103 109 Artificial Neural Networks 111 layers 113 Neurons 115 items 117 point cloud 200 Manufacturing equipment

Claims

1. A method for recognizing a dental object (100-1), comprising: - providing (S101) a digital dental object (100-1) in a coordinate system describing the shape of a dental object (100-2) to be manufactured; and automatically assigning (S102) said digital dental object (100-1) to a specific class based on its shape by a self-learning algorithm; Detecting a number of points on a surface of the digital dental object (100-1); randomly selecting said points on the surface of said digital dental object (100-1); A method of recognition, wherein the coordinates of said detected points form input to an artificial neural network (109).

2. 2. The method of claim 1, wherein the artificial neural network (109) is trained with a plurality of training data sets.

3. 3. The method of claim 1, wherein said classes are output by said artificial neural network (109).

4. A digital reference object is assigned to the digital dental object (100-1) based on the assigned class. The recognition method according to any one of claims 1 to 3.

5. The method according to any one of claims 1 to 4, wherein the digital dental object (100-1) is transformed based on the assigned class and / or the reference object in the coordinate system.

6. The method of recognition according to any one of the preceding claims, further comprising assigning a manufacturing method to the digital dental object (100-1) based on the assigned class.

7. The method of claim 6, wherein further spatial structure is added to the digital dental object (100-1) based on the associated manufacturing method.

8. The method of claim 6 or 7, wherein the dental object (100-2) is manufactured by the manufacturing method.

9. The method of claim 8, wherein the manufacturing method is an additive or subtractive manufacturing method.

10. Method according to any of the preceding claims, wherein the correctness of the assignment is checked by geometrical features of the digital dental object (100-1).

11. A manufacturing device (200) for dental objects (100-2) adapted to carry out the recognition method according to any one of the preceding claims.

12. A computer program comprising instructions which, when said computer program is executed by a computer, cause said computer program to carry out the method according to any one of claims 1 to 10.

Images