Apparatus for configuring occlusal plane by using computed tomographic image and method therefor

Abstract

A method for configuring an occlusal plane comprises the steps in which: a data processing unit obtains a tomographic image including a hard tissue image and a soft issue image of a facial area including a tooth of a patient through cone beam-computed tomography (CBCT); a landmark detection unit detects a landmark in the tomographic image through a trained detection model; and an occlusal plane definition unit defines an occlusal plane with reference to the landmark detected in the tomographic image.

Description

Device for setting occlusal plane using computed tomography image and method therefor

[0001] The present invention relates to a technology for setting an occlusal plane, and more particularly, to a device for setting an occlusal plane using a computed tomography image and a method therefor.

[0002] Meanwhile, the present invention was supported by the national research and development project in Table 1 below.

[0003] Project number S1402-23-1001 Ministry name Ministry of Science and ICT Project management (specialized) organization name National IT Industry Promotion Agency Research project name AI-based medical device commercialization demonstration support project Research project name AI diagnostic assistance virtual surgery and digital surgical guide for dental implant restoration in a super-aged society: multi-center clinical demonstration Contribution rate 100% Project implementing organization name Dentium ICT Gwanggyo Factory Research period July 1, 2023 ~ December 31, 2025

[0004] When designing prosthetics for edentulous patients, the position of the occlusal plane and vertical dimension are crucial. Conventionally, a wax rim is created using a plaster model to establish the occlusal plane. However, this method fails to reflect the patient's anatomical information, leaving the occlusal plane unclear. Consequently, the occlusal plane position of the wax rim must be adjusted within the patient's mouth, increasing the procedure time.

[0005] The purpose of the present invention is to provide a device for setting an occlusal plane using a computed tomography image and a method therefor.

[0006] A method for setting an occlusal plane according to a preferred embodiment of the present invention for achieving the above-described purpose includes a step in which a data processing unit acquires a tomographic image including hard tissue images and soft tissue images of a facial area including teeth of a patient through cone beam computed tomography (CBCT), a step in which a landmark detection unit detects a landmark from the tomographic image through a learned detection model, and a step in which an occlusal plane definition unit defines an occlusal plane based on the landmark detected from the tomographic image.

[0007] The above landmarks are characterized by including the points at both ends of the Orbitale, Porion, ANS, A-point, Xi, and Mid-palatal suture line in the case of hard tissue images, and the pupil, ala, and tragus in the case of soft tissue images.

[0008] The step of defining the occlusal plane includes a step of the occlusal plane definition unit deriving four planes including the ANS-Porion plane, the Orbitale plane, the mid-palatal suture plane, and the ala-tragus plane using the landmarks, and the occlusal plane definition unit deriving a plane that is parallel to the ANS-Porion plane based on a lateral (sagittal) image among hard tissue images, or is parallel to the ala-tragus plane among soft tissue images, is parallel to the Orbitale plane based on a frontal (coronal) image among hard tissue images, is perpendicular to the mid-palatal suture plane based on a frontal (coronal) image among hard tissue images, is spaced downward from the ANS-Porion plane by a preset distance based on a lateral (sagittal) image among hard tissue images, and is spaced downward from the ala-tragus plane by the above-described distance based on a lateral (sagittal) image among soft tissue images. It includes a step of deriving the occlusal plane.

[0009] The method further includes a step in which a prosthesis design unit generates a prosthesis model based on the occlusal plane according to a user's input, a step in which a prosthesis placement unit places the prosthesis model on a registration image obtained by joining the hard tissue image and the soft tissue image, and a step in which the prosthesis placement unit corrects the position of the prosthesis based on a landmark of the soft tissue image.

[0010] The step of correcting the position of the prosthesis includes a step of the prosthesis placement unit setting an inter-pupillary plane, a step of the prosthesis placement unit setting a midline plane which is a plane perpendicular to the inter-pupillary plane while passing through the center of an inter-pupillary line which is a line on a planar image of the inter-pupillary plane, and a step of the prosthesis placement unit adjusting the position of the prosthesis to align with the midline plane.

[0011] The method further includes, before the step of acquiring the tomographic image, a step of loading learning data including a tomographic image and a target image, which is an image in which a landmark is marked on the tomographic image, by a model learning unit; a step of inputting the tomographic image into a detection model by the model learning unit; a step of deriving a marking image, which is an image in which a landmark is marked on the tomographic image, by the detection model performing a plurality of operations to which weights for which learning has not yet been completed are applied to the tomographic image; a step of deriving a loss representing a difference between the target image and the marking image through a loss function by the model learning unit; and a step of performing optimization by the model learning unit to update the weights of the detection model so that the loss is minimized.

[0012] The above loss function is

[0013] Mathematical formula

[0014]

[0015] , and the above θ is a loss, and the above is the target image, and the above is the transpose matrix of the target image, and is a marking image, and i is characterized in that it is an index of learning data.

[0016] The step of defining the occlusal plane comprises the steps of: setting a first plane perpendicular to the lateral image while the occlusal plane definition part connects the Orbitale and the Porion; setting a second plane parallel to the first plane and passing through A-point by the occlusal plane definition part; setting a third plane parallel to the first plane and spaced apart from the second plane by a predetermined distance from the distance between the first plane and the second plane by the occlusal plane definition part; setting a fourth plane parallel to the first plane and dividing the distance from the second plane to the third plane by a predetermined ratio by the occlusal plane definition part; setting a top point of the mandibular anterior teeth on the fourth plane by the occlusal plane definition part; setting a fifth plane perpendicular to the lateral image while passing through the top point of the mandibular anterior teeth and Xi by the occlusal plane definition part; and setting a fifth plane parallel to the fifth plane and spaced apart from the fifth plane by a predetermined distance downward by the occlusal plane definition part. It includes a step of setting an occlusal plane at a location.

[0017] In order to achieve the above-described purpose, a device for setting an occlusal plane according to a preferred embodiment of the present invention comprises a data processing unit for obtaining a tomographic image including a hard tissue image and a soft tissue image of a facial area including teeth of a patient through cone beam computed tomography (CBCT), a landmark detection unit for detecting a landmark in the tomographic image through a learned detection model, and an occlusal plane definition unit for defining an occlusal plane based on the landmark detected in the tomographic image.

[0018] The above landmarks are characterized by including the points at both ends of the Orbitale, Porion, ANS, A-point, Xi, and Mid-palatal suture line in the case of hard tissue images, and the pupil, ala, and tragus in the case of soft tissue images.

[0019] The above occlusal plane definition unit derives four planes including an ANS-Porion plane, an Orbitale plane, a mid-palatal suture plane, and an ala-tragus plane using the landmarks, and is characterized in that a plane that is parallel to the ANS-Porion plane based on a lateral (sagittal) image among hard tissue images, or parallel to the ala-tragus plane among soft tissue images, is parallel to the Orbitale plane based on a frontal (coronal) image among hard tissue images, is perpendicular to the mid-palatal suture plane based on a frontal (coronal) image among hard tissue images, is spaced downward from the ANS-Porion plane by a preset distance based on a lateral (sagittal) image among hard tissue images, and is spaced downward from the ala-tragus plane by the above-mentioned distance based on a lateral (sagittal) image among soft tissue images is derived as an occlusal plane.

[0020] The device further includes a prosthesis design unit that creates a prosthesis model based on the occlusal plane according to a user's input, and a prosthesis placement unit that places the prosthesis model on a registration image obtained by joining the hard tissue image and the soft tissue image, and corrects the position of the prosthesis based on a landmark of the soft tissue image.

[0021] The above prosthesis placement unit is characterized by setting an inter-pupillary plane, setting a midline plane which is a plane perpendicular to the inter-pupillary plane while passing through the center of the inter-pupillary line which is a line on a planar image of the inter-pupillary plane, and adjusting the position of the prosthesis to match the midline plane.

[0022] The device further includes a model learning unit that loads learning data including a tomographic image and a target image, which is an image in which a landmark is marked on the tomographic image, inputs the tomographic image into a detection model, and when the detection model performs a plurality of operations to which weights that have not yet been learned are applied to the tomographic image to derive a marking image, which is an image in which a landmark is marked on the tomographic image, derives a loss representing a difference between the target image and the marking image through a loss function, and performs optimization to update the weights of the detection model so that the loss is minimized.

[0023] The above loss function is a mathematical formula

[0024]

[0025] , and the above θ is a loss, and the above is the target image, and the above is the transpose matrix of the target image, and is a marking image, and i is characterized in that it is an index of learning data.

[0026] The above occlusal plane definition unit is characterized in that it sets a first plane that is perpendicular to the lateral image while connecting the Orbitale and the Porion, sets a second plane that is parallel to the first plane and passes through the A-point, sets a third plane at a position that is parallel to the first plane and spaced apart from the second plane by a predetermined distance from the distance between the first plane and the second plane, sets a fourth plane at a position that divides the distance from the second plane to the third plane by a predetermined ratio while being parallel to the first plane, sets the uppermost point of the lower anterior teeth on the fourth plane, sets a fifth plane that is perpendicular to the lateral image while passing through the uppermost point of the lower anterior teeth and Xi, and sets an occlusal plane at a position that is parallel to the fifth plane and spaced apart from the fifth plane by a predetermined distance downward.

[0027] According to the present invention, in the case of the present invention, the occlusal plane can be more accurately obtained by using the patient's anatomical landmarks using cone beam computed tomography (CBCT) image data. In particular, even in the case of an edentulous patient without teeth, a more accurate occlusal plane can be obtained by using anatomical landmarks other than teeth. Furthermore, the present invention has the advantage of being able to set and design the prosthesis position with an undistorted image by utilizing the soft tissue image of the CBCT image to replace facial scan information.

[0028] FIG. 1 is a drawing for explaining the configuration of a device for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0029] FIGS. 2 and 3 are screen examples for explaining landmarks for setting an occlusal plane using computed tomography images according to an embodiment of the present invention.

[0030] FIG. 4 is a drawing for explaining a detection model for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0031] Figure 5 is a flowchart for explaining a learning method of a detection model (DM) according to an embodiment of the present invention.

[0032] FIG. 6 is a flowchart illustrating a method for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0033] FIGS. 7 to 10 are screen examples for explaining a method for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0034] FIG. 11 is a flowchart illustrating a method for setting an occlusal plane using a computed tomography image according to another embodiment of the present invention.

[0035] FIG. 12 is an example screen for explaining a method for setting an occlusal plane using a computed tomography image according to another embodiment of the present invention.

[0036] FIG. 13 is a drawing showing a computing device according to an embodiment of the present invention.

[0037] Before going into a detailed description of the present invention, it should be understood that the embodiments described in this specification and the configurations illustrated in the drawings are merely the most preferred embodiments of the present invention and do not represent all of the technical ideas of the present invention, and therefore, there may be various equivalents and modified examples that can replace them at the time of filing this application.

[0038] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. It should be noted that, where possible, identical components are represented by identical reference numerals throughout the drawings. Furthermore, detailed descriptions of well-known functions and structures that may obscure the gist of the present invention will be omitted. For the same reason, some components in the attached drawings are exaggerated, omitted, or schematically depicted, and the sizes of each component do not fully reflect their actual sizes.

[0039] In particular, the terms and words used in the present specification and claims described below should not be interpreted as limited to their usual or dictionary meanings, but should be interpreted as meanings and concepts that conform to the technical idea of ​​the present invention based on the principle that the inventor can appropriately define the concept of the term in order to explain his or her own invention in the best way.

[0040] In particular, in the embodiments of the present invention, the term "segmentation" will be used to mean specifying a target region through artificial neural network operations. For example, "segmenting" a tooth means specifying the tooth by distinguishing it from other regions. Those skilled in the art will understand the meaning of "segmentation" achieved through artificial neural network operations.

[0041] First, a device for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention will be described. Fig. 1 is a drawing for explaining the configuration of a device for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention. Figs. 2 and 3 are screen examples for explaining landmarks for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0042] Referring to FIG. 1, an image processing device (10) according to an embodiment of the present invention includes a model learning unit (100), a data processing unit (200), a landmark detection unit (300), an occlusal plane definition unit (400), a prosthesis design unit (500), and a prosthesis placement unit (600).

[0043] The model learning unit (100) is for generating a detection model (DM) that detects landmarks in a tomographic image captured through cone beam computed tomography (CBCT) according to an embodiment of the present invention. The model learning unit (100) can generate the detection model (DM) through learning (Deep Learning). When the model learning unit (100) generates the detection model (DM), it can provide the generated detection model (DM) to the landmark detection unit (300).

[0044] The data processing unit (200) is for acquiring a tomographic image. In the embodiment of the present invention, the tomographic image is a collection of multiple images captured through cone beam computed tomography (CBCT), and the captured area includes the facial area including the patient's dental area. In addition, the tomographic image includes hard tissue images and soft tissue images. These tomographic images can be input by the user or directly input from a connected radiographic imaging device (not shown).

[0045] The landmark detection unit (300) is for detecting landmarks in a tomographic image using a detection model (DM) for which learning has been completed. At this time, the landmark detection unit (300) performs a weight operation in which learned weights are applied to the tomographic image using the detection model (DM) for which learning has been completed, thereby deriving a marking image, which is an image in which landmarks are marked, from the tomographic image, and can detect landmarks through the marking image.

[0046] As shown in Fig. 2, for hard tissue images, landmarks include the left and right Orbitale, Porion, ANS, A-point, Xi, and points at both ends of the mid-palatal suture line. As shown in Fig. 3, for soft tissue images, landmarks include the left and right pupil, ala, and tragus.

[0047] The occlusal plane definition unit (400) is intended to define the occlusal plane based on landmarks detected from a single-layer image. To this end, the occlusal plane definition unit (400) first derives four reference planes using landmarks, and can derive an occlusal plane that satisfies predetermined conditions based on the four reference planes.

[0048] The prosthesis design department (500) generates a prosthesis model based on the occlusal plane according to user input. Here, the prosthesis model may be a three-dimensional model composed of multiple voxels. Alternatively, the prosthesis model may be a three-dimensional mesh model.

[0049] The prosthesis placement unit (600) is for placing a prosthesis model on a matched image that combines hard tissue images and soft tissue images according to user input, or for correcting the position of the prosthesis model on the matched image.

[0050] The specific operation of the image processing device (10) including the aforementioned model learning unit (100), data processing unit (200), landmark detection unit (300), occlusal plane definition unit (400), prosthesis design unit (500), and prosthesis placement unit (600) will be described in more detail below.

[0051] Next, a detection model for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention will be described. FIG. 4 is a diagram for explaining a detection model for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0052] Referring to FIG. 4, a detection model (DM) according to an embodiment of the present invention includes multiple layers. The detection model (DM) includes an encoder (EN) and a decoder (DE). That is, the detection model (DM) includes multiple layers, and the multiple layers can be divided into an encoder (EN) and a decoder (DE).

[0053] The encoder (EN) includes an input layer (IL), multiple convolutional layers (CL), and multiple pooling layers (PL).

[0054] The decoder (DE) includes multiple concatenated layers (NL), multiple convolutional layers (CL), multiple upsampling layers (UL), and an output layer (OL).

[0055] Multiple layers of a detection model (DM) generate feature maps (FMs) through multiple operations where weights are applied between the multiple layers. In Figure 4, the hexahedrons represent feature maps (FMs) for each of the multiple layers.

[0056] The convolutional layer (CL) performs convolution operations and operations using an activation function. Activation functions include, but are not limited to, sigmoid, hyperbolic tangent (tanh), exponential linear unit (ELU), rectified linear unit (ReLU), leaky ReLU, maxout, minout, and softmax. In the embodiment of the present invention, it is preferable to use rectified linear unit (ReLU) for operations using an activation function.

[0057] The pooling layer (PL) is for down-sampling and performs a pooling (max pooling) operation.

[0058] The upsampling layer (UL) is used for upsampling and can perform up-convolution operations.

[0059] The concatenation layer (NL) combines the feature map (FM) of the convolutional layer (CL) of the encoder (EN) and the feature map (FM) of the upsampling layer (UL) of the decoder (DE) to form a feature map (FM).

[0060]

[0061] *As described above, the detection model (DM) includes multiple layers, and the multiple layers include multiple operations. Furthermore, the multiple layers are connected by weights (W). The computational results of one layer are weighted and input to the next layer. In other words, a layer of the detection model (DM) receives a weighted value from the previous layer, performs an operation on it, and passes the result of the operation as the input to the next layer. This computational procedure of the detection model (DM) is referred to as "weighted computation."

[0062] As illustrated in Fig. 4, when a tomographic image (CI) is input, the encoder (EN) performs encoding to compress the features of the tomographic image (CI) to derive a latent vector. This latent vector can become a feature map of the second convolutional layer of the encoder (EN). The decoder (DE) performs decoding to restore the latent vector to the size of the tomographic image (CI), thereby deriving a marking image (MI).

[0063] The more specific weighting calculation procedures for these detection models (DMs) are listed sequentially as follows. In the explanation below, the feature map, which is the output of the previous layer, is weighted and input to the calculations of the next layer.

[0064] When a single-layer image (CI) is input to the input layer (IL), the first pooling layer (PL1) of the encoder (EN) performs a pooling operation on the single-layer image to derive a first feature map (FM1).

[0065] The first convolutional layer (CL1) performs a convolution operation and an operation using an activation function on the first feature map (FM1) to derive the second feature map (FM2).

[0066] The second pooling layer (PL2) performs a pooling operation on the second feature map (FM2) to derive the third feature map (FM3).

[0067] The second convolutional layer (CL2) performs a convolution operation and an operation using an activation function on the third feature map (FM3) to derive the fourth feature map (FM4).

[0068] Next, the third convolutional layer (CL3) of the decoder (DE) performs a convolution operation and an operation using an activation function on the fourth feature map (FM4) to derive a fifth feature map (FM5).

[0069] The first upsampling layer (UL1) performs an upconvolution operation on the fifth feature map (FM5) to derive the sixth feature map (FM6).

[0070] The first concatenated layer (NL1) combines the second feature map (FM2) and the sixth feature map (FM6) to form the seventh feature map (FM7).

[0071] The fourth convolutional layer (CL4) performs a convolution operation and an activation function operation on the seventh feature map (FM7) to derive the eighth feature map (FM8).

[0072] The second upsampling layer (UL2) performs an upconvolution operation on the eighth feature map (FM8) to derive the ninth feature map (FM9).

[0073] The second combined layer (NL2) combines the single-layer image (CI) and the ninth feature map (FM9) to form the tenth feature map (FM10).

[0074] The fifth convolutional layer (CL5) performs a convolution operation and an operation using an activation function on the tenth feature map (FM10) to derive the eleventh feature map (FM11).

[0075] The output layer (OL) performs an operation using an activation function on the eleventh feature map (FM11) to derive a marking image (MI). The marking image (MI) is an image in which landmarks predicted from the single-layer image (CI) are marked.

[0076] Next, a method for learning a detection model according to an embodiment of the present invention will be described. Fig. 5 is a flowchart illustrating a method for learning a detection model (DM) according to an embodiment of the present invention.

[0077] Referring to FIGS. 4 and 5, the model learning unit (100) loads pre-saved learning data in step S110.

[0078] The training data includes a tomographic image (CI) and a target image (TI) corresponding to the tomographic image. In the embodiment of the present invention, the tomographic image is a set of multiple images captured using cone beam computed tomography (CBCT), and the captured area includes the facial area including the patient's dental area. In addition, the tomographic image (CI) includes a hard tissue image and a soft tissue image. In addition, the target image (TI) is an image in which landmarks are marked on the tomographic image. As illustrated in FIG. 2, in the case of the hard tissue image, the landmarks include the left and right Orbitale, Porion, ANS, A-point, Xi, and points at both ends of the mid-palatal suture line. In addition, as illustrated in FIG. 3, in the case of the soft tissue image, the landmarks include the left and right pupil, ala, and tragus. This training data is pre-generated and stored.

[0079] Once the learning data is prepared, the model learning unit (100) inputs the single-layer image (CI) into the detection model (DM) whose learning has not been completed in step S120.

[0080] Then, the detection model (DM) derives a marking image (MI) by performing multiple operations on the single-layer image (CI) at step S130, where multiple weighted operations are assigned that have not yet completed learning across multiple layers. The specific procedure by which the detection model (DM) derives the marking image (MI) is as described above in Figure 4.

[0081] Accordingly, the model learning unit (100) can calculate a loss representing the difference between the target image (TI) and the marking image (MI) through a loss function at step S140. At this time, the model learning unit (100) can calculate the loss through a loss function according to the following mathematical expression 1.

[0082]

[0083] Here, θ represents loss. is the target image, and the above represents the transpose matrix of the target image. Also, represents the marking image. And i is the index of the learning data.

[0084] Next, the model learning unit (100) performs optimization to modify the weights of the detection model (DM) so that the loss derived through the loss function is minimized in step S150.

[0085] Steps S120 to S150 described above are repeated using different training data until a predetermined training termination condition is satisfied. Here, the training termination condition may be when, when the detection model (DM) derives a marking image (MI), the loss representing the difference between the target image (TI) and the marking image (MI) of the training data converges and falls below a preset threshold.

[0086] Accordingly, the model learning unit (100) determines whether the learning termination condition is satisfied in step S160, and if the learning termination condition is satisfied, the learning is terminated in step S170.

[0087] Next, a method for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention will be described. Fig. 6 is a flowchart for explaining a method for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention. Figs. 7 to 10 are screen examples for explaining a method for setting an occlusal plane using a computed tomography image according to an embodiment of the present invention.

[0088] Referring to FIG. 6, the data processing unit (200) acquires a tomographic image at step S210. In the embodiment of the present invention, the tomographic image is a collection of multiple images captured through cone beam computed tomography (CBCT), and the captured area includes the facial area including the patient's dental area. In addition, the tomographic image includes hard tissue images and soft tissue images. These tomographic images may be input by a user or directly input from a connected radiographic imaging device (not shown).

[0089] Next, the landmark detection unit (300) detects landmarks in the tomographic image using the detection model (DM) for which learning has been completed in step S220. At this time, the landmark detection unit (300) performs a weight operation in which the learned weights are applied to the tomographic image using the detection model (DM) for which learning has been completed, thereby deriving a marking image, which is an image in which landmarks are marked, from the tomographic image, and can detect landmarks through the marking image.

[0090] Referring to FIGS. 2 and 3, in the embodiment of step S220, for a hard tissue image, landmarks including points at both ends of the left and right Orbitale, Porion, ANS, and Mid-palatal suture line are used, and for a soft tissue image, landmarks including the left and right Pupil, ala, and tragus are used.

[0091] When a landmark is detected, the occlusal plane definition unit (400) can define the occlusal plane based on the landmark detected in the single-layer image.

[0092] To this end, the occlusal plane definition unit (400) first derives four reference planes using landmarks in step S230. These reference planes include the ANS-Porion plane, the Orbitale plane, the mid-palatal suture plane, and the ala-tragus plane. Here, the ANS-Porion plane connects the landmarks ANS and Porion and is a plane perpendicular to the lateral (sagittal plane) image. The Orbitale plane connects the left and right Orbitale and is a plane perpendicular to the frontal (coronal plane) image. The mid-palatal suture plane connects the points at both ends of the mid-palatal suture line and is a plane perpendicular to the transverse plane image. And the ala-tragus plane connects the ala and tragus and is a plane perpendicular to the lateral (sagittal plane) image.

[0093] And the occlusal plane definition unit (400) derives an occlusal plane that satisfies the following conditions based on four reference planes in step S240. That is, as illustrated in FIGS. 7 and 8, the occlusal plane definition unit (400) derives a plane as the occlusal plane that is parallel to the ANS-Porion plane based on the lateral (sagittal plane) image among hard tissue images, or parallel to the ala-tragus plane among soft tissue images, parallel to the Orbitale plane based on the frontal (coronal plane) image among hard tissue images, perpendicular to the mid-palatal suture plane based on the frontal (coronal plane) image among hard tissue images, spaced downward by a preset distance from the ANS-Porion plane based on the lateral (sagittal plane) image among hard tissue images, and spaced downward by the above-described distance from the ala-tragus plane based on the lateral (sagittal plane) image among soft tissue images.

[0094] Once the occlusal plane is derived, the prosthesis design unit (500) generates a prosthesis model based on the occlusal plane according to the user's input at step S250. Here, the prosthesis model may be a three-dimensional model composed of multiple voxels. Alternatively, the prosthesis model may be a three-dimensional model as a mesh model.

[0095] Next, the prosthesis placement unit (600) places the prosthesis model on the aligned image obtained by joining the hard tissue image and the soft tissue image according to the user's input at step S260.

[0096] Then, the prosthesis placement unit (600) corrects the position of the prosthesis based on a plane derived from the landmark of the soft tissue image in step S270. Here, the plane includes the derived occlusal plane.

[0097] According to one embodiment, as illustrated in FIG. 9, a user may input information to correct the position of the prosthesis while viewing the face in a matching image in which the prosthesis model is placed. Based on this input, the prosthesis placement unit (600) may correct the position of the prosthesis.

[0098] According to another embodiment, as illustrated in FIG. 10, the prosthesis placement unit (600) first sets an inter-pupillary plane (IPP). The inter-pupillary plane (IPP) is a plane that connects the points at both ends of the mid-palatal suture line and is perpendicular to a transverse plane image. Then, the prosthesis placement unit (600) sets a mid-line plane (MLP), which is a plane that passes through the center of the inter-pupillary line and is perpendicular to the inter-pupillary plane (IPP). Here, the inter-pupillary line represents a line on a transverse plane image of the inter-pupillary plane (IPP). Subsequently, the prosthesis placement unit (600) can adjust the position of the prosthesis to align with the mid-line plane (ML).

[0099] Next, a method for setting an occlusal plane using computed tomography images according to another embodiment of the present invention will be described. Fig. 11 is a flowchart for explaining a method for setting an occlusal plane using computed tomography images according to another embodiment of the present invention. Fig. 12 is an example screen for explaining a method for setting an occlusal plane using computed tomography images according to another embodiment of the present invention.

[0100] Referring to FIGS. 11 and 12, the data processing unit (200) acquires a tomographic image at step S310. In the embodiment of the present invention, the tomographic image is a collection of multiple images captured using cone beam computed tomography (CBCT), and the captured area includes the facial area including the patient's dental area. This tomographic image may be input by a user or directly input from a connected radiographic imaging device (not shown).

[0101] Next, the landmark detection unit (300) detects landmarks in the tomographic image using the detection model (DM) for which learning has been completed in step S320. At this time, the landmark detection unit (300) performs a weight operation in which the learned weights are applied to the tomographic image using the detection model (DM) for which learning has been completed, thereby deriving a marking image, which is an image in which landmarks are marked, from the tomographic image, and can detect landmarks through the marking image. In this embodiment, landmarks use Porion, Orbitale, A-point, and Xi.

[0102] Once a landmark is detected, the occlusal plane definition unit (400) can define the occlusal plane using the landmark detected in the single-layer image. This is described in detail as follows.

[0103] The occlusal plane definition unit (400) sets a first plane (PL1) perpendicular to the lateral (sagittal plane) image while connecting the Orbitale and Porion in step S330.

[0104] Next, the occlusal plane definition unit (400) sets a second plane (PL2) that passes through A-point while being parallel to the first plane (PL1) in step S340.

[0105] Next, the occlusal plane definition unit (400) sets a third plane (PL3) at a position (a:b=1:1.618) that is parallel to the first plane (PL1) and is spaced apart from the second plane by 1.618 times the distance between the first plane and the second plane in step S350. The present invention provides a guideline that can set the mandibular inflection point (Pm) of the mandible through the third plane (PL3). The mandibular inflection point (Pm) is for setting the position of the mandible.

[0106] Then, the occlusal plane definition unit (400) sets a fourth plane (PL4) at a position that divides the distance from the second plane (PL2) to the third plane (PL3) into 1:1.618 while being parallel to the first plane (PL1) in step S360 (a:b=1:1.618).

[0107] Next, the occlusal plane definition unit (400) sets the uppermost point (L1) of the mandibular anterior teeth on the fourth plane (PL4) in step S370.

[0108] Next, the occlusal plane definition unit (400) sets a fifth plane (PL5) perpendicular to the lateral (sagittal plane) image while passing through the uppermost point (L1) of the lower anterior teeth and Xi at step S380.

[0109] Next, the occlusal plane definition unit (400) sets the occlusal plane (OL) at a position parallel to the fifth plane (PL5) and spaced downward from the fifth plane (PL5) by a predetermined distance in step S390. Here, the distance between the fifth plane (PL5) and the occlusal plane (OL) is preferably 2 to 4 mm.

[0110]

[0111] Fig. 13 is a diagram illustrating a computing device according to an embodiment of the present invention. The computing device (TN100) of Fig. 13 may be a device described in this specification, for example, an image processing device (10).

[0112] In the embodiment of FIG. 13, the computing device (TN100) may include at least one processor (TN110), a transceiver (TN120), and a memory (TN130). In addition, the computing device (TN100) may further include a storage device (TN140), an input interface device (TN150), an output interface device (TN160), and the like. The components included in the computing device (TN100) may be connected by a bus (TN170) to communicate with each other.

[0113] The processor (TN110) can execute program commands stored in at least one of the memory (TN130) and the storage device (TN140). The processor (TN110) may refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor in which methods according to embodiments of the present invention are performed. The processor (TN110) may be configured to implement procedures, functions, methods, etc. described in relation to embodiments of the present invention. The processor (TN110) may control each component of the computing device (TN100).

[0114] The memory (TN130) and the storage device (TN140) can each store various information related to the operation of the processor (TN110). The memory (TN130) and the storage device (TN140) can each be configured with at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory (TN130) can be configured with at least one of a read-only memory (ROM) and a random access memory (RAM).

[0115] The transceiver (TN120) can transmit or receive wired or wireless signals. The transceiver (TN120) can be connected to a network to perform communication.

[0116] In particular, the model learning unit (100), data processing unit (200), landmark detection unit (300), occlusal plane definition unit (400), prosthesis design unit (500), and prosthesis placement unit (600) according to an embodiment of the present invention may be implemented in the form of a program readable by a computer means and stored in a memory (TN130) and then executed by a processor (TN110), or may become a sub-module of the processor (TN110).

[0117] Meanwhile, the method according to the embodiment of the present invention described above may be implemented in the form of a program readable by various computer means and recorded on a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, etc., singly or in combination. The program commands recorded on the recording medium may be those specially designed and configured for the present invention, or may be those known and usable by those skilled in the art of computer software. For example, the recording medium includes magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of program commands may include not only machine language wires generated by a compiler, but also high-level language wires that can be executed by a computer using an interpreter, etc. These hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

[0118] While the present invention has been described using several preferred embodiments, these embodiments are illustrative and not limiting. As such, those skilled in the art will appreciate that various changes and modifications can be made in accordance with the doctrine of equivalents without departing from the spirit of the invention and the scope of the claims.

Claims

1. In a method for setting an occlusal plane, A step of a data processing unit obtaining a cross-sectional image including hard tissue images and soft tissue images of a facial area including the patient's teeth through cone beam computed tomography (CBCT); A step of detecting a landmark in the above-described single-layer image using a learned detection model by a landmark detection unit; and A step of defining the occlusal plane based on landmarks detected from the above-mentioned single-layer image; characterized by including A method for establishing an occlusal plane.

2. In paragraph 1, The above landmark is For hard tissue images, Includes points at both ends of the Orbitale, Porion, ANS, A-point, Xi and Mid-palatal suture lines. For soft tissue imaging, characterized by including the pupil, ala and tragus. A method for establishing an occlusal plane.

3. In paragraph 1, The steps for defining the above occlusal plane are The step of deriving four planes including the ANS-Porion plane, the Orbitale plane, the mid-palatal suture plane, and the ala-tragus plane by using the landmarks of the occlusal plane definition part; and The above occlusal plane definition Parallel to the ANS-Porion plane based on the sagittal image among hard tissue images, or parallel to the ala-tragus plane among soft tissue images, Among the hard tissue images, the coronal image is parallel to the orbital plane, Among the hard tissue images, it is perpendicular to the mid-palatal suture plane based on the coronal image, Among the hard tissue images, the sagittal image is spaced downwards by a preset distance from the ANS-Porion plane. Among the soft tissue images, the distance from the ala-tragus plane to the lower part of the sagittal image is the above distance. A step of deriving a plane as an occlusal plane; characterized by including A method for establishing an occlusal plane.

4. In paragraph 1, A step in which the prosthesis design department creates a prosthesis model based on the occlusal plane according to user input; A step of placing the prosthesis model on a registration image that combines the hard tissue image and the soft tissue image by the prosthesis placement unit; and A step of correcting the position of the prosthesis based on the landmark of the soft tissue image by the prosthesis placement unit; characterized by further including A method for establishing an occlusal plane.

5. In paragraph 4, The step of correcting the position of the above prosthesis is A step of setting the inter-pupillary plane by the above prosthesis placement unit; A step of setting a midline plane, which is a plane perpendicular to the inter-pupillary plane while passing through the center of the inter-pupillary line, which is a line on a plane image of the inter-pupillary plane, by the prosthesis placement part; A step of adjusting the position of the prosthesis by aligning the prosthesis placement unit with the midline plane; characterized by including A method for establishing an occlusal plane.

6. In paragraph 1, Before the step of acquiring the above-mentioned single-layer image, A step of loading learning data including a single-layer image and a target image, which is an image in which landmarks are marked on the single-layer image, by a model learning unit; A step in which the model learning unit inputs the above-mentioned single-layer image into the detection model; The above detection model performs multiple operations to which weights that have not yet been learned are applied to the above single-layer image. A step of deriving a marking image, which is an image in which landmarks are marked from the above-mentioned single-layer image; A step in which the above model learning unit derives a loss representing the difference between the target image and the marking image through a loss function; and A step in which the model learning unit performs optimization to update the weights of the detection model so that the loss is minimized; characterized by further including A method for establishing an occlusal plane.

7. In paragraph 6, The above loss function is Mathematical formula And, The above θ is the loss, Above is the target image, Above is the transpose matrix of the target image, Above is a marking image, The above i is characterized as being an index of learning data. A method for establishing an occlusal plane.

8. In paragraph 1, The steps for defining the above occlusal plane are A step of establishing a first plane perpendicular to the lateral image while connecting the Orbitale and Porion by the occlusal plane definition part; A step of setting a second plane that passes through A-point while being parallel to the first plane by defining the occlusal plane; A step for setting a third plane at a position parallel to the first plane and spaced apart from the second plane by a predetermined distance between the first plane and the second plane, wherein the occlusal plane definition portion is; A step of setting a fourth plane at a position that divides the distance from the second plane to the third plane at a predetermined ratio while the occlusal plane definition portion is parallel to the first plane; A step of setting the uppermost point of the mandibular incisor on the fourth plane by the occlusal plane definition part; The step of setting a fifth plane perpendicular to the lateral image while passing through the uppermost point of the mandibular incisor and Xi by the occlusal plane definition part; and A step of setting the occlusal plane at a position spaced downward from the fifth plane while the occlusal plane definition part is parallel to the fifth plane; characterized by including A method for establishing an occlusal plane.

9. In a device for setting the occlusal plane, A data processing unit that obtains a cross-sectional image including hard tissue images and soft tissue images of the facial area including the patient's teeth using cone beam computed tomography (CBCT); A landmark detection unit that detects landmarks in the above-described single-layer image using a learned detection model; and An occlusal plane definition unit that defines an occlusal plane based on landmarks detected in the above-mentioned single-layer image; characterized by including A device for setting the occlusal plane.

10. In paragraph 9, The above landmark is For hard tissue images, Includes points at both ends of the Orbitale, Porion, ANS, A-point, Xi and Mid-palatal suture lines. For soft tissue imaging, characterized by including the pupil, ala and tragus. A device for setting the occlusal plane.

11. In paragraph 9, The above occlusal plane definition part Using the above landmarks, four planes including the ANS-Porion plane, the Orbitale plane, the mid-palatal suture plane, and the ala-tragus plane are derived. Parallel to the ANS-Porion plane based on the sagittal image among hard tissue images, or parallel to the ala-tragus plane among soft tissue images, Among the hard tissue images, the coronal image is parallel to the orbital plane, Among the hard tissue images, it is perpendicular to the mid-palatal suture plane based on the coronal image, Among the hard tissue images, the sagittal image is spaced downwards by a preset distance from the ANS-Porion plane. Among the soft tissue images, the distance from the ala-tragus plane to the lower part of the sagittal image is the above distance. It is characterized by deriving the plane as an occlusal plane. A device for setting the occlusal plane.

12. In paragraph 9, A prosthesis design department that creates a prosthesis model based on the occlusal plane according to user input; and The prosthesis model is placed on the aligned image that combines the above hard tissue image and the above soft tissue image, A prosthesis placement unit that corrects the position of the prosthesis based on the landmark of the soft tissue image; characterized by further including A device for setting the occlusal plane.

13. In paragraph 12, The above prosthesis placement part Establish the inter-pupillary plane, A midline plane is set as a plane perpendicular to the inter-pupillary plane while passing through the center of the inter-pupillary line, which is a line on the plane image of the inter-pupillary plane, and characterized by adjusting the position of the prosthesis to the above midline plane. A device for setting the occlusal plane.

14. In paragraph 9, Load training data including a cross-sectional image and a target image, which is an image with landmarks marked on the cross-sectional image, Input the above single-layer image into the detection model, If the above detection model performs multiple operations to which weights that have not yet been learned are applied to the above-mentioned single-layer image, and derives a marking image, which is an image in which landmarks are marked in the above-mentioned single-layer image, A loss function is used to derive a loss that represents the difference between the target image and the marking image. A model learning unit that performs optimization to update the weights of the detection model so that the loss is minimized; characterized by further including A device for setting the occlusal plane.

15. In paragraph 14, The above loss function is Mathematical formula And, The above θ is the loss, Above is the target image, Above is the transpose matrix of the target image, Above is a marking image, The above i is characterized as being an index of learning data. A device for setting the occlusal plane.

16. In paragraph 9, The above occlusal plane definition part Connecting the Orbitale and Porion, we establish a first plane perpendicular to the side image, A second plane is set parallel to the first plane and passes through point A, A third plane is set at a position parallel to the first plane and spaced apart from the second plane by a predetermined distance from the first plane and the second plane, A fourth plane is set at a position that divides the distance from the second plane to the third plane at a predetermined ratio while being parallel to the first plane, Set the uppermost point of the mandibular incisor on the above fourth plane, A fifth plane is set perpendicular to the lateral image, passing through the uppermost point of the mandibular incisor and Xi, It is characterized by setting the occlusal plane at a position parallel to the fifth plane and spaced downward from the fifth plane at a predetermined distance. A device for setting the occlusal plane.

Images