An occlusion algorithm for an intraoral three-dimensional scanning system

Abstract

The present invention discloses a bite algorithm for an intraoral three-dimensional scanning system. The bite algorithm mainly includes determination of bite contact points, an automatic recognition algorithm for bite contact points, and a non-linear bite numerical optimization algorithm. The bite algorithm for the intraoral three-dimensional scanning system obtains more accurate bite contact points and records them through an artificial intelligence matching (AI) algorithm. Moreover, through a non-linear bite numerical optimization algorithm, a corresponding optimization equation is established to improve the global optimization bite accuracy. Additionally, the bite algorithm for the intraoral three-dimensional scanning system combines the artificial intelligence matching algorithm with the non-linear bite numerical optimization algorithm to obtain a more accurate bite relationship between the upper and lower jaw models, helping clinicians improve the bite accuracy during the bite treatment process, reducing the workload while improving the efficiency of oral treatment, and enabling patients to receive better and faster treatment.

Description

Technical Field

[0001] The present invention relates to the field of optoelectronic information technology, and particularly to a bite algorithm for an intraoral three-dimensional scanning system. Background Art

[0002] With the improvement of the living standards of the people in our country and the growth of the population base, the demand for oral health has gradually increased. At the same time, in recent years, the domestic computer-aided design and manufacturing (CAD / CAM) technology has gradually matured to commercial use. In order to pursue a better appearance image, new demands for oral beauty have also been put forward. The digital impression technology has been a research hotspot in recent years, especially in the treatment of oral clinical indications, such as orthodontics, restoration, implantation, etc. The digital impression technology can greatly improve the accuracy and treatment efficiency of oral treatment. The intraoral scanner device is the terminal device for data acquisition in the digital impression process, with characteristics such as non-contact measurement, real restoration of the intraoral three-dimensional shape, and good experience, so it has become the main device for digital treatment such as orthodontics, restoration, and implantation in the current oral industry. At the same time, due to the complex clinical oral environment, this technology also faces many problems in solving intraoral optical imaging, scanning data accuracy, patient experience, etc.

[0003] The intraoral three-dimensional digital technology is a technology that integrates optoelectronic devices and visual algorithms. It projects onto the tooth surface through optoelectronic devices, collects two-dimensional image data of the tooth surface with the help of sensing devices, and performs three-dimensional modeling on the intraoral tooth data through visual algorithms to meet the needs of clinical dental design. Among them, for the 3D model data obtained by the intraoral digital scanner, the occlusal relationship between the upper and lower jaws is the key to the success of the treatment plan design. However, due to the complexity of the actual measurement environment (problems such as occlusal looseness and severe tooth loss), the occlusal relationship may be inaccurate and subsequent processing design cannot be carried out. Therefore, whether the correct occlusal relationship between the upper and lower jaws can be obtained affects the success rate of the entire oral digital process and has also become a difficulty in current oral digital treatment. Therefore, in view of the above problems, this application will provide a bite algorithm for an intraoral three-dimensional scanning system to solve them. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] Aiming at the deficiencies of the prior art, the present invention provides a bite algorithm for an intraoral three-dimensional scanning system, which solves the problems raised in the above background art.

[0006] (2) Technical Solutions

[0007] 1. To achieve the above object, the present invention provides the following technical solutions: A bite algorithm for an intraoral three-dimensional scanning system. The bite algorithm mainly includes bite contact point determination, an automatic bite contact point recognition algorithm, and a non-linear bite numerical optimization algorithm. The specific operations are as follows:

[0008] The first step: Bite contact point determination

[0009] (1) Use an oral scanning device to obtain the upper and lower jaw models of the dental arch, and identify the accurate bite relationship established between the upper and lower jaws;

[0010] (2) By observing the bite relationship between the upper and lower jaws during the biting process, and at the same time using an acquisition imaging device to obtain 2D images of the dental arch, thereby determining the position of the bite contact point and making a mark;

[0011] The second step: Automatic bite contact point recognition algorithm

[0012] (1) Model learning: Manually annotate the bite contact points of the model through a large amount of clinical bite data, and then input them into a general AI learning algorithm library. Utilize the powerful experience prediction ability of the AI algorithm to complete the learning of the bite contact points by the model;

[0013] (2) Identify contact points: Identify the bite contact points of the upper and lower jaw 3D models through the model trained in the above model learning step;

[0014] (3) Intelligent matching: Intelligently match the bite contact points obtained in the above identify contact points step to complete the intelligent matching process of the bite contact points;

[0015] The third step: Non-linear bite numerical optimization algorithm

[0016] (1) Selection of the optimization solver: This algorithm uses the Newton algorithm to solve the optimization problem. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations;

[0017] (2) Establish the solution equation: The solution equations of this algorithm are mainly divided into two types, the bite equation and the upper and lower jaw equation. The definition of the loss function is as follows:

[0018]

[0019] Where Merit is the overall loss function value of the current optimization model, merit i is the loss function value of each equation in the current optimization model, and n is the number of equations.

[0020] The fourth step,

[0021] The obtained occlusal contact points are substituted into a globally optimized algorithm to establish a solver and a solution equation, and then an optimal occlusal transformation matrix RT is obtained. Then, it acts on the maxillomandibular models established at the time when the occlusal contact points are determined in the first step to obtain the best occlusal relationship. If the occlusal accuracy is not satisfied, return to re-obtain the occlusal contact points and continue to perform global optimization to obtain the best occlusal relationship.

[0022] Preferably, in the occlusal contact point determination step of the first step, multiple determined occlusal contact points can be respectively marked as A1 - A2, B1 - B2, C1 - C2 to form three corresponding groups of occlusal contact points.

[0023] Preferably, the occlusal contact point automatic recognition algorithm in the second step is based on an artificial intelligence matching algorithm, and the artificial intelligence matching algorithm includes existing open-source deep learning algorithm libraries, such as 3D data processing models PointNet++, KCNet, SO-Net, PointCNN, A-CNN, and PointConv.

[0024] Preferably, for the non-linear occlusal numerical optimization algorithm in the third step, it solves the optimization problem through algorithms such as the quasi-Newton algorithm. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations.

[0025] Preferably, for the non-linear occlusal numerical optimization algorithm in the third step, the occlusal equation in the step of establishing the solution equation optimizes the relationship between the occlusal data and the upper and lower jaws, while the upper and lower jaw equations in the step of establishing the solution equation optimize the relationship between the upper and lower jaws.

[0026] Preferably, the general occlusal algorithm of the present application can refer to currently commercialized intraoral scanning products, such as the currently used intraoral scanners.

[0027] (III)Advantages

[0028] The present invention provides an occlusal algorithm for an intraoral three-dimensional scanning system, having the following advantages:

[0029] (1) For the occlusal algorithm for the intraoral three-dimensional scanning system, by using an artificial intelligence matching (AI) algorithm, more accurate occlusal contact points are obtained and recorded, and through a non-linear occlusal numerical optimization algorithm, a corresponding optimization equation is established to improve the global optimization occlusal accuracy.

[0030] (2) For the occlusal algorithm for the intraoral three-dimensional scanning system, by combining the artificial intelligence matching (AI) algorithm with the non-linear occlusal numerical optimization algorithm, a more accurate occlusal relationship between the upper and lower jaw models is obtained, which helps clinicians improve the occlusal accuracy during the occlusal treatment process and improves the efficiency of oral treatment. Brief Description of the Drawings

[0031] Figure 1 It is a three-dimensional structural schematic diagram of the upper and lower dental arches 3D model of the present invention;

[0032] Figure 2 It is a schematic diagram of the principle flow of the present invention. Detailed Description of the Invention

[0033] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0034] The present invention provides a technical solution: a bite algorithm for an intraoral three-dimensional scanning system. The bite algorithm mainly includes determination of bite contact points, automatic recognition algorithm for bite contact points, and non-linear bite numerical optimization algorithm. The specific operations are as follows:

[0035] The first step: Determination of bite contact points

[0036] (1) Use an intraoral scanning device to obtain the upper and lower jaw models of the dental arch, and identify the accurate bite relationship established between the upper and lower jaws;

[0037] (2) By observing the bite relationship between the upper and lower jaws during the biting process, and at the same time using an imaging acquisition device to obtain 2D images of the dental arch, thereby determining the positions of the bite contact points and making marks. Multiple determined bite contact points can be respectively marked as A1 - A2, B1 - B2, C1 - C2, forming three groups of corresponding bite contact point groups for subsequent recording and identification. The general bite algorithm of this application can refer to currently commercialized intraoral scanning products, such as the currently used intraoral scanner.

[0038] The second step: Automatic recognition algorithm for bite contact points

[0039] (1) Model learning: Manually annotate the bite contact points of the model through a large amount of clinical bite data, and then input it into a general AI learning algorithm library. Utilize the powerful experience prediction ability of the AI algorithm to complete the learning of the bite contact points by the model. The automatic recognition algorithm for bite contact points is based on an artificial intelligence matching algorithm, and the artificial intelligence matching algorithm includes existing open-source deep learning algorithm libraries, such as 3D data processing models PointNet++, KCNet, SO-Net, PointCNN, A-CNN, and PointConv, with multiple options;

[0040] (2) Identify contact points: Through the model trained in the above model learning step, identify the bite contact points of the upper and lower jaw 3D models;

[0041] (3); Intelligent matching: intelligently match the occlusal contact points obtained in the above-mentioned step of identifying contact points to complete the intelligent matching process of occlusal contact points;

[0042] Step 3: Nonlinear occlusal numerical optimization algorithm

[0043] (1) Selection of the optimization solver: This algorithm uses the Newton algorithm to solve the optimization problem. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations;

[0044] This algorithm solves the optimization problem through the quasi-Newton algorithm, etc. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations;

[0045] (2) Establishment of the solution equation: The solution equations of this algorithm are mainly divided into two types, the occlusal equation and the maxillomandibular equation. The occlusal equation optimizes the relationship between the occlusal data and the upper and lower jaws, while the maxillomandibular equation optimizes the relationship between the upper and lower jaws. The definition of the loss function is as follows:

[0046]

[0047] where Merit is the overall loss function value of the current optimization model, and merit i is the loss function value of each equation in the current optimization model, and n is the number of equations.

[0048] Step 4

[0049] Substitute the obtained occlusal contact points into the globally optimized algorithm to establish a solver and a solution equation, and then obtain the optimal occlusal transformation matrix RT, and then apply it to the maxillomandibular model established at the time of determining the occlusal contact points in the first step to obtain the best occlusal relationship. If the occlusal accuracy is not satisfied, return to re-obtain the occlusal contact points and continue to perform global optimization to obtain the best occlusal relationship.

[0050] In summary, the occlusal algorithm for the intraoral three-dimensional scanning system selects occlusal contact points and performs global occlusal optimization. Substitute the obtained occlusal contact points into the globally optimized algorithm to establish a solver and a solution equation to obtain the optimal occlusal transformation matrix RT, and finally apply it to the maxillomandibular model to obtain the best occlusal relationship. If the occlusal accuracy is not satisfied, return to re-obtain the occlusal contact points and continue to perform global optimization to obtain the best occlusal relationship.

[0051] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0052] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An occlusion algorithm for an intraoral three-dimensional scanning system, characterized in that: The occlusion algorithm mainly includes the determination of occlusion contact points, the automatic recognition algorithm of occlusion contact points, and the non-linear occlusion numerical optimization algorithm. The specific operations are as follows: The first step: Determination of occlusion contact points (1) Use an intraoral scanner to obtain the upper and lower jaw models of the dental arch, and identify the accurate occlusion relationship established between the upper and lower jaws; (2) By observing the occlusion relationship between the upper and lower jaws during the occlusion process, and at the same time using an imaging device to obtain 2D images of the dental arch, the position of the occlusion contact points can be judged and marked; The second step: Automatic recognition algorithm of occlusion contact points (1) Model learning: Manually label the occlusion contact points of the model through a large amount of clinical occlusion data, and then input them into a general AI learning algorithm library to utilize the powerful experience prediction ability of the AI algorithm to complete the learning of the occlusion contact points by the model; (2) Identify contact points: Identify the occlusion contact points of the upper and lower jaw 3D models through the model trained in the above model learning step; (3) Intelligent matching: Intelligently match the occlusion contact points obtained in the above step of identifying contact points to complete the intelligent matching process of the occlusion contact points; The third step: Non-linear occlusion numerical optimization algorithm (1) Selection of the optimization solver: This algorithm uses the Newton algorithm to solve the optimization problem. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations; (2) Establishment of the solution equation: The solution equations of this algorithm are mainly divided into two types, the occlusion equation and the upper and lower jaw equation. The definition of the loss function is as follows: where Merit is the overall loss function value of the current optimized model, and merit i is the loss function value of each equation in the current optimized model, and n is the number of equations; The fourth step Substitute the obtained occlusion contact points into the global optimization algorithm to establish a solver and a solution equation, and then obtain the optimal occlusion transformation matrix RT, and then apply it to the upper and lower jaw models established at the time of determining the occlusion contact points in the first step to obtain the best occlusion relationship. If the occlusion accuracy is not satisfied, return to obtain the occlusion contact points again and continue to perform global optimization to obtain the best occlusion relationship.

2. The occlusal algorithm for an intraoral three-dimensional scanning system according to claim 1, wherein: In the step of determining the occlusion contact points in the first step, multiple determined occlusion contact points can be respectively marked as A1 - A2, B1 - B2, C1 - C2 to form three groups of corresponding occlusion contact point groups.

3. The occlusal algorithm for an intraoral three-dimensional scanning system according to claim 1, characterized in that: The automatic recognition algorithm of occlusion contact points in the second step is based on the artificial intelligence matching algorithm, and the artificial intelligence matching algorithm includes 3D data processing models PointNet++, KCNet, SO-Net, PointCNN, A-CNN, and PointConv.

4. The occlusal algorithm for an intraoral three-dimensional scanning system according to claim 1, characterized in that: The non-linear occlusion numerical optimization algorithm in the third step solves the optimization problem through the quasi-Newton algorithm. This algorithm optimizes the memory and iteration process of the solver according to the change of the gradient between two consecutive iterations.

5. The occlusal algorithm for an intraoral three-dimensional scanning system according to claim 1, characterized in that: In the non-linear occlusion numerical optimization algorithm in the third step, the occlusion equation in the step of establishing the solution equation optimizes the relationship between the occlusion data and the upper and lower jaws, and the upper and lower jaw equation in the step of establishing the solution equation optimizes the relationship between the upper and lower jaws.

Images