Gingival recession automatic generation method and system based on oral scanning model and denture manufacturing method

Abstract

The present application relates to the technical field of oral medical image processing, and particularly relates to a gingival sinking automatic generation method and system based on an oral scanning model and a false tooth manufacturing method. After data preprocessing, the oral scanning grid model identifies key features and extracts constraint conditions, and simultaneously constructs a regional weight map of gingival deformation and a spatial distance field based on a designed edge line. Then, in a collaborative optimization framework, the gingival grid is synchronously driven to deform in accordance with the laws of biomechanics, and the tooth base is reasonably extended in a root direction. Finally, the deformed model is smoothed, repaired and optimized through post-processing, and a gingival sinking three-dimensional model is finally output. The present application realizes automatic, accurate and efficient sinking of the gingival, reduces the rework rate and clinical adjustment time in the design of a digital oral scanning model, and provides a high-precision data basis for subsequent prosthesis design.

Description

Technical Field

[0001] This invention relates to the field of oral medical image processing technology, and in particular to a method, system, and method for automatically generating gingival depression based on an oral scanning model, as well as a method for fabricating dentures. Background Technology

[0002] Traditional oral scanning methods, such as those using alginate / silicone rubber impressions, suffer from cumbersome procedures, patient discomfort, and easy model wear. While the precision of intraoral scanners has improved, 3D digital models of the oral cavity are gradually replacing traditional methods. However, current digital intraoral scanning model processing still relies on manual editing, especially during gingival depression, which is typically done manually by technicians in CAD software. Mainstream dental CAD software (such as 3Shape Dental System, Exocad, and DentalCAD) generally employs interactive semi-automatic methods for gingival editing in intraoral scanning models. The core workflow of this method is as follows: the designer first manually outlines the estimated gingival margin around the target abutment tooth in the model view; the software then projects the drawn 2D curve onto the 3D gingival surface using a "magnetic" or "nearest point adsorption" algorithm, forming a spatial edge line; the system finally offsets this edge line by a fixed distance (usually 0.3–0.6 mm) along the gingival surface normal or the global Z-axis, thereby generating a visually "depressed" gingival morphology. This process relies entirely on geometric manipulation, lacks biomechanical constraints, and the "missing" areas of the tooth structure must be manually repaired by the designer using filling tools. The entire process takes approximately 3–10 minutes per tooth, and the results are heavily dependent on the operator's experience and skill. This can lead to significant differences in shoulder exposure morphology in the same case handled by different designers, affecting the accuracy of subsequent restoration margin fit. Therefore, there is an urgent need for an automated gingival retraction method to solve the problems of low efficiency and poor accuracy. Summary of the Invention

[0003] To address the problems of time-consuming processing of gingival depression using existing digital intraoral scanning models, significant influence from operator experience, and difficulty in ensuring the accuracy of restoration margins, this invention provides an automatic gingival depression generation method, system, and denture fabrication method based on intraoral scanning models. Through feature recognition, constraint-driven processing, and deformation optimization, it achieves automatic, precise, and efficient gingival depression, reducing the rework rate and clinical adjustment time in digital intraoral scanning model design.

[0004] According to the design scheme provided by the present invention, on the one hand, a method for automatically generating gingival depression based on an intraoral scanning model is provided, comprising:

[0005] The oral scan mesh model is acquired and preprocessed to obtain a local model containing the region of interest, in which initial tooth and gingival labels are assigned;

[0006] Based on the local model, key features are identified and gingival region constraints are extracted. A regional weight map and a spatial distance field are constructed according to the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region in the local model to the edge line and to distinguish whether the vertex is located on the gingival side or the tooth side by positive and negative values.

[0007] Based on the regional weight map and spatial distance field, a co-optimized energy function for gingival subsidence and root extension of tooth structure is constructed, and the co-optimized energy function is solved to obtain the final displacement of all vertices after the deformation of the gingival region in the local model.

[0008] The deformed local model is post-processed to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

[0009] As part of the automatic gingival depression generation method based on an oral scanning model of the present invention, the oral scanning mesh model is further preprocessed, including:

[0010] Anisotropic bilateral filtering is used to denoise the mesh in the cross-scan mesh model;

[0011] The target is selected based on a tooth, and the local model is obtained by spatially cropping the oral scanning mesh model according to the tooth size, with the target tooth as the center.

[0012] Assign tooth and gingival labels to each triangular facet of the local model and output the label mask.

[0013] As an automatic gingival depression generation method based on an oral scanning model of the present invention, further comprising identifying key features and extracting gingival region constraints based on a local model, including:

[0014] Based on the tooth and gingival labels in the local model, extract the set of triangular facets of the teeth respectively. Combined with the triangular facets of the gums ;

[0015] Those belonging to the same set of triangular facets and The grid edges are used as the initial gingival margin edges, and the initial gingival margin edges are smoothed by B-spline fitting to obtain the initial gingival margin curve;

[0016] Based on clinical experience and prosthesis design requirements, the initial gingival margin curve is adjusted in three-dimensional space to obtain the interactive design edge line;

[0017] The gingival mass surrounding the target abutment tooth was separated from the set of tooth triangular facets using the region growing method. The tooth contact area was determined by the Euclidean distance from each vertex to the nearest neighbor tooth surface. After removing the tooth contact area from the gingival mass, the free gingival region was determined by the normal plane projection distance from the remaining vertices to the edge line of the interaction design. The tooth contact area is the gingival papilla region, and the free gingival region is the buccal / lingual central region.

[0018] As part of the automatic gingival depression generation method based on an oral scanning model of the present invention, a spatial distance field is further constructed according to key features and gingival region constraints, including:

[0019] Using the edge line of the interaction design as the source, obtain the symbolic distance of each vertex on the model surface. The symbolic distance is used to describe whether the point is located on the gingival side or the tooth side.

[0020] If the vertex belongs to the gingival block and the symbol distance indicates that the vertex is located on the gingival side, then the basic target displacement of the vertex is calculated according to the preset displacement formula. The displacement formula is set according to the maximum sinking amount, the Gaussian decay coefficient of the control deformation influence range, and the displacement direction.

[0021] Otherwise, set the base target displacement of the vertex to 0.

[0022] As part of the automatic gingival depression generation method based on an oral scanning model of this invention, a regional weight map is further constructed according to key features and gingival region constraints, including:

[0023] Deformation weights are assigned to each vertex in the region according to the vertex deformation weight allocation rules to achieve differentiated deformation control in different regions. The vertex deformation weight allocation rules include: if the vertex belongs to the adjacent tooth contact area, the vertex weight is set to 0; if the vertex belongs to the free gingival area, the vertex weight is set to 1; if the vertex belongs to the transition area node, the vertex weight is set according to the smooth interpolation function.

[0024] As an automatic gingival depression generation method based on an oral scanning model of the present invention, a collaboratively optimized energy function for gingival depression and root extension of tooth structure is constructed based on a regional weight map and a spatial distance field, comprising:

[0025] The data terms for the movement of the gingival vertex to the weighted target position are obtained based on the regional weight map and spatial distance field. The mesh is deformed by Laplace constraint and the smoothing term is obtained. Constraint terms are continuously set in position and normal direction based on the gingival deformed boundary and the extended tooth boundary. The tooth extension term is obtained based on the extension amount of the tooth surface vertex along the root direction.

[0026] The data terms, smoothing terms, constraint terms, and tooth extension terms are weighted and fused to obtain the collaborative optimization energy function.

[0027] As part of the automatic gingival depression generation method based on an oral scanning model of this invention, the method further includes solving the collaborative optimization energy function, which includes:

[0028] The collaborative optimization energy function is expressed as a positive definite and sparse quadratic expression with respect to the displacement variables;

[0029] The optimal displacement is obtained by solving the positive definite and sparse quadratic expression using the preconditional conjugate gradient method.

[0030] Furthermore, this invention also provides an automatic gingival depression generation system based on an intraoral scanning model, comprising: a preprocessing module, a feature recognition module, a collaborative deformation module, and a model optimization module, wherein...

[0031] The preprocessing module is used to acquire the oral scan mesh model and perform preprocessing to obtain a local model containing the region of interest, wherein the local model is assigned initial tooth and gingival labels;

[0032] The feature recognition module allows users to identify key features and extract gingival region constraints based on a local model. It then constructs a regional weight map and a spatial distance field based on the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region in the local model to the edge line and distinguishes whether the vertex is located on the gingival side or the tooth side by using positive and negative values.

[0033] The collaborative deformation module is used to construct a collaborative optimization energy function for gingival subsidence and root extension of the tooth based on the regional weight map and spatial distance field, and solve the collaborative optimization energy function to obtain the final displacement of all vertices in the local model after the deformation of the gingival region.

[0034] The model optimization module is used to post-process the deformed local model to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

[0035] In another aspect, the present invention also provides a method for manufacturing dentures, comprising:

[0036] The gingival deformation changes were simulated using the above-mentioned automatic gingival depression generation method to obtain the corresponding digital braces model;

[0037] Dentures are generated and fabricated based on digital dental model.

[0038] The beneficial effects of this invention are:

[0039] 1. This invention achieves automatic, precise, and efficient gingival sinking through feature recognition, constraint-driven, and deformation optimization, greatly improving editing efficiency, ensuring accurate sinking position, avoiding abrupt changes in gingival morphology, and supporting flexible control of sinking amount, region, and smoothness by adjusting the weight parameters in the function.

[0040] 2. To ensure that the deformation of the adjacent tooth contact area is restricted while the free gingival area sinks sufficiently, this invention employs a gingival mesh displacement control method with regional deformation weight constraints. This ensures that the deformation of the adjacent tooth contact area is restricted while the free gingival area sinks sufficiently. To maintain the continuity and integrity of the deformed model, a collaborative optimization function for gingival sinking and root extension of the tooth structure is constructed and solved to achieve the continuity and integrity of the deformed model. This ensures that the gingival sinking morphology is natural and continuous, the shoulder exposure is accurate, and the process is fully automatic, precise, controllable, and conforms to clinical anatomical constraints. This provides a high-precision data foundation for subsequent restoration design, reducing rework rates and clinical adjustment time. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the automatic generation process of gingival depression based on an oral scanning model in the embodiment.

[0042] Figure 2 This is a schematic diagram illustrating the difference between automatic gingival subsidence before and after in the 3D model design software of the embodiment;

[0043] Figure 3 This is a schematic diagram illustrating the principle of the automatic gingival subsidence algorithm in the example. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will be further described in detail below with reference to the accompanying drawings and technical solutions.

[0045] Gingival recession is a common oral health problem, mainly manifested as the gum tissue receding towards the tooth root, exposing the tooth root, which may cause a cottony feeling in the teeth, damage to aesthetics, or even loosening of the teeth. Figure 2 The illustration shows a comparison before and after the automatic gingival retraction. Existing interactive methods for digital oral models rely on manual drawing and pure geometric offset, ignoring the biomechanical properties of the gingiva, resulting in deformation distortion and low efficiency. They lack regional constraints, often producing unnatural shapes and boundary gaps. The entire process is time-consuming, subjective, and the results are uncontrollable, making it difficult to meet the clinical needs of precise digital restoration. Therefore, this invention provides an embodiment, see [link to embodiment]. Figure 1 As shown, an automatic method for generating gingival depression based on an intraoral scanning model is provided, comprising:

[0046] S101. Obtain the oral scan mesh model and perform preprocessing to obtain a local model containing the region of interest. The local model is assigned initial tooth and gingival labels.

[0047] Specifically, the preprocessing of the scanned mesh model can be designed to include:

[0048] Anisotropic bilateral filtering is used to denoise the mesh in the cross-scan mesh model;

[0049] The target is selected based on a tooth, and the local model is obtained by spatially cropping the oral scanning mesh model according to the tooth size, with the target tooth as the center.

[0050] Assign tooth and gingival labels to each triangular facet of the local model and output the label mask.

[0051] like Figure 3 The algorithm shown starts with a high-precision intraoral scan of a 3D mesh model. After data preprocessing, key features are identified and constraints are extracted. At the same time, a regional weight map of gingival deformation and a spatial distance field based on the design edge line are constructed. Subsequently, in a collaborative optimization framework, the gingival mesh is synchronously driven to undergo biomechanical deformation, and the tooth matrix is ​​reasonably extended radicularly. Finally, a series of post-processing steps are used to smooth, repair, and optimize the deformed model, ultimately outputting a 3D model of gingival subsidence that clearly exposes the prepared body shoulder and can be directly used for crown design.

[0052] Among them, the original three-dimensional triangular mesh model obtained by the intraoral scanner It includes the teeth, gums, and soft tissues on the lingual and buccal sides.

[0053] First, mesh denoising and smoothing are performed using anisotropic bilateral filtering to remove scan noise while preserving feature edges. For vertices... Its location has been updated to:

[0054]

[0055] in, It is the vertex The single-ring neighborhood, It is the vertex normal. and These are Gaussian weighting functions based on spatial distance and normal difference, respectively. This is the iteration step size.

[0056] Secondly, the region of interest is extracted by selecting a target abutment tooth. The system then uses this abutment tooth as the center and, based on the average tooth size, performs [analysis / analysis]. Spatial clipping is performed to obtain a local model. This reduces the amount of subsequent calculations.

[0057] Then, the initial label settings are... Each triangular facet is assigned a label: tooth, gum, and the output is a label mask L.

[0058] S102. Identify key features based on the local model and extract gingival region constraints. Construct a regional weight map and a spatial distance field based on the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region to the edge line in the local model and distinguish whether the vertex is located on the gingival side or the tooth side by positive and negative values.

[0059] Specifically, based on identifying key features and extracting gingival region constraints using a local model, it can be designed to include:

[0060] Based on the tooth and gingival labels in the local model, extract the set of triangular facets of the teeth respectively. Combined with the triangular facets of the gums ;

[0061] Those belonging to the same set of triangular facets and The grid edges are used as the initial gingival margin edges, and the initial gingival margin edges are smoothed by B-spline fitting to obtain the initial gingival margin curve;

[0062] Based on clinical experience and prosthesis design requirements, the initial gingival margin curve is adjusted in three-dimensional space to obtain the interactive design edge line;

[0063] The gingival mass surrounding the target abutment tooth was separated from the set of tooth triangular facets using the region growing method. The tooth contact area was determined by the Euclidean distance from each vertex to the nearest neighbor tooth surface. After removing the tooth contact area from the gingival mass, the free gingival region was determined by the normal plane projection distance from the remaining vertices to the edge line of the interaction design. The tooth contact area is the gingival papilla region, and the free gingival region is the buccal / lingual central region.

[0064] Based on the label mask L, extract the set of all triangular faces labeled as teeth and gums. and Find all those that simultaneously belong to and The grid edges are defined as the initial gingival margin edges. .right The resulting polygonal line is fitted with a B-spline to obtain a smooth initial gingival margin curve. .

[0065] Will As suggested, the recommendations can be made based on clinical experience and prosthesis design requirements. Based on this, perform 3D spatial editing (drag control points) to obtain the final design edge lines. . It is a smooth, closed spatial curve located on the gingival surface and surrounding the neck of the abutment tooth, and is the core driving force for all subsequent deformation calculations.

[0066] Using the regional growth method from Separate the gingival mass surrounding the target abutment tooth. Calculate the Euclidean distance from each vertex v to the surface of the nearest neighbor tooth (non-target abutment tooth). ,like If the vertex belongs to the physical contact constraint region, its deformation should be suppressed; this region is the gingival papilla region. This refers to the tooth contact area. Empirical values ​​based on clinical measurements. Set to 0.3mm.

[0067] exist Remove After the region is defined, calculate the remaining vertices to... The projected distance on the normal plane of the curve is calculated, and the local curvature is analyzed. The distance is... Vertices within a certain range and located on the buccal or lingual convex surface are classified as This area of ​​soft tissue is the main area of ​​deformation during gingival retraction.

[0068] The spatial distance field constructed based on key features and gingival region constraints may include:

[0069] Using the edge line of the interaction design as the source, obtain the symbolic distance of each vertex on the model surface. The symbolic distance is used to describe whether the point is located on the gingival side or the tooth side.

[0070] If the vertex belongs to the gingival block and the symbol distance indicates that the vertex is located on the gingival side, then the basic target displacement of the vertex is calculated according to the preset displacement formula. The displacement formula is set according to the maximum sinking amount, the Gaussian decay coefficient of the control deformation influence range, and the displacement direction.

[0071] Otherwise, set the base target displacement of the vertex to 0.

[0072] Among them, a regional weight map is constructed based on key features and gingival region constraints, including:

[0073] Deformation weights are assigned to each vertex in the region according to the vertex deformation weight allocation rules to achieve differentiated deformation control in different regions. The vertex deformation weight allocation rules include: if the vertex belongs to the adjacent tooth contact area, the vertex weight is set to 0; if the vertex belongs to the free gingival area, the vertex weight is set to 1; if the vertex belongs to the transition area node, the vertex weight is set according to the smooth interpolation function.

[0074] To design edge lines As the source, compute each vertex on the surface of the model. Sign distance The application should be on the gingival side (the side where the target tooth is lowered). On the side of the teeth .

[0075] Define vertices Basic target displacement :

[0076]

[0077] in This is the maximum subsidence amount set by the user, with a default value of 0.3mm; To control deformation Gaussian attenuation coefficient of the affected range. This indicates the direction of displacement.

[0078] To better conform to the natural orientation of the tooth root surface, The direction is determined by Pointing to its corresponding The inner normal direction of the surface of the nearest tooth By query The local normal of the tooth surface is estimated.

[0079] For each vertex Assign a deformation weight ,when This indicates that the vertex belongs to the contact area between adjacent teeth. The gingiva in the gap between the abutment tooth and the adjacent tooth hardly sinks, so it is set to prohibit sinking operations; when This indicates that the vertex belongs to the free gingival region and is fully depressed. When the vertex does not belong to the contact area of ​​the adjacent tooth... It does not belong to the free gingival area. If it is a transition zone node, it will be determined to be a smooth transition node.

[0080]

[0081] in, It is a smooth interpolation function (cubic Hermite interpolation). and It is from the vertex to and The shortest geodesic distance in the region.

[0082] S103. Construct a co-optimized energy function for gingival subsidence and root extension of the tooth based on the regional weight map and spatial distance field, and solve the co-optimized energy function to obtain the final displacement of all vertices after the deformation of the gingival region in the local model.

[0083] Specifically, a collaborative optimization energy function for gingival subsidence and radicular extension of the tooth structure, constructed based on a regional weight map and a spatial distance field, can be designed to include:

[0084] The data terms for the movement of the gingival vertex to the weighted target position are obtained based on the regional weight map and spatial distance field. The mesh is deformed by Laplace constraint and the smoothing term is obtained. Constraint terms are continuously set in position and normal direction based on the gingival deformed boundary and the extended tooth boundary. The tooth extension term is obtained based on the extension amount of the tooth surface vertex along the root direction.

[0085] The data terms, smoothing terms, constraint terms, and tooth extension terms are weighted and fused to obtain the collaborative optimization energy function.

[0086] function It consists of four parts, data items Smoothing terms Constraints and tooth extension Specifically, it is expressed as:

[0087]

[0088] Encourage the gingival apex to move towards the weighted target position, data item Represented as

[0089]

[0090] Use Laplacian smoothing constraints to maintain the uniformity of the mesh after deformation:

[0091]

[0092] in It is the vertex The uniform Laplace coordinates.

[0093]

[0094] Constraints Constraints are divided into fixed constraints and continuity constraints. Fixed constraints ensure that the boundary vertices are far from the target tooth. Do not move Continuity constraints guarantee that... On and near the curve, the boundary after forced gingival deformation and the extended tooth boundary are continuous in position and in their first derivative (normal). This is achieved by adding strong spring constraints to the corresponding vertex pairs.

[0095]

[0096] in, Indicates the line between the gums and teeth. The set of corresponding vertex pairs at the location; These represent the vertices corresponding to the gingival and dental sides, respectively. and Normal at the corresponding vertex; The normal continuity weighting coefficient is represented.

[0097] Tooth extension For the apex of the tooth surface (labeled as tooth), especially The upper part encourages it to follow the calculated root direction. extend.

[0098]

[0099] in: It is the extension quantity, and arrive The vertical distance is positively correlated.

[0100] Solving for the collaborative optimization energy function may include:

[0101] The collaborative optimization energy function is expressed as a positive definite and sparse quadratic expression with respect to the displacement variables;

[0102] The optimal displacement is obtained by solving the positive definite and sparse quadratic expression using the preconditional conjugate gradient method.

[0103] The total energy function Expressed in quadratic form with respect to the displacement variable ΔV, it can be specifically represented as:

[0104]

[0105] Since the system is positive definite and sparse, the preconditional conjugate gradient method is used to solve the sparse linear system. To obtain the optimal displacement, obtain the final displacement of all vertices. Applying displacement .

[0106] S104. Post-process the deformed local model to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

[0107] In regions of severe deformation, triangular facets may be stretched or compressed. An adaptive mesh refinement algorithm based on edge folding, vertex splitting, and edge flipping is used to ensure mesh quality. Adaptive mesh refinement is an intelligent and dynamic discretization strategy, the core of which lies in an automated cycle of "error estimation - dynamic adjustment - information synchronization".

[0108] Despite continuity constraints, after numerical solution, the gingiva and extended tooth structure are... There may still be micrometer-level gaps. The following steps are recommended:

[0109] Search for distances less than a threshold However, gaps between unconnected vertex pairs are eliminated by moving adjacent vertices. A light Laplacian smoothing algorithm is applied to smooth the mesh while preserving the original features well and preventing over-smoothing.

[0110] Finally, the processed local model Seamlessly replace back to the original global model Based on the corresponding positions in the image, generate the final target oral 3D model that includes the effect of gingival retraction. The model clearly exposes the complete shoulder morphology of the preparatory body, which can be directly used by the CAD design module for the restoration.

[0111] Furthermore, based on the above method, this embodiment of the invention also provides an automatic gingival depression generation system based on an intraoral scanning model, comprising: a preprocessing module, a feature recognition module, a collaborative deformation module, and a model optimization module, wherein,

[0112] The preprocessing module is used to acquire the oral scan mesh model and perform preprocessing to obtain a local model containing the region of interest, wherein the local model is assigned initial tooth and gingival labels;

[0113] The feature recognition module allows users to identify key features and extract gingival region constraints based on a local model. It then constructs a regional weight map and a spatial distance field based on the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region in the local model to the edge line and distinguishes whether the vertex is located on the gingival side or the tooth side by using positive and negative values.

[0114] The collaborative deformation module is used to construct a collaborative optimization energy function for gingival subsidence and root extension of the tooth based on the regional weight map and spatial distance field, and solve the collaborative optimization energy function to obtain the final displacement of all vertices in the local model after the deformation of the gingival region.

[0115] The model optimization module is used to post-process the deformed local model to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

[0116] Furthermore, based on the above method, embodiments of the present invention also provide a method for fabricating dentures, comprising:

[0117] The gingival deformation changes were simulated using the above-mentioned automatic gingival depression generation method to obtain the corresponding digital braces model;

[0118] Dentures are generated and fabricated based on digital dental model.

[0119] In the fabrication of dentures, denture materials are selected according to oral cavity type, and the materials are processed into solid crowns through computer cutting or 3D printing technology, thereby improving precision and efficiency through digital technology.

[0120] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0121] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0122] The units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations are not considered to be beyond the scope of this invention.

[0123] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in hardware or as a software functional module. This invention is not limited to any particular combination of hardware and software.

[0124] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for automatically generating gingival depression based on an intraoral scanning model, characterized in that, Include: The oral scan mesh model is acquired and preprocessed to obtain a local model containing the region of interest, in which initial tooth and gingival labels are assigned; Based on the local model, key features are identified and gingival region constraints are extracted. A regional weight map and a spatial distance field are constructed according to the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region in the local model to the edge line and to distinguish whether the vertex is located on the gingival side or the tooth side by positive and negative values. Based on the regional weight map and spatial distance field, a co-optimized energy function for gingival subsidence and root extension of tooth structure is constructed, and the co-optimized energy function is solved to obtain the final displacement of all vertices after the deformation of the gingival region in the local model. The deformed local model is post-processed to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

2. The method for automatically generating gingival depression based on an intraoral scanning model according to claim 1, characterized in that, Preprocessing of the scanned mesh model includes: Anisotropic bilateral filtering is used to denoise the mesh in the cross-scan mesh model; The target is selected based on a tooth, and the local model is obtained by spatially cropping the oral scanning mesh model according to the tooth size, with the target tooth as the center. Assign tooth and gingival labels to each triangular facet of the local model and output the label mask.

3. The method for automatically generating gingival depression based on an oral scanning model according to claim 1, characterized in that, Key features are identified and gingival region constraints are extracted based on a local model, including: Based on the tooth and gingival labels in the local model, extract the set of triangular facets of the teeth respectively. Combined with the triangular facets of the gums ; Those belonging to the same set of triangular facets and The grid edges are used as the initial gingival margin edges. The initial gingival margin edges are smoothed by B-spline fitting to obtain the initial gingival margin curve. Based on clinical experience and prosthesis design requirements, the initial gingival margin curve is adjusted in three-dimensional space to obtain the interactive design edge line; The gingival mass surrounding the target abutment tooth was separated from the set of tooth triangular facets using the region growing method. The tooth contact area was determined by the Euclidean distance from each vertex to the nearest neighbor tooth surface. After removing the tooth contact area from the gingival mass, the free gingival region was determined by the normal plane projection distance from the remaining vertices to the edge line of the interaction design. The tooth contact area is the gingival papilla region, and the free gingival region is the buccal / lingual central region.

4. The method for automatically generating gingival depression based on an intraoral scanning model according to claim 3, characterized in that, A spatial distance field is constructed based on key features and gingival region constraints, including: Using the edge line of the interaction design as the source, obtain the symbolic distance of each vertex on the model surface. The symbolic distance is used to describe whether the point is located on the gingival side or the tooth side. If the vertex belongs to the gingival block and the symbol distance indicates that the vertex is located on the gingival side, then the basic target displacement of the vertex is calculated according to the preset displacement formula. The displacement formula is set according to the maximum sinking amount, the Gaussian decay coefficient of the control deformation influence range, and the displacement direction. Otherwise, set the base target displacement of the vertex to 0.

5. The method for automatically generating gingival depression based on an oral scanning model according to claim 3, characterized in that, A regional weighted map is constructed based on key features and gingival region constraints, including: Deformation weights are assigned to each vertex in the region according to the vertex deformation weight allocation rules to achieve differentiated deformation control in different regions. The vertex deformation weight allocation rules include: if the vertex belongs to the adjacent tooth contact area, the vertex weight is set to 0; if the vertex belongs to the free gingival area, the vertex weight is set to 1; if the vertex belongs to the transition area node, the vertex weight is set according to the smooth interpolation function.

6. The method for automatically generating gingival depression based on an oral scanning model according to claim 1, characterized in that, A collaborative optimization energy function for gingival subsidence and radicular extension of tooth structure is constructed based on regional weight maps and spatial distance fields, including: The data terms for the movement of the gingival vertex to the weighted target position are obtained based on the regional weight map and spatial distance field. The mesh is deformed by Laplace constraint and the smoothing term is obtained. Constraint terms are continuously set in position and normal direction based on the gingival deformed boundary and the extended tooth boundary. The tooth extension term is obtained based on the extension amount of the tooth surface vertex along the root direction. The data terms, smoothing terms, constraint terms, and tooth extension terms are weighted and fused to obtain the collaborative optimization energy function.

7. The method for automatically generating gingival depression based on an oral scanning model according to claim 1 or 6, characterized in that, Solving the co-optimization energy function includes: The collaborative optimization energy function is expressed as a positive definite and sparse quadratic expression with respect to the displacement variables; The optimal displacement is obtained by solving the positive definite and sparse quadratic expression using the preconditional conjugate gradient method.

8. An automatic gingival depression generation system based on an intraoral scanning model, characterized in that, It includes: a preprocessing module, a feature recognition module, a cooperative deformation module, and a model optimization module. The preprocessing module is used to acquire the oral scan mesh model and perform preprocessing to obtain a local model containing the region of interest, wherein the local model is assigned initial tooth and gingival labels; The feature recognition module allows users to identify key features and extract gingival region constraints based on a local model. It then constructs a regional weight map and a spatial distance field based on the key features and gingival region constraints. The key features include the tooth-gingival junction line and the interaction design edge line. The regional weight map is used to assign deformation weights to each vertex of the gingival region in the local model. The spatial distance field is used to describe the distance from the vertex of the gingival region in the local model to the edge line and distinguishes whether the vertex is located on the gingival side or the tooth side by using positive and negative values. The collaborative deformation module is used to construct a collaborative optimization energy function for gingival subsidence and root extension of the tooth based on the regional weight map and spatial distance field, and solve the collaborative optimization energy function to obtain the final displacement of all vertices in the local model after the deformation of the gingival region. The model optimization module is used to post-process the deformed local model to obtain a three-dimensional model of gingival depression. The post-processing includes smoothing of the deformed area, filling of gaps, and model integrity processing.

9. A method for manufacturing dentures, characterized in that, include: The automatic gingival subsidence generation method according to any one of claims 1 to 7 simulates gingival deformation changes to obtain a corresponding digital dental brace model; Dentures are generated and fabricated based on digital dental model.

10. An electronic device, characterized in that, include: At least one processor, and a memory coupled to said at least one processor; The memory stores a computer program that can be executed by the at least one processor to implement the method as described in any one of claims 1 to 7.

Images