Method for generating three-dimensional model for predicting soft tissue of maxillofacial region after occlusal change

Abstract

This invention provides a method for generating a three-dimensional model for predicting soft tissue morphology of the maxillofacial region after changes in oral occlusion. It relates to the field of three-dimensional modeling and includes: obtaining a postoperative jawbone model by acquiring preoperative three-dimensional data and an occlusion adjustment plan; performing rigid body transformation on the jawbone mesh to obtain a postoperative jawbone model; dividing the soft tissue into layers and calculating a thickness distribution map to determine the deformation transfer coefficient of each layer; calculating the deformation of each layer based on the differences in jawbone position and the transfer coefficient; applying muscle tension and gravity compensation to the chin soft tissue deformation to generate a correction amount; and finally reconstructing and rendering the postoperative soft tissue prediction model based on the deformation of each layer and the chin correction amount, achieving a visual effect that accurately predicts the soft tissue morphology after occlusion adjustment.

Description

Technical Field

[0001] This invention relates to three-dimensional modeling technology, and more particularly to a method for generating three-dimensional models for predicting soft tissue changes in the maxillofacial region after changes in oral occlusion. Background Technology

[0002] With the development of digital oral diagnosis and treatment and orthodontic technology, preoperative three-dimensional scanning of the maxillofacial region and occlusal adjustment plans have become an important foundation for clinical treatment. However, existing technologies still suffer from insufficient accuracy and limited visualization effects in predicting maxillofacial soft tissues after occlusal changes. On the one hand, traditional methods rely heavily on simple empirical models or two-dimensional images, making it difficult to accurately reflect the complex three-dimensional deformation effects of jawbone displacement on soft tissues. On the other hand, existing three-dimensional modeling technologies lack effective calculation and compensation for soft tissue thickness distribution, interlayer deformation transmission, and the effects of chin muscle tension and gravity, resulting in inaccurate postoperative soft tissue predictions and making it difficult to provide reliable references for clinical surgical plans. In addition, existing methods are inefficient in handling interference areas, jawbone rigid body transformation, and mesh reconstruction, failing to meet the need for rapid and accurate three-dimensional soft tissue prediction.

[0003] Therefore, there is an urgent need for a three-dimensional modeling method that can combine preoperative maxillofacial three-dimensional scanning data and occlusal adjustment plan, and achieve high-precision and visualized postoperative maxillofacial soft tissue prediction through layered soft tissue modeling, deformation transfer calculation and chin compensation. Summary of the Invention

[0004] This invention provides a method for generating a three-dimensional model for predicting soft tissue changes in the maxillofacial region after oral occlusion, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for generating a three-dimensional model for predicting soft tissue changes in the maxillofacial region after oral occlusion alteration, comprising:

[0006] Acquire preoperative three-dimensional scanning data of the maxillofacial region and occlusal position adjustment scheme data. Based on the occlusal position adjustment scheme data, perform rigid body space transformation and coordinate reconstruction on the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scanning data of the maxillofacial region to obtain a postoperative three-dimensional model of the jawbone.

[0007] The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region is divided into multiple layers by voxel segmentation in the depth direction. The soft tissue thickness distribution map is calculated by image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue is determined based on the soft tissue thickness distribution map.

[0008] Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated.

[0009] For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin correction deformation variables. Based on the deformation variables of each layer of the multi-layered soft tissue and the chin correction deformation variables, a postoperative three-dimensional model of the maxillofacial soft tissue is reconstructed and rendered.

[0010] Based on the occlusal position adjustment scheme data, the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scan data of the maxillofacial region is subjected to rigid body space transformation and coordinate reconstruction to obtain the postoperative three-dimensional model of the jawbone, including:

[0011] Extract the maxillary and mandibular rotation parameters from the occlusal position adjustment scheme data;

[0012] Based on the maxillary bone rotation parameters and the mandibular bone rotation parameters, construct the maxillary transformation matrix and the mandibular transformation matrix respectively. Apply the maxillary transformation matrix and the mandibular transformation matrix to the vertex coordinates of the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region, and perform rigid body space transformation on the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region to generate the jaw mesh after preliminary transformation.

[0013] The minimum distance distribution between the maxillary and mandibular vertices in the initially transformed jawbone mesh is calculated. Interference regions where the minimum distance is less than a preset distance threshold are identified. The position of the interference vertices is offset along the vertex normal vector direction within the interference regions. The coordinates of the initially transformed jawbone mesh are reconstructed to generate the postoperative three-dimensional jawbone model.

[0014] The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region was segmented into multiple layers using voxels along the depth direction. A soft tissue thickness distribution map was calculated through image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue layers was determined based on the soft tissue thickness distribution map, including:

[0015] Three-dimensional voxel data of the soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. A depth direction vector field is constructed from the surface of the soft tissue region to the bottom of the soft tissue region. Based on the depth direction vector field, the three-dimensional voxel data of the soft tissue region is divided into voxel layers according to a preset layer spacing, and the soft tissue is divided into multiple layers.

[0016] The vector field along the depth direction is traced from the surface of the soft tissue region to the bottom of the soft tissue region. The length of the tracing path is calculated as the thickness value. A spatial thickness mapping relationship between the spatial coordinates of the soft tissue region surface and the thickness value is established. The soft tissue region surface is projected onto a two-dimensional plane. According to the spatial thickness mapping relationship, the thickness value is assigned to the corresponding spatial coordinate position on the projection plane to generate a soft tissue thickness distribution map.

[0017] Based on the spatial thickness mapping relationship, the thickness values ​​corresponding to each layer in the multi-layer soft tissue are extracted from the soft tissue thickness distribution map. The depth values ​​from each layer in the multi-layer soft tissue to the surface of the soft tissue region are calculated. The remaining thickness value is calculated based on the thickness value and the depth value. The ratio of the remaining thickness value to the thickness value is used as the deformation transfer coefficient of each layer in the multi-layer soft tissue.

[0018] Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated, including:

[0019] Extract the maxillary adjustment vertices and mandibular adjustment vertices from the postoperative three-dimensional model of the jawbone, and extract the mesh vertex coordinates of the maxilla and mandible from the preoperative three-dimensional scan data of the maxillofacial region;

[0020] Calculate the maxillary position offset between the maxillary adjustment vertex and the grid vertex coordinates of the maxilla, and the mandibular position offset between the mandibular adjustment vertex and the grid vertex coordinates of the mandible;

[0021] A jawbone position difference field is constructed based on the maxillary position offset and the mandibular position offset;

[0022] The coordinates of the center of each layer of soft tissue in the multi-layered soft tissue are projected onto the surface vertex coordinates of the jawbone structure data in the preoperative three-dimensional scan data of the maxillofacial region to obtain the jawbone reference point coordinates corresponding to each layer of soft tissue.

[0023] Based on the coordinates of the jawbone reference points corresponding to each layer of soft tissue, spatial interpolation calculations are performed in the jawbone position difference field to obtain the jawbone displacement vector corresponding to each layer of soft tissue.

[0024] The deformation transfer coefficient of each layer in the multi-layered soft tissue is weighted and calculated with the jawbone displacement vector corresponding to each layer of soft tissue to obtain the deformation of each layer in the multi-layered soft tissue.

[0025] For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin-corrected deformation variables, including:

[0026] The boundary coordinates of the chin soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. Based on the boundary coordinates, the deformations belonging to the chin soft tissue region are selected from the deformations of each layer in the multi-layer soft tissue.

[0027] Extract the coordinates of the muscle attachment points within the soft tissue region of the chin, calculate the spatial relationship between the coordinates of the muscle attachment points and the boundary coordinates of the soft tissue region of the chin, and determine the range of influence of muscle tension based on the spatial relationship;

[0028] The degree of muscle stretching is calculated based on the deformation of soft tissue within the range of muscle tension influence, and a muscle tension compensation vector is generated based on the degree of stretching.

[0029] The gravity direction component is calculated based on the center coordinates of each layer of soft tissue in the chin soft tissue region. The degree of tissue sagging is calculated based on the gravity direction component and the deformation of each layer of soft tissue in the chin soft tissue region. A gravity compensation vector is generated based on the degree of tissue sagging.

[0030] The muscle tension compensation vector and the gravity compensation vector are superimposed on the deformation of the soft tissue region of the chin to generate the chin correction deformation.

[0031] Based on the deformation variables of each layer in the multi-layered soft tissue and the chin correction deformation, the postoperative maxillofacial soft tissue prediction 3D model is reconstructed and rendered, including:

[0032] The boundary coordinates of the chin soft tissue region are extracted from the chin correction deformation. A transition region is constructed between the deformation of each layer in the multi-layer soft tissue and the chin correction deformation based on the boundary coordinates. The deformation of each layer in the multi-layer soft tissue and the chin correction deformation are spatially weighted and fused within the transition region to generate the fused multi-layer soft tissue deformation.

[0033] Extract the vertex coordinates of the surface mesh of the soft tissue region from the preoperative three-dimensional scan data of the maxillofacial region, and superimpose the deformation of the corresponding soft tissue region surface position in the fused multi-layer soft tissue deformation variables onto the vertex coordinates of the surface mesh to generate the postoperative soft tissue surface mesh.

[0034] The fused multi-layer soft tissue deformation is spatially correlated with the center coordinates of each layer of soft tissue in the multi-layer soft tissue, and three-dimensional spatial interpolation is performed between the center coordinates of the layers along the depth direction to generate the internal deformation field of the soft tissue.

[0035] The deformation field inside the soft tissue is mapped to the voxel space coordinates inside the postoperative soft tissue surface mesh. The mapped deformation field inside the soft tissue is combined with the postoperative soft tissue surface mesh. The surface mesh normal vector of the combined model is calculated. Lighting calculation and three-dimensional graphics rendering are performed based on the surface mesh normal vector to obtain a predicted three-dimensional model of the postoperative maxillofacial soft tissue.

[0036] The fused multi-layer soft tissue deformation variable is spatially correlated with the layer center coordinates of each layer in the multi-layer soft tissue. Three-dimensional spatial interpolation is then performed between the layer center coordinates along the depth direction to generate the internal deformation field of the soft tissue, including:

[0037] Extract the deformation of each layer of soft tissue from the fused multi-layer soft tissue deformation variables, extract the layer center coordinates of each layer of soft tissue in the multi-layer soft tissue, and establish a spatial binding relationship between the deformation variables and the layer center coordinates;

[0038] The coordinates of the center of adjacent layers are identified along the direction of the depth vector field. An interpolation path is constructed between the coordinates of the center of adjacent layers. Three-dimensional spatial sampling is performed along the interpolation path to generate the coordinates of the interpolation nodes between layers.

[0039] Calculate the spatial distance between the inter-layer interpolation node coordinates and the center coordinates of adjacent layers, calculate the interpolation weight coefficients based on the spatial distance, and perform weighted calculations based on the interpolation weight coefficients and the deformation variables in the spatial binding relationship to obtain the interpolation deformation variables corresponding to the inter-layer interpolation node coordinates;

[0040] By establishing a correspondence between interpolated deformation variables and inter-layer interpolation node coordinates, and combining the spatial binding relationship between deformation variables and layer center coordinates, as well as the correspondence between interpolated deformation variables and inter-layer interpolation node coordinates in three-dimensional space, a deformation field inside the soft tissue is generated.

[0041] A second aspect of the present invention provides an electronic device, comprising:

[0042] processor;

[0043] Memory used to store processor-executable instructions;

[0044] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0045] A third aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0046] In this embodiment, preoperative three-dimensional scanning data of the maxillofacial region can be combined with occlusal adjustment schemes. Rigid body space transformation and interference vertex correction are performed on the three-dimensional mesh of the jawbone to accurately construct the postoperative jawbone model. Through soft tissue voxel layering and thickness distribution analysis, multi-layer soft tissue deformation variables are calculated, and muscle tension and gravity compensation are applied to the chin to achieve accurate prediction of soft tissue deformation. Based on the soft tissue deformation variables of each layer and the chin correction deformation variables, the postoperative three-dimensional model of the maxillofacial soft tissue is reconstructed, achieving high-precision and visualized soft tissue prediction and improving the accuracy of three-dimensional modeling. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating the method for generating a three-dimensional model of maxillofacial soft tissue after changes in oral occlusion, according to an embodiment of the present invention.

[0048] Figure 2 This is a flowchart illustrating the generation process of a three-dimensional model for predicting postoperative maxillofacial soft tissue in an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0051] Figure 1 This is a flowchart illustrating the method for generating a three-dimensional model of maxillofacial soft tissue after changes in oral occlusion, as described in an embodiment of the present invention. Figure 1 As shown, the method includes:

[0052] Acquire preoperative three-dimensional scanning data of the maxillofacial region and occlusal position adjustment scheme data. Based on the occlusal position adjustment scheme data, perform rigid body space transformation and coordinate reconstruction on the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scanning data of the maxillofacial region to obtain a postoperative three-dimensional model of the jawbone.

[0053] The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region is divided into multiple layers by voxel segmentation in the depth direction. The soft tissue thickness distribution map is calculated by image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue is determined based on the soft tissue thickness distribution map.

[0054] Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated.

[0055] For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin correction deformation variables. Based on the deformation variables of each layer of the multi-layered soft tissue and the chin correction deformation variables, a postoperative three-dimensional model of the maxillofacial soft tissue is reconstructed and rendered.

[0056] Based on the occlusal position adjustment scheme data, the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scan data of the maxillofacial region is subjected to rigid body space transformation and coordinate reconstruction to obtain the postoperative three-dimensional model of the jawbone, including:

[0057] Extract the maxillary and mandibular rotation parameters from the occlusal position adjustment scheme data;

[0058] Based on the maxillary bone rotation parameters and the mandibular bone rotation parameters, construct the maxillary transformation matrix and the mandibular transformation matrix respectively. Apply the maxillary transformation matrix and the mandibular transformation matrix to the vertex coordinates of the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region, and perform rigid body space transformation on the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region to generate the jaw mesh after preliminary transformation.

[0059] The minimum distance distribution between the maxillary and mandibular vertices in the initially transformed jawbone mesh is calculated. Interference regions where the minimum distance is less than a preset distance threshold are identified. The position of the interference vertices is offset along the vertex normal vector direction within the interference regions. The coordinates of the initially transformed jawbone mesh are reconstructed to generate the postoperative three-dimensional jawbone model.

[0060] After obtaining preoperative three-dimensional facial scan data and occlusal position adjustment plan data, precise spatial transformation of the jaw structure is required. The occlusal position adjustment plan data typically originates from a surgical plan developed by an orthodontist or maxillofacial surgeon based on a clinical diagnosis. This data includes the spatial movement parameters required for the maxilla and mandible. First, maxillary and mandibular rotation parameters are extracted from the occlusal position adjustment plan data. These rotation parameters include rotation angles around three spatial coordinate axes and corresponding translation vectors. Rotation angles are usually given in Euler angles, including pitch, yaw, and roll angles, while translation vectors describe the displacement of the jawbone in the anterior-posterior, lateral, and superior-inferior directions.

[0061] Transformation matrices are constructed based on the extracted rotation parameters. The maxillary transformation matrix is ​​achieved by combining the rotation matrix and translation vectors into a homogeneous transformation matrix with a dimension of 4×4. The rotation matrix is ​​calculated using Euler angles to obtain a 3×3 matrix, and the translation vector is the first three elements of the fourth column. The mandibular transformation matrix is ​​constructed in the same way. These two transformation matrices are applied to the coordinates of all vertices of the 3D mesh of the maxilla and mandible in the preoperative 3D facial scan data. Specifically, the 3D coordinates of each vertex are expanded into homogeneous coordinates by adding the value 1 to the original coordinates to form a four-dimensional vector, which is then multiplied by the corresponding transformation matrix to obtain the transformed new coordinates. This rigid space transformation preserves the shape and topology of the jawbone mesh, only changing its position and orientation in space, generating the initially transformed jawbone mesh.

[0062] Because occlusal adjustments can lead to overlap or excessive proximity between the maxilla and mandible in space, collision detection and interference elimination are necessary. The minimum distance from each maxillary vertex to the mandibular mesh surface in the initially transformed jawbone mesh is calculated. A spatial indexing structure such as an octree or KD-tree is used to accelerate the distance calculation process. For each maxillary vertex, the nearest triangular facet is quickly located in the mandibular mesh, and the minimum distance value is calculated using the point-to-triangle distance formula, thus obtaining the overall minimum distance distribution. A preset distance threshold of 2 mm is set. When the minimum distance is less than this threshold, the area is determined to be an interference region, indicating unreasonable spatial overlap between the maxilla and mandible.

[0063] For the identified interference regions, coordinate reconstruction is performed to eliminate interference. For each interference vertex within the interference region, the normal vector of that vertex is calculated. The normal vector is obtained by weighted averaging of the normal vectors of the adjacent triangular facets of that vertex. The position of the interference vertex is offset along the direction of the vertex normal vector. The offset distance is determined based on the difference between the current minimum distance and a preset distance threshold, ensuring that the distance between the offset vertex and the contralateral jawbone is greater than the threshold. To maintain the smoothness of the mesh, the coordinates of the offset vertex are smoothed using Laplacian smoothing to allow a natural transition between the interference region and the surrounding non-interference region. After adjusting the positions of all interference vertices, the vertex coordinates and facet connectivity of the jawbone mesh are updated, ultimately generating a postoperative 3D model of the jawbone without interference and with a reasonable spatial position. This model accurately reflects the spatial configuration of the jawbone after occlusion adjustment.

[0064] The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region was segmented into multiple layers using voxels along the depth direction. A soft tissue thickness distribution map was calculated through image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue layers was determined based on the soft tissue thickness distribution map, including:

[0065] Three-dimensional voxel data of the soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. A depth direction vector field is constructed from the surface of the soft tissue region to the bottom of the soft tissue region. Based on the depth direction vector field, the three-dimensional voxel data of the soft tissue region is divided into voxel layers according to a preset layer spacing, and the soft tissue is divided into multiple layers.

[0066] The vector field along the depth direction is traced from the surface of the soft tissue region to the bottom of the soft tissue region. The length of the tracing path is calculated as the thickness value. A spatial thickness mapping relationship between the spatial coordinates of the soft tissue region surface and the thickness value is established. The soft tissue region surface is projected onto a two-dimensional plane. According to the spatial thickness mapping relationship, the thickness value is assigned to the corresponding spatial coordinate position on the projection plane to generate a soft tissue thickness distribution map.

[0067] Based on the spatial thickness mapping relationship, the thickness values ​​corresponding to each layer in the multi-layer soft tissue are extracted from the soft tissue thickness distribution map. The depth values ​​from each layer in the multi-layer soft tissue to the surface of the soft tissue region are calculated. The remaining thickness value is calculated based on the thickness value and the depth value. The ratio of the remaining thickness value to the thickness value is used as the deformation transfer coefficient of each layer in the multi-layer soft tissue.

[0068] When extracting 3D voxel data of soft tissue regions from preoperative maxillofacial 3D scan data, the first step is to convert the obtained point cloud data into a voxelized representation. The voxelization process uses a uniform mesh, with voxel sizes set to 0.5mm × 0.5mm × 0.5mm to ensure the accuracy of subsequent segmentation. After voxelization, a labeling algorithm is used to identify the boundaries between soft tissue and bone. Specifically, the CT or grayscale values ​​of the voxels are calculated, and a threshold range is set to distinguish soft tissue voxels from bone voxels.

[0069] The depth-direction vector field is constructed using gradient descent. Starting with each voxel on the soft tissue surface, a normal vector pointing inwards towards the soft tissue is calculated, initially perpendicular to the surface tangent plane. Through iterative optimization, the normal vector is guided along the shortest path from the soft tissue to the underlying bone surface, forming a smooth and continuous vector field. After the vector field is constructed, voxels are segmented according to a preset interlayer spacing. The preset interlayer spacing is adaptively determined based on the total soft tissue thickness, typically set to one-tenth of the average soft tissue thickness, ensuring the number of layers is between 5 and 15. During the segmentation process, voxels with the same depth value are grouped into the same layer, thus dividing the soft tissue region into a multi-layered structure.

[0070] Calculating soft tissue thickness distribution maps requires establishing a spatial thickness mapping relationship. In practice, each vertex of the soft tissue surface is selected as the starting point for tracking, and the corresponding depth vector is traced step-by-step with a step size of 0.2 mm until the bone surface or the bottom of the soft tissue region is reached. The step distance is accumulated during the tracking process, and the final accumulated value is the thickness value at that location. A mapping relationship is established between the three-dimensional coordinates (x, y, z) of the soft tissue surface and the corresponding thickness value t, and stored as a data structure. To generate a two-dimensional thickness distribution map, the soft tissue surface is projected onto a specific plane using orthogonal projection, typically a plane parallel to the midsagittal plane of the face. After projection, the thickness value is assigned to the corresponding position on the projection plane according to the mapping relationship. Bilinear interpolation is used to fill in pixels that are not directly mapped, ultimately generating a color-coded soft tissue thickness distribution map.

[0071] The deformation transfer coefficient is calculated based on the ratio of remaining thickness to total thickness. For the i-th layer in a multilayer soft tissue, its depth d to the soft tissue surface is... iThe thickness can be obtained by multiplying the layer number by the interlayer spacing. Based on the spatial location of the layer, the corresponding total thickness value t is extracted from the soft tissue thickness distribution map. The remaining thickness value is calculated as td. i This represents the distance of that layer from the bottom surface of the bone. The deformation transfer coefficient is defined as the ratio of the remaining thickness to the total thickness, i.e., (td) i The coefficient (t) reflects the degree to which different layers of soft tissue are affected by changes in bone position. The closer a layer is to the bone, the smaller the deformation transfer coefficient, indicating a larger deformation. Conversely, the closer a layer is to the surface, the larger the deformation transfer coefficient, indicating a gradual decrease in deformation. This coefficient allows for layered control of soft tissue deformation, making predictions more consistent with biomechanical characteristics.

[0072] Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated, including:

[0073] Extract the maxillary adjustment vertices and mandibular adjustment vertices from the postoperative three-dimensional model of the jawbone, and extract the mesh vertex coordinates of the maxilla and mandible from the preoperative three-dimensional scan data of the maxillofacial region;

[0074] Calculate the maxillary position offset between the maxillary adjustment vertex and the grid vertex coordinates of the maxilla, and the mandibular position offset between the mandibular adjustment vertex and the grid vertex coordinates of the mandible;

[0075] A jawbone position difference field is constructed based on the maxillary position offset and the mandibular position offset;

[0076] The coordinates of the center of each layer of soft tissue in the multi-layered soft tissue are projected onto the surface vertex coordinates of the jawbone structure data in the preoperative three-dimensional scan data of the maxillofacial region to obtain the jawbone reference point coordinates corresponding to each layer of soft tissue.

[0077] Based on the coordinates of the jawbone reference points corresponding to each layer of soft tissue, spatial interpolation calculations are performed in the jawbone position difference field to obtain the jawbone displacement vector corresponding to each layer of soft tissue.

[0078] The deformation transfer coefficient of each layer in the multi-layered soft tissue is weighted and calculated with the jawbone displacement vector corresponding to each layer of soft tissue to obtain the deformation of each layer in the multi-layered soft tissue.

[0079] After obtaining the postoperative three-dimensional model of the jawbone and the divided multi-layered soft tissue structure, the spatial changes in the position of the jawbone before and after surgery are compared and analyzed. Combined with the deformation transfer coefficients of each layer of soft tissue, the deformation of the soft tissue is accurately calculated.

[0080] Key vertices involved in positional adjustments were extracted from the postoperative 3D model of the jawbone. These vertices, including maxillary and mandibular adjustment vertices, represent the new positions of the jawbone structures after the implementation of the occlusal position adjustment plan. Simultaneously, the original mesh vertex coordinates of the maxilla and mandible were extracted from preoperative 3D facial scan data as reference benchmarks for positional changes. During extraction, a one-to-one matching process was performed using vertex identifiers to ensure accurate pairing of vertices at the same anatomical location.

[0081] For successfully paired vertices, the three-dimensional spatial offset between the maxillary adjustment vertex and the original maxillary mesh vertex coordinates is calculated. This offset is denoted as the maxillary position offset and represented as a three-dimensional vector containing x, y, and z directional components. The mandibular position offset between the mandibular adjustment vertex and the original mandibular mesh vertex coordinates is calculated using the same method. These two sets of position offsets comprehensively describe the spatial displacement of the jawbone structure during occlusal adjustment.

[0082] Based on the maxillary and mandibular positional offsets, a jawbone positional difference field is constructed. This difference field is a three-dimensional spatial vector field, where each position corresponds to a displacement vector, representing the positional change of the jawbone structure at that position. The difference field is constructed using a three-dimensional spatial interpolation method, expanding the discrete vertex displacement data into a continuous spatial field, enabling the acquisition of displacement information at any position on and around the jawbone surface.

[0083] For each layer of multi-layered soft tissue, the center coordinates of each layer are calculated. These center coordinates are obtained by averaging the coordinates of all voxels within that layer, representing its spatial location. The center coordinates of each layer are then projected along the normal direction onto the surface of the jawbone structure data from the preoperative 3D facial scan, obtaining the vertex coordinates of the projection point. These coordinates are the jawbone reference point coordinates for each soft tissue layer. The projection process uses a nearest neighbor search algorithm to find the jawbone surface vertex closest to the center coordinates of the layers.

[0084] Based on the coordinates of the jawbone reference points corresponding to each layer of soft tissue, spatial interpolation is performed in the jawbone position difference field. Radial basis function interpolation or trilinear interpolation is used. Based on the spatial position of the jawbone reference point coordinates, the displacement vector corresponding to that position is extracted from the difference field and used as the jawbone displacement vector for each soft tissue layer. This displacement vector reflects the positional changes of the jawbone region adjacent to the soft tissue layer.

[0085] The deformation transfer coefficients of each layer are weighted and calculated with the corresponding jawbone displacement vectors. The deformation transfer coefficient, acting as a weighting factor, is multiplied by each component of the jawbone displacement vector to obtain the deformation of that soft tissue layer. A larger deformation transfer coefficient indicates a more significant influence of jawbone displacement on that soft tissue layer, and the calculated deformation is closer to the amount of jawbone displacement; a smaller deformation transfer coefficient results in a correspondingly smaller deformation of that layer. This weighted calculation achieves differentiated transmission of jawbone positional changes to different soft tissue layers, allowing for the acquisition of deformation values ​​for each layer within a multi-layered soft tissue system.

[0086] For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin-corrected deformation variables, including:

[0087] The boundary coordinates of the chin soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. Based on the boundary coordinates, the deformations belonging to the chin soft tissue region are selected from the deformations of each layer in the multi-layer soft tissue.

[0088] Extract the coordinates of the muscle attachment points within the soft tissue region of the chin, calculate the spatial relationship between the coordinates of the muscle attachment points and the boundary coordinates of the soft tissue region of the chin, and determine the range of influence of muscle tension based on the spatial relationship;

[0089] The degree of muscle stretching is calculated based on the deformation of soft tissue within the range of muscle tension influence, and a muscle tension compensation vector is generated based on the degree of stretching.

[0090] The gravity direction component is calculated based on the center coordinates of each layer of soft tissue in the chin soft tissue region. The degree of tissue sagging is calculated based on the gravity direction component and the deformation of each layer of soft tissue in the chin soft tissue region. A gravity compensation vector is generated based on the degree of tissue sagging.

[0091] The muscle tension compensation vector and the gravity compensation vector are superimposed on the deformation of the soft tissue region of the chin to generate the chin correction deformation.

[0092] After acquiring preoperative three-dimensional facial scan data, the boundary coordinates of the chin soft tissue region were extracted from this data. During the extraction process, the chin region was determined using anatomical landmark localization methods. Specifically, the anterior chin point, submental point, and the line connecting the two mental foramina were selected as boundary references, and the x, y, and z coordinate values ​​of these boundary points were marked in a three-dimensional spatial coordinate system. Based on the marked boundary coordinates, a closed polygonal boundary of the chin region was constructed. The deformation data of each layer of soft tissue was traversed, and it was determined whether the spatial coordinates of each deformation data point were located within the polygonal boundary. The deformation variables located within the boundary were selected as the deformation variables of the chin soft tissue region.

[0093] For the soft tissue region of the chin, the coordinates of muscle attachment points are extracted. The attachment locations of the geniohyoid, geniohyoid, and depressor labii inferioris muscles in the chin of the mandible are primarily identified. Muscle texture features are located in the 3D scan data using image segmentation algorithms to determine the 3D coordinates of each muscle attachment point. The Euclidean distance between these muscle attachment points and the chin boundary coordinates is calculated. Regions with a distance less than a preset threshold around the attachment point are marked as directly affected areas, while regions with a distance between the threshold and twice the threshold are marked as indirectly affected areas. Both constitute the range of influence of muscle tension. This range of influence is determined using a distance decay function, ensuring that locations farther from the attachment point are less affected by muscle tension.

[0094] Within the range of muscle tension influence, deformation data of each grid vertex is read, and the distance change between the muscle attachment point and the surrounding soft tissue vertices before and after deformation is calculated. When the position of the jawbone changes postoperatively, causing the attachment point to move, the angle and modulus difference between the attachment point movement vector and the soft tissue deformation vector are calculated, and the degree of muscle stretching is obtained through the tensile strain calculation formula. Based on the degree of stretching, a muscle tension compensation vector is generated along the muscle fiber direction. This vector points in the direction of muscle contraction, and its magnitude is proportional to the degree of stretching. For areas where stretching exceeds a threshold, the compensation coefficient is increased to reflect the muscle elastic recoil effect.

[0095] For each layer of soft tissue within the chin region, the center coordinates of each layer are read; these center coordinates represent the centroid of all vertices within that layer. With the direction of gravity as the negative z-axis, the projected component of the deformation of each layer in the direction of gravity is calculated. When the jawbone moves anteriorly or posteriorly postoperatively, the chin soft tissue experiences additional downward displacement. By comparing the vertical positional changes of each layer before and after deformation, and combining soft tissue density and thickness parameters, the degree of tissue sagging is calculated. The degree of tissue sagging is related to soft tissue thickness, the vertical component of the deformation, and the distance of the layer from the bone surface; the farther the soft tissue layer is from the bone surface, the greater the sagging. Based on the calculated sagging degree, a gravity compensation vector along the direction of gravity is generated. This vector is used to correct for the additional displacement of the soft tissue under gravity.

[0096] The generated muscle tension compensation vector and gravity compensation vector are superimposed, taking into account their spatial directional differences. For each vertex within the chin soft tissue region, the original deformation of that vertex is added to the corresponding muscle tension compensation vector and gravity compensation vector to obtain the chin correction deformation of that vertex. During the superposition process, a weighting factor is applied to the compensation vector. This weighting factor is dynamically adjusted based on the relative position of the vertex within the influence range to ensure a smooth transition of the compensation effect. After superposition, the chin correction deformation includes the basic deformation caused by changes in jawbone position, as well as the physiological compensation deformation under the combined effects of muscle tension and gravity.

[0097] like Figure 2 As shown, Figure 2 This is a flowchart illustrating the generation process of a three-dimensional model for predicting postoperative maxillofacial soft tissue in an embodiment of the present invention.

[0098] Based on the deformation variables of each layer in the multi-layered soft tissue and the chin correction deformation, the postoperative maxillofacial soft tissue prediction 3D model is reconstructed and rendered, including:

[0099] The boundary coordinates of the chin soft tissue region are extracted from the chin correction deformation. A transition region is constructed between the deformation of each layer in the multi-layer soft tissue and the chin correction deformation based on the boundary coordinates. The deformation of each layer in the multi-layer soft tissue and the chin correction deformation are spatially weighted and fused within the transition region to generate the fused multi-layer soft tissue deformation.

[0100] Extract the vertex coordinates of the surface mesh of the soft tissue region from the preoperative three-dimensional scan data of the maxillofacial region, and superimpose the deformation of the corresponding soft tissue region surface position in the fused multi-layer soft tissue deformation variables onto the vertex coordinates of the surface mesh to generate the postoperative soft tissue surface mesh.

[0101] The fused multi-layer soft tissue deformation is spatially correlated with the center coordinates of each layer of soft tissue in the multi-layer soft tissue, and three-dimensional spatial interpolation is performed between the center coordinates of the layers along the depth direction to generate the internal deformation field of the soft tissue.

[0102] The deformation field inside the soft tissue is mapped to the voxel space coordinates inside the postoperative soft tissue surface mesh. The mapped deformation field inside the soft tissue is combined with the postoperative soft tissue surface mesh. The surface mesh normal vector of the combined model is calculated. Lighting calculation and three-dimensional graphics rendering are performed based on the surface mesh normal vector to obtain a predicted three-dimensional model of the postoperative maxillofacial soft tissue.

[0103] After obtaining the deformation variables of each layer of soft tissue and the chin correction deformation variables, spatial fusion processing is required. First, the three-dimensional boundary of the chin soft tissue region is identified from the chin correction deformation variable data, and the set of three-dimensional coordinate points at the boundary is extracted. This boundary is typically located within approximately 5-15 mm lateral to the lower edge of the mandible, and the spatial position of the boundary points is determined through geometric distance calculation. After determining the boundary, a transition zone with a set width is extended outward from this boundary, typically set to 10-20 mm. Within the transition zone, spatial interpolation fusion is performed on the soft tissue deformation variables without muscle tension compensation and gravity compensation, and the compensated chin correction deformation variables. Specifically, a distance-weighted function is used, calculating the normalized distance from any point in the transition zone to the chin boundary as a weighting coefficient. This coefficient is close to 1 in the central region of the chin and close to 0 at the outer edge of the transition zone. This weighting coefficient is used to linearly weight and combine the two deformation variables to ensure the spatial continuity and smoothness of the deformation field, avoiding surface wrinkles caused by abrupt deformation changes.

[0104] After fusion, the mesh structure of the soft tissue surface in the preoperative three-dimensional scan data of the maxillofacial region is extracted. This mesh consists of triangular facets, with each vertex having a unique three-dimensional coordinate. All surface mesh vertices are traversed, and the deformation vector at the corresponding position in the fused multi-layer soft tissue deformation variables is queried based on the vertex's spatial location. The three components of the obtained deformation vector are then superimposed onto the X, Y, and Z components of the vertex coordinates to update the vertex positions and generate the postoperative soft tissue surface mesh. This surface mesh describes the predicted morphology of the soft tissue outer surface after occlusal adjustment.

[0105] To construct complete deformation information within soft tissue, an internal deformation field needs to be established. The fused multi-layer soft tissue deformation is then correlated one-to-one with the center coordinates of each layer. The center coordinates are typically located at the midpoint of the depth range of each soft tissue layer. Three-dimensional spatial interpolation is performed along the depth direction between the center coordinates of adjacent layers. Cubic spline interpolation or radial basis function interpolation can be used to generate a continuous three-dimensional deformation field. This deformation field describes the deformation distribution from the soft tissue surface to various deep locations.

[0106] The constructed internal deformation field of soft tissue is mapped onto the postoperative soft tissue surface mesh. Specifically, a voxel mesh is created within the three-dimensional space enclosed by the surface mesh, with each voxel having spatial coordinates. The deformation value at the corresponding location in the internal deformation field is retrieved based on the voxel coordinates, completing the mapping process. The mapped internal deformation field data is then combined with the postoperative soft tissue surface mesh to form a complete model containing both surface geometry and internal deformation information.

[0107] To achieve 3D visualization, the normal vector of each triangular facet of the surface mesh is calculated. The normal vector is obtained by the cross product of the two edge vectors formed by the three vertices of the triangular facet, and then normalized. Based on the calculated surface normal vectors, the position of the virtual light source and lighting parameters are set, and lighting calculations are performed using the Phong lighting model or a physically based rendering algorithm. The lighting calculations consider ambient light, diffuse reflection, and specular reflection components, and the brightness of each facet is determined based on the angle between the normal vector and the light source direction. Finally, the calculation results are output as an interactive 3D predictive model of postoperative maxillofacial soft tissue using a 3D graphics rendering engine. This model can be used for preoperative assessment and patient communication.

[0108] The fused multi-layer soft tissue deformation variable is spatially correlated with the layer center coordinates of each layer in the multi-layer soft tissue. Three-dimensional spatial interpolation is then performed between the layer center coordinates along the depth direction to generate the internal deformation field of the soft tissue, including:

[0109] Extract the deformation of each layer of soft tissue from the fused multi-layer soft tissue deformation variables, extract the layer center coordinates of each layer of soft tissue in the multi-layer soft tissue, and establish a spatial binding relationship between the deformation variables and the layer center coordinates;

[0110] The coordinates of the center of adjacent layers are identified along the direction of the depth vector field. An interpolation path is constructed between the coordinates of the center of adjacent layers. Three-dimensional spatial sampling is performed along the interpolation path to generate the coordinates of the interpolation nodes between layers.

[0111] Calculate the spatial distance between the inter-layer interpolation node coordinates and the center coordinates of adjacent layers, calculate the interpolation weight coefficients based on the spatial distance, and perform weighted calculations based on the interpolation weight coefficients and the deformation variables in the spatial binding relationship to obtain the interpolation deformation variables corresponding to the inter-layer interpolation node coordinates;

[0112] By establishing a correspondence between interpolated deformation variables and inter-layer interpolation node coordinates, and combining the spatial binding relationship between deformation variables and layer center coordinates, as well as the correspondence between interpolated deformation variables and inter-layer interpolation node coordinates in three-dimensional space, a deformation field inside the soft tissue is generated.

[0113] After fusing multi-layer soft tissue deformation data, it is necessary to establish a precise correspondence between the deformation variables and their three-dimensional spatial positions. Deformation information is extracted layer by layer from the fused multi-layer soft tissue deformation data structure. Each layer of soft tissue corresponds to a three-dimensional vector-like deformation variable, which describes the displacement components of that layer along the x, y, and z coordinate axes. Simultaneously, the center coordinates of each layer are extracted from the soft tissue layering results. These coordinates represent the representative location of each layer of soft tissue in the depth direction. A spatial binding relationship is established between the extracted deformation variables and the corresponding layer center coordinates, forming a key-value pair data structure to ensure that each layer center coordinate uniquely corresponds to a deformation variable vector.

[0114] Based on the normal vector field along the depth direction of soft tissue, adjacent layer center coordinate pairs are identified along the direction from the surface to the depth. For any adjacent layer center coordinate pair, the connecting vector between them is calculated, which forms the basic direction of the interpolation path. Three-dimensional spatial sampling is performed between adjacent layer center coordinates at equal intervals or according to an adaptive density. The number of sampling points is determined based on the spatial distance between adjacent layers; when the distance between adjacent layers is large, the number of sampling points is increased to ensure interpolation accuracy. The generated sampling points serve as interlayer interpolation nodes, and their three-dimensional coordinates are recorded.

[0115] For each inter-layer interpolation node coordinate, calculate its Euclidean distance to the center coordinates of the two adjacent layers above and below. Let the distance between the interpolation node and the center of the upper layer be d1, and the distance to the center of the lower layer be d2. Calculate the interpolation weight coefficients using the inverse distance weighting method: upper layer weight w1 = d2 / (d1 + d2), lower layer weight w2 = d1 / (d1 + d2), and the sum of the two weights is 1. Obtain the corresponding shape variable vectors of the upper and lower layers from the spatial binding relationship. Multiply the upper layer shape variable by w1, and the lower layer shape variable by w3. Superimpose the two weighted results to obtain the interpolation shape variable corresponding to that interpolation node. Repeat this calculation process for all inter-layer interpolation nodes to obtain the complete set of interpolation shape variables.

[0116] A one-to-one mapping relationship is established between the calculated interpolated deformation variables and the corresponding interlayer interpolation node coordinates. Two types of data are integrated in 3D space: the spatial binding relationship between the original deformation variables and the layer center coordinates serves as discrete control points, while the correspondence between the interpolated deformation variables and the interlayer interpolation node coordinates fills the spatial region between layers. Using a 3D scattered data organization method, all coordinate points and their corresponding deformation vectors are uniformly stored, forming a continuous deformation field data structure covering the entire depth range of the soft tissue. This deformation field records the deformation information at every spatial location within the soft tissue from the surface to the depth, providing fine-grained driving data for subsequent soft tissue mesh deformation.

[0117] A second aspect of the present invention provides an electronic device, comprising:

[0118] processor;

[0119] Memory used to store processor-executable instructions;

[0120] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0121] A third aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0122] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a three-dimensional model to predict soft tissue changes in the maxillofacial region after changes in oral occlusion, characterized in that, include: Acquire preoperative three-dimensional scanning data of the maxillofacial region and occlusal position adjustment scheme data. Based on the occlusal position adjustment scheme data, perform rigid body space transformation and coordinate reconstruction on the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scanning data of the maxillofacial region to obtain a postoperative three-dimensional model of the jawbone. The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region is divided into multiple layers by voxel segmentation in the depth direction. The soft tissue thickness distribution map is calculated by image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue is determined based on the soft tissue thickness distribution map. Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated. For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin correction deformation variables. Based on the deformation variables of each layer of the multi-layered soft tissue and the chin correction deformation variables, a postoperative three-dimensional model of the maxillofacial soft tissue is reconstructed and rendered.

2. The method according to claim 1, characterized in that, Based on the occlusal position adjustment scheme data, the three-dimensional mesh of the jawbone structure in the preoperative three-dimensional scan data of the maxillofacial region is subjected to rigid body space transformation and coordinate reconstruction to obtain the postoperative three-dimensional model of the jawbone, including: Extract the maxillary and mandibular rotation parameters from the occlusal position adjustment scheme data; Based on the maxillary bone rotation parameters and the mandibular bone rotation parameters, construct the maxillary transformation matrix and the mandibular transformation matrix respectively. Apply the maxillary transformation matrix and the mandibular transformation matrix to the vertex coordinates of the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region, and perform rigid body space transformation on the three-dimensional mesh of the jaw structure in the preoperative three-dimensional scan data of the maxillofacial region to generate the jaw mesh after preliminary transformation. The minimum distance distribution between the maxillary and mandibular vertices in the initially transformed jawbone mesh is calculated. Interference regions where the minimum distance is less than a preset distance threshold are identified. The position of the interference vertices is offset along the vertex normal vector direction within the interference regions. The coordinates of the initially transformed jawbone mesh are reconstructed to generate the postoperative three-dimensional jawbone model.

3. The method according to claim 1, characterized in that, The soft tissue in the preoperative three-dimensional scan data of the maxillofacial region was segmented into multiple layers using voxels along the depth direction. A soft tissue thickness distribution map was calculated through image analysis, and the deformation transfer coefficient of each layer in the multiple soft tissue layers was determined based on the soft tissue thickness distribution map, including: Three-dimensional voxel data of the soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. A depth direction vector field is constructed from the surface of the soft tissue region to the bottom of the soft tissue region. Based on the depth direction vector field, the three-dimensional voxel data of the soft tissue region is divided into voxel layers according to a preset layer spacing, and the soft tissue is divided into multiple layers. The vector field along the depth direction is traced from the surface of the soft tissue region to the bottom of the soft tissue region. The length of the tracing path is calculated as the thickness value. A spatial thickness mapping relationship between the spatial coordinates of the soft tissue region surface and the thickness value is established. The soft tissue region surface is projected onto a two-dimensional plane. According to the spatial thickness mapping relationship, the thickness value is assigned to the corresponding spatial coordinate position on the projection plane to generate a soft tissue thickness distribution map. Based on the spatial thickness mapping relationship, the thickness values ​​corresponding to each layer in the multi-layer soft tissue are extracted from the soft tissue thickness distribution map. The depth values ​​from each layer in the multi-layer soft tissue to the surface of the soft tissue region are calculated. The remaining thickness value is calculated based on the thickness value and the depth value. The ratio of the remaining thickness value to the thickness value is used as the deformation transfer coefficient of each layer in the multi-layer soft tissue.

4. The method according to claim 1, characterized in that, Based on the positional differences of the jawbone structure in the postoperative three-dimensional model of the jawbone and the preoperative three-dimensional scan data of the maxillofacial region, and in conjunction with the deformation transfer coefficient, the deformation of each layer in the multi-layered soft tissue is calculated, including: Extract the maxillary adjustment vertices and mandibular adjustment vertices from the postoperative three-dimensional model of the jawbone, and extract the mesh vertex coordinates of the maxilla and mandible from the preoperative three-dimensional scan data of the maxillofacial region; Calculate the maxillary position offset between the maxillary adjustment vertex and the grid vertex coordinates of the maxilla, and the mandibular position offset between the mandibular adjustment vertex and the grid vertex coordinates of the mandible; A jawbone position difference field is constructed based on the maxillary position offset and the mandibular position offset; The coordinates of the center of each layer of soft tissue in the multi-layered soft tissue are projected onto the surface vertex coordinates of the jawbone structure data in the preoperative three-dimensional scan data of the maxillofacial region to obtain the jawbone reference point coordinates corresponding to each layer of soft tissue. Based on the coordinates of the jawbone reference points corresponding to each layer of soft tissue, spatial interpolation is performed in the jawbone position difference field to obtain the jawbone displacement vector corresponding to each layer of soft tissue. The deformation transfer coefficient of each layer in the multi-layered soft tissue is weighted and calculated with the jawbone displacement vector corresponding to each layer of soft tissue to obtain the deformation of each layer in the multi-layered soft tissue.

5. The method according to claim 1, characterized in that, For the deformation variables belonging to the chin soft tissue region in each layer of the multi-layered soft tissue, muscle tension compensation and gravity compensation are applied to generate chin-corrected deformation variables, including: The boundary coordinates of the chin soft tissue region are extracted from the preoperative three-dimensional scan data of the maxillofacial region. Based on the boundary coordinates, the deformations belonging to the chin soft tissue region are selected from the deformations of each layer in the multi-layer soft tissue. Extract the coordinates of the muscle attachment points within the soft tissue region of the chin, calculate the spatial relationship between the coordinates of the muscle attachment points and the boundary coordinates of the soft tissue region of the chin, and determine the range of influence of muscle tension based on the spatial relationship; The degree of muscle stretching is calculated based on the deformation of soft tissue within the range of muscle tension influence, and a muscle tension compensation vector is generated based on the degree of stretching. The gravity direction component is calculated based on the center coordinates of each layer of soft tissue in the chin soft tissue region. The degree of tissue sagging is calculated based on the gravity direction component and the deformation of each layer of soft tissue in the chin soft tissue region. A gravity compensation vector is generated based on the degree of tissue sagging. The muscle tension compensation vector and the gravity compensation vector are superimposed on the deformation of the soft tissue region of the chin to generate the chin correction deformation.

6. The method according to claim 1, characterized in that, Based on the deformation variables of each layer in the multi-layered soft tissue and the chin correction deformation, the postoperative maxillofacial soft tissue prediction 3D model is reconstructed and rendered, including: The boundary coordinates of the chin soft tissue region are extracted from the chin correction deformation. A transition region is constructed between the deformation of each layer in the multi-layer soft tissue and the chin correction deformation based on the boundary coordinates. The deformation of each layer in the multi-layer soft tissue and the chin correction deformation are spatially weighted and fused within the transition region to generate the fused multi-layer soft tissue deformation. Extract the vertex coordinates of the surface mesh of the soft tissue region from the preoperative three-dimensional scan data of the maxillofacial region, and superimpose the deformation of the corresponding soft tissue region surface position in the fused multi-layer soft tissue deformation variables onto the vertex coordinates of the surface mesh to generate the postoperative soft tissue surface mesh. The fused multi-layer soft tissue deformation is spatially correlated with the center coordinates of each layer of soft tissue in the multi-layer soft tissue, and three-dimensional spatial interpolation is performed between the center coordinates of the layers along the depth direction to generate the internal deformation field of the soft tissue. The deformation field inside the soft tissue is mapped to the voxel space coordinates inside the postoperative soft tissue surface mesh. The mapped deformation field inside the soft tissue is combined with the postoperative soft tissue surface mesh. The surface mesh normal vector of the combined model is calculated. Lighting calculation and three-dimensional graphics rendering are performed based on the surface mesh normal vector to obtain a predicted three-dimensional model of the postoperative maxillofacial soft tissue.

7. The method according to claim 6, characterized in that, The fused multi-layer soft tissue deformation variable is spatially correlated with the layer center coordinates of each layer in the multi-layer soft tissue. Three-dimensional spatial interpolation is then performed between the layer center coordinates along the depth direction to generate the internal deformation field of the soft tissue, including: Extract the deformation of each layer of soft tissue from the fused multi-layer soft tissue deformation variables, extract the layer center coordinates of each layer of soft tissue in the multi-layer soft tissue, and establish a spatial binding relationship between the deformation variables and the layer center coordinates; The coordinates of the center of adjacent layers are identified along the direction of the depth vector field. An interpolation path is constructed between the coordinates of the center of adjacent layers. Three-dimensional spatial sampling is performed along the interpolation path to generate the coordinates of the interpolation nodes between layers. Calculate the spatial distance between the inter-layer interpolation node coordinates and the center coordinates of adjacent layers, calculate the interpolation weight coefficients based on the spatial distance, and perform weighted calculations based on the interpolation weight coefficients and the deformation variables in the spatial binding relationship to obtain the interpolation deformation variables corresponding to the inter-layer interpolation node coordinates; By establishing a correspondence between interpolated deformation variables and inter-layer interpolation node coordinates, and combining the spatial binding relationship between deformation variables and layer center coordinates, as well as the correspondence between interpolated deformation variables and inter-layer interpolation node coordinates in three-dimensional space, a deformation field inside the soft tissue is generated.

8. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images