Method and system for predicting postoperative facial change model
By constructing local geometric relationships and features, combining bone and facial features, predicting postoperative facial change models, the problem of low prediction accuracy in the prior art is solved, and higher accuracy and fewer artifacts are achieved.
Patent Information
- Application Number
- CN202410944724.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-12
AI Technical Summary
Existing postoperative facial change model prediction algorithms, such as PointNet++, ignore the local geometric relationship of point clouds, resulting in low prediction accuracy and serious facial distortion.
By collecting facial soft tissue models, preoperative bone models and postoperative bone models, local geometric relationships were constructed, facial and bone points were extracted, facial and bone points were combined, and the displacement characteristics of facial points were determined, and the facial displacement characteristics were predicted based on facial displacement characteristics and soft tissue models. Postoperative facial change model, using local geometric features to encode facial topological shapes, establishing an interactive relationship between bone and soft tissue point movement.
The prediction accuracy of the postoperative facial change model is improved, artifacts are reduced, and the accuracy and reliability of prediction are enhanced.
Smart Images

Figure CN118942631A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of prediction of postoperative facial change models, and in particular to a prediction method and system for postoperative facial change models. Background Art
[0002] In some actual field scenarios, patients need to undergo postoperative simulation before surgery. In orthognathic surgery, deep learning technology is used to synthesize the pre- and postoperative facial soft tissue 3D point cloud models and jaw 3D point cloud models of patients who have undergone orthognathic surgery in the past. However, the algorithm obtains point-by-point features through PointNet++, which ignores the local geometric relationship of the point cloud, which can easily cause the predicted face to be distorted, resulting in low prediction accuracy of the postoperative facial change model. Summary of the invention
[0003] Based on this, it is necessary to provide a prediction method and system for a postoperative facial change model that can improve prediction accuracy in response to the above technical problems.
[0004] A method for predicting a postoperative facial change model, comprising:
[0005] During the simulation of the surgery, facial soft tissue models, preoperative bone models, and postoperative bone models were collected;
[0006] Two local geometric relationships are constructed based on the facial soft tissue model, and facial point features are extracted;
[0007] Extract bone point features based on the preoperative bone model, and determine the displacement of the bone points based on the postoperative bone model;
[0008] Determine the displacement characteristics of facial points by combining bone point characteristics, bone point displacement, and facial point characteristics;
[0009] A model for predicting postoperative facial changes based on the displacement characteristics of facial points and facial soft tissue model.
[0010] A prediction system for a postoperative facial change model, wherein the prediction system for a postoperative facial change model is applied to the above-mentioned prediction method for a postoperative facial change model, and the prediction system for a postoperative facial change model comprises:
[0011] An acquisition module, used to acquire facial soft tissue models, preoperative bone models, and postoperative bone models during the simulation of the surgery;
[0012] An extraction module, used to construct two local geometric relationships based on the facial soft tissue model and extract facial point features;
[0013] A first displacement module is used to extract bone point features based on a preoperative bone model and determine the displacement of the bone point based on a postoperative bone model;
[0014] The second displacement module is used to determine the displacement characteristics of the facial points by combining the bone point characteristics, the bone point displacement, and the facial point characteristics;
[0015] The prediction module is used to predict the postoperative facial change model based on the displacement characteristics of facial points and the facial soft tissue model.
[0016] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:
[0017] During the simulation of the surgery, facial soft tissue models, preoperative bone models, and postoperative bone models were collected;
[0018] Two local geometric relationships are constructed based on the facial soft tissue model, and facial point features are extracted;
[0019] Extract bone point features based on the preoperative bone model, and determine the displacement of the bone points based on the postoperative bone model;
[0020] Determine the displacement characteristics of facial points by combining bone point characteristics, bone point displacement, and facial point characteristics;
[0021] A model for predicting postoperative facial changes based on the displacement characteristics of facial points and facial soft tissue model.
[0022] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:
[0023] During the simulation of the surgery, facial soft tissue models, preoperative bone models, and postoperative bone models were collected;
[0024] Two local geometric relationships are constructed based on the facial soft tissue model, and facial point features are extracted;
[0025] Extract bone point features based on the preoperative bone model, and determine the displacement of the bone points based on the postoperative bone model;
[0026] Determine the displacement characteristics of facial points by combining bone point characteristics, bone point displacement, and facial point characteristics;
[0027] A model for predicting postoperative facial changes based on the displacement characteristics of facial points and facial soft tissue model.
[0028] The above-mentioned prediction method and system of the postoperative facial change model construct two local geometric relationships based on the facial soft tissue model. The local geometric features encode the facial topological shape to obtain more effective facial point features, and combine the bone point features, bone point displacement, and facial point features to determine the displacement characteristics of the facial points. The postoperative facial change model is predicted based on the displacement characteristics of the facial points and the facial soft tissue model, so as to better establish the interactive relationship between the movement of bones and the movement of facial soft tissue points, thereby improving the prediction accuracy, and the obtained postoperative facial change model has fewer artifacts. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 is a flow chart of a method for predicting a postoperative facial change model in an embodiment of the present invention;
[0030] Figure 2 is a flow chart of S11 in the method for predicting a postoperative facial change model in an embodiment of the present invention;
[0031] Figure 3 is a flow chart of S12 in the method for predicting a postoperative facial change model in an embodiment of the present invention;
[0032] Figure 4 is a schematic flow chart of S13 in the method for predicting a postoperative facial change model in an embodiment of the present invention;
[0033] Figure 5 is a flow chart of S14 in the method for predicting a postoperative facial change model in an embodiment of the present invention;
[0034] Figure 6 is a flow chart of S15 in the method for predicting a postoperative facial change model in an embodiment of the present invention;
[0035] Figure 7 is a schematic diagram of a framework of a method for predicting a postoperative facial change model in an embodiment of the present invention;
[0036] Figure 8 Schematic diagram of the structure of the prediction system of the postoperative facial change model in the embodiment of the present invention;
[0037] Fig. 9 The figure is a hardware diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION
[0038] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0039] In one embodiment, see Figures 1 to 9 A method for predicting a postoperative facial change model is applied to a prediction scenario of a postoperative facial change model; the method for predicting a postoperative facial change model comprises:
[0040] Step S11: during the simulation of the surgery, a facial soft tissue model, a preoperative bone model, and a postoperative bone model are collected;
[0041] Step S12: constructing two local geometric relationships based on the facial soft tissue model and extracting facial point features;
[0042] Step S13: extracting bone point features based on the preoperative bone model, and determining the displacement of the bone point based on the postoperative bone model;
[0043] Step S14: Determine the displacement features of the facial points by combining the bone point features, the bone point displacement amounts, and the facial point features;
[0044] Step S15: Predicting a postoperative facial change model based on the displacement characteristics of facial points and the facial soft tissue model.
[0045] In an embodiment of the present invention, through the method in the embodiment of the present invention, two local geometric relationships are constructed based on the facial soft tissue model. The local geometric features encode the facial topological shape to obtain more effective facial point features. The bone point features, the bone point displacement, and the facial point features are combined to determine the displacement features of the facial points. The postoperative facial change model is predicted based on the displacement features of the facial points and the facial soft tissue model, so as to better establish the interactive relationship between the movement of the bones and the movement of the facial soft tissue points, thereby improving the prediction accuracy, and the obtained postoperative facial change model has fewer artifacts.
[0046] See also Figure 2 , in step S11, during the simulation of the surgery, a facial soft tissue model and a bone model are collected;
[0047] In the specific implementation process of the present invention, the specific steps may be:
[0048] S111: Collect the patient's preoperative examination images;
[0049] S112: constructing a facial model and a bone model based on the inspection image;
[0050] S113: determining a facial soft tissue model according to the cutting of the facial model;
[0051] S114: Determine the postoperative bone model based on the bone model and the amount of movement of the jawbone.
[0052] In an embodiment of the present application, Proplan software is used to segment and construct a facial model and a bone model from the CBCT of the patient's preoperative examination. At this time, the patient's preoperative examination image is collected, and the examination image is an image in the CBCT of the preoperative examination, so that the facial model and the bone model can be constructed based on the examination image, thereby achieving the freezing of the facial model and the bone model, and the facial model and the bone model are used as the preoperative model.
[0053] Furthermore, the facial soft tissue model is determined according to the cropping of the facial model, so that the facial soft tissue model can be constructed based on the facial model, thereby further controlling the facial soft tissue model. At the same time, the postoperative bone model is determined based on the bone model and the movement of the jaw. Therefore, the doctor performs surgical planning and simulates the jaw movement in orthognathic surgery to obtain a bone model after moving the jaw, and the bone model after moving the jaw is used as the postoperative bone model.
[0054] In addition, the areas that will be affected by the surgery in the facial soft tissue model and the bone model are cut out, and the displacement of the postoperative bone point features on the bone model is calculated through the registration algorithm based on the planned bone movement.
[0055] The registration algorithm here uses the ICP registration algorithm, which can be specifically expressed as the following formula:
[0056] set up is the preoperative bone point cloud, is the postoperative bone point cloud, R and t are the rotation matrix and translation matrix, and the registration algorithm iteratively performs the steps of pairing point calculation and optimization of rotation and translation matrix, specifically optimizing the following expression:
[0057]
[0058] The registration algorithm registers the postoperative bone model to the preoperative bone model to obtain the corresponding R and t. After the transformation of R and t, it is recorded as The displacement vector of the bone point feature after surgery can be expressed as follows:
[0059]
[0060] read Figure 3 , in step S12, two local geometric relationships are constructed based on the facial soft tissue model, and facial point features are extracted;
[0061] In the specific implementation process of the present invention, the specific steps may be:
[0062] S121: Perform function operations based on the facial soft tissue model to construct two local geometric relationships;
[0063] S122: Construct a dual-graph convolution based on two local geometric relationships to extract features of each facial point.
[0064] In an embodiment of the present application, function operations are performed based on the facial soft tissue model to further solve the facial soft tissue model, thereby constructing two local geometric relationships, so as to use dual graph convolution based on the two local geometric relationships, and then extract each facial point feature, thereby ensuring the accuracy of the facial point features.
[0065] At this time, the facial soft tissue model, the Euclidean distance function and the geodesic distance function are associated; and two local geometric relationships between points in the facial soft tissue model are established through the Euclidean distance function and the geodesic distance function.
[0066] The Euclidean distance between two points x and y in N-dimensional space can be expressed as follows:
[0067]
[0068] Geodesic distance is expressed as the shortest path length between two points in a graph.
[0069] Therefore, two local geometric relationships are constructed based on the facial soft tissue model. The local geometric features encode the facial topological shape to obtain more effective facial point features. The displacement features of the postoperative bone points and the facial point features are combined to determine the displacement features of the facial points. The postoperative facial change model is predicted based on the displacement features of the facial points and the facial soft tissue model, so as to better establish the interactive relationship between the movement of the bones and the movement of the facial soft tissue points, thereby improving the prediction accuracy. The resulting postoperative facial change model has fewer artifacts.
[0070] See also Figure 4 , in step S13, the displacement of the bone point feature is determined based on the postoperative bone model;
[0071] In the specific implementation process of the present invention, the specific steps may be:
[0072] S131: freeze the postoperative bone model and construct the postoperative bone point cloud;
[0073] S132: determining the features of each bone point based on the preoperative bone point cloud;
[0074] S133: Determine the displacement of the postoperative bone point according to the postoperative bone point and the characteristics of each bone point.
[0075] In an embodiment of the present application, the postoperative bone model is frozen and a postoperative bone point cloud is constructed to facilitate feature extraction based on the preoperative bone point cloud, and the displacement of the postoperative bone points is determined based on the preoperative and postoperative bone points, thereby ensuring the accuracy of the displacement of the bone point features.
[0076] See also Figure 5 , S14: combining the bone point features, the bone point displacements, and the facial point features, determining the displacement features of the facial points;
[0077] In the specific implementation process of the present invention, the specific steps may be:
[0078] S141: associating the postoperative bone points and the facial points based on the local geometric relationships;
[0079] S142: Combining the postoperative bone point and the bone point displacement to form a postoperative bone point displacement feature;
[0080] S143: Migrate the postoperative bone point displacement features to the facial points to determine the displacement features of the facial points.
[0081] In an embodiment of the present application, postoperative displacement features of bone points and facial point features are introduced to facilitate associating the postoperative displacement features of bone points and facial point features. At this time, the postoperative displacement features of bone points and facial point features are associated based on each local geometric relationship to facilitate combining the displacement features of bone points and facial point features, thereby migrating the displacement features of bone points to facial point features to determine the displacement features of facial points, thereby facilitating subsequent processing of the displacement features of facial points.
[0082] At this time, the migration process of the displacement features of the bone points relative to the facial point features is monitored; the displacement features of the bone points are migrated to the facial point features to obtain the first prediction result; in each prediction process, the subsequent prediction result is compared with the first prediction result, and the erroneous result generated by the first prediction is corrected, wherein the subsequent prediction is to make a more fine-grained adjustment based on the first prediction.
[0083] The present invention designs a soft tissue displacement prediction module (DGCFP). The coarse-to-fine prediction refinement module in the soft tissue displacement prediction module (DGCFP) takes more points into consideration through upsampling based on the first prediction result, strengthens the constraints on point prediction, avoids the slow convergence speed caused by weak supervision training, and easily generates outliers, making the algorithm more robust.
[0084] In addition, facial models, facial point features, bone models, and postoperative bone models are collected; multi-order loss functions are trained based on the facial models, facial point features, bone models, and postoperative bone models to build a prediction model. Optionally, a two-stage loss function consisting of Chamferloss, density loss, and landmark restricted laplacian regularization is used for training. The specific loss function formulas include:
[0085] Chamfer Loss is:
[0086]
[0087] Where Y is the actual postoperative facial point cloud, is the predicted postoperative facial point cloud, and d is the L2 distance.
[0088] Density Loss is:
[0089]
[0090] Landmark Restricted Laplacian Regularization is:
[0091]
[0092] in, It is the soft tissue of the face before surgery. is the set of key points, δ v are the Laplace coordinates of point v, is the Laplace coordinate after rough prediction, is the Laplace coordinate after fine prediction, p is the real key point, and It is the key point of prediction.
[0093] See also Figure 6 ,S15: Model for predicting postoperative facial changes based on displacement characteristics of facial points and facial soft tissue model;
[0094] In the specific implementation process of the present invention, the specific steps may be:
[0095] S151: Collect the displacement characteristics of facial points and facial soft tissue model;
[0096] S152: Combining the displacement features of facial points with the facial soft tissue model and merging the postoperative unaffected areas to predict the postoperative facial change model.
[0097] In an embodiment of the present application, the displacement characteristics of facial points and the facial soft tissue model are collected to facilitate combining the displacement characteristics of facial points with the facial soft tissue model. At the same time, a postoperative unaffected area is introduced, and the postoperative unaffected area is further merged to predict the postoperative facial change model.
[0098] The specific steps include:
[0099] The prediction algorithm of the postoperative facial change model is composed of a data preprocessing module, a soft tissue displacement prediction module (DGCFP), and a post-processing module in series.
[0100] The data preprocessing module is used to generate 3D data that can be used to perform prediction from the original patient examination data. The module includes:
[0101] CBCT reconstruction of facial soft tissue model and bone model (Bone & facial soft tissue segmentation and reconstruction);
[0102] Calculate jaw displacement vector (Bone movement planning);
[0103] Cropping the affected area and downsampling (Cropping & Downsampling);
[0104] The soft tissue displacement prediction module (DGCFP) can solve the above technical problems faced by previous algorithms by considering local geometric relationships when extracting facial features and adopting a two-stage prediction model from coarse to fine. The DGCFP module includes:
[0105] Multi-scale Dualconv Face Encoder;
[0106] Bone point cloud feature encoding module (Pointwise Bone Encoder);
[0107] Dual-space Movement Transfer module;
[0108] Coarse-to-fine prediction refinement modules (Coarse Prediction, Mesh Unpooling, FinePrediction).
[0109] The present invention can construct two local geometric relationships based on the facial soft tissue model. The local geometric features encode the facial topological shape to obtain more effective facial point features, and combine the displacement features of bone points and facial point features to determine the displacement features of facial points. The postoperative facial change model is predicted based on the displacement features of facial points and the facial soft tissue model, so as to better establish the interactive relationship between the movement of bones and the movement of facial soft tissue points, thereby improving the prediction accuracy, and the obtained postoperative facial change model has fewer artifacts.
[0110] Example
[0111] See also Figure 8 , Figure 8 It is a schematic diagram of the structural composition of the prediction system of the postoperative facial change model in an embodiment of the present invention.
[0112] like Figure 8 As shown, a prediction system for a postoperative facial change model, the prediction system for a postoperative facial change model comprises:
[0113] The acquisition module 21 is used to acquire the facial soft tissue model, the preoperative bone model and the postoperative bone model during the simulation of the operation;
[0114] An extraction module 22, used to construct two local geometric relationships based on the facial soft tissue model and extract facial point features;
[0115] A first displacement module 23, used for extracting bone point features based on a preoperative bone model, and determining the displacement of the bone point based on a postoperative bone model;
[0116] The second displacement module 24 is used to determine the displacement characteristics of the facial points by combining the bone point characteristics, the bone point displacement, and the facial point characteristics;
[0117] The prediction module 25 is used to predict the postoperative facial change model based on the displacement characteristics of facial points and the facial soft tissue model.
[0118] Example
[0119] See also Fig. 9 , refer to the following Fig. 9 The electronic device 40 according to this embodiment of the present invention will be described. Fig. 9 The electronic device 40 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0120] like Fig. 9 As shown, the electronic device 40 is in the form of a general computing device. The components of the electronic device 40 may include but are not limited to: at least one processing unit 41, at least one storage unit 42, and a bus 43 connecting different system components (including the storage unit 42 and the processing unit 41).
[0121] The storage unit stores program codes, which can be executed by the processing unit 41, so that the processing unit 41 executes the steps according to various exemplary embodiments of the present invention described in the above “Embodiment Method” section of this specification.
[0122] The storage unit 42 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit 421 and / or a cache storage unit 422 , and may further include a read-only storage unit (ROM) 423 .
[0123] The storage unit 42 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.
[0124] Bus 43 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
[0125] The electronic device 40 may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device 40, and / or any device that enables the electronic device 40 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 44. Furthermore, the electronic device 40 may also communicate with one or more networks (e.g., local area networks, wide area networks, and / or public networks, such as the Internet) via a network adapter 45. Fig. 9 As shown, the network adapter 45 communicates with other modules of the electronic device 40 via the bus 43. It should be understood that although Fig. 9 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, tape drives, and data backup planning systems.
[0126] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.
[0127] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium can include: a read-only memory, a random access memory, a disk or an optical disk, etc. In addition, it stores computer program instructions, and when the computer program instructions are executed by a computer, the computer executes the above method.
[0128] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.
[0129] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.
[0130] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0131] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0132] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A method for predicting a postoperative facial change model, characterized in that: include: During the simulation of the surgery, facial soft tissue models, preoperative bone models, and postoperative bone models were collected; Two local geometric relationships are constructed based on the facial soft tissue model, and facial point features are extracted; Extract bone point features based on the preoperative bone model, and determine the displacement of the bone points based on the postoperative bone model; Determine the displacement characteristics of facial points by combining bone point characteristics, bone point displacement, and facial point characteristics; A model for predicting postoperative facial changes based on the displacement characteristics of facial points and facial soft tissue model.
2. The method for predicting a postoperative facial change model according to claim 1, characterized in that: During the simulation of the surgery, the facial soft tissue model, the preoperative bone model and the postoperative bone model are collected, including: Collect the patient's preoperative examination images; constructing a facial model and a bone model based on the inspection images; Determining a facial soft tissue model based on the cutting of the facial model; The postoperative bone model is determined based on the bone model and the amount of jaw bone movement.
3. The method for predicting a postoperative facial change model according to claim 1, characterized in that: The method of constructing two local geometric relationships based on the facial soft tissue model and extracting facial point features includes: Perform function operations based on the facial soft tissue model and construct two local geometric relationships; A dual-graph convolution is constructed based on two local geometric relationships to extract the features of each facial point.
4. The method for predicting a postoperative facial change model according to claim 3, characterized in that: The function operation is performed based on the facial soft tissue model to construct two local geometric relationships, including: Correlating facial soft tissue models, Euclidean distance functions, and geodesic distance functions; Two local geometric relationships between points in the facial soft tissue model are established through the Euclidean distance function and the geodesic distance function.
5. The method for predicting a postoperative facial change model according to claim 1, characterized in that: The step of extracting bone point features based on the preoperative bone model and determining the displacement of the bone point based on the postoperative bone model includes: Freeze the postoperative bone model and construct the postoperative bone point cloud; Determine the features of each bone point based on the preoperative bone point cloud; The displacement of the postoperative bone points is determined according to the postoperative bone points and the characteristics of each bone point.
6. The method for predicting a postoperative facial change model according to claim 1, characterized in that: The step of combining the bone point features, the bone point displacement amounts, and the facial point features to determine the facial point displacement features includes: Correlate postoperative bone points and facial points based on local geometric relationships; Combining the postoperative bone point and the bone point displacement, the postoperative bone point displacement characteristics are formed; The postoperative displacement characteristics of bone points are transferred to facial points to determine the displacement characteristics of facial points.
7. The method for predicting a postoperative facial change model according to claim 6, characterized in that: The step of combining the bone point features, the bone point displacement amounts, and the facial point features to determine the facial point displacement features also includes: Monitor the postoperative migration of bone point displacement features relative to facial points; The postoperative bone point displacement features are transferred to the facial points to obtain the first prediction results; In each prediction process, the next prediction result is compared with the first prediction result, and the erroneous result produced by the first prediction is corrected. The purpose of the next prediction is to make more fine-grained adjustments based on the first prediction.
8. The method for predicting a postoperative facial change model according to claim 7, characterized in that: The step of combining the bone point features, the bone point displacement amounts, and the facial point features to determine the facial point displacement features also includes: Collect facial models, facial point features, bone models, and postoperative bone models; Multi-order loss functions are trained based on facial models, facial point features, bone models, and postoperative bone models to build a prediction model.
9. The method for predicting a postoperative facial change model according to claim 1, characterized in that: The model for predicting postoperative facial changes based on the displacement characteristics of facial points and the facial soft tissue model includes: Collect the displacement characteristics of facial points and facial soft tissue model; The displacement characteristics of facial points are combined with the facial soft tissue model, and the postoperative unaffected areas are merged to predict the postoperative facial change model.
10. A prediction system for a postoperative facial change model, characterized in that: The prediction system of the postoperative facial change model is applied to the prediction method of the postoperative facial change model as claimed in any one of claims 1 to 9, and the prediction system of the postoperative facial change model comprises: An acquisition module, used to acquire facial soft tissue models, preoperative bone models, and postoperative bone models during the simulation of the surgery; An extraction module, used to construct two local geometric relationships based on the facial soft tissue model and extract facial point features; A first displacement module is used to extract bone point features based on a preoperative bone model and determine the displacement of the bone point based on a postoperative bone model; The second displacement module is used to determine the displacement characteristics of the facial points by combining the bone point characteristics, the bone point displacement, and the facial point characteristics; The prediction module is used to predict the postoperative facial change model based on the displacement characteristics of facial points and the facial soft tissue model.
Citation Information
Cited By
Postoperative facial soft tissue form prediction method and system based on conditional deformation field supplementation
CN120809259A
Postoperative facial soft tissue morphology prediction method and system based on conditional deformation field completion
CN120809259B
