Implant data processing method and system for edentulous jaw guide plate, product and medium

Abstract

The invention discloses an implantation data processing method and system of an edentulous jaw guide plate, a product and a medium, and relates to the technical field of tooth implantation, and the method comprises the following steps: obtaining facial static feature data, facial motion video data and jaw bone tomography data of a target patient, and carrying out space registration on a facial static three-dimensional model and a jaw bone three-dimensional model to obtain a three-dimensional model; obtaining a position corresponding relation between the bony mark point and the jaw position; mark point movement track data in the face movement video data are extracted, dynamic displacement data of each mark point are obtained, the relative distance variation between the mark points is calculated according to the dynamic displacement data, and a preset number of mark points with the minimum relative distance variation are selected as guide plate positioning datum points; and generating a guide plate digital design scheme according to the three-dimensional coordinate information of the guide plate positioning datum point and the jaw bone three-dimensional model. By implementing the method, the positioning stability of the edentulous jaw guide plate can be improved.

Description

Technical Field

[0001] The present application relates to the field of dental implant technology, and in particular to an implant data processing method, system, product and medium for an edentulous jaw guide. Background Art

[0002] With the development of oral medical technology, edentulous jaw implant restoration has become a common treatment option in dental clinics. Edentulous jaw patients usually face problems such as chewing dysfunction and facial morphology changes, so accurate implant restoration is of great significance to improving the quality of life of patients. In implant surgery, the precise design and positioning of the implant guide directly affects the accuracy of implant placement and the surgical effect.

[0003] In related technologies, the guide plate can be designed by collecting the patient's jaw CT data and combining it with soft tissue markers. The doctor attaches markers to the patient's face, collects the patient's facial data and jaw data in a closed state, and then determines the positioning reference of the guide plate based on the position information of these fixed markers, thereby completing the design of the guide plate.

[0004] However, in real life, patients make various facial expressions such as talking and smiling, which can cause deformation of facial soft tissue. As the existing technology only collects data from patients in a static state, it cannot reflect the movement law of soft tissue during dynamic deformation. Therefore, the guide plate positioning reference point determined based on static data is easily affected by changes in facial expressions, which reduces the stability of guide plate positioning. Summary of the invention

[0005] The present application provides an implant data processing method, system, product and medium for an edentulous jaw guide plate, which are used to improve the stability of the positioning of the edentulous jaw guide plate.

[0006] In a first aspect, the present application provides an implant data processing method for an edentulous jaw guide plate, which is applied to an implant data processing system, the method comprising: acquiring facial static feature data, facial motion video data and jaw tomography data of a target patient, the facial feature data including three-dimensional spatial position information of bony landmarks; performing feature extraction on the facial static feature data to obtain a facial static three-dimensional model; performing feature extraction on the jaw tomography data to obtain a jaw three-dimensional model; spatially aligning the facial static three-dimensional model with the jaw three-dimensional model to obtain a positional correspondence between bony landmarks and jaw positions; extracting the landmark point motion trajectory data in the facial motion video data to obtain dynamic displacement data of each landmark point, the dynamic displacement data including three-dimensional spatial coordinate change information of the landmark point under different expression states; calculating the relative distance change between each of the landmark points based on the dynamic displacement data, and selecting a preset number of landmark points with the smallest relative distance change as guide plate positioning reference points; generating a digital design scheme for the guide plate based on the three-dimensional coordinate information of the guide plate positioning reference points and the jaw three-dimensional model.

[0007] In the above embodiment, facial static feature data including three-dimensional spatial position information of bony landmarks, facial motion video data and jaw tomography data of the target patient are obtained, and a facial static three-dimensional model and a jaw three-dimensional model are obtained through feature extraction. After spatial alignment, the guide positioning reference point is determined based on the facial motion video data, and then a digital design scheme for the guide is generated in combination with the jaw three-dimensional model. The comprehensive use and precise processing of multi-source data enables the guide design to fully consider the static and dynamic characteristics of the face, improve the accuracy and stability of the guide positioning, ensure a more precise implant placement, and effectively improve the implant restoration effect of the edentulous jaw.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the static three-dimensional facial model and the three-dimensional jaw model are spatially aligned to obtain the positional correspondence between the bony landmark points and the jaw position, specifically including: using the bony landmark points in the static three-dimensional facial model as reference points to establish a coordinate system for the static three-dimensional facial model; extracting jaw feature points based on the anatomical features of the three-dimensional jaw model, and establishing a jaw coordinate system; aligning and transforming the coordinate system of the static three-dimensional facial model with the jaw coordinate system, and calculating the three-dimensional coordinates of the aligned bony landmark points in the jaw coordinate system to obtain the positional correspondence between the bony landmark points and the jaw position.

[0009] In the above embodiment, the bony landmarks in the static three-dimensional facial model are used as reference points to establish a coordinate system, and the jaw coordinate system is established by extracting jaw feature points based on the anatomical features of the three-dimensional jaw model, and then a registration transformation is performed to obtain the positional correspondence between the bony landmarks and the jaw positions. By using the reference function of the bony landmarks and the anatomical positioning of the jaw feature points, the coordinate system is accurately established and registered to ensure that the face and the jaw model are accurately corresponding in space, providing an accurate spatial reference for the subsequent determination of the guide plate positioning reference points and the guide plate design, so that the guide plate can better adapt to the individual characteristics of the patient and improve the accuracy and success rate of the implant surgery.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the step of extracting the motion trajectory data of the marker points in the facial motion video data to obtain the dynamic displacement data of each marker point specifically includes: performing frame decomposition on the facial motion video data to obtain a continuous video frame sequence; using a feature point tracking algorithm to track the motion trajectory of each bony marker point in the video frame sequence; performing three-dimensional reconstruction on the motion trajectory to obtain the three-dimensional spatial coordinates of each bony marker point at different times; and based on the three-dimensional spatial coordinates, calculating the displacement of each bony marker point relative to the initial position to obtain the dynamic displacement data.

[0011] In the above embodiment, the facial motion video data frame is decomposed to obtain a video frame sequence, and the feature point tracking algorithm is used to track the motion trajectory of the bony landmarks. The three-dimensional reconstruction obtains the three-dimensional spatial coordinates at different times, and then the dynamic displacement data is calculated. The frame decomposition and feature point tracking of the video data accurately capture the motion trajectory of the bony landmarks during the facial expression changes. The three-dimensional reconstruction and displacement calculation provide accurate dynamic displacement information, which can deeply analyze the deformation law of facial soft tissues and provide a reliable basis for selecting the landmark with the smallest relative distance change as the reference point for the guide plate positioning, thereby enhancing the stability of the guide plate positioning and ensuring the implant restoration effect.

[0012] In combination with some embodiments of the first aspect, in some embodiments, the steps of calculating the relative distance change between each of the landmark points based on the dynamic displacement data, and selecting a preset number of landmark points with the smallest relative distance change as the guide plate positioning reference points specifically include: calculating the Euclidean distance between any two of the bony landmark points in different expression states based on the dynamic displacement data; obtaining the maximum change in the relative distance between each pair of bony landmark points during the entire movement process; sorting the relative distance change from small to large, and selecting the top N bony landmark points with the smallest change as the guide plate positioning reference points.

[0013] In the above embodiment, the Euclidean distance between the bony landmarks is calculated based on the dynamic displacement data, the relative distance change is obtained, and the preset number of landmarks with the smallest change are sorted and selected as the guide plate positioning reference points. Calculating the Euclidean distance can accurately quantify the change in the spatial relationship between the landmarks. By analyzing the relative distance change during the entire movement process, a combination of landmarks with relatively stable positions when the facial expression changes can be screened out. These stable landmarks serve as positioning reference points to provide a reliable positioning reference for the guide plate, so that the guide plate can maintain an accurate position under different facial conditions, effectively improving the stability of the guide plate positioning and the accuracy of the implant surgery, and ensuring a good implant restoration effect.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the steps of generating a digital design scheme for the guide plate based on the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the mandible specifically include: according to the position correspondence, projecting the guide plate positioning reference point onto the surface of the three-dimensional model of the mandible to obtain a reference projection point; calculating the offset area of ​​the reference projection point within the corresponding range of the dynamic displacement data, and constructing a positioning pillar at the boundary of the area; generating a support structure connecting each of the positioning pillars based on the spatial distribution of the three-dimensional model of the mandible and the positioning pillars; constructing a guide plate base based on the surface contour of the three-dimensional model of the mandible and the position of the positioning pillars; and constructing the digital design scheme for the guide plate based on the guide plate base, the positioning pillars and the support structure.

[0015] In the above embodiment, the guide plate positioning reference point is projected onto the surface of the three-dimensional model of the mandible according to the position correspondence to obtain the reference projection point, the offset area is calculated to construct the positioning pillar, the support structure is generated by combining the mandible model and the pillar distribution, the guide plate matrix is ​​constructed based on the mandible contour and the pillar position, and then the guide plate digital design scheme is constructed. The reference projection point determines the support point, and the pillars and support structure constructed considering the dynamic displacement can adapt to facial movement, and the matrix that fits the mandible ensures overall stability. The various components cooperate with each other, and the guide plate structure designed based on precise data is reasonable, stable and reliable, which can accurately guide the implantation of the implant, improve the success rate of implant restoration, and provide better treatment solutions for edentulous patients.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the step of generating a digital design plan of the guide plate based on the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw, the method also includes: establishing a patient facial feature database based on the static three-dimensional facial model; and storing the distribution characteristics of the guide plate positioning reference point in the feature database.

[0017] In the above embodiment, after the digital design of the guide plate is generated, a patient facial feature database is established based on the static three-dimensional facial model and the distribution characteristics of the guide plate positioning reference points are stored. The static three-dimensional facial model provides basic data for the database, and storing the distribution characteristics of the positioning reference points is helpful for subsequent analysis. The database can be used for data comparison and effect evaluation during the patient's treatment process, providing doctors with comprehensive information, facilitating doctors to further understand the individual characteristics of patients, helping to optimize treatment plans and improve the personalization of treatment. At the same time, it also accumulates data for medical research and promotes the development and improvement of edentulous jaw implant technology.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of generating a digital design plan of the guide plate based on the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw, the method further includes: when receiving a plan acquisition instruction, sending the digital design plan of the guide plate to the target client.

[0019] In the above embodiment, after the digital design plan of the guide is generated, it is sent to the target client when the plan acquisition instruction is received. The digital design plan of the guide is an accurate result based on multi-source data processing, and contains detailed information of each component of the guide. The target client can be a doctor's end or other related equipment end. Sending the plan in time can ensure that relevant personnel can obtain accurate guide design data at the first time, so that the preparation work for the implant surgery can be carried out quickly and accurately. This ensures the efficient connection of the treatment process, avoids time waste due to data transmission delays, improves the efficiency of the entire edentulous jaw implant restoration process, and ensures that patients can receive appropriate treatment more quickly.

[0020] In a second aspect, an embodiment of the present application provides a planting data processing system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the planting data processing system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when the above-mentioned computer program product is run on a planting data processing system, enables the above-mentioned planting data processing system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions. When the instructions are executed on a planting data processing system, the planting data processing system executes the method described in the first aspect and any possible implementation method of the first aspect.

[0023] It can be understood that the planting data processing system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. This application obtains the target patient's facial static feature data including the three-dimensional spatial position information of bony landmarks, facial motion video data and jaw tomography data, and obtains a facial static three-dimensional model and a jaw three-dimensional model through feature extraction. After spatial alignment, the guide positioning reference point is determined based on the facial motion video data, and then a digital design plan for the guide is generated in combination with the jaw three-dimensional model. The comprehensive use and precise processing of multi-source data enables the guide design to fully consider the static and dynamic characteristics of the face, improve the accuracy and stability of the guide positioning, ensure a more precise implant placement, and effectively improve the implant restoration effect of the edentulous jaw.

[0025] 2. This application establishes a coordinate system by using the bony landmarks in the static three-dimensional facial model as reference points, extracts the jaw feature points based on the anatomical features of the three-dimensional jaw model to establish a jaw coordinate system, and then performs a registration transformation to obtain the positional correspondence between the bony landmarks and the jaw positions. By using the reference role of the bony landmarks and the anatomical positioning of the jaw feature points, the coordinate system is accurately established and registered to ensure that the face and the jaw model correspond precisely in space, providing an accurate spatial reference for the subsequent determination of the guide positioning reference points and the guide design, so that the guide can better adapt to the individual characteristics of the patient and improve the accuracy and success rate of the implant surgery.

[0026] 3. This application obtains a video frame sequence by decomposing the facial motion video data frame, and uses a feature point tracking algorithm to track the motion trajectory of the bony landmarks, and three-dimensional reconstruction to obtain the three-dimensional spatial coordinates at different times, and then calculates the dynamic displacement data. The frame decomposition and feature point tracking of the video data accurately capture the motion trajectory of the bony landmarks during the facial expression changes. The three-dimensional reconstruction and displacement calculation provide accurate dynamic displacement information, which can deeply analyze the deformation law of facial soft tissues, and provide a reliable basis for selecting the landmark with the smallest relative distance change as the reference point for the guide plate positioning, thereby enhancing the stability of the guide plate positioning and ensuring the implant restoration effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of a method for processing implant data of an edentulous jaw guide plate in an embodiment of the present application; Figure 2It is another flowchart of the implantation data processing method of the edentulous jaw guide plate in the embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a physical device of the planting data processing system in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] As the population ages, the number of edentulous patients continues to increase. These patients generally have problems such as chewing dysfunction, unclear pronunciation, and changes in facial morphology, which seriously affect their quality of life. Full-mouth implant restoration is currently the most ideal treatment option, but it faces the technical difficulty of ensuring implant accuracy. Especially when determining the position of the implant guide, the facial soft tissue of elderly patients is very loose and produces obvious deformation when their expressions change, which makes the guide designed based on static data prone to position displacement during use. Clinical practice has shown that the accuracy of implant position is directly related to the stability of guide positioning. Therefore, how to ensure the stability of guide positioning under dynamic facial soft tissue deformation conditions has become a key technical issue affecting the treatment effect.

[0031] Traditional implant guide design methods mainly rely on static data. Clinically, titanium alloy markers are attached to the patient's face and facial CT scan data in a closed state are collected for guide design. This method has obvious limitations in practical applications: when patients engage in daily activities such as speaking and swallowing, the facial soft tissue deforms, causing the guide positioning pillars designed based on static positions to shift. Practical data show that this displacement may cause a 3-5 degree deviation in the implant placement angle, affecting the distribution and arrangement of the implants and reducing the stability and service life of the restoration.

[0032] The newly developed dynamic guide design system uses a multi-source data fusion method to collect static feature data, dynamic expression videos, and jaw CT data at the same time. The system analyzes the movement trajectory of the marker points in daily expression activities, identifies relatively stable areas of the face, and uses these areas as the focus points of the guide positioning pillars. Experimental data show that the guide designed based on dynamic data can remain stable when facial expressions change, and the implant placement deviation is controlled within 0.5mm.

[0033] For ease of understanding, the following describes the process of the method provided by this implementation in combination with the above scenario. Figure 1 , which is a flow chart of the implant data processing method of the edentulous jaw guide plate in the embodiment of the present application.

[0034] S101, acquiring facial static feature data, facial motion video data, and jaw tomography data of a target patient, wherein the facial feature data includes three-dimensional spatial position information of bony landmarks.

[0035] Among them, the target patient refers to a patient who needs edentulous jaw implant restoration, which is used to represent the object of data collection. Facial static feature data refers to the collected patient facial surface morphology information, including three-dimensional point cloud data and the position of bony landmark points. Facial motion video data refers to a continuous image sequence recording the patient completing standard facial expressions. Jaw tomography data refers to a sequence of jaw cross-sectional images obtained using CT and other equipment. Bone landmark points refer to spherical markers attached to specific anatomical positions on the patient's face, which are used to establish the spatial correspondence between the face and the jaw. Three-dimensional spatial position information represents the X, Y, and Z coordinate values ​​of the landmark points in the three-dimensional coordinate system.

[0036] This step is performed when the patient first visits the doctor to collect the basic data required for the subsequent guide plate design. Specifically, the system first uses a three-dimensional optical scanner to scan the patient's face and collect facial surface point cloud data with a resolution of 0.1mm. At the same time, a titanium alloy spherical marker with a diameter of 3mm is attached to the obvious anatomical landmarks on the patient's face. A binocular high-speed camera system is used to record the process of the patient completing standard facial expressions such as smiling and chewing. The video sampling rate is 120fps and the resolution is 4096×3072 pixels. Finally, cone beam CT is used to scan the patient's jaw area with a layer thickness of 0.125mm, and the scanning range covers the entire alveolar bone area.

[0037] In some embodiments, the data collection process can be implemented in a variety of ways: optionally, using a structured light 3D scanner to collect static facial data, using a robotic arm to control the scanning path, and using a multi-angle scanning method to splice a complete facial model; using an industrial camera array to synchronously collect facial motion videos, and performing time and space synchronization calibration between cameras; using spiral CT to scan the jaws and perform motion artifact correction. Optionally, using a handheld laser scanner to collect facial data, with real-time registration and splicing during the scanning process; using a monocular high-speed camera in conjunction with a depth camera to collect facial motion videos; using CBCT to scan the jaws, and using a metal artifact suppression algorithm to optimize image quality. It is understandable that other data acquisition devices and methods can also be used to achieve the acquisition of the required data, which is not limited here.

[0038] S102: Extract the facial static feature data to obtain a facial static three-dimensional model.

[0039] Feature extraction refers to the process of extracting key information required to build a three-dimensional model from raw data. Features include geometric features such as spatial coordinates, normal vectors, and curvature in point cloud data. A static three-dimensional facial model refers to a digital model that expresses the patient's facial appearance and is reconstructed using computer graphics methods. It contains facial contours, surface details, and marker point location information.

[0040] This step is performed after acquiring the facial static feature data to generate a standardized 3D model for subsequent processing. Specifically, the system first preprocesses the original point cloud data, including outlier removal and noise smoothing. Then the point cloud is downsampled and the mesh is reconstructed to generate a regular triangular mesh model. The reconstructed mesh is optimized, including filling holes, optimizing topology, and simplifying the mesh. Finally, the precise position of the bony landmarks is extracted and marked, and the local geometric features around each marker are calculated.

[0041] In some embodiments, feature extraction can be achieved in a variety of ways: optionally, using statistical outlier detection to remove abnormal points; using moving least squares method to smooth point clouds; using Poisson reconstruction algorithm to generate a watertight triangular mesh model; identifying the location of marker points based on local curvature analysis. Optionally, using a density-based clustering algorithm to filter noise points; using bilateral filtering to smooth point clouds; using a greedy projection triangulation algorithm to generate a mesh model; locating marker points based on a template matching method. It is understandable that other feature extraction and model reconstruction algorithms can also be used to achieve the construction of a three-dimensional model, which is not limited here.

[0042] S103, extracting features from the mandibular tomography data to obtain a three-dimensional model of the mandibular bone.

[0043] Among them, jaw tomography data refers to a series of images of jaw cross sections obtained using imaging equipment such as CT, which is used to represent the internal structure information of the jaw. Feature extraction refers to the process of identifying and extracting jaw tissue from continuous two-dimensional tomographic images. The jaw three-dimensional model refers to a digital anatomical model constructed through image processing and three-dimensional reconstruction technology, which is used to represent the shape and internal structure of the jaw. The jaw tissue includes two types of bone tissue with different densities, cortical bone and cancellous bone, which are represented by different grayscale value ranges in CT images.

[0044] This step is performed after acquiring the mandibular tomography data and is used to generate an accurate three-dimensional model of the mandible. Specifically, the system first preprocesses the CT image sequence, including image enhancement and noise suppression, to improve the clarity of tissue boundaries. Then, a multi-threshold segmentation method is used to extract the cortical bone and cancellous bone areas respectively, and the threshold range is determined based on the Hounsfield value. The segmentation results are morphologically processed, including boundary smoothing and cavity filling. Finally, a three-dimensional reconstruction algorithm is used to reconstruct the continuous segmentation results into a high-precision three-dimensional mesh model while retaining bone density information.

[0045] In some embodiments, the construction of the three-dimensional model of the jaw can be achieved in a variety of ways: optionally, using adaptive histogram equalization to enhance image contrast; using a multi-threshold segmentation method based on region growing to separate different bone tissues; using morphological closing operations to fill small holes; using a marching cube algorithm to reconstruct the three-dimensional surface. Optionally, using anisotropic diffusion filtering to eliminate image noise; using a level set-based active contour segmentation method to extract bone tissue boundaries; using distance transforms to smooth segmentation boundaries; using a patch reconstruction algorithm to generate a three-dimensional model. It is understandable that other image processing and model reconstruction methods can also be used to achieve the construction of the three-dimensional model of the jaw, which is not limited here.

[0046] S104: spatially aligning the static three-dimensional facial model with the three-dimensional model of the jaw to obtain a positional correspondence between the bony landmarks and the jaw positions.

[0047] Among them, spatial registration refers to the process of transforming the static 3D facial model and the 3D jaw model into the same coordinate system. The position correspondence relationship represents the spatial position description of the bony landmarks in the jaw coordinate system. Registration transformation includes two basic transformations, rotation and translation, which are used to adjust the spatial position and posture of the model. As the reference point for registration, the bony landmarks have clear spatial coordinates in both models.

[0048] This step is performed after the reconstruction of the facial and jaw three-dimensional models is completed, and is used to establish a spatial mapping relationship between the two models. Specifically, the system first extracts the three-dimensional coordinates of all bony landmarks from the facial model and extracts the corresponding landmark positions from the jaw model. The facial coordinate system and the jaw coordinate system are established, and the transformation matrix between the two coordinate systems is calculated. The model is aligned using a rigid body transformation algorithm to ensure that the shape of the model does not change before and after the transformation. Finally, the alignment accuracy is verified, and the residuals between all corresponding points are calculated to ensure that the alignment error is within a clinically acceptable range.

[0049] In some embodiments, the spatial registration process can be implemented in a variety of ways: optionally, use principal component analysis to initialize the model direction; use iterative closest point algorithm for coarse registration; use improved quaternion method to optimize the registration result; calculate the registration error and make position fine-tuning. Optionally, use feature point matching to determine the initial position; use gradient descent-based registration algorithm to optimize transformation parameters; use robust estimation method to eliminate the influence of abnormal matching points; calculate the registration quality score and optimize. It is understandable that other registration algorithms and optimization methods can also be used to achieve spatial alignment of two models, which is not limited here.

[0050] S105, extracting the motion trajectory data of the marker points in the facial motion video data to obtain dynamic displacement data of each marker point, wherein the dynamic displacement data includes three-dimensional space coordinate change information of the marker point under different expression states.

[0051] Among them, facial motion video data refers to a sequence of facial motion images of patients recorded with a high-speed camera, which is used to represent the dynamic process of facial expression changes. The landmark motion trajectory data represents the continuous position change of the bony landmarks in the video sequence. Dynamic displacement data refers to the spatial displacement of the landmarks relative to the initial position, which is used to describe the deformation characteristics of facial soft tissue. The three-dimensional spatial coordinate change information represents the displacement components of the landmarks in the X, Y, and Z directions. The expression state refers to the standard expression movements made by the patient, including daily facial activities such as smiling and chewing.

[0052] This step is performed after the video data acquisition is completed to obtain the motion characteristics of the marker points. Specifically, the system first decomposes the video data into frames to generate a continuous image sequence of 120fps. Distortion correction and noise suppression are performed on each frame to improve image quality. The deep learning target detection algorithm is used to identify the position of the marker points in each frame and output sub-pixel two-dimensional coordinates. Based on the principle of binocular stereo vision, the three-dimensional coordinates of the marker points are restored by parallax calculation. The three-dimensional coordinate sequence is filtered in the time domain to eliminate tracking noise and measurement errors. Finally, the displacement of each marker point relative to the initial frame is calculated to form a complete dynamic displacement data set.

[0053] In some embodiments, the extraction of marker trajectory can be achieved in a variety of ways: optionally, using an improved YOLOv5 network to detect markers; using Kalman filtering to predict the position of markers; combining template matching algorithms to improve positioning accuracy; using triangulation to calculate three-dimensional coordinates; using Butterworth filters to smooth the trajectory. Optionally, using FasterR-CNN detectors to locate markers; using optical flow to track marker motion; using feature point matching to optimize tracking results; reconstructing three-dimensional positions based on multi-view geometry; and using wavelet transforms for denoising. It is understandable that other target detection and trajectory tracking algorithms can also be used to extract the motion trajectory of markers, which is not limited here.

[0054] S106. Calculate the relative distance change between each of the marking points based on the dynamic displacement data, and select a preset number of marking points with the smallest relative distance change as guide plate positioning reference points.

[0055] The relative distance change refers to the change in the Euclidean distance between any two landmarks over time, which is used to indicate the stability of the positional relationship between the landmark pairs. The preset number indicates the number of reference points determined according to the design requirements of the guide. The guide positioning reference points refer to the reference points used to determine the spatial position of the implant guide. These points have high spatial stability during changes in facial expressions.

[0056] This step is performed after obtaining the dynamic displacement data of the marker points, and is used to screen the most stable combination of marker points. Specifically, the system first calculates the Euclidean distance of each pair of marker points in the entire motion sequence to generate a distance-time curve. Perform extreme value analysis on each curve, and calculate the difference between the maximum and minimum values ​​as the relative distance change. Sort the changes of all pairs of marker points in ascending order, and establish a sorted table of changes. Starting from the top of the sorted table, select the point pair with the smallest change, while considering the spatial distribution characteristics of the points to ensure that the selected reference points can form a stable spatial configuration. Finally, select a specified number of marker points as the reference points for guide plate positioning.

[0057] In some embodiments, the selection of reference points can be achieved in a variety of ways: optionally, using Euclidean distance to calculate the distance between point pairs; using the sliding window method to detect the extreme value of the distance; calculating the relative change and constructing a sorted heap; screening candidate points based on spatial distribution constraints; verifying the stability of the selected point set. Optionally, using Mahalanobis distance to measure the relationship between point pairs; using the peak detection algorithm to find the range of distance changes; using quick sorting to construct a sequence of changes; combining convex hull analysis to select a point set with a reasonable spatial distribution; and evaluating the geometric constraint capability of the point set. It is understandable that other distance metrics and point set selection methods can also be used to determine the guide plate positioning reference point, which is not limited here.

[0058] S107, generating a digital design scheme for the guide plate according to the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw.

[0059] Among them, the three-dimensional coordinate information of the guide plate positioning reference point refers to the spatial position data of the selected reference point in the mandibular coordinate system, which is used to determine the spatial positioning of the guide plate. The three-dimensional model of the mandible refers to a digital model that expresses the patient's mandibular anatomical structure, including the outer contour and internal structure information. The digital design plan of the guide plate represents the complete three-dimensional model of the implant guide, including the geometric parameters and spatial relationships of components such as the guide plate base, positioning pillars and supporting structure. The guide plate base refers to the main structure that fits the surface of the mandible. The positioning pillar refers to the supporting pillar connecting the base and the reference point. The supporting structure refers to the reinforcing structure that connects the pillars to form an overall framework.

[0060] This step is performed after the guide plate positioning reference point is determined, and is used to generate a complete guide plate design plan. Specifically, the system first projects the reference point onto the mandibular surface to determine the position of the pillar's fulcrum. A variable-section pillar is constructed at each projection point, and the pillar height is determined according to the local anatomical gap. The mandibular surface mesh is extracted and locally optimized to generate a guide plate matrix that fits the mandibular surface precisely. Connecting beams are designed between the pillars to form a support frame, and the frame adopts an I-shaped cross-section to increase strength. Finite element analysis is performed on the overall structure to verify its mechanical properties under the use load. Finally, a 3D model file in a standard format is generated, containing parametric design information of all components.

[0061] In some embodiments, the digital design of the guide plate can be achieved in a variety of ways: optionally, the position of the pillar is determined using normal projection; the surface mesh of the jaw is extracted based on the scan line algorithm; the guide plate matrix is ​​generated using the offset algorithm; the pillar model is created using parametric modeling; a B-spline surface is constructed to connect adjacent pillars; and structural strength and stability analysis is performed. Optionally, the pillar is positioned using the nearest point projection; surface features are extracted based on octree partitioning; a matrix is ​​generated using hierarchical Boolean operations; pillars are constructed using feature templates; support structures are designed using stress-guided topology optimization; and dynamic load simulation analysis is performed. It is understandable that other three-dimensional modeling and structural optimization methods can also be used to achieve the digital design of the guide plate, which is not limited here.

[0062] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the implant data processing method of the edentulous jaw guide plate in the embodiment of the present application.

[0063] S201, acquiring facial static feature data, facial motion video data, and jaw tomography data of a target patient, wherein the facial feature data includes three-dimensional spatial position information of bony landmarks.

[0064] It can be understood that this step is similar to step S101 and will not be described in detail here.

[0065] S202: Extract features from the static facial feature data to obtain a static three-dimensional facial model.

[0066] It can be understood that this step is similar to step S102 and will not be described in detail here.

[0067] S203, extracting features from the jaw bone tomography data to obtain a three-dimensional model of the jaw bone.

[0068] It can be understood that this step is similar to step S103 and will not be described in detail here.

[0069] S204: Use the bony landmarks in the static three-dimensional facial model as reference points to establish a coordinate system for the static three-dimensional facial model.

[0070] Bony landmarks are radiopaque markers attached to the patient's facial surface. These markers are located on bony prominences that are fixed in anatomical positions and easily identifiable. Benchmarks are reference points used to establish a coordinate system. Three or more non-collinear bony landmarks are usually selected as benchmarks. The facial static three-dimensional model coordinate system is a three-dimensional rectangular coordinate system based on selected benchmarks, which is used to describe the spatial position of each point in the facial model.

[0071] The system first extracts the spatial coordinates of all bony landmarks from the static three-dimensional facial model, and selects the left and right points of the supraorbital margin and the nasion as the main reference points. The origin and main axis direction of the coordinate system are determined by these three points: the nasion is set as the origin of the coordinates, the direction of the line connecting the two points of the supraorbital margin is defined as the positive direction of the X axis, the direction of the line connecting the nasion to the midpoint of the two supraorbital margins is defined as the positive direction of the Y axis, and the positive direction of the Z axis is determined according to the right-hand rule. The system uses the least squares method to optimize the coordinates of the reference points to ensure that the established coordinate system has the best orientation accuracy.

[0072] S205. Extract jaw feature points based on the anatomical features of the jaw three-dimensional model, and establish a jaw coordinate system.

[0073] The jaw feature points refer to the characteristic positions of the jaw anatomical structure, including the highest point of the alveolar ridge, mandibular angle, mandibular foramen and other anatomical landmarks. The jaw coordinate system is a three-dimensional rectangular coordinate system established based on the extracted jaw feature points, which is used to describe the spatial position relationship of the jaw anatomical structure.

[0074] The system uses curvature analysis and morphological feature extraction methods to automatically identify anatomical feature points on the surface of the three-dimensional model of the jaw. For the maxilla, the deepest point of the left and right incisor fossa and the posterior end point of the midpalatal suture are selected as reference points; for the mandible, the left and right mandibular angles and the mental foramen are selected as reference points. Based on these feature points, the system establishes the jaw coordinate system: the midpoint of the incisor fossa or the midpoint of the mental foramen is used as the origin, the direction of the line connecting the symmetrical points on both sides is defined as the X-axis, and the direction perpendicular to the occlusal plane is defined as the Z-axis, thereby determining the complete jaw coordinate system.

[0075] S206, performing registration transformation on the coordinate system of the static three-dimensional facial model and the jaw coordinate system, and calculating the three-dimensional coordinates of the registered bony landmarks in the jaw coordinate system to obtain the positional correspondence between the bony landmarks and the jaw positions.

[0076] Registration transformation refers to the mathematical process of converting points in the facial static 3D model coordinate system to the jaw coordinate system, including rotation and translation transformation. Position correspondence refers to the spatial coordinates of bony landmarks in the jaw coordinate system, reflecting the relative position of the landmarks to the jaw anatomical structure.

[0077] The system uses a rigid transformation algorithm for coordinate system registration. The specific steps include: first, calculating the rotation matrix and translation vector between the two coordinate systems; then transforming the coordinates of all bony landmarks to obtain their new coordinates in the jaw coordinate system; and finally establishing a database of spatial correspondence between landmarks and jaw anatomical structures. The system uses the ICP (Iterative Closest Point) algorithm for registration optimization, with a registration accuracy of better than 0.5 mm. The calculation of the transformation matrix uses the SVD (Singular Value Decomposition) method to ensure the rigid body characteristics of the transformation and avoid model deformation.

[0078] S207: Decompose the facial motion video data into frames to obtain a continuous video frame sequence.

[0079] Facial motion video data refers to a continuous dynamic image sequence that records the changes in the patient's facial expressions, which is collected at a fixed frame rate using high-speed video equipment. Video frame sequence refers to the decomposition of continuous video data into discrete image frames in chronological order, each frame representing the facial state at a specific moment. Frame decomposition is the process of converting video data into a sequence of images that can be processed individually.

[0080] The system uses a video decoder to decompose the original video data according to the acquisition frame rate, generating a 120-frame image sequence per second. The decomposition process includes: first reading the encoding format and resolution information of the video file; then dividing the data stream into independent frames according to the video timestamp; finally converting each frame into a standard image format and saving it. The system uses the H.264 decoding algorithm to process the video data and generate an RGB image sequence with a resolution of 1920×1080 pixels to ensure that the image quality meets the requirements of subsequent processing. The time interval between video frames is maintained at 8.33 milliseconds, ensuring the continuity of motion capture.

[0081] S208: Using a feature point tracking algorithm, track the motion trajectory of each bony landmark point in the video frame sequence.

[0082] The feature point tracking algorithm refers to a computational method that automatically identifies and tracks the position of a target point in a sequence of continuous video frames. The motion trajectory of a bony landmark refers to the path of the spatial position change of the landmark during the entire facial expression change process. The feature point tracking process needs to solve problems such as target recognition, position prediction, and trajectory smoothing.

[0083] The system uses a multi-feature fusion tracking algorithm to achieve marker tracking. First, the positions of all markers are detected in the initial frame through color and shape features; then the Kalman filter is used to predict the approximate position of the marker in the next frame; then the template matching method is used in the predicted area to accurately locate the marker; finally, the B-spline curve is used to smooth the obtained trajectory. The system integrates SIFT feature matching and optical flow method, and can maintain stable tracking effect under lighting changes and fast movement, and the tracking accuracy reaches sub-pixel level.

[0084] The tracking module of the system first detects the positions of all markers in the first frame using the FasterR-CNN network trained with marker data. For each detected marker, a 16×16 pixel area centered on it is extracted as a feature template. In the processing of subsequent frames, the system uses a Kalman filter to predict the motion state of the marker, and the prediction search range is a 32×32 pixel area. The normalized cross-correlation algorithm is used for template matching in the predicted area, and the matching threshold is set to 0.85 to ensure accuracy. The matching position is accurately located at the sub-pixel level using the centroid calculation method, and the positioning accuracy reaches 0.1 pixel. The system dynamically updates the Kalman filter state and template features, and the feature update adopts the exponential sliding average method, with a historical feature weight of 0.9. When the marker is occluded, the trained LSTM network is used to predict its position until the marker is redetected. Finally, the Savitzky-Golay filter is used to smooth the trajectory, with a filter window length of 21 frames and a polynomial order of 3.

[0085] S209, performing three-dimensional reconstruction on the motion trajectory to obtain the three-dimensional spatial coordinates of each of the bony landmarks at different times.

[0086] It refers to the position of the marker point in the world coordinate system, including the position information in three directions: X, Y, and Z. The coordinate sequence at different times constitutes a complete spatial motion trajectory.

[0087] The system uses binocular stereo vision method for 3D reconstruction. First, the binocular camera is calibrated to obtain the internal and external parameters of the camera; then the polar line constraint is used to determine the correspondence between the markers in the left and right views; then the 3D coordinates of the markers are calculated using the triangulation principle; finally, the reconstruction results are time-series filtered to eliminate reconstruction noise. The system uses Zhang Zhengyou calibration method for camera calibration, and the reconstruction accuracy is better than 0.1 mm. For occlusion, the system performs trajectory compensation through multi-view geometric constraints and motion continuity constraints to ensure the integrity of the reconstruction results.

[0088] In the reconstruction process, the system first uses a standard 9×12 chessboard calibration plate to collect 20 sets of calibration images of different poses, and calculates the camera intrinsic parameter matrix and lens distortion coefficient through the Zhang Zhengyou calibration algorithm. Then, the rotation matrix and translation vector between the two cameras are calculated according to the principle of epipolar geometry, and the geometric relationship model of the binocular system is established. Distortion correction and epipolar correction are performed on each pair of collected images to generate a corrected standard image pair. The binocular matching process uses a region matching algorithm based on normalized cross-correlation, with a matching window size of 7×7 pixels, combined with a dynamic programming algorithm to ensure the global consistency of the matching process. After obtaining the disparity data, the three-dimensional coordinates of the marker points are calculated by the triangulation principle, and the coordinates are transformed into the world coordinate system. The system uses a Butterworth low-pass filter with a cutoff frequency of 30Hz to perform time domain filtering on the reconstruction results to eliminate high-frequency noise. For data missing caused by occlusion, the cubic spline interpolation method is used to complete the trajectory. Finally, 120Hz three-dimensional trajectory data is output, which contains the temporal spatial coordinate information of all marker points.

[0089] S210. Based on the three-dimensional spatial coordinates, calculate the displacement of each bony landmark point relative to the initial position to obtain the dynamic displacement data, which includes the three-dimensional spatial coordinate change information of the landmark point under different expression states.

[0090] The three-dimensional spatial coordinates refer to the absolute position of the bony landmark in the world coordinate system, expressed in a rectangular coordinate system (X, Y, Z). The initial position refers to the spatial position of the landmark when the patient maintains a natural state, which serves as a reference point for calculating displacement. The dynamic displacement data records the relative movement of the landmark relative to the initial position, including the displacement components in three directions and the composite displacement.

[0091] The system first synchronizes the reconstructed 3D trajectory data and aligns the timestamps of all frames to a unified reference. For each marker point, the first frame position is taken as the initial reference point P0 (x0, y0, z0). The difference between the marker point position Pi (xi, yi, zi) and the initial position in each subsequent frame is calculated to obtain the displacement vector D (dx, dy, dz). The system uses 32-bit floating point numbers to store displacement data with an accuracy of 0.001mm. The calculated displacement data is fitted by least squares to obtain the displacement-time curves in each direction. The curve fitting uses a 5th-order polynomial model, and the fitting residual is controlled within 0.01mm. The final output dynamic displacement data contains a displacement time series with a sampling rate of 120Hz, which records the complete facial expression change process.

[0092] S211. Based on the dynamic displacement data, calculate the Euclidean distance between any two bony landmarks in different expression states.

[0093] Euclidean distance refers to the straight-line distance between two points in three-dimensional space, reflecting the relative position relationship between two markers in space. For any two markers, their Euclidean distance changes with facial expressions, and this change reflects the deformation characteristics of facial soft tissue.

[0094] The system establishes an N×N distance matrix (N is the total number of markers) and calculates the Euclidean distance between all pairs of markers in each frame. For markers Pi (xi, yi, zi) and Pj (xj, yj, zj), calculate their Euclidean distance dij = √ [(xi-xj)² + (yi-yj)² + (zi-zj)²]. The calculation uses the CUDA parallel processing architecture to process the distance calculation of all point pairs at the same time. The system generates a distance-time curve for each pair of points, and the sampling rate remains consistent with the original data (120Hz). To improve the calculation accuracy, the system uses the Kahan summation algorithm to eliminate the accumulated error of floating-point operations to ensure that the distance calculation accuracy is better than 0.005mm.

[0095] S212, obtaining the maximum change in the relative distance between each pair of bony landmarks during the entire movement process.

[0096] Relative distance refers to the Euclidean distance between two marked points, and the maximum change refers to the difference between the maximum and minimum distances between two points during the entire movement. This indicator reflects the stability of the relative position relationship between the two marked points. The smaller the change, the more stable the relative position between the two points.

[0097] The system analyzes the distance-time curve of each pair of markers. First, the local maximum and minimum values ​​of the distance are found through the peak detection algorithm, and the detection threshold is set to 0.05mm to filter out noise fluctuations. The detected extreme values ​​are sorted, the maximum value dmax and the minimum value dmin are taken, and the maximum change Δd=dmax-dmin is calculated. The system generates an N×N change matrix, where N is the total number of markers, and the matrix element (i, j) stores the maximum distance change between the i-th and j-th markers. At the same time, the time points when the maximum and minimum distances occur are recorded for subsequent analysis of the timing characteristics of deformation. The calculation process uses double-precision floating point numbers to ensure that the calculation accuracy of the change reaches 0.001mm.

[0098] S213, sorting the relative distance changes from small to large, and selecting the first N bony landmarks with the smallest changes as the guide plate positioning reference points, where N is equal to the preset number.

[0099] The relative distance change refers to the maximum change in the Euclidean distance between two marking points, in millimeters. The guide plate positioning reference point refers to the reference point used to determine the spatial position of the implant guide. The preset number N is the number of reference points determined based on the geometric dimensions and support requirements of the guide plate, usually set to 4-6.

[0100] The system first constructs an N×N change matrix (N is the number of all marked points), and each matrix element represents the change in the relative distance between the corresponding two points. Use the quick sort algorithm to sort the changes of all non-repeating point pairs in ascending order. Select the point pairs with the smallest change from the sorting results in turn, and check whether the points in the newly selected point pairs are already included in the selected point set. Stop selecting when the cumulative number of non-repeating points selected reaches the preset number. The system evaluates the spatial distribution of the selected reference points, requiring that any three points are not collinear and any four points are not coplanar, to ensure that the spatial configuration of the reference points has sufficient constraint capabilities. If the spatial distribution does not meet the requirements, continue to select the next qualifying point from the sorted list until a set of reference points that meets the geometric constraints is obtained.

[0101] S214. According to the position correspondence, the guide plate positioning reference point is projected onto the surface of the mandibular three-dimensional model to obtain a reference projection point.

[0102] Position correspondence refers to the coordinate mapping relationship between facial landmarks and the jaw model established after spatial registration. Reference projection points refer to the spatial points obtained by projecting the guide plate positioning reference points along the normal direction onto the jaw surface. These points are the actual points of support for the guide plate support structure.

[0103] The system first reads the surface mesh data of the three-dimensional model of the mandible, and uses an octree structure to divide the model space to accelerate the search for the nearest point. For each selected guide positioning reference point, the system calculates the direction of the shortest distance to the mandibular surface, which is the projection direction. The intersection point on the mandibular surface is searched along the projection direction as the reference projection point. To ensure the local smoothness of the projection point, the system extracts the surface mesh within a range of 10mm around the projection point, and uses the bicubic B-spline function to locally reconstruct the surface. The final position of the projection point is located on the reconstructed surface. The system also records the surface normal vector and principal curvature direction at each projection point for the subsequent design of the support structure.

[0104] S215, calculating the offset area of ​​the reference projection point within the corresponding range of the dynamic displacement data, and constructing a positioning pillar at the boundary of the area.

[0105] The offset area refers to the displacement range of the reference projection point during facial movement, which is determined by the dynamic displacement data. The positioning pillar is a supporting structure connecting the guide base and the jaw surface to ensure the precise positioning of the guide.

[0106] The system first analyzes the dynamic displacement data corresponding to each reference projection point and calculates its spatial envelope surface during the entire motion cycle. The displacement data is projected into the local coordinate system of the projection point, with the Z axis of the coordinate system along the surface normal and the X axis along the direction of the maximum principal curvature. The extreme range of displacement is calculated in the local coordinate system to determine the boundary of the offset area. The system designs the geometric parameters of the pillar according to the size of the offset area: the diameter of the pillar base is 1.2 times the diameter of the circumscribed circle of the offset area; the pillar height is determined according to the local anatomical gap, generally 5-8mm; the cross-section of the pillar adopts a variable cross-section design, with a circular bottom and a transition to a square top, which is convenient for connection with the guide plate base. The material of the pillar is medical-grade photosensitive resin, and the surface is specially treated to increase biocompatibility. The system uses a parametric modeling method to generate a three-dimensional model of the pillar and adds a rounded corner (radius 0.5mm) on the surface of the pillar to reduce stress concentration.

[0107] S216. Generate a support structure connecting each positioning pillar according to the three-dimensional model of the jaw and the spatial distribution of the positioning pillars.

[0108] The support structure is the frame part that connects adjacent positioning pillars and is used to enhance the overall rigidity and stability of the guide plate. The spatial distribution of the positioning pillars refers to the relative position relationship of each pillar on the surface of the jaw, and these positions determine the geometric shape of the support structure. The three-dimensional model of the jaw provides the anatomical boundary conditions for the design of the support structure.

[0109] The system uses a parametric design method to generate the support structure. First, a Delaunay triangulation network is constructed to connect the tops of all positioning pillars into closed polygons. For each connecting edge, the system calculates the distance and relative height difference between the pillars at both ends to generate a variable cross-section connecting beam. The connecting beam adopts an I-shaped cross-section, and the cross-sectional parameters change adaptively with the span: the flange width is 4-6mm, the web thickness is 2-3mm, and the total height is 1 / 10 of the span. The path of the connecting beam takes into account the undulations of the jaw surface, and a cubic Bezier curve is used for transition to ensure that the minimum gap with the jaw surface is not less than 2mm. The system performs finite element analysis on the support structure to ensure that the maximum deformation under a vertical load of 100N is less than 0.1mm. The support structure material uses medical-grade PEEK material, which has good mechanical properties and biocompatibility.

[0110] S217, constructing a guide plate base based on the surface contour of the three-dimensional model of the jaw and the position of the positioning support.

[0111] The guide base is the main structure covering the surface of the jaw, used to support the implant guide and fix the entire guide system. The surface contour refers to the geometric features of the surface of the jaw 3D model, including concave and convex changes and anatomical landmarks. The position of the positioning pillar determines the boundary range of the base and the distribution of the fixing points.

[0112] The system first marks the surgical area on the surface of the mandible, and expands 5mm outside this area as the coverage range of the guide base. The surface mesh within this range is extracted, and the adaptive resampling algorithm is used to optimize the mesh quality. The side length of the triangular facet is controlled between 0.5-1mm. The inner surface of the guide base is generated by offsetting 2mm outward along the surface normal, and the outer surface is uniformly thickened by 3mm according to the inner surface. At the location of the positioning pillar, the system automatically generates a locking structure with a dovetail groove design with a groove width of 4mm, a depth of 3mm, and an inclination angle of 15 degrees to ensure a reliable connection between the pillar and the base. The system arranges drainage and ventilation holes on the surface of the base with a hole diameter of 1mm and a spacing of 5mm, avoiding the positions of pillars and reinforcement ribs. The guide base material is made of light-cured transparent resin with a hardness of ShoreD85 or above.

[0113] S218, constructing a digital design scheme for the guide plate based on the guide plate base, the positioning pillar and the supporting structure.

[0114] The digital design of the guide plate is a complete three-dimensional digital model, which contains all the geometric information and material parameters of the guide plate base, positioning pillars and supporting structures. This design will be used for subsequent 3D printing manufacturing and clinical application.

[0115] The system uses parametric modeling software to integrate all components. First, the guide plate base is used as the main body, and the reference surface of the pillar installation position is arranged on its surface. The positioning pillar is precisely docked with the base through a dovetail groove, and the contact surface adopts an interference fit of 0.05mm to ensure stability after assembly. The support structure is connected to the pillar and the base by chemical bonding, and the contact surface is designed with a positioning boss. The system automatically generates engineering drawings and assembly drawings of all parts, including key dimensions and tolerance requirements. At the same time, an STL file for 3D printing is generated, and the model accuracy is set to 0.01mm to ensure the dimensional accuracy of the print. The system performs finite element analysis and verification on the complete guide, including static analysis and fatigue analysis, to ensure its safety in the oral environment. The final design solution is output as a 3D model file in a standard format, containing a complete parametric history record, which is convenient for subsequent modification and optimization.

[0116] In some embodiments, the step further includes the following steps: Based on the static three-dimensional facial model, a patient facial feature database is established.

[0117] Specifically, the facial feature database refers to a structured data set used to store the patient's facial anatomical features and morphological parameters. The static three-dimensional facial model represents the digital expression of the patient's facial surface morphology. Features include quantitative parameters such as facial contour curves, anatomical landmark locations, and surface curvature distribution. The database structure contains multiple data tables such as patient basic information, feature data, and index relationships.

[0118] This step is performed after the reconstruction of the facial 3D model is completed, and is used to establish a standardized feature storage system. Specifically, the system first extracts key anatomical features from the 3D model, including facial contour lines, nasolabial grooves, orbital margins and other characteristic curves. The main curvature distribution and shape index of the facial surface are calculated to generate surface morphological feature descriptors. A relational database based on PostgreSQL is established, and the patient information table, feature parameter table and index table are designed. The extracted feature data is stored in the database in a standard format, and the association relationship between the feature data is established.

[0119] In some embodiments, the establishment of a feature database can be achieved in a variety of ways: optionally, using differential geometry methods to extract surface features; establishing a feature index structure based on B-tree; using JSON format to store feature data; designing data integrity constraints; and implementing database backup and recovery mechanisms. Optionally, using deep learning methods to extract semantic features; establishing hash indexes to accelerate queries; using binary format compression storage; designing triggers to maintain data consistency; and implementing distributed data synchronization. It is understandable that other database technologies and feature storage methods can also be used to manage facial features, which are not limited here.

[0120] The distribution characteristics of the guide plate positioning reference points are stored in the feature database.

[0121] Specifically, the distribution characteristics of the guide plate positioning reference points refer to the spatial arrangement pattern of the reference points, including the relative position relationship between the points and the geometric configuration characteristics. The distribution characteristics include the three-dimensional coordinates of the reference points, the distance matrix between the point pairs, the geometric invariants of the configuration, and other information. The feature database is used to store and manage these distribution feature data.

[0122] This step is performed after the guide plate positioning reference points are determined, and is used to record the typical pattern of reference point selection. Specifically, the system calculates the spatial distribution characteristics of the reference point set, including geometric invariants such as point-to-point distance matrix, triangle area ratio, and dihedral angle. Generate feature descriptors of the reference point distribution, including position information and configuration parameters. The feature descriptors are stored in a special table structure in the database to establish an association with the patient information.

[0123] In some embodiments, the storage of distribution features can be implemented in a variety of ways: optionally, calculating the Delaunay triangulation of the reference points; extracting the geometric invariants of the configuration; generating standardized feature vectors; designing feature similarity metrics; and implementing feature retrieval functions. Optionally, calculating the convex hull features of the reference points; extracting the main directions of the spatial configuration; generating a compact representation of the features; designing feature clustering analysis; and implementing pattern recognition functions. It is understandable that other feature representation and storage methods can also be used to implement the management of the reference point distribution features, which are not limited here.

[0124] When the solution acquisition instruction is received, the guide plate digital design solution is sent to the target client.

[0125] Specifically, the scheme acquisition instruction refers to the command information sent by the client to request the guide design data. The target client refers to the authorized terminal device used to receive and display the guide design scheme. The guide digital design scheme includes the guide 3D model and related parameter information, using a standard data exchange format.

[0126] This step is executed after receiving the client request and is used to achieve the secure transmission of the design plan. Specifically, the system first verifies the legitimacy of the request instruction and checks the client's access rights. The guide plate design plan is packaged into a standard format, including a 3D model file, a parameter configuration file, and a description document. The data packet is encrypted using an encryption algorithm to ensure the security of the transmission process. The encrypted data is sent to the target client through the network protocol and the receipt confirmation information is verified.

[0127] The following describes the planting data processing system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of a physical device structure of a planting data processing system in an embodiment of the present application.

[0128] It should be noted that Figure 3 The structure of the planting data processing system 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.

[0129] like Figure 3 As shown, the planting data processing system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, the ROM 302 and the RAM 303 are connected to each other through the bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0130] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0131] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0132] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0133] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0134] Specifically, the implant data processing system of the present embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the implant data processing method of the edentulous jaw guide plate provided in the above embodiment is implemented.

[0135] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the implant data processing system described in the above embodiment; or may exist independently without being assembled into the implant data processing system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the implant data processing system, the implant data processing system implements the implant data processing method of the edentulous jaw guide provided in the above embodiment.

[0136] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements 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 application.

[0137] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0138] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A method for processing implant data of an edentulous jaw guide, characterized in that: Applied to a planting data processing system, the method comprises: Acquire static facial feature data, facial motion video data, and jaw tomography data of a target patient, wherein the facial feature data includes three-dimensional spatial position information of bony landmarks; Extracting features from the facial static feature data to obtain a facial static three-dimensional model; Extracting features from the jaw tomography data to obtain a three-dimensional model of the jaw; Performing spatial registration of the static three-dimensional facial model with the three-dimensional model of the jaw to obtain a positional correspondence between bony landmarks and jaw positions; Extracting the motion trajectory data of the marker points in the facial motion video data to obtain dynamic displacement data of each marker point, wherein the dynamic displacement data includes three-dimensional space coordinate change information of the marker point under different expression states; Calculate the relative distance change between each of the marking points according to the dynamic displacement data, and select a preset number of marking points with the smallest relative distance change as guide plate positioning reference points; A digital design scheme for the guide plate is generated based on the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw.

2. The method according to claim 1, characterized in that The step of spatially registering the static facial three-dimensional model with the jaw three-dimensional model to obtain the position correspondence between the bony landmarks and the jaw positions specifically includes: Using the bony landmarks in the static three-dimensional facial model as reference points, establishing a coordinate system for the static three-dimensional facial model; Extracting jaw feature points based on the anatomical features of the jaw three-dimensional model and establishing a jaw coordinate system; The coordinate system of the static three-dimensional facial model is registered and transformed with the jaw coordinate system, and the three-dimensional coordinates of the registered bony landmarks in the jaw coordinate system are calculated to obtain the positional correspondence between the bony landmarks and the jaw positions.

3. The method according to claim 1, characterized in that The step of extracting the motion trajectory data of the marker points in the facial motion video data to obtain the dynamic displacement data of each marker point specifically includes: Performing frame decomposition on the facial motion video data to obtain a continuous video frame sequence; Using a feature point tracking algorithm to track the motion trajectory of each bony landmark point in the video frame sequence; Performing three-dimensional reconstruction on the motion trajectory to obtain the three-dimensional spatial coordinates of each of the bony landmarks at different times; Based on the three-dimensional spatial coordinates, the displacement of each bony landmark point relative to the initial position is calculated to obtain the dynamic displacement data.

4. The method according to claim 1, characterized in that: The step of calculating the relative distance change between each of the marking points according to the dynamic displacement data, and selecting a preset number of marking points with the smallest relative distance change as the guide plate positioning reference points specifically includes: Based on the dynamic displacement data, calculating the Euclidean distance between any two of the bony landmarks in different expression states; Obtain the maximum change in the relative distance between each pair of bony landmarks during the entire movement process; The relative distance changes are sorted from small to large, and the first N bony landmarks with the smallest changes are selected as the guide plate positioning reference points, where N is equal to the preset number.

5. The method according to claim 1, characterized in that The step of generating a digital design scheme of the guide plate according to the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw specifically comprises: According to the position correspondence, the guide plate positioning reference point is projected onto the surface of the jaw three-dimensional model to obtain a reference projection point; Calculate the offset area of ​​the reference projection point within the corresponding range of the dynamic displacement data, and construct a positioning pillar at the boundary of the area; Generating a support structure connecting each of the positioning pillars according to the three-dimensional model of the jaw and the spatial distribution of the positioning pillars; Constructing a guide plate base based on the surface contour of the three-dimensional jaw model and the position of the positioning pillar; A digital design scheme for the guide plate is constructed based on the guide plate base, the positioning pillars and the supporting structure.

6. The method according to claim 1, characterized in that After the step of generating a digital design scheme of the guide plate according to the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw, the method further comprises: Based on the static three-dimensional facial model, establishing a patient facial feature database; The distribution characteristics of the guide plate positioning reference points are stored in the feature database.

7. The method according to claim 1, characterized in that After the step of generating a digital design scheme of the guide plate according to the three-dimensional coordinate information of the guide plate positioning reference point and the three-dimensional model of the jaw, the method further comprises: When the solution acquisition instruction is received, the guide plate digital design solution is sent to the target client.

8. A planting data processing system, characterized in that: The planting data processing system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the planting data processing system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a planting data processing system, the planting data processing system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a planting data processing system, the planting data processing system is caused to execute the method according to any one of claims 1 to 7.

Images