Integrated surgical guide plate and obturator design method and related device
By registering CBCT and intraoral scan data and using AI analysis, an integrated surgical guide and occluder were designed, which solved the problems of insufficient precision in pediatric jaw cyst surgery and long postoperative management cycles. This enabled precise and personalized treatment plans, reduced the number of follow-up visits, reduced the suffering of children, and improved treatment outcomes.
Patent Information
- Application Number
- CN202511521249.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-23
- Publication Date
- 2025-12-16
AI Technical Summary
Traditional pediatric jaw cyst surgery suffers from insufficient surgical precision, long postoperative occluder fabrication time, and inability to dynamically adapt to changes in cyst shrinkage and permanent tooth eruption, leading to increased suffering for children and delays in treatment.
By registering CBCT data and intraoral scan data, and combining AI automated analysis and 3D printing technology, an integrated surgical guide and occluder are designed to achieve a precise digital model of the cyst and adjacent anatomical structures. This model dynamically adapts to the healing of the cyst and the eruption trend of the permanent tooth. Surgical planning and postoperative management are optimized through multimodal image fusion and deep learning algorithms.
Significantly improve surgical precision and efficiency, reduce the number of follow-up visits, reduce children's suffering, improve treatment effects and prognosis, and promote the innovation of pediatric jaw cyst diagnosis and treatment towards intelligence, precision and standardization.
Smart Images
Figure CN121145484A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of oral and maxillofacial surgery and pediatric dentistry, and in particular to a design method and related device for an integrated surgical guide and occluder. Background Technology
[0002] The following problems exist in existing techniques for the sequential treatment of pediatric jaw cysts.
[0003] Insufficient surgical precision: Traditional window opening positions rely on the doctor's experience and are prone to damaging the permanent tooth germ or nerve canal.
[0004] Postoperative management is passive: the occluder is mostly made in a single session and cannot dynamically adapt to changes in cyst shrinkage and permanent tooth eruption, requiring frequent follow-up visits for adjustment.
[0005] Difficulty in cooperating with the child: Repeated mold removal or adjustment of the occluder increases the child's pain, and each production requires a production cycle of 1-2 weeks, delaying the treatment time.
[0006] Moreover, the existing solutions have the following drawbacks.
[0007] Conventional occluders are statically designed and cannot predict jawbone development trends.
[0008] There is a lack of quantitative prediction methods for the healing process of cysts, and the interval between follow-up visits depends on subjective experience.
[0009] Therefore, traditional surgical methods are limited in clinical application due to their reliance on the surgeon's experience, insufficient accuracy in cyst localization, and long production cycle of postoperative occluders. Summary of the Invention
[0010] The purpose of this application is to provide an integrated surgical guide and occluder design method and related device, which can systematically solve the technical bottlenecks of the three core links of surgical planning, intraoperative navigation and postoperative drainage, significantly reduce the learning curve and have significant clinical application value.
[0011] To achieve the above objectives, this application provides the following solution.
[0012] In a first aspect, this application provides a design method for an integrated surgical guide and occluder, the integrated surgical guide and occluder design method comprising the following steps.
[0013] Acquire CBCT data and intraoral scan data.
[0014] The CBCT data and the intraoral scan data are registered to obtain a digital model of the cyst and adjacent anatomical structures.
[0015] AI-automated analysis was performed on the digital model of the cyst and adjacent anatomical structures to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption.
[0016] Based on the spatial characteristics, the geometric topology settings, and the trends of cyst healing and permanent tooth eruption, multiple integrated surgical guides and occluders are parametrically designed using 3D printing technology.
[0017] Optionally, the CBCT data and the intraoral scan data are registered to obtain a digital model of the cyst and adjacent anatomical structures, specifically including the following steps.
[0018] The CBCT data is preprocessed to obtain preprocessed data; the preprocessing includes foreground intensity value cropping and normalization.
[0019] The preprocessed data was segmented using a CBCT segmentation model to obtain teeth, jawbones, and nerve canals; the CBCT segmentation model was a model trained based on the nnU-Net network.
[0020] Using the jawbone as a template, the cystic lesions of the jawbone were obtained by segmentation.
[0021] The connected components of jaw cystic lesions are calculated based on the 6-neighborhood property of voxel space. The volume of the connected components is obtained by the summation method, and connected components with volumes smaller than a preset value are filtered to obtain the filtered connected components.
[0022] The intraoral scan data is sampled to obtain a point cloud.
[0023] The point cloud is segmented using an oral scanning point cloud segmentation model to obtain the crown point cloud; the oral scanning point cloud segmentation model is a model trained based on a point cloud segmentation neural network.
[0024] The teeth are converted into a curved surface model and sampled to obtain several sampling points.
[0025] The RANSAC algorithm is used to register the sampling points with the crown point cloud to obtain the Euler transformation matrix.
[0026] The Euler transformation matrix is applied to the intraoral scanning model to obtain a digital model of the cyst and adjacent anatomical structures.
[0027] Optionally, the determination of the CBCT segmentation model specifically includes the following steps.
[0028] Obtain historical CBCT datasets.
[0029] The historical CBCT dataset was annotated using ITK-SNAP to obtain an annotated historical dataset; the annotated historical dataset includes: teeth, jawbone, nerve canal and jawbone cystic lesions.
[0030] Using historical datasets as input and labeled historical datasets as output, the nnU-Net model is trained to obtain the CBCT segmentation model.
[0031] Optionally, the determination of the point cloud segmentation model includes the following steps.
[0032] Obtain historical intraoral scan data.
[0033] The historical intraoral scan data is sampled to obtain historical point clouds.
[0034] The historical point cloud is annotated to obtain the annotated crown point cloud.
[0035] Using historical point clouds as input and labeled crown point clouds as output, a point cloud segmentation neural network is trained to obtain an oral scan point cloud segmentation model.
[0036] Optionally, the expression for the RANSAC algorithm is as follows.
[0037] .
[0038] in, p For sampling points; q Add cloud-like dots to the dental crown; For a pair of nearest neighbors; It is a rotation matrix; It is a translation matrix.
[0039] Optionally, the digital model of the cyst and adjacent anatomical structures is subjected to AI automated analysis to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption. This includes the following steps.
[0040] The digital model of the cyst and adjacent anatomical structures is labeled based on the neural tube and the filtered connected components to obtain a labeled dataset; the labeling includes: the location of the permanent tooth germ, the boundary of the cyst, and the orientation of the neural tube.
[0041] Based on the labeled dataset, an AI simulation algorithm is used to predict the output, obtaining spatial features, geometric features, topological structure setting parameters, cyst healing, and permanent tooth eruption trends.
[0042] Optionally, the geometric feature topology setting parameters are as follows: the opening window is greater than or equal to 1 cm or greater than 2 tooth positions, located on the labial / buccal side of the lesion closest to the alveolar ridge, the vertical depth of the occluder should be less than the vertical distance from the normal alveolar ridge plane to the bottom of the cyst, located on the labial / buccal side or above the permanent tooth germ, the base plate occluder should avoid buccal retention devices, and the lingual / palatal base plate extends above the central part of the abutment tooth.
[0043] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the integrated surgical guide and occluder design method described in any one of the first aspects.
[0044] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the integrated surgical guide and occluder design method described in any one of the first aspects.
[0045] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the integrated surgical guide and occluder design method described in any of the first aspects.
[0046] Based on the specific embodiments provided in this application, the following technical effects are disclosed.
[0047] This application provides a design method and related device for an integrated surgical guide and occluder. The method includes: acquiring CBCT data and intraoral scan data; this step provides accurate and comprehensive raw data support for subsequent digital model construction. Registering the CBCT data and the intraoral scan data yields a digital model of the cyst and adjacent anatomical structures; this step enables accurate digital representation of the cyst and adjacent anatomical structures, laying the foundation for subsequent analysis. Performing AI-automated analysis on the digital model of the cyst and adjacent anatomical structures yields spatial features, geometric topological structure setting parameters, and trends in cyst healing and permanent tooth eruption; this step improves analysis efficiency and accuracy, providing a scientific basis for design. Based on the spatial features, the geometric topological structure setting parameters, and the trends in cyst healing and permanent tooth eruption, multiple integrated surgical guides and occluders are parametrically designed using 3D printing technology; this step enables personalized and precise design of surgical instruments to meet surgical needs. This application, through the coordinated efforts of each step, enables precise, efficient, and personalized design of integrated surgical guides and occluders, providing reliable technical support for cyst surgical treatment and helping to improve surgical outcomes and prognosis. Attached Figure Description
[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0049] Figure 1 This is an application environment diagram of an integrated surgical guide and occluder design method according to an embodiment of this application.
[0050] Figure 2 This is a flowchart illustrating an integrated surgical guide and occluder design method provided in one embodiment of this application.
[0051] Figure 3 This is a diagram of the CBCT segmentation model architecture provided in an embodiment of this application.
[0052] Figure 4 This is a diagram of the oral scanning segmentation model architecture provided in an embodiment of this application.
[0053] Figure 5 This is a schematic diagram of a unified model for CBCT and intraoral scan alignment provided in an embodiment of this application.
[0054] Figure 6 This is a schematic diagram of the equipment required for one embodiment of this application.
[0055] Figure 7 A schematic diagram comparing the conventional solution provided in one embodiment of this application with the solution of this application.
[0056] Figure 8 This is an illustration illustrating the innovation of this application as an embodiment.
[0057] Figure 9 This is a schematic diagram of a highly adaptive medical image segmentation algorithm provided in one embodiment of this application.
[0058] Figure 10 This is a schematic diagram illustrating the combination of bone tissue information with dental crown and soft tissue information provided in an embodiment of this application.
[0059] Figure 11 This is a schematic diagram of a multimodal image that integrates bone and soft tissue information, provided as an embodiment of this application.
[0060] Figure 12 This is a schematic diagram illustrating the setting parameters of the geometric feature topology structure provided in an embodiment of this application.
[0061] Figure 13 This is a schematic diagram illustrating a three-dimensional visualization example of a pediatric jaw cyst after time registration, provided as an embodiment of this application.
[0062] Figure 14 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0064] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0065] The integrated surgical guide and occluder design method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send acquired CBCT data and intraoral scan data to server 104. After receiving the CBCT data and intraoral scan data, server 104 registers the CBCT data and intraoral scan data to obtain a digital model of the cyst and adjacent anatomical structures. It then performs AI automated analysis on the digital model of the cyst and adjacent anatomical structures to obtain spatial features, geometric topology setting parameters, and trends in cyst healing and permanent tooth eruption. Based on the spatial features, geometric topology setting parameters, and trends in cyst healing and permanent tooth eruption, it uses 3D printing technology to parametrically design multiple integrated surgical guides and occluders. Server 104 can then feed back the obtained multiple integrated surgical guides and occluders to terminal 102. In addition, in some embodiments, the integrated surgical guide and occluder design method can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly design the integrated surgical guide and occluder based on CBCT data and intraoral scan data, or the server 104 can obtain CBCT data and intraoral scan data from the data storage system and design the integrated surgical guide and occluder based on the CBCT data and intraoral scan data.
[0066] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, and tablets. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers, or it can be a cloud server.
[0067] In one exemplary embodiment, such as Figure 2 As shown, an integrated surgical guide and occluder design method is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the following steps are included.
[0068] S1: Acquire CBCT data and intraoral scan data.
[0069] S2: Register the CBCT data and the intraoral scan data to obtain a digital model of the cyst and adjacent anatomical structures.
[0070] S3: Perform AI-automated analysis on the digital model of the cyst and adjacent anatomical structures to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption.
[0071] S4: Based on the spatial features, the geometric features, topological structure settings, and the trends of cyst healing and permanent tooth eruption, multiple integrated surgical guides and occluders are parametrically designed using 3D printing technology.
[0072] By implementing steps S1 to S4 above, this application systematically solves the technical bottlenecks of the three core aspects of surgical planning, intraoperative navigation, and postoperative drainage through AI-assisted multimodal image segmentation, intelligent guide plate design, and 3D printing for real-time manufacturing, significantly reducing the learning curve and demonstrating significant clinical application value.
[0073] As an optional implementation, in step S2, the CBCT data and the intraoral scan data are registered to obtain a digital model of the cyst and adjacent anatomical structures, which specifically includes the following steps.
[0074] S21: Preprocess the CBCT data to obtain preprocessed data; the preprocessing includes: foreground intensity value cropping and normalization.
[0075] S22: The preprocessed data is segmented using a CBCT segmentation model to obtain teeth, jawbone, and nerve canal; the CBCT segmentation model is a model trained based on the nnU-Net network.
[0076] S23: Using the jawbone as a template, segmentation is performed to obtain cystic lesions of the jawbone.
[0077] S24: Calculate the connected components of cystic lesions of the jawbone based on the 6-neighborhood property of voxel space, obtain the volume of the connected components by the summation method, and filter the connected components with volumes smaller than the preset value to obtain the filtered connected components.
[0078] S25: Sample the intraoral scanning data to obtain a point cloud.
[0079] S26: The point cloud is segmented using an oral scanning point cloud segmentation model to obtain a crown point cloud; the oral scanning point cloud segmentation model is a model trained based on a point cloud segmentation neural network.
[0080] S27: Convert the tooth into a curved surface model and sample it to obtain several sampling points.
[0081] S28: The RANSAC algorithm is used to register the sampling points with the crown point cloud to obtain the Euler transformation matrix.
[0082] S29: Apply the Euler transformation matrix to the intraoral scanning model to obtain a digital model of the cyst and adjacent anatomical structures.
[0083] Specifically, in step one, a multi-task deep learning framework based on the nnU-Net algorithm is designed to construct a segmentation model for important oral anatomical structures and cystic lesions of the jawbone. The model is then iteratively trained using a CBCT dataset annotated with ITK-SNAP to achieve accurate identification of lesions and key anatomical structures such as tooth germs, neural tubes, and jawbones.
[0084] The CBCT segmentation steps are shown below.
[0085] (1) Crop the foreground intensity value of the image to the range of [0.5, 99.5].
[0086] (2) Subsequently, based on the mean and standard deviation of the intensity values of the foreground of the entire dataset, z-socre is applied to normalize the intensity to the interval [0,1].
[0087] (3) Use the nnU-Net network to segment teeth, jawbone and nerve canal in CBCT. The segmentation model architecture is as follows: Figure 3 As shown.
[0088] (4) Using the nnU-Net network, the jawbone obtained in (3) is used as a mask to segment the cystic lesions of the jawbone.
[0089] (5) Calculate the connected domain of jaw cystic lesions based on the 6-neighborhood property of voxel space, and obtain the volume of the connected domain by the summation method, filtering out connected domains with a volume less than 1000.
[0090] Step 2: Based on the connected domains obtained in Step 1, fuse the intraoral scan data to establish a multimodal 3D model, breaking through the soft tissue resolution limitations of traditional single images.
[0091] The steps for fusing intraoral scan data with CBCT data are as follows.
[0092] (1) The intraoral scanning data was sampled into 16,000 points to form a point cloud.
[0093] (2) The point cloud is segmented using a point cloud segmentation neural network to obtain the point cloud of the tooth crown. The architecture of the oral scanning point cloud segmentation model is as follows: Figure 4 As shown.
[0094] (3) Use the Marching Cubes algorithm to convert the teeth segmented by CBCT into a surface model.
[0095] (4) Sample 2000 points on the surface model.
[0096] (5) Use the RANSAC algorithm to register the sampling points. And crown dot cloud This yields the Euler transformation matrix. The RANSAC algorithm can be expressed as follows.
[0097] .
[0098] in, For a pair of nearest neighbors; It is a rotation matrix; It is a translation matrix.
[0099] (6) Apply the Euler transformation matrix to the intraoral scanning model to obtain an intraoral scanning model aligned with CBCT (digital model of the cyst and adjacent anatomical structures). The final result is as follows: Figure 5 As shown, the unified model obtained provides a reference for subsequent analysis.
[0100] Step 3: Based on the digital model of the cyst and adjacent anatomical structures obtained in Step 2, AI-automated analysis is used to automatically analyze and extract the spatial features, geometric features, topological structure settings, and cyst healing and permanent tooth eruption trends of the lesion. The surgical guide and personalized implant are then parametrically designed using the Mimics / 3-matic platform, and applied intraoperatively using SLA / DLP light-curing 3D printing. The necessary equipment includes... Figure 6 As shown.
[0101] In summary, the three-dimensional digital treatment system obtained in this application includes a surgical guide model, an occluder model, and a printing parameter file.
[0102] As an optional implementation method, the determination of the CBCT segmentation model specifically includes the following steps.
[0103] S221: Obtain historical CBCT datasets.
[0104] S222: The historical CBCT dataset is annotated using ITK-SNAP to obtain an annotated historical dataset; the annotated historical dataset includes: teeth, jawbone, nerve canal and jawbone cystic lesions.
[0105] S223: Using the historical dataset as input and the labeled historical dataset as output, train the nnU-Net model to obtain the CBCT segmentation model.
[0106] As an optional implementation method, the determination of the point cloud segmentation model includes the following steps.
[0107] S261: Obtain historical intraoral scan data.
[0108] S262: Sample the historical intraoral scan data to obtain historical point cloud.
[0109] S263: Annotate the historical point cloud to obtain the annotated crown point cloud.
[0110] S264: Using historical point clouds as input and labeled crown point clouds as output, a point cloud segmentation neural network is trained to obtain an oral scan point cloud segmentation model.
[0111] As an optional implementation, in step S3, the digital model of the cyst and adjacent anatomical structures is subjected to AI automated analysis to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption. This specifically includes the following steps.
[0112] S31: Based on the neural tube and the filtered connected components, the digital model of the cyst and adjacent anatomical structures is labeled to obtain the labeled dataset; the labeling includes: the location of the permanent tooth germ, the boundary of the cyst, and the orientation of the neural tube.
[0113] S32: Based on the labeled dataset, an AI simulation algorithm is used to predict the output, and the spatial features, geometric features, topological structure setting parameters, cyst healing, and permanent tooth eruption trends are obtained.
[0114] The geometric feature topology setting parameters are as follows: the window opening is ≥1cm or more than 2 tooth positions, located on the labial / buccal side of the lesion closest to the alveolar ridge, the vertical depth of the occluder should be less than the vertical distance from the normal alveolar ridge plane to the bottom of the cyst, located on the labial / buccal side or above the permanent tooth germ, the base plate occluder should avoid buccal retention devices, and the lingual / palatal base plate should extend above the central part of the abutment tooth.
[0115] This application specifically relates to a sequential treatment system and method for pediatric jaw cysts based on multimodal image fusion, AI simulation prediction, and 3D printing technology, used to optimize occluder fitting and prognostic management after cyst fenestration decompression surgery. A comparative diagram of the traditional approach and the proposed approach is shown below. Figure 7 As shown, fenestration decompression surgery is the preferred treatment for pediatric jaw cysts, its core advantage being the preservation of jaw function and developmental potential through minimally invasive intervention. However, the traditional treatment process involves multidisciplinary collaboration, requiring multiple visits for the child and parents, with procedures such as fenestration surgery, occluder fabrication and placement, and prognosis observation performed in different departments, resulting in a long treatment cycle and limited clinical application. This application aims to combine the diagnosis, treatment, and prognosis processes, with the ultimate goal of integrated treatment throughout the entire process. Deep learning algorithms will be used to segment the cyst area and surrounding important anatomical structures, establishing a multimodal model. AI parametric design combined with 3D printing technology will enable the pre-operative fabrication of the surgical guide and occluder. Furthermore, by enhancing the integrated adaptability of both, different models of occluders will be pre-made to suit the child's prognosis. This optimizes the treatment process while reducing surgical difficulty, improving surgical precision, and reducing the difficulty of patient cooperation, promoting the intelligent, precise, and standardized treatment of pediatric jaw cysts, and forming a standardized and scalable clinical treatment protocol.
[0116] Specifically, the technical objective of this application is to achieve "one surgery, dynamic adaptation" through preoperative AI simulation and serialized occluder design, thereby reducing the number of follow-up visits and ensuring therapeutic efficacy.
[0117] The technical solution is summarized below.
[0118] Step 1: Multimodal data fusion.
[0119] Register CBCT (3D bone tissue imaging) and oral scan data (soft tissue morphology) to construct a digital model of the cyst and adjacent anatomical structures.
[0120] Specifically, CBCT scan parameters (0.25mm voxel, covering the cyst area and at least two surrounding teeth); intraoral scan data must include the soft tissue contour of the fenestrated surgical area. Registration is performed to obtain a multimodal model integrating soft tissue and bone tissue (a digital model of the cyst and adjacent anatomical structures).
[0121] Step 2: AI simulation and surgical planning.
[0122] Deep learning algorithms were used to predict the shrinkage trend of cysts and the eruption path of permanent teeth, optimize the fenestration position, and generate surgical guide design parameters.
[0123] Specifically, the training dataset consists of a CBCT image library that labels the location of the permanent tooth germ, the boundary of the cyst, and the direction of the neural tube; the predicted output includes the cyst volume shrinkage rate and shrinkage dimension, and the permanent tooth displacement vector (e.g., how many mm it erupts per month and in which direction).
[0124] Step 3: Sequence Blocker Design and Manufacturing.
[0125] Based on the simulation results, multiple sequences of occluders (such as initial large-volume occluders → mid-term adaptable occluders → late-term small occluders) were parametrically designed to cover different healing stages.
[0126] Surgical guides and occluder magazines are manufactured simultaneously using 3D printing technology to ensure mechanical compatibility between instruments.
[0127] Specifically, the initial blocker has a thickness of ≥2mm and a pre-reserved decompression channel; the final blocker has a thickness of ≥2mm and its margins avoid the eruption area of the permanent tooth (distance ≥2mm).
[0128] Step 4: Dynamic management of follow-up visits.
[0129] During follow-up visits, the system automatically recommends a suitable occluder model through image comparison (such as overlaying pre-operative or previous CBCT with the current CBCT), which is then confirmed by the doctor before being worn.
[0130] Specifically, imaging and intraoral scanning data acquisition and guide plate and occluder fabrication are completed one week before the operation; follow-up visits are conducted at 1, 2, 3, 6, 9 and 12 months after the operation, and the timing of occluder replacement is automatically prompted by image AI comparison (such as when the cyst cavity shrinks by 50% or the permanent tooth moves more than 2mm).
[0131] The innovative aspects and beneficial effects of this application are as follows.
[0132] 1. Innovation points.
[0133] Dynamic sequence design: For the first time, the cyst healing process is quantified and pre-programmed into a multi-stage occluder, replacing the traditional "trial and error" adjustment.
[0134] AI-driven prognosis: By simulating the biomechanical relationship between permanent tooth eruption and cyst shrinkage, anatomical risks can be proactively avoided.
[0135] like Figure 8 As shown, the innovation of this application is reflected in the following three aspects.
[0136] Process Innovation: For the first time, the integrated design of the entire process is applied to the diagnosis and treatment of pediatric jaw cysts. With the assistance of AI, the integrated design of precise diagnosis and treatment throughout the entire process of "preoperative planning - intraoperative navigation - postoperative drainage" is realized.
[0137] Algorithm innovation: The latest nnU-Net algorithm is applied, and a multi-stage, multi-task framework is used for the first time in the intelligent diagnosis and treatment of pediatric jaw cysts.
[0138] Technological innovation: Time series modeling is introduced to enable AI to predict cyst volume changes, and based on this, an automated design system for an integrated surgical guide / occluder is innovatively designed.
[0139] 2. Beneficial effects.
[0140] The number of follow-up visits has decreased by more than 50% (from an average of 4 times to 2 times).
[0141] The risk of permanent tooth damage is reduced by 10% (through AI path prediction).
[0142] Improved patient comfort (avoiding repeated mold taking).
[0143] The following example, a patient with a jaw cyst, is used to illustrate this application.
[0144] ① Perform an oral scan and take a CBCT scan, then register the CBCT (three-dimensional bone tissue image) and the oral scan data (soft tissue morphology) to construct a digital model of the cyst and adjacent anatomical structures.
[0145] like Figure 9 As shown, a highly adaptive medical image segmentation algorithm called nn-Unet is presented. Based on this algorithm, this embodiment uses the CBCT dataset annotated with ITK-SNAP for iterative training. For the complex oral and maxillofacial anatomy of children in the mixed dentition stage, this embodiment formulates a multi-task cascaded nnU-Net framework: the permanent tooth germ, deciduous teeth, neural tube and other anatomical structures at different developmental stages are segmented at the same time. First, the jawbone region is located by coarse segmentation, and then the high-resolution sub-network is used to complete the accurate recognition.
[0146] Because the clarity of CBCT images cannot meet the requirements for precise digital design of surgical guides and occluders, this embodiment attempts to register the intraoral scan model with the dental crowns, thereby combining complex bone tissue information with precise crown and soft tissue information, such as... Figure 10 As shown.
[0147] like Figure 11 As shown, step 1: Through the multi-task cascaded nnU-Net framework, the automatic labeling and segmentation of tooth crowns on the oral scanning model is completed using the dataset.
[0148] Step 2: Replace and dynamically register the crown data of the CBCT segmentation model with the crown data of the intraoral scan model.
[0149] Step 3: Combine the CBCT data of the jawbone, lesions, and tooth roots with the results obtained in step 2 to export a multimodal image that integrates bone and soft tissue information.
[0150] ②Automatically analyze and extract spatial features, and parametrically design an integrated surgical guide and occluder.
[0151] By using AI to automatically analyze and extract the geometric features and topological structure of multimodal images, and combining it with 3D printing technology, the automatic design and fabrication of an integrated surgical guide and sequential occluder can be achieved by setting the following parameters.
[0152] like Figure 12 As shown, the geometric feature topology settings parameters are as follows: the opening window is greater than or equal to 1 cm or greater than 2 tooth positions, located on the labial / buccal side of the lesion closest to the alveolar ridge, the vertical depth of the occluder should be less than the vertical distance from the normal alveolar ridge plane to the bottom of the cyst, located on the labial / buccal side or above the permanent tooth germ, the base plate occluder should avoid buccal retention devices, and the lingual / palatal base plate should extend above the central part of the abutment tooth.
[0153] The model was used to simulate the healing of the cyst and the eruption of the permanent tooth. Multiple sequential occluders were designed before the operation and worn by the doctor after examination during the patient's follow-up visit to avoid the eruption channel of the permanent tooth.
[0154] ③ The patient undergoes surgery according to the guide plate, and the initial occluder is worn immediately after the operation.
[0155] ④ The patient returns for a follow-up visit one month later, and a CBCT scan is taken. This scan is then time-coordinated with the preoperative CBCT scan to determine the volume and dimensions of the cyst reduction. For example, it can be determined which dimension (top, bottom, front, back, left, right) the reduction occurred in, by how many millimeters in each dimension, the total volume reduction, and in which dimension the permanent tooth contained within the cyst was moved. When the cyst volume shrinks by more than 50% or the permanent tooth moves by more than 2mm, the occluder replacement mechanism is triggered. An appropriate occluder model (intermediate occluder) is automatically recommended, and the occluder is fitted after the doctor's confirmation.
[0156] like Figure 13 As shown in Figure A: In the visualization, the cyst appears grayish-blue preoperatively and red postoperatively. The jawbone appears grayish-white preoperatively and grayish-green postoperatively, visually demonstrating the process of cyst shrinkage and jawbone remodeling. Figure B: Example of a three-dimensional structural integration model of preoperative and postoperative CBCT images of children. The gray areas in the figure represent the jawbone structure, and the colored areas represent the teeth and cyst regions, respectively. The top row is the maxillary model, with different colors corresponding to different teeth and cysts; the bottom row is the mandibular model, showing the spatial relationship between teeth, cysts, and jawbone.
[0157] The results of several pre- and post-operative follow-up examinations show that the cyst has shrunk and the permanent tooth is moving.
[0158] ⑤ The patient returns for a follow-up visit 2 months later. The process is the same as above. If the volume has shrunk by more than 50% or the permanent tooth has moved by more than 2mm compared to the previous visit, the occluder replacement mechanism is triggered. The system automatically recommends a suitable occluder model (end-stage occluder), which is then fitted after confirmation by the dentist.
[0159] ⑥ Patients should return for follow-up visits at 3, 6, 9, and 12 months, following the same procedure. If the volume shrinks again or the cyst closes, the obstructor should be removed, and the eruption of the permanent tooth should be observed. The entire treatment process is then complete.
[0160] This application focuses on AI-assisted design, integrating CBCT and intraoral scan data. Through deep learning algorithms, it achieves automatic segmentation and 3D reconstruction of the cyst region and adjacent key anatomical structures such as the permanent tooth germ and neural canal. Personalized surgical guides and multiple sequential occluders are simultaneously manufactured using AI parametric design and 3D printing technology. By enhancing the integration and adaptability of these components, different models of occluders are pre-fabricated to suit the child's prognosis, reducing surgical difficulty, optimizing surgical precision, shortening the treatment cycle, and reducing the difficulty of patient cooperation. This promotes innovation in the diagnosis and treatment of pediatric jaw cysts towards intelligence, precision, and standardization, forming a standardized and scalable clinical treatment plan.
[0161] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 14 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores CBCT data and intraoral scan data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements an integrated surgical guide and occluder design method.
[0162] Those skilled in the art will understand that Figure 14The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0163] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method embodiments.
[0164] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the above-described method embodiments.
[0165] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described method embodiments.
[0166] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0167] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0168] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0169] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0170] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A design method for an integrated surgical guide and occluder, characterized in that, The integrated surgical guide and occluder design method includes: Acquire CBCT data and intraoral scan data; The CBCT data and the intraoral scan data are registered to obtain a digital model of the cyst and adjacent anatomical structures. AI-automated analysis was performed on the digital model of the cyst and adjacent anatomical structures to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption. Based on the spatial characteristics, the geometric topology settings, and the trends of cyst healing and permanent tooth eruption, multiple integrated surgical guides and occluders are parametrically designed using 3D printing technology.
2. The integrated surgical guide and occluder design method according to claim 1, characterized in that, The CBCT data and the intraoral scan data are registered to obtain a digital model of the cyst and adjacent anatomical structures, specifically including: The CBCT data is preprocessed to obtain preprocessed data; the preprocessing includes: foreground intensity value cropping and normalization; The preprocessed data was segmented using a CBCT segmentation model to obtain teeth, jawbones, and nerve canals; the CBCT segmentation model was a model trained based on the nnU-Net network. Using the jawbone as a template, segmentation was performed to obtain cystic lesions of the jawbone; The connected components of cystic lesions of the jawbone are calculated based on the 6-neighborhood property of voxel space. The volume of the connected components is obtained by the summation method, and connected components with a volume smaller than a preset value are filtered to obtain the filtered connected components. The intraoral scan data is sampled to obtain a point cloud; The point cloud is segmented using an intraoral scanning point cloud segmentation model to obtain a crown point cloud; the intraoral scanning point cloud segmentation model is a model trained based on a point cloud segmentation neural network. The teeth are converted into a curved surface model and sampled to obtain several sampling points; The RANSAC algorithm is used to register the sampling points with the crown point cloud to obtain the Euler transform matrix; The Euler transformation matrix is applied to the intraoral scanning model to obtain a digital model of the cyst and adjacent anatomical structures.
3. The integrated surgical guide and occluder design method according to claim 2, characterized in that, The determination of the CBCT segmentation model specifically includes: Obtain historical CBCT datasets; The historical CBCT dataset was annotated using ITK-SNAP to obtain an annotated historical dataset; the annotated historical dataset includes: teeth, jawbone, neural canal and jawbone cystic lesions; Using historical datasets as input and labeled historical datasets as output, the nnU-Net model is trained to obtain the CBCT segmentation model.
4. The integrated surgical guide and occluder design method according to claim 2, characterized in that, The determination of the point cloud segmentation model specifically includes: Acquire historical intraoral scan data; The historical intraoral scan data is sampled to obtain historical point clouds; The historical point cloud is annotated to obtain the annotated crown point cloud; Using historical point clouds as input and labeled crown point clouds as output, a point cloud segmentation neural network is trained to obtain an oral scan point cloud segmentation model.
5. The integrated surgical guide and occluder design method according to claim 2, characterized in that, The expression for the RANSAC algorithm is: ； in, p For sampling points; q Add cloud-like dots to the dental crown; For a pair of nearest neighbors; It is a rotation matrix; It is a translation matrix.
6. The integrated surgical guide and occluder design method according to claim 2, characterized in that, AI-automated analysis was performed on the digital model of the cyst and adjacent anatomical structures to obtain spatial features, geometric features, topological structure setting parameters, and trends in cyst healing and permanent tooth eruption, specifically including: The digital model of the cyst and adjacent anatomical structures is labeled based on the neural tube and the filtered connected components to obtain a labeled dataset; the labeling includes: the location of the permanent tooth germ, the boundary of the cyst, and the orientation of the neural tube. Based on the labeled dataset, an AI simulation algorithm is used to predict the output, obtaining spatial features, geometric features, topological structure setting parameters, cyst healing, and permanent tooth eruption trends.
7. The integrated surgical guide and occluder design method according to claim 6, characterized in that, The geometric feature topology setting parameters are as follows: the opening window is greater than or equal to 1 cm or greater than 2 tooth positions, located on the labial and buccal side of the lesion closest to the alveolar ridge, the vertical depth of the occluder should be less than the vertical distance from the normal alveolar ridge plane to the bottom of the cyst, located on the labial and buccal side or above the permanent tooth germ, the base plate occluder should avoid buccal retention devices, and the lingual / palatal base plate should extend above the central part of the abutment tooth.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the integrated surgical guide and occluder design method according to any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the integrated surgical guide and occluder design method as described in any one of claims 1-7.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the integrated surgical guide and occluder design method as described in any one of claims 1-7.