A digital guide plate preparation method and system for extracting completely impacted teeth

Abstract

The present application relates to a method and system for preparing a digital guide plate for extracting completely impacted teeth, which solves the problem that the existing digital guide plate preparation methods are mostly based on universal design parameters and fail to fully consider the individual differences of patients, resulting in limited applicability and accuracy of the guide plate. The method includes: performing in-depth analysis on the data after three-dimensional registration and fitting to obtain basic information of the patient's oral and maxillofacial regions; using the obtained basic information of the patient's oral and maxillofacial regions as input objects, inputting them into a trained deep neural network model for predicting optimal guide plate design parameters, and outputting the predicted optimal guide plate design parameters as the guide plate design parameters for actual application; printing out the guide plate according to the guide plate design parameters for actual application. The present application has the following effects: improving the accuracy of guide plate preparation and surgical efficiency.

Description

Technical Field

[0001] The invention relates to the field of stomatology and digital medical technology, and in particular to a method and system for preparing a digital guide plate for extracting completely impacted teeth. Background Art

[0002] In recent years, due to economic development and improved living standards, the number of people seeking medical treatment for oral health problems has increased. With the popularization of oral imaging equipment (such as panoramic machines, cone beam CT, etc.), various completely impacted teeth (except third molars) are becoming more and more common. Because completely impacted teeth occupy the normal dentition space, they often cause the establishment of normal occlusal relationship, displacement of adjacent teeth, external root resorption or development of dental cysts, which seriously affects the patient's oral function and aesthetics. Therefore, this type of completely impacted teeth should be extracted as soon as possible.

[0003] At present, the three-dimensional position of completely impacted teeth is mainly determined clinically by cone beam CT. Clinicians transfer the information seen on cone beam CT directly to the patient's mouth without using an intermediate medium, relying only on their experience, which can easily cause position deviation of completely impacted teeth. In actual surgery, it is often impossible to find completely impacted teeth in a short period of time, resulting in too large a flap, too much bone removal, damage to adjacent teeth and other important anatomical structures, and long tooth extraction time, which are difficult for patients to accept.

[0004] With the development of digital technology, digital technology has gradually been introduced into the preparation of surgical guides. Existing digital guide preparation methods are mostly based on universal design parameters and fail to fully consider the individual differences of patients, resulting in limited applicability and accuracy of guides. Summary of the invention

[0005] In order to improve the accuracy of guide plate preparation and surgical efficiency, the present application provides a digital guide plate preparation method and system for extracting completely impacted teeth.

[0006] In a first aspect, the present application provides a method for preparing a digital guide plate for extracting a completely impacted tooth, using the following technical solution:

[0007] A method for preparing a digital guide plate for extracting a completely impacted tooth, comprising:

[0008] Obtain the patient's maxillofacial continuous scanning data and the digital tooth model data of the upper and lower mandibular hard plaster models;

[0009] Perform three-dimensional registration and fitting of the digital tooth model data and the maxillofacial continuous scanning data;

[0010] Performing in-depth analysis on the data after 3D registration fitting to obtain basic information of the patient's oral and maxillofacial regions, including but not limited to anatomical structure information, bone condition information, tooth position information, and neurovascular information;

[0011] The basic information of the patient's oral and maxillofacial regions is used as input to a trained deep neural network model for predicting optimal guide plate design parameters, and the predicted optimal guide plate design parameters are output as guide plate design parameters for actual application;

[0012] The guide plate is printed out according to the guide plate design parameters of actual application, wherein the guide plate includes a tooth-supported retention plate, a surgical incision guide ring, a positioning and depth guide ring, and a tooth separation positioning guide ring.

[0013] By adopting the above technical solution, the digital guide preparation method for extracting completely impacted teeth integrates the continuous scanning data of the maxillofacial region with the digital tooth model to perform three-dimensional precise registration, deeply analyze the anatomical structure, bone, tooth position and neurovascular information, and provide comprehensive data support for personalized guide design. The deep neural network model is used to predict the optimal guide design parameters to ensure that the guide design accurately matches the patient's oral and maxillofacial features and improve surgical accuracy. The printed guide, including the tooth-supported retention plate, surgical incision guide ring, positioning and depth guide ring and tooth separation positioning guide ring, effectively guides the surgical path, reduces surgical risks, shortens the operation time, improves patient comfort, and realizes personalized, safe and efficient tooth extraction surgery. At the same time, it provides doctors with intuitive and accurate surgical guidance to improve the success rate of the operation.

[0014] Optionally, a trained deep neural network model for predicting optimal guide plate design parameters is obtained including:

[0015] Obtain basic information of oral and maxillofacial areas of historical patients and guide design parameters;

[0016] Perform data preprocessing on the basic information of the patient's oral and maxillofacial areas and the guide plate design parameters, including outlier removal, normalization, and feature selection;

[0017] Input the preprocessed data into the deep neural network model of the preset architecture, and complete the initialization setting of the deep neural network model, including the initial value setting of weights and biases;

[0018] According to the preset learning rate, number of iterations and L2 regularization coefficient of the deep neural network model and the deep neural network model with initialization settings, the deep neural network model is offline trained using historical data, wherein the training set and the validation set of the offline training are divided according to a preset ratio;

[0019] When the prediction error rate of the trained deep neural network model on the validation set is lower than the preset error rate, the deep neural network model is confirmed to have completed training, and the confirmed deep neural network model is used as a trained deep neural network model for predicting the optimal guide plate design parameters.

[0020] By adopting the above technical solution, the deep neural network model collects basic information of oral and maxillofacial areas of historical patients and guide design parameters, and ensures data quality through data preprocessing and feature selection. After the model is initialized, it is trained offline using historical data, and the model performance is optimized by adjusting the learning rate, number of iterations and L2 regularization coefficient. When the prediction error rate of the model on the validation set is lower than the preset value, the model training is confirmed to be completed. This model can accurately predict the optimal guide design parameters, improve surgical accuracy, reduce surgical risks, enhance patient comfort, and realize personalized, safe and efficient surgical guide design.

[0021] Optionally, the method further includes steps after outputting the predicted optimal guide plate design parameters and before printing out the guide plate according to the guide plate design parameters actually applied, specifically as follows:

[0022] Process the data collected from the basic information of the patient's oral and maxillofacial areas, including data cleaning, data segmentation, and data labeling;

[0023] The model is established based on the basic information of the patient's oral and maxillofacial regions after data processing;

[0024] Convert the established model and input guide parameters into finite element mesh, apply boundary conditions and loads, and simulate various operations during surgery;

[0025] By solving the finite element equation, the result of finite element analysis is obtained, wherein the result of finite element analysis includes stress analysis result, strain analysis result and displacement analysis result;

[0026] Analyze whether the key indicator values ​​contained in the results of the finite element analysis exceed the thresholds of the corresponding key indicators;

[0027] If no, continue with the next steps;

[0028] If yes, the key indicator that exceeds the threshold of the corresponding key indicator is taken as the indicator to be optimized, and the difference between the value of the indicator to be optimized and the threshold of the corresponding key indicator is analyzed to see whether it falls within the difference threshold interval;

[0029] If not, it is determined that the design parameters of the guide plate actually used have obvious deviations, and the subsequent steps are stopped;

[0030] If yes, then according to the correspondence between the difference interval range in which the difference between the value of the indicator to be optimized and the threshold value of the corresponding key indicator falls and the adjustment range schemes of various guide plate associated parameters, analyze and determine the adjustment range schemes of various guide plate associated parameters, and execute the adjustment range schemes of various guide plate associated parameters;

[0031] The guide plate parameters after executing the adjustment range schemes of various guide plate associated parameters are used as input guide plate parameters, and the step of converting the established model and the input guide plate parameters into a finite element mesh is jumped to be executed;

[0032] If and only if the key indicator values ​​contained in the results of the finite element analysis do not exceed the threshold values ​​of the corresponding key indicators, continue with the subsequent steps; otherwise, continue to execute the adjustment range plan of various guide plate associated parameters.

[0033] By adopting the above technical solution, the finite element analysis is used to optimize the design of tooth extraction surgical guides, which can effectively evaluate the key indicators in the surgical process and ensure that they are within the safety threshold. If the key indicator exceeds the threshold, the relationship between the difference and the threshold is analyzed, and the associated parameters of the guide are intelligently adjusted until the key indicator meets the requirements. This method significantly improves the accuracy and safety of guide design, reduces surgical risks, and realizes personalized surgical plans. In the design of guides, finite element analysis approximates the real physical system through mathematical simulation methods, uses simple and interacting elements (units) to approximate the real system of infinite unknown quantities, decomposes the solution domain into a finite number of subdomains, and assumes an approximate solution for each subdomain, thereby obtaining a solution for the entire domain. Although this method provides an approximate solution, it can effectively predict the physical behavior of complex systems and structures, helping engineers make more informed design decisions, especially in the medical field, such as the design of tooth extraction surgical guides, which can significantly improve the accuracy and safety of surgery, ensure that the surgical process is more controllable, reduce complications, and improve patient satisfaction.

[0034] Optionally, the method further includes a step of continuing the subsequent steps if and only if the key indicator value included in the result of the finite element analysis does not exceed the threshold value of the corresponding key indicator, as follows:

[0035] Analyze whether the number of executions of adjustment range schemes of various guide plate associated parameters exceeds a preset threshold number;

[0036] If yes, then stop the subsequent adjustments, import the most recently adjusted guide plate parameters into the virtual reality or augmented system, and create a virtual surgical scene that matches the actual patient situation for the doctor to simulate the surgery and adjust the guide plate parameters;

[0037] If and only if the doctor's simulated surgery process conforms to the preset process and the simulated surgery is successful, the adjusted guide plate parameters will be used as the guide plate design parameters for this actual application.

[0038] By adopting the above technical solutions, optimizing the guide design parameters through finite element analysis, and combining with virtual reality or augmented reality systems, doctors can simulate surgery in a safe virtual environment and evaluate the adjusted guide parameters. If the number of adjustments exceeds the preset threshold, the system automatically imports the latest parameters and creates a virtual scene that matches the patient. The doctor simulates the surgery to ensure that the process meets the preset standards and the surgery is successful, and finally determines the guide design parameters. This process improves the efficiency of surgical preparation, enhances surgical safety, improves the doctor's proficiency, ensures the accurate implementation of the surgical plan, and reduces the surgical risks of patients to achieve better surgical results.

[0039] Optionally, create a virtual surgical scene that matches the patient's actual situation for the doctor to simulate the surgery, including:

[0040] Obtain relevant data on patient-specific factors and pre-process the data, wherein the patient-specific factors include but are not limited to overall health status, age, gender, and lifestyle habits;

[0041] Based on the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial area, a personalized oral and maxillofacial model is constructed using biomechanical and biophysical simulation techniques.

[0042] By adopting the above technical solutions, using biomechanical and biophysical simulation technologies, combined with patient-specific factors such as overall health status, age, gender, living habits, etc., a personalized oral and maxillofacial model is constructed to provide doctors with a virtual surgical scene that matches the actual situation of the patient. This not only enhances the personalization and accuracy of the surgery, but also allows doctors to rehearse the surgery in a virtual environment, assess risks, and optimize the surgical plan, thereby improving the success rate of the surgery, reducing complications, and enhancing the patient's treatment experience and safety.

[0043] Optional parameters for the doctor to adjust the guide include:

[0044] Obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, and obtain force-related results based on detailed data analysis, where the detailed data include but are not limited to stress distribution, strain, displacement, and force-related results include force characteristics and deformation trends;

[0045] Input the design parameters of the current guide plate into the simulation system, compare the interaction forces between the guide plate and tissues and bones in the simulation results, and determine the problems existing in the current guide plate according to the preset judgment range of the problem-related parameters, where the problems include local stress concentration, excessive compression of tissues, and mismatch with anatomical structures;

[0046] Based on the problems existing in the current guide plate and the corresponding relationship between the problems and the optimization direction, the optimization direction is formed and provided as a reference for the doctor to adjust the guide plate parameters.

[0047] By adopting the above technical solutions, the simulation system can obtain detailed data on the impact of patient-specific factors on tissues and bones, and analyze force-related results such as stress distribution, strain and displacement. Doctors can accurately identify problems in guide design, such as local stress concentration, excessive compression of tissues or mismatch with anatomical structures. Based on the correspondence between problems and optimization directions, the system provides optimization direction reference prompts to guide doctors to adjust guide parameters to ensure that the guide design is more in line with patient needs, improve surgical safety and success rate, reduce surgical risks, and improve patient treatment effects and satisfaction.

[0048] Optionally, the doctor simulates the surgery in accordance with the preset process including:

[0049] Obtain the accuracy of the doctor's surgical simulation path and surgical operation, and analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and analyze whether the values ​​of surgical operation-related parameters are all within the preset parameter value range and whether the damage degree values ​​of key structures during the operation fall within the preset damage degree value range;

[0050] The doctor's simulated surgery process is judged to be in accordance with the preset process if and only if the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and the values ​​of the surgical operation-related parameters fall within the preset parameter value range and the degree of damage to the key structure during the operation falls within the preset damage degree value range.

[0051] By adopting the above technical solution, the deviation between the doctor's surgical simulation path and the preset planning path, as well as the surgical operation parameters and the degree of damage to key structures, is accurately analyzed to ensure that the simulated surgical process is consistent with the preset standards. This method effectively verifies the feasibility of the surgical plan, improves surgical accuracy, reduces surgical risks, ensures patient safety, and improves the doctor's surgical skills to ensure that the surgical effect achieves the expected goal.

[0052] Optional, successful simulated surgery includes:

[0053] Obtain the result values ​​of the simulated surgery with respect to different set targets, including the surgical effect, surgical time and efficiency;

[0054] Analyze whether the result values ​​of the simulated surgery with respect to different set targets fall within the reasonable result value range of the corresponding set targets;

[0055] If yes, the simulated surgery is judged to be successful;

[0056] If not, the simulated operation is judged to have failed.

[0057] By adopting the above technical solution, the result values ​​of the simulated surgery on the set goals such as surgical effect, surgical time and efficiency are analyzed to determine whether they meet the preset reasonable range. When all target result values ​​are met, the simulated surgery is confirmed to be successful, which shows that the surgical plan is effective, efficient, and the risk is controllable, providing reliable guidance for the actual surgery; conversely, if any goal is not met, the surgical plan needs to be adjusted until the simulated surgery is successful to ensure the smooth progress of the actual surgery and the safety of the patient.

[0058] In a second aspect, the present application provides a digital guide plate preparation system for extracting completely impacted teeth, which adopts the following technical solution:

[0059] A digital guide plate preparation system for extracting completely impacted teeth comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement the digital guide plate preparation method for extracting completely impacted teeth as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the overall process of a method for preparing a digital guide plate for extracting a completely impacted tooth according to an embodiment of the present application.

[0061] Figure 2 It is a flow chart of obtaining a deep neural network model that has been trained for predicting optimal guide plate design parameters in an embodiment of the present application.

[0062] Figure 3 It is a flowchart diagram of the steps after outputting the predicted optimal guide plate design parameters and before printing out the guide plate according to the guide plate design parameters actually applied in another embodiment of the present application.

[0063] Figure 4 It is a flowchart diagram of an embodiment of the present application and the steps of continuing the subsequent steps in parallel if and only if the key indicator values ​​included in the results of the finite element analysis do not exceed the threshold values ​​of the corresponding key indicators. DETAILED DESCRIPTION

[0064] The present application is further described in detail below in conjunction with the accompanying drawings.

[0065] Reference Figure 1 , a digital guide plate preparation method for extracting completely impacted teeth disclosed in the present application, comprising:

[0066] Step S100, obtaining the patient's maxillofacial continuous scanning data and the digitized tooth model data of the upper and lower jaw hard plaster models.

[0067] Among them, the acquisition of the patient's maxillofacial continuous scanning data is as follows: First, the patient's maxillofacial region is continuously scanned by a high-precision three-dimensional scanning device, such as CBCT (cone beam computed tomography) or a laser scanner, to obtain three-dimensional image data of the maxillofacial region. These data contain detailed information on structures such as the jaw, teeth, soft tissue, and possible impacted teeth. Specifically, the CBCT (cone beam computed tomography) involved in this application can use the New Tom Giano (Italy) three-dimensional scanning device, whose parameters are set to 120V and 60mA, and the layer thickness is 0.3mm, which can provide high-resolution maxillofacial three-dimensional image data. Through this scan, DICOM data of the upper and lower jaws can be obtained, which contains detailed information on structures such as the jaw, teeth, soft tissue, and possible impacted teeth.

[0068] The digital tooth model data of the upper and lower mandibular anhydrite models are obtained as follows: the upper and lower mandibular anhydrite models of the patient are made, and then these models are scanned using a 3D scanning device to obtain digital tooth model data. These data accurately reflect the morphology, position and arrangement of the patient's teeth. Specifically, a warehouse scanner (Xianlin, China) can be used to scan the upper and lower mandibular anhydrite models of the patient. The warehouse scanner is a high-precision 3D scanning device that can accurately scan the anhydrite model and obtain digital tooth model data. After scanning, high-precision digital tooth models can be obtained by trimming through software. These model data can be stored in STL file format for subsequent 3D registration fitting and in-depth analysis.

[0069] Step S200, performing three-dimensional registration fitting on the digitized tooth model data and the maxillofacial continuous scanning data.

[0070] Specifically, the implementation process of step S200 is as follows: by using professional 3D registration software, the digitized tooth model data is compared and adjusted with the maxillofacial continuous scanning data, so that the two can achieve the best matching state in the 3D space. In this application, the 3D registration fitting can be processed by TRIOS Design Studio (3shape, Denmark) software. The digitized tooth model STL file and the CBCT DICOM file are imported into the software TRIOS Design Studio (3shape, Denmark) to achieve 3D data registration fitting.

[0071] Step S300, performing in-depth analysis on the data after three-dimensional registration fitting to obtain basic information of the patient's oral and maxillofacial regions.

[0072] The basic information includes but is not limited to anatomical structure information, bone condition information, tooth position information, and neurovascular information.

[0073] The implementation process of deep analysis is as follows: Based on the data after 3D registration fitting, the implementation of deep analysis usually involves the use of professional 3D visualization and analysis software. These software provide various tools, such as measurement tools, segmentation tools, and visualization tools, to help analysts accurately identify and quantify anatomical structures, bone density, tooth position, and the distribution of nerves and blood vessels.

[0074] The analysis of different basic information is as follows: 1. Regarding anatomical structure information, deep analysis can accurately identify and measure the anatomical structure of the maxillofacial region, including teeth, jaws, soft tissues, etc., provide detailed anatomical structure information, and provide accurate reference for guide design. 2. Regarding bone condition information, through deep analysis, the density, thickness, and quality of the bone can be evaluated, which is crucial for guide design and surgical planning to ensure the safety and success rate of the operation. 3. Regarding tooth position information, deep analysis can accurately determine the position of the teeth, including the specific position and direction of impacted teeth, providing key information for guide design and ensuring the accuracy of the operation. 4. Regarding neurovascular information, deep analysis can identify the distribution and direction of neurovascular vessels, avoid damage to important neurovascular vessels during surgery, and improve the safety of surgery.

[0075] Step S400, taking the acquired basic information of the patient's oral and maxillofacial areas as input objects, inputting them into a trained deep neural network model for predicting optimal guide plate design parameters, and outputting the predicted optimal guide plate design parameters as guide plate design parameters for actual application.

[0076] Regarding the training of the deep neural network model, before step S400, the deep neural network model has been fully trained to predict the optimal guide plate design parameters. The training data set usually includes a large amount of known basic information of the patient's oral and maxillofacial regions, and the corresponding guide plate design parameters. Through training, the model learns the complex relationship between the basic information and the guide plate design parameters, and can predict the optimal guide plate design parameters based on the input basic information.

[0077] For example, in step S300, detailed basic information of the patient's oral and maxillofacial areas has been obtained, including the precise location of the anatomical structure, the density and quality of the bone, the specific location of the teeth, and the distribution of nerves and blood vessels. In step S400, these basic information are used as input objects and input into the deep neural network model that has been trained. Based on the input basic information, the model predicts the optimal guide design parameters, including the size, shape, position, angle, etc. of the guide through calculations of a multi-layer neural network. These parameters ensure that the guide design is highly matched to the actual situation of the patient, improving the accuracy and safety of the surgery. The predicted guide design parameters will be used as the guide design parameters for actual application, and will be used for printing the guide in step S500 to ensure the personalization and accuracy of the guide design.

[0078] Step S500: Printing the guide plate according to the guide plate design parameters actually applied.

[0079] The guide plate includes a tooth-supported retention plate, a surgical incision guide ring, a positioning depth guide ring, and a tooth separation positioning guide ring. Specifically, the tooth-supported retention plate is used to fix the position of the guide plate in the oral cavity to ensure the stability of the guide plate; the surgical incision guide ring is used to guide the position and direction of the surgical incision to ensure the accuracy of the operation; the positioning depth guide ring is used to control the depth of the surgical instrument to avoid damage to the surrounding tissues; the tooth separation positioning guide ring is used to accurately position the teeth to ensure the accuracy and safety of the operation.

[0080] The printing technology of the guide plate usually adopts 3D printing technology, which is a fast and precise manufacturing method that can quickly generate complex guide plate structures according to the design parameters. 3D printing technology can use a variety of materials, such as plastic, metal or biocompatible materials, to meet different surgical needs. Regarding the material of the guide plate, resin can be selected.

[0081] exist Figure 1 In step S400, further consideration is given to how to construct a deep neural network model for predicting the optimal guide plate design parameters. The specific construction process is referred to Figure 2 The embodiment shown.

[0082] Reference Figure 2 The deep neural network model that has been trained to predict the optimal guide plate design parameters includes:

[0083] Step SA00, obtaining basic information of the oral and maxillofacial regions of historical patients and guide plate design parameters.

[0084] Specifically, in step SA00, a large amount of basic information about the oral and maxillofacial regions of historical patients needs to be collected, including anatomical structure information, bone condition information, tooth position information, and neurovascular information. This information usually comes from the patient's maxillofacial continuous scan data and the digital tooth model data of the upper and lower mandibular hard plaster models. At the same time, the corresponding guide plate design parameters need to be collected, including the size, shape, position, angle, etc. of the guide plate. These parameters are designed according to the actual situation of the patient and are used in the operation of extracting completely impacted teeth.

[0085] Step SB00, preprocessing the acquired basic information of the patient's oral and maxillofacial regions and the guide plate design parameters.

[0086] Among them, data preprocessing includes outlier removal, normalization and feature selection.

[0087] A specific example is as follows: Assume that in step SB00, the basic information of the patient's oral and maxillofacial areas and the guide plate design parameters obtained need to be preprocessed. First, outliers are identified and eliminated through statistical methods or domain knowledge to ensure the accuracy and reliability of the data. Then, the data is normalized and scaled to between 0 and 1 to improve the training speed and stability of the model. Finally, the most relevant features are selected from the original features through feature selection methods for model training, reducing the complexity of the model and improving the generalization ability of the model.

[0088] Step SC00, input the preprocessed data into a deep neural network model of a preset architecture, and complete the initialization setting of the deep neural network model, including the initial value setting of weights and biases.

[0089] The model architecture selection is as follows: In step SC00, a preset deep neural network model architecture needs to be selected first. The selection of the deep neural network model architecture should be based on the complexity of the problem, the characteristics of the data, and the limitations of computing resources. In this application, a convolutional neural network (CNN) suitable for processing images and structured data may be selected as the model architecture.

[0090] Assume that in step SC00, the preprocessed data needs to be input into a deep neural network model of a preset architecture, and the initialization setting of the deep neural network model is completed. First, a convolutional neural network (CNN) suitable for processing images and structured data is selected as the model architecture. Then, Xavier initialization or He initialization is selected as the initialization method of the model to set the initial values ​​of weights and biases. Finally, the preprocessed data is input into the deep neural network model of the preset architecture to train the model.

[0091] Step SD00, based on the preset learning rate, number of iterations and L2 regularization coefficient of the deep neural network model and the initialized deep neural network model, the deep neural network model is trained offline using historical data.

[0092] Among them, the training set and validation set of offline training are divided according to the preset ratio.

[0093] Among them, the learning rate is set as follows. In this application, an adaptive learning rate algorithm may be selected, such as Adam, RMSprop, etc., to automatically adjust the learning rate and improve the stability and efficiency of model training.

[0094] The number of iterations is set as follows. In this application, it is possible to choose to dynamically adjust the number of iterations according to the prediction error rate on the validation set to avoid overfitting and underfitting.

[0095] The L2 regularization coefficient is set as follows. In this application, a cross-validation method, such as k-fold cross-validation, may be used to select the optimal L2 regularization coefficient to improve the generalization ability of the model.

[0096] The training set and validation set are divided as follows. In this application, 70% of the data may be selected as the training set and 30% of the data as the validation set to balance the training and validation of the model.

[0097] A specific example is as follows: Assume that in step SD00, it is necessary to use historical data to perform offline training on the deep neural network model based on the preset learning rate, number of iterations, L2 regularization coefficient, and initialization setting of the deep neural network model. First, select an adaptive learning rate algorithm, such as Adam, RMSprop, etc., to automatically adjust the learning rate to improve the stability and efficiency of model training. Then, choose to dynamically adjust the number of iterations based on the prediction error rate on the validation set to avoid overfitting and underfitting. Next, choose to use a cross-validation method, such as k-fold cross-validation, to select the optimal L2 regularization coefficient to improve the generalization ability of the model. Finally, select 70% of the data as the training set and 30% of the data as the validation set to train and verify the model.

[0098] Step SE00, when the prediction error rate of the trained deep neural network model on the validation set is lower than the preset error rate, it is confirmed that the deep neural network model has completed training, and the confirmed deep neural network model is used as a trained deep neural network model for predicting the optimal guide plate design parameters.

[0099] The preset error rate is set as follows. In this application, you may choose to set a reasonable preset error rate based on domain knowledge and experience to determine whether the model has completed training. The preset error rate in this application can be 90%.

[0100] exist Figure 1 After step S400, it is also necessary to consider that the output predicted optimal guide plate design parameters may not necessarily pass the simulated surgery, so further analysis is required. Figure 3 The illustrated embodiment is described in detail.

[0101] Reference Figure 3 A digital guide preparation method for extracting a completely impacted tooth also includes steps after outputting the best predicted guide design parameters and before printing out the guide according to the guide design parameters actually applied, specifically as follows:

[0102] Step Sa00, processing the data contained in the collected basic information of the patient's oral and maxillofacial areas.

[0103] Among them, data processing includes data cleaning, data segmentation, and data labeling.

[0104] Step Sb00, establishing a model based on the basic information of the oral and maxillofacial regions of the patient after data processing.

[0105] The process of model building is as follows: the model is built based on the basic information of the patient's oral and maxillofacial regions after data processing. The implementation methods of model building include three-dimensional reconstruction, finite element modeling, neural network modeling, etc. The purpose of model building is to provide an accurate model for subsequent finite element analysis to evaluate the rationality of the guide plate design parameters. During the model building process, the parameters of the model need to be set, including geometric parameters, material parameters, boundary conditions, etc. The setting of model parameters should be based on the basic information of the patient's oral and maxillofacial regions to improve the accuracy and reliability of the model.

[0106] For example, assume that in step Sb00, a model needs to be established based on the basic information of the patient's oral and maxillofacial regions after data processing. First, three-dimensional reconstruction is performed to establish a three-dimensional model based on the basic information of the patient's oral and maxillofacial regions. Then, finite element modeling is performed to establish a finite element model based on the three-dimensional model. Next, neural network modeling is performed to establish a neural network model based on the finite element model.

[0107] In step Sc00, the established model and the input guide plate parameters are converted into a finite element mesh, boundary conditions and loads are applied, and various operations during the surgery are simulated.

[0108] The process of finite element mesh conversion is as follows: In step Sc00, the established model and the input guide plate parameters need to be converted into finite element mesh. The implementation methods of finite element mesh conversion include STL model to finite element mesh, CAD model to finite element mesh, etc. The purpose of finite element mesh conversion is to convert the model and parameters into the mesh form required for finite element analysis so as to perform subsequent finite element analysis.

[0109] Boundary condition setting: After the finite element mesh conversion is completed, boundary conditions need to be set. The boundary conditions should be set according to the actual situation during the operation, including fixed boundaries, sliding boundaries, contact boundaries, etc. The setting of boundary conditions has an important impact on the results of finite element analysis and should be based on the actual situation during the operation.

[0110] The settings for load application are as follows: After the boundary conditions are set, loads need to be applied. The application of loads should be based on the actual situation during the operation, including force loads, pressure loads, temperature loads, etc. The application of loads has an important impact on the results of finite element analysis and should be based on the actual situation during the operation.

[0111] The surgical process simulation is as follows: After the finite element mesh conversion, boundary condition setting and load application are completed, it is necessary to simulate various operations during the surgical process. The implementation methods of surgical process simulation include finite element analysis, neural network analysis, etc. The purpose of surgical process simulation is to evaluate the rationality of the guide plate design parameters and provide accurate simulation results for subsequent finite element analysis.

[0112] Specific examples are as follows:

[0113] Assume that in step Sc00, it is necessary to convert the established model and the input guide plate parameters into a finite element mesh, apply boundary conditions and loads, and simulate various operations during the operation. First, convert the STL model to a finite element mesh, and convert the model and parameters into a finite element mesh. Then, set the boundary conditions according to the actual situation during the operation. Next, apply the load according to the actual situation during the operation. Finally, simulate the operation process and evaluate the rationality of the guide plate design parameters.

[0114] Step Sd00, obtaining the result of finite element analysis by solving the finite element equation.

[0115] Among them, the results of finite element analysis include stress analysis results, strain analysis results and displacement analysis results.

[0116] The solution of the finite element equation is as follows: In step Sd00, the finite element equation needs to be solved to obtain the result of the finite element analysis. The implementation methods of the finite element equation solution include direct solution, iterative solution, etc. The purpose of the finite element equation solution is to obtain the result of the finite element analysis, including the stress analysis result, the strain analysis result and the displacement analysis result.

[0117] Stress analysis is as follows: Stress analysis is an important part of finite element analysis, which is used to evaluate the rationality of the guide plate design parameters. The implementation methods of stress analysis include linear stress analysis, nonlinear stress analysis, etc. The purpose of stress analysis is to evaluate the rationality of the guide plate design parameters and provide accurate stress analysis results for subsequent finite element analysis.

[0118] Strain analysis is as follows: Strain analysis is an important part of finite element analysis, which is used to evaluate the rationality of the guide plate design parameters. The implementation methods of strain analysis include linear strain analysis, nonlinear strain analysis, etc. The purpose of strain analysis is to evaluate the rationality of the guide plate design parameters and provide accurate strain analysis results for subsequent finite element analysis.

[0119] Displacement analysis is as follows: Displacement analysis is an important part of finite element analysis, which is used to evaluate the rationality of the guide plate design parameters. The implementation methods of displacement analysis include linear displacement analysis, nonlinear displacement analysis, etc. The purpose of displacement analysis is to evaluate the rationality of the guide plate design parameters and provide accurate displacement analysis results for subsequent finite element analysis.

[0120] Specific examples are as follows:

[0121] Assume that in step Sd00, it is necessary to obtain the results of finite element analysis by solving finite element equations. First, solve the finite element equations to obtain the results of finite element analysis, including stress analysis results, strain analysis results, and displacement analysis results. For example, the stress analysis results may show that the stress value of the guide plate at the maximum stress point is 150MPa, the strain analysis results may show that the strain value of the guide plate at the maximum strain point is 0.002, and the displacement analysis results may show that the displacement value of the guide plate at the maximum displacement point is 0.5mm. These values ​​are calculated by finite element analysis software. The software obtains them by solving finite element equations based on factors such as the material properties, geometry, boundary conditions, and loads of the guide plate. The software uses high-precision meshing and precise material property parameters to ensure the accuracy of the analysis results.

[0122] Step Se00, analyzing whether the key indicator value contained in the result of finite element analysis exceeds the threshold value of the corresponding key indicator. If not, execute step Sf00; if yes, execute step Sg00.

[0123] In step Se00, it is necessary to set the threshold of the key indicator to evaluate the rationality of the finite element analysis results. The methods for setting the threshold of the key indicator include setting based on experience, setting based on standards, setting based on safety factors, etc. The purpose of setting the threshold of the key indicator is to evaluate the rationality of the finite element analysis results and provide accurate thresholds of the key indicator for subsequent verification of the finite element analysis results.

[0124] Specific examples are as follows:

[0125] Assume that in step Se00, it is necessary to analyze whether the key indicator value contained in the result of finite element analysis exceeds the threshold of the corresponding key indicator. First, set the threshold of the key indicator, for example, the stress threshold is 100MPa, the strain threshold is 0.001, and the displacement threshold is 0.3mm. Then, perform key indicator value verification, for example, the maximum stress value in the stress analysis result is 150MPa, the maximum strain value in the strain analysis result is 0.002, and the maximum displacement value in the displacement analysis result is 0.5mm. By comparing the key indicator value and the key indicator threshold, the rationality of the finite element analysis result can be evaluated. If the key indicator value exceeds the key indicator threshold, for example, the maximum stress value in the stress analysis result exceeds the stress threshold, the maximum strain value in the strain analysis result exceeds the strain threshold, and the maximum displacement value in the displacement analysis result exceeds the displacement threshold, then execute step Sg00 to adjust the parameters until the surgical requirements are met. If the key indicator value does not exceed the key indicator threshold, execute step Sf00 and continue to execute subsequent steps, such as printing of the guide plate and surgical preparation.

[0126] Step Sf00, continue to execute subsequent steps.

[0127] Step Sg00, taking the key indicator exceeding the threshold of the corresponding key indicator as the indicator to be optimized, and analyzing whether the difference between the value of the indicator to be optimized and the threshold of the corresponding key indicator falls within the difference threshold interval. If not, execute step Sh00; if yes, execute step Si00.

[0128] In step Sg00, the key indicators that exceed the threshold of the corresponding key indicators need to be used as indicators to be optimized, which are used to evaluate the rationality of the design parameters of the guide plate. The implementation methods for determining the indicators to be optimized include direct comparison, statistical analysis, etc. The purpose of determining the indicators to be optimized is to evaluate the rationality of the design parameters of the guide plate and provide accurate indicators to be optimized for subsequent key indicator optimization.

[0129] Specific examples are as follows:

[0130] Assume that in step Sg00, it is necessary to take the key indicator that exceeds the threshold of the corresponding key indicator as the indicator to be optimized, and analyze whether the difference between the value of the indicator to be optimized and the threshold of the corresponding key indicator falls within the difference threshold interval range. First, determine the indicator to be optimized, for example, the maximum stress value in the stress analysis result is 150MPa, the maximum strain value in the strain analysis result is 0.002, and the maximum displacement value in the displacement analysis result is 0.5mm. Then, set the threshold of the key indicator, for example, the stress threshold is 100MPa, the strain threshold is 0.001, and the displacement threshold is 0.3mm. Next, perform a difference threshold interval range analysis, for example, the difference between the maximum stress value in the stress analysis result and the stress threshold is 50MPa, the difference between the maximum strain value in the strain analysis result and the strain threshold is 0.001, and the difference between the maximum displacement value in the displacement analysis result and the displacement threshold is 0.2mm. Set the difference threshold interval range, for example, the stress difference threshold interval range is 0-50MPa, the strain difference threshold interval range is 0-0.001, and the displacement difference threshold interval range is 0-0.2mm. By comparing the difference and the difference threshold interval range, the rationality of the guide plate design parameters can be evaluated. If the difference falls within the difference threshold interval range, for example, the difference between the maximum stress value and the stress threshold in the stress analysis result falls within the stress difference threshold interval range, the difference between the maximum strain value and the strain threshold in the strain analysis result falls within the strain difference threshold interval range, and the difference between the maximum displacement value and the displacement threshold in the displacement analysis result falls within the displacement difference threshold interval range, then step Si00 is executed to adjust the parameters until the surgical requirements are met. If the difference does not fall within the difference threshold interval range, step Sh00 is executed to determine that there is a significant deviation in the guide plate design parameters actually applied, and subsequent steps are stopped.

[0131] Step Sh00, determining that there is an obvious deviation in the design parameters of the guide plate in actual application, and stopping the subsequent steps.

[0132] Step Si00, analyze and determine the adjustment amplitude schemes of various guide plate associated parameters according to the correspondence between the difference interval range of the difference between the numerical value of the indicator to be optimized and the corresponding key indicator threshold and the adjustment amplitude schemes of various guide plate associated parameters, and execute the adjustment amplitude schemes of various guide plate associated parameters.

[0133] The purpose of the analysis of the correspondence between the difference interval range and the adjustment range scheme is to analyze and determine the adjustment range scheme of various guide plate associated parameters, so as to provide an accurate adjustment range scheme for the subsequent guide plate associated parameter adjustment. The analysis of the adjustment range scheme of various guide plate associated parameters is determined as follows: taking the difference interval range in which the difference between the value of the indicator to be optimized and the corresponding key indicator threshold falls as the query object, query and obtain the adjustment range scheme of various guide plate associated parameters from the preset database storing the correspondence between the difference interval range in which the difference between the value of the indicator to be optimized and the corresponding key indicator threshold falls and the adjustment range scheme of various guide plate associated parameters.

[0134] For example, the guide plate includes a tooth-supported retaining plate, a surgical incision guide ring, a positioning and depth-setting guide ring, and a tooth-separating positioning guide ring. For a tooth-supported retaining plate, if the difference between the maximum stress value and the stress threshold in the stress analysis result falls within the stress difference interval, the thickness of the tooth-supported retaining plate is adjusted, for example, the thickness is adjusted from 1.5 mm to 1.65 mm. For a surgical incision guide ring, if the difference between the maximum strain value and the strain threshold in the strain analysis result falls within the strain difference interval, the diameter of the surgical incision guide ring is adjusted, for example, the diameter is adjusted from 10 mm to 10.1 mm. For a positioning and depth-setting guide ring, if the difference between the maximum displacement value and the displacement threshold in the displacement analysis result falls within the displacement difference interval, the depth of the positioning and depth-setting guide ring is adjusted, for example, the depth is adjusted from 3 mm to 3.1 mm. For the tooth separation positioning guide ring, if the difference between the maximum stress value and the stress threshold in the stress analysis result falls within the stress difference interval, the width of the tooth separation positioning guide ring is adjusted, for example, the width is adjusted from 2 mm to 2.2 mm.

[0135] Step Sj00, taking the guide plate parameters after executing the adjustment range schemes of various guide plate associated parameters as input guide plate parameters, and jumping to execute the step of converting the established model and the input guide plate parameters into a finite element mesh.

[0136] Step Sk00: If and only if the key indicator value included in the result of the finite element analysis does not exceed the threshold value of the corresponding key indicator, continue with the subsequent steps; otherwise, continue to execute the adjustment range plan of various guide plate associated parameters.

[0137] exist Figure 3 In step Sk00, the possibility of too many adjustments should also be considered. In this case, the doctor should be considered to intervene in the active simulation adjustment to obtain the adjustment plan as soon as possible. For details, refer to Figure 4 The illustrated embodiment is described in detail.

[0138] Reference Figure 4A method for preparing a digital guide plate for extracting a completely impacted tooth also includes a step parallel to continuing the subsequent steps if and only if the key indicator value included in the result of the finite element analysis does not exceed the threshold value of the corresponding key indicator, specifically as follows:

[0139] Step SF00, analyzing whether the number of times the adjustment amplitude schemes of various guide plate associated parameters are executed exceeds a preset threshold number of times.

[0140] In step SF00, it is necessary to analyze whether the number of times the adjustment range schemes of various guide plate associated parameters are executed exceeds the preset threshold number. The implementation methods of the adjustment number monitoring include counting, statistical analysis, etc. The purpose of the adjustment number monitoring is to monitor the number of times the guide plate parameters are adjusted, to ensure that the number of times the guide plate parameters are adjusted does not exceed the preset threshold number, and to avoid unreasonable guide plate design parameters caused by excessive adjustment.

[0141] For example, assume that in step SF00, it is necessary to analyze whether the number of times the adjustment amplitude schemes of various guide plate associated parameters are executed exceeds the preset threshold number. First, the preset threshold number is set, for example, the preset threshold number is 10 times. Then, the number of adjustments is monitored, for example, the number of times the adjustment amplitude schemes of various guide plate associated parameters are executed is 8 times. By comparing the number of adjustments and the preset threshold number, the number of times the guide plate parameters are adjusted can be monitored. If the number of adjustments does not exceed the preset threshold number, for example, the number of times the adjustment amplitude schemes of various guide plate associated parameters are executed is 8 times, which does not exceed the preset threshold number of 10 times, then the adjustment amplitude schemes of various guide plate associated parameters continue to be executed. If the number of adjustments exceeds the preset threshold number, for example, the number of times the adjustment amplitude schemes of various guide plate associated parameters are executed is 12 times, which exceeds the preset threshold number of 10 times, then step SG00 is executed to stop subsequent adjustments, and the latest adjusted guide plate parameters are imported into the virtual reality or augmented system, and a virtual surgical scene matching the actual situation of the patient is created for the doctor to simulate the surgery and adjust the guide plate parameters.

[0142] Step SG00, if yes, stop subsequent adjustments, import the most recently adjusted guide plate parameters into the virtual reality or augmented system, and create a virtual surgical scene that matches the patient's actual situation for the doctor to simulate surgery and adjust the guide plate parameters.

[0143] exist Figure 4 In step SG00, creating a virtual surgical scene that matches the actual patient's situation for the doctor to simulate the surgery includes: step SG10, obtaining relevant data of the patient's specific factors and preprocessing the data, wherein the patient's specific factors include but are not limited to general health status, age, gender, and living habits. Step SG20, based on the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial regions, using biomechanical and biophysical simulation technology, construct a personalized oral and maxillofacial model.

[0144] Among them, the method for obtaining the relevant data of the specific factors involved in step SG10 is as follows: questionnaire survey, medical record query, biomarker detection, etc. The purpose of obtaining patient-specific factor data is to obtain the relevant data of the patient's specific factors and provide accurate patient-specific factor data for the subsequent creation of virtual surgical scenes.

[0145] The process of constructing a personalized oral and maxillofacial model in step SG20 is as follows: The construction of a personalized oral and maxillofacial model is an important part of the application of biomechanical and biophysical simulation technology, which is used to construct a personalized oral and maxillofacial model, including teeth, alveolar bones, soft tissues, etc. The implementation methods of constructing a personalized oral and maxillofacial model include three-dimensional modeling, finite element meshing, model parameterization, etc. The purpose of constructing a personalized oral and maxillofacial model is to construct a personalized oral and maxillofacial model to provide an accurate oral and maxillofacial model for the subsequent creation of a virtual surgical scene, among which the implementation methods of the application of biomechanical and biophysical simulation technology include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc.

[0146] Specific examples are as follows:

[0147] Assume that in step SG20, it is necessary to construct a personalized oral and maxillofacial model based on the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial regions using biomechanical and biophysical simulation technology. First, obtain the relevant data of the patient's specific factors, such as good general health, age of 30 years old, male gender, and lifestyle of no smoking and no drinking. Then, obtain the basic information of the patient's oral and maxillofacial regions, such as the number of teeth is 32, the thickness of the alveolar bone is 3mm, and the thickness of the soft tissue is 5mm. Next, use biomechanical and biophysical simulation technology to construct a personalized oral and maxillofacial model, such as using finite element analysis technology to construct a personalized tooth model, using fluid dynamics analysis technology to construct a personalized alveolar bone model, and using biomaterial mechanics analysis technology to construct a personalized soft tissue model. By using biomechanical and biophysical simulation technology to construct a personalized oral and maxillofacial model based on the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial regions, an accurate oral and maxillofacial model can be provided for the subsequent creation of virtual surgical scenes. If the model construction results show that the model is accurate and reliable, subsequent steps such as virtual surgical scene creation will be performed. If the model construction results indicate that the model is inaccurate and unreliable, the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial regions will be retrieved, and a personalized oral and maxillofacial model will be constructed using biomechanical and biophysical simulation techniques until the model is accurate and reliable.

[0148] exist Figure 4 In step SG00, the doctor adjusts the guide plate parameters including: step SGa0, obtaining detailed data on the impact of patient-specific factors on tissues and bones in the simulation, and obtaining force-related results based on detailed data analysis. Among them, the detailed data include but are not limited to stress distribution, strain, displacement, and force-related results include force characteristics and deformation trends. Step SGb0, input the design parameters of the current guide plate into the simulation system, compare the interaction force between the guide plate and tissues and bones in the simulation results, and determine the problems existing in the current guide plate based on the preset discrimination interval range of the problem-related parameters. Among them, the problems include local stress concentration, excessive compression of tissues, and mismatch with anatomical structures. Step SGc0, based on the problems existing in the current guide plate determined and the corresponding relationship between the problems and the optimization direction, an optimization direction is formed and a reference prompt is provided for the doctor to adjust the guide plate parameters.

[0149] Among them, in step SGa0, it is necessary to obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, including but not limited to stress distribution, strain, displacement, etc. The implementation methods of detailed data acquisition include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc. The purpose of detailed data acquisition is to obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, and to provide accurate detailed data for the subsequent acquisition of force-related results.

[0150] The acquisition of force-related results is an important part of detailed data acquisition, which is used to obtain force-related results, including force characteristics and deformation trends, etc. The implementation methods of force-related results acquisition include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc. The purpose of force-related results acquisition is to obtain force-related results and provide accurate force-related results for subsequent guide plate parameter adjustment.

[0151] Assume that in step SGa0, it is necessary to obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, and obtain force-related results based on detailed data analysis. First, obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, such as stress distribution of 0.1MPa, strain of 0.01, and displacement of 0.1mm. Then, obtain force-related results based on detailed data analysis, such as uniform force characteristics and linear deformation trend. By obtaining detailed data on the effects of patient-specific factors on tissues and bones in the simulation, and obtaining force-related results based on detailed data analysis, accurate force-related results can be provided for subsequent guide plate parameter adjustments. If the force-related result acquisition results show that the results are accurate and reliable, step SGb0 will be executed, the design parameters of the current guide plate will be input into the simulation system, the interaction forces between the guide plate and tissues and bones in the simulation results will be compared, and the problems existing in the current guide plate will be determined based on the preset discrimination interval range of the problem-related parameters. If the results of force-related result acquisition indicate that the results are inaccurate and unreliable, detailed data on the effects of patient-specific factors on tissue and bone in the simulation will be reacquired, and force-related results will be acquired based on detailed data analysis until the results are accurate and reliable.

[0152] In step SGb0, the design parameters of the current guide plate need to be input into the simulation system for comparison of the interaction forces between the guide plate and tissues and bones in the simulation results. The implementation methods of inputting the guide plate design parameters into the simulation system include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc. The purpose of inputting the guide plate design parameters into the simulation system is to input the design parameters of the current guide plate into the simulation system, so as to provide accurate guide plate design parameters for subsequent interaction force comparison.

[0153] Interaction force comparison is an important part of the simulation system for inputting design parameters of the guide plate, and is used to compare the interaction forces between the guide plate and tissues and bones in the simulation results. The implementation methods of interaction force comparison include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc. The purpose of interaction force comparison is to compare the interaction forces between the guide plate and tissues and bones in the simulation results, and to provide accurate interaction force comparison results for subsequent problem judgment.

[0154] Problem judgment is an important part of the interaction force comparison, which is used to judge the problems existing in the current guide plate according to the preset judgment range of the problem-related parameters. The implementation methods of problem judgment include finite element analysis, fluid dynamics analysis, biomaterial mechanics analysis, etc. The purpose of problem judgment is to judge the problems existing in the current guide plate and provide accurate problem judgment results for subsequent guide plate parameter adjustment.

[0155] For example, assuming that in step SGb0, the design parameters of the current guide plate need to be input into the simulation system, the interaction forces between the guide plate and tissues and bones in the simulation results are compared, and the problems existing in the current guide plate are determined based on the preset discrimination interval range of the problem-related parameters. First, the design parameters of the current guide plate are input into the simulation system, for example, the guide plate thickness is 1mm, the guide plate length is 10mm, and the guide plate width is 5mm. Then, the interaction forces between the guide plate and tissues and bones in the simulation results are compared, for example, the interaction force between the guide plate and tissues is 0.1N, and the interaction force between the guide plate and bones is 0.2N. Next, according to the preset discrimination interval range of the problem-related parameters, the problems existing in the current guide plate are determined, for example, the interaction force between the guide plate and tissues exceeds 0.05N, and the interaction force between the guide plate and bones exceeds 0.1N, and it is determined that the current guide plate has problems such as local stress concentration, excessive compression of tissues, and mismatch with anatomical structures. By inputting the design parameters of the current guide into the simulation system, comparing the interaction forces between the guide and tissues and bones in the simulation results, and judging the problems existing in the current guide according to the preset discrimination interval range of the problem-related parameters, an accurate problem judgment result can be provided for the subsequent guide parameter adjustment. If the problem judgment result shows that the result is accurate and reliable, step SGc0 will be executed to form an optimization direction and provide a reference prompt for the doctor to adjust the guide parameters according to the problems existing in the current guide and the corresponding relationship between the problems and the optimization direction. If the problem judgment result shows that the result is inaccurate and unreliable, the design parameters of the current guide will be re-input into the simulation system, comparing the interaction forces between the guide and tissues and bones in the simulation results, and judging the problems existing in the current guide according to the preset discrimination interval range of the problem-related parameters, until the result is accurate and reliable.

[0156] Step SH00, if and only if the doctor's simulated surgery process conforms to the preset process and the simulated surgery is successful, the adjusted guide plate parameters are used as the guide plate design parameters for this actual application.

[0157] The process of the doctor simulating surgery in accordance with the preset process includes: step SH10, obtaining the accuracy of the doctor's surgical simulation path and the surgical operation, and analyzing whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and analyzing whether the values ​​of the surgical operation-related parameters are all within the preset parameter value range and whether the damage degree of the key structure during the operation falls within the preset damage degree value range. In step SH10, it is necessary to obtain the accuracy of the doctor's surgical simulation path and the surgical operation. The implementation method of surgical simulation path acquisition includes virtual reality technology, augmented reality technology, three-dimensional reconstruction technology, etc. The purpose of surgical simulation path acquisition is to obtain the accuracy of the doctor's surgical simulation path and the surgical operation, and to provide an accurate surgical simulation path for subsequent deviation analysis.

[0158] Deviation analysis is an important part of surgical simulation path acquisition, which is used to analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation. The implementation methods of deviation analysis include three-dimensional reconstruction technology, image processing technology, data statistics technology, etc. The purpose of deviation analysis is to analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, so as to provide accurate deviation analysis results for the subsequent surgical process to judge whether it meets the preset process.

[0159] The numerical analysis of surgical operation-related parameters is an important part of deviation analysis, which is used to analyze whether the numerical values ​​of surgical operation-related parameters are all within the preset parameter value range. The implementation methods of the numerical analysis of surgical operation-related parameters include three-dimensional reconstruction technology, image processing technology, data statistics technology, etc. The purpose of the numerical analysis of surgical operation-related parameters is to analyze whether the numerical values ​​of surgical operation-related parameters are all within the preset parameter value range, and to provide accurate numerical analysis results of surgical operation-related parameters for subsequent surgical process judgments that meet the preset process.

[0160] The analysis of the damage degree value of key structures is an important part of the numerical analysis of surgical operation-related parameters. It is used to analyze whether the damage degree value of key structures during surgery falls within the preset damage degree value range. The implementation methods of the analysis of the damage degree value of key structures include three-dimensional reconstruction technology, image processing technology, data statistics technology, etc. The purpose of the analysis of the damage degree value of key structures during surgery is to analyze whether the damage degree value of key structures during surgery falls within the preset damage degree value range, and to provide accurate analysis results of the damage degree value of key structures for the subsequent surgical process to determine whether it meets the preset process.

[0161] Assume that in step SH10, it is necessary to obtain the accuracy of the doctor's surgical simulation path and the surgical operation, and analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and analyze whether the values ​​of the surgical operation-related parameters all fall within the preset parameter value range, and whether the degree of damage to the key structure during the operation falls within the preset damage degree value range. First, obtain the accuracy of the doctor's surgical simulation path and the surgical operation, for example, the surgical simulation path is a straight path, and the accuracy of the surgical operation is 0.1mm. Then, analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, for example, the deviation between the doctor's surgical simulation path and the preset surgical planning path is 0.05mm, and the preset deviation is 0.1mm. Next, analyze whether the values ​​of the surgical operation-related parameters all fall within the preset parameter value range, for example, the surgical operation-related parameter value is 0.1N, and the preset parameter value range is 0.05N-0.15N. Finally, analyze whether the damage degree value of the key structure during the operation falls within the preset damage degree value range, for example, the damage degree value of the key structure during the operation is 0.01, and the preset damage degree value range is 0.005-0.015. By obtaining the accuracy of the doctor's surgical simulation path and the surgical operation, and analyzing whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and analyzing whether the values ​​of the surgical operation-related parameters are all within the preset parameter value range, and whether the damage degree value of the key structure during the operation falls within the preset damage degree value range, it is possible to provide accurate deviation analysis results, surgical operation-related parameter value analysis results, and key structure damage degree value analysis results for the subsequent surgical process to meet the preset process judgment. If the deviation analysis results, surgical operation-related parameter value analysis results, and key structure damage degree value analysis results show that the results are accurate and reliable, step SH20 will be executed, and the doctor's simulated surgery process is judged to meet the preset process only when and only when the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and the surgical operation-related parameter values ​​fall within the preset parameter value range, and the damage degree value of the key structure during the operation falls within the preset damage degree value range. If the deviation analysis results, the numerical analysis results of surgical operation-related parameters and the analysis results of the degree of damage to key structures indicate that the results are inaccurate and unreliable, the doctor's surgical simulation path and the accuracy of the surgical operation will be re-acquired, and it will be analyzed whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and whether the numerical values ​​of surgical operation-related parameters all fall within the preset parameter numerical range, and whether the degree of damage to key structures during the operation falls within the preset damage degree value range, until the results are accurate and reliable.

[0162] Step SH20, if and only if the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and the values ​​of the surgical operation-related parameters fall within the preset parameter value range and the degree of damage to the key structure during the operation falls within the preset damage degree value range, it is judged that the doctor's simulated surgery process conforms to the preset process.

[0163] Successful simulation surgery includes:

[0164] Step SHa0, obtaining the result values ​​of the simulated surgery regarding different set targets, where the set targets include the surgical effect, surgical time and efficiency.

[0165] In step SHa0, the set goals of the simulated surgery need to be determined, including the surgical effect, surgical time and efficiency. The implementation methods of setting goals include clinical experience, surgical planning, surgical simulation, etc. The purpose of setting goals is to determine the set goals of the simulated surgery and provide accurate set goals for subsequent result value acquisition.

[0166] Result value acquisition is an important part of setting goals, and is used to obtain the result values ​​of simulated surgery for different set goals. The implementation methods of result value acquisition include clinical experience, surgical planning, surgical simulation, etc. The purpose of result value acquisition is to obtain the result values ​​of simulated surgery for different set goals, and provide accurate result values ​​for subsequent reasonable result value range analysis.

[0167] Specific examples are as follows:

[0168] Assume that in step SHa0, it is necessary to obtain the result values ​​of the simulated surgery for different set targets, and the set targets include surgical effects, surgical time and efficiency. First, determine the set targets of the simulated surgery, for example, the surgical effect is to completely remove the impacted teeth, the surgical time and efficiency are that the surgical time is less than 1 hour, and the surgical efficiency is greater than 80%. Then, obtain the result values ​​of the simulated surgery for different set targets, for example, the surgical effect is to completely remove the impacted teeth, the surgical time is 0.5 hours, and the surgical efficiency is 90%. By determining the set targets of the simulated surgery, including the surgical effect, the surgical time and efficiency, and obtaining the result values ​​of the simulated surgery for different set targets, accurate result values ​​can be provided for the subsequent reasonable result value range analysis. If the result value acquisition result shows that the result is accurate and reliable, step SHb0 will be executed to analyze whether the result values ​​of the simulated surgery for different set targets fall into the reasonable result value range of the corresponding set targets. If the result value acquisition result shows that the result is inaccurate and unreliable, the set targets of the simulated surgery will be re-determined, including the surgical effect, the surgical time and efficiency, and the result values ​​of the simulated surgery for different set targets will be obtained until the result is accurate and reliable.

[0169] Step SHb0, analyzing whether the result values ​​of the simulated surgery with respect to different set targets fall within the reasonable result value range of the corresponding set targets. If yes, execute step SHc0; if no, execute step SHd0.

[0170] Step SHc0, determine whether the simulated operation is successful.

[0171] Step SHd0, determining whether the simulated operation has failed.

[0172] Based on the same inventive concept, an embodiment of the present invention provides a digital guide plate preparation system for extracting completely impacted teeth, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following steps: Figures 1 to 4 Procedure for either method.

[0173] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for preparing a digital guide plate for extracting completely impacted teeth, characterized in that: include: Obtain the patient's maxillofacial continuous scanning data and the digital tooth model data of the upper and lower mandibular hard plaster models; Perform three-dimensional registration and fitting of the digital tooth model data and the maxillofacial continuous scanning data; Performing in-depth analysis on the data after 3D registration fitting to obtain basic information of the patient's oral and maxillofacial regions, including but not limited to anatomical structure information, bone condition information, tooth position information, and neurovascular information; The basic information of the patient's oral and maxillofacial regions is used as input to a trained deep neural network model for predicting optimal guide plate design parameters, and the predicted optimal guide plate design parameters are output as guide plate design parameters for actual application; Printing out a guide plate according to the design parameters of the guide plate for actual application, wherein the guide plate includes a tooth-supported retention plate, a surgical incision guide ring, a positioning and depth guide ring, and a tooth separation positioning guide ring; The trained deep neural network model for predicting the optimal guide plate design parameters includes: Obtain basic information of oral and maxillofacial areas of historical patients and guide design parameters; Perform data preprocessing on the basic information of the patient's oral and maxillofacial areas and the guide plate design parameters, including outlier removal, normalization, and feature selection; Input the preprocessed data into the deep neural network model of the preset architecture, and complete the initialization setting of the deep neural network model, including the initial value setting of weights and biases; According to the preset learning rate, number of iterations and L2 regularization coefficient of the deep neural network model and the deep neural network model with initialization settings, the deep neural network model is offline trained using historical data, wherein the training set and the validation set of the offline training are divided according to a preset ratio; When the prediction error rate of the trained deep neural network model on the validation set is lower than the preset error rate, the deep neural network model is confirmed to have completed training, and the confirmed deep neural network model is used as a deep neural network model that has completed training and is used to predict optimal guide plate design parameters; The method further includes steps after outputting the predicted optimal guide plate design parameters and before printing out the guide plate according to the guide plate design parameters actually applied, which are as follows: Process the data collected from the basic information of the patient's oral and maxillofacial areas, including data cleaning, data segmentation, and data labeling; The model is established based on the basic information of the patient's oral and maxillofacial regions after data processing; Convert the established model and input guide parameters into finite element mesh, apply boundary conditions and loads, and simulate various operations during surgery; By solving the finite element equation, the result of finite element analysis is obtained, wherein the result of finite element analysis includes stress analysis result, strain analysis result and displacement analysis result; Analyze whether the key indicator values ​​contained in the results of the finite element analysis exceed the thresholds of the corresponding key indicators; If no, continue with the next steps; If yes, the key indicator that exceeds the threshold of the corresponding key indicator is taken as the indicator to be optimized, and the difference between the value of the indicator to be optimized and the threshold of the corresponding key indicator is analyzed to see whether it falls within the difference threshold interval; If not, it is determined that the design parameters of the guide plate actually used have obvious deviations, and the subsequent steps are stopped; If yes, then according to the correspondence between the difference interval range in which the difference between the value of the indicator to be optimized and the threshold value of the corresponding key indicator falls and the adjustment range schemes of various guide plate associated parameters, analyze and determine the adjustment range schemes of various guide plate associated parameters, and execute the adjustment range schemes of various guide plate associated parameters; The guide plate parameters after executing the adjustment range schemes of various guide plate associated parameters are used as input guide plate parameters, and the step of converting the established model and the input guide plate parameters into a finite element mesh is jumped to be executed; If and only if the key indicator values ​​contained in the results of the finite element analysis do not exceed the threshold values ​​of the corresponding key indicators, continue with the subsequent steps; otherwise, continue to execute the adjustment range plan of various guide plate associated parameters.

2. A method for preparing a digital guide plate for extracting a completely impacted tooth according to claim 1, characterized in that: Also included are steps parallel to continuing with the subsequent steps if and only if the key indicator value included in the result of the finite element analysis does not exceed the threshold value of the corresponding key indicator, as follows: Analyze whether the number of executions of adjustment range schemes of various guide plate associated parameters exceeds a preset threshold number; If yes, stop the subsequent adjustment, import the latest adjusted guide plate parameters into the virtual reality or augmented system, and create a virtual surgical scene matching the actual situation of the patient for the doctor to simulate the surgery and adjust the guide plate parameters; If and only if the doctor's simulated surgery process conforms to the preset process and the simulated surgery is successful, the adjusted guide plate parameters will be used as the guide plate design parameters for this actual application.

3. A method for preparing a digital guide plate for extracting a completely impacted tooth according to claim 2, characterized in that: Creating a virtual surgical scene that matches the patient's actual situation for doctors to simulate surgery includes: Obtain relevant data on patient-specific factors and pre-process the data, wherein the patient-specific factors include but are not limited to overall health status, age, gender, and lifestyle habits; Based on the relevant data of the patient's specific factors and the basic information of the patient's oral and maxillofacial area, a personalized oral and maxillofacial model is constructed using biomechanical and biophysical simulation techniques.

4. A method for preparing a digital guide plate for extracting a completely impacted tooth according to claim 3, characterized in that: The parameters for doctors to adjust the guide include: Obtain detailed data on the effects of patient-specific factors on tissues and bones in the simulation, and obtain force-related results based on detailed data analysis, where the detailed data include but are not limited to stress distribution, strain, displacement, and force-related results include force characteristics and deformation trends; Input the design parameters of the current guide plate into the simulation system, compare the interaction forces between the guide plate and tissues and bones in the simulation results, and determine the problems existing in the current guide plate according to the preset judgment range of the problem-related parameters, where the problems include local stress concentration, excessive compression of tissues, and mismatch with anatomical structures; Based on the problems existing in the current guide plate and the corresponding relationship between the problems and the optimization direction, the optimization direction is formed and provided as a reference for the doctor to adjust the guide plate parameters.

5. A method for preparing a digital guide plate for extracting a completely impacted tooth according to claim 4, characterized in that: The doctor's simulated surgery process complies with the preset process including: Obtain the accuracy of the doctor's surgical simulation path and surgical operation, and analyze whether the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and analyze whether the values ​​of surgical operation-related parameters are all within the preset parameter value range and whether the damage degree values ​​of key structures during the operation fall within the preset damage degree value range; The doctor's simulated surgery process is judged to be in accordance with the preset process if and only if the deviation between the doctor's surgical simulation path and the preset surgical planning path is less than the preset deviation, and the values ​​of the surgical operation-related parameters fall within the preset parameter value range and the degree of damage to the key structure during the operation falls within the preset damage degree value range.

6. A method for preparing a digital guide plate for extracting a completely impacted tooth according to claim 4, characterized in that: Successful simulation surgery includes: Obtain the result values ​​of the simulated surgery with respect to different set targets, including the surgical effect, surgical time and efficiency; Analyze whether the result values ​​of the simulated surgery with respect to different set targets fall within the reasonable result value range of the corresponding set targets; If yes, the simulated surgery is judged to be successful; If not, the simulated operation is judged to have failed.

7. A digital guide plate preparation system for extracting completely impacted teeth, characterized in that: The invention comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein the program can be loaded and executed by the processor to implement a digital guide plate preparation method for extracting a completely impacted tooth as described in any one of claims 1 to 6.

Images