Oral and maxillofacial implant restoration method and system based on visual analysis

Abstract

The invention relates to the technical field of oral implantation, in particular to an oral maxillofacial implantation restoration method and system based on visual analysis, and the method comprises the steps: determining preoperative planning information by adopting a mode of combining data collection, state analysis and dynamic occlusion simulation; during the surgery, performing real-time navigation on the surgery operation, and performing early warning on wrong operation based on the preoperative planning information; after the operation is completed, personalized review schemes such as matters needing attention and review time which are fit for the condition of a patient are generated according to monitoring data of the operation in the operation and patient detection data, full-period management is achieved, and the probability of occurrence of complications can be reduced. Moreover, through the mode, the problem that the review scheme is lack of personalization due to the fact that the prediction of the complications is mainly based on empirical judgment and cannot be accurately analyzed in combination with the intraoperative operation data and the individual characteristics of the patient is solved.

Description

Technical Field

[0001] This invention relates to the field of oral implant technology, specifically to a method and system for oral and maxillofacial implant restoration based on visual analysis. Background Technology

[0002] With the rapid development and widespread application of dental implant technology, dental implant restoration has gradually replaced traditional restoration techniques as the preferred treatment for tooth loss and edentulism. However, due to varying technical conditions, complications in dental implant restoration are increasingly becoming a significant factor affecting the long-term stability of patients' implant treatment. Therefore, completing standardized and precise dental implant restoration treatment is crucial to reducing complications and ensuring long-term stability. Currently, dental implant restoration faces the following challenges.

[0003] First, preoperative planning has limitations. Traditional preoperative assessment relies on two-dimensional X-rays or simple three-dimensional CBCT, which makes it difficult to comprehensively quantify bone mineral density distribution, soft tissue morphology, and dynamic characteristics of occlusal function. Although CAD / CAM technology has been developed to assist in implant placement design, it is mostly based on static anatomical data and lacks biomechanical analysis of the interaction between the implant and surrounding tissues during dynamic occlusion, leading to postoperative occlusal trauma or implant loosening in some cases.

[0004] Secondly, intraoperative care relies heavily on the surgeon's experience and lacks early warning mechanisms. Existing navigation systems mostly rely on robotic arms or optical markers for tracking, which suffers from large registration errors and poor robustness due to obstruction by blood / soft tissue. Intraoperative monitoring largely depends on the surgeon's subjective judgment and lacks real-time early warning mechanisms for risks such as neurovascular damage and implant placement angle deviations. Surgical precision is highly dependent on the surgeon's experience.

[0005] In addition, postoperative efficacy assessment mainly relies on qualitative observation through clinical examination and postoperative imaging, lacking quantitative comparison between preoperative planning and actual postoperative location; complication prediction is mostly based on empirical judgment, and cannot be accurately analyzed by combining intraoperative operation data and individual patient characteristics, resulting in a lack of personalization in follow-up plans.

[0006] Therefore, the present invention provides a method and system for oral and maxillofacial implant restoration based on visual analysis to solve the above problems. Summary of the Invention

[0007] In order to overcome the shortcomings of the existing technology, the present invention provides a method and system for oral and maxillofacial implant restoration based on visual analysis, which solves the problem that the prediction of the above-mentioned complications is mostly based on empirical judgment, and cannot be accurately analyzed by combining intraoperative operation data and individual patient characteristics, resulting in a lack of personalized follow-up plans.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for oral and maxillofacial implant restoration based on visual analysis, comprising: acquiring the patient's detection data; analyzing the bone tissue state and soft tissue state based on the detection data to obtain oral cavity state data; and analyzing the dynamic functional data in the detection data to obtain occlusal parameter data; adjusting a preset occlusal simulation model based on the oral cavity state data and occlusal parameter data to obtain a first occlusal simulation model; and performing simulation analysis based on the first occlusal simulation model to obtain occlusal simulation data. A multi-objective optimization algorithm was used to analyze the occlusal simulation data to obtain initial implant planning information; the initial implant planning information was then filtered based on the obtained patient feedback information to obtain preoperative planning information. The operation is based on preoperative planning information and the operation information is acquired in real time; the deviation information between the preoperative planning information and the operation information is analyzed, and an early warning information is issued when the deviation information is not met. After the procedure is completed, a postoperative assessment model is used to analyze the complete operation information and test data to obtain a comprehensive assessment result; a maintenance strategy is then generated based on the comprehensive assessment result.

[0009] Preferably, the step of analyzing bone tissue and soft tissue conditions based on the detection data to obtain oral cavity condition data includes: preprocessing the three-dimensional voxel data in the detection data using a Gaussian filtering algorithm to obtain preprocessed data; segmenting the preprocessed data using a U-Net deep learning model to obtain segmentation marker data; traversing and analyzing the potential implantation areas in the segmentation marker data to obtain bone information; comparing and grading the bone information based on preset analysis standards, and displaying it as a heatmap to obtain a bone condition heatmap. The Poisson surface reconstruction method was used to process the dentition data and point cloud data in the detection data to generate a soft tissue surface mesh model. The shortest distance algorithm was used to analyze the soft tissue surface mesh model to obtain the mucosal thickness data of the potential implant area. The preset ResNet50 model was used to extract features and classify the soft tissue surface image data in the detection data to determine the classification results of the potential implant area. The bone condition heatmap, mucosal thickness data and classification results were integrated to obtain oral cavity status data.

[0010] Preferably, the analysis of dynamic functional data in the detection data to obtain occlusal parameter data includes: extracting edge information from the dynamic functional data using the Canny edge detection method; identifying contact point coordinates in the edge information using the Hough transform method; calculating the contact area and distribution density based on the contact point coordinates; smoothing the dynamic functional data using the Kalman filter method to obtain motion trajectory data; fitting open and closed motion curves based on the motion trajectory data; calculating the condyle movement range based on the motion curves; and integrating the contact point coordinates, contact area, distribution density, motion curves, and condyle movement range to obtain occlusal parameter data.

[0011] Preferably, the step of adjusting the preset occlusal simulation model according to oral state data and occlusal parameter data to obtain the first occlusal simulation model includes: matching the oral state data with a preset three-dimensional oral template, and replacing the current three-dimensional oral template in the preset occlusal simulation model with the preset three-dimensional oral template with the highest matching degree; and adjusting the corresponding parameter data in the preset occlusal simulation model according to the occlusal parameter data to obtain the first occlusal model.

[0012] Preferably, the step of performing simulation analysis based on the first occlusal simulation model to obtain occlusal simulation data includes: using the first occlusal simulation model to simulate oral functional movement information after implantation to obtain occlusal simulation data.

[0013] Preferably, the step of using a multi-objective optimization algorithm to analyze occlusal simulation data to obtain initial implant planning information includes: analyzing the occlusal simulation data to determine the characteristic data and corresponding constraints of each of the implant stability target, occlusal function recovery target, soft tissue aesthetic target, and surgical safety target; using a non-dominated sorting genetic algorithm to iteratively optimize the characteristic data and corresponding constraints of each target to output a multi-objective balanced optimal solution; and forming initial implant planning information based on the multi-objective balanced optimal solution.

[0014] Preferably, the step of using a non-dominated sorting genetic algorithm to iteratively optimize the feature data and corresponding constraints of each objective and output a multi-objective equilibrium optimal solution includes: generating an initial scheme population based on the implant parameter range using a random sampling method; verifying the initial scheme population based on occlusion simulation data and constraints to obtain multiple feasible initial schemes; performing a crossover operation on the initial schemes to obtain new schemes; iteratively executing non-dominated sorting and crowding calculation on the initial and new schemes based on each objective, retaining the scheme with the highest crowding in this comparison; and when the change range of each objective is less than a preset range for three consecutive times and the proportion of feasible schemes in the population is greater than the corresponding preset value, the loop is exited, and multiple multi-objective equilibrium optimal solutions are output.

[0015] Preferably, the step of filtering initial implant planning information based on the obtained patient opinion information to obtain preoperative planning information includes: analyzing the patient opinion information and extracting key opinion features; selecting the planning information with the highest matching degree with the key opinion features from multiple initial implant planning information to obtain preoperative planning information.

[0016] Preferably, the postoperative evaluation model is used to analyze complete operational information and test data to obtain a comprehensive evaluation result, including: comparing bone tissue state data in the operational information and bone tissue state data in the test data to obtain bone tissue analysis results; determining postoperative occlusal parameter information based on the operational information; performing simulation analysis based on the postoperative occlusal parameter information to obtain occlusal recovery results; examining soft tissue morphology to obtain soft tissue evaluation results; and integrating bone tissue analysis results, occlusal recovery results, and soft tissue evaluation results to obtain a comprehensive evaluation result.

[0017] Secondly, the present invention provides a visual analysis-based oral and maxillofacial implant restoration system, which is used to perform the implant restoration method described in any of the above technical solutions.

[0018] The beneficial effects of this invention are as follows: 1. This invention employs a combination of data acquisition, status analysis, and dynamic occlusal simulation to determine preoperative planning information. During surgery, real-time navigation of the surgical procedure is provided, and warnings for erroneous operations are issued based on the preoperative planning information. After the procedure, a personalized follow-up plan tailored to the patient's condition is generated based on intraoperative monitoring data and patient test data, including precautions and follow-up schedules, achieving full-cycle management and helping to reduce the probability of complications. Furthermore, through the above methods, this invention addresses the problem that complication prediction is often based on empirical judgment, failing to combine intraoperative data and individual patient characteristics for precise analysis, resulting in a lack of personalized follow-up plans.

[0019] 2. During the surgery, the system acquires operational information in real time through the intraoperative navigation system, sensors, and other devices. When the calculated deviation exceeds the preset allowable range, i.e. when the deviation does not meet the requirements, an early warning message is immediately issued. The doctor will then adjust the surgical operation in a timely manner to correct the deviation, ensuring that the surgical operation meets the preoperative planning requirements, improving the accuracy and safety of the surgery. At the same time, it solves the problem of relying heavily on the doctor's experience during the operation and lacking an early warning mechanism.

[0020] 3. This invention does not pursue the optimization of a single objective, but rather balances four major objectives—stability, function, aesthetics, and safety—through a multi-objective optimization algorithm to generate a comprehensive and optimal implant planning scheme that conforms to clinical practice. Compared to traditional methods where doctors subjectively select implant parameters based on experience, this invention can process occlusal simulation data, avoiding subjective bias; it can simultaneously handle multi-dimensional objectives, taking into account both short-term surgical safety and long-term implant outcomes; and the generated initial planning information can be directly integrated with the intraoperative navigation system, providing data support for precise surgery, ultimately reducing the incidence of postoperative complications and improving the success rate of implant restoration.

[0021] 4. After obtaining the initial implant planning information, the present invention can make appropriate adjustments to the initial implant planning information based on the patient's feedback, and find preoperative planning information that meets the patient's needs. Attached Figure Description

[0022] Figure 1 This is a schematic flowchart of a visual analysis-based oral and maxillofacial implant restoration method according to the present invention; Figure 2 This is a schematic diagram of a visual analysis-based oral and maxillofacial implant restoration system according to the present invention. Detailed Implementation

[0023] The following will refer to the attached reference. Figure 1 and attached Figure 2 The various embodiments of the present invention will be described in detail below. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0024] In one embodiment of the present invention, a method for oral and maxillofacial implant restoration based on visual analysis is provided, as shown in the attached figure. Figure 1 As shown, it includes the following steps: Step S11: Acquire patient's test data. The test data includes static anatomical data and dynamic functional data. Static anatomical data includes: 3D voxel data of the jawbone, teeth, and nerves / blood vessels obtained via CBCT, including key parameters such as bone density distribution and bone thickness; and high-precision 3D point cloud data of the dentition and soft tissue surfaces obtained via an intraoral scanner. Dynamic functional data includes recording the patient's occlusal data using a high-speed camera combined with marker-based tracking technology, including maximum opening, protrusion, and lateral movement trajectories; and acquiring static occlusal contact point distribution information through a combination of pressure-sensitive occlusal films and visual recognition.

[0025] Step S12: Based on the detection data, analyze the bone tissue state and soft tissue state to obtain oral cavity state data; and analyze the dynamic functional data in the detection data to obtain occlusal parameter data; based on the oral cavity state data and occlusal parameter data, adjust the preset occlusal simulation model to obtain the first occlusal simulation model; perform simulation analysis based on the first occlusal simulation model to obtain occlusal simulation data.

[0026] Specifically, when analyzing bone tissue status, adaptive threshold segmentation is performed on the three-dimensional voxel data in the detection data. The U-Net deep learning model is used to segment the cancellous bone and cortical bone regions, calculate the bone density value, and classify it according to the preset threshold. Then, the bone width and bone height of the potential dental implant area are calculated through a three-dimensional distance transformation algorithm, and the areas with insufficient bone volume are automatically marked, thereby obtaining bone status data.

[0027] When analyzing soft tissue status, the Poisson surface reconstruction method was used to process the point cloud data from the oral cavity scan to generate a soft tissue mesh. The mucosal thickness of the implant area was calculated using the shortest distance algorithm. A pre-set ResNet50 model was used to classify soft tissue surface features and identify risk types such as thin gingival biotype. Soft tissue surface features include gingival margin curve and keratinized gingival width. The ICP algorithm was used to register the point cloud data and three-dimensional voxel data, extract the surface data of the soft tissue in the implant area, calculate the thickness distribution and morphological parameters, and output soft tissue status data.

[0028] The dynamic functional data in the test data are analyzed to obtain occlusal parameter data, including: using a combination of Canny edge detection and Hough transform methods to extract the contact point coordinates in the dynamic functional data, and calculating the contact area and distribution density; the dynamic functional data are smoothed using Kalman filtering, and the motion curves of the open and closed occlusal regions are fitted to calculate the condylar movement range, providing a data foundation for subsequent dynamic occlusal simulation based on bone and soft tissue constraints and occlusal motion parameters.

[0029] Based on oral cavity status data and occlusal parameter data, a pre-set occlusal simulation model is adjusted to obtain a first occlusal simulation model. Simulation analysis is then performed on this first occlusal simulation model to obtain occlusal simulation data. Specifically, the pre-set occlusal simulation model is a template model built based on a large amount of historical data. After adjusting some key parameters in the pre-set occlusal simulation model according to oral cavity status data and occlusal parameter data, the first occlusal simulation model is obtained, making the occlusal simulation data simulated by the first occlusal simulation model more closely reflect the patient's actual situation. The occlusal simulation data includes parameters such as the maximum equivalent force of the bone tissue surrounding the implant and the frequency of occlusal collisions. This allows for early prediction of the implant's stability during dynamic occlusion, avoiding postoperative occlusal trauma.

[0030] Step S13: Analyze the occlusal simulation data using a multi-objective optimization algorithm to obtain initial implant planning information; filter the initial implant planning information based on the obtained patient feedback information to obtain preoperative planning information.

[0031] Step S14: Perform the operation based on the preoperative planning information and obtain the operation information in real time; analyze the deviation information between the preoperative planning information and the operation information, and issue an early warning information when the deviation information is not met.

[0032] Step S15: After the operation is completed, the postoperative assessment model is used to analyze the complete operation information and test data to obtain a comprehensive assessment result; a maintenance strategy is generated based on the comprehensive assessment result.

[0033] In summary, this invention employs a combination of data acquisition, status analysis, and dynamic occlusal simulation to determine preoperative planning information. During surgery, real-time navigation of the surgical procedure is provided, and warnings for erroneous operations are issued based on the preoperative planning information. After the procedure, a personalized follow-up plan tailored to the patient's condition is generated based on intraoperative monitoring data and patient test data, including precautions and follow-up schedules, achieving full-cycle management and helping to reduce the probability of complications. Furthermore, through the above methods, this invention addresses the problem that complication prediction is often based on empirical judgment, failing to combine intraoperative data and individual patient characteristics for precise analysis, resulting in a lack of personalization in follow-up plans.

[0034] In one embodiment of the present invention, the step of analyzing bone tissue state and soft tissue state based on detection data to obtain oral cavity state data includes: preprocessing the three-dimensional voxel data in the detection data using a Gaussian filtering algorithm to obtain preprocessed data; segmenting the preprocessed data using a U-Net deep learning model to obtain segmentation marker data; traversing and analyzing the potential implantation areas in the segmentation marker data to obtain bone information; comparing and grading the bone information based on preset analysis standards, and displaying it as a heatmap to obtain a bone condition heatmap.

[0035] The Poisson surface reconstruction method was used to process the dentition data and point cloud data in the detection data to generate a soft tissue surface mesh model. The shortest distance algorithm was used to analyze the soft tissue surface mesh model to obtain the mucosal thickness data of the potential implant area. The preset ResNet50 model was used to extract features and classify the soft tissue surface image data in the detection data to determine the classification results of the potential implant area. The bone condition heatmap, mucosal thickness data and classification results were integrated to obtain oral cavity status data.

[0036] Specifically, the acquired 3D voxel data is first preprocessed using a Gaussian filtering algorithm. Gaussian filtering effectively removes noise from the 3D voxel data by weighted averaging of the pixel values ​​surrounding each pixel, making the data smoother and improving the accuracy of subsequent analysis.

[0037] The U-Net deep learning model was used to segment the preprocessed 3D voxel data, resulting in segmentation labeled data. Trained on a large amount of labeled 3D oral and maxillofacial images, the U-Net model accurately identifies and segments key anatomical structures such as the mandibular canal and maxillary sinus. After segmentation, the system automatically marks the location and extent of these structures, providing warnings of potential danger zones for subsequent implant assessment.

[0038] While avoiding marked dangerous structures such as the mandibular canal and maxillary sinus, all possible implantation areas were explored. For each potential implantation area, its bone mineral density (BMD) value was calculated by analyzing the grayscale values ​​of voxels within the area and comparing them with a standard BMD database. Simultaneously, a three-dimensional distance transformation algorithm was used to calculate the bone width and height of the potential implantation area. This algorithm, by calculating the distance from each pixel to the nearest boundary, can quickly and accurately measure the dimensional parameters of the bone tissue.

[0039] Based on calculated parameters such as bone mineral density, bone width, and bone height, bone tissue conditions are graded according to a pre-defined grading standard. Then, the bone tissue conditions are visually displayed in the form of a bone condition heatmap, with different colors representing different grades of bone tissue condition, thus obtaining bone status data; for example, red indicates areas with sufficient bone volume and high bone density, while blue indicates areas with insufficient bone volume or low bone density. Bone condition heatmaps allow dentists to quickly understand the bone tissue condition in different areas of the patient's oral cavity, providing a clear reference for implant placement selection.

[0040] When analyzing the soft tissue condition, the Poisson surface reconstruction method is used to process the high-precision 3D point cloud data of the dental arches and soft tissue surfaces in the detection data to generate a soft tissue surface mesh model. By solving the Poisson equation, the Poisson surface reconstruction method can generate a continuous and smooth soft tissue surface mesh model from discrete point cloud data, which accurately reflects the morphological characteristics of the soft tissue.

[0041] Based on the generated soft tissue mesh model, the mucosal thickness of the implantation area was calculated using the shortest distance algorithm. The shortest distance algorithm obtains mucosal thickness data by calculating the shortest distance from a point on the surface of the implantation area to a point on the surface of the deep bone tissue within the mesh model. Mucosal thickness is an important indicator for assessing soft tissue health and implant stability; an excessively thin mucosa may increase the risk of postoperative complications.

[0042] A pre-defined ResNet50 model was used to classify soft tissue surface features. The ResNet50 model is a deep convolutional neural network with strong feature extraction and classification capabilities. Trained on a large amount of historical soft tissue surface image data, it can identify and classify soft tissue surface features such as gingival margin curves and keratinized gingival width. Furthermore, based on the classification results, risk types such as thin gingival biotypes were identified. Patients with thin gingival biotypes are more prone to gingival recession and other problems after implant surgery, requiring special attention in preoperative planning.

[0043] The ICP (Iterative Closest Point) algorithm was used to register 3D point cloud data and 3D voxel data. The ICP algorithm iteratively searches for the optimal transformation relationship between the two sets of point cloud data, achieving precise spatial alignment. After registration, surface data of the soft tissue in the implantation area was extracted and combined with previously calculated mucosal thickness distribution and morphological parameters to generate soft tissue status data. This data details the health status of the soft tissue in the implantation area, existing risks, and corresponding recommendations, providing a soft tissue-related basis for implant planning.

[0044] Finally, the bone tissue status data and soft tissue status data are integrated to obtain oral cavity status data.

[0045] Through the above analysis methods, the present invention can provide comprehensive and accurate oral cavity status data, providing a precise data foundation for subsequent simulation analysis and determination of preoperative planning information.

[0046] In one embodiment of the present invention, the analysis of dynamic functional data in the detection data to obtain occlusal parameter data includes: extracting edge information from the dynamic functional data using the Canny edge detection method; identifying contact point coordinates in the edge information using the Hough transform method; calculating the contact area and distribution density based on the contact point coordinates; smoothing the dynamic functional data using the Kalman filter method to obtain motion trajectory data; fitting open and closed motion curves based on the motion trajectory data; calculating the condyle movement range based on the motion curves; and integrating the contact point coordinates, contact area, distribution density, motion curves, and condyle movement range to obtain occlusal parameter data.

[0047] Specifically, a combination of the Canny edge detection method and the Hough transform method is used to process static occlusal contact point image information from dynamic functional data. The Canny edge detection method accurately extracts edge information from the image, while the Hough transform method further processes this edge information to identify the location of the occlusal contact points, i.e., the contact point coordinates. By combining these two methods, the contact point coordinates are extracted from the dynamic functional data, and then the contact area and distribution density are calculated based on these coordinates. The contact area and distribution density reflect the extent and concentration of tooth contact during occlusion and are important parameters for assessing occlusal rationality.

[0048] Kalman filtering was employed to smooth the dynamic functional data. Kalman filtering effectively removes random noise from the data, resulting in smoother and more accurate motion trajectory data. Then, opening and closing motion curves were fitted based on the smoothed motion trajectory data. These curves visually reflect the patterns and characteristics of mandibular movement. Finally, the condylar range of motion was calculated through analysis of the motion curves. Condylar range of motion is an important indicator for evaluating temporomandibular joint function, and its data provides the foundation for subsequent dynamic occlusal simulation, including bone and soft tissue constraints and occlusal motion parameters.

[0049] In one embodiment of the present invention, the step of adjusting the preset occlusal simulation model according to oral state data and occlusal parameter data to obtain a first occlusal simulation model includes: matching oral state data with a preset three-dimensional oral template, and replacing the current three-dimensional oral template in the preset occlusal simulation model with a preset three-dimensional oral template with the highest matching degree; and adjusting the corresponding parameter data in the preset occlusal simulation model according to the occlusal parameter data to obtain the first occlusal model.

[0050] Specifically, the preset occlusal simulation model is a template model built based on a large amount of historical oral implant restoration data. This model includes information such as common oral anatomical structures, occlusal relationships, and implant parameters. The previously obtained oral condition data (bone tissue condition data and soft tissue condition data) is matched with the preset 3D oral template, and the current 3D oral template in the preset occlusal simulation model is replaced with the preset 3D oral template with the highest matching degree. Then, based on the occlusal parameter data, some key parameters in the preset occlusal simulation model are adjusted. For example, the mechanical parameters of bone tissue in the model are adjusted according to the bone density distribution and bone thickness; the mechanical parameters of soft tissue in the model are adjusted according to the mucosal thickness and morphological characteristics; and the trajectory and force of occlusal movements in the model are adjusted according to the occlusal parameter data. Through these adjustments, the constructed first occlusal simulation model more closely reflects the patient's actual oral condition.

[0051] In one embodiment of the present invention, the step of performing simulation analysis based on the first occlusal simulation model to obtain occlusal simulation data includes: using the first occlusal simulation model to simulate oral functional movement information after implantation to obtain occlusal simulation data.

[0052] Specifically, a pre-constructed first occlusal simulation model is used for simulation analysis. This analysis simulates the occlusal process after implant placement, including common oral functional movements such as chewing and swallowing. Through simulation analysis, occlusal simulation data is obtained, including parameters such as the maximum equivalent stress of the bone tissue surrounding the implant and the frequency of occlusal collisions. The maximum equivalent stress reflects the maximum stress borne by the bone tissue surrounding the implant during occlusion; excessive stress may lead to bone resorption and implant loosening. The frequency of occlusal collisions reflects the frequency of collisions between teeth during occlusion; excessively high collision frequencies may lead to tooth wear and uneven force distribution on the implant. Through the setup of this embodiment, the present invention can obtain this occlusal simulation data in advance, predict the stability of the implant during dynamic occlusion, and avoid postoperative occlusal trauma.

[0053] In one embodiment of the present invention, the step of using a multi-objective optimization algorithm to analyze occlusal simulation data to obtain initial implant planning information includes: analyzing the occlusal simulation data to determine the characteristic data and corresponding constraints of each of the implant stability target, occlusal function recovery target, soft tissue aesthetic target, and surgical safety target; using a non-dominated sorting genetic algorithm to iteratively optimize the characteristic data and corresponding constraints of each target to output a multi-objective balanced optimal solution; and forming initial implant planning information based on the multi-objective balanced optimal solution.

[0054] Specifically, the implant stability target uses the maximum equivalent stress of the bone tissue around the implant and the contact area between the bone and the implant as core indicators in the occlusal simulation data. The goal is to ensure that the maximum equivalent stress of the bone tissue around the implant is ≤30MPa, while the contact area between the bone and the implant is ≥70%, to guarantee long-term implant stability. For example, if the occlusal simulation data shows low bone density (200-300HU) in a potential implant area, the algorithm will prioritize "reducing the maximum equivalent stress" as a high-weight target to prevent bone tissue damage due to overload. The occlusal function recovery target is based on the occlusal collision frequency, the uniformity of occlusal contact point distribution, and the matching degree of condylar movement trajectory in the occlusal simulation data. Requirements: occlusal collision frequency ≤2 times / minute to avoid frequent collisions leading to implant loosening; uniformity of occlusal contact point distribution ≥80% to ensure uniform transmission of chewing force; and the deviation between the condylar movement trajectory after implantation and the preoperative dynamic functional data ≤0.5mm, in order to restore normal occlusal movement. Soft tissue aesthetic goals are set by combining mucosal thickness, gingival margin curve, keratinized gingival width from soft tissue condition analysis data with implant neck exposure from occlusal simulation data. For example, for anterior aesthetic zone implants, the mucosal thickness at the implant neck should be ≥2mm to avoid implant exposure due to gingival recession, and the deviation of the gingival margin curve from adjacent teeth should be ≤0.3mm to ensure visual harmony. Surgical safety targets are based on the spatial location data of key anatomical structures such as the mandibular nerve canal and the maxillary sinus floor. The targets are: a distance ≥2mm between the implant and the mandibular nerve canal to avoid nerve damage; and a distance ≥1mm between the implant and the maxillary sinus floor to avoid perforation of the sinus cavity. If occlusal simulation data shows that a planned location may be close to a dangerous structure, the algorithm will set the safety distance as a hard constraint target, prioritizing the elimination of risky solutions.

[0055] Constraints include anatomical constraints: the implant must not intrude into the mandibular nerve canal, maxillary sinus, or adjacent tooth roots; when using 3D anatomical data segmented based on the U-Net model, spatial no-go zones are defined. Bone volume constraints: the bone width of the implantation area must be ≥ the implant diameter + 2mm, leaving 1mm of bone wall on each side to ensure bone support; the bone height must be ≥ the implant length + 1mm to prevent the implant from penetrating the bone wall. Occlusal movement constraints are: after implantation, during mandibular protrusion and lateral movements, the gap between the implant and the opposing tooth must be ≥ 1mm to avoid occlusal interference during movement.

[0056] In one embodiment of the present invention, a non-dominated sorting genetic algorithm is used to iteratively optimize the feature data and corresponding constraints of each objective, and output a multi-objective equilibrium optimal solution. This includes: generating an initial scheme population based on implant parameter ranges using a random sampling method; verifying the initial scheme population based on occlusion simulation data and constraints to obtain multiple feasible initial schemes; performing a crossover operation on the initial schemes to obtain new schemes; iteratively executing non-dominated sorting and crowding calculation on the initial and new schemes based on each objective, retaining schemes with high crowding in the comparison; and exiting the loop when the change in each objective is less than a preset range for three consecutive times and the proportion of feasible schemes in the population is greater than a corresponding preset value, outputting multiple multi-objective equilibrium optimal solutions.

[0057] Specifically, the population is initialized using a random sampling method based on the implant parameter range. For each initial scheme, a feasibility test is performed using preprocessed occlusion simulation data and constraints, eliminating schemes that violate hard constraints and retaining multiple feasible initial schemes. Then, a crossover operation is performed on the initial schemes to obtain new schemes. The process is iteratively executed to perform non-dominated sorting of the initial and new schemes according to four optimization objectives, calculate crowding degree, and retain schemes with high crowding degree. The loop exits when the change in the objectives is less than a preset range for three consecutive times and the proportion of feasible schemes in the population is greater than the corresponding preset value. Multiple multi-objective equilibrium optimal solutions are output. Multiple initial implant planning information is formed based on these multi-objective equilibrium optimal solutions.

[0058] In this embodiment, the multi-objective optimization algorithm comprehensively considers multiple objective factors, such as implant stability, bone tissue stress, occlusal function recovery, and aesthetics, seeking the optimal balance among these factors. By analyzing occlusal simulation data, parameters such as implant type, diameter, length, implantation location, and angle are determined, thereby obtaining initial implant planning information. For example, when selecting the implant diameter and length, the bone thickness and density of the implantation area are considered to ensure sufficient bone support; when determining the implantation location and angle, the occlusal motion trajectory and occlusal contact point distribution are considered to ensure uniform force distribution on the implant during occlusion, while also guaranteeing aesthetic results.

[0059] Through the configuration method of this embodiment, the present invention does not pursue the optimization of a single objective, but rather generates a comprehensive optimal implant planning scheme that conforms to clinical practice by balancing four major objectives: stability, function, aesthetics, and safety. Compared with traditional methods where doctors subjectively select implant parameters based on experience, the present invention can simulate occlusal data, avoiding subjective bias; it can simultaneously handle multi-dimensional objectives, taking into account both short-term surgical safety and long-term implant outcomes; and the generated initial planning information can be directly integrated with the intraoperative navigation system, providing data support for precise surgery, ultimately reducing the incidence of postoperative complications and improving the success rate of implant restoration.

[0060] In one embodiment of the present invention, the step of filtering initial implant planning information based on the acquired patient opinion information to obtain preoperative planning information includes: analyzing the patient opinion information and extracting key opinion features; selecting the planning information with the highest matching degree with the key opinion features from multiple initial implant planning information to obtain preoperative planning information.

[0061] Specifically, after obtaining the initial implant planning information, thorough communication is conducted with the patient to understand their needs, expectations, and opinions regarding implant type, restoration plan, etc. Patient feedback may include preferences for implant brands, treatment timeframes, and specific aesthetic expectations. Based on this feedback, the initial implant planning information is adjusted accordingly. For example, if the patient has high aesthetic requirements, the implant placement and angle may need to be adjusted to ensure the restored tooth harmonizes with the surrounding natural teeth in shape and alignment; if the patient wishes to shorten the treatment time, it may be necessary to select a suitable implant type and restoration plan, optimizing the treatment process while ensuring treatment effectiveness. After these adjustments, the final preoperative planning information is determined.

[0062] In one embodiment of the invention, deviation information between preoperative planning information and operational information is analyzed, and a warning is issued when the deviation information is not met. Specifically, the doctor performs procedures such as implant placement in the patient's oral cavity based on the determined preoperative planning information. During the operation, it is necessary to strictly follow the surgical operation specifications to ensure that every step is accurate. For example, during implant placement, the depth and angle of placement must be controlled to avoid damage to surrounding nerves, blood vessels, and bone tissue.

[0063] Intraoperative navigation systems and sensors acquire operational information in real time. The intraoperative navigation system provides real-time guidance to the surgeon based on preoperative planning information and real-time images of the patient's oral cavity, while simultaneously recording information such as the position, angle, and depth of the surgical procedure. Sensors, mounted on surgical instruments or inside the patient's mouth, collect data on forces and displacements during the surgical procedure, obtaining operational information. This real-time operational information is compared and analyzed with the preoperative planning information to calculate deviations. Deviations include deviations in position, angle, and depth. When a deviation exceeds a preset allowable range, i.e., when the deviation does not meet requirements, an immediate warning is issued. This warning can be communicated to the surgeon through sound, light, or screen prompts. Upon receiving the warning, the surgeon will promptly adjust the surgical procedure to correct deviations, ensuring the surgical operation meets preoperative planning requirements and improving the accuracy and safety of the surgery.

[0064] In one embodiment of the present invention, the method of using a postoperative evaluation model to analyze complete operational information and test data to obtain a comprehensive evaluation result includes: comparing bone tissue state data in the operational information and bone tissue state data in the test data to obtain bone tissue analysis results; determining postoperative occlusal parameter information based on the operational information; performing simulation analysis based on the postoperative occlusal parameter information to obtain occlusal recovery results; examining soft tissue morphology to obtain soft tissue evaluation results; and integrating bone tissue analysis results, occlusal recovery results, and soft tissue evaluation results to obtain a comprehensive evaluation result.

[0065] Specifically, complete operational information and test data are collected and input into the postoperative evaluation model. This model is an analytical framework built upon extensive postoperative follow-up data and clinical experience. It can comprehensively evaluate the effectiveness of implant surgery, including: assessing the stability and osseointegration of the bone tissue around the implant by comparing preoperative and postoperative bone tissue status data; evaluating the recovery of occlusal function by adjusting the current preset occlusal model based on postoperative occlusal parameter data and performing simulation analysis based on the current preset occlusal simulation model; and assessing the healing and aesthetic effects of soft tissues by examining their morphology and health condition.

[0066] The postoperative evaluation model analyzes and processes the input operational information and test data, and outputs evaluation results. The evaluation results will be presented in the form of a report, which will detail the success of the implant surgery, existing problems, such as slight bone resorption around the implant, slight deviation in occlusion, and corresponding improvement suggestions.

[0067] Based on the postoperative assessment results, a personalized maintenance strategy is generated for the patient. This strategy includes postoperative care instructions, a schedule of regular follow-up appointments, and dietary recommendations. For example, if the assessment shows good stability of the bone tissue around the implant but mild soft tissue inflammation, the maintenance strategy will include specific methods to enhance oral hygiene, such as using specific mouthwashes and dental floss, scheduling short-term follow-up appointments, and advice to avoid irritating foods. If the assessment shows a slight deviation in occlusion, the strategy will include a plan and schedule for subsequent occlusal adjustments. By developing a scientifically sound maintenance strategy, patients can better maintain the stability of their implants and oral health, extending the lifespan of the implants.

[0068] In one embodiment of the present invention, as shown in the appendix Figure 2 As shown, the present invention also provides a visual analysis-based oral and maxillofacial implant restoration system, which is used to perform the implant restoration method described in any of the above embodiments.

[0069] Specifically, the visual analysis-based oral and maxillofacial implant restoration system includes a data acquisition module, an occlusal simulation module, a preoperative planning module, a monitoring and early warning module, and a postoperative evaluation module. The data acquisition module is used to obtain the patient's test data.

[0070] The occlusion simulation module analyzes the bone and soft tissue states based on the detection data to obtain oral cavity state data; it also analyzes the dynamic functional data in the detection data to obtain occlusion parameter data; based on the oral cavity state data and occlusion parameter data, it adjusts the preset occlusion simulation model to obtain the first occlusion simulation model; and it performs simulation analysis based on the first occlusion simulation model to obtain occlusion simulation data.

[0071] The preoperative planning module uses a multi-objective optimization algorithm to analyze occlusal simulation data to obtain initial implant planning information; based on the obtained patient feedback, the initial implant planning information is filtered to obtain preoperative planning information.

[0072] The monitoring and early warning module operates based on preoperative planning information and acquires operational information in real time; it analyzes the deviation information between preoperative planning information and operational information, and issues early warning information when the deviation information is not met.

[0073] The postoperative assessment module analyzes complete operational information and test data using a postoperative assessment model after the operation is completed to obtain a comprehensive assessment result; and generates a maintenance strategy based on the comprehensive assessment result.

[0074] The oral and maxillofacial implant restoration system of this invention achieves data acquisition, status analysis, and dynamic occlusal simulation through the cooperation of the various modules mentioned above, thus determining preoperative planning information. During surgery, it provides real-time navigation for the surgical procedure and issues warnings for erroneous operations based on the preoperative planning information. After the procedure, it generates personalized follow-up plans tailored to the patient's condition, including precautions and follow-up schedules, based on intraoperative monitoring data and patient test data, achieving full-cycle management and helping to reduce the probability of complications. Furthermore, through the above methods, this invention solves the problem that complication prediction is often based on empirical judgment, which cannot combine intraoperative data and individual patient characteristics for accurate analysis, resulting in a lack of personalized follow-up plans.

[0075] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0076] It should be noted that in the description of this invention, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0077] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0078] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0079] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0080] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0081] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0082] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for oral and maxillofacial implant restoration based on visual analysis, characterized in that, include: Obtain patient test data; Based on the test data, the condition of bone tissue and soft tissue is analyzed to obtain oral cavity condition data; The dynamic functional data in the test data is analyzed to obtain occlusal parameter data; based on the oral cavity state data and occlusal parameter data, the preset occlusal simulation model is adjusted to obtain the first occlusal simulation model; simulation analysis is performed based on the first occlusal simulation model to obtain occlusal simulation data. A multi-objective optimization algorithm was used to analyze the occlusal simulation data to obtain initial implant planning information; Initial implant planning information is filtered based on the obtained patient feedback information to obtain preoperative planning information; The procedure is performed based on preoperative planning information, and operational information is obtained in real time. Analyze the deviation information between preoperative planning information and operational information, and issue an early warning when the deviation information is not met; After the procedure is completed, a postoperative evaluation model is used to analyze the complete operation information and test data to obtain a comprehensive evaluation result; a maintenance strategy is then generated based on the comprehensive evaluation result.

2. The planting remediation method according to claim 1, characterized in that, The process of analyzing bone and soft tissue conditions based on test data to obtain oral cavity condition data includes: The Gaussian filtering algorithm is used to preprocess the three-dimensional voxel data in the detection data to obtain preprocessed data; the U-Net deep learning model is used to segment the preprocessed data to obtain segmentation label data; the potential implantation areas in the segmentation label data are traversed and analyzed to obtain bone information; the bone information is compared and graded based on preset analysis standards and displayed as a heat map to obtain a bone condition heat map. The Poisson surface reconstruction method was used to process the dentition data and point cloud data in the detection data to generate a soft tissue surface mesh model. The shortest distance algorithm was used to analyze the soft tissue surface mesh model to obtain the mucosal thickness data of the potential implant area. The preset ResNet50 model was used to extract features and classify the soft tissue surface image data in the detection data to determine the classification results of the potential implant area. The bone condition heatmap, mucosal thickness data and classification results were integrated to obtain oral cavity status data.

3. The planting remediation method according to claim 1, characterized in that, The analysis of dynamic functional data in the aforementioned detection data to obtain occlusal parameter data includes: extracting edge information from the dynamic functional data using the Canny edge detection method; identifying contact point coordinates in the edge information using the Hough transform method; calculating the contact area and distribution density based on the contact point coordinates; smoothing the dynamic functional data using the Kalman filter method to obtain motion trajectory data; fitting open and closed motion curves based on the motion trajectory data; calculating the condyle movement range based on the motion curves; and integrating the contact point coordinates, contact area, distribution density, motion curves, and condyle movement range to obtain occlusal parameter data.

4. The planting remediation method according to claim 1, characterized in that, The step of adjusting a preset occlusal simulation model based on oral cavity state data and occlusal parameter data to obtain a first occlusal simulation model includes: matching oral cavity state data with a preset three-dimensional oral cavity template, and replacing the current three-dimensional oral cavity template in the preset occlusal simulation model with a preset three-dimensional oral cavity template with the highest matching degree; and adjusting the corresponding parameter data in the preset occlusal simulation model based on the occlusal parameter data to obtain the first occlusal model.

5. The planting remediation method according to claim 1, characterized in that, The aforementioned simulation analysis based on the first occlusal simulation model to obtain occlusal simulation data includes: using the first occlusal simulation model to simulate oral functional movement information after implantation to obtain occlusal simulation data.

6. The planting remediation method according to claim 1, characterized in that, The method of using a multi-objective optimization algorithm to analyze occlusal simulation data and obtain initial implant planning information includes: analyzing the occlusal simulation data to determine the characteristic data and corresponding constraints of each of the implant stability target, occlusal function recovery target, soft tissue aesthetic target, and surgical safety target; using a non-dominated sorting genetic algorithm to iteratively optimize the characteristic data and corresponding constraints of each target and output the optimal solution for multi-objective balance; and forming the initial implant planning information based on the optimal solution for multi-objective balance.

7. The planting remediation method according to claim 6, characterized in that, The method employing a non-dominated sorting genetic algorithm to iteratively optimize the feature data and corresponding constraints of each objective, and output a multi-objective equilibrium optimal solution, includes: generating an initial scheme population based on the implant parameter range using a random sampling method; verifying the initial scheme population based on occlusion simulation data and constraints to obtain multiple feasible initial schemes; performing a crossover operation on the initial schemes to obtain new schemes; iteratively executing non-dominated sorting and crowding calculation on the initial and new schemes based on each objective, retaining the scheme with the highest crowding in this comparison; and exiting the loop when the change range of each objective is less than a preset range for three consecutive times and the proportion of feasible schemes in the population is greater than the corresponding preset value, outputting multiple multi-objective equilibrium optimal solutions.

8. The planting remediation method according to claim 1, characterized in that, The process of filtering initial implant planning information based on acquired patient feedback information to obtain preoperative planning information includes: analyzing patient feedback information and extracting key feedback features; selecting the planning information with the highest matching degree with the key feedback features from multiple initial implant planning information to obtain preoperative planning information.

9. The planting remediation method according to claim 1, characterized in that, The aforementioned postoperative evaluation model analyzes complete operational information and test data to obtain a comprehensive evaluation result, including: comparing bone tissue state data in the operational information and bone tissue state data in the test data to obtain bone tissue analysis results; determining postoperative occlusal parameter information based on the operational information; performing simulation analysis based on the postoperative occlusal parameter information to obtain occlusal recovery results; examining soft tissue morphology to obtain soft tissue evaluation results; and integrating bone tissue analysis results, occlusal recovery results, and soft tissue evaluation results to obtain a comprehensive evaluation result.

10. A visual analysis-based oral and maxillofacial implant restoration system, characterized in that, The oral and maxillofacial implant restoration system is used to perform the implant restoration method as described in any one of claims 1 to 9.

Images