Virtual-real fusion system and teaching method for periodontal subosseous pocket teaching

Abstract

The invention provides a virtual-real fusion system for periodontal subbone pocket teaching and a teaching method. The virtual-real fusion system comprises a learning task construction module, a virtual simulation module, an identification module, an entity operation module, a deconstruction module, an image acquisition module, an image processing module and an intelligent analysis module. A course is issued through a learning task construction module, and a user learns the virtual model under the guidance of an identification module; and then under the guidance of the deconstruction module, physical splicing is carried out by utilizing the entity operation module. The system collects operation data of a user in real time through an image processing or sensor technology, and comprehensively calculates a grasp degree score by combining the correct rate and time consumption of the user in the recognition stage and the splicing similarity, hesitation time and taking sequence of the user in the deconstruction stage. According to the system, through a virtuality and reality combined teaching mode, a user can more clearly master the structures of different types of periodontal subosseous pockets, and the learning efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of periodontology, and particularly to a virtual-real integrated system and teaching method for teaching periodontal subosseous pockets. Background Technology

[0002] Periodontitis is the leading cause of tooth loss in adults, with a prevalence rate exceeding 50%. It has become the most common and most pressing oral health problem.

[0003] For dentists, especially periodontists and general dentists, in-depth learning and mastery of the diagnosis, classification, and treatment techniques for periodontal infrabone pockets are crucial for improving clinical outcomes and preserving affected teeth. However, current teaching on periodontal infrabone pockets primarily relies on traditional methods such as images, video data, and clinical instruction. Trainees lack a three-dimensional understanding and struggle to construct a three-dimensional spatial structure of the infrabone pocket. Their understanding of the depth, morphology, and relationship with adjacent teeth and important anatomical structures is abstract. Furthermore, inconsistent clinical instruction levels and a lack of real-time and quantitative feedback result in varying levels of trainees' understanding of infrabone pockets. This leads to a long transition period from theoretical learning to clinical practice and low training efficiency.

[0004] Therefore, there is a need for a teaching system that can deeply integrate the accuracy and repeatability of virtual information with real clinical operation scenarios and provide objective quantitative assessments. Summary of the Invention

[0005] In view of this, this application provides a virtual-real fusion system and teaching method for teaching subperiodontal pockets, which improves the user's learning efficiency through virtual-real fusion teaching.

[0006] Firstly, this application provides a virtual-real fusion system for teaching infrapermanent periodontal pockets, including: The learning task building module is used to publish learning courses at different stages based on the user's learning progress; The virtual simulation module stores several virtual models of the periodontal bone pocket, including virtual models reconstructed based on real data and virtual building block models. Each of the virtual building block models can be broken down into several virtual building blocks. The recognition module is used to control the virtual simulation module to display virtual models based on the learning courses selected by the user, and to guide the user to perform observation and classification recognition operations. The physical operation module includes several physical block assembly models corresponding to the virtual block assembly model, and each of the physical block assembly models can be broken down into several blocks. The deconstruction module is used to prompt the user to assemble several building blocks in the entity operation module based on the learning courses selected by the user. The image acquisition module is used to acquire images of the assembled block model obtained by the user. The image processing module is used to compare the image of the physical block model obtained by the user after assembly with the image of the virtual block model from the same perspective and calculate the similarity. The intelligent analysis module is used to calculate the user's mastery score based on the accuracy and time taken by the user in selecting the virtual building block model during the recognition stage, as well as the number of attempts and similarity during the deconstruction stage.

[0007] Optionally, the virtual block model and the physical block model are formed according to different types of periodontal bone pocket structures.

[0008] Optionally, the identification operation includes: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

[0009] Optionally, the learning task building module is also used for: Based on the user's mastery score, supplementary learning tasks are created to address the user's weaknesses.

[0010] Optionally, it also includes: The positioning module is set on each entity's block; The operation feedback module is used to determine the information of the building block picked up by the user based on the positioning module, and then provide feedback to the user.

[0011] Optionally, the intelligent analysis module is further used for: The positioning module records the user's block-building process, hesitation time, and block-taking order. This data is then used to further evaluate the user's skill mastery.

[0012] Optionally, it also includes: The error correction module overlays the image of the user-assembled block model with a standard outline image to display the error and correct areas in the image of the user-assembled block model.

[0013] Secondly, this application provides a teaching method for teaching infraperineal periodontal pockets using a virtual-real integrated system, including: In response to the learning courses selected by the user, load the virtual model corresponding to the learning courses; Based on the learning course, users are guided to perform recognition operations on the virtual model; The learning course prompts the user to assemble several solid wooden blocks. Acquire images of the solid block building model obtained after the user assembles the blocks; The similarity is calculated by comparing the image of the block model obtained by the user after assembly with the image of the virtual block model from the same viewpoint. The user's mastery score is calculated based on the accuracy and time taken by the user in selecting the virtual building block model during the recognition phase, as well as the number of attempts and similarity during the deconstruction phase.

[0014] Optionally, the identification operation includes: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

[0015] Optionally, it also includes: Record the user's block-building process, hesitation time, and block-taking order, and combine the block-building process, hesitation time, and block-taking order to further evaluate the user's skill mastery depth.

[0016] The beneficial effects of the technical solution provided in this application include: This application provides a virtual-real fusion system for teaching infrapermanent periodontal pockets, including a learning task construction module, a virtual simulation module, a recognition module, a physical operation module, a deconstruction module, an image acquisition module, an image processing module, and an intelligent analysis module. Through learning tasks in the learning task construction module, users learn via the virtual simulation module under the guidance of the recognition module, obtaining the user's accuracy and time taken in the recognition stage. Subsequently, under the guidance of the deconstruction module, users assemble physical block models using the physical operation module, and the similarity is calculated based on the image of the assembled physical block model. Finally, based on the similarity, the number of attempts, the user's accuracy and time taken in the recognition process, the user's mastery score is determined. This system, through a virtual-real fusion teaching mode, allows users to gain a clearer understanding of the structure of different types of infrapermanent periodontal pockets, improving learning efficiency. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a structural block diagram of a virtual-real fusion system for teaching infradentine pockets provided in an embodiment of this application; Figure 2This is a structural block diagram of a virtual simulation module provided in one embodiment of this application; Figure 3 This is a structural schematic diagram of a solid block-building model provided in an embodiment of this application; Figure 4 A structural block diagram of a virtual-real fusion system for teaching subperiodontal pockets provided in another embodiment of this application; Figure 5 A flowchart illustrating the teaching method of a virtual-real fusion system for teaching infradentine periodontal pockets, provided in an embodiment of this application.

[0019] The attached figures are labeled as follows: 1: Learning Task Construction Module; 2: Virtual Simulation Module; 21: 3D Model Library Unit; 22: Scene Rendering Engine; 23: Interactive Instruction Unit; 3: Recognition Module; 4: Entity Operation Module; 41: Base Alveolar Bone; 42: First Molar; 43: Second Premolar; 44: Middle Alveolar Bone; 45: Upper Alveolar Bone; 451: Small Building Blocks; 46: Gingival Components; 461: First Gingival Component; 462: Second Gingival Component; 463: Biological Barrier Membrane Component; 47: Groove; 48: Through Hole; 49: Trench; 5: Deconstruction Module; 6: Image Acquisition Module; 7: Image Processing Module; 8: Intelligent Analysis Module; 9: Positioning Module; 10: Operation Feedback Module; 11: Error Correction Module. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] Figure 1 This is a structural block diagram of a virtual-real fusion system for teaching infrapermanent periodontal pockets according to an embodiment of this application. See also... Figure 1 ,include: Learning task building module 1 is used to publish learning courses at different stages according to the user's learning progress; The virtual simulation module 2 stores several virtual models of the periodontal bone pocket, including virtual models reconstructed based on real data and virtual building block models. Each of the virtual building block models can be broken down into several virtual building blocks. The recognition module 3 is used to control the virtual simulation module 2 to display the virtual model according to the learning course selected by the user, and to guide the user to perform observation and classification recognition operations. The entity operation module 4 includes several real block assembly models corresponding to the virtual block assembly model, and each of the real block assembly models can be broken down into several blocks. Deconstruction module 5 is used to prompt the user to assemble several building blocks in entity operation module 4 based on the learning course selected by the user. Image acquisition module 6 is used to acquire images of the building block model obtained by the user after assembly; Image processing module 7 is used to compare the image of the physical block assembly model obtained by the user with the image of the virtual block assembly model from the same perspective and calculate the similarity. The intelligent analysis module 8 is used to calculate the user's mastery score based on the user's accuracy and time in selecting the virtual building block model during the recognition stage, as well as the number of attempts and similarity during the deconstruction stage.

[0022] In some examples, learning task building module 1 is also used for: Based on the user's mastery score, supplementary learning tasks are created to address the user's weaknesses.

[0023] In some examples, the learning task is based on the structure of each different type of periodontal infrabone pocket. The types of periodontal infrabone pocket structures include one-walled, two-walled, three-walled, four-walled, and mixed-walled pockets.

[0024] In some examples, a question bank is first created. Then, based on the types of infraperineal pocket structures, initial learning questions are developed. Each learning question then includes multiple-choice questions with images to guide users in identifying the different types of infraperineal pocket structures. Furthermore, if a user's error rate is high after completing all the initial learning questions, targeted learning questions will be developed based on the user's weaknesses.

[0025] In some examples, the learning task building module 1 can also create exclusive learning accounts for users, assign different learning permissions and course access levels according to the student's identity; and be responsible for the design, storage and publication of the teaching curriculum system, and track and visualize the learning status of each student in real time.

[0026] In some examples, the virtual simulation module 2 is used to build and render high-precision virtual scenes that are applicable to multiple terminals (such as mobile phones, tablets, or computers).

[0027] See Figure 2 In some examples, the virtual simulation module 2 includes a 3D model library unit 21, a scene rendering engine 22, and an interactive instruction unit 23.

[0028] The 3D model library unit 21 stores virtual building blocks and virtual real models of the periodontal bone pocket (such as one wall, two walls, three walls, four walls, and mixed pockets).

[0029] The three-dimensional virtual building block model of the infraperitoneal periodontium breaks down the overall structure of the infraperitoneal periodontium into virtual small building blocks, which can be assembled or disassembled according to the absorption rate and type of periodontitis.

[0030] Scene rendering engine 22 is used to load and display 3D models, allowing users to rotate, scale, and section the models at will, and observe the three-dimensional spatial structure of the subosseous bag from multiple angles.

[0031] The interactive instruction unit 23 is used to receive user input instructions (such as via stylus or gestures) and complete operations such as bone defect classification in a virtual environment.

[0032] In some examples, virtual models and physical modular models are formed based on different types of periodontal pocket structures.

[0033] In some examples, the 3D model library unit 21 can be developed based on the Unity 2022 LTS engine. The 3D model library imports clinically collected CBCT data into Mimics Innovation Suite software, performs 3D reconstruction, mesh optimization, and smoothing, generates high-fidelity ".FBX" format model files, and imports them into the Unity project for categorized storage.

[0034] Model Classification: In Unity, two types of model assets are created using ScriptableObject: (a) Realistic Models: Different types of sub-bags. (b) Decomposed Models: Sub-bags of the same type are assembled as the "standard answer".

[0035] Scene rendering engine 22 can utilize Unity's AR Foundation framework to call the device's native AR capabilities. Gesture interactions are implemented by writing C# scripts: RotateGesture (single-finger swipe to rotate), PinchGesture (two-finger pinch to zoom), and ToggleSlicePlane (UI button triggered slice).

[0036] Interactive instruction unit 23 responds to screen touch events and implements measurement tools (calculating distance by detecting collision points via Raycast) and classification multiple-choice questions (implemented through the UI system) on the virtual model. Users can learn and identify different types of subosseous sacs on the virtual simulation software.

[0037] In some examples, the recognition module 3 is used to guide the user to perform a recognition operation on the virtual model based on the learning task, the recognition operation including: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

[0038] By first recognizing a virtual model of the infraorbital pocket, users gain a preliminary understanding of its structure. For example, if the learning course focuses on the structure of one side of the infraorbital pocket, the recognition module guides the user to identify the structure of that side based on a virtual modular model of the infraorbital pocket.

[0039] In some examples, the solid block model in the solid manipulation module 4 is manufactured using 3D printing technology.

[0040] See Figure 3 As an example, solid modular building models include: The base alveolar bone 41 has several grooves 47 provided thereon; The first molar 42 is set to correspond to the groove 47 on the alveolar bone 41 of the base; The second premolar 43 is positioned to correspond to the groove 47 on the alveolar bone 41 of the base; The middle alveolar bone 44 is provided with several through holes 48 and subosseous pocket components. The through holes 48 are provided in a one-to-one correspondence with the grooves 47 on the base alveolar bone 41. The upper alveolar bone 45 is provided with several through holes 48 and several subosseous pocket components. The through holes 48 are provided in a one-to-one correspondence with the grooves 47 on the base alveolar bone 41. The gingival component 46 includes a first gingival element 461, a second gingival element 462, and a bio-barrier membrane element 463. The first gingival element 461 is provided with a plurality of through holes 48, which correspond one-to-one with the grooves on the alveolar bone 41 of the base. A groove 49 is provided between any two adjacent through holes on the first gingival element 461. The second gingival element 462 and the bio-barrier membrane element 463 are stacked in the groove 49.

[0041] By adjusting the combination of subosseous pocket components, different types of periodontal subosseous pocket structures can be demonstrated.

[0042] certainly, Figure 3 This is merely an example of a physical building block provided in this application. In reality, building blocks can be smaller structures, and each physical building block model can be broken down into a greater number of building blocks.

[0043] In some examples, Figure 3 The upper alveolar bone 45 has six small building blocks 451, and the middle alveolar bone 44 also has one small building block 451. Of course, Figure 3 The positions of the small building blocks in this application are only examples provided in this application. Small building blocks can also be set on other structures. This application does not limit the position and number of building blocks.

[0044] In some examples, magnets can be further provided on the top and bottom of the blocks to attract and hold the blocks in place during assembly.

[0045] In some examples, after the user clicks the "Submit" button in the deconstruction module 5, the camera of the image acquisition module 6 captures a top-view photo containing the assembled model. The image processing module 7 then calculates the similarity based on the top-view photo of the assembled model, as follows: The first step, preprocessing: convert the photo to grayscale, apply Gaussian blur (kernel size 5x5), and perform Canny edge detection.

[0046] The second step is contour extraction: use the FindContours function to extract the outermost contour of the solid model.

[0047] The third step is virtual-real registration: the extracted contour is compared with the "standard answer" contour image pre-rendered from the same viewpoint in the virtual simulation module.

[0048] Step 4: Similarity Calculation: The Structural Similarity Index (SSIM) algorithm is used to calculate the similarity between the two contour images. A threshold of SSIM ≥ 0.90 is set as acceptable.

[0049] After calculating the similarity score, the intelligent analysis module then calculates the mastery score, as follows: The intelligent analysis module uses machine learning algorithms (such as decision trees) to form a mastery score prediction model. By recording the student's choices and time spent in the "recognition" stage (virtual model classification), as well as the number of attempts and the final SSIM score in the "deconstruction" stage (physical assembly), a mastery score is calculated using the mastery score prediction model.

[0050] See Figure 4 In some examples, it also includes: Positioning module 9 is set on each entity's block; The operation feedback module 10 is used to determine the information of the building block picked up by the user based on the positioning module 9, and then provide feedback to the user.

[0051] In some examples, the positioning module determines the order of the blocks selected by the user, and if the user selects the wrong block, the user is reminded via voice.

[0052] In some examples, the positioning module can be an ultra-wideband (UWB) tag embedded inside the block.

[0053] The system constructs a high-precision 3D positioning space by embedding ultra-wideband (UWB) tags within each building block and deploying a UWB base station array on the control panel base. The system tracks the coordinates (X, Y, Z) and orientation of each module in real time at a frequency of 10 times per second. When a student picks up or moves a module, the data processing module (deployed on Alibaba Cloud ECS) compares the real-time streaming data with the virtual standard model. If the system detects that the module's shape differs from the target by more than a threshold, or that the movement trajectory deviates from the efficient path, it will immediately issue a voice prompt through the mobile terminal's audio system, such as, "The module you picked up appears to be incorrect; please refer to the virtual model on the left," enabling real-time error correction and guidance during operation.

[0054] In some examples, the intelligent analysis module 8 is also used for: The positioning module 9 records the user's block assembly process, hesitation time, and block retrieval order. The user's skill mastery depth is further evaluated by combining the block assembly process, hesitation time, and block retrieval order.

[0055] Specifically, the system goes beyond simply recording the final result and focuses on collecting multi-dimensional process data, including the operation path trajectory reconstructed through UWB data, the hesitation time determined when the module is stationary or moving at low speed, and the retrieval order defined by the timing of when the module is picked up.

[0056] After this data is uploaded to the cloud, it is analyzed using machine learning algorithms (such as XGBoost).

[0057] Specifically, by recording the length of the movement path of several building blocks during the assembly process, as well as the curvature of the movement path, and combining this with a machine learning model, the user's operational smoothness is evaluated.

[0058] By recording the movement time of each block (as hesitation time), the user's decision-making bottlenecks are identified based on the hesitation time of the blocks. Through the user's decision-making bottlenecks, further feedback is provided on the depth of the user's skill mastery.

[0059] By analyzing the order in which blocks are picked up, it's determined whether the learner is using a "structured strategy" or a "trial and error strategy." The specific block the user picks is determined by the coordinate changes of the blocks. Each block has a specific assembly priority; blocks with the same priority can be arranged in any order, but the key is to determine the assembly order of blocks with different priorities, and then adjust accordingly.

[0060] Based on this, the system can more accurately assess each student's operating habits, spatial cognition strategies, and depth of skill mastery, rather than relying solely on the final score.

[0061] In some examples, it also includes: Error correction module 11 is used to overlay the image of the block model obtained by the user after assembly with the standard outline image to display the error area and the correct area in the image of the block model obtained by the user after assembly.

[0062] In some examples, the error correction module 11 summarizes the results of the student's subosseous pouch type identification in the virtual software and points out the subosseous pouch types that need to be reviewed in detail. It then overlays the student's photograph of the actual object with the standard outline in a semi-transparent manner. Matching areas are highlighted in green, while error areas are highlighted in flashing red, with text prompts such as "Move the red highlighted module to the right, please adjust."

[0063] Figure 3 This is a flowchart illustrating a teaching method using a virtual-real fusion system for teaching infradentine pockets, as provided in an embodiment of this application. See also... Figure 3 ,include: S101. In response to the learning course selected by the user, load the virtual building block model corresponding to the learning course.

[0064] S102. Based on the learning course, guide the user to perform recognition operations on the virtual building block model.

[0065] In some examples, the identification operation includes: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

[0066] S103. Based on the learning course prompts, the user assembles several solid wooden blocks.

[0067] S104. Obtain an image of the solid block model assembled by the user.

[0068] S105. Compare the image of the building block model obtained by the user after assembly with the image of the virtual building block model from the same viewpoint, and calculate the similarity.

[0069] In some examples, step S105 includes: The first step, preprocessing: convert the photo to grayscale, apply Gaussian blur (kernel size 5x5), and perform Canny edge detection.

[0070] The second step is contour extraction: use the FindContours function to extract the outermost contour of the solid model.

[0071] The third step is virtual-real registration: the extracted contour is compared with the "standard answer" contour image pre-rendered from the same viewpoint in the virtual simulation module.

[0072] Step 4: Similarity Calculation: The Structural Similarity Index (SSIM) algorithm is used to calculate the similarity between the two contour images. A threshold of SSIM ≥ 0.90 is set as acceptable.

[0073] S106. Calculate the user's mastery score based on the accuracy and time taken by the user in selecting the virtual building block model during the recognition stage, as well as the number of attempts and similarity during the deconstruction stage.

[0074] In some examples, step S106 includes: The intelligent analysis module uses machine learning algorithms (such as decision trees) to form a mastery score prediction model. By recording the student's choices and time spent in the "recognition" stage (virtual model classification), as well as the number of attempts and the final SSIM score in the "deconstruction" stage (physical assembly), a mastery score is calculated using the mastery score prediction model.

[0075] In some examples, it also includes: Record the user's block-building process, hesitation time, and block-taking order, and combine the block-building process, hesitation time, and block-taking order to further evaluate the user's skill mastery depth.

[0076] Specifically, by recording the length of the movement path of several building blocks during the assembly process, as well as the curvature of the movement path, and combining this with a machine learning model, the user's operational smoothness is evaluated.

[0077] By recording the movement time of each block (as hesitation time), the user's decision-making bottlenecks are identified based on the hesitation time of the blocks. Through the user's decision-making bottlenecks, further feedback is provided on the depth of the user's skill mastery.

[0078] By analyzing the order in which blocks are picked up, it's determined whether the learner is using a "structured strategy" or a "trial and error strategy." The specific block the user picks is determined by the coordinate changes of the blocks. Each block has a specific assembly priority; blocks with the same priority can be arranged in any order, but the key is to determine the assembly order of blocks with different priorities, and then adjust accordingly.

[0079] Based on this, the system can more accurately assess each student's operating habits, spatial cognition strategies, and depth of skill mastery, rather than relying solely on the final score.

[0080] Finally, it should be noted that, compared with the prior art, the virtual-real fusion system for teaching infraperineal periodontal pockets provided by this invention has the following significant advantages and beneficial effects: (1) Significantly improves students' understanding of infrabone pockets, doubling teaching efficiency. This invention transforms abstract, two-dimensional knowledge of infrabone pockets into an intuitive, three-dimensional spatial structure through a high-precision virtual 3D model and a tactile, assembleable, and realistically enlarged physical model, completely solving the core pain point of students' "inability to imagine and understand". Among fourth-year undergraduate students, compared with traditional teaching that relies on static pictures and verbal descriptions, the recognition rate of infrabone pockets using the virtual-real fusion system for teaching infrabone pockets increased from 78.27% to 98.98% in the same teaching time.

[0081] (2) Achieve standardized, repeatable, and low-cost training, breaking the limitations of teaching resources. The system enables each trainee to practice different types of subosseous pouches repeatedly and multiple times without real patients and specific instructors. It overcomes the drawbacks of traditional clinical teaching, which relies on luck for special cases and has limited hands-on opportunities, and greatly improves the efficiency of training resource utilization and educational equity.

[0082] (3) Provide personalized teaching and achieve individualized teaching through objective, real-time and quantitative assessment and feedback. Through the information input of physical models and intelligent comparison algorithms, the system can capture the operational details of students (such as the accuracy of the assembly position) and immediately provide visual and data-driven feedback (such as "one bone bag module is placed incorrectly").

[0083] (4) The system analyzes the performance of students at each stage, automatically diagnoses their knowledge weaknesses, and dynamically adjusts the content and difficulty of subsequent training to ensure the accuracy and efficiency of the investment of teaching resources.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A virtual-real fusion system for teaching infrapermanent periodontal pockets, characterized in that, include: The learning task building module is used to publish learning courses at different stages based on the user's learning progress; The virtual simulation module stores several virtual models of the periodontal bone pocket, including virtual models reconstructed based on real data and virtual building block models. Each of the virtual building block models can be broken down into several virtual building blocks. The recognition module is used to control the virtual simulation module to display virtual models based on the learning courses selected by the user, and to guide the user to perform observation and classification recognition operations. The physical operation module includes several physical block assembly models corresponding to the virtual block assembly model, and each of the physical block assembly models can be broken down into several blocks. The deconstruction module is used to prompt the user to assemble several building blocks in the entity operation module based on the learning courses selected by the user. The image acquisition module is used to acquire images of the assembled block model obtained by the user. The image processing module is used to compare the image of the physical block model obtained by the user after assembly with the image of the virtual block model from the same perspective and calculate the similarity. The intelligent analysis module is used to calculate the user's mastery score based on the accuracy and time taken by the user in selecting the virtual building block model during the recognition stage, as well as the number of attempts and similarity during the deconstruction stage.

2. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 1, characterized in that, Virtual models and physical modular assembly models are formed based on different types of periodontal bone pocket structures.

3. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 1, characterized in that, The identification operation includes: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

4. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 1, characterized in that, The learning task building module is also used for: Based on the user's mastery score, supplementary learning tasks are created to address the user's weaknesses.

5. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 1, characterized in that, Also includes: The positioning module is set on each entity's block; The operation feedback module is used to determine the information of the building block picked up by the user based on the positioning module, and then provide feedback to the user.

6. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 5, characterized in that, The intelligent analysis module is also used for: The positioning module records the user's block-building process, hesitation time, and block-taking order. This data is then used to further evaluate the user's skill mastery.

7. The virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 1, characterized in that, Also includes: The error correction module overlays the image of the user-assembled block model with a standard outline image to display the error and correct areas in the image of the user-assembled block model.

8. A teaching method using a virtual-real integrated system for teaching infraperineal periodontal pockets, characterized in that, include: In response to the learning courses selected by the user, load the virtual model corresponding to the learning courses; Based on the learning course, users are guided to perform recognition operations on the virtual model; The learning course prompts the user to assemble several solid wooden blocks. Acquire images of the solid block building model obtained after the user assembles the blocks; The similarity is calculated by comparing the image of the block model obtained by the user after assembly with the image of the virtual block model from the same viewpoint. The user's mastery score is calculated based on the accuracy and time taken by the user in selecting the virtual building block model during the recognition phase, as well as the number of attempts and similarity during the deconstruction phase.

9. The teaching method of the virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 8, characterized in that, The identification operation includes: The learning course generates several multiple-choice questions. Each multiple-choice question uses several images of the periodontal pockets for the user to select from, and the user's selection time and accuracy are recorded.

10. The teaching method of the virtual-real fusion system for teaching infraperineal periodontal pockets according to claim 9, characterized in that, Also includes: Record the user's block-building process, hesitation time, and block-taking order, and combine the block-building process, hesitation time, and block-taking order to further evaluate the user's skill mastery depth.

Images