Medullary space retention crown manufacturing method and system based on artificial intelligence

Abstract

The invention relates to the technical field of dental crown manufacturing, in particular to an artificial intelligence-based medullary space retention crown manufacturing method and system, and the artificial intelligence-based medullary space retention crown manufacturing method comprises the following steps: obtaining an oral cavity CBCT image and a mouth scanning model; synthesizing the oral cavity CBCT image and the oral scanning model into a synthetic model; inputting the synthetic model into the trained AI model, and generating a medullary space retention crown model by the AI model based on the synthetic model; performing a stress simulation test on the medullary space retention crown model, and regenerating the medullary space retention crown model when the stress of any region of the medullary space retention crown model is greater than a preset value; performing a simulation try-on test on the medullary space retention crown model qualified in the simulation stress test; adjusting the medullary space retention crown model according to the result of the simulation try-on test; inputting the adjusted medullary space retention crown model into a printing device, and printing to generate a semi-finished medullary space retention crown; and curing the semi-finished medullary space retention crown to generate a finished medullary space retention crown. The patient experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of crown manufacturing, and specifically relates to a method and system for manufacturing a pulp-retained crown based on artificial intelligence. Background Art

[0002] In the field of oral restoration, the pulp-retained crown is an important tooth restoration method. Because it can utilize the pulp cavity structure of the remaining tooth tissue to provide retention force, it plays a key role in the restoration of tooth defects. The traditional manufacturing of pulp-retained crowns mainly relies on manual manufacturing methods. In this method, a series of processes such as tooth preparation, impression taking, model pouring, and wax pattern making are completed through the manual operation of doctors.

[0003] The existing manual manufacturing method of pulp-retained crowns has significant defects, which are specifically as follows: 1. During the manual manufacturing process, many links rely on the personal experience and operation proficiency of professional dental technicians, and it is difficult to achieve precise quantitative control; 2. The manual manufacturing process is cumbersome. For example, the wax pattern needs to be repeatedly modified, the entire manufacturing cycle is long, and the production efficiency is low. As a result, patients need to make multiple trips to medical institutions, increasing the time cost of patients and affecting the patient's medical experience; 3. Due to the lack of precise digital design guidance, aspects such as the shape of the restoration and the fit between the restoration and the pulp cavity are subjectively judged by doctors alone, which is prone to human errors, resulting in poor fit between the restoration and the patient's tooth tissue, and often requires more on-site adjustments, prolonging the patient's visit time.

[0004] Therefore, how to solve the above deficiencies of the existing technology has become the subject to be studied and solved by the present invention. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for manufacturing a pulp-retained crown based on artificial intelligence.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is: A method for manufacturing a pulp-retained crown based on artificial intelligence includes: Step 1, obtaining oral CBCT images and intraoral scan models; Step 2, synthesizing the oral CBCT images and the intraoral scan models into a synthesized model; Step 3, inputting the synthesized model into a trained AI model, and generating a pulp-retained crown model by the AI model based on the synthesized model; Step 4, performing a simulated stress test on the pulp-retained crown model. When the stress in any area of the pulp-retained crown model is greater than a predetermined value, regenerating the pulp-retained crown model; Step 5: Perform a simulated fitting test on the endodontic retention crown model that has passed the simulated stress test; Step 6: Adjust the endodontic retention crown model according to the results of the simulated fitting test; Step 7: Input the adjusted endodontic retention crown model into a printing device and print to generate a semi-finished endodontic retention crown; Step 8: Cure the semi-finished endodontic retention crown to generate a finished endodontic retention crown.

[0007] The CBCT image and the intraoral scan model are existing settings and are briefly introduced here: The CBCT image (cone beam computed tomography) belongs to radiological imaging technology and provides hard tissue information such as the morphology of the jawbone, the course of the nerve canal, and the position of the maxillary sinus through three-dimensional X-ray scanning; the intraoral scan model (oral scan data) belongs to optical three-dimensional surface scanning technology, which is radiation-free and can quickly obtain the surface morphology of the tooth crown and gingiva, and is suitable for prosthetic design and orthodontic simulation; the two are synergistically applied through software fusion. The crown data of the intraoral scan is combined with the intraosseous structure data of the CBCT to generate a complete three-dimensional model; in addition, two-dimensional radiological images such as periapical films and panoramic films can be introduced.

[0008] The training method of the AI model is as follows: Collect the three-dimensional data of the jawbone scanned by CBCT and the optical data of the intraoral scan; Have oral experts mark the pulp cavity boundary, margin design type (butt joint type / wrap-around type), and material selection (such as lithium disilicate ceramic); Use a dual-channel 3D CNN to process the CBCT voxel data and the intraoral scan point cloud data, and fuse the features through cross-attention; Build a built-in mechanical simulation module to calculate the stress distribution (such as von Mises stress) in real time to constrain the rationality of the design; Use historical restoration cases for supervised learning to optimize the design error and mechanical compliance; Perform reinforcement learning optimization based on the clinical success rate and provide manual correction feedback for complex cases.

[0009] The simulated stress test is mainly realized through finite element analysis (FEA), which will not be elaborated here. When the stress in any area of the endodontic retention crown model is greater than the predetermined value, a new endodontic retention crown model is regenerated, specifically by updating the surrogate model and calculating after marking the defective part.

[0010] The simulated fitting test is realized through virtual fitting technology, and the technical implementation principle is as follows: The restoration model (endodontic retention crown model) is superimposed on the patient's oral digital model (synthetic model) through augmented reality technology.

[0011] When printing the semi-finished endodontic retention crown, composite resin can be used for printing.

[0012] During the curing process, a dual-curing resin cement (such as All Ceram Core) is used. It ensures complete curing of the restoration (pulp cavity-retained crown) even in areas with insufficient light transmission through a dual mechanism of rapid photo-initiated curing (triggered by light) and deep chemical-initiated curing (continuous reaction in the dark), providing reliable bonding strength.

[0013] In the prior art, a series of processes such as tooth preparation, impression taking, model pouring, and wax pattern making are completed manually by professional dental technicians. In this application, generally speaking, a pulp cavity-retained crown model is generated with the help of a trained AI model, and then the (semi-finished) pulp cavity-retained crown is printed by a printing device, shortening the manufacturing time of the pulp cavity-retained crown.

[0014] The method for manufacturing the pulp cavity-retained crown set in this application does not rely on professional dental technicians and can achieve precise quantitative control.

[0015] The pulp cavity-retained crown model generated in this application can be manufactured relatively quickly by a 3D printer. Therefore, the restoration can be completed in one go, reducing the costs for doctors and patients and improving the patient's medical experience.

[0016] Before manufacturing the finished pulp cavity-retained crown, a simulated stress test and a simulated try-on test are performed on the pulp cavity-retained crown model. Through these two tests, the risk that the finished pulp cavity-retained crown is not suitable for the patient is reduced, avoiding the extension of the patient's consultation time due to excessive on-site adjustments to the finished pulp cavity-retained crown, and avoiding the shortening of the service life of the finished pulp cavity-retained crown due to its stress-bearing capacity being lower than expected.

[0017] In the dual tests, the simulated stress test is carried out before the simulated try-on test to avoid the need to re-perform the simulated try-on test due to regenerating the pulp cavity-retained crown model, thus avoiding the extension of the manufacturing time of the finished pulp cavity-retained crown.

[0018] By improving the manufacturing method of the pulp cavity-retained crown, the manufacturing difficulty can be reduced and the manufacturing speed can be increased, enabling places such as clinics to be more capable of manufacturing the finished pulp cavity-retained crown independently (especially for simple cases), and avoiding the need for patients to visit multiple times.

[0019] This application involves the design of the pulp cavity-retained crown with the participation of artificial intelligence, avoiding mistakes caused by over-relying on the experiential judgment of professional dental technicians.

[0020] The present application provides a customizable endodontic retention crown, which can be customized individually according to the specific conditions of the affected tooth, minimizing the removal of healthy tooth structure, and achieving good retention and sufficient strength to meet future use. Based on this method, minimally invasive restoration of the crown can be achieved, retaining approximately 2 mm of dentin shoulder collar, and without placing a metal post in the root canal, avoiding damage to the root during root canal preparation, maximizing the retention of the remaining tooth structure, reducing the risk of tooth fracture, and extending the service life of the tooth.

[0021] In a further technical solution, in step three, the step of generating the endodontic retention crown model includes: S1. Generating a retention part for endodontic retention; S2. Generating an occlusal part for cooperation with the opposing tooth and adjacent teeth; S3. Generating a marginal part for cooperation with the remaining tooth structure.

[0022] In S1, on the one hand, a three-dimensional spiral retention structure is designed. Specifically, the bionic anchoring principle is adopted to design a spiral groove structure with a gradient pitch, enhancing the interfacial bonding strength while ensuring the removal force. On the other hand, a porous reinforcement structure is designed. Specifically, a structure with multiple supports (such as honeycomb, columnar, or dendritic) is integrated inside the retention part, improving the strength of the crown while reducing its weight.

[0023] In S2, multi-dimensional occlusal contact optimization is achieved. Specifically, a virtual occlusal analysis system (such as Dentalocclusion Analyzer 4.0) is applied, and the best occlusal contact point distribution model is generated through machine learning algorithms to avoid interference between the occlusal part and other teeth.

[0024] In S3, a 3D U-Net is used to generate the marginal contour.

[0025] In a further technical solution, the outer wall of the retention part forms an adhesive surface, and the surface roughness of the adhesive surface ranges from 0.2 μm to 0.5 μm.

[0026] In this embodiment, the surface roughness refers to the surface roughness of the outer wall of the retention part.

[0027] The surface roughness affects the mechanical interlocking effect between the adhesive and the substrate (retention part). When the adhesive surface is not smooth, from a microscopic perspective, the adhesive penetrates into the concave and convex structures on the adhesive surface and cures, forming an anchoring effect and enhancing the adhesive force.

[0028] The surface roughness affects the bonding strength of the adhesive as follows: When the surface roughness is too low (<0.2 μm), the adhesive surface is too smooth, resulting in insufficient mechanical interlocking force and a significant decrease in bonding strength; When the surface roughness is too high (>0.5μm), the surface of the bonding interface is too rough, resulting in the bonding agent being unable to fully penetrate into the concave and convex structures on the bonding surface, leading to stress concentration or microcracks, and the bonding strength decreases instead.

[0029] The surface roughness also affects the plaque control effect, as follows: When the surface roughness is too high (>0.5μm), bacteria are likely to adhere to the bonding surface, thus accelerating the speed of plaque formation and increasing the risk of secondary caries; When the surface roughness is too low (<0.2μm), the number of bacterial attachment points decreases, which is beneficial to controlling the speed of plaque formation.

[0030] In summary, when the surface roughness is in the range of 0.2μm - 0.5μm, it not only ensures the bonding strength of the bonding agent but also limits the speed of plaque formation, achieving a balance between bonding strength and plaque control.

[0031] For a further technical solution, in S1 of step three, the surface roughness of each region of the bonding surface is determined based on the trained morphology feature - surface roughness correlation model.

[0032] The purpose of this embodiment is to dynamically adjust the surface roughness of each region of the bonding surface. One feasible way is as follows: Based on the scanned data, use CAD software to analyze the shape features (such as grooves, sharp edges, flat areas) to determine the demand differences of each region for bonding strength and plaque control.

[0033] The operation mode of the morphology feature - surface roughness correlation model is supplemented here: The input features are matched with the clinical standard values in the database through a deep neural network (such as the SSA - BiLSTM model), and the recommended roughness range is output.

[0034] The training process of the morphology feature - surface roughness correlation model refers to the training process of the above - mentioned AI model.

[0035] For a further technical solution, in step four, the steps of the simulated stress test include: S1. Apply a vertical load of magnitude A to the pulp - retained crown model; S2. Apply an oblique load of magnitude B to the pulp - retained crown model; S3. Obtain the stress change data of each region of the pulp - retained crown model; S4. Compare the maximum stress value during the simulated stress test with the tensile strength of dentin. When the maximum stress value is greater than the tensile strength of dentin, regenerate the pulp - retained crown model.

[0036] In this embodiment, all parts of the medullary cavity retaining crown model are first generated, and then a simulated force test is performed together, thereby shortening the manufacturing process of the finished medullary cavity retaining crown.

[0037] In a further technical solution, in step six, the content of adjusting the medullary cavity retention crown model includes: adjusting the degree of aggregation of the shaft wall of the retaining portion; adjusting the surface roughness of the retaining portion; The distribution of the bite contact points on the bite portion is adjusted.

[0038] The degree of convergence of the shaft wall can be understood as the inclination angle.

[0039] This implementation method clarifies the adjustment content in step six, and further ensures the compatibility of the medullary cavity retention crown model with the patient through multiple adjustments, thereby avoiding prolonged patient consultation time due to multiple on-site adjustments to the finished medullary cavity retention crown.

[0040] Here we also provide an artificial intelligence-based intramedullary crown manufacturing system, including: Acquisition module, used to obtain oral CBCT images and oral scan models; A processing module, used for synthesizing the oral CBCT image and the oral scan model into a synthetic model; A generation module, used to generate a medullary cavity retention crown model based on the synthetic model using the trained AI model; An analysis module, used for performing a simulated stress test on the medullary cavity retention crown model; The review module is used to perform a simulated trial-wear test on the medullary cavity retention crown model that has passed the simulated force test.

[0041] The relevant descriptions in this embodiment can refer to the descriptions of the relevant parts in the above embodiment. It should be noted that there is no limitation on the specific structure of each module, as long as it meets the requirements.

[0042] In a further technical solution, the medullary cavity retention crown model comprises: The retaining part is used for retaining with the medullary cavity; The occlusal part is used to cooperate with the opposing teeth and adjacent teeth; The edge portion is used to fit with the remaining tooth.

[0043] In this embodiment, the medullary cavity retention crown model is divided into a retention part, an occlusal part and a marginal part, so that different schemes can be selected according to this division to generate the medullary cavity retention crown model to meet various needs.

[0044] According to a further technical solution, the outer wall of the retaining portion forms a bonding surface, and the surface roughness of the bonding surface is in the range of 0.2 μm-0.5 μm.

[0045] When the surface roughness is within the range of 0.2μm-0.5μm, it not only ensures the bonding strength of the adhesive, but also limits the speed of plaque formation, achieving a balance between bonding strength and plaque control.

[0046] The terms “include,” “including,” “have,” etc. used in this document are open-ended terms, meaning including but not limited to.

[0047] The terms used in this document generally have the ordinary meaning of each term used in this field, in the context of the case and in the specific context unless otherwise noted. Certain terms used to describe the case will be discussed below or elsewhere in this specification to provide additional guidance to those skilled in the art on the description of the case.

[0048] The working principle and advantages of the present invention are as follows: In the prior art, a series of processes such as tooth preparation, mold taking, model infusion, and wax model making are completed manually by professional dental technicians. In this application, in summary, a medullary cavity-retained crown model is generated with the help of a trained AI model, and then a medullary cavity-retained crown is printed by a printing device, thereby shortening the manufacturing time of the medullary cavity-retained crown.

[0049] The method of manufacturing medullary cavity retention crowns provided in the present application does not rely on professional dental technicians and can achieve precise quantitative control.

[0050] The medullary cavity retention crown model generated by the present application can be manufactured relatively quickly by a 3D printer, so that the repair can be completed in one go, reducing the costs for doctors and patients (such as time costs) and improving the patient's medical experience.

[0051] Before manufacturing the finished medullary cavity-retaining crown, the medullary cavity-retaining crown model is subjected to a simulated force test and a simulated trial-wear test. The double test reduces the risk that the finished medullary cavity-retaining crown is not suitable for the patient, avoids prolonged patient consultation time due to frequent on-site adjustments to the finished medullary cavity-retaining crown, and avoids shortening the service life of the finished medullary cavity-retaining crown due to its force-bearing capacity being lower than expected.

[0052] In the double test, the simulated force test is performed before the simulated trial-fitting test, so as to avoid the need to re-perform the simulated trial-fitting test due to the regeneration of the medullary cavity-retained crown model, thereby avoiding prolonging the manufacturing time of the finished medullary cavity-retained crown.

[0053] By improving the manufacturing method of medullary cavity-retained crowns, the manufacturing difficulty can be reduced and the manufacturing speed can be increased, making clinics and other places more capable of independently manufacturing medullary cavity-retained crown products (especially for simple cases) and avoiding patients having to visit the doctor multiple times.

[0054] This application designs the pulp cavity-retained crown with the participation of artificial intelligence, avoiding mistakes caused by over-relying on the experience and judgment of professional dental technicians.

[0055] This application can customize the pulp cavity-retained crown. Therefore, it can be individually customized according to the specific conditions of the affected tooth, minimizing the removal of healthy tooth structure, and can also achieve good retention and sufficient strength to meet future use. Brief Description of the Drawings

[0056] Figure 1 It is a schematic flow chart of the method for manufacturing the pulp cavity-retained crown according to the embodiment of the present invention; Figure 2 It is a schematic structural diagram of the system for manufacturing the pulp cavity-retained crown according to the embodiment of the present invention. Detailed Embodiments

[0057] The present invention will be further described below in conjunction with the drawings and embodiments: Embodiment: The present case will be clearly described below with diagrams and detailed descriptions. After any person skilled in the art understands the embodiments of the present case, the techniques taught by the present case can be changed and modified without departing from the spirit and scope of the present case.

[0058] The terms used in this article are only for describing specific embodiments and are not intended to limit the present case. Singular forms such as "a", "this", "this", "the present", and "the" also include plural forms as used herein.

[0059] See Figure 1 , a method for manufacturing a pulp cavity-retained crown based on artificial intelligence, includes: Step 1: Obtain oral CBCT images and intraoral scan models; Step 2: Synthesize the oral CBCT images and the intraoral scan models into a synthesized model; Step 3: Input the synthesized model into a trained AI model, and the AI model generates a pulp cavity-retained crown model based on the synthesized model; Step 4: Conduct a simulated stress test on the pulp cavity-retained crown model. When the stress in any area of the pulp cavity-retained crown model is greater than a predetermined value, regenerate the pulp cavity-retained crown model; Step 5: Conduct a simulated try-on test on the pulp cavity-retained crown model that passes the simulated stress test; Step 6: Adjust the pulp cavity-retained crown model according to the results of the simulated try-on test; Step 7: Input the adjusted pulp cavity-retained crown model into a printing device and print to generate a semi-finished pulp cavity-retained crown; Step 8: Cure the semi-finished pulp cavity-retained crown to generate a finished pulp cavity-retained crown.

[0060] The CBCT image and the intraoral scan model are existing setups, which are briefly introduced here: The CBCT image (Cone Beam Computed Tomography) belongs to radiological imaging technology and provides hard tissue information such as the morphology of the jawbone, the course of the nerve canal, and the position of the maxillary sinus through three-dimensional X-ray scanning; the intraoral scan model (oral scan data) belongs to optical three-dimensional surface scanning technology, which is radiation-free and can quickly obtain the morphology of the tooth crown and the gingival surface, and is suitable for prosthetic design and orthodontic simulation; the two are used in cooperation through software fusion. The crown data of the intraoral scan is combined with the intraosseous structure data of the CBCT to generate a complete three-dimensional model; in addition, two-dimensional radiological images such as periapical films and panoramic films can be introduced.

[0061] In some embodiments, the format of the CBCT image is DICOM, and the format of the intraoral scan model is STL. The ICP algorithm is used for comparison during the synthesis process.

[0062] This application is implemented based on artificial intelligence. With the help of an AI model, the training method of the AI model is as follows: Collect the three-dimensional data of the jawbone scanned by CBCT and the optical data of the intraoral scan. The dental experts label the pulp cavity boundary, the edge design type (butt joint type / wrap-around type), and the material selection (such as lithium disilicate ceramic). Use a dual-channel 3D CNN to process the CBCT voxel data and the intraoral scan point cloud data, and fuse the features through cross-attention. Build a mechanical simulation module to calculate the stress distribution (such as von Mises stress) in real time to constrain the rationality of the design. Use historical restoration cases for supervised learning to optimize the design error and mechanical compliance. Based on the clinical success rate, perform reinforcement learning optimization and provide manual correction feedback for complex cases.

[0063] The simulated stress test is mainly realized through finite element analysis (FEA), which will not be elaborated here. When the stress in any area of the pulp cavity-retained crown model is greater than the predetermined value, regenerate the pulp cavity-retained crown model, specifically by updating the surrogate model and calculating after marking the defective part.

[0064] The simulated trial fitting test is realized through virtual trial fitting technology. The technical implementation principle is as follows: Superimpose the restoration model (pulp cavity-retained crown model) on the patient's oral digital model (synthetic model) through augmented reality technology. Skip step six if there is no need to adjust the pulp cavity-retained crown model.

[0065] When printing the semi-finished pulp cavity-retained crown, composite resin can be used for printing.

[0066] During the curing process, a dual-curing resin cement (such as All Ceram Core) is used. It ensures the complete curing of the restoration (pulpal-retained crown) even in areas with insufficient light transmission through a dual mechanism of rapid photo-initiated curing (triggered by light) and deep chemical-initiated curing (continuous reaction in the dark), providing reliable bonding strength.

[0067] In the prior art, a series of processes such as tooth preparation, impression taking, model pouring, and wax pattern making are completed by professional dental technicians manually. In this application, generally speaking, a pulpal-retained crown model is generated with the help of a trained AI model, and then the pulpal-retained crown is printed by a printing device, shortening the manufacturing time of the pulpal-retained crown.

[0068] The method for manufacturing the pulpal-retained crown set in this application does not rely on professional dental technicians and can achieve precise quantitative control.

[0069] The pulpal-retained crown model generated in this application can be manufactured relatively quickly by a 3D printer. Therefore, the restoration can be completed in one go, reducing the costs (such as time costs) for both doctors and patients and improving the patient's medical experience.

[0070] Before manufacturing the finished pulpal-retained crown, a simulated stress test and a simulated try-in test are performed on the pulpal-retained crown model. Through these dual tests, the risk that the finished pulpal-retained crown is not suitable for the patient is reduced, avoiding the extension of the patient's consultation time due to excessive on-site adjustment of the finished pulpal-retained crown and avoiding the shortening of its service life due to the lower-than-expected stress-bearing capacity of the finished pulpal-retained crown.

[0071] In the dual tests, the simulated stress test is carried out before the simulated try-in test to avoid the need to re-perform the simulated try-in test due to the regeneration of the pulpal-retained crown model, thus avoiding the extension of the manufacturing time of the finished pulpal-retained crown.

[0072] By improving the manufacturing method of the pulpal-retained crown, the manufacturing difficulty can be reduced and the manufacturing speed can be increased, enabling places such as clinics to be more capable of manufacturing the finished pulpal-retained crown independently (especially for simple cases), and avoiding the need for patients to visit multiple times.

[0073] This application designs the pulpal-retained crown with the participation of artificial intelligence, avoiding mistakes caused by over-relying on the empirical judgment of professional dental technicians.

[0074] The present application provides a customizable endodontic retention crown, which can be customized individually according to the specific conditions of the affected tooth, minimizing the removal of healthy tooth structure, and can also achieve good retention and sufficient strength to meet future use. Based on this method, minimally invasive restoration of the crown can be achieved, retaining approximately 2 mm of dentin shoulder, and there is no need to place a metal post in the root canal, avoiding damage to the root during root canal preparation, maximizing the retention of the remaining tooth structure, reducing the risk of tooth fracture, and extending the service life of the tooth.

[0075] In this embodiment, in step three, the step of generating the endodontic retention crown model includes: S1. Generate a retention part for endodontic retention; S2. Generate an occlusal part for cooperating with the opposing tooth and adjacent teeth; S3. Generate an edge part for cooperating with the remaining tooth structure (such as the dentin shoulder).

[0076] In S1, on the one hand, a three-dimensional spiral retention structure is designed. Specifically, the bionic anchoring principle is adopted to design a spiral groove structure with a gradient pitch, enhancing the interfacial bonding strength while ensuring the removal force. On the other hand, a porous reinforcement structure is designed. Specifically, a structure with multiple supports (such as honeycomb, columnar, or dendritic) is integrated inside the retention part, improving the strength of the crown while reducing its weight.

[0077] In S2, multi-dimensional occlusal contact optimization is achieved. Specifically, a virtual occlusal analysis system (such as Dentalocclusion Analyzer 4.0) is applied, and the best occlusal contact point distribution model is generated through machine learning algorithms to avoid interference between the occlusal part and other teeth.

[0078] In S3, a 3D U-Net is used to generate the edge contour.

[0079] In this embodiment, the outer wall of the retention part forms an adhesive surface, and the surface roughness of the adhesive surface ranges from 0.2 μm to 0.5 μm.

[0080] The surface roughness mentioned in this embodiment refers to the surface roughness of the outer wall of the retention part.

[0081] The surface roughness affects the mechanical interlocking effect between the adhesive and the substrate (retention part). When the adhesive surface is not smooth, from a microscopic perspective, the adhesive penetrates into the concave and convex structures on the adhesive surface and cures, forming an anchoring effect and enhancing the adhesive force.

[0082] The surface roughness affects the bonding strength of the adhesive as follows: When the surface roughness is too low (<0.2 μm), the adhesive surface is too smooth, resulting in insufficient mechanical interlocking force and a significant decrease in bonding strength; When the surface roughness is too high (> 0.5 μm), the surface of the bonding surface is too rough, resulting in the adhesive not being able to fully penetrate into the concave-convex structure on the bonding surface, leading to stress concentration or microcracks, and the bonding strength decreases instead.

[0083] The surface roughness also affects the plaque control effect, as follows: When the surface roughness is too high (> 0.5 μm), bacteria are easily attached to the bonding surface, thus accelerating the speed of plaque formation and increasing the risk of secondary caries; When the surface roughness is too low (< 0.2 μm), the number of bacterial attachment points decreases, which is beneficial to controlling the speed of plaque formation.

[0084] In summary, when the surface roughness is in the range of 0.2 μm - 0.5 μm, it not only ensures the bonding strength of the adhesive but also limits the speed of plaque formation, achieving a balance between bonding strength and plaque control.

[0085] In this embodiment, in S1 of step three, the surface roughness of each region of the bonding surface is determined based on the trained morphology feature - surface roughness correlation model.

[0086] The purpose of this embodiment is to dynamically adjust the surface roughness of each region of the bonding surface. One feasible way is as follows: Based on the scanned data, use CAD software to analyze the shape features (such as grooves, sharp edges, flat areas) to determine the difference in the requirements for bonding strength and plaque control in each region.

[0087] The operation method of the morphology feature - surface roughness correlation model is supplemented here: Through a deep neural network (such as the SSA - BiLSTM model), match the input features with the clinical standard values in the database and output the recommended roughness range (such as 0.5 μm recommended for flat areas).

[0088] The training process of the morphology feature - surface roughness correlation model refers to the training process of the above AI model.

[0089] In this embodiment, in step four, the steps of the simulated stress test include: S1. Apply a vertical load of magnitude A to the pulp - retained crown model; S2. Apply an oblique load of magnitude B to the pulp - retained crown model; S3. Obtain the stress change data of each region of the pulp - retained crown model; S4. Compare the maximum stress value in the simulated stress test process with the tensile strength of dentin. When the maximum stress value is greater than the tensile strength of dentin, regenerate the pulp - retained crown model.

[0090] In this embodiment, all parts of the endodontic retention crown model are first generated, and then simulated stress tests are performed together, thereby shortening the manufacturing progress of the finished endodontic retention crown. The specific settings of the vertical load and the oblique load (such as the load magnitude) are adjusted according to actual requirements.

[0091] In some embodiments, A and B are 100 N.

[0092] In some embodiments, when applying an oblique load of magnitude B, the angle is set to 45.

[0093] In some embodiments, based on electromyogram (EMG) and chewing movement trajectory data, an asymmetric dynamic load spectrum (frequency range of 0.5 - 30 Hz) is constructed.

[0094] In some embodiments, the penalty function method + Lagrange multiplier hybrid algorithm is used to handle the dynamic contact between the crown and the opposing tooth, simulating the stick-slip friction behavior.

[0095] In some embodiments, a saliva viscoelastic lubrication model (generalized Maxwell constitutive) is introduced to analyze the effects of interface energy reduction and capillary effects on marginal microleakage in a wet environment.

[0096] In some embodiments, thermo-mechanical coupling simulation is performed to simulate the thermal expansion mismatch stress at the ceramic-metal interface caused by thermal cycling (5°C ↔ 55°C).

[0097] In this embodiment, in step six, the content of adjusting the endodontic retention crown model includes: Adjust the axial wall convergence angle of the retention part; Adjust the surface roughness of the retention part; Adjust the distribution of the occlusal contact points on the occlusal part.

[0098] The axial wall convergence angle can be understood as the inclination angle.

[0099] This embodiment clarifies the adjustment content in step six. Through multiple adjustments, the adaptation of the endodontic retention crown model to the patient is further ensured, avoiding the extension of the patient's visit time due to (more) adjustments to the finished endodontic retention crown on-site.

[0100] See Figure 2 , and an endodontic retention crown manufacturing system based on artificial intelligence is also provided here, including: An acquisition module for obtaining oral CBCT images and intraoral scan models; A processing module for synthesizing the oral CBCT images and the intraoral scan models into a synthetic model; A generation module for generating an endodontic retention crown model based on the synthetic model using a trained AI model; An analysis module for performing a simulated stress test on the endodontic retention crown model; An approval module for performing a simulated fitting test on the endodontic retention crown model that has passed the simulated stress test.

[0101] In this embodiment, relevant descriptions can refer to the descriptions of relevant parts in the above embodiments. It should be noted that for each module, the specific structure adopted is not limited, as long as the requirements are met.

[0102] In this embodiment, the endodontic retention crown model includes: A retention part for retaining with the pulp cavity; An occlusal part for cooperating with the opposing tooth and adjacent teeth; An edge part for cooperating with the remaining tooth structure.

[0103] In this embodiment, the endodontic retention crown model is divided into a retention part, an occlusal part, and an edge part, so that different schemes can be selected according to this division to generate the endodontic retention crown model to meet various requirements.

[0104] In this embodiment, an adhesive surface is formed on the outer wall of the retention part, and the surface roughness of the adhesive surface ranges from 0.2 μm to 0.5 μm.

[0105] When the surface roughness ranges from 0.2 μm to 0.5 μm, it not only ensures the bonding strength of the adhesive but also limits the plaque formation rate, achieving a balance between bonding strength and plaque control.

[0106] The above embodiments are only for illustrating the technical concept and characteristics of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly, and shall not be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit and essence of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A method for manufacturing a pulp chamber-retained crown based on artificial intelligence, characterized in that: Including: Step 1: Obtain oral CBCT images and oral scan models; Step 2: Synthesize the oral CBCT images and the oral scan models into a synthesized model; Step 3: Input the synthesized model into a trained AI model, and generate a pulp-retained crown model by the AI model based on the synthesized model; Step 4: Conduct a simulated stress test on the pulp-retained crown model. When the stress in any area of the pulp-retained crown model is greater than a predetermined value, regenerate the pulp-retained crown model; Step 5: Conduct a simulated trial fitting test on the pulp-retained crown model that passes the simulated stress test; Step 6: Adjust the pulp-retained crown model according to the results of the simulated trial fitting test; Step 7: Input the adjusted pulp-retained crown model into a printing device and print a semi-finished pulp-retained crown; Step 8: Cure the semi-finished pulp-retained crown to generate a finished pulp-retained crown.

2. The method for manufacturing a pulp cavity-retained crown based on artificial intelligence according to claim 1, wherein: In Step 3, the steps of generating the pulp-retained crown model include: S1: Generate a retention part for pulp retention; S2: Generate an occlusal part for cooperation with the opposing tooth and adjacent teeth; S3: Generate a marginal part for cooperation with the remaining tooth structure.

3. The method for manufacturing a pulp cavity-retained crown based on artificial intelligence according to claim 2, wherein: The outer wall of the retention part forms an adhesive surface, and the surface roughness of the adhesive surface ranges from 0.2 μm to 0.5 μm.

4. The method for manufacturing a pulp cavity-retained crown based on artificial intelligence according to claim 3, wherein: In S1 of Step 3, determine the surface roughness of each area of the adhesive surface based on a trained morphology feature-surface roughness correlation model.

5. The method for manufacturing a pulp cavity-retained crown based on artificial intelligence according to claim 1, wherein: In Step 4, the steps of the simulated stress test include: S1: Apply a vertical load of magnitude A to the pulp-retained crown model; S2: Apply an oblique load of magnitude B to the pulp-retained crown model; S3: Obtain the stress change data of each area of the pulp-retained crown model; S4: Compare the maximum stress value during the simulated stress test with the tensile strength of dentin. When the maximum stress value is greater than the tensile strength of dentin, regenerate the pulp-retained crown model.

6. The method for manufacturing a pulp cavity-retained crown based on artificial intelligence according to claim 2, wherein: In Step 6, the content of adjusting the pulp-retained crown model includes: Adjust the axial wall polymerization degree of the retention part; Adjust the surface roughness of the retention part; Adjust the distribution of occlusal contact points on the occlusal part.

7. Artificial intelligence-based endocrown manufacturing system, characterized in that: Including: An acquisition module for obtaining oral CBCT images and oral scan models; A processing module for synthesizing the oral CBCT images and the oral scan models into a synthesized model; A generation module for generating a pulp-retained crown model by using a trained AI model based on the synthesized model; An analysis module for conducting a simulated stress test on the pulp-retained crown model; An auditing module for conducting a simulated trial fitting test on the pulp-retained crown model that passes the simulated stress test.

8. The artificial intelligence-based endocrown manufacturing system according to claim 7, wherein: The pulp-retained crown model includes: A retention part for pulp retention; An occlusal part for cooperation with the opposing tooth and adjacent teeth; A marginal part for cooperation with the remaining tooth structure.

9. The artificial intelligence-based pulp cavity-retained crown manufacturing system according to claim 8, characterized in that: The outer wall of the retention part forms an adhesive surface, and the surface roughness of the adhesive surface ranges from 0.2 μm to 0.5 μm.

Images