Digital removable denture blank preparation process based on AI guide plate design

Abstract

The invention discloses a digital removable denture blank preparation process based on AI guide plate design, which comprises the following steps: firstly, placing a finished tooth at a corresponding position of an incisor bottom guide plate, and physically polishing the part of the finished tooth, which protrudes out of the incisor bottom guide plate, to obtain a finished tooth conforming to the oral cavity of a patient; then designing three-dimensional data of a false tooth through CAD software, generating a glue-filled guide plate model in cooperation with a pre-expansion compensation algorithm, and obtaining a glue-filled guide plate through 3D printing; finally, the finished tooth polished in the step T1 is placed in the glue filling guide plate generated in the step T2, and then resin is injected into the glue filling guide plate; and putting the glue-filled guide plate filled with the resin and the finished tooth product into a constant-temperature pressurizing device or performing polymerization reaction in a normal-temperature state to obtain a false tooth blank. The machining precision can be improved, and the beneficial effects of being convenient to produce and high in structural stability are achieved.

Description

Technical Field

[0001] This invention relates to the field of denture manufacturing, and in particular to a digital process for preparing removable denture blanks based on AI guide plate design. Background Technology

[0002] Removable dentures utilize natural teeth and a denture base covering the mucosa and bone tissue for support. Relying on the retention of the denture and the base, the artificial teeth restore the shape and function of missing teeth, and the base material restores the morphology of the damaged alveolar ridge and soft tissue. Patients can remove and insert the denture themselves. Traditional denture fabrication involves using a silicone rubber impression to obtain the patient's oral structure, followed by impression taking, plaster casting and embedding, wax casting, alloy solution injection, cooling, and demolding. During the plaster casting process, cooling shrinkage of the plaster can cause deformation of the internal wax-shaped framework, occlusal elevation, and marginal closure, affecting the overall fabrication accuracy. Therefore, traditional denture filling techniques suffer from relatively low fabrication accuracy. Summary of the Invention

[0003] The purpose of this invention is to provide a digital fabrication process for removable denture blanks based on an AI guide plate design. This invention has the advantage of high processing precision.

[0004] The technical solution of this invention: A digital removable denture preform fabrication process based on AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software or AI, and a pre-expansion compensation algorithm is used to generate a filling guide plate model, which is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0005] In the aforementioned digital removable denture preform fabrication process based on AI guide plate design, the pre-expansion compensation algorithm in T2 is based on a displacement vector field inverse mapping algorithm. This involves globally aligning the design model and the finished product scanning model to establish a corresponding three-dimensional spatial vector field, calculating the deformation displacement vector of each vertex, and then applying the formula: "Compensated coordinates = Original coordinates - (Compensation gain coefficient)". The coordinates that need to be pre-compensated for in the design model are calculated using the deformation displacement vector. Finally, the finished glue-filled guide plate model is obtained based on the calculated pre-compensation coordinates.

[0006] In the aforementioned digital removable denture preform fabrication process based on AI guide plate design, the pre-expansion compensation algorithm in T2 is a nonlinear expansion algorithm based on local wall thickness weighting. It utilizes the characteristic that resin polymerization shrinkage is positively correlated with material volume and thickness, and leverages the principle that the new vertex position equals the original vertex position plus the normal direction. (Local wall thickness mapping weight) Linear shrinkage rate), to obtain the local wall thickness value at each point of the denture.

[0007] In the aforementioned digital removable denture preform fabrication process based on AI guide plate design, the pre-expansion compensation algorithm in T2 is a radial basis function spatial deformation algorithm with regional constraints. It uses radial basis functions to construct a globally smooth deformation field and introduces a "regional locking coefficient". The "regional locking coefficient" is achieved by setting zero-weight masks in high-precision areas such as artificial alveolar sockets and rests to ensure that the geometric features of key functional areas remain unchanged during the global deformation process.

[0008] In the aforementioned digital removable denture preform fabrication process based on AI guide plate design, the pre-expansion compensation algorithm in T2 is a semantic feature compensation algorithm based on anatomical partitions. The denture model is imported into the computer and semantic feature recognition is performed on the denture model to determine the palatal dome region of the maxilla and the horseshoe-shaped posterior tooth region of the mandible. Then, a vertical downward "anti-lift" displacement compensation is applied to the palatal dome region of the maxilla, and an "anti-centripetal contraction" tension compensation that expands outward from the midline to both sides is applied to the horseshoe-shaped posterior tooth region of the mandible.

[0009] In the aforementioned digital removable denture preform fabrication process based on AI guide plate design, the pre-expansion compensation algorithm in T2 is a simulated shrinkage iterative closed-loop compensation algorithm. It uses a computer system to generate an initial compensation model and perform shrinkage simulation, and then compares the residual between the simulation results and the target model. If the residual exceeds a set threshold, the compensation amount is readjusted and the compensation model is regenerated, and the shrinkage simulation is repeated until the residual between the simulation results and the target model is within the threshold, thus obtaining the final glue-filled guide plate model.

[0010] In the aforementioned digital removable denture preform preparation process based on AI guide plate design, the glue-filling guide plate generated in T2 has glue injection holes on its side and positioning pins on the mold-closing part of the glue-filling guide plate.

[0011] Compared with existing technologies, this invention improves the resin filling process in the current denture manufacturing process. During the processing of the finished denture, an incisor floor guide is used to assist in the physical polishing of the finished denture, ensuring the occlusal positioning accuracy of the finished denture within the resin filling guide and guaranteeing processing precision. By utilizing a pre-expansion compensation algorithm during the preparation of the resin filling guide, pre-space is provided for the shrinkage deformation of the resin during the filling process. This prevents deformation of the denture base, occlusal elevation, and poor marginal seal due to shrinkage during subsequent cooling, ensuring the accuracy of the final denture. Furthermore, the incisor floor guide and the resin filling guide work together to achieve precise processing of the denture, guaranteeing processing precision. Simultaneously, AI automatically identifies the occlusal plane of the patient's teeth, and CAD software is used to design and generate a resin filling guide with injection holes, positioning pins, and shrinkage compensation pre-space, eliminating the need for manual intervention in geometric deformation and facilitating production.

[0012] Furthermore, this invention sets the pre-expansion compensation algorithm to a displacement vector field inverse mapping algorithm. By aligning the design model and the finished product scanning model across the entire domain, a three-dimensional spatial displacement vector field is established, the deformation offset of each vertex is calculated, and during the filling cavity design stage, the vertices of the original model are pre-compensated and moved along the opposite direction of the deformation vector. This can provide precise and personalized displacement correction for the non-uniform shrinkage of various parts of the denture, maximizing the accuracy of the denture. By adopting a nonlinear expansion algorithm based on local wall thickness weighting and automatically identifying denture thickness differences, the problem of large shrinkage in thick areas and warping deformation in thin areas caused by uneven polymerization heat generation can be effectively solved. By employing a radial basis function spatial deformation algorithm with regional constraints, global deformation compensation is achieved while ensuring precise implantation of the artificial tooth, eliminating the need for secondary grinding and simplifying processing. The semantic feature compensation algorithm based on anatomical partitions fully incorporates the deformation patterns in oral clinics, improving the fit and retention of the prosthesis tissue surfaces. Furthermore, a simulated shrinkage iterative closed-loop compensation algorithm generates an initial compensation model using a computer system and performs shrinkage simulations. The residuals between the simulation results and the target model are then compared, and the structure of the compensation model is continuously adjusted based on the comparison between the residuals and a threshold, achieving physical processing accuracy that approximates the ideal model to the greatest extent possible, further enhancing processing precision. Therefore, this invention not only improves processing precision but also offers advantages such as convenient production and high structural stability. Detailed Implementation

[0013] The present invention will be further described below with reference to embodiments, but these embodiments are not intended to limit the scope of the invention.

[0014] Example 1. A process for fabricating a digital removable denture preform based on AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0015] The pre-expansion compensation algorithm in T2 is based on the displacement vector field inverse mapping algorithm. It establishes a corresponding three-dimensional spatial vector field by aligning the design model and the finished product scanning model globally, calculating the deformation displacement vector of each vertex, and then applying the formula: "Compensated coordinates = Original coordinates - (Compensation gain coefficient)". The deformation displacement vector) is used to calculate the coordinates that need to be pre-compensated for the design model. Finally, the finished glue-filled guide plate model is obtained based on the calculated pre-compensated coordinates. The glue-filled guide plate generated in T2 has glue injection holes on its side and positioning pins on the film-closing part of the glue-filled guide plate.

[0016] Example 2. A process for fabricating a digital removable denture preform based on AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0017] The pre-expansion compensation algorithm in T2 is a nonlinear expansion algorithm based on local wall thickness weighting. It utilizes the characteristic that the shrinkage of resin polymerization is positively correlated with the material volume and thickness, and uses the formula: new vertex position = original vertex position + normal direction. (Local wall thickness mapping weight) The linear shrinkage rate is used to obtain the local wall thickness value at each point of the denture; the side of the filling guide plate generated in T2 is provided with a glue injection hole, and the mating part of the filling guide plate is provided with a positioning pin.

[0018] Example 3. A process for fabricating a digital removable denture preform based on AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0019] The pre-expansion compensation algorithm in T2 is a radial basis function spatial deformation algorithm with regional constraints. It uses radial basis functions to construct a globally smooth deformation field and introduces a "regional locking coefficient". The "regional locking coefficient" is achieved by setting zero-weight masks in high-precision areas such as artificial alveolar sockets and supports to ensure that the geometric features of key functional areas remain unchanged during the global deformation process. The glue filling guide plate generated in T2 has glue injection holes on its side and positioning pins on the mold closing part of the glue filling guide plate.

[0020] Example 4. A process for fabricating a digital removable denture preform based on an AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0021] The pre-expansion compensation algorithm in T2 is a semantic feature compensation algorithm based on anatomical partitions. The denture model is imported into the computer and semantic feature recognition is performed on the denture model to determine the palatal dome region of the maxilla and the horseshoe-shaped posterior tooth region of the mandible. Then, a vertical downward "anti-lift" displacement compensation is applied to the palatal dome region of the maxilla, and an "anti-centripetal contraction" tension compensation that expands outward from the midline to both sides is applied to the horseshoe-shaped posterior tooth region of the mandible. The glue filling guide plate generated in T2 has glue injection holes on its side and positioning pins on the mating part of the glue filling guide plate.

[0022] Example 5. A process for fabricating a digital removable denture preform based on an AI guide plate design, comprising the following steps: T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Synergistic Polymerization: The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate; the filling guide plate containing resin and finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture preform.

[0023] The pre-expansion compensation algorithm in T2 is a simulated shrinkage iterative closed-loop compensation algorithm. It uses a computer system to generate an initial compensation model and perform shrinkage simulation. Then, it compares the residual between the simulation results and the target model. If the residual exceeds a set threshold, the compensation amount is readjusted and the compensation model is regenerated, and the shrinkage simulation is repeated. This continues until the residual between the simulation results and the target model is within the threshold, thus obtaining the final glue-filled guide plate model. The glue-filled guide plate generated in T2 has glue injection holes on its side and positioning pins at the film-closing point of the glue-filled guide plate.

Claims

1. A process for fabricating digital removable denture preforms based on AI guide plate design, characterized in that: Includes the following steps T1. Physical space mapping: The finished tooth is placed on the corresponding position of the incisor floor guide plate, and the part of the finished tooth protruding from the incisor floor guide plate is physically polished to obtain a finished tooth that conforms to the patient's oral cavity. T2. Preparation of the filling guide plate: The three-dimensional data of the denture is designed using CAD software, and a pre-expansion compensation algorithm is used to generate the filling guide plate model. The filling guide plate is then obtained by 3D printing. T3, Collaborative Aggregation; The finished tooth, which has been polished in T1, is placed in the filling guide plate generated in T2, and then resin is injected into the filling guide plate. The filling guide plate containing resin and the finished tooth is then placed in a constant temperature and pressure device or at room temperature for polymerization reaction to obtain the denture blank.

2. The process for fabricating a digital removable denture preform based on AI guide plate design according to claim 1, characterized in that: The pre-expansion compensation algorithm in T2 is based on the displacement vector field inverse mapping algorithm. It establishes a corresponding three-dimensional spatial vector field by aligning the design model and the finished product scanning model globally, calculating the deformation displacement vector of each vertex, and then applying the formula: "Compensated coordinates = Original coordinates - (Compensation gain coefficient)". The coordinates that need to be pre-compensated for in the design model are calculated using the deformation displacement vector. Finally, the finished glue-filled guide plate model is obtained based on the calculated pre-compensation coordinates.

3. The digital removable denture preform fabrication process based on AI guide plate design according to claim 1, characterized in that: The pre-expansion compensation algorithm in T2 is a nonlinear expansion algorithm based on local wall thickness weighting. It utilizes the characteristic that the shrinkage of resin polymerization is positively correlated with the material volume and thickness, and uses the formula: new vertex position = original vertex position + normal direction. (Local wall thickness mapping weight) Linear shrinkage rate), to obtain the local wall thickness value at each point of the denture.

4. The process for fabricating a digital removable denture preform based on AI guide plate design according to claim 1, characterized in that: The pre-expansion compensation algorithm in T2 is a radial basis function spatial deformation algorithm with regional constraints. It uses radial basis functions to construct a globally smooth deformation field and introduces a "regional locking coefficient". The "regional locking coefficient" is achieved by setting zero-weight masks in high-precision areas such as artificial alveolar sockets and supports to ensure that the geometric features of key functional areas remain unchanged during the global deformation process.

5. The process for fabricating a digital removable denture preform based on AI guide plate design according to claim 1, characterized in that: The pre-expansion compensation algorithm in T2 is a semantic feature compensation algorithm based on anatomical partitioning; The denture model was imported into the computer and semantic feature recognition was performed on the denture model to determine the palatal dome region of the maxilla and the horseshoe-shaped posterior tooth region of the mandible. Then, a vertical downward "anti-lift" displacement compensation was applied to the palatal dome region of the maxilla, and an "anti-centripetal contraction" tension compensation was applied to the horseshoe-shaped posterior tooth region of the mandible, which expanded outward from the midline to both sides.

6. The fabrication process of a digital removable denture preform based on AI guide plate design according to claim 1, characterized in that: The pre-expansion compensation algorithm in T2 is a simulated contraction iterative closed-loop compensation algorithm. It uses a computer system to generate an initial compensation model and perform contraction simulation, and then compares the residuals between the simulation results and the target model. If the residual exceeds the set threshold, the compensation amount is readjusted and the compensation model is regenerated, and the shrinkage simulation is repeated until the residual between the simulation results and the target model is within the threshold, thus obtaining the final glue-filled guide plate model.

7. A digital removable denture preform fabrication process based on AI guide plate design according to any one of claims 1 to 6, characterized in that: The glue-filling guide plate generated in T2 has glue injection holes on its side and positioning pins on the film-closing part of the glue-filling guide plate.