A method, system, and electronic device for automatically generating a design for a removable denture

Abstract

The application relates to the technical field of oral image processing, and in particular to a method, system and electronic device for automatically generating a design scheme of a removable denture, the method comprising the following steps: acquiring an oral curved surface tomography panoramic film of a target user; performing a data enhancement operation on the oral curved surface tomography panoramic film; identifying the oral curved surface tomography panoramic film subjected to the data enhancement operation by using a trained network model to obtain a lesion condition of a tooth to be repaired; constructing a decision tree corresponding to the tooth to be repaired according to the lesion condition of the tooth to be repaired; matching a target decision tree with the maximum similarity to the decision tree corresponding to the tooth to be repaired from a preset database; and when the maximum similarity is greater than a preset similarity threshold, determining a removable denture design scheme corresponding to the target decision tree as a removable denture design scheme corresponding to the tooth to be repaired. The application can quickly and accurately determine the removable denture design scheme corresponding to the tooth to be repaired, thereby saving medical resources and reducing the medical cost of the user.

Description

Technical Field

[0001] This invention relates to the field of oral image processing technology, and in particular to a method, system, and electronic device for automatically generating removable denture design schemes. Background Technology

[0002] With rapid economic development and rising living standards, more and more people are paying attention to dental health. People's quality of life is closely related to their dental health. As people age, more and more people experience tooth loss, which can significantly impact their digestive system, speech, and facial aesthetics. Therefore, restoring missing teeth is crucial, and removable partial dentures are one of the main treatment methods for tooth loss. They utilize natural teeth and the underlying mucosa and bone tissue for support, relying on the denture's retainers and base for retention. Artificial teeth restore the shape and function of the missing teeth, while the base material restores the damaged alveolar ridge and soft tissue morphology. Removable partial dentures not only have a wide range of applications but also offer advantages such as easy removal and cleaning for maintaining good oral hygiene, minimal tooth reduction, ease of fabrication, repair, and addition, and lower cost. Therefore, removable partial dentures are one of the most commonly used methods for restoring missing teeth.

[0003] Currently, traditional partial removable denture design requires a doctor's diagnosis, followed by drawing up a design plan on a paper form, and then having the dentures manufactured by staff. This process requires numerous examinations to make a more accurate design, which is inefficient and increases the patient's medical costs. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a method, system and electronic device for automatically generating removable denture design schemes.

[0005] The technical solution of the method for automatically generating removable denture design schemes according to the present invention is as follows:

[0006] Obtain panoramic radiographs of the target user's oral cavity;

[0007] Data augmentation was performed on the panoramic radiograph of the oral cavity.

[0008] The trained network model is used to identify the lesions of the teeth to be repaired by using data-augmented panoramic radiographs of the oral cavity.

[0009] Based on the lesion condition of the tooth to be repaired, construct a decision tree corresponding to the tooth to be repaired;

[0010] From the preset database, the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired is selected;

[0011] When the maximum similarity is greater than the preset similarity threshold, the removable denture design scheme corresponding to the target decision tree is determined as the removable denture design scheme corresponding to the tooth to be repaired.

[0012] The beneficial effects of the method for automatically generating removable denture design schemes according to the present invention are as follows:

[0013] It can quickly and accurately determine the design of removable dentures for the teeth to be restored, which can save medical resources and reduce medical costs for users.

[0014] Based on the above scheme, the method for automatically generating a removable denture design according to the present invention can be further improved as follows.

[0015] Furthermore, data augmentation operations are performed on the aforementioned panoramic radiograph of the oral cavity, including:

[0016] The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

[0017] Furthermore, the process of obtaining the trained network model includes:

[0018] The Mask Transfiner is trained based on a preset sample set to obtain the trained network model.

[0019] Furthermore, the aforementioned pathological conditions include at least one of the following: tooth loss, periodontal disease, carious damage to the tooth structure, periapical abnormalities, pulp abnormalities, gingival abnormalities, and periodontal and alveolar bone abnormalities.

[0020] The technical method of the present invention for automatically generating a removable denture design scheme is as follows:

[0021] It includes an acquisition module, an enhancement module, an identification module, a construction module, a matching module, and a determination module;

[0022] The acquisition module is used to: acquire panoramic radiographs of the target user's oral cavity;

[0023] The enhancement module is used to perform data enhancement operations on the panoramic radiograph of the oral cavity.

[0024] The recognition module is used to: use a trained network model to recognize the data-augmented panoramic radiograph of the oral cavity to obtain the lesion status of the tooth to be repaired;

[0025] The construction module is used to: construct a decision tree corresponding to the tooth to be repaired based on the lesion condition of the tooth to be repaired;

[0026] The matching module is used to: match the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired from a preset database;

[0027] The determining module is used to: when the maximum similarity is greater than a preset similarity threshold, determine the removable denture design scheme corresponding to the target decision tree as the removable denture design scheme corresponding to the tooth to be restored.

[0028] The beneficial effects of the system for automatically generating removable denture designs according to the present invention are as follows:

[0029] It can quickly and accurately determine the design of removable dentures for the teeth to be restored, which can save medical resources and reduce medical costs for users.

[0030] Based on the above solution, the system for automatically generating removable denture designs according to the present invention can be further improved as follows.

[0031] Furthermore, the enhancement module is specifically used for:

[0032] The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

[0033] Furthermore, it also includes a training module, which is used for:

[0034] The Mask Transfiner is trained based on a preset sample set to obtain the trained network model.

[0035] Furthermore, the aforementioned pathological conditions include at least one of the following: tooth loss, periodontal disease, carious damage to the tooth structure, periapical abnormalities, pulp abnormalities, gingival abnormalities, and periodontal and alveolar bone abnormalities.

[0036] The present invention provides a storage medium storing instructions that, when read by a computer, cause the computer to execute a method for automatically generating a removable denture design as described above.

[0037] An electronic device according to the present invention includes a processor and the above-described storage medium, wherein the processor executes instructions in the storage medium. Attached Figure Description

[0038] Figure 1 This is a flowchart illustrating a method for automatically generating a removable denture design according to an embodiment of the present invention.

[0039] Figure 2 This is a schematic diagram of the structure of a system for automatically generating removable denture design schemes according to an embodiment of the present invention. Detailed Implementation

[0040] like Figure 1 As shown in the figure, a method for automatically generating a removable denture design scheme according to an embodiment of the present invention includes the following steps:

[0041] S1. Obtain panoramic radiographs of the target user's oral cavity;

[0042] Among them, the panoramic radiograph of the oral cavity can display anatomical structures such as the upper and lower jaws, all teeth, maxillary sinuses and temporomandibular joint, which can comprehensively show the oral condition of the target user.

[0043] S2. Perform data enhancement on panoramic radiographs of the oral cavity, specifically:

[0044] The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

[0045] S3. Use the trained network model to identify the data-augmented panoramic radiographs of the oral cavity to obtain the lesion status of the teeth to be repaired.

[0046] The pathological conditions include at least one of the following: missing teeth, periodontal disease, carious damage to the tooth structure, periapical abnormalities, pulp abnormalities, gingival abnormalities, and periodontal and alveolar bone abnormalities.

[0047] S4. Based on the condition of the tooth to be repaired, construct a decision tree corresponding to the tooth to be repaired;

[0048] S5. From the preset database, find the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired;

[0049] The pre-set database contains multiple removable denture design schemes and a decision tree corresponding to each removable denture design scheme. The similarity between the decision tree corresponding to the tooth to be restored and each decision tree pre-set in the pre-set database is calculated, and the decision tree with the highest similarity is determined as the target decision tree.

[0050] Among them, teeth to be restored refer to teeth that require the fabrication of removable dentures.

[0051] S6. When the maximum similarity is greater than the preset similarity threshold, the removable denture design scheme corresponding to the target decision tree is determined as the removable denture design scheme corresponding to the tooth to be restored, so as to facilitate the fabrication of the removable denture according to the removable denture design scheme corresponding to the tooth to be restored.

[0052] When the maximum similarity is not greater than the preset similarity threshold, the human intervention determines the removable denture design scheme corresponding to the tooth to be restored, and stores the removable denture design scheme corresponding to the tooth to be restored and the decision tree corresponding to the tooth to be restored in the preset database. The specific value of the preset similarity threshold can be set according to the actual situation.

[0053] This invention can quickly and accurately determine the design scheme of the removable denture corresponding to the tooth to be restored, which can save medical resources and reduce the medical costs for users.

[0054] Optionally, in the above technical solution, the process of obtaining the trained network model includes:

[0055] The Mask Transfiner is trained based on a pre-defined sample set to obtain a trained network model.

[0056] The sample in the preset sample set refers to a pre-existing panoramic film dataset with corresponding partial removable denture design schemes.

[0057] The network model used in this invention to extract data features from panoramic images is the Mask Transfiner, a high-quality instance segmentation algorithm based on Transformer. Unlike conventional methods, Transfiner first identifies error-prone pixel regions that require optimization and represents these pixels using a quadtree structure. These pixels are calculated based on the information loss of the downsampled object mask and are mainly distributed in object boundaries or high-frequency regions. They are spatially discontinuous and are called information loss regions. They are sparse, accounting for only a small portion of the total pixels. However, this part of the region is very important for the final segmentation performance. This also allows Transfiner to process only a small portion of the high-resolution feature map during the prediction of the object mask.

[0058] Transfiner transforms these discrete nodes into an unordered sequence of pixels. It comprises three modules: a Node Encoder, a Sequence Encoder, and a Pixel Decoder. The Node Encoder first enriches the feature representation of each point, such as supplementing the point's location encoding and local details of neighboring points. To model the relationships between points, the multi-head attention module in the Sequence Encoder performs feature fusion and updates between points in the input sequence. The Sequence Encoder takes the node sequence as the query input in the Transformer-based Sequence Encoder. Each layer of the Sequence Encoder consists of a multi-head self-attention module and a fully connected feedforward network (FFN). To supplement the input sequence with sufficient foreground and background information, we also input 196 feature points of size 14×14 from the lowest layer of the RoI pyramid. The Pixel Decoder predicts the instance label corresponding to each point. Unlike the standard Transformer decoder, Transfiner's pixel decoder is a simple two-layer MLP without a multi-head attention module. It decodes the output query of each node in the tree to predict the final instance label. The serialized input of feature points and multi-head attention enable Transfiner to flexibly model sparse feature points of multiple scales simultaneously, and to perform long-distance comparisons and cross-level associations of the relationships between each feature point.

[0059] After extracting data features such as tooth loss, periodontal pathology, gingival abnormalities, and pulpal abnormalities using Mask Transfiner, a classification decision tree is constructed using gain rate as the index for these features and the corresponding partial removable denture design schemes from panoramic radiographs. Furthermore, the panoramic radiographs and the corresponding partial removable denture design schemes together form the case database of this invention.

[0060] Preliminary matching module: In the algorithm processing module, this invention utilizes a pre-trained neural network model to extract data features such as missing teeth, periodontal pathology, gingival abnormalities, and pulp abnormalities from the panoramic image. These features are then matched against a pre-generated classification decision tree. After matching, the probability of the corresponding category is obtained. If the probability value is greater than a pre-set threshold, the input panoramic image and the matched solution are considered to have a high similarity, and the image is added to the existing case database. If the probability value is less than the pre-set threshold, the input panoramic image needs to be sent to the case reasoning module for partial removable denture design according to empirical rules.

[0061] In another embodiment, based on the tooth loss situation, edentulism can be classified into the following four categories according to the currently popular Ken's classification: Ken's Class 1, bilateral free tooth loss; Ken's Class 2, unilateral free tooth loss; and Ken's Classes 3 and 4, non-free tooth loss. Once the tooth loss situation is determined, other dental pathological features of the patient can be obtained through a pre-trained Mask transfiner. These features can then be used to identify healthy teeth near the missing teeth, which can be used as abutment teeth for constructing partial removable dentures, where rests, clasps, and bases can be placed. Combining these two methods allows for the design of the partial removable denture corresponding to the panoramic radiograph of the oral cavity, and the design is then populated into an existing case database, i.e., a pre-set database.

[0062] This invention enables the rapid generation of partial removable denture designs using only panoramic radiographs, saving medical costs, reducing diagnosis time, and lowering patient expenses. Furthermore, this invention employs a pre-trained Mask Transfiner model when extracting features from panoramic radiograph data. The mask transfiner captures rich positional information features from the panoramic radiographs, allowing the network model to focus on both global and local information features while reducing computational load, thus improving classification accuracy. Simultaneously, this invention calculates the similarity to other cases in a case database when generating the design, making the generated design more accurate and objective.

[0063] In the above embodiments, although the steps are numbered S1, S2, etc., they are only specific embodiments given in this application. Those skilled in the art can adjust the execution order of S1, S2, etc. according to the actual situation, which is also within the protection scope of this invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0064] like Figure 2 As shown, a system 200 for automatically generating removable denture design schemes according to an embodiment of the present invention includes an acquisition module 210, an enhancement module 220, an identification module 230, a construction module 240, a matching module 250, and a determination module 260.

[0065] The acquisition module 210 is used to: acquire panoramic radiographs of the target user's oral cavity;

[0066] Enhancement module 220 is used to perform data enhancement operations on panoramic radiographs of the oral cavity.

[0067] The recognition module 230 is used to: use a trained network model to recognize the data-augmented panoramic radiograph of the oral cavity to obtain the lesion status of the tooth to be repaired;

[0068] The construction module 240 is used to: construct a decision tree corresponding to the tooth to be repaired based on the lesion condition of the tooth to be repaired;

[0069] The matching module 250 is used to: match the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired from the preset database;

[0070] The determination module 260 is used to: determine the removable denture design scheme corresponding to the target decision tree as the removable denture design scheme corresponding to the tooth to be restored when the maximum similarity is greater than the preset similarity threshold.

[0071] This invention can quickly and accurately determine the design scheme of the removable denture corresponding to the tooth to be restored, which can save medical resources and reduce the medical costs for users.

[0072] Optionally, in the above technical solution, the enhancement module is specifically used for:

[0073] The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

[0074] Optionally, the above technical solution also includes a training module, which is used for:

[0075] The Mask Transfiner is trained based on a pre-defined sample set to obtain a trained network model.

[0076] Optionally, in the above technical solution, the pathological condition includes at least one of the following: missing teeth, periodontal disease, carious damage to the tooth structure, periapical abnormalities, pulp abnormalities, gingival abnormalities, and periodontal and alveolar bone abnormalities.

[0077] The parameters and steps of each unit module in the system 200 for automatically generating a removable denture design scheme of the present invention described above can be referred to the parameters and steps in the embodiments of the method for automatically generating a removable denture design scheme described above, and will not be repeated here.

[0078] An embodiment of the present invention provides a storage medium storing instructions, which, when read by a computer, cause the computer to execute any of the above-mentioned methods for automatically generating a removable denture design scheme.

[0079] An electronic device according to an embodiment of the present invention includes a processor and the aforementioned storage medium. The processor executes instructions in the storage medium. The electronic device may be a computer, a mobile phone, or the like.

[0080] Those skilled in the art will know that this invention can be implemented as a system, method, or computer program product.

[0081] Therefore, this disclosure can be implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product in one or more computer-readable media, the computer-readable medium containing computer-readable program code.

[0082] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0083] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for automatically generating removable denture design schemes, characterized in that, include: Obtain panoramic radiographs of the target user's oral cavity; Data augmentation was performed on the panoramic radiograph of the oral cavity. A trained network model is used to identify the lesions in a data-augmented panoramic radiograph of the oral cavity to determine the condition of the tooth to be repaired; wherein, the process of acquiring the trained network model includes: The Mask Transfiner is trained based on a preset sample set to obtain the trained network model; the samples in the preset sample set refer to a pre-existing panoramic image dataset corresponding to the local active denture design scheme. Based on the pathological condition of the tooth to be repaired, a decision tree is constructed corresponding to the tooth to be repaired; the pathological condition includes at least one of the following: tooth loss, carious damage to the tooth, periapical abnormality, pulp abnormality, gingival abnormality, and alveolar bone abnormality; From the preset database, the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired is selected; The preset database contains multiple removable denture design schemes and decision trees corresponding to each removable denture design scheme. The similarity between the decision tree corresponding to the tooth to be restored and each decision tree stored in the preset database is calculated, and the decision tree with the highest similarity is determined as the target decision tree. When the maximum similarity is greater than the preset similarity threshold, the removable denture design scheme corresponding to the target decision tree is determined as the removable denture design scheme corresponding to the tooth to be repaired.

2. The method for automatically generating a removable denture design scheme according to claim 1, characterized in that, Data augmentation operations are performed on the aforementioned panoramic radiographs of the oral cavity, including: The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

3. A system for automatically generating removable denture design schemes, characterized in that, It includes an acquisition module, an enhancement module, an identification module, a construction module, a matching module, and a determination module; it also includes a training module. The acquisition module is used to: acquire panoramic radiographs of the target user's oral cavity; The enhancement module is used to perform data enhancement operations on the panoramic radiograph of the oral cavity. The training module is used to: train Mask Transfiner based on a preset sample set to obtain a trained network model; the samples in the preset sample set refer to a pre-existing panoramic image dataset corresponding to the local active denture design scheme. The recognition module is used to: use the trained network model to recognize the data-augmented panoramic radiograph of the oral cavity to obtain the lesion status of the tooth to be repaired; The construction module is used to: construct a decision tree corresponding to the tooth to be repaired based on the lesion condition of the tooth to be repaired; the lesion condition includes at least one of the following: tooth loss, carious damage to the tooth, periapical abnormality, pulp abnormality, gingival abnormality and alveolar bone abnormality; The matching module is used to: match the target decision tree with the highest similarity to the decision tree corresponding to the tooth to be repaired from a preset database; The preset database contains multiple removable denture design schemes and decision trees corresponding to each removable denture design scheme. The similarity between the decision tree corresponding to the tooth to be restored and each decision tree stored in the preset database is calculated, and the decision tree with the highest similarity is determined as the target decision tree. The determining module is used to: when the maximum similarity is greater than a preset similarity threshold, determine the removable denture design scheme corresponding to the target decision tree as the removable denture design scheme corresponding to the tooth to be restored.

4. The system for automatically generating removable denture design schemes according to claim 3, characterized in that, The enhancement module is specifically used for: The panoramic radiographs of the oral cavity were sequentially subjected to horizontal flipping, normalization, standardization, random adjustment of contrast, random adjustment of brightness, and histogram equalization.

5. A storage medium, characterized in that, The storage medium stores instructions that, when read by a computer, cause the computer to execute a method for automatically generating a removable denture design as described in any one of claims 1 to 2.

6. An electronic device, characterized in that, It includes a processor and the storage medium of claim 5, wherein the processor executes instructions in the storage medium.

Images