Method for detecting wearing fitting degree of shell-shaped dental instrument
Through the deep neural network detection method performed by computers, the wear fit of shell-shaped dental devices is detected, solving the problems of difficulty in detection and frequent follow-up visits in the prior art, and achieving convenient and efficient fit detection.
Patent Information
- Application Number
- CN202311389755.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-24
- Publication Date
- 2025-05-02
AI Technical Summary
During the orthodontic treatment, it is difficult to wear and fit the shell-shaped dental equipment to be easily tested, resulting in frequent visits to the clinic, which takes up a lot of time and energy.
Using a computer-executed method, by obtaining photos of the patient's teeth wearing shell-shaped dental instruments, using the trained first deep neural network to detect the unfit area, and using the second deep neural network to segment the teeth and assign the tooth numbers to calculate the final fit detection result.
It significantly reduces the number of times patients visit the clinic, reduces the burden on patients and dental professionals, and provides a more convenient method of fit detection.
Smart Images
Figure CN119919332A_ABST
Abstract
Description
Technical Field
[0001] The present application generally relates to a method for improving the fit of a shell-shaped dental appliance. Background Art
[0002] Shell-shaped dental appliances are becoming more and more popular due to their advantages in appearance, convenience, and ease of cleaning. Orthodontic treatment using shell-shaped dental appliances usually requires dozens of shell-shaped dental appliances in succession. During orthodontic treatment using shell-shaped dental appliances, the actual orthodontic treatment may not be consistent with the design plan, and in severe cases, the shells may come off. Therefore, patients need to return to the dental clinic regularly to have dental professionals check whether there are any areas where their shell-shaped dental appliances do not fit. However, multiple return visits by patients take up a lot of time and energy for both patients and dental professionals, which is a heavy burden for both.
[0003] Therefore, it is necessary to provide a new and more convenient method for detecting the fit of shell-shaped dental appliances. Summary of the invention
[0004] One aspect of the present application provides a computer-implemented method for detecting the fit of a shell-shaped dental appliance, comprising: obtaining a photograph of a patient's teeth wearing the shell-shaped dental appliance; using a trained first deep neural network to detect a non-fitting area between the teeth in the photograph and the shell-shaped dental appliance; using a trained second deep neural network to segment the teeth in the photograph and assign a tooth number to each segmented tooth; and calculating a final detection result based on the results output by the first and second deep neural networks, which includes the detected non-fitting area and its corresponding tooth number.
[0005] In some implementations, each of the detected misfit regions is a misfit region corresponding to a single tooth.
[0006] In some embodiments, the first deep neural network classifies the detected non-fitting areas according to the degree of non-fitting, and the final detection result includes the classification of the degree of non-fitting of the detected non-fitting areas.
[0007] In some implementations, the first deep neural network is an instance segmentation network.
[0008] In some implementations, the second deep neural network is an instance segmentation network.
[0009] In some embodiments, the calculation is based on a distance or an intersection-over-union ratio between the detected misaligned area and the segmented teeth.
[0010] In some embodiments, the shell-shaped dental appliance is a shell-shaped dental appliance, and the photograph is taken when the user starts to wear or finishes wearing the shell-shaped dental appliance.
[0011] Another aspect of the present application provides a computer system for detecting the fit of a shell-shaped dental appliance, which includes a storage device and a processor. The storage device stores a computer program. When the computer program is run, the processor will execute the method for detecting the fit of a shell-shaped dental appliance. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The above and other features of the present application will be further described below in conjunction with the accompanying drawings and detailed descriptions thereof. It should be understood that these drawings only illustrate several exemplary embodiments according to the present application and should not be considered as limiting the scope of protection of the present application. Unless otherwise specified, the drawings are not necessarily to scale and similar reference numerals represent similar components.
[0013] Figure 1 It is a schematic flow chart of a method for detecting the wearing fit of a shell-shaped dental appliance in one embodiment of the present application;
[0014] Figure 2 A photograph of a patient's teeth wearing a shell-shaped dental appliance in one example;
[0015] Figure 3 Demonstrated the use of a second deep neural network based on Figure 2 The result of teeth segmentation on the photo shown; and
[0016] Figure 4 The results based on the output of the first and second deep neural networks are shown. Figure 2 Results of the fit test between the teeth and the shell-shaped dental appliance in the photos shown. DETAILED DESCRIPTION
[0017] The following detailed description refers to the drawings that form part of this specification. The illustrative embodiments mentioned in the specification and drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present application. Under the guidance of the present application, those skilled in the art will understand that many other embodiments can be adopted and various changes can be made to the described embodiments without departing from the subject matter and scope of protection of the present application. It should be understood that the various aspects of the present application described and illustrated herein can be arranged, replaced, combined, separated and designed according to many different configurations, and these different configurations are within the scope of protection of the present application.
[0018] One aspect of the present application provides a method for detecting the fit of a shell-shaped dental appliance, which uses a trained deep neural network to detect the non-fitting area between the shell-shaped dental appliance and the teeth based on photos of the patient's teeth wearing the shell-shaped dental appliance. Since the photos can be taken by the patient himself or his relatives and friends using appropriate equipment anywhere, this significantly reduces the number of times the patient returns to the clinic, alleviating the burden on patients and dental professionals.
[0019] Please refer to Figure 1 , which is a schematic flow chart of a method 100 for detecting the wearing fit of a shell-shaped dental appliance in one embodiment of the present application.
[0020] At 101, a photograph of a patient's teeth wearing a shell-shaped dental appliance is obtained.
[0021] In one embodiment, the photo can be taken by the patient himself or others using appropriate equipment, for example, the MOOELI product of Hangzhou Chaohou Information Technology Co., Ltd.
[0022] In one embodiment, the shell-shaped dental appliance can be a shell-shaped dental appliance for repositioning the patient's teeth from a first tooth configuration to a second tooth configuration. In another embodiment, the shell-shaped dental appliance can also be a retainer for retaining the patient's teeth in a current tooth configuration. In addition to the shell-shaped dental appliance and retainer, the shell-shaped dental appliance can also be a dental appliance with other uses.
[0023] The shell-shaped dental appliance is made of a transparent material, and is an integrated shell that forms a cavity for accommodating a plurality of teeth. The tooth receiving cavity of a more common shell-shaped dental appliance can accommodate a complete dentition, for example, the maxillary dentition or the mandibular dentition.
[0024] Since the dental arch is U-shaped, in order to detect the fit of the shell-shaped dental appliance at each tooth, it is usually necessary to take multiple photos of the teeth wearing the shell-shaped dental appliance from different angles (for example, one photo each from the left, middle, and right angles), and then analyze each photo to obtain the fit of the shell-shaped dental appliance at each tooth.
[0025] In one embodiment, the photograph of the patient's teeth wearing the shell-shaped dental appliance can be taken at a predetermined node during the orthodontic treatment of the patient, for example, when starting to wear or completing wearing each of several pre-specified shell-shaped dental appliances.
[0026] Please refer to Figure 2 , is a photograph of a patient's teeth wearing a shell-shaped dental appliance in one example.
[0027] In 103, a mismatch area between each tooth and the shell-shaped dental appliance in the photo is detected using the trained first deep neural network.
[0028] The non-fit area is a gap or cavity between a tooth and the shell-like dental appliance.
[0029] In one embodiment, the first deep neural network is an instance segmentation network, such as a YOLO (You Only Look Once) network or a Mask R-CNN network.
[0030] In one embodiment, the first deep neural network can be trained with more than 2,000 annotated photos, and more preferably, can be trained with more than 5,000 annotated photos.
[0031] Each of the misfit regions detected by the first deep neural network corresponds to a single tooth. If there is a connected misfit region across two teeth in the photo, the first deep neural network detects the parts belonging to their respective misfit regions for the two teeth. Accordingly, each of the misfit regions annotated in the annotated photos used to train the first deep neural network corresponds to a single tooth.
[0032] In one embodiment, when the first deep neural network detects a non-fitting area, it can also classify the detected non-fitting area according to the degree of non-fitting. For example, the non-fitting area where the distance between the tooth and the shell-shaped dental appliance is less than 0.75mm is classified into the first category, the non-fitting area where the distance between the tooth and the shell-shaped dental appliance is greater than 0.75mm and less than 1.5mm is classified into the second category, and the non-fitting area where the distance between the tooth and the shell-shaped dental appliance is greater than 1.5mm is classified into the third category. It can be understood that the above categories and their thresholds are only exemplary, and the number of categories and their thresholds can be determined according to specific needs.
[0033] In some cases, the size of the gap (cavitation) between the tooth and the shell-shaped dental appliance is significantly larger than the reasonable value. This abnormal gap is not caused by the difference between the geometry of the tooth receiving cavity of the shell-shaped dental appliance and the geometry of the tooth, and does not belong to the case of poor fit. For example, the growth space left for the erupting tooth, and another example is the cavitation of the jaw plate and other structures.
[0034] In one embodiment, the first deep neural network may classify such detected abnormal gaps as abnormalities, and the gaps classified as abnormalities will not be included in the detection results of the non-fitting areas.
[0035] In 105, each tooth in the photo is segmented using the trained second deep neural network.
[0036] In one embodiment, the second deep neural network is an instance segmentation network, such as a YOLO (You Only Look Once) network, a Mask R-CNN network, or a DETR network (Detection Transformer).
[0037] The second deep neural network segments each tooth in the photo and assigns a tooth number to the tooth.
[0038] Please refer to Figure 3 , demonstrating the use of the second deep neural network Figure 2 The results of segmenting each tooth in the photo shown.
[0039] In 107, based on the detection result output by the first deep neural network and the segmentation result output by the second deep neural network, the fit detection result of each single tooth is calculated.
[0040] Although the first deep neural network has segmented the non-fitting areas corresponding to each tooth in the photo, it is still unknown to which tooth number these non-fitting areas specifically correspond. Therefore, based on the detection results of the non-fitting areas output by the first deep neural network and the tooth segmentation results output by the second deep neural network, the fit degree detection results of each single tooth can be calculated, that is, a tooth number is assigned to each detected non-fitting area.
[0041] In one embodiment, for a detected misfit region, the tooth number of the tooth closest to the misfit region may be assigned to the misfit region.
[0042] In another embodiment, for a detected mismatching area, an intersection-over-union ratio between it and each segmented tooth may be calculated, and the tooth number of the tooth with the highest intersection-over-union ratio may be assigned to the mismatching area.
[0043] Please refer to Figure 4 , demonstrating the use of the method of the present application Figure 2 The results of the fit test of the shell-shaped dental appliance on the tooth photo shown, which divides the non-fitting area, classifies the non-fitting area according to the degree of non-fitting, and assigns a tooth number to each non-fitting area.
[0044] In yet another embodiment, in addition to the graphical form, the test result of the wearing fit of the shell-shaped dental appliance of the present application can also be expressed in text form.
[0045] Another aspect of the present application provides a computer system for detecting the fit of a shell-shaped dental appliance, which includes a processor and a storage device. The storage device stores a computer program. When the computer program is run, the processor executes the method 100 for detecting the fit of a shell-shaped dental appliance.
[0046] Although various aspects and embodiments of the present application are disclosed herein, other aspects and embodiments of the present application will be apparent to those skilled in the art in light of the present application. The various aspects and embodiments disclosed herein are for illustrative purposes only and not for limiting purposes. The scope and subject matter of the present application are determined solely by the appended claims.
[0047] Likewise, various diagrams may illustrate exemplary architectures or other configurations of the disclosed methods and systems that aid in understanding the features and functions that may be included in the disclosed methods and systems. The claimed content is not limited to the exemplary architectures or configurations shown, and the desired features may be implemented with various alternative architectures and configurations. In addition, for flow charts, functional descriptions, and method claims, the order of blocks presented herein should not be limited to various embodiments that are implemented in the same order to perform the described functions, unless otherwise clearly indicated in the context.
[0048] Unless otherwise expressly noted, the terms and phrases used herein and their variations should be interpreted as open ended, not restrictive. In some instances, the appearance of broad words and phrases such as "one or more", "at least", "but not limited to", or other similar terms should not be understood as an intent or need to indicate a narrowing of the example where such broad terms may not be present.
Claims
1. A computer-implemented method for detecting the fit of a shell-shaped dental appliance, comprising: Obtain a photograph of the patient's teeth with the shell-shaped dental appliance; Detecting a non-fitting area between the teeth in the photo and the shell-shaped dental appliance using a trained first deep neural network; Segmenting the teeth in the photo using the trained second deep neural network, and assigning a tooth number to each segmented tooth; as well as Based on the results output by the first and second deep neural networks, a final detection result is calculated, which includes the detected non-fitting area and its corresponding tooth number.
2. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: Each of the detected mismatching regions is a mismatching region corresponding to a single tooth.
3. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: The first deep neural network classifies the detected non-fitting areas according to the degree of non-fitting, and the final detection result includes the classification of the degree of non-fitting of the detected non-fitting areas.
4. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: The first deep neural network is an instance segmentation network.
5. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: The second deep neural network is an instance segmentation network.
6. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: The calculation is based on a distance or an intersection-over-union ratio between the detected misaligned area and the segmented teeth.
7. The method for detecting the wearing fit of a shell-shaped dental appliance according to claim 1, characterized in that: The shell-shaped dental appliance is a shell-shaped dental appliance, and the photo is taken when the shell-shaped dental appliance begins to be worn or is finished being worn.
8. A computer system for detecting the fit of a shell-shaped dental appliance, comprising a storage device and a processor, wherein the storage device stores a computer program, and when the computer program is run, the processor executes the method for detecting the fit of a shell-shaped dental appliance as claimed in claim 1.