Pneumatic gingival retraction control method based on artificial intelligence and intraoral three-dimensional scanner using control method
By constructing an improved convolutional neural network model and using the Laplacian operator to detect image sharpness, and combining it with an autonomously controlled air blowing device, the problem of autonomous control of angle and airflow intensity during pneumatic gum removal was solved, thereby improving the scanner's operating efficiency and image quality.
Patent Information
- Application Number
- CN202411147884.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-29
AI Technical Summary
Existing pneumatic gingival retraction 3D scanners require operators to adjust the blowing angle and airflow intensity based on their experience during clinical operation, which is inconvenient and affects the scanning quality. Furthermore, the airflow can easily cause mirror splashing, increasing the operation time.
An artificial intelligence-based pneumatic gingival retraction control method is adopted. By constructing an improved convolutional neural network model and Laplacian operator to calculate image sharpness, the system achieves autonomous control of the optimal blowing angle and airflow intensity during pneumatic gingival retraction. It is also equipped with an autonomously adjustable blowing pipe and cleaning device.
It enables autonomous control of the optimal blowing angle and airflow intensity during pneumatic gingival retraction, reducing the inconvenience of clinical operation, improving the accuracy and stability of intraoral scanning images, and avoiding the influence of mirror splash.
Smart Images

Figure CN119033493A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of oral medicine, and in particular to an artificial intelligence-based pneumatic gingival retraction control method and an intraoral three-dimensional scanner utilizing the control method. Background Art
[0002] Currently, the method of using pneumatic gingival retraction to obtain intraoral three-dimensional scanning data of the edge of the tooth preparation has begun to be tested in clinical practice. This method can effectively blow away the free gingival tissue and obtain a clear edge morphology of the tooth preparation.
[0003] As a device that uses digital imaging technology to perform three-dimensional scanning and measurement of the oral area, the intraoral 3D scanner mainly records the surface data of teeth and gums to make digital impressions, and copies the intraoral situation to the outside of the mouth for restoration production, implant location design, orthodontic treatment plan formulation, etc. At present, pneumatic gingival retraction technology relies on intraoral 3D scanners. In the process of using pneumatic gingival retraction intraoral 3D scanners, the blowing direction, airflow intensity and blowing channel distance have a great influence on the distance of the gums being blown open. Close distance, high flow rate and appropriate angle can blow open the gingival tissue to the greatest extent, fully expose the edge morphology of the tooth preparation, and make it easier for the intraoral 3D scanner to capture the edge information of the preparation and obtain higher quality edge morphology data. However, the pneumatic gingival retraction 3D intraoral scanner has a fixed blowing channel, and the operator needs to try many times based on experience to find a suitable blowing angle to fully expose the edge of the preparation, which brings inconvenience to clinical operation. In addition, the airflow can easily cause the mirror surface to be splashed with saliva and blood, affecting the scanning quality, and the operator needs to clean it in time, which greatly increases the clinical operation time.
[0004] Therefore, how to provide an artificial intelligence-based pneumatic gingival retraction control method and an intraoral 3D scanner using the control method that can autonomously control the optimal blowing angle and airflow intensity during pneumatic gingival retraction, reduce the inconvenience of clinical operation, and effectively improve the accuracy of intraoral scanning images is an urgent problem that technical personnel in this field need to solve. Summary of the invention
[0005] In view of this, the present invention proposes an artificial intelligence-based pneumatic gingival retraction control method and an intraoral three-dimensional scanner using the control method.
[0006] In order to achieve the above object, the present invention adopts the following technical solution:
[0007] A pneumatic gingival retraction control method based on artificial intelligence, comprising:
[0008] Step 1: Obtain an intraoral scan image and the air outlet angle and blowing intensity corresponding to the intraoral scan image, and mark whether the gums are effectively blown open to obtain a training set;
[0009] Step 2: construct an improved convolutional neural network structure and train the training set to obtain a pneumatic gingival retraction control output model; wherein each convolution layer of the improved convolutional neural network structure includes: a convolution layer, an attention module, and a spatial pyramid pooling layer connected in sequence, and the output of the first convolution layer is connected to the output of the last convolution layer through a residual connection;
[0010] Step 3: Use the Laplace operator to calculate the second-order derivative of the image to be tested, determine the clarity of the image to be tested, input the clear image to be tested into the pneumatic gingival retraction control output model, obtain the air outlet angle and blowing intensity that effectively blows open the gums, and perform pneumatic gingival retraction control.
[0011] Optionally, in step 1, before constructing a training set based on the annotated intraoral scanned image, the step further includes: performing image grayscale transformation and histogram correction on the annotated intraoral scanned image, and converting the color image into a grayscale image using binarization.
[0012] Optionally, in step 2, the convolutional layer is as follows:
[0013]
[0014] Among them, y is the output of the convolution calculation; i and j are the positions of the output; k is the size of the convolution kernel; m and n are the positions of the convolution kernel; x is the input of the convolution calculation; and w is the convolution kernel.
[0015] Optionally, in step 2, the attention module is as follows:
[0016] Attention(x)=σ(MLP(AvgPool(x))+MLP(MaxPool(x)));
[0017] Among them, Attention is the output of attention; σ is the sigmoid function; MLP is the fully connected layer; AvgPool and MaxPool are average pooling and maximum pooling operations respectively; x is the input of attention.
[0018] Optionally, in step 3, the second-order derivative of the image to be tested is calculated using the Laplace operator to determine the clarity of the image to be tested, specifically:
[0019] Convert the input color image to a gray image, and use the Laplace operator to process the gray image and calculate the second-order derivative as follows:
[0020]
[0021] The variance operation is performed on the Laplace transform result, and it is determined whether the variance operation result is greater than a preset threshold. If it is greater than, it means that the image is clear, and if it is less than, it means that the image is blurred.
[0022] Optionally, in step 3, after performing pneumatic gingival retraction control based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model, it also includes: detecting whether the shoulder edge line of the shoulder preparation is clearly exposed, and the depth of the gums blown open in the non-shoulder area according to the intraoral scanning image after the pneumatic gingival retraction control, and based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model, adaptively adjusting the air outlet tube angle and the air blowing intensity to perform supplementary scanning until a complete subgingival edge is obtained.
[0023] Optionally, it also includes: updating the pneumatic gingival retraction control output model in real time based on the training set and the supplementary scanning data.
[0024] The present invention also provides an intraoral three-dimensional scanner using an artificial intelligence-based pneumatic gingival retraction control method, comprising: an air blowing pipeline that can be autonomously regulated, a reflective lens, an air blowing pipe direction control device, an air blowing pipe flow control device, a scanner main body lens, an integrated control module, an external independent air pump, and an automatic wiper;
[0025] The integrated control module controls the scanner main body lens to obtain the intraoral scan image reflected by the reflective lens, and uses the Laplace operator to calculate the second-order derivative of the intraoral scan image to determine the clarity of the intraoral scan image; when the intraoral scan image is blurred, the integrated control module controls the direction of the self-adjustable air blowing pipe to align with the reflective lens through the air blowing pipe direction control device, and controls the air flow of the self-adjustable air blowing pipe through the air outlet pipe flow control device, and cooperates with the automatic wiper to clean the reflective lens;
[0026] The integrated control module uses the pre-trained pneumatic gingival retraction control output model to process the clear intraoral scan image, obtains the air outlet angle and air blowing intensity that can effectively blow open the gums, and controls the pneumatic gingival retraction of the self-regulatory air pipe through the air outlet direction control device and the air outlet flow control device to obtain the subgingival edge information of the tooth preparation.
[0027] An external independent air pump is connected to the air blowing pipe flow control device to provide a compressed air source.
[0028] Optionally, the integrated control module is also used to perform pneumatic gingival retraction control based on the air outlet pipe angle and air blowing intensity output by the pneumatic gingival retraction control output model, and after obtaining the subgingival margin information of the tooth preparation, detect whether the shoulder margin line of the shoulder preparation is clearly exposed, and the depth of the gingiva blown open in the non-shoulder area, and based on the air outlet pipe angle and air blowing intensity output by the pneumatic gingival retraction control output model, perform adaptive adjustment of the air outlet pipe angle and air blowing intensity to supplement the sweep until a complete subgingival margin is obtained.
[0029] Optionally, it also includes: updating the pneumatic gingival retraction control output model in real time based on the training set and the supplementary scanning data.
[0030] It can be seen from the above technical solutions that, compared with the prior art, the present invention proposes a pneumatic gingival retraction control method based on artificial intelligence and an intraoral three-dimensional scanner using the control method. Based on artificial intelligence, a pneumatic gingival retraction control output model is trained to output the optimal blowing angle and airflow intensity under different conditions during the pneumatic gingival retraction process, thereby realizing autonomous control of the optimal blowing angle and airflow intensity during the pneumatic gingival retraction process, reducing the inconvenience of clinical operation, and effectively improving the accuracy of intraoral scanned images. In addition, by calculating the second-order derivative of the image to be tested based on the Laplace operator to detect the clarity of the intraoral scanned image, real-time detection and processing of whether the reflective lens is affected by splashing saliva or blood during the pneumatic gingival retraction process is realized, further improving the acquisition accuracy of the intraoral scanned image. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0032] Figure 1 It is a schematic diagram of the method flow of the present invention.
[0033] Figure 2 Schematic diagram of the improved convolutional neural network structure of the present invention. DETAILED DESCRIPTION
[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0035] Embodiment 1:
[0036] Embodiment 1 of the present invention discloses a pneumatic gingival retraction control method based on artificial intelligence, such as Figure 1 As shown, including:
[0037] Step 1: Obtain the intraoral scan image and the air outlet angle and blowing intensity corresponding to the intraoral scan image (including the gum status under various conditions, such as being blown open and not blown open, gum thickness, whether there is gingival sulcus fluid and blood, etc.), and mark whether the gums are effectively blown open to obtain a training set.
[0038] Before constructing a training set based on the annotated intraoral scanned images, the process also includes: performing grayscale transformation and histogram correction on the annotated intraoral scanned images to reduce the impact of noise, and using binarization to convert color images into grayscale images to reduce the amount of calculation, making it easier for subsequent algorithms to extract key features related to the gum state, such as edges, colors, textures, and shapes, etc.
[0039] Step 2: Construct an improved convolutional neural network structure, such as Figure 2 As shown in the figure, the training set is trained to obtain the pneumatic gingival retraction control output model; wherein, each convolution layer of the improved convolutional neural network structure includes: sequentially connected convolution layers, attention modules (in the convolution process, it is necessary to pay attention to the gingival depth of a specific area (the area without shoulder), so an attention module (Attention) is added, and in the process of extracting features from image convolution, more weights are given to important features, thereby improving the network prediction effect) and spatial pyramid pooling layers (considering that observing gingival images at different scales may have different effects, spatial pyramid pooling (SPP) is introduced to aggregate contextual information of different scales, so that the model can better understand and identify features in the image), and the output of the first convolution layer is connected to the output of the last convolution layer through a residual connection (which helps the model learn more complex features).
[0040] It also includes: using the test set data to test the model performance and adjusting the model structure or parameters as needed. After that, performance evaluation is performed, using the validation data set that is not involved in the training to evaluate the accuracy and generalization ability of the model. The accuracy index is used to evaluate the model performance. According to the results of the performance evaluation, the learning rate, optimizer and other parameters are adjusted.
[0041] The convolutional layer is as follows:
[0042]
[0043] Among them, y is the output of the convolution calculation; i and j are the positions of the output; k is the size of the convolution kernel; m and n are the positions of the convolution kernel; x is the input of the convolution calculation; and w is the convolution kernel.
[0044] The attention module is as follows:
[0045] Attention(x)=σ(MLP(AvgPool(x))+MLP(MaxPool(x)));
[0046] Among them, Attention is the output of attention; σ is the sigmoid function; MLP is the fully connected layer; AvgPook and MaxPool are the average pooling and maximum pooling operations respectively; x is the input of attention. The global maximum value and average value in the image are extracted through the pooling operation, and then they are fused through the fully connected layer to obtain an attention coefficient, which will weight the input feature map and emphasize the most useful information, thereby improving the performance of the model.
[0047] Step 3: Use the Laplace operator to calculate the second-order derivative of the image to be tested, determine the clarity of the image to be tested, input the clear image to be tested into the pneumatic gingival retraction control output model, obtain the air outlet angle and blowing intensity that effectively blows open the gums, and perform pneumatic gingival retraction control.
[0048] The Laplace operator is used to calculate the second-order derivative of the image to be tested to determine the clarity of the image to be tested, specifically:
[0049] Convert the input color image to a gray image, and use the Laplace operator to process the gray image and calculate the second-order derivative as follows:
[0050]
[0051] The variance operation is performed on the Laplace transform result. The larger the variance, the more high-frequency components in the image, that is, the clearer the image; the smaller the variance, the blurrier the image. It is also determined whether the variance operation result is greater than the preset threshold. If it is greater, it means the image is clear; if it is less, it means the image is blurry.
[0052] By using the Laplace operator to process the clarity of the image in real time, it is possible to detect in real time whether the reflective lens is affected by splashing saliva or blood during the actual pneumatic gingival retraction process, so that it can be cleaned in time to improve the accuracy of intraoral scanning image acquisition.
[0053] After performing pneumatic gingival retraction control based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model, the method also includes: detecting whether the shoulder edge line of the shoulder preparation is clearly exposed, and the depth of the gingiva blown open in the non-shoulder area according to the intraoral scanning image after the pneumatic gingival retraction control, and adaptively adjusting the air outlet tube angle and air blowing intensity based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model to perform supplementary scanning until a complete subgingival edge is obtained.
[0054] Also includes: real-time update of pneumatic gingival retraction control output model based on training set and supplementary scanning data.
[0055] Embodiment 2:
[0056] Embodiment 2 of the present invention discloses an intraoral three-dimensional scanner using an artificial intelligence-based pneumatic gingival retraction control method described in the embodiment, comprising: an air blowing pipe that can be autonomously regulated, a reflective lens, an air blowing pipe direction control device, an air blowing pipe flow control device, a scanner main body lens, an integrated control module, an external independent air pump, and an automatic wiper;
[0057] The integrated control module controls the scanner main body lens to obtain the intraoral scan image reflected by the reflective lens, and uses the Laplace operator to calculate the second-order derivative of the intraoral scan image to determine the clarity of the intraoral scan image; when the intraoral scan image is blurred, the integrated control module controls the direction of the self-adjustable air blowing pipe to align with the reflective lens through the air blowing pipe direction control device, and controls the air flow of the self-adjustable air blowing pipe through the air outlet pipe flow control device, and cooperates with the automatic wiper to clean the reflective lens;
[0058] The integrated control module uses the pre-trained pneumatic gingival retraction control output model to process the clear intraoral scan image, obtains the air outlet angle and air blowing intensity that can effectively blow open the gums, and controls the pneumatic gingival retraction of the self-regulatory air pipe through the air outlet direction control device and the air outlet flow control device to obtain the subgingival edge information of the tooth preparation.
[0059] An external independent air pump is connected to the air blowing pipe flow control device to provide a compressed air source.
[0060] The integrated control module is also used to perform pneumatic gingival retraction control based on the air outlet pipe angle and air blowing intensity output by the pneumatic gingival retraction control output model. After obtaining the subgingival margin information of the tooth preparation, it is used to detect whether the shoulder margin line of the preparation with a shoulder is clearly exposed, and the depth of the gingiva to be blown open in the part without a shoulder. Based on the air outlet pipe angle and air blowing intensity output by the pneumatic gingival retraction control output model, the air outlet pipe angle and the air blowing intensity are adaptively adjusted to perform supplementary scanning until a complete subgingival margin is obtained.
[0061] Also includes: real-time update of pneumatic gingival retraction control output model based on training set and supplementary scanning data.
[0062] Also included: Provides a user interface that allows the physician to manually adjust the airflow intensity when needed. Manual and automatic mode switching: Allows the physician to switch between automatic and manual modes to respond to different clinical situations.
[0063] The embodiment of the present invention discloses a pneumatic gingival retraction control method based on artificial intelligence and an intraoral three-dimensional scanner using the control method. Based on artificial intelligence, a pneumatic gingival retraction control output model is trained to output the optimal blowing angle and airflow intensity under different conditions during the pneumatic gingival retraction process, thereby realizing autonomous control of the optimal blowing angle and airflow intensity during the pneumatic gingival retraction process, reducing the inconvenience of clinical operation, and effectively improving the accuracy of intraoral scanned images. In addition, by calculating the second-order derivative of the image to be tested based on the Laplace operator to detect the clarity of the intraoral scanned image, real-time detection and processing of whether the reflective lens is affected by splashing saliva and blood during the pneumatic gingival retraction process is realized, further improving the acquisition accuracy of the intraoral scanned image.
[0064] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0065] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A pneumatic gingival retraction control method based on artificial intelligence, characterized in that: include: Step 1: Obtain an intraoral scan image and the air outlet angle and blowing intensity corresponding to the intraoral scan image, and mark whether the gums are effectively blown open to obtain a training set; Step 2: constructing an improved convolutional neural network structure, and training the training set to obtain a pneumatic gingival retraction control output model; wherein each convolution layer of the improved convolutional neural network structure includes: a convolution layer, an attention module, and a spatial pyramid pooling layer connected in sequence, and the output of the first convolution layer is connected to the output of the last convolution layer through a residual connection; Step 3: Use the Laplace operator to calculate the second-order derivative of the image to be tested, determine the clarity of the image to be tested, input the clear image to be tested into the pneumatic gingival retraction control output model, obtain the air outlet angle and blowing intensity that effectively blows open the gums, and perform pneumatic gingival retraction control.
2. The artificial intelligence-based pneumatic gingival retraction control method according to claim 1, characterized in that: In step 1, before constructing the training set based on the annotated intraoral scanned image, the method further includes: performing image grayscale transformation and histogram correction on the annotated intraoral scanned image, and converting the color image into a grayscale image using binarization.
3. The artificial intelligence-based pneumatic gingival retraction control method according to claim 1, characterized in that: In step 2, the convolutional layer is as follows: Among them, y is the output of the convolution calculation; i and j are the positions of the output; k is the size of the convolution kernel; m and n are the positions of the convolution kernel; x is the input of the convolution calculation; and w is the convolution kernel.
4. The artificial intelligence-based pneumatic gingival retraction control method according to claim 1, characterized in that: In step 2, the attention module is as follows: Attention(x)=σ(MLP(AvgPool(x))+MLP(MaxPool(x))); Among them, Attention is the output of attention; σ is the sigmoid function; MLP is the fully connected layer; AvgPool and MaxPool are average pooling and maximum pooling operations respectively; x is the input of attention.
5. The method for controlling pneumatic gingival retraction based on artificial intelligence according to claim 1, characterized in that: In step 3, the second-order derivative of the image to be tested is calculated using the Laplace operator to determine the clarity of the image to be tested, specifically: Convert the input color image to a gray image, and use the Laplace operator to process the gray image and calculate the second-order derivative as follows: The variance operation is performed on the Laplace transform result, and it is determined whether the variance operation result is greater than a preset threshold. If it is greater than, it means that the image is clear, and if it is less than, it means that the image is blurred.
6. The artificial intelligence-based pneumatic gingival retraction control method according to claim 1, characterized in that: In step 3, after performing pneumatic gingival retraction control based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model, it also includes: detecting whether the shoulder edge line of the shoulder preparation is clearly exposed, and the depth of the jaw gums of the gums blown open in the non-shoulder area according to the intraoral scanning image after the pneumatic gingival retraction control, and based on the air outlet tube angle and air blowing intensity output by the pneumatic gingival retraction control output model, adaptively adjusting the air outlet tube angle and the air blowing intensity to perform supplementary scanning until a complete subgingival edge is obtained.
7. The artificial intelligence-based pneumatic gingival retraction control method according to claim 6, characterized in that: Also includes: Based on the training set and the supplementary scanning data, the pneumatic gingival retraction control output model is updated in real time.
8. An intraoral three-dimensional scanner using the artificial intelligence-based pneumatic gingival retraction control method according to any one of claims 1 to 7, characterized in that: include: It can independently control the air blowing pipe, reflective lens, air blowing pipe direction control device, air blowing pipe flow control device, scanner main body lens, integrated control module, external independent air pump, and automatic wiper; The integrated control module controls the scanner main body lens to obtain the intra-oral scan image reflected by the reflective lens, and uses the Laplace operator to calculate the second-order derivative of the intra-oral scan image to determine the clarity of the intra-oral scan image; when the intra-oral scan image is blurred, the integrated control module controls the direction of the self-adjustable air blowing pipe to align with the reflective lens through the air blowing pipe direction control device, and controls the air flow of the self-adjustable air blowing pipe through the air outlet pipe flow control device, and cooperates with the automatic wiper to clean the reflective lens; The integrated control module processes the clear intraoral scan image using the pre-trained pneumatic gingival retraction control output model to obtain the air outlet pipe angle and air blowing intensity that effectively blows the gums open, and controls the pneumatic gingival retraction of the self-regulatory air pipe through the air outlet pipe direction control device and the air outlet pipe flow control device to obtain the subgingival edge information of the tooth preparation; The external independent air pump is connected to the air blowing pipe flow control device to provide a compressed air source.
9. The intraoral three-dimensional scanner according to claim 8, characterized in that: The integrated control module is also used to perform the pneumatic gingival retraction control based on the air outlet pipe angle and the air blowing intensity output by the pneumatic gingival retraction control output model, and after obtaining the subgingival edge information of the tooth preparation, detect whether the shoulder edge line of the shoulder preparation is clearly exposed, and the depth of the gums blown open in the non-shoulder area, and based on the air outlet pipe angle and the air blowing intensity output by the pneumatic gingival retraction control output model, perform adaptive adjustment and supplementary scanning of the air outlet pipe angle and the air blowing intensity until a complete subgingival edge is obtained.
10. The intraoral three-dimensional scanner according to claim 9, characterized in that: Also includes: Based on the training set and the supplementary scanning data, the pneumatic gingival retraction control output model is updated in real time.
Citation Information
Cited By
Data processing method, device and equipment for assisting oral cavity scanning
CN120072211A
Data processing method, device and equipment for assisting oral scanning
CN120072211B
Artificial intelligence-based pneumatic gingival retraction control method and intraoral three-dimensional scanner using same
WO2026040730A1
Data processing method, apparatus, and device for assisting oral scanning
WO2026156939A1
