Electric tooth cleaning system based on personalized dentition digital information intelligent planning

Abstract

An electric tooth cleaning system based on personalized dentition digital information intelligent planning is provided with a dentition digital information acquisition module, and intraoral tooth dentition information of an object is obtained to obtain a three-dimensional model; the visual acquisition module is used for acquiring tooth cleaning data of the operation module in real time during tooth cleaning and sending the tooth cleaning data to the processing module; the operation module is used for cleaning teeth; the processing module is used for generating a three-dimensional model according to the dentition information of the intraoral teeth, then identifying the tooth condition and generating a target tooth cleaning scheme; performing real-time judgment according to the tooth cleaning data of the visual acquisition module and a target tooth cleaning scheme to obtain a judgment result and performing prompting; the target tooth cleaning scheme comprises a target cleaning route, a target tooth cleaning parameter for cleaning the heavy and difficult point area, and a target tooth cleaning parameter for non-cleaning the heavy and difficult point area. According to the electric tooth cleaning system, by digitally collecting personal oral cavity information, the important and difficult areas of oral cavity cleaning are recognized, and personalized tooth brushing planning is carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of oral health care equipment, and in particular to an electric teeth cleaning system based on intelligent planning of personalized dentition digital information. Background Art

[0002] Brushing teeth is an important means of personal oral health care, and mastering a reasonable way of brushing teeth is essential to maintaining oral health. Scientific brushing methods and treatments can effectively and comprehensively clean the oral cavity while minimizing damage to the soft and hard tissues of the oral cavity, and effectively help individuals prevent the occurrence of caries and periodontal disease. This is also in line with the requirements of the National Healthy Oral Action Plan (2019-2025).

[0003] Although the public's oral health awareness has improved in recent years, the overall situation of oral health care in my country is not optimistic, especially in the education of correct brushing. The most effective method of brushing teeth is the "Bass brushing method", but more than 80% of Chinese people use the horizontal brushing method, which is harmful to teeth; for areas with uneven teeth, large interproximal gaps, and undercuts, most people do not effectively focus on cleaning the dead corners of personal oral hygiene; the reasonable use of brushing time and frequency can also effectively improve the quality of brushing. Especially in the field of oral health care for infants and children in my country, if you master the correct brushing method and awareness as early as possible, it will benefit you for life. In response to this, many oral cleaning devices have been launched on the market, such as electric toothbrushes, water flossers, etc. Oral cleaning devices are usually equipped with multiple working modes to adapt to different cleaning needs of users. However, it is found in practice that users usually need to manually switch working modes, with low intelligence and cumbersome operation. In addition, traditional oral cleaning devices lack cleaning modes for non-tooth areas, resulting in inadequate cleaning of the oral cavity and reducing the cleaning effect of the oral cavity.

[0004] Therefore, in view of the shortcomings of the existing technology, it is necessary to provide an electric teeth cleaning system based on intelligent planning of personalized dentition digital information to solve the shortcomings of the existing technology. Summary of the invention

[0005] The purpose of the present invention is to avoid the shortcomings of the prior art and provide an electric tooth cleaning system based on personalized dentition digital information intelligent planning. The electric tooth cleaning system based on personalized dentition digital information intelligent planning can directly identify the fracture classification prediction results of fresh or old vertebral compression fractures based on CT images.

[0006] The above-mentioned purpose of the present invention is achieved by the following technical measures:

[0007] Provided is an electric teeth cleaning system based on intelligent planning of personalized dentition digital information, comprising:

[0008] Denture digital information acquisition module - obtains the dental information of the subject's mouth to obtain a three-dimensional model;

[0009] Visual acquisition module - collects the teeth cleaning data of the operation module in real time during teeth cleaning, and sends the teeth cleaning data to the processing module;

[0010] Operation module: perform teeth cleaning operation;

[0011] Processing module - generates a three-dimensional model based on the dental arch information in the mouth, then identifies the tooth condition and generates a target cleaning plan; makes real-time judgments based on the cleaning data of the visual acquisition module and the target cleaning plan, obtains the judgment results, and gives prompts; the target cleaning plan includes a target cleaning route, target cleaning parameters for cleaning key and difficult areas, and target cleaning parameters for non-key and difficult areas.

[0012] Preferably, the method for obtaining the target teeth cleaning plan of the processing module is performed by the following steps:

[0013] A1, a three-dimensional model generated according to the intraoral dentition information of the dentition digital information acquisition module;

[0014] A2, segment each tooth and the periodontal tissue disease area according to the three-dimensional model of A1;

[0015] A3. Get the dental arch line and cleaning route based on the centroid of each tooth;

[0016] A4. Determine whether there is tooth crowding based on the positional relationship between the tooth and the adjacent teeth according to the dental arch line. If there is tooth crowding, mark it as a tooth crowding area; the cleaning key and difficult area is at least one of a tooth crowding area or a periodontal tissue disease area;

[0017] A5. Obtain the target teeth cleaning plan according to the key and difficult cleaning areas and the target cleaning route.

[0018] Preferably, the target cleaning route is a cleaning route that takes the dental arch line as a reference line and covers the lingual side and labial side of each tooth back and forth.

[0019] Preferably, the target teeth cleaning parameter is at least one of a target pressure value, a target teeth cleaning time or a target circumscribed rectangular frame reduction value of the operation module.

[0020] Preferably, the above A2 is obtained by the following steps:

[0021] A2.1, segment each tooth and periodontal tissue diseased area in the 3D model of A1 through neural network segmentation;

[0022] A2.2. The segmented three-dimensional model is converted into RGB format data and detected using a neural network to obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area.

[0023] Preferably, the periodontal tissue disease area is at least one of a tartar area, a dental plaque area or a caries area.

[0024] Preferably, the teeth cleaning data are image data and current teeth cleaning parameters.

[0025] Preferably, the current teeth cleaning parameter is at least one of a reduction value of an external rectangular frame, a teeth cleaning time, or a pressure value of an operating module.

[0026] Preferably, the method for determining the brushing result of the processing module is performed by the following steps:

[0027] B1. Collecting tooth cleaning data in real time according to the visual acquisition module, and segmenting each tooth and periodontal tissue disease area in the image data of the tooth cleaning data in real time through a neural network;

[0028] B2. Convert the segmented 3D model into RGB format data and use a neural network to detect it, and obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area;

[0029] B3, aligning the circumscribed rectangular frame of the periodontal tissue disease area obtained in B2 with the circumscribed rectangular frame of the periodontal tissue disease area in the target tooth cleaning scheme;

[0030] B4, judging in real time whether the current cleaning parameters corresponding to the current position have reached the target cleaning parameters, if not, marking it as inadequate cleaning and prompting to enter B5; if yes, returning to B1;

[0031] B5. Determine the cumulative number of times the current position is not cleaned properly. When the cumulative number is less than or equal to the prompt setting value, return to B1; when the cumulative number is greater than the prompt setting value, a prompt is given to enter manual cleaning.

[0032] When the cleaning key and difficult area is a crowded tooth area, the target teeth cleaning parameter is a target teeth cleaning time, and the target teeth cleaning time is 1.1 to 2 times the target teeth cleaning time of a non-cleaning key and difficult area.

[0033] When the cleaning key and difficult area is a tooth decay area, the target tooth cleaning parameter is a target pressure value of the operation module, and the target pressure value is 1.1 to 2 times the target pressure value of the non-cleaning key and difficult area.

[0034] When the cleaning key and difficult area is a calculus area or a dental plaque area, the target cleaning parameter is a target circumscribed rectangular frame reduction value.

[0035] The neural network in B1 and A2.1 is a Unet neural network.

[0036] The neural network in B2 and A2.2 is the YoLov3 neural network.

[0037] Preferably, the above-mentioned visual acquisition module is a stereo camera composed of binocular or multi-eye cameras.

[0038] Preferably, the operating module is an oral irrigator, an electric dental flosser or an electric toothbrush.

[0039] Preferably, the operating module is also provided with a somatosensory gyroscope and a pressure sensor.

[0040] When the processing module is a Unity game engine, the teeth cleaning path of the operation module is set as a real-time clearance path, and the target cleaning route is set as a target clearance path. The processing module sends the data of the real-time clearance path and the data of the target clearance path to the display medium.

[0041] The electric tooth cleaning system based on personalized dentition digital information intelligent planning of the present invention is provided with a dentition digital information acquisition module, which obtains the dentition information of the subject's mouth to obtain a three-dimensional model; a visual acquisition module, which collects the tooth cleaning data of the operation module in real time during tooth cleaning, and sends the tooth cleaning data to the processing module; the operation module performs tooth cleaning operations; the processing module generates a three-dimensional model according to the dentition information of the teeth in the mouth, then identifies the tooth condition, and generates a target tooth cleaning plan; the tooth cleaning data of the visual acquisition module and the target tooth cleaning plan are judged in real time to obtain the judgment result and prompt; the target tooth cleaning plan includes a target cleaning route, target tooth cleaning parameters for cleaning key and difficult areas, and target tooth cleaning parameters for non-cleaning key and difficult areas. The electric tooth cleaning system based on personalized dentition digital information intelligent planning, by digitally collecting personal oral information, identifies the key and difficult areas of oral cleaning and performs personalized brushing planning. The present invention identifies the position of the operation module through the visual acquisition module, so as to correctly guide the operator to effectively master the brushing method and complete the cleaning process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention is further described with the aid of the accompanying drawings, but the contents in the accompanying drawings do not constitute any limitation to the present invention.

[0043] Figure 1 The dental arch wire for the teeth.

[0044] Figure 2 It is the circumscribed rectangular frame and centroid coordinates of the tooth decay area.

[0045] Figure 3 It is the circumscribed rectangular frame and centroid coordinates of the dental calculus area. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is further described in conjunction with the following embodiments.

[0047] Example 1

[0048] An electric teeth cleaning system based on intelligent planning of personalized dentition digital information, comprising:

[0049] Denture digital information acquisition module - obtains the dental information of the subject's mouth to obtain a three-dimensional model;

[0050] Visual acquisition module - collects the teeth cleaning data of the operation module in real time during teeth cleaning, and sends the teeth cleaning data to the processing module;

[0051] Operation module: perform teeth cleaning operation;

[0052] Processing module - generates a three-dimensional model based on the dental arch information in the mouth, then identifies the tooth condition and generates a target cleaning plan; makes real-time judgments based on the cleaning data of the visual acquisition module and the target cleaning plan, obtains the judgment results, and gives prompts; the target cleaning plan includes target cleaning routes, target cleaning parameters for cleaning key and difficult areas, and target cleaning parameters for non-key and difficult areas.

[0053] It should be noted that the dentition digital information collection module of the present invention collects the dentition information of the subject's mouth after a certain number of cleanings or regularly used, and updates the three-dimensional model.

[0054] The method for obtaining the target tooth cleaning scheme of the processing module of the present invention is performed by the following steps:

[0055] A1, a three-dimensional model generated based on the intraoral tooth dentition information of the dentition digital information acquisition module;

[0056] A2, segment each tooth and the periodontal tissue disease area according to the three-dimensional model of A1;

[0057] A3. Get the dental arch line and cleaning route based on the centroid of each tooth;

[0058] A4. Determine whether there is tooth crowding based on the positional relationship between the teeth and adjacent teeth according to the dental arch line. If there is tooth crowding, mark it as a tooth crowding area. The key and difficult area for cleaning is at least one of the tooth crowding area and the periodontal tissue disease area.

[0059] A5. Get the target cleaning plan based on the key and difficult areas of cleaning and the target cleaning route.

[0060] Specifically, A2 is obtained by the following steps:

[0061] A2.1, segment each tooth and periodontal tissue diseased area in the 3D model of A1 through neural network segmentation;

[0062] A2.2. The segmented three-dimensional model is converted into RGB format data and detected using a neural network to obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area.

[0063] The target cleaning route is a cleaning route that covers the lingual and labial sides of each tooth back and forth with the dental arch line as the reference line. The target cleaning parameter is at least one of the target pressure value, target cleaning time, or target circumscribed rectangular frame reduction value of the operation module. The periodontal tissue disease area is at least one of the tartar area, dental plaque area, or tooth decay area. The cleaning data is the image data and the current cleaning parameter. The current cleaning parameter is at least one of the circumscribed rectangular frame reduction value, the cleaning time, or the pressure value of the operation module.

[0064] It should be noted that the dental arch line of the present invention is fitted by the least square method to obtain a three-dimensional curve of the centroid of all teeth, such as Figure 1 The black dot in the figure is the centroid. The tooth crowding situation of the present invention determines the tooth crowding area according to the distance between the centroids of different teeth, and the determination of tooth crowding belongs to the common knowledge in the field and will not be repeated here.

[0065] The method for determining the toothbrushing result of the processing module of the present invention is performed by the following steps:

[0066] B1. Collecting tooth cleaning data in real time according to the visual acquisition module, and segmenting each tooth and periodontal tissue disease area in the image data of the tooth cleaning data in real time through the neural network;

[0067] B2. Convert the segmented 3D model into RGB format data and use a neural network to detect it, and obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area;

[0068] B3, aligning the circumscribed rectangular frame of the periodontal tissue disease area obtained in B2 with the circumscribed rectangular frame of the periodontal tissue disease area in the target tooth cleaning scheme;

[0069] B4, judging in real time whether the current cleaning parameters corresponding to the current position have reached the target cleaning parameters, if not, marking it as inadequate cleaning and prompting to enter B5; if yes, returning to B1;

[0070] B5. Determine the cumulative number of times the current position is not cleaned properly. When the cumulative number is less than or equal to the prompt setting value, return to B1; when the cumulative number is greater than the prompt setting value, a prompt is given to enter manual cleaning.

[0071] When the cleaning key and difficult area is the crowded tooth area, the target cleaning parameter is the target cleaning time, and the target cleaning time is 1.1 to 2 times the target cleaning time of the non-key and difficult cleaning area. When the cleaning key and difficult area is the caries area, the target cleaning parameter is the target pressure value of the operation module, and the target pressure value is 1.1 to 2 times the target pressure value of the non-key and difficult cleaning area. When the cleaning key and difficult area is the tartar area or the plaque area, the target cleaning parameter is the target circumscribed rectangular frame reduction value.

[0072] It should be noted that the target cleaning time for the non-cleaning difficult areas, the current pressure value for the non-cleaning difficult areas, and the target circumscribed rectangular frame reduction value of the present invention are determined according to actual conditions. For example, the target cleaning time for the non-cleaning difficult areas is 5s to 10s, the current pressure value for the non-cleaning difficult areas is 0.5Mpa, and the target circumscribed rectangular frame reduction value can be 50%, etc.

[0073] The neural network in B1 and A2.1 is the Unet neural network; the neural network in B2 and A2.2 is the YoLov3 neural network.

[0074] It should be noted that the Unet neural network and YoLov3 neural network of the present invention are both trained networks, and the loss function is the cross entropy loss function (Cross Entropy Loss), and the training end condition is that the verification set intersection over Union (IoU) is no longer improved, and the degree of transformation is less than (0.01). The training of the Unet neural network and the training of the YoLov3 neural network are both based on the collection of oral dentition information data of 2000 teeth in the early stage, and the oral experts manually outline and mark the teeth, tartar areas, dental plaque areas and caries areas, and then use the neural network for training and learning.

[0075] The YoLov3 neural network of the present invention detects target detection information of tartar, caries, and dental plaque on RGB format data. The YoLov3 neural network mainly detects whether the tooth surface in this area is intact by identifying the black / gray area in the image, thereby identifying the condition of the teeth. The length, width, and upper left corner coordinate information of the circumscribed rectangular frame of the tartar, dental plaque, or caries area are used to generate the centroid of the outer rectangle, which is recorded as Pos2D (x, y). The centroid coordinate Pos2D is the depth coordinate sampled from the depth image of the RGB binocular depth camera, recorded as the Z coordinate, thereby obtaining the three-dimensional coordinate Pos2D (Pos2D, Z) of the tartar area, such as Figure 2and Figure 3 , the black dot in the figure is the centroid.

[0076] The visual acquisition module of the present invention is a stereo camera composed of binocular or multi-cameras; the operating module is a water flosser, an electric dental flosser or an electric toothbrush; the operating module is also provided with a somatosensory gyroscope and a pressure sensor.

[0077] The electric tooth cleaning system based on personalized dentition digital information intelligent planning can identify the key and difficult areas of oral cleaning and carry out personalized tooth brushing planning by digitally collecting personal oral information. The present invention can correctly guide the operator to effectively master the tooth brushing method and complete the cleaning process by identifying the position of the operation module through the visual collection module.

[0078] Example 2

[0079] An electric teeth cleaning system based on intelligent planning of personalized dental arch digital information, other features are the same as those of Example 1, and also has the following features: when the processing module is the Unity game engine, the teeth cleaning path of the operation module is set as a real-time clearance path, and the target cleaning route is set as a target clearance path. The processing module sends the data of the real-time clearance path and the data of the target clearance path to a display medium.

[0080] It should be noted that the Unity game engine of the present invention can also set the target teeth cleaning parameters to pass the level. When the current teeth cleaning parameters reach the target teeth cleaning parameters, virtual rewards are generated, such as gold coins.

[0081] For example, the present invention constructs the target clearance path into a flying path similar to that of cartoon characters such as butterflies, and the shape of the operation module is constructed into the image of a fishing net. Through mimetic animation, it simulates and guides the physical toothbrush to complete teeth cleaning in the real space. At the same time, it also promotes and checks the cleaning effect through force feedback and specific perception of operation time.

[0082] The processing module of this embodiment sends the data of the real-time clearance path and the data of the target clearance path to the display medium, where the display medium can be a screen, a projection, VR / AR glasses, etc., so that the game process of the processing module can be presented to the operator in real time, guiding the operator to perform planning operations.

[0083] Compared with the first embodiment, the electric teeth cleaning system of the present embodiment has multiple visual scenes, thereby increasing the interest of the teeth cleaning process.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. An electric tooth cleaning system based on intelligent planning of personalized dentition digital information, characterized in that: The settings are: Denture digital information acquisition module - obtains the subject's intraoral dental information; Visual acquisition module - collects the cleaning data of the operation module in real time during teeth cleaning, and sends the cleaning data to the processing module; Operation module: perform teeth cleaning operation; Processing module: Generate a three-dimensional model based on the dental information in the mouth, then identify the tooth condition and generate a target cleaning plan; Make a real-time judgment based on the teeth cleaning data of the visual acquisition module and the target teeth cleaning plan, obtain the judgment result, and give a prompt; The target tooth cleaning scheme includes a target cleaning route, target tooth cleaning parameters for cleaning key and difficult areas, and target tooth cleaning parameters for non-key and difficult areas.

2. The electric tooth cleaning system based on intelligent planning of personalized dentition digital information according to claim 1, characterized in that: The method for obtaining the target tooth cleaning plan is carried out by the following steps: A1, a three-dimensional model generated according to the intraoral dentition information of the dentition digital information acquisition module; A2, segment each tooth and the periodontal tissue disease area according to the three-dimensional model of A1; A3. Get the dental arch line and cleaning route based on the centroid of each tooth; A4. Determine whether there is dental crowding by judging the positional relationship between the teeth and adjacent teeth based on the dental arch line. If there is dental crowding, mark it as the dental crowding area. The cleaning key and difficult area is at least one of a crowded tooth area or a periodontal tissue disease area; A5. Obtain the target teeth cleaning plan according to the key and difficult cleaning areas and the target cleaning route.

3. The electric tooth cleaning system based on intelligent planning of personalized dentition digital information according to claim 2, characterized in that: The A2 is obtained by the following steps: A2.1, segment each tooth and periodontal tissue diseased area in the 3D model of A1 through neural network segmentation; A2.

2. The segmented three-dimensional model is converted into RGB format data and detected using a neural network to obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area.

4. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to claim 3, characterized in that: The method for determining the toothbrushing result of the processing module is performed by the following steps: B1. Collecting tooth cleaning data in real time according to the visual acquisition module, and segmenting each tooth and periodontal tissue diseased area in the image data of the tooth cleaning data in real time through a neural network; B2. Convert the segmented 3D model into RGB format data in real time and use a neural network to detect it, so as to obtain the centroid of each tooth, the circumscribed rectangular frame of the periodontal tissue disease area, the centroid of the circumscribed rectangular frame, and the centroid of the periodontal tissue disease area; B3, aligning the circumscribed rectangular frame of the periodontal tissue disease area obtained in B2 with the circumscribed rectangular frame of the periodontal tissue disease area in the target tooth cleaning scheme; B4, judging in real time whether the current cleaning parameters corresponding to the current position have reached the target cleaning parameters, if not, marking it as inadequate cleaning and prompting to enter B5; if yes, returning to B1; B5. Determine the cumulative number of times the current position is not cleaned properly. When the cumulative number is less than or equal to the prompt setting value, return to B1; when the cumulative number is greater than the prompt setting value, a prompt is given to enter manual cleaning.

5. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to claim 4, characterized in that: The target cleaning route is a cleaning route that covers the lingual and labial sides of each tooth back and forth with the dental arch line as the reference line; The target teeth cleaning parameter is at least one of a target pressure value, a target teeth cleaning time, or a target circumscribed rectangular frame reduction value of the operation module.

6. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to claim 5, characterized in that: The periodontal tissue disease area is at least one of a calculus area, a dental plaque area or a caries area; The teeth cleaning data are image data and current teeth cleaning parameters; The current teeth cleaning parameter is at least one of a reduction value of an external rectangular frame, a teeth cleaning time, or a pressure value of an operation module.

7. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to claim 6, characterized in that: When the cleaning key and difficult area is a crowded tooth area, the target cleaning parameter is a target cleaning time, and the target cleaning time is 1.1 to 2 times the target cleaning time of a non-cleaning key and difficult area; When the cleaning key and difficult area is a tooth decay area, the target tooth cleaning parameter is a target pressure value of the operation module, and the target pressure value is 1.1 to 2 times the target pressure value of the non-cleaning key and difficult area; When the cleaning key and difficult area is a calculus area or a dental plaque area, the target cleaning parameter is a target circumscribed rectangular frame reduction value.

8. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to claim 7, characterized in that: The neural network in B1 and A2.1 is a Unet neural network; The neural network in B2 and A2.2 is the YoLov3 neural network.

9. The electric tooth cleaning system based on intelligent planning of personalized dentition digital information according to any one of claims 1 to 8, characterized in that: The visual acquisition module is a stereo camera composed of binocular or multi-cameras; The operating module is a water flosser, an electric dental flosser or an electric toothbrush; The operation module is also provided with a somatosensory gyroscope and a pressure sensor.

10. The electric teeth cleaning system based on intelligent planning of personalized dentition digital information according to any one of claims 1 to 8, characterized in that: When the processing module is the Unity game engine, the teeth cleaning path of the operation module is set as the real-time clearance path, the target cleaning route is set as the target clearance path, and the processing module sends the data of the real-time clearance path and the data of the target clearance path to the display medium.

Images