A method and device for generating a dentition alveolar bone model

Abstract

Embodiments of the present application relate to a kind of dentition alveolar bone model generation method and device, the method comprises the following steps: obtaining the scanning data of target oral cavity;Using the scanning data, three-dimensional reconstruction is carried out to obtain reconstruction data;The reconstruction data is segmented and marked to obtain the three-dimensional spatial information of different tissues;The corresponding ROI region in the three-dimensional spatial information is internally filled, closed and data smoothed, and STL format data is exported;The STL format data is 3D printing.The method provided in the embodiments of the present application generates model original cone beam CT data directly from target oral cavity, so its dentition and alveolar bone morphology truly reflects the actual situation of individual dentition and alveolar bone;And the three-dimensional segmentation of dentition and alveolar bone data is carried out by segmentation network, so that the accuracy of the generated model is higher.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of medical data processing, in particular to a method and device for generating a dental alveolar bone model. BACKGROUND

[0002] Absorption of alveolar bone is one of the main features of periodontitis. After suffering from periodontitis for several years, the morphology of the remaining alveolar bone of different patients and different teeth varies greatly after years of alveolar bone absorption, and the morphology of some types of alveolar bone is not conducive to the control of periodontal inflammation and the healing of periodontal tissue. In clinical practice, oral physicians often use periodontal surgery to correct the morphology of the alveolar bone to promote the healing of the periodontal tissue and the control of periodontal inflammation. However, different types of alveolar bone morphology are suitable for different types of periodontal surgery. Therefore, oral physicians often need to master the morphology of the remaining alveolar bone through cone beam CT images when formulating individualized surgical plans. However, using a computer to read cone beam CT images to obtain three-dimensional structural information of the target region requires high-level software operation ability and spatial imagination of the physician. How to more visually obtain the three-dimensional structure of the dentition and the alveolar bone has become a problem to be solved. SUMMARY

[0003] Based on the above situation of the prior art, the purpose of the embodiment of the present application is to provide a method and device for generating a dental alveolar bone model, which generates a model according to the actual morphology of an individual's oral cavity to more visually display the anatomical morphology of the individual's dentition and alveolar bone.

[0004] To achieve the above-mentioned purpose, according to one aspect of the present application, a method for generating a dental alveolar bone model is provided, comprising the steps of:

[0005] obtaining scanning data of a target oral cavity;

[0006] using the scanning data to perform three-dimensional reconstruction to obtain reconstruction data;

[0007] performing tissue segmentation and labeling on the reconstruction data to obtain three-dimensional spatial information of different tissues;

[0008] performing internal filling, closing and data smoothing on the part corresponding to the ROI region in the reconstruction data and the three-dimensional spatial information, and exporting STL format data;

[0009] performing 3D printing on the STL format data to generate a 3D printed model of the teeth and the alveolar bone;

[0010] performing splicing and assembling on the 3D model of the printed teeth and alveolar bone to form a dental alveolar bone model matching the target oral cavity.

[0011] Further, the ROI region includes the region where the teeth and the alveolar bone are located.

[0012] Further, the cone beam CT is used to obtain the scanning data of the target oral cavity.

[0013] Further, the segmentation network is used for tissue segmentation.

[0014] The segmentation network is established according to the following steps:

[0015] An original segmentation network is constructed, which includes an encoding module, a feature extraction module, a first decoding module, a second decoding module and a comprehensive decoding module.

[0016] The original segmentation network is trained by using an oral cavity data set to obtain the segmentation network.

[0017] Further, the oral cavity data set is subjected to mask processing or annotation processing before being used for training of the original segmentation network.

[0018] Further, the annotation processing includes:

[0019] The tooth region in the original oral cavity data set is annotated as 1X, wherein X represents the serial number of each tooth, and the region outside the tooth is annotated as 0.

[0020] The alveolar bone region in the original oral cavity data set is annotated as 2Y, wherein Y represents the serial number of the alveolar bone region, and the region outside the alveolar bone is annotated as 0.

[0021] The serial numbers X and Y correspond to each other, and the correspondence relationship is consistent with the correspondence relationship of the identified tooth and alveolar bone region.

[0022] Further, the encoding module is used for encoding the input reconstructed data to obtain a semantic feature map.

[0023] Further, the feature extraction module is used for processing each semantic feature map obtained by the encoding module to obtain semantic feature maps of different resolutions.

[0024] Further, the first decoding module is used for obtaining a tooth prediction probability map according to the semantic feature map.

[0025] The second decoding module is used for obtaining an alveolar bone prediction probability map according to the semantic feature map.

[0026] The comprehensive decoding module is used for comprehensively decoding the tooth prediction probability map and the alveolar bone prediction probability map to obtain an oral cavity segmentation probability map.

[0027] According to another aspect of the present application, a dental alveolar bone model generation device is provided, comprising a data acquisition module, a data reconstruction module, a segmentation and labeling module, an STL data export module, a 3D printing module and a splicing and assembling module, wherein,

[0028] The data acquisition module acquires scanning data of a target oral cavity.

[0029] The data reconstruction module performs three-dimensional reconstruction using the scanning data to obtain reconstruction data.

[0030] The segmentation and labeling module performs tissue segmentation and labeling on the reconstruction data to obtain three-dimensional spatial information of different tissues.

[0031] The STL data export module internally fills, closes and data smooths the part corresponding to the ROI region in the reconstruction data and three-dimensional spatial information, and exports STL format data.

[0032] The 3D printing module performs 3D printing on the STL format data.

[0033] The splicing and assembling module splices and assembles the printed 3D model of teeth and alveolar bone to form a dental alveolar bone model matching the target oral cavity.

[0034] In summary, the present application provides a dental alveolar bone model generation method and device, which comprises the following steps: acquiring scanning data of a target oral cavity; performing three-dimensional reconstruction using the scanning data to obtain reconstruction data; performing tissue segmentation and labeling on the reconstruction data to obtain three-dimensional spatial information of different tissues; internally filling, closing and data smoothing the corresponding ROI region in the three-dimensional spatial information, and exporting STL format data; and performing 3D printing on the STL format data.

[0035] The technical solution provided by the present application has the following beneficial technical effects:

[0036] (1) The model generated by the method provided by the present application directly takes the original cone beam CT data from the target oral cavity, so the morphology of the dental arch and alveolar bone truly reflects the actual situation of the individual dental arch and alveolar bone.

[0037] (2) The method provided by the present application generates a dental alveolar bone model according to the scanning data of the target oral cavity, which has a more intuitive effect when used, and solves the problem of the prior art that a computer needs to read cone beam CT images to obtain three-dimensional structure information of the target region, thereby requiring high-level software operation ability and spatial imagination.

[0038] (3) The model generation method provided by the embodiment of the present application has good segmentation effect on the dental arch and alveolar bone data through the segmentation network, and can more accurately reflect the respective positions of the teeth and alveolar bone in the image data, so that the generated model has higher accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of one or more embodiments of the present application or the prior art, the drawings needed to be used in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only one or more embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0040] Figure 1 is a flow chart of the dental arch and alveolar bone model generation method of the embodiment of the present application;

[0041] Figure 2 is a block diagram of the dental arch and alveolar bone model generation device of the embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the objects, technical solutions and advantages of the present application clearer, further detailed description will be made below in combination with specific embodiments and with reference to the drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present application. In addition, in the following description, the description of known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present application.

[0043] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present application should be understood as the usual meaning understood by those skilled in the art to which the present application belongs. The terms "first", "second" and similar terms used in one or more embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. The terms "include" or "contain" and similar terms mean that the elements or objects before the term cover the elements or objects listed after the term and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connect" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0044] The technical solutions of the present application will be described in detail below in combination with the drawings. According to an optional embodiment of the present application, a dental arch and alveolar bone model generation method is provided, Figure 1A flowchart of the generation method is shown in FIG. 1, which includes the following steps:

[0045] S1, obtaining scan data of a target oral cavity. In this step, cone beam CT can be used to obtain scan data of the target oral cavity, and cone beam CT image data of the dentition and alveolar bone in the target oral cavity can be obtained. According to some embodiments, multiple sets of scan data can be obtained from multiple angles with different parameters to obtain more comprehensive information. The embodiments of the present application use cone beam CT data to directly extract three-dimensional data from the target oral cavity, so that the morphology of the dentition and alveolar bone truly reflects the actual situation of the dentition and alveolar bone in the target oral cavity, which helps to generate a model with higher accuracy.

[0046] According to some embodiments, the obtained scan data can also be corrected in three-dimensional space. Due to different CT shooting angles, the imaging effect can be different. By making rigid changes to the three-dimensional data, such as but not limited to rotating and translating the image, and then using a registration method to realize the same spatial coordinate system for homologous markers or marker regions.

[0047] S2, using the scan data to perform three-dimensional reconstruction to obtain reconstruction data. Three-dimensional reconstruction software can be used to set the corresponding threshold value as needed, or the Boolean operation function and model separation function can be used to realize three-dimensional reconstruction of the target scan data.

[0048] S3, organizing, segmenting and labeling the reconstruction data to obtain three-dimensional spatial information of different tissues. In the embodiments of the present application, the three-dimensional reconstruction data is segmented based on a convolutional neural network, and is identified and labeled. The segmentation method based on the convolutional neural network can include three processes: (1) preparing a training set, (2) training a convolutional neural network, and (3) testing the trained neural network. The embodiments of the present application use an improved U-net network to segment the tissues. The U-net neural network belongs to a convolutional neural network, and its network structure resembles the letter U, hence the name U-net. The entire network is usually composed of two parts, encoding layer and decoding layer, and its outstanding advantage is that it can realize the segmentation of images of any size. In the present embodiment, the segmentation network can be established according to the following steps:

[0049] S3.1 Constructing an original segmentation network, the original segmentation network includes an encoding module, a feature extraction module, a first decoding module, a second decoding module, and a comprehensive decoding module. The encoding module encodes the input reconstructed data to obtain semantic feature maps; the feature extraction module processes each semantic feature map obtained by the encoding module to obtain semantic feature maps of different resolutions; the first decoding module obtains a tooth prediction probability map based on the semantic feature maps; the second decoding module obtains an alveolar bone prediction probability map based on the semantic feature maps; and the comprehensive decoding module performs comprehensive decoding on the tooth prediction probability map and the alveolar bone prediction probability map to obtain an oral cavity segmentation probability map.

[0050] S3.2. Train the original segmentation network using the oral cavity dataset to obtain the segmentation network. Specifically, the oral cavity dataset is processed by masking or annotation before being used to train the original segmentation network. The original oral cavity dataset is processed by manual annotation, labeling the tooth regions as 1X, where X represents the sequence number of each tooth, such as 1A, 1B, 1C… and labeling the regions outside the teeth as 0; and labeling the alveolar bone regions as 2Y, where Y represents the sequence number of the alveolar bone region, and this sequence number corresponds to the sequence number of its corresponding tooth, such as 2a, 2b, 2c… and labeling the regions outside the alveolar bone as 0. Alternatively, the original oral cavity dataset can be processed by masking. After manual annotation, the original oral cavity dataset is masked, resulting in each image in the dataset having a mask for each tooth region and a mask for the corresponding alveolar bone region.

[0051] S4. The Region of Interest (ROI) corresponding to the three-dimensional spatial information is filled, closed, and smoothed to export STL format data. In this embodiment, the ROI includes the areas containing teeth and alveolar bone. Alternatively, the exported STL format data can be meshed, redundant sharp corner features removed, and smoothed to eliminate unwanted undulations. Surface fitting is then performed using a quadrilateral point layout principle to obtain STL format data with better fitting results. During filling, the slice layer thickness can be determined based on the model's minimum wall thickness, and slices can be obtained. The corresponding inner contour is obtained sequentially from top to bottom based on the input contour of each slice layer, and filling and closing are then performed based on this inner contour.

[0052] S5. 3D print the STL format data to obtain 3D printed models of each tooth and alveolar bone. The STL format data of teeth and alveolar bone obtained in the preceding steps are cropped and denoised before being 3D printed to obtain their respective 3D printed models.

[0053] S6. Assemble the printed 3D-printed tooth and alveolar bone models to form a dentition alveolar bone model that matches the target oral cavity. The dentition alveolar bone model obtained using the method provided in this embodiment can realistically reflect the actual situation of an individual's dentition and alveolar bone, is highly targeted, and has high accuracy. It solves the problem in the prior art that it requires a high level of software operation skills and spatial imagination to obtain the three-dimensional structural information of the target area by reading cone-beam CT images using a computer.

[0054] According to another optional embodiment of the present invention, an apparatus for generating a dental alveolar bone model is provided. Figure 2 The diagram shows a block diagram of a device for generating a dental alveolar bone model according to an embodiment of the present invention. The device includes a data acquisition module, a data reconstruction module, a segmentation and marking module, an STL data export module, and a 3D printing module.

[0055] The data acquisition module acquires scan data of the target oral cavity;

[0056] The data reconstruction module uses the scanned data to perform three-dimensional reconstruction to obtain reconstructed data;

[0057] The segmentation and labeling module organizes, segments, and labels the reconstructed data to obtain three-dimensional spatial information of different tissues;

[0058] The STL data export module performs internal filling, closure, and data smoothing on the segmented three-dimensional data of teeth and alveolar bone, and exports the data in STL format.

[0059] The 3D printing module performs 3D printing on the STL format data.

[0060] According to some embodiments, the device may also include a splicing and assembly module that splices and assembles the printed teeth and alveolar bone 3D printed models to form a dental arch and alveolar bone model that matches the target oral cavity.

[0061] The specific processes by which each module in this embodiment of the present invention implements its function are the same as the steps involved in the method in the above embodiments of the present invention, and will not be repeated here.

[0062] In summary, the embodiments of the present invention relate to a method and apparatus for generating a dental alveolar bone model. The method includes the following steps: acquiring scan data of a target oral cavity; using the scan data to perform three-dimensional reconstruction to obtain reconstructed data; performing tissue segmentation and labeling on the reconstructed data to obtain three-dimensional spatial information of different tissues; performing internal filling, closure, and data smoothing on the corresponding ROI regions in the three-dimensional spatial information to export STL format data; and performing 3D printing on the STL format data. The model generated using the method provided in this embodiment of the invention uses original cone-beam CT data directly taken from the target oral cavity. Therefore, the morphology of the dentition and alveolar bone truly reflects the actual situation of the individual's dentition and alveolar bone. The method provided in this embodiment of the invention generates a dentition and alveolar bone model based on the scan data of the target oral cavity, which is more intuitive to use. It solves the problem in the prior art that it is necessary to use a computer to read cone-beam CT images to obtain the three-dimensional structural information of the target area, which requires a high level of software operation skills and spatial imagination. In the model generation method provided in this embodiment of the invention, the dentition and alveolar bone data are three-dimensionally segmented through a segmentation network. The segmentation effect is good and can more accurately reflect the position of the teeth and alveolar bone in the image data, making the generated model more accurate.

[0063] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, technical features of the above embodiments or different embodiments can also be combined, steps can be implemented in any order, and many other variations exist regarding different aspects of the embodiments of the invention as described above, which are not provided in the details for the sake of brevity. The embodiments of the invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of the invention should be included within the protection scope of the invention.

Claims

1. A method for generating a dental alveolar bone model, characterized in that, Including the following steps: Acquire scan data of the target oral cavity; Using the scanned data, three-dimensional reconstruction is performed to obtain reconstructed data; The reconstructed data is segmented and labeled to obtain three-dimensional spatial information of different tissues; The reconstructed data and the ROI region in the three-dimensional spatial information are filled, closed and smoothed internally, and then exported as STL format data. The STL format data is 3D printed to generate 3D printed models of teeth and alveolar bone. The 3D models of printed teeth and alveolar bone are spliced ​​and assembled to form a dental arch and alveolar bone model that matches the target oral cavity; The ROI region includes the area containing teeth and alveolar bone; The steps of performing tissue segmentation and labeling on the reconstructed data to obtain three-dimensional spatial information of different tissues specifically include: The oral cavity dataset is subjected to masking or annotation processing for training the segmentation network. The annotation processing includes: labeling the tooth regions in the original oral cavity dataset as 1X, where X identifies the sequence number of each tooth and the regions outside the teeth are labeled as 0; labeling the alveolar bone regions in the original oral cavity dataset as 2Y, where Y identifies the sequence number of the alveolar bone region and the regions outside the alveolar bone are labeled as 0; the sequence numbers X and Y correspond to each other, and their correspondence is consistent with the correspondence between the labeled teeth and alveolar bone regions. Then, a segmentation network is used for organizational segmentation.

2. The method according to claim 1, characterized in that, Cone-beam CT was used to acquire scan data of the target oral cavity.

3. The method according to claim 1 or 2, characterized in that, The segmentation network is established according to the following steps: Construct an original segmentation network, which includes an encoding module, a feature extraction module, a first decoding module, a second decoding module, and a comprehensive decoding module; The original segmentation network was trained using the oral cavity dataset to obtain the segmentation network.

4. The method according to claim 3, characterized in that, The encoding module is used to encode the input reconstructed data to obtain a semantic feature map.

5. The method according to claim 4, characterized in that, The feature extraction module is used to process each semantic feature map obtained by the encoding module to obtain semantic feature maps of different resolutions.

6. The method according to claim 5, characterized in that, The first decoding module is used to obtain a tooth prediction probability map based on the semantic feature map; The second decoding module is used to obtain a predicted probability map of alveolar bone based on the semantic feature map; The integrated decoding module is used to perform integrated decoding of the tooth prediction probability map and the alveolar bone prediction probability map to obtain the oral cavity segmentation probability map.

7. A device for generating a dental alveolar bone model, characterized in that, It includes a data acquisition module, a data reconstruction module, a segmentation and marking module, an STL data export module, a 3D printing module, and a splicing and assembly module; among them, The data acquisition module acquires scan data of the target oral cavity; The data reconstruction module uses the scanned data to perform three-dimensional reconstruction to obtain reconstructed data; The segmentation and labeling module performs organizational segmentation and labeling on the reconstructed data to obtain three-dimensional spatial information of different tissues; The STL data export module performs internal filling, closure and data smoothing on the part of the ROI region corresponding to the reconstructed data and the three-dimensional spatial information, and exports STL format data. The 3D printing module performs 3D printing on the STL format data; The splicing and assembly module splices and assembles the printed 3D models of teeth and alveolar bone to form a dental arch and alveolar bone model that matches the target oral cavity. The ROI region includes the area containing teeth and alveolar bone; Furthermore, the segmentation marker module is also used for, The oral cavity dataset is masked or labeled for training the segmentation network. The labeling process includes: labeling the tooth regions in the original oral cavity dataset as 1X, where X identifies the sequence number of each tooth and the regions outside the teeth are labeled as 0; labeling the alveolar bone regions in the original oral cavity dataset as 2Y, where Y identifies the sequence number of the alveolar bone region and the regions outside the alveolar bone are labeled as 0; the sequence numbers X and Y correspond to each other, and their correspondence is consistent with the correspondence between the labeled teeth and alveolar bone regions. Then, a segmentation network is used for organizational segmentation.

Images