Root-bone feature determination method, display method, apparatus, device, and storage medium

Abstract

The present application discloses a root-bone feature determination method, a display method, an apparatus, a device, and a storage medium. The method comprises: obtaining a root-bone scan image; on the basis of the root-bone scan image, performing jawbone segmentation to determine a jawbone segmentation image; determining cortical bone feature information on the basis of the jawbone segmentation image; inputting the root-bone scan image into a localization model to obtain first localization information of N teeth, the first localization information being used for determining position information of the teeth in the root-bone scan image; identifying center information of different teeth in the root-bone scan image to obtain second localization information of M teeth, wherein N and M are positive integers greater than 1; obtaining a tooth segmentation result on the basis of the first localization information of the N teeth and the second localization information of the M teeth; and determining root-bone feature information on the basis of the cortical bone feature information and the tooth segmentation result. The method provided in the present application can improve the accuracy of root-bone feature determination.

Description

Methods, display methods, devices, equipment and storage media for determining calcaneal features

[0001] This application claims priority to Chinese Patent Application No. 202411998923.0, filed on December 31, 2024, entitled "Method, Display Method, Apparatus, Device and Storage Medium for Determining Jawbone Features", and Chinese Patent Application No. 202411998951.2, filed on December 31, 2024, entitled "Method, Display Method, Apparatus, Device and Storage Medium for Tooth Segmentation", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of medical image processing technology, and in particular to a method for determining radicular bone features, a method for displaying radicular bone features, an apparatus, a device, and a storage medium. Background Technology

[0003] With societal development and rising demands for quality of life, oral hygiene and aesthetic oral health maintenance have gradually become key concerns. During related treatments, root bone characteristic information serves as crucial intermediate data, revealing the user's current dental and jaw condition and assisting the operator in determining the treatment plan.

[0004] When determining root bone features, it is necessary to determine the state information of the teeth and / or jawbone structures. Existing technologies include methods for root bone feature analysis based on deep neural networks. However, when using deep neural networks for tooth differentiation, it relies on a large amount of training data, making it difficult to achieve high segmentation accuracy and tooth detection rate. Furthermore, when using deep neural networks for jawbone structure analysis, because they are applied to the entire jawbone, it is difficult to determine the specific structural features of the jawbone. Summary of the Invention

[0005] One of the purposes of this application is to provide a method for determining calcaneal bone characteristics, so as to solve the technical problems of poor accuracy of calcaneal bone characteristic analysis, analysis results that do not conform to reality, and inability to be used as an adjunct to treatment in the prior art.

[0006] One embodiment of this application provides a method for determining radicular bone features, comprising: obtaining a radicular bone scan image; performing jawbone segmentation based on the radicular bone scan image to determine a segmented jawbone image; determining cortical bone feature information based on the segmented jawbone image; inputting the radicular bone scan image into a positioning model to obtain first positioning information for N teeth, the first positioning information being used to determine the position information of the teeth in the radicular bone scan image; identifying the center information of different teeth in the radicular bone scan image to obtain second positioning information for M teeth; N and M being positive integers greater than 1; obtaining a tooth segmentation result based on the first positioning information of the N teeth and the second positioning information of the M teeth; and determining radicular bone feature information based on the cortical bone feature information and the tooth segmentation result.

[0007] One embodiment of this application provides a method for determining root bone features, comprising: obtaining a root bone scan image; inputting the root bone scan image into a localization model to obtain first localization information for N teeth, wherein the first localization information is used to determine the position information of the teeth in the root bone scan image; identifying the center information of different teeth in the root bone scan image to obtain second localization information for M teeth; wherein N and M are positive integers greater than 1; determining regions of interest for tooth segmentation based on the first localization information of N teeth and the second localization information of M teeth; and inputting the image corresponding to the segmented regions of interest into a segmentation model to obtain tooth segmentation results.

[0008] One embodiment of this application provides a method for determining radicular bone features, comprising: obtaining a radicular bone scan image; performing jawbone segmentation based on the radicular bone scan image to determine a jawbone segmentation image; using a neural network, taking both the radicular bone scan image and the jawbone segmentation image representing the same jawbone location as input, to determine a jawbone feature image and a cancellous bone feature image; comparing the jawbone feature image and the cancellous bone feature image, and determining a cortical bone feature image based on the differences; and determining and outputting cortical bone feature information based on the cortical bone feature image.

[0009] One embodiment of this application provides a method for displaying radicular bone features. The method, executing any technical solution of this application, includes at least one of the following: presenting a jawbone feature image in a graphical user interface, wherein the jawbone feature image is determined by a neural network based on both a radicular bone scan image representing the same jawbone location and a segmented jawbone image as input, and the jawbone feature image is used to determine a cortical bone feature image based on its difference from a cancellous bone feature image; presenting a cancellous bone feature image in a graphical user interface, wherein the cancellous bone feature image is determined by a neural network based on both a radicular bone scan image representing the same jawbone location and a segmented jawbone image as input, and the cancellous bone feature image is used to determine a cortical bone feature image based on its difference from a cancellous bone feature image. The difference between the bone feature image and the jawbone feature image is used to determine the cortical bone feature image; the cortical bone feature image is presented in the graphical user interface. The cortical bone feature image is determined by comparing the differences between the jawbone feature image and the cancellous bone feature image. The jawbone feature image and the cancellous bone feature image are determined by a neural network based on both a root bone scan image representing the same jawbone location and a jawbone segmentation image as input; the tooth segmentation result is presented in the graphical user interface; a root bone scan image is presented in the graphical user interface, which is superimposed with the tooth segmentation result; a three-dimensional model of the tooth segmented tooth is presented in the graphical user interface, which is generated by three-dimensional reconstruction based on the tooth segmentation result.

[0010] The method for determining root bone characteristics provided in this application can achieve tooth segmentation, or determine cortical bone characteristics, or both tooth segmentation and determination of cortical bone characteristics. It can achieve accurate analysis of root bone characteristics, and the analysis results are consistent with reality. The analysis results can be used as intermediate data to effectively assist in treatment.

[0011] On the one hand, using the segmented jawbone image obtained by segmenting the radicular bone scan image and the radicular bone scan image itself as dual-channel inputs to the neural network can enhance the feature extraction capability of the neural network and more accurately determine the overall jawbone features and cancellous bone features. The method provided in this application also determines cortical bone features based on the differences between cancellous bone features and overall jawbone features. Compared with existing technologies, the cortical bone feature extraction is more accurate and computationally more efficient.

[0012] On the one hand, by determining the region of interest (ROI) for tooth segmentation based on two types of tooth localization information, the accuracy of tooth localization and detection is improved. When using the obtained ROI for tooth segmentation, complete and accurate tooth segmentation can be achieved, avoiding the omission of special cases such as impacted teeth and supernumerary teeth. One type of localization information is used to determine the position of the tooth in the root bone scan image, enabling the tooth segmentation result to have the accuracy of single-tooth segmentation for conventional teeth. The other type of localization information is obtained by identifying the center information of the root bone scan image, ensuring that the tooth segmentation result includes all teeth of the subject without omission.

[0013] One embodiment of this application provides an electronic device, including a processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other through the communication bus; the memory is used to store application programs; and the processor is used to implement the steps of any method of this application when executing the application programs stored in the memory.

[0014] One embodiment of this application provides a storage medium on which an application program is stored, wherein when the application program is executed, it implements the steps of any method of this application. Attached Figure Description

[0015] Figure 1 is a schematic diagram of the steps of the method for determining the characteristics of the root bone in this application.

[0016] Figure 2 is a schematic diagram of the jawbone features in this application.

[0017] Figure 3 is a schematic diagram of the steps of a method for determining root bone characteristics in one embodiment of this application.

[0018] Figure 4 is a schematic diagram of a pedicle scan image and sequence in one embodiment of this application.

[0019] Figure 5 is a schematic diagram of an embodiment of the jawbone segmentation process in one implementation of this application.

[0020] Figure 6 is a schematic diagram of another embodiment of the jawbone segmentation process in one embodiment of this application.

[0021] Figure 7 is a schematic diagram of an embodiment of the neural network processing procedure in one implementation of this application.

[0022] Figure 8 is a schematic diagram of another embodiment of the neural network processing procedure in one implementation of this application.

[0023] Figure 9 is a schematic diagram of an embodiment of the comparison difference process in one implementation of this application.

[0024] Figure 10 is a schematic diagram of another embodiment of the comparison difference process in one implementation of this application.

[0025] Figure 11 is a schematic diagram of a first embodiment of establishing a three-dimensional model according to one implementation of this application.

[0026] Figure 12 is a schematic diagram of a second embodiment of establishing a three-dimensional model in one implementation of this application.

[0027] Figure 13 is a schematic diagram of a third embodiment of establishing a three-dimensional model in one implementation of this application.

[0028] Figure 14 is a schematic diagram of a fourth embodiment of establishing a three-dimensional model in one implementation of this application.

[0029] Figure 15 is a schematic diagram of the cutting process in one embodiment of this application.

[0030] Figure 16 is a schematic diagram of the cutting process in one embodiment of this application.

[0031] Figure 17 is a three-dimensional model of the teeth in this application.

[0032] Figure 18 is a schematic diagram of the steps of a method for determining root bone characteristics in one embodiment of this application.

[0033] Figure 19 is a schematic diagram of the process of obtaining the segmentation result of N teeth in one embodiment of this application.

[0034] Figure 20 is a schematic diagram of the process of obtaining the center reconstruction results of M teeth in one embodiment of this application.

[0035] Figure 21 is a schematic diagram of the process for determining differential teeth in one embodiment of this application.

[0036] Figure 22 is a schematic diagram of the process of generating differential teeth in one embodiment of this application.

[0037] Figure 23 is a schematic diagram of the steps of a method for displaying radicular features according to this application.

[0038] Figure 24 is a schematic diagram of the steps of another method for displaying pedicle features according to this application.

[0039] Figure 25 is a schematic diagram of the electronic device in this application.

[0040] Figure 26 is a schematic diagram of the structure of the device for determining the characteristics of the root bone in this application. Detailed Implementation

[0041] The present application will now be described in detail with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present application, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are included within the scope of protection of this application.

[0042] It should be noted that the term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0043] The terms "first," "second," and "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. There is no necessary correlation between the terms "first," etc.; the inclusion of a feature related to "second" in one embodiment does not necessarily mean that the embodiment must include a feature related to "first."

[0044] This application provides a method for determining the characteristics of the calcaneus, as shown in Figure 1.

[0045] Methods for determining calcaneal features may include at least one of the following steps.

[0046] Step S01: Obtain a calcaneal scan image.

[0047] Step S02: Based on the root bone scan image, perform jawbone segmentation to determine the jawbone segmentation image; based on the jawbone segmentation image, determine the cortical bone feature information.

[0048] Step S03: Input the root bone scan image into the localization model to obtain the first localization information of N teeth. The first localization information is used to determine the position information of the teeth in the root bone scan image. Identify the center information of different teeth in the root bone scan image to obtain the second localization information of M teeth. N and M are positive integers greater than 1. Based on the first localization information of N teeth and the second localization information of M teeth, the tooth segmentation result is obtained.

[0049] Step S04: Determine the root bone feature information based on the cortical bone feature information and the tooth segmentation results.

[0050] In this way, the characteristics of the root bone can be accurately determined, and the characteristics of the cortical bone and the segmentation of the teeth can be obtained, which can be used as intermediate information to assist or participate in the treatment.

[0051] In this application, the root bone feature information can be jawbone features, tooth segmentation results, or both jawbone features and tooth segmentation results.

[0052] Jawbone features may include cortical bone features, cancellous bone features, or information generated during the determination of cortical bone features, or information generated during the determination of cancellous bone features.

[0053] In one embodiment, step S02 may specifically include at least one of the following steps.

[0054] Step S021: Using a neural network, the root bone scan image and the jawbone segmentation image representing the same jawbone location are used as inputs to determine the jawbone feature image and the cancellous bone feature image.

[0055] Step S022: Compare the jawbone feature image and the cancellous bone feature image, and determine the cortical bone feature image based on the differences.

[0056] Step S023: Determine the bone cortex feature information based on the bone cortex feature image.

[0057] When identifying jawbone features within the radicular bone features, overall segmentation of the target image often leads to poor accuracy in distinguishing detailed features, resulting in jawbone features that do not accurately reflect the actual condition of the subject and are unusable for medical assistance. Therefore, one embodiment provides a method for determining radicular bone features. First, preliminary jawbone segmentation is performed on the radicular bone scan image. The preliminary segmentation result and the radicular bone scan image itself are used as inputs to a neural network. Based on the distribution characteristics of the overall jawbone and cancellous bone, or the relationship between the two, a dual-channel approach determines the overall jawbone features and cancellous bone features, significantly improving the accuracy of feature determination. This application also utilizes the differences between overall jawbone features and cancellous bone features to determine cortical bone features, ensuring feature determination is entirely based on the actual condition of the scanned subject, simplifying the processing steps, and balancing speed and accuracy.

[0058] Jawbone features refer to the characteristics corresponding to the jawbone structure, which can include overall jawbone features and local jawbone features. When the analysis object includes the maxilla and mandible, jawbone features can also include features corresponding to the maxilla or features corresponding to the mandible, as shown in Figure 2. Figure 2 is a tomographic image of the corresponding transverse section of the jawbone. The black and gray areas shown in Figure 2 can be used to distinguish different parts. The root bone scan image obtained by implementing this application can be a grayscale image or a color image.

[0059] Classified by tooth location, local jawbone features can include characteristics of the jawbone region corresponding to the anterior teeth and the jawbone region corresponding to the posterior teeth. Classified by the physiological structure of the jawbone, local jawbone features can include characteristics of the cortical bone and the cancellous bone.

[0060] Figure 2 shows the maxillary region J1 and the mandibular region J2. The maxillary region J1 includes the cortical bone CO1 and cancellous bone sp1 of the maxilla. The mandibular region J2 includes the cortical bone CO2 and cancellous bone sp2 of the mandible.

[0061] The cortical bone, located on the outer layer of the jawbone, provides protection and support; the cancellous bone, located on the inner layer of the jawbone, is typically spongy and responsible for hematopoiesis. The bone density of the cancellous bone is lower than that of the cortical bone.

[0062] During orthodontic treatment, external forces are applied to move the teeth, and these forces are transmitted to the jawbone. At this time, the cancellous bone is responsible for absorbing and dispersing the force and is highly active in bone remodeling, while the cortical bone provides support and protection. Based on this, the practitioner can adjust the treatment plan according to the different responses of the cancellous and cortical bone to the orthodontic process. For example, the practitioner can adjust the magnitude of the applied force to prevent damage to the cortical bone.

[0063] Jawbone features can be characterized by images. Images can be two-dimensional images or three-dimensional models. Two-dimensional images can be color renderings along various directions or masked images along various directions (specifically, binary images). Three-dimensional models can be stereoscopic models that can represent the distribution features of the jawbone.

[0064] Jawbone features can be characterized using data. This data can be dimensional information corresponding to different dimensions of the jawbone as a whole, positional information of different local parts of the jawbone, or information about the relationships between different parts of the jawbone.

[0065] As shown in Figure 3, one embodiment of this application provides a method for determining the characteristics of the root bone.

[0066] The application program corresponding to this method can be mounted on an electronic device, a device for determining calcaneal features and / or a storage medium, or it can be mounted on a carrier of the method for displaying calcaneal features provided below, or it can be mounted on a display device to achieve the corresponding technical effect.

[0067] Methods for determining calcaneal features may include at least one of the following steps.

[0068] Step S11: Obtain a radicular bone scan image.

[0069] A root bone scan image can be an image used to characterize bones and / or tooth roots, which can be obtained by scanning the subject.

[0070] Periatomy scans can be used to determine jawbone features. Periatomy scans can be images used to characterize the jawbone. Periatomy scans can also be used to determine tooth root features. Periatomy scans can also be images used to characterize tooth roots.

[0071] A calcaneal bone scan image can be a tomographic image. A tomographic image can be a two-dimensional image or a combination of two-dimensional images. Specifically, CT (Computed Tomography) technology can be used to scan the subject layer by layer using X-rays, obtaining multiple tomographic images corresponding to different locations. The tomographic sections can correspond to the coronal plane (from anterior to posterior), sagittal plane (from left to right), or cross-section plane (from top to bottom).

[0072] When the slice is a transverse section, the calcaneal scan image can be obtained by scanning layer by layer along the vertical direction. In this case, the calcaneal scan image can be shown in Figure 4(a), illustrating the tissue structure at a certain transverse section in the vertical direction. The pixel values ​​of hard tissues such as bones and teeth in the calcaneal scan image can be higher than the pixel values ​​of soft tissues within them; the grayscale values ​​of hard tissues such as bones and teeth in the calcaneal scan image can also be higher than the grayscale values ​​of soft tissues within them.

[0073] The calcaneal scan image can be in the form of a two-dimensional image sequence or a two-dimensional image set. The image sequence or image set can include multiple images, each corresponding to a tomographic scan image at a different position in the scanning direction, as shown in Figure 4(b). When the calcaneal scan image is further processed, the processed image can be in one-to-one correspondence with the original calcaneal scan image; the processed images can form another image sequence or image set, with the number of images before and after processing corresponding, and the positions of the tested object represented by the images before and after processing corresponding.

[0074] The radicular bone scan image can be a sequence of two-dimensional tomographic images of the jawbone region.

[0075] Periatomy images include cone-beam computed tomography (CBCT) images. Periatomy images can be composed of cone-beam computed tomography (CBCT) images. CBCT images are obtained through cone-beam X-ray scanning and are particularly suitable for examining hard tissue structures such as the jawbone and tooth roots. CBCT images provide three-dimensional image information of the jawbone structure of the subject being examined.

[0076] When the calcaneal scan image is a CBCT image, the method for determining calcaneal features provided in this application can also overcome the complex situation of uneven gray-level distribution, artifacts and low signal-to-noise ratio in CBCT images, and improve the accuracy of jawbone feature determination.

[0077] The calcaneal scan image obtained in step S11 can be a normalized CBCT image.

[0078] Step S12: Based on the root bone scan image, perform jawbone segmentation to determine the jawbone segmentation image.

[0079] Jawbone segmentation can be performed by distinguishing the region representing the jawbone in a radicular bone scan image from regions representing other parts of the jawbone in the same image. Alternatively, jawbone segmentation can involve identifying the jawbone region as the region of interest and segmenting that region within the radicular bone scan image.

[0080] Jaw segmentation can also be performed by separating the region representing the maxilla from the region representing the mandible in the radicular bone scan image.

[0081] Compared to segmenting specific tissue structures in the jawbone (e.g., cortical bone, cancellous bone, etc.), the regional segmentation or differentiation of the upper and lower jaws performed in step S12 can be understood as coarse segmentation.

[0082] The jaw segmentation image can include a mask image corresponding to the maxillary region and a mask image corresponding to the background. The mask image can be a binary image, where 1 represents the foreground and 0 represents the background.

[0083] The jaw segmentation image can include a mask image corresponding to the mandibular region and a mask image corresponding to the background.

[0084] The method for determining calcaneal features provided in this application may include: segmenting the jawbone using a neural network.

[0085] The input to this neural network can be a root bone scan image. The neural network can include three output channels: the first channel outputs a mask image corresponding to the maxilla, the second channel outputs a mask image corresponding to the mandible, and the third channel outputs a mask image corresponding to the background.

[0086] This neural network can include a fully convolutional deep neural network. The neural network is used to achieve global segmentation of pedicle scan images. For example, the neural network can be a UNet, Atten-UNet, UNet++, or similar architecture.

[0087] The jaw segmentation image can simultaneously include information for the corresponding maxillary region (lighter colored area) and information for the corresponding mandibular region (darker colored area), as shown in Figure 5.

[0088] The calcaneal scan image can be a two-dimensional image. Based on this, each calcaneal scan image in the calcaneal scan image sequence can be segmented into the jawbone separately to obtain multiple two-dimensional jawbone segmentation images. These multiple jawbone segmentation images can then be combined in the same order as the calcaneal scan image sequence to obtain a jawbone segmentation image sequence that represents the three-dimensional jawbone segmentation result.

[0089] At this point, if a neural network is used for jawbone segmentation, the neural network can be a two-dimensional deep fully convolutional neural network.

[0090] The calcaneal scan image can be a three-dimensional image. Based on this, the jawbone can be segmented from the calcaneal scan image to obtain a three-dimensional segmented jawbone image.

[0091] At this point, if a neural network is used for jawbone segmentation, the neural network can be a three-dimensional deep fully convolutional neural network.

[0092] In one embodiment, the method for determining radicular bone features provided in this application may include the steps of: using a neural network to segment the maxilla and mandible based on radicular bone scan images, and determining a segmented image of the maxilla representing the maxilla and a segmented image of the mandible representing the mandible.

[0093] This step can be included in step S12; when implementing step S12, this step can be implemented in detail.

[0094] The determined segmented images of the maxilla and mandible can be presented in the same segmented image, as shown in Figure 5.

[0095] The determined segmented images of the maxilla and mandible can be presented as two separate segmented images. When the radicular bone scan image sequence contains multiple radicular bone scan images, segmenting the jawbone can yield a segmented image sequence for the corresponding maxilla and a segmented image sequence for the corresponding mandible, as shown in Figure 6.

[0096] When calcaneal scan images are used to compose a calcaneal scan image sequence, and the calcaneal scan images are cross-sectional tomographic images, multiple calcaneal scan images have a predetermined order in the scanning direction (e.g., vertical direction). After jawbone segmentation of the calcaneal scan images (included in the calcaneal scan image sequence), the determined jawbone segmentation images correspond one-to-one with the calcaneal scan images in that scanning direction, and correspond to different jawbone positions with respect to that scanning direction. The number of images in the calcaneal scan image sequence is equal to the number of images in the jawbone segmentation images.

[0097] Step S13: Using a neural network, the root bone scan image and the jaw segmentation image representing the same jawbone location are used as inputs to determine the jawbone feature image and the cancellous bone feature image.

[0098] The segmented image of the jawbone is determined based on the radicular bone scan image. The segmented image of the jawbone corresponds to the radicular bone scan image; the segmented image of the jawbone and the radicular bone scan image are located in the same coordinate system.

[0099] The radicular bone scan image and the jawbone segmentation image represent the same jawbone location. This can be: the radicular bone scan image corresponding to the maxilla of the subject, together with the jawbone segmentation image also corresponding to the maxilla, are used to determine the jawbone feature image and cancellous bone feature image of the corresponding maxilla; or, the radicular bone scan image and the jawbone segmentation image corresponding to the mandible are used together to determine the jawbone feature image and cancellous bone feature image of the corresponding mandible.

[0100] The root bone scan image and the jawbone segmentation image represent the same jawbone location. They can be: a root bone scan image corresponding to a first region of the jawbone of the subject, and a jawbone segmentation image corresponding to the same first region, which are used together to determine the jawbone feature image and cancellous bone feature image corresponding to the first region.

[0101] The radicular bone scan images constitute a radicular bone scan image sequence, and the jawbone segmentation images constitute a jawbone segmentation image sequence. The radicular bone scan image corresponding to the first position in the scanning direction in the radicular bone scan image sequence, and the jawbone segmentation image corresponding to the same first position in the jawbone segmentation image sequence, are used together to determine the jawbone feature image and cancellous bone feature image corresponding to the first position.

[0102] This neural network can be a fully convolutional deep neural network. Specifically, it can be a multi-label fully convolutional deep neural network, which can predict multiple labels simultaneously for each pixel or voxel using a fully convolutional network (FCN). Fully convolutional neural networks process inputs of arbitrary size within a certain range and output results with the same spatial resolution, facilitating further processing. Combined with its multi-label feature, it can predict the probabilities of different labels through multiple output channels or branches.

[0103] In this application, when the neural network is a fully convolutional deep neural network, the neural network can simultaneously output jawbone feature images and cancellous bone feature images through multiple output channels, and make the jawbone feature images and cancellous bone feature images have the same spatial structure.

[0104] This neural network can be trained using binary cross-entropy as a loss function.

[0105] The neural network can include two input channels. The first input channel is used to input either a radicular bone scan image or a processed radicular bone scan image; the second input channel is used to input either a segmented jawbone image or a processed segmented jawbone image.

[0106] The processed radicular bone scan image can be a cropped radicular bone scan image containing only the region of interest; similarly, the processed jawbone segmentation image can also be a cropped jawbone segmentation image containing only the region of interest. The region of interest in the radicular bone scan image and the jawbone segmentation image are consistent.

[0107] The jawbone segmentation image can be the image obtained by coarse segmentation of the corresponding root bone scan image.

[0108] Compared to the segmented image of the jawbone, step S13 distinguishes the features of the cancellous bone, which can be understood as fine segmentation.

[0109] Specifically, the root bone scan image input to the first input channel (e.g., Figure 7, included in root bone scan image sequence set11) can be a normalized root bone scan image containing a corresponding local region of interest in the maxilla; the jaw segmentation image input to the second input channel (e.g., Figure 7, included in jaw segmentation image sequence set21) can be a normalized jaw segmentation image containing a corresponding local region of interest in the maxilla. Alternatively, the root bone scan image input to the first input channel corresponds to the mandibular region (e.g., Figure 8, the root bone scan image included in root bone scan image sequence set12), and the jaw segmentation image input to the second input channel corresponds to the mandibular region (e.g., Figure 8, the jaw segmentation image included in jaw segmentation image sequence set22).

[0110] The neural network can include two output channels. The first output channel is used to output a jawbone feature image, and the second output channel is used to output a cancellous bone feature image.

[0111] Specifically, the jawbone feature image output by the first output channel can be a jawbone feature image corresponding to the maxilla and in the form of a binary masked image (e.g., Figure 7, included in the jawbone feature image sequence set31); the cancellous bone feature image output by the second output channel can be a cancellous bone feature image corresponding to the maxilla and in the form of a binary masked image (e.g., Figure 7, included in the cancellous bone feature image sequence set41). Alternatively, the jawbone feature image output by the first output channel corresponds to the mandibular region (e.g., Figure 8, the jawbone feature image is included in the jawbone feature image sequence set32), and the cancellous bone feature image output by the second input channel corresponds to the mandibular region (e.g., Figure 8, the cancellous bone feature image is included in the cancellous bone feature image sequence set42).

[0112] In the masked image, 1 represents the foreground and 0 represents the background.

[0113] The calcaneal scan images and jawbone segmentation images can be two-dimensional images. Based on this, each calcaneal scan image in the calcaneal scan image sequence set1, and its corresponding jawbone segmentation image in the jawbone segmentation image sequence set2, can be processed separately or sequentially using a neural network to obtain multiple two-dimensional jawbone feature images and corresponding cancellous bone feature images. Then, following the same order as the calcaneal scan image sequence and jawbone segmentation image sequence, these multiple jawbone feature images can be combined to obtain the jawbone feature image sequence set3, which represents the result of three-dimensional feature extraction of the jawbone; similarly, following the same order as the calcaneal scan image sequence and jawbone segmentation image sequence, the cancellous bone feature images can be combined to obtain the cancellous bone feature image sequence set4, which represents the result of three-dimensional feature extraction of the cancellous bone.

[0114] At this point, the neural network can be a two-dimensional deep fully convolutional neural network.

[0115] Both the radicular bone scan image and the jawbone segmentation image can be three-dimensional images. Based on this, a neural network can be used to process both to obtain a three-dimensional jawbone feature image and a corresponding cancellous bone feature image. Alternatively, the three-dimensional radicular bone scan image and the jawbone segmentation image can be divided into several overlapping or non-overlapping three-dimensional sub-regions. Each three-dimensional sub-region of the radicular bone scan image, along with its corresponding three-dimensional sub-region of the jawbone segmentation image, is input into a neural network for processing to obtain a three-dimensional sub-region of the jawbone feature image corresponding to that sub-region, and a corresponding three-dimensional sub-region of the cancellous bone feature image. Then, the three-dimensional sub-regions of the jawbone feature image are combined to obtain a jawbone feature image representing the result of three-dimensional feature extraction; similarly, the three-dimensional sub-regions of the cancellous bone feature image are combined to obtain a cancellous bone feature image representing the result of three-dimensional feature extraction.

[0116] At this point, the neural network can be a three-dimensional deep fully convolutional neural network.

[0117] Thus, by using a "coarse-to-fine" feature determination method, the accuracy of determining jawbone features and cancellous bone features can be significantly improved.

[0118] The differences between radicular bone scan images, jawbone segmentation images, and jawbone feature images will be explained in detail below.

[0119] A radicular bone scan image is an image obtained by scanning the subject and containing the most complete hard and / or soft tissue features of the jawbone. The jawbone structural features are mixed with other features and require professional analysis or further processing to extract the specific features of the jawbone.

[0120] Jaw segmentation images can be "coarse segmentation" results obtained by directly processing root bone scan images based on global jaw segmentation. Jaw segmentation images at least distinguish the corresponding maxilla and mandible parts in the root bone scan images, indicating the distribution area and relationship of the maxilla and mandible. However, the specific structures in each group of jaws still need further analysis or processing.

[0121] Jawbone feature images can be the result of "fine segmentation" obtained after processing radicular bone scan images. The distribution area of ​​the jawbone region determined in the jawbone feature image is more accurate than that in the jawbone segmentation image, and the jawbone feature image includes the corresponding cancellous bone feature image.

[0122] In step S12, jawbone segmentation is performed. After separating the region corresponding to the maxilla and the region corresponding to the mandible in the root bone scan image, the steps of determining the structural features of the corresponding maxilla and the steps of determining the structural features of the corresponding mandible can be performed simultaneously or sequentially.

[0123] For example, step S12 includes the steps of "determining the segmented image of the jawbone corresponding to the maxillary region" and "determining the segmented image of the jawbone corresponding to the mandibular region". If the two steps are executed sequentially, then step S13 includes the steps of "using a neural network, taking the root bone scan image and jawbone segmentation image representing the same jawbone position in the maxillary region as input, to determine the jawbone feature image and cancellous bone feature image of the corresponding maxilla", and "using the same or another neural network, taking the root bone scan image and jawbone segmentation image representing the same jawbone position in the mandibular region as input, to determine the jawbone feature image and cancellous bone feature image of the corresponding mandible"; and the two steps are executed sequentially.

[0124] The neural network used to process images of the corresponding maxillary region and the neural network used to process images of the corresponding mandibular region can have the same configuration or different configurations. Based on the structural differences between the maxillary and mandibular regions, it is preferable to use two different neural networks and train them separately.

[0125] In one embodiment, the method for determining radicular bone features provided in this application may include the steps of: using a neural network to extract features from a radicular bone scan image with reference to a segmented jawbone image, and determining a jawbone feature image and a cancellous bone feature image.

[0126] Considering that the location and distribution area of ​​the jawbone have been determined in the segmented jawbone image, using the segmented jawbone image as a reference for feature extraction can enhance the neural network's ability to extract features from the foreground region and improve the accuracy of the determined jawbone feature image and cancellous bone feature image.

[0127] Specifically, the method for determining radicular bone features provided in this application enables a neural network to extract features from radicular bone scan images using jawbone segmentation images as prior knowledge.

[0128] This step can be included in step S13; when implementing step S13, this step can be implemented in detail.

[0129] Step S14: Compare the jawbone feature image and the cancellous bone feature image, and determine the cortical bone feature image based on the differences.

[0130] The jawbone structure mainly consists of the cancellous bone and the cortical bone. When determining the characteristics of the cancellous bone and jawbone, the cortical bone characteristics can be determined based on the differences between the jawbone characteristics and the cancellous bone characteristics. Since the cortical bone is usually thin (especially the maxillary cortex), directly predicting the cortical bone results in class imbalance, leading to low segmentation accuracy. However, cortical bone characteristics determined by comparing differences have higher accuracy, and the determination process is computationally less and more efficient.

[0131] For the maxillary region, as shown in Figure 9, step S14 can be to compare the corresponding maxillary bone feature image fig31 with the corresponding maxillary cancellous bone feature image fig41, and determine the corresponding maxillary cortical bone feature image fig51 based on the differences.

[0132] The maxillary bone feature image fig31 can be included in the maxillary bone feature image sequence set31; the maxillary cancellous bone feature image fig41 can be included in the maxillary cancellous bone feature image sequence set41.

[0133] For the mandibular region, as shown in Figure 10, step S14 can be to compare the corresponding mandibular bone feature image fig32 with the corresponding mandibular cancellous bone feature image fig42, and determine the corresponding mandibular cortical bone feature image fig52 based on the difference.

[0134] The corresponding mandibular bone feature image fig32 can be included in the corresponding mandibular bone feature image sequence set32; the corresponding mandibular bone cancellous feature image fig42 can be included in the corresponding mandibular bone cancellous feature image sequence set42.

[0135] The jawbone feature image sequence set3 contains multiple jawbone feature images fig3, and the cancellous bone feature image sequence set4 contains multiple cancellous bone feature images fig4. Jawbone feature images fig3 and cancellous bone feature images fig4 representing the same jawbone location are compared, and the corresponding cortical bone feature image fig5 is determined based on the differences.

[0136] In one embodiment, the method for determining calcaneal features provided in this application may include the steps of: performing pixel-by-pixel subtraction between a jawbone feature image and a cancellous bone feature image to determine a cortical bone feature image.

[0137] This step can be included in step S14; when implementing step S14, this step can be implemented in detail.

[0138] The jawbone feature image and the cancellous bone feature image can be a masked image; the masked image can be a binary image or a grayscale image. Specifically, the jawbone feature image and the cancellous bone feature image can be a binary masked image, where the pixel value in the image is 0 or 255.

[0139] For each pixel position (x, y), the following operation can be performed:

[0140] fig5(x,y)=fig3(x,y)–fig4(x,y).

[0141] In this way, the cortical bone feature image can be determined based on the difference between the jawbone feature image and the cancellous bone feature image.

[0142] Step S15: Determine and output the bone cortex feature information based on the bone cortex feature image.

[0143] The method for determining calcaneal features provided in this application can directly output the cortical bone feature image obtained in step S14 as the cortical bone feature information to present the cortical bone features; or it can output at least one of the cortical bone feature image, cancellous bone feature image, and jawbone feature image.

[0144] The method for determining calcaneal features provided in this application can determine the data information contained in a cortical bone feature image and output the data information as the cortical bone feature information. For example, at least one of the following can be determined from the cortical bone feature image: cortical bone size information (e.g., shape, thickness), resorption status, etc. Alternatively, the cortical bone density information can be determined by combining the cortical bone feature image and the calcaneal scan image.

[0145] In one embodiment, the method for determining calcaneal features provided in this application may include the steps of: determining and outputting cortical bone size information based on a cortical bone feature image.

[0146] This step can be included in step S15; when implementing step S15, this step can be implemented in detail.

[0147] The method for determining calcaneal bone characteristics provided in this application can also determine the data information contained in a cancellous bone feature image and output the data information. For example, at least one of the following cancellous bone structural information (e.g., trabecular morphology, trabecular arrangement, trabecular spacing), thickness information, etc., can be determined from the cancellous bone feature image.

[0148] In one embodiment, the method for determining calcaneal bone characteristics provided in this application may include the steps of: determining and outputting bone density information based on a cancellous bone feature image and a calcaneal bone scan image.

[0149] This step can be included in step S15; when implementing step S15, this step can be implemented in detail.

[0150] When the cortical bone feature image obtained in step S14 is a two-dimensional image, the method for determining the features of the calcaneus provided in this application can establish a three-dimensional model containing cortical bone feature information based on a sequence of cortical bone feature images composed of multiple cortical bone feature images, and output this as cortical bone feature information.

[0151] In one embodiment, the method for determining calcaneal features provided in this application may include the step of: establishing a three-dimensional model of the cortical bone based on several cortical bone feature images corresponding to several jawbone locations.

[0152] This step can be included in step S15; when implementing step S15, this step can be implemented in detail.

[0153] The bone cortical feature images used to build a three-dimensional model of the bone cortex can be all or part of a sequence of bone cortical feature images.

[0154] When the jawbone feature image is a two-dimensional image, the method for determining the root bone features provided in this application can establish and output a three-dimensional model 3d-3 containing jawbone features based on a jawbone feature image sequence set3 composed of multiple jawbone feature images.

[0155] As shown in Figure 11, a 3D model 3d-31 of the corresponding maxillary jawbone features can be obtained by 3D reconstruction based on the corresponding maxillary jawbone feature image sequence set31. As shown in Figure 12, a 3D model 3d-32 of the corresponding mandibular jawbone features can be obtained by 3D reconstruction based on the corresponding mandibular jawbone feature image sequence set32.

[0156] In one embodiment, the method for determining the features of the radicle provided in this application may include the steps of: establishing a three-dimensional model of the jawbone based on several jawbone feature images corresponding to several jawbone positions.

[0157] This step can be included in step S15; when implementing step S15, this step can be implemented in detail.

[0158] The jawbone feature images used to create a three-dimensional model of the jawbone can be all or part of a sequence of jawbone feature images.

[0159] When the cancellous bone feature image is a two-dimensional image, the method for determining the radicular bone features provided in this application can establish and output a three-dimensional model 3d-4 containing cancellous bone features based on a sequence of multiple cancellous bone feature images set4.

[0160] As shown in Figure 13, a 3D model 3d-41 of the cancellous bone features of the corresponding maxilla can be obtained by 3D reconstruction based on the image sequence set41 of the cancellous bone features of the corresponding maxilla. As shown in Figure 14, a 3D model 3d-42 of the cancellous bone features of the corresponding mandible can be obtained by 3D reconstruction based on the image sequence set42 of the cancellous bone features of the corresponding mandible.

[0161] In one embodiment, the method for determining radicular bone features provided in this application may include the step of: establishing a three-dimensional model of cancellous bone based on several cancellous bone feature images corresponding to several jawbone locations.

[0162] This step can be included in step S15; when implementing step S15, this step can be implemented in detail.

[0163] The cancellous bone feature images used to build a three-dimensional model of cancellous bone can be all or part of a sequence of cancellous bone feature images.

[0164] At least one of the jawbone features, cortical bone features, and cancellous bone features can also be displayed on the oral scan model or other three-dimensional models to serve as a reminder or indication.

[0165] The method for determining calcaneal features provided in this application may include the steps of: obtaining an oral scan model and displaying cortical bone feature information on the oral scan model.

[0166] An oral scan model can correspond to the same test subject as a calcaneal scan image; an oral scan model can correspond to the same test subject as cortical bone feature information.

[0167] In one embodiment, the method for determining calcaneal features provided in this application may include the steps of:

[0168] Step P11: Obtain the oral scan model;

[0169] Step P12 involves performing coordinate transformation on the cortical bone feature information to display it on the oral scan model.

[0170] Step P12 can be included in step S15; when step S15 is implemented, step P12 can be specifically implemented. Step P11 can be implemented at any time before step P12.

[0171] In step P12, coordinate transformation can also be performed on the cortical bone feature image.

[0172] The coordinate transformation method may include at least one of the following steps: determining the calcaneal model based on the calcaneal scan image; registering the calcaneal model and the oral scan model to obtain coordinate transformation information; and performing coordinate transformation on the cortical bone feature information based on the coordinate transformation information to display the cortical bone feature information at the oral scan model.

[0173] In step P12, the cortical bone feature information can be combined with the intraoral scan model for display. Specifically, the cortical bone feature information can be combined and displayed according to the correspondence between the position of the crowns, gingiva, etc., in the intraoral scan model. In this way, the cortical bone feature information will be displayed within the area of ​​the intraoral scan model, providing a reference for comparison.

[0174] By implementing step P12, the areas of cortical bone features in the oral scan model can be highlighted, or information such as cortical bone size can be annotated in the oral scan model to provide guidance.

[0175] The method for determining calcaneal bone characteristics provided in this application can obtain an oral scan model and display the cancellous bone characteristic information at the oral scan model.

[0176] The characteristics of cancellous bone can be determined based on images of cancellous bone features.

[0177] An oral scan model can correspond to the same test subject as a calcaneal scan image; an oral scan model can correspond to the same test subject as cancellous bone feature information.

[0178] In one embodiment, the method for determining calcaneal features provided in this application may include the steps of:

[0179] Step P21: Determine the characteristic information of cancellous bone based on the cancellous bone characteristic image;

[0180] Step P22, obtain the oral scan model;

[0181] Step P23: Perform coordinate transformation on the cancellous bone feature information to display the cancellous bone feature information at the oral scan model.

[0182] Step P21 can be included in step S14; when implementing step S14, step P21 can be specifically implemented.

[0183] Step P23 can be included in step S15; when step S15 is implemented, step P23 can be specifically implemented. Step P22 can be implemented at any time before step P23.

[0184] In step P23, coordinate transformation can also be performed on the cancellous bone feature image.

[0185] The coordinate transformation method may include at least one of the following steps: determining the calcaneal model based on the calcaneal scan image; registering the calcaneal model and the oral scan model to obtain coordinate transformation information; and performing coordinate transformation on the cancellous bone feature information based on the coordinate transformation information to display the cancellous bone feature information at the oral scan model.

[0186] In step P23, the cancellous bone feature information can be combined with the intraoral scan model for display. Specifically, the combination can be based on the correspondence between the cancellous bone feature information and the positions of the crowns, gingiva, etc., in the intraoral scan model. In this way, the cancellous bone feature information will be displayed within the area of ​​the intraoral scan model, providing a reference for comparison.

[0187] By implementing step P23, the areas of cancellous bone features in the oral scan model can be highlighted, or bone density information such as cancellous bone can be annotated in the oral scan model to provide indication.

[0188] In step S13 above, the corresponding radicular bone scan image and jawbone segmentation image can be images containing the jawbone region obtained through cropping. Step S13 may include a cropping step of the radicular bone scan image, or may include a cropping step of both the radicular bone scan image and the jawbone segmentation image.

[0189] The jawbone segmentation image obtained in step S12 can be used to determine the region where the jawbone is located. Since the jawbone segmentation image corresponds to the root bone scan image, the region where the jawbone is located can be determined in the root bone scan image and cropped accordingly.

[0190] In one embodiment, as shown in FIG15, the method for determining calcaneal features provided in this application may include the following steps:

[0191] Step M11: Obtain a root bone scan image and a segmented image of the jawbone at the same jawbone location in the first direction;

[0192] Step M12: Based on the segmented jawbone image, determine the region representing the jawbone area in the root bone scan image;

[0193] Step M13: Crop the radicular bone scan image.

[0194] Referring to Figure 16, in step M11, the first direction D1 is perpendicular to the display plane of the radicular bone scan image and the jawbone segmentation image. The radicular bone scan image is contained in the original radicular bone scan image sequence set01, and the jawbone segmentation image is contained in the original jawbone segmentation image sequence set02.

[0195] The radicular bone scan image and the segmented jawbone image can be two-dimensional images.

[0196] The display plane can be the distribution plane of the area shown in the root bone scan image and the jawbone segmentation image in the actual jawbone area.

[0197] The calcaneal scan image can specifically be a tomographic scan image. Based on this, the first direction D1 can be a direction perpendicular to the tomographic plane or the scanning direction of the tomographic scan. The display plane can be the tomographic plane of the tomographic scan image. The tomographic plane can be a transverse section, and the first direction D1 can be along the vertical direction.

[0198] The jawbone segmentation image is determined based on the root bone scan image. The first direction D1 is perpendicular to the plane where the root bone scan image is located, therefore the first direction D1 is perpendicular to the plane where the jawbone segmentation image is located.

[0199] The original calcaneal scan image sequence set01 includes several original calcaneal scan images. Different original calcaneal scan images correspond to different positions on the first direction D1, and also to different jawbone positions. Similarly, the original jawbone segmentation image sequence set02 contains different original jawbone segmentation images that correspond to different positions on the first direction D1.

[0200] Original root bone scan images and original jaw segmentation images corresponding to the same jawbone position and the same position on the first direction D1 exist in pairs. When executing step M11, multiple sets of original root bone scan images and original jaw segmentation images are confirmed.

[0201] After obtaining the original jawbone segmentation image based on the original radicular bone scan image, the jawbone segmentation image records the distribution area of ​​the jawbone, which can be shown as the selection box of each image in the original jawbone segmentation image sequence set02 in Figure 16.

[0202] Based on the regions representing the jawbone area determined by the original jawbone segmentation image, and the correspondence between the original jawbone segmentation image and the original radicular bone scan image, the radicular bone scan image can be cropped (as shown in the first cropping in Figure 16) to increase the proportion of the jawbone area in the radicular bone scan image, or to make the radicular bone scan image include only the regions representing the jawbone area.

[0203] The cropped radicular bone scan images can be used to form a radicular bone scan image sequence set1. The radicular bone scan images in the processed radicular bone scan image sequence set1 can be used to perform steps S11 to S15, for example, to perform jawbone segmentation to obtain a jawbone segmentation image with a larger proportion of the jawbone area, or as input to a neural network to obtain better jawbone feature images, cancellous bone feature images, and cortical bone feature images.

[0204] Taking the mandible as an example, the original root bone scan image sequence set012 and the original jawbone segmentation image sequence set022 of the corresponding mandible can be obtained. The region representing the jawbone is determined using the original jawbone segmentation image sequence set022, and the first cropping is performed on the original root bone scan image sequence set012 to obtain the cropped root bone scan image sequence set12 of the corresponding mandible.

[0205] The procedure for the maxilla is similar and will not be repeated here.

[0206] The region representing the jawbone can also be replaced with other regions of interest.

[0207] In one embodiment, the method for determining radicular bone features provided in this application may include the steps of: determining the region representing the jawbone area in a segmented jawbone image; and determining the region representing the jawbone area in a radicular bone scan image based on the region representing the jawbone area in the segmented jawbone image.

[0208] In one specific embodiment, the region representing the jawbone can be characterized using a selection box. The selection box can be a two-dimensional selection box or a three-dimensional selection box as shown in Figure 16.

[0209] In one specific embodiment, the method for determining radicular bone features provided in this application may include the steps of: determining the region representing the jawbone in a segmented jawbone image and obtaining a first selection box corresponding to the jawbone region; adjusting the size of the first selection box according to a preset expansion value to obtain a second selection box; determining the region representing the jawbone in a radicular bone scan image according to the second selection box, and cropping the radicular bone scan image accordingly.

[0210] Adjusting the size of the first selection box can expand it to create a second selection box with a larger volume or area than the first selection box.

[0211] When the first selection box is a two-dimensional rectangle, it can be determined based on the coordinates of its diagonal vertices. Adjusting the size of the first selection box can be achieved by adjusting the coordinates of at least one of its two diagonal vertices.

[0212] When the first selection box is a 3D rectangle, it can be determined based on the coordinates of its diagonal vertices. Adjusting the size of the first selection box can be achieved by adjusting the coordinates of at least one of its two diagonal vertices.

[0213] For example, the first selection box includes a first vertex v1 and a second vertex v2. The coordinates of the first vertex v1 are (x1, y1, z1), and the coordinates of the second vertex v2 are (x2, y2, z2). Based on this, the coordinates of the two vertices can be adjusted to expand the first selection box into a second selection box. The coordinates of the new first vertex v1' are (x1-margin_x, y1-margin_y, z1-margin_z), and the coordinates of the new second vertex v2' are (x2+margin_x, y2+margin_y, z2+margin_z).

[0214] The reason for adjusting the size of the selection box to expand it is to include part of the area around the jawbone (background), which helps improve the recognition accuracy of the jawbone (foreground) area.

[0215] The magnitude of the expansion (i.e., the aforementioned margin_x, margin_y, and margin_z) can be preset. In one embodiment, the preset expansion value is set as follows: the magnitude by which the first selection box is expanded is a preset multiple of the corresponding side length of the first selection box. Specifically, the preset value is 0.2.

[0216] For example, when adjusting the size of the x-coordinate direction, the expansion is equal to 0.2 times the side length of the x-coordinate direction of the first x-selection box (e.g., |x2-x1|). The same applies to other coordinate directions.

[0217] For example, the default extended value can be set as follows:

[0218] margin_x = 0.2 × (x2 - x1);

[0219] margin_y = 0.2 × (y2 - y1);

[0220] margin_z = 0.2 × (z2 - z1).

[0221] Once the second selection box is determined, it can characterize the region of the jawbone formed after the boundary expansion. Based on the second selection box, the region of the jawbone can be determined in the radicular bone scan image, and then a first cropping can be performed on the radicular bone scan image based on the determined region.

[0222] In some embodiments, the segmented jawbone image may also be cropped. By performing the first and second cropping, the resulting processed radicular bone scan image and jawbone segmentation image can be directly used in the aforementioned step S13 to directly determine the jawbone feature image and cancellous bone feature image.

[0223] In one specific embodiment, the method for determining radicular bone features provided in this application may include the steps of: determining the region representing the jawbone area in a segmented jawbone image; cropping the segmented jawbone image; and determining the region representing the jawbone area in a radicular bone scan image based on the region representing the jawbone area in the segmented jawbone image.

[0224] The same bounding box method described above can be used to determine the regions representing the jawbone area in a segmented jawbone image. These bounding boxes can be obtained through connected component analysis.

[0225] Based on the determined second selection box, the original jawbone segmentation image can be cropped a second time, and the generated jawbone segmentation image is used to form the jawbone segmentation image sequence set2. The jawbone segmentation images in the processed jawbone segmentation image sequence set2 can be used to perform steps S11 to S15.

[0226] Taking the mandible as an example, the original mandibular segmentation image sequence set022 of the corresponding mandible can be obtained. The second selection box is determined using it, and the second cropping is performed on the mandibular segmentation image sequence set022 of the corresponding mandible to obtain the cropped mandibular segmentation image sequence set22 of the corresponding mandible.

[0227] The procedure for the maxilla is similar and will not be repeated here.

[0228] In summary, the method for determining radicular bone features provided in this application uses the segmented jawbone image obtained by segmenting the radicular bone scan image and the radicular bone scan image itself as dual-channel inputs to a neural network. This enhances the feature extraction capability of the neural network and more accurately determines the overall jawbone features and cancellous bone features. Furthermore, the method provided in this application determines cortical bone features based on the differences between cancellous bone features and overall jawbone features. Compared to existing technologies, the cortical bone feature extraction is more accurate and computationally more efficient.

[0229] One implementation method

[0230] One implementation method can be implemented independently as a method for determining root bone characteristics, or it can be included in the aforementioned method for confirming root bone characteristics. It can also be used to explain other technical solutions in this application, or it can be combined with other technical solutions in this application.

[0231] One implementation method addresses the following: During tooth segmentation in the process of determining root bone features, whether segmenting a 2D image or a 3D model, the teeth are typically located first based on their conventional arrangement before their morphology is determined. However, in cases of supernumerary teeth or impacted teeth, these special teeth cannot be located, leading to segmentation results that fail to balance completeness and accuracy, and do not accurately reflect reality. This implementation method for determining root bone features, based on the positional information of the teeth in the root bone scan image, further references the center information of the teeth in the root bone scan image. This improves the accuracy of tooth detection and segmentation while maintaining both completeness and accuracy. The reference to the root bone scan image is not simply using both images as input for calculation; instead, it uses the center information of the teeth in the root bone scan image to obtain another set of locations for the same dentition. These two sets of location information are then cross-validated and referenced to identify any unconventional or missed teeth, thereby improving the accuracy of tooth detection.

[0232] The calcaneal scan image can be a CBCT (cone beam computed tomography) image.

[0233] Tooth segmentation refers to the process of segmenting teeth into instances within an image, classifying each pixel into either a tooth category or the background. The segmentation results can assist operators in diagnosis and treatment planning. These results can be presented as a 3D model.

[0234] Tooth segmentation does not necessarily lead to the physical differentiation of each tooth. The physical positional relationship of the segmented teeth can be adjusted, such as by setting different teeth at intervals; the segmented teeth can also be distinguished from each other through labels, colors, grayscale, edge depiction, arrow markings, etc. For example, Figure 17 shows the segmented teeth displayed in grayscale.

[0235] Figure 17(a) shows one three-dimensional model corresponding to a tooth, and Figure 17(b) shows another three-dimensional model corresponding to a tooth. Comparing the two three-dimensional models, it can be seen that the three-dimensional model shown in Figure 17(a) is missing the tooth selected by the dashed rectangular frame. This tooth may be a supernumerary tooth or an impacted tooth. By implementing the technical solution provided in this application, this tooth can be identified as the differential tooth in the following text, thereby achieving segmentation together with other teeth.

[0236] Dividing teeth into segments helps in setting different orthodontic forces for different teeth during orthodontic treatment. For example, smaller forces are needed for corresponding incisors to avoid root resorption, while larger forces are needed for corresponding molars to ensure their movement.

[0237] Distinguishing between impacted teeth, supernumerary teeth, and other special teeth helps in identifying the causes of adjacent root lesions or abnormal interdental spaces during orthodontic treatment, and assists the operator in deciding whether to add special tooth treatment steps (such as extracting special teeth) to the treatment plan.

[0238] As shown in Figure 26, one embodiment of this application provides a device 200 for determining the characteristics of the root bone.

[0239] The device 200 for determining calcaneal features includes at least one of the following components:

[0240] The first module 21 is used to obtain pedicle scan images;

[0241] The second module 22 is used to input the root bone scan image into the positioning model to obtain the first positioning information of N teeth;

[0242] The third module 23 is used to identify the center information of different teeth in the root bone scan image and obtain the second positioning information of M teeth, where M is a positive integer greater than 1;

[0243] The fourth module 24 is used to determine the region of interest for tooth segmentation based on the first localization information of N teeth and the second localization information of M teeth.

[0244] The fifth module 25 is used to process the image corresponding to the region of interest in tooth segmentation using a segmentation model to obtain the tooth segmentation result.

[0245] The first positioning information is used to determine the position of the tooth in the root bone scan image.

[0246] The localization model can be implemented using at least one neural network. The at least one neural network is pre-trained to determine the positional information of the tooth in the root bone scan image, and / or to determine first localization information. The second module 22 may carry the localization model.

[0247] The second positioning information may include the center information of different teeth in the root bone scan image. The center information may be the information of the two-dimensional center point, the three-dimensional center point, the long axis, or other anatomical features of the tooth. The information of the two-dimensional center point may be the information of the geometric center point, centroid point, or centroid point of the tooth in the two-dimensional image of each cross section, or it may be a set of multiple center points in multiple cross sections, or information of the three-dimensional structure fitted based on multiple center points. The third module 23 can use morphological features such as boundary distance to identify the center information of the tooth, specifically by using a boundary distance regression model or other neural network models to achieve identification. In other embodiments, the third module 13 may also use edge detection and contour extraction, minimum bounding rectangle and centroid calculation, ellipse fitting, template matching, etc.

[0248] The fourth module 24 can determine the region of interest (ROI) for tooth segmentation by comparing the second and first localization information. This ROI, used to guide tooth segmentation, can be all teeth in the root bone scan image or only a portion of the teeth. Specifically, a portion of the teeth can be a differentially expressed tooth included in one of the first and second localization information but excluded from the other, or an overlapping tooth included in both localization information but with different data in the two localization information. Differentially expressed teeth can be supernumerary teeth, impacted teeth, or omitted teeth.

[0249] The tooth segmentation result is determined based on the region of interest (ROI) of the tooth segmentation, which can be all teeth or a portion of the teeth. The tooth segmentation result can be in the form of a two-dimensional image, a three-dimensional image, or a three-dimensional model. When the tooth segmentation result is a three-dimensional model, the fifth module 25 can also be used to perform three-dimensional reconstruction based on the ROI of the tooth segmentation; similarly, the second module 21 can also be used to perform three-dimensional reconstruction based on the first positioning information.

[0250] The first module 21 can obtain pedicle scan images acquired by external devices, or it can perform acquisition operations itself to obtain pedicle scan images.

[0251] In one embodiment, the device 200 for determining pedicle features may also be connected to the first device 101.

[0252] The first device 101 can be disposed outside the device 200 for determining calcaneal features, serving as an external device to the device 200 for determining calcaneal features. The first device 101 and the device 200 for determining calcaneal features can also be integrated together.

[0253] The first device 101 is used to perform scanning to obtain radicular bone scanning information. The radicular bone scanning information can be the radicular bone scanning image or data information related to the radicular bone scanning image. The first device 101 can scan the jawbone area or perform an oral cavity scan.

[0254] The first device 101 can be used to implement a cone-beam ring hood to obtain calcaneal scan images.

[0255] The calcaneal scan image can be used to determine initial localization information. For example, the calcaneal scan image can be processed using one or more convolutional neural networks to determine the initial localization information.

[0256] Peripheral scan images can be used to determine secondary localization information. For example, one or more convolutional neural networks can be used to process peripheral scan images to determine secondary localization information.

[0257] Root bone scan images can be used to determine 3D models. For example, one or more convolutional neural networks can be used to process root bone scan images to obtain first localization information, and then tooth segmentation and 3D reconstruction can be performed based on the first localization information to generate a 3D model. Alternatively, tooth segmentation and 3D reconstruction can be performed based on both the first localization information and information about the first tooth to generate a 3D model. The information used to generate the 3D model can be in the form of a binary masked image.

[0258] The apparatus 200 for determining calcaneal features can also be configured based on the method for determining calcaneal features in any of the technical solutions provided below. Specifically, depending on the relationship between the steps, related steps can be implemented in the same or different modules.

[0259] Electrical or communication connections can be established between the first module 21, the second module 22, the third module 23, the fourth module 24, and the fifth module 25 to enable data transmission.

[0260] One embodiment of this application may also provide a display device.

[0261] The display device includes at least one of the following:

[0262] The sixth module is used to present the tooth segmentation results in the graphical user interface;

[0263] The seventh module is used to present the root bone scan image in a graphical user interface, wherein the root bone scan image is superimposed with the tooth segmentation result;

[0264] The eighth module is used to present a three-dimensional model of the teeth after segmentation in a graphical user interface. The three-dimensional model is generated by three-dimensional reconstruction based on the teeth segmentation results.

[0265] The tooth segmentation result is obtained by inputting the image corresponding to the region of interest of the segmented tooth into the segmentation model.

[0266] The region of interest for tooth segmentation is determined based on the first localization information of N teeth and the second localization information of M teeth. N and M are positive integers greater than 1.

[0267] The first positioning information is obtained by inputting the root bone scan image into the positioning model. The first positioning information is used to determine the position of the tooth in the root bone scan image.

[0268] The second positioning information is obtained by identifying the center information of different teeth in the root bone scan image.

[0269] The display device can also be configured based on the method for displaying pedicle features in any of the technical solutions provided below. Specifically, depending on the relationship between the steps, related steps can be implemented in the same or different modules.

[0270] Electrical or communication connections can be established between the sixth, seventh, and eighth modules to enable data transmission.

[0271] The device, or its modules or units, may be implemented by a computer chip or physical entity, or by a product with corresponding functions. While the device is described in terms of multiple modules, in some embodiments, the functions of the modules may be implemented in one or more software or hardware components.

[0272] As shown in Figure 18, one embodiment of this application provides a method for determining the characteristics of the root bone.

[0273] The application corresponding to this method can be mounted in an electronic device, a device for determining calcaneal features and / or a storage medium, can be mounted in the carrier of the method for displaying calcaneal features provided later, or can be mounted in a display device to achieve the corresponding technical effect.

[0274] Methods for determining calcaneal features may include at least one of the following steps.

[0275] Step S21: Obtain a radicular bone scan image.

[0276] A root bone scan image can be an image used to characterize bones and / or tooth roots, which can be obtained by scanning the subject.

[0277] Periatomy scans can be used to determine jawbone features. Periatomy scans can be images used to characterize the jawbone. Periatomy scans can also be used to determine tooth root features. Periatomy scans can also be images used to characterize tooth roots.

[0278] A calcaneal bone scan image can be a tomographic image. A tomographic image can be a two-dimensional image or a combination of two-dimensional images. Specifically, CT (Computed Tomography) technology can be used to scan the subject layer by layer using X-rays, obtaining multiple tomographic images corresponding to different locations. The tomographic sections can correspond to the coronal plane (from anterior to posterior), sagittal plane (from left to right), or cross-section plane (from top to bottom).

[0279] Root bone scan images can be a sequence of two-dimensional tomographic images of the dental area.

[0280] When the slice is a transverse section, a calcaneal scan image can be obtained by scanning layer by layer along the vertical direction. In this case, the calcaneal scan image can show the tissue structure at a specific transverse section in the vertical direction. The pixel or grayscale values ​​of hard tissues such as bones and teeth in the calcaneal scan image can be higher than those of soft tissues.

[0281] The calcaneal scan image can be in the form of a two-dimensional image sequence or a two-dimensional image set. The image sequence or image set can include multiple images, each of which corresponds to a tomographic scan image at a different position in the scanning direction.

[0282] Root bone scan images can be a sequence of two-dimensional tomographic images of the dental area.

[0283] Periatomy images include cone-beam computed tomography (CBCT) images. Periatomy images can be composed of cone-beam computed tomography (CBCT) images. CBCT images are obtained through cone-beam X-ray scanning and are particularly suitable for examining hard tissue structures such as the jawbone and tooth roots.

[0284] CBCT images can provide three-dimensional image information of the teeth and soft tissues of the tested object. The three-dimensional model of the teeth provides important auxiliary information, which can help doctors diagnose and formulate treatment plans.

[0285] When the calcaneal scan image is a CBCT image, the method for determining calcaneal features provided in this application can also overcome the complex situation of uneven gray-scale distribution, artifacts and low signal-to-noise ratio in CBCT images.

[0286] The calcaneal scan images in this application can also be normalized CBCT images.

[0287] Step S22: Input the root bone scan image into the positioning model to obtain the first positioning information of N teeth.

[0288] The localization model can be pre-trained to determine the positional information of the tooth in the root bone scan image, and / or determine the first localization information.

[0289] N teeth represent the teeth that the localization model can locate based on the root bone scan image. The actual number of teeth included in the root bone scan image can be greater than or equal to N. N is a positive integer greater than 1.

[0290] The standard number and distribution of teeth can be set according to the characteristics of the subject. For example, when the subject is in the deciduous dentition stage (child), there are usually 20 teeth, 5 on each side of the upper and lower jaws. The teeth on each side, from front to back, are: central incisor, lateral incisor, canine, first molar, and second molar. For example, when the subject is in the permanent dentition stage (adult), there are usually 28 to 32 teeth, 16 on each side of the upper and lower jaws, symmetrically distributed. The teeth on each side, from front to back, are: central incisor, lateral incisor, canine, first premolar, second premolar, first molar, second molar, and third molar (wisdom tooth).

[0291] The localization model may include at least one neural network. The neural network may be a deep neural network.

[0292] The first neural network localization model can be used to locate teeth one by one in a root bone scan image, determining the number and distribution of teeth. The localization information determined by the localization model can be used to determine the region of interest (ROI) of each tooth, and the image corresponding to the ROI can be cropped from the root bone scan image using a bounding box surrounding the tooth.

[0293] The localization model can be or includes a deep convolutional neural network. For example, it can be an object detection network, such as Faster-RCNN, YOLO, or RetinaNet (at least one of these). The second neural network can be a fully convolutional deep neural network, such as a medical image segmentation network, such as UNet, Atten-UNet, or UNet++ (at least one of these).

[0294] The first location information can be used to generate a three-dimensional model. In this case, the method for determining root bone features provided in this application may include the steps of: segmenting N teeth according to the first location information, performing surface reconstruction using the Marching Cubes algorithm, and obtaining a three-dimensional model corresponding to the N teeth.

[0295] The first location information and the second location information (described later) both point to at least the same or nearby test site on the same subject. This ensures that missed tooth information can be identified based on both.

[0296] The initial localization information can be used to determine the regions of interest (ROIs) for N teeth in the root bone scan image. Based on this initial localization information, images corresponding to the segmented ROIs of the N teeth can be cropped from the root bone scan image. These images can then be used for 3D reconstruction to obtain 3D models of the corresponding N teeth.

[0297] As shown in Figure 19, the regions of interest (ROIs) of different teeth can be determined in the root bone scan image set0 based on tooth localization using a localization model. The root bone scan image set0 can be a three-dimensional image. The ROIs of N teeth can be determined based on the first localization information, or the first localization information can characterize the ROIs of N teeth. From the root bone scan image set0, the corresponding image set111 for each tooth's ROI can be determined. Each tooth has a set of images set111 corresponding to its ROIs. The images corresponding to the ROIs of the N teeth of the test subject, determined by the localization model, constitute the image set set11 corresponding to the ROIs. The first localization information loc11 can include information about the bounding boxes surrounding the N teeth, as shown in Figure 19.

[0298] In one specific embodiment, the method for determining root bone features provided in this application may include the following steps: cropping the root bone scan image based on the bounding boxes of N teeth to obtain images corresponding to the regions of interest segmented from the N teeth.

[0299] This step can be included in the step: determining the region of interest for tooth segmentation based on the first localization information of N teeth and the second localization information of M teeth; when this step is implemented, the above steps can be specifically implemented.

[0300] In this specific embodiment, the localization model can be pre-trained to take a root bone scan image as input and the bounding box information surrounding each tooth as output. The bounding box information surrounding each tooth can be coordinate information; for example, it can be the coordinate information of the vertices of the bounding box, where the vertices can be diagonal vertices or all vertices of the bounding box; it can also be the coordinate information of the center point of the bounding box and the size information of the bounding box. The bounding box can be a two-dimensional bounding box or a three-dimensional bounding box.

[0301] The bounding box information can be either the first bounding box information for the corresponding tooth or the information of a second bounding box that has been expanded. The first bounding box can be the smallest rectangular bounding box surrounding a single tooth, and it can be three-dimensional or two-dimensional. The second bounding box can be a larger rectangular bounding box formed by expanding the smallest rectangular bounding box, and it can also be three-dimensional or two-dimensional.

[0302] For example, the minimum bounding box includes diagonal vertices: a first vertex v1 (which could be the top-left front vertex of the 3D rectangle) and a second vertex v2 (which could be the bottom-right back vertex of the 3D rectangle). The coordinates of the first vertex v1 are (x1, y1, z1), and the coordinates of the second vertex v2 are (x2, y2, z2). Based on this, the coordinates of the two vertices can be adjusted to expand the first bounding box into a second bounding box. The coordinates of the new first vertex v1' are (x1-margin_x, y1-margin_y, z1-margin_z), and the coordinates of the new second vertex v2' are (x2+margin_x, y2+margin_y, z2+margin_z).

[0303] The reason for adjusting the size of the bounding box to expand it is to include part of the area around a single tooth (background), which helps improve the recognition accuracy of the tooth (foreground).

[0304] The magnitude of the expansion (i.e., the aforementioned margin_x, margin_y, and margin_z) can be preset. In one embodiment, the preset expansion value is set as follows: the magnitude by which the first bounding box is expanded is a preset multiple of the corresponding side length within the first bounding box. Specifically, the preset value is 0.2.

[0305] For example, when adjusting the size of the x-coordinate direction, the expansion is equal to 0.2 times the side length of the first bounding box in the x-coordinate direction (e.g., |x2-x1|). The same applies to other coordinate directions.

[0306] For example, the default extended value can be set as follows:

[0307] margin_x = 0.2 × (x2 - x1);

[0308] margin_y = 0.2 × (y2 - y1);

[0309] margin_z = 0.2 × (z2 - z1).

[0310] The localization model can be pre-trained to output either the second bounding box or the first bounding box. For the latter, an additional step of expanding the first bounding box into the second bounding box can be performed.

[0311] Once the second bounding box is determined, it can characterize the regions of different teeth formed after the boundary expansion. Based on the second bounding box, N regions of interest can be segmented in the root bone scan image, and then the root bone scan image can be cropped to obtain the image corresponding to the region of interest.

[0312] The image corresponding to the region of interest can be a two-dimensional image sequence or a three-dimensional image.

[0313] If the region of interest (ROI) of N teeth is taken as at least one of the objects of tooth segmentation in this application, then the N teeth can be segmented after the ROI is determined. Specifically, a segmentation model can be used for segmentation.

[0314] The segmentation model may include at least one neural network. The neural network may be a deep neural network.

[0315] The segmentation model can segment the tooth based on the image corresponding to the region of interest of the tooth, and determine the morphological features of the tooth (including the crown and / or root).

[0316] The segmentation model can segment N teeth separately, and then combine the segmentation results of the N teeth to obtain a segmentation result of N teeth containing N tooth morphological features and positional distribution.

[0317] In one specific embodiment, the method for determining root bone features provided in this application may include the following steps: inputting the images corresponding to the regions of interest segmented from N teeth into a segmentation model to obtain the segmentation results of N teeth.

[0318] If the segmentation results of N teeth are used as the final tooth segmentation result obtained by determining the root bone features, then the step can be: inputting the images corresponding to the regions of interest of the N teeth into the segmentation model to obtain the tooth segmentation results.

[0319] This step can be included in the step: inputting the image corresponding to the region of interest of the teeth segmentation into the segmentation model to obtain the teeth segmentation result; when this step is implemented, the above steps can be specifically implemented.

[0320] The image set111 corresponding to the region of interest for a single tooth segmentation is input into the segmentation model for tooth segmentation, resulting in the corresponding tooth segmentation result info111. The tooth segmentation result can be a 3D image or a 2D image sequence. The tooth segmentation results info111 corresponding to N teeth of the test subject are combined to form the segmentation result info111 for N teeth.

[0321] The image corresponding to the region of interest input to the segmentation model can be a normalized image corresponding to the region of interest of each of the N teeth. For normalization, the pixel value range of the root bone scan image can be normalized to [-1,1] or [0,1] before cropping, or the pixel range of the cropped local region of interest image can be normalized to [-1,1] or [0,1] before being input to the segmentation model.

[0322] The output of the segmentation model can be a two-dimensional image, a two-dimensional image sequence, or a three-dimensional image, specifically a mask image. The mask image can be a binary image, where 1 represents the foreground and 0 represents the background.

[0323] The image corresponding to the region of interest can be a two-dimensional image. Based on this, each image in the image sequence corresponding to the region of interest can be segmented into teeth, resulting in multiple two-dimensional output images corresponding to that tooth. These multiple two-dimensional output images can then be combined in the same order as the image sequence to obtain a sequence of two-dimensional output images representing the segmentation result of that tooth. Combining N sets of two-dimensional output image sequences corresponding to N teeth yields the segmentation result for N teeth.

[0324] At this point, the segmentation model can include a two-dimensional deep fully convolutional neural network.

[0325] The image corresponding to the region of interest can be a 3D image. Based on this, tooth segmentation can be performed on the image corresponding to the region of interest to obtain a 3D output image corresponding to the segmentation result of that tooth. The N 3D output images corresponding to N teeth are combined to obtain the segmentation result of N teeth.

[0326] In this case, the segmentation model can be a three-dimensional deep fully convolutional neural network.

[0327] Step S23: Identify the center information of different teeth in the root bone scan image to obtain the second positioning information of M teeth.

[0328] M teeth represent the teeth whose corresponding positioning information can be obtained by identifying their center information. The actual number of teeth included in a root bone scan image can be equal to M. M is a positive integer greater than 1.

[0329] The second positioning information may include the center information of different teeth in the root bone scan image. The center information can be obtained by first determining the center of the tooth in each root bone scan image, and then fitting a three-dimensional structure based on multiple centers corresponding to the same tooth; or it can be obtained by directly using the center information of the corresponding tooth in the root bone scan image as the center information.

[0330] The center information of a tooth can be determined based on the geometric center or anatomical features of the tooth in the root bone scan image. It is used to indicate the core position of the tooth in the root bone scan image, which facilitates the localization, classification or comparison of teeth.

[0331] The center information can be information about the center point. The center point of a tooth can be its geometric center, centroid, or centroid, and can be determined by calculating the volume or two-dimensional boundary. The center point of a tooth can also be an anatomical feature point such as the cusp or the midpoint of the cervical line. The center point of a tooth can also be calculated and determined by annotating the tooth boundary through three-dimensional reconstruction.

[0332] Obtaining second positioning information determined by identifying the center information of teeth breaks away from the limitations of conventional tooth number or morphology, so that the obtained second positioning information can include information of all existing teeth and avoid omissions.

[0333] Center information can be determined by identifying the boundary distance information of the tooth. The reason is that boundary distance information can be used to characterize the distance between pixels within the tooth area and the tooth's boundary point, and / or the distance between pixels and the tooth's center in a root bone scan image. The closer a pixel is to the center point, the farther it is from its boundary; therefore, in these cases, center information can be indirectly characterized through boundary-related features.

[0334] The process of identifying the center information of different teeth to determine secondary positioning information can be achieved using a boundary distance regression model. Boundary distance regression models can include deep neural networks.

[0335] The boundary distance regression model can be used to output the distance from each pixel within the tooth's range to the nearest boundary point on that tooth. Based on the characteristic that "pixels closer to the tooth's center have larger distance values, and pixels closer to the tooth's boundary have smaller distance values," the center information of the tooth is determined. In some embodiments, step S23 can also be: inputting the root bone scan image into the boundary distance regression model to identify the center points of different teeth and obtain the second positioning information of M teeth.

[0336] The boundary distance regression model can include a deep convolutional neural network. For example, it can be a fully convolutional deep neural network. For example, it can be at least one of UNet, Atten-UNet, and UNet++.

[0337] The second location information can be a two-dimensional image, a two-dimensional image sequence, or a three-dimensional image. In an embodiment where the second location information is a three-dimensional image, the three-dimensional image can be determined based on an image that can characterize the boundary distances of pixels within the tooth area, using a region growing algorithm with the center point of the tooth as the seed point to determine the three-dimensional mask of the tooth, thereby determining the second location information in the form of a three-dimensional image.

[0338] In embodiments where the second positioning information is a two-dimensional image or a sequence of two-dimensional images, the second positioning information may be a boundary distance image that can characterize the boundary distance of points on the tooth, or a center mask image that can characterize the center point of the tooth.

[0339] In one embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: determining the boundary information of the teeth based on the root bone scan image; and determining the boundary distance information of the teeth based on the shape and boundary information of the teeth.

[0340] This step can be included in step S23; when implementing step S23, this step can be implemented in detail.

[0341] Boundary distance information can be used to determine the second positioning information of the M teeth. The step of identifying center information in step S23 may include or include the step of determining boundary distance information.

[0342] Boundary distance information is used to characterize the distance between pixels within the tooth region and the tooth's boundary points in a root bone scan image, and / or the distance between pixels within the tooth region and the tooth's center. This boundary distance can be characterized by pixel values, grayscale values, colors, or numerical values. Based on the principle that "points closer to the tooth's center have larger distance values, and points closer to the tooth's boundary points have smaller distance values," center information can be determined based on boundary distance information, or boundary distance information can be determined based on the tooth's center information.

[0343] A boundary distance regression model can be pre-trained to take a root bone scan image as input and tooth boundary distance information as output. This model can be used to detect tooth morphological features, determine the tooth's extent, and then determine the boundary distances of pixels within that tooth's extent based on the boundary information. The root bone scan image can be a CBCT image, a two-dimensional image sequence, or a normalized image. The tooth boundary distance information can be in image form or a normalized image.

[0344] As shown in Figure 20, the root bone scan image fig0 is included in the root bone scan image sequence set0. Based on the root bone scan image fig0, the boundary distance information of the tooth can be determined, resulting in a boundary distance image fig21. In one embodiment, the boundary distance information of the tooth can be represented by the boundary distance image fig21. In this case, the boundary distance information can be represented by the grayscale of the pixels. For example, in the boundary distance image fig21, pixels closer to the center of the tooth within the tooth area are brighter, corresponding to larger grayscale values; pixels closer to the boundary of the tooth (away from the center) within the tooth area are darker, corresponding to smaller grayscale values.

[0345] In this embodiment, the method for determining root bone features provided in this application may include the step of: determining a boundary distance image of the teeth as boundary distance information based on a root bone scan image. In the boundary distance image, the boundary distance information is represented by the grayscale of pixels. The boundary distance information can be used to identify center information. This step may be included in the step "Identifying the center information of different teeth in the root bone scan image".

[0346] Boundary distance regression models can be used to determine the boundary distance images of teeth based on root bone scan images. The boundary distance regression model can be pre-trained to take a root bone scan image (fig0) as input and output the boundary distance image (fig21) corresponding to the tomographic location pointed to by fig0. The boundary distance regression model is used to determine boundary distances.

[0347] The root bone scan image can be a two-dimensional image. In this case, the boundary distance regression model can include a two-dimensional deep fully convolutional neural network. Boundary distance images corresponding to the same tooth are combined to represent the boundary distance information of that tooth, while boundary distance information corresponding to different teeth is used to determine the secondary localization information of the teeth.

[0348] The root bone scan image can be a three-dimensional image. In this case, the boundary distance regression model can include a three-dimensional deep fully convolutional neural network. The three-dimensional boundary distance image corresponding to the same tooth represents the boundary distance information of that tooth, while the boundary distance information corresponding to different teeth is used to determine the secondary localization information of the teeth.

[0349] The second localization information may include a central mask image. The central mask image is used to characterize the center positions of the M teeth. The central mask image can be used directly as the second localization information, or the second localization information can be determined based on the central mask image.

[0350] In one embodiment, the center mask image can be determined based on the boundary distance image. As shown in Figure 20, in the center mask image fig22, pixels with boundary distance information less than a preset threshold have a first grayscale value. For example, in the center mask image fig22, pixels on a tooth that are far from the center of the tooth and have boundary distance information less than the preset threshold are set to the first grayscale value; correspondingly, pixels on a tooth that are close to the center of the tooth and have a boundary distance greater than the preset threshold can be set to a second grayscale value. The first grayscale value is 0, and the second grayscale value is 1. The center mask image fig22 can be determined based on the boundary distance image fig21 using a threshold segmentation method.

[0351] In this embodiment, the method for determining root bone features provided in this application may include the steps of: determining boundary distance information based on tooth boundary information; determining boundary distance information based on tooth shape and boundary information; and determining a center mask image of the tooth based on the tooth boundary distance information.

[0352] Boundary distance information is used to characterize the distance between pixels within the tooth region in a root bone scan image and the boundary points of the tooth, and / or the distance between pixels within the tooth region in a root bone scan image and the center of the tooth.

[0353] In the central mask image, pixels with boundary distance information less than a preset threshold have a first gray value. Pixels with boundary distance information greater than the preset threshold have a second gray value. The central mask image is used to represent the center point of the tooth. This step may be included in the step of "identifying the center information of different teeth in the root bone scan image to obtain the second positioning information of M teeth".

[0354] A neural network can be used to determine the central mask image of the tooth based on the root bone scan image.

[0355] In one embodiment, the threshold segmentation method may be: set a preset threshold T = 0.8, and judge the numerical relationship between the gray value I(x,y) of the pixel in the boundary distance image (x,y) of the tooth and the preset threshold T. When I(x,y) < T, set I(x,y) = 0, and when I(x,y) ≥ T, set I(x,y) = 1.

[0356] The boundary distance image of the tooth may be a grayscale image with gray values distributed between 0 and 255. The central mask image of the tooth may be a binary image.

[0357] The central mask image fig22 is used to determine the second positioning information. In one embodiment, the central mask image fig22 serves as the second positioning information.

[0358] The boundary distance information can also be used to determine the image of the region of interest of the first tooth. The first tooth may be all or part of the M teeth included in the boundary distance information. In one embodiment, the method for determining the root bone characteristics provided in this application may include the following steps: according to the boundary distance information, determine the image corresponding to the region of interest of the first tooth segmentation in the root bone scan image.

[0359] This step may be included in determining the region of interest of the tooth segmentation based on the first positioning information of N teeth and the second positioning information of M teeth; when this step is implemented, the above steps may be specifically implemented.

[0360] Step S24, determine the region of interest of the tooth segmentation according to the first positioning information of N teeth and the second positioning information of M teeth.

[0361] Step S25, input the image corresponding to the region of interest of the tooth segmentation into the segmentation model to obtain the tooth segmentation result.

[0362] N and M may be the same or different. If they are the same, it means that the number of teeth determined by the two positioning information is the same, and the tooth segmentation result can be obtained based on one of them. Preferably, use the first positioning information M to determine the region of interest of the tooth segmentation, input the corresponding image into the segmentation model, and obtain the tooth segmentation result.

[0363] In one embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: determining N regions of interest for tooth segmentation based on first positioning information; inputting the images corresponding to the N regions of interest for tooth segmentation into a segmentation model to obtain tooth segmentation results.

[0364] At this point, the segmentation results of the teeth are shown in Figures 19, 21, and 22 (info12 for the segmentation of N teeth).

[0365] If N and M are different, it indicates that one of the two localization information sets is missing, or that there is a discrepancy between them. In this case, they can be referenced and supplemented. For example, based on M teeth and N teeth, the differing teeth can be identified, and the region of interest for tooth segmentation can be determined based on the differing teeth. The corresponding image is then input into the segmentation model to obtain the tooth segmentation result.

[0366] The segmentation model may include at least one neural network. The neural network may be a deep neural network. The segmentation model may be configured as described above. The segmentation model for segmenting the image corresponding to the region of interest of N teeth, and the segmentation model for segmenting the image corresponding to the region of interest of the teeth segmented based on two localization information, may be one or two separate models.

[0367] In one embodiment, the region of interest for tooth segmentation includes regions of interest corresponding to differential teeth, which are determined based on M teeth and N teeth.

[0368] Based on this, in the step of determining the region of interest corresponding to the first tooth segmentation in the root bone scan image, the first tooth may include the difference between N teeth and M teeth.

[0369] A differential tooth can be a tooth included in the second positioning information but not included in the first positioning information. In this case, N can be less than M. For a tooth, if the second positioning information includes the center information of the tooth, but the first positioning information does not include the position information of the tooth, then the tooth can be identified as a differential tooth or a first tooth.

[0370] Difference teeth reflect the limitations of the first positioning information determined based on location. Difference teeth can be teeth that were missed by the first positioning information, or they can be supernumerary teeth, impacted teeth, etc.

[0371] The image corresponding to the region of interest (ROI) of the differentially segmented teeth can be input into the segmentation model to obtain the segmentation results of the differentially segmented teeth. The segmentation results of the differentially segmented teeth can be in the form of data information such as position coordinates, in the form of two-dimensional images, or in the form of three-dimensional models.

[0372] Determining the difference between N teeth and M teeth can be achieved by checking if a tooth included in the second positioning information exists in the first positioning information, and then assigning pixel values ​​to teeth that meet the criteria for a difference. Specifically, if a tooth included in the second positioning information exists in the first positioning information, its pixel value is assigned to 0; if a tooth included in the second positioning information does not exist in the first positioning information, its pixel value is assigned to the value found in the second positioning information.

[0373] Abstracting this mathematically, it can be understood as selecting teeth from the second localization information that are not present in the first localization information. For example, using the center mask image as the second localization information, assuming the center mask of M teeth is C, assuming the first localization information is S1, and assuming the pixel value of the corresponding difference tooth is Cm, then the pixel value Cm(x,y) at (x,y) of the pixel value Cm of the corresponding difference tooth satisfies:

[0374] C(x,y) is the pixel value at (x,y) in the second positioning information, and S1(x,y) is the pixel value at (x,y) in the first positioning information. S1(x,y)>0 indicates that the pixel at (x,y) belongs to the foreground teeth; S1(x,y)=0 indicates that the pixel at (x,y) belongs to the background area.

[0375] Based on this, we can obtain the set of pixel values ​​Cm of all the differential teeth in the second positioning information. Cm can be the set of center masks of all the differential teeth. Each connected component in Cm is the center mask of the differential tooth, as shown in Figure 21.

[0376] In one embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: determining regions of interest for N tooth segments based on first positioning information; and determining regions of interest for differentially segmented teeth based on the first positioning information and second positioning information. This step may be included in step S24; when implementing step S24, this step may be specifically implemented.

[0377] In one embodiment, N teeth and differentially segmented teeth can be segmented together. In this embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: determining regions of interest (ROIs) for the N teeth segmentation based on first localization information; inputting the images corresponding to the ROIs for the N teeth segmentation and the images corresponding to the ROIs for the differentially segmented teeth into a segmentation model to obtain tooth segmentation results.

[0378] In one embodiment, the N teeth and the differentially segmented teeth can be segmented separately, and then the two segmentation results can be combined to obtain the final tooth segmentation result. In this embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: inputting the image corresponding to the region of interest of the N teeth segmentation into a segmentation model to obtain the segmentation result of the N teeth; inputting the image corresponding to the region of interest of the differentially segmented teeth into a segmentation model to obtain the segmentation result of the differentially segmented teeth; and obtaining the final tooth segmentation result based on the segmentation result of the N teeth and the segmentation result of the differentially segmented teeth. This step may be included in step S25; when implementing step S25, this step can be specifically implemented.

[0379] The region of interest for segmenting differential teeth can be the area where the differential teeth are located in the root bone scan image (after cropping, the image corresponding to the region of interest for segmenting differential teeth can be obtained).

[0380] In one specific embodiment, the method for determining root bone features provided in this application may include the steps of: determining second positioning information of the differential tooth in second positioning information; and determining the image corresponding to the region of interest of the differential tooth in the root bone scan image based on boundary distance information and the second positioning information of the differential tooth. The boundary distance information is used to characterize the distance between pixels within the tooth area in the root bone scan image and the boundary point of the tooth, and / or the distance between pixels within the tooth area in the root bone scan image and the center of the tooth.

[0381] This step can be included in step S24; when implementing step S24, this step can be implemented in detail.

[0382] To identify the region of interest for differentially expressed teeth in root bone scan images, region growing can be employed.

[0383] The second localization information used to determine the difference in teeth can be the second localization information itself, or the center reconstruction result of M teeth reconstructed based on the second localization information.

[0384] In one embodiment, the method for determining root bone features provided in this application may include at least one of the following steps: determining regions of interest (ROIs) for N teeth segmentation based on first positioning information; inputting the images corresponding to the ROIs of the N teeth segmentation into a segmentation model to obtain segmentation results for the N teeth; performing three-dimensional reconstruction based on second positioning information to obtain center reconstruction results for M teeth; comparing the segmentation results for the N teeth with the center reconstruction results for the M teeth to determine the teeth with differences. This step may be included in step S24; when implementing step S24, this step may be specifically implemented.

[0385] For example, as shown in Figure 19, after obtaining the first localization information loc11, the image set111 corresponding to the region of interest for N teeth segmentation in the root bone scan image can be determined based on the first localization information, and then input into the segmentation model to obtain the segmentation result info12 for the N teeth. As shown in Figure 20, after obtaining the second localization information in the form of the center mask image fig22, three-dimensional reconstruction can be performed based on the second localization information to obtain the center reconstruction result info2 for M teeth. As shown in Figure 21, the segmentation result info12 for the N teeth can be compared with the center reconstruction result info2 for the M teeth to obtain the center reconstruction result info3 for the differential teeth as the second localization information for the differential teeth, thus identifying the differential teeth.

[0386] In other embodiments, the difference teeth can also be determined by comparing the first positioning information loc11 with the central mask image fig22, and the central mask image of the difference teeth can be used as the second positioning information of the difference teeth.

[0387] As mentioned above, one embodiment of this application further includes the steps of: determining the image corresponding to the region of interest of the first tooth segmentation in the root bone scan image based on the boundary distance information; and inputting the image corresponding to the region of interest of the tooth segmentation into the segmentation model to obtain the tooth segmentation result.

[0388] When the first tooth includes a differential tooth, the tooth segmentation result shall include at least the segmentation result of the differential tooth.

[0389] As shown in Figure 22, after determining the center reconstruction result info3 of the differential tooth as the second localization information of the differential tooth, the position of the differential tooth can be determined in the boundary distance image (as boundary distance information, it is included in the boundary distance image sequence set21), and the image corresponding to the region of interest of the differential tooth segmentation can be determined in the root bone scan image based on its position in the boundary distance image.

[0390] When the boundary distance image is in the form of a two-dimensional image, the central mask image or central reconstruction result corresponding to the differential tooth can be determined first, and the boundary distance image corresponding to the differential tooth can be determined accordingly. In the boundary distance image corresponding to the differential tooth, the center point of the differential tooth can be used as the seed point to segment it using a two-dimensional region growing algorithm to obtain the two-dimensional distribution region of the differential tooth in the boundary distance image (which can be represented in the form of a two-dimensional mask image). Multiple boundary distance images corresponding to the differential tooth contain multiple two-dimensional distribution regions, which can be combined to obtain the three-dimensional distribution region of the corresponding differential tooth. Based on this three-dimensional distribution region, the local region of interest image can be determined in the root bone scan image.

[0391] When the boundary distance image has the form of a three-dimensional image, the central mask image or central reconstruction result corresponding to the differential teeth can be determined first. Then, the center point of the differential teeth can be used as the seed point to segment the region using a three-dimensional region growing algorithm to obtain the three-dimensional region distribution of the corresponding differential teeth, thereby determining the local region of interest image.

[0392] The images corresponding to the regions of interest (ROIs) of differentially segmented teeth are cropped from the root bone scan images based on the ROIs of differentially segmented teeth. The root bone scan images constitute the root bone scan image sequence set0, and the cropped ROI images constitute the image sequence set3 corresponding to the ROIs of differentially segmented teeth. The number of images corresponding to the ROIs in the image sequence set3 is less than or equal to the number of root bone scan images in the image sequence set0. Each image corresponding to the ROI of differentially segmented teeth has a corresponding root bone scan image, and each image has a corresponding tomographic region of the subject.

[0393] The region of interest corresponding to the segmentation of differential teeth can be represented by a bounding box surrounding the differential teeth. The bounding box can be a two-dimensional bounding box or a three-dimensional bounding box.

[0394] In one specific embodiment, the method for determining root bone features provided in this application may include the steps of: determining the region of interest for segmented differential teeth and obtaining a third bounding box for the corresponding differential teeth; adjusting the size of the third bounding box according to a preset expansion value to obtain a fourth bounding box; and determining the region of interest for segmented differential teeth in the root bone scan image according to the fourth bounding box.

[0395] The region of interest for the differential tooth segmentation can be either a two-dimensional or three-dimensional distribution region as described above. Adjusting the size of the third bounding box can involve expanding it to obtain a fourth bounding box with a larger volume or area. When the third bounding box is a two-dimensional rectangle, it can be determined based on the coordinates of its diagonal vertices. Adjusting the size of the third bounding box can be achieved by adjusting the coordinates of at least one of its two diagonal vertices. When the third bounding box is a three-dimensional rectangle, it can be determined based on the coordinates of its diagonal vertices. Adjusting the size of the third bounding box can be achieved by adjusting the coordinates of at least one of its two diagonal vertices.

[0396] For example, consider the third vertex v3 (which could be the top-left front vertex of the 3D rectangle enclosing the teeth with differences) and the fourth vertex v4 (which could be the bottom-right back vertex of the 3D rectangle enclosing the teeth with differences). The coordinates of the third vertex v3 are (x3, y3, z3), and the coordinates of the fourth vertex v4 are (x4, y4, z4). Based on this, the coordinates of the two vertices can be adjusted to expand the third bounding box into a fourth bounding box. The coordinates of the new third vertex v3' are (x3-margin_x, y3-margin_y, z3-margin_z), and the coordinates of the new fourth vertex v4' are (x4+margin_x, y4+margin_y, z4+margin_z).

[0397] The reason for adjusting the size of the bounding box to expand it is to include part of the area around the differential teeth (background), which helps improve the recognition accuracy of differential teeth (foreground).

[0398] The magnitude of the expansion (i.e., the aforementioned margin_x, margin_y, and margin_z) can be preset. In one embodiment, the preset expansion value is set as follows: the magnitude by which the third bounding box is expanded is a preset multiple of the corresponding side length within the third bounding box. Specifically, the preset value is 0.2.

[0399] For example, when adjusting the size of the x-coordinate direction, the expansion is equal to 0.2 times the side length of the third bounding box in the x-coordinate direction (e.g., |x4-x3|). The same applies to other coordinate directions.

[0400] For example, the default extended value can be set as follows:

[0401] margin_x = 0.2 × (x4 - x3);

[0402] margin_y = 0.2 × (y4 - y3);

[0403] margin_z = 0.2 × (z4 - z3).

[0404] Once the fourth bounding box is determined, it can characterize the region of the differentially defined teeth formed after the boundary expansion. Based on the fourth bounding box, the region of the differentially defined teeth can be identified in the root bone scan image. Then, based on the identified region of interest, the root bone scan image can be cropped to obtain the image corresponding to the segmented region of interest of the differentially defined teeth.

[0405] Based on the boundary distance information and the second localization information of the differential teeth, after determining the image corresponding to the region of interest for segmentation of the differential teeth in the root bone scan image, tooth segmentation can be performed based on the image (e.g., inputting the image into the segmentation model) to obtain the segmentation result 3d-3 of the differential teeth.

[0406] As shown in Figure 22, the segmentation result 3d-3 of the differentially segmented teeth can be merged with the segmentation result info12 of the N teeth to obtain the final tooth segmentation result 3d-0. The segmentation result 3d-3 of the differentially segmented teeth, the segmentation result info12 of the N teeth, and the final tooth segmentation result 3d-0 can be two-dimensional images, three-dimensional images, or a three-dimensional model generated based on three-dimensional reconstruction of two-dimensional images.

[0407] In this application, a segmentation model can be used for tooth segmentation to obtain segmentation results of 3d-3 for teeth with differences. The segmentation model may include at least one neural network, and the segmentation model can be configured as described above. The segmentation model used to segment the image corresponding to the region of interest of N teeth and the segmentation model used to segment the image corresponding to the region of interest of the teeth with differences can be one or two separate models.

[0408] The segmentation model can be pre-trained to take as input an image the region of interest (ROI) corresponding to the segmented teeth with differences in morphological features, and as output a two-dimensional image. The ROI input to the segmentation model can be a normalized image corresponding to the ROI of the teeth with differences in morphology. If the output of the segmentation model is a two-dimensional image, this image can specifically be a mask image. The mask image can be a binary image, where 1 represents the foreground and 0 represents the background.

[0409] The region of interest (ROI) image can be a two-dimensional image. A sequence of ROI images for segmenting differentially segmented teeth is input into a segmentation model. Multiple processed two-dimensional images can be combined to obtain a two-dimensional output image sequence. In this case, the segmentation model can include a two-dimensional deep fully convolutional neural network. The two-dimensional output image sequence can be used to obtain the segmentation results for differentially segmented teeth. Alternatively, the ROI image can be a three-dimensional image. The ROI images for segmenting differentially segmented teeth are input into a segmentation model. After processing, the segmentation results for differentially segmented teeth are obtained. In this case, the segmentation model can include a three-dimensional deep fully convolutional neural network. The neural network can be at least one of UNet, Atten-UNet, or UNet++.

[0410] In summary, the method for determining root bone features provided in this application improves the accuracy of tooth localization and detection by determining the region of interest (ROI) for tooth segmentation based on two types of tooth localization information. When using the obtained ROI for tooth segmentation, it achieves complete and accurate tooth segmentation, avoiding the omission of special cases such as impacted teeth and supernumerary teeth. One type of localization information is used to determine the position of the tooth in the root bone scan image, enabling the tooth segmentation result to have the accuracy of single-tooth segmentation for conventional teeth. The other type of localization information is obtained by identifying the center information of the root bone scan image, ensuring that the tooth segmentation result includes all teeth of the subject without omission.

[0411] Methods for displaying calcaneal features

[0412] As shown in Figure 23, this application provides a method for displaying radicular bone features.

[0413] The application or instructions corresponding to this method can be mounted on an electronic device, a device for determining calcaneal features and / or a storage medium, or on a carrier for implementing the method of displaying calcaneal features, so as to achieve the corresponding technical effect.

[0414] The method for displaying calcaneal features may specifically include at least one of the following steps.

[0415] Step S31: Present the jawbone feature image in the graphical user interface.

[0416] Jawbone feature images can be three-dimensional models or two-dimensional images.

[0417] When the jawbone feature image is a three-dimensional model, it can be a three-dimensional model of the jawbone features corresponding to the maxilla, as shown in Figure 11 (3d-31), or a three-dimensional model of the jawbone features corresponding to the mandible, as shown in Figure 12 (3d-32), or a combination of the three-dimensional models corresponding to the maxilla and mandible.

[0418] When the jawbone feature image is a three-dimensional model, it can also be a three-dimensional model formed by combining multiple jawbone feature images in the jawbone feature image sequence set3 in Figure 11 or Figure 12.

[0419] When the jawbone feature image is a two-dimensional image, it can be one or more jawbone feature images from the jawbone feature image sequence set3 in Figure 11 or Figure 12. The jawbone feature image can be a masked image, especially a binary image.

[0420] The presented jawbone feature image can be the image itself, the information contained in the image, or a combination of the image itself and the information. The jawbone feature image may include portions indicating the shape of the jawbone, and / or portions indicating other feature information of the corresponding jawbone; other feature information may be size information, etc.

[0421] Jawbone feature images can be determined by a neural network based on both root bone scan images representing the same jawbone location and jawbone segmentation images as input.

[0422] Cancellous bone feature images can be used to determine cortical bone feature images based on the differences between cancellous bone feature images and jawbone feature images.

[0423] When performing step S31, the corresponding root bone scan image, jaw segmentation image, cancellous bone feature image, or cortical bone feature image is not necessarily displayed.

[0424] Jawbone feature images can also be presented together with other two-dimensional images or three-dimensional models. Two-dimensional images or three-dimensional models are used to present more complete structural information within the oral cavity. In this case, jawbone feature images can be presented as part of the two-dimensional image or the three-dimensional model to indicate the features of the jawbone region in the two-dimensional image or the three-dimensional model.

[0425] The other three-dimensional model can be an oral scan model. The oral scan model and the jawbone feature image correspond to the same test object.

[0426] In the two-dimensional image or the three-dimensional model, the jawbone feature image can be distinguished from other parts by changing the color, edge lines, strobe, etc., so as to indicate the features of the jawbone in the two-dimensional image or the three-dimensional model.

[0427] In one embodiment, the method for displaying calcaneal features provided in this application may include the steps of: presenting an oral scan model and a jawbone feature image corresponding to the oral scan model in a graphical user interface.

[0428] Jawbone feature images are used to characterize the jawbone region. Specifically, the jawbone feature images can be displayed at the locations corresponding to the intraoral scan model.

[0429] This step can be included in step S31; when implementing step S31, this step can be implemented in detail.

[0430] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the maxillary region in a graphical user interface. This step may be included in step S31; when implementing step S31, this step may be specifically implemented.

[0431] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the mandibular region in a graphical user interface. This step may be included in step S31; when implementing step S31, this step may be specifically implemented.

[0432] The selection box can be a two-dimensional selection box or a three-dimensional selection box.

[0433] Referring to Figure 16, the two-dimensional selection box can be shown as the rectangular box in the original jawbone segmentation image in the original jawbone segmentation image sequence set022. The three-dimensional selection box can be shown as the first selection box corresponding to diagonal vertices v1 and v2, or as the second selection box corresponding to diagonal vertices v1' and v2'.

[0434] The selection box can be determined based on the region representing the jawbone area (e.g., the maxilla or mandible) in the segmented jawbone image. The jawbone segmentation image can be determined based on a root bone scan image. Jawbone segmentation can be performed using a neural network.

[0435] When the graphical user interface displays a jawbone feature image, if the jawbone feature image corresponds to either the maxilla or the mandible, the presented selection box will surround the area where the jawbone part of the image is located; if the jawbone feature image includes both the maxilla and the mandible, the presented selection box is equivalent to dividing the jawbone part into the maxilla and the mandible.

[0436] Step S32: Present the bone spongy features image in the graphical user interface.

[0437] Images of cancellous bone features can be three-dimensional models or two-dimensional images.

[0438] When the bone cancellous feature image is a three-dimensional model, it can be a three-dimensional model of bone cancellous features corresponding to the maxilla as shown in Figure 13, or a three-dimensional model of bone cancellous features corresponding to the mandible as shown in Figure 14, or a combination of three-dimensional models corresponding to the maxilla and mandible.

[0439] When the cancellous bone feature image is a three-dimensional model, it can also be a three-dimensional model formed by combining multiple cancellous bone feature images in the cancellous bone feature image sequence set4 in Figure 13 or Figure 14.

[0440] When the cancellous bone feature image is a two-dimensional image, it can be one or more cancellous bone feature images from the cancellous bone feature image sequence set4 in Figure 13 or Figure 14. The cancellous bone feature image can be a masked image, especially a binary image.

[0441] The presented images of cancellous bone features can be information contained in the image, the image itself, or a combination of image and information.

[0442] The feature image of cancellous bone can be determined by a neural network based on both a root bone scan image representing the same jawbone location and a segmented jawbone image as input.

[0443] Cancellous bone feature images can be used to determine cortical bone feature images based on the differences between cancellous bone feature images and jawbone feature images.

[0444] When performing step S32, the corresponding root bone scan image, jaw segmentation image, jaw feature image, or bone cortex feature image is not necessarily displayed.

[0445] Images of cancellous bone features can also be presented together with other two-dimensional images or three-dimensional models.

[0446] The other three-dimensional model can be an oral scan model.

[0447] In one embodiment, the method for displaying calcaneal features provided in this application may include the steps of: presenting an oral scan model and a corresponding cancellous bone feature image in a graphical user interface.

[0448] The cancellous bone feature image is used to characterize the cancellous bone region. Specifically, the cancellous bone feature image can be displayed at the location corresponding to the intraoral scan model.

[0449] This step can be included in step S32; when implementing step S32, this step can be implemented in detail.

[0450] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the maxillary region in a graphical user interface. This step may be included in step S32; when implementing step S32, this step may be specifically implemented.

[0451] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the mandibular region in a graphical user interface. This step may be included in step S32; when implementing step S32, this step may be specifically implemented.

[0452] The selection box can be a two-dimensional selection box or a three-dimensional selection box.

[0453] The selection box can be determined based on the region representing the jawbone area (e.g., the maxilla or mandible) in the segmented jawbone image. The jawbone segmentation image can be determined based on a root bone scan image. Jawbone segmentation can be performed using a neural network.

[0454] Step S33: Present the cortical bone feature image in the graphical user interface.

[0455] Bone cortical feature images can be three-dimensional models or two-dimensional images.

[0456] When the cortical bone feature image is a three-dimensional model, it can be a three-dimensional model of the cortical bone features corresponding to the maxilla, a three-dimensional model of the cortical bone features corresponding to the mandible, or a combination of corresponding three-dimensional models of the maxilla and mandible.

[0457] When a bone cortical feature image is a three-dimensional model, it can also be a three-dimensional model formed by combining multiple bone cortical feature images in a sequence of bone cortical feature images.

[0458] When the cortical bone feature image is a two-dimensional image, it can be a single or multiple cortical bone feature images in a sequence. The cortical bone feature image can be a masked image, and particularly a binary image.

[0459] The presented images of bone cortical features can be information contained in the image, the image itself, or a combination of image and information.

[0460] Cortical bone feature images can be determined by comparing the differences between jawbone feature images and cancellous bone feature images.

[0461] Jawbone feature images can be determined by a neural network based on both root bone scan images representing the same jawbone location and jawbone segmentation images as input.

[0462] The feature image of cancellous bone can be determined by a neural network based on both a root bone scan image representing the same jawbone location and a segmented jawbone image as input.

[0463] When performing step S32, the corresponding root bone scan image, jaw segmentation image, jaw feature image, or cancellous bone feature image is not necessarily displayed.

[0464] Bone cortical feature images can also be presented together with other two-dimensional images or three-dimensional models.

[0465] The other three-dimensional model can be an oral scan model.

[0466] In one embodiment, the method for displaying calcaneal features provided in this application may include the steps of: presenting an oral scan model and a cortical bone feature image corresponding to the oral scan model in a graphical user interface.

[0467] Cortical bone feature images are used to characterize the cortical bone region. Specifically, the cortical bone feature images can be displayed at the locations corresponding to the intraoral scan model.

[0468] This step can be included in step S33; when implementing step S33, this step can be implemented in detail.

[0469] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the maxillary region in a graphical user interface. This step may be included in step S33; when implementing step S33, this step may be specifically implemented.

[0470] In one embodiment, the method for displaying calcaneal features provided in this application may include the step of presenting a selection box surrounding the mandibular region in a graphical user interface. This step may be included in step S33; when implementing step S33, this step may be specifically implemented.

[0471] The selection box can be a two-dimensional selection box or a three-dimensional selection box.

[0472] The selection box can be determined based on the region representing the jawbone area (e.g., the maxilla or mandible) in the segmented jawbone image. The jawbone segmentation image can be determined based on a root bone scan image. Jawbone segmentation can be performed using a neural network.

[0473] The parts of the method for displaying calcaneal features provided in this application that are not fully described can be implemented as corresponding parts in other technical solutions of this application, such as the method for determining calcaneal features. Similarly, the various technical solutions, implementation methods, and embodiments provided in this application can be combined with or explained with each other.

[0474] As shown in Figure 24, this application provides a method for displaying radicular bone features.

[0475] The application or instructions corresponding to this method can be mounted on an electronic device, a device for determining calcaneal features and / or a storage medium, or on a carrier for implementing the method of displaying calcaneal features, so as to achieve the corresponding technical effect.

[0476] The method for displaying calcaneal features may specifically include at least one of the following steps.

[0477] Step S41: Present the final tooth segmentation result on the graphical user interface.

[0478] The tooth segmentation result is obtained by inputting the image corresponding to the region of interest of the segmented tooth into the segmentation model. The final tooth segmentation result can be a two-dimensional image, a three-dimensional model, or a three-dimensional image; for example, in Figure 22, the final tooth segmentation result is 3d-0.

[0479] The region of interest for tooth segmentation is determined based on the first localization information of N teeth and the second localization information of M teeth. N and M are positive integers greater than 1.

[0480] The first positioning information is obtained by inputting the root bone scan image into the positioning model. The first positioning information is used to determine the position information of the tooth in the root bone scan image.

[0481] The second positioning information is obtained by identifying the center information of different teeth in the root bone scan image.

[0482] The second localization information can be a center mask image. The center mask image can be determined based on the boundary distance information, which can be represented by a boundary distance image.

[0483] The first and second location information point to the same object being measured.

[0484] The first localization information includes the localization information of N teeth, and the second localization information includes the localization information of M teeth. The N teeth and M teeth may include teeth that differ from each other. The region of interest for tooth segmentation may include the regions of interest of the differing teeth.

[0485] When performing step S41, the corresponding first positioning information, second positioning information, segmentation results of N teeth, segmentation results of M teeth, or segmentation results of differential teeth are not necessarily displayed.

[0486] In one embodiment, the method for displaying radicular features provided in this application may include the step of: presenting a bounding box surrounding each tooth in a graphical user interface. This step may be included in step S41; when implementing step S41, this step may be specifically implemented.

[0487] The bounding box can be a two-dimensional or three-dimensional bounding box. The bounding box is determined by a neural network based on the location of different teeth in the root bone scan image.

[0488] Step S42: Present the radicular bone scan image in the graphical user interface.

[0489] The calcaneal scan image can be a CBCT image.

[0490] The root bone scan image is overlaid with the final tooth segmentation result.

[0491] The features related to the final tooth segmentation result can be implemented by referring to the description corresponding to step S41, or by referring to the method for determining root bone features in any of the preceding embodiments.

[0492] In one embodiment, the method for displaying radicular features provided in this application may include the step of: presenting a bounding box surrounding each tooth in a graphical user interface. This step may be included in step S42; when implementing step S42, this step may be specifically implemented.

[0493] The bounding box can be a two-dimensional or three-dimensional bounding box. The bounding box is determined by a neural network based on the location of different teeth in the root bone scan image.

[0494] Step S43: Present the 3D model of the teeth segmented in the graphical user interface.

[0495] The three-dimensional model is generated by three-dimensional reconstruction based on the final tooth segmentation results.

[0496] The features related to the final tooth segmentation result can be implemented by referring to the description corresponding to step S41, or by referring to the method for determining root bone features in any of the preceding embodiments.

[0497] In one embodiment, the method for displaying radicular features provided in this application may include the step of presenting a bounding box surrounding each tooth in a graphical user interface. This step may be included in step S43; when implementing step S43, this step may be specifically implemented.

[0498] The bounding box can be a two-dimensional or three-dimensional bounding box. The bounding box is determined by a neural network based on the location of different teeth in the root bone scan image.

[0499] The parts of the method for displaying calcaneal features provided in this application that are not fully described can be implemented as corresponding parts in other technical solutions of this application, such as the method for determining calcaneal features. Similarly, the various technical solutions, implementation methods, and embodiments provided in this application can be combined with or explained with each other.

[0500] As shown in Figure 25, one embodiment of this application provides a device 200 for determining pedicle bone characteristics.

[0501] The device 200 for determining calcaneal features includes at least one of the following components:

[0502] The first module 21 is used to obtain pedicle scan images;

[0503] The second module 22 is used to segment the jawbone based on the root bone scan image and determine the segmented jawbone image;

[0504] The third module 23 is used to determine the jawbone feature image and the cancellous bone feature image by using a neural network with both the root bone scan image and the jawbone segmentation image representing the same jawbone location as input.

[0505] Module 4, 24, is used to compare jawbone feature images and cancellous bone feature images, and determine cortical bone feature images based on the differences.

[0506] Module 5, 25, is used to determine and output cortical bone feature information based on the cortical bone feature image.

[0507] The first module 21 can obtain pedicle scan images acquired by external devices, or it can perform acquisition operations itself to obtain pedicle scan images.

[0508] In one embodiment, the apparatus 200 for determining calcaneal features may also be connected to the first device 101. In another embodiment, the method for determining calcaneal features may be implemented by connecting the first device 101.

[0509] The first device 101 can be disposed outside the device 200 for determining calcaneal features, serving as an external device to the device 200 for determining calcaneal features. The first device 101 and the device 200 for determining calcaneal features can also be integrated together.

[0510] The first device 101 is used to perform scanning to obtain radicular bone scanning information. The radicular bone scanning information can be the radicular bone scanning image or data information related to the radicular bone scanning image. The first device 101 can scan the jawbone area or perform an oral cavity scan.

[0511] The first device 101 can be used to implement a cone-beam annular hood to obtain calcaneal scan images. The calcaneal scan images can be CBCT (cone beam computed tomography) images.

[0512] The imaging process of CBCT images involves scanning a target volume area with cone-beam X-rays, followed by a single rotation to obtain three-dimensional data. CBCT images can display the structure of teeth, bones, and soft tissues, with high pixel or grayscale values, making them suitable for imaging smaller anatomical areas. CBCT images offer advantages such as high accuracy, ease of assessment of tooth roots and bone, convenient location of impacted teeth, and convenient temporomandibular joint assessment.

[0513] CBCT images can include multiple two-dimensional images corresponding to multiple slices. Slices can correspond to the coronal plane (from front to back), sagittal plane (from left to right), or cross section plane (from top to bottom).

[0514] Periatomy scan images can be used to obtain jawbone feature images. For example, one or more convolutional neural networks can be used to process periatomy scan images to extract jawbone features and obtain jawbone feature images.

[0515] Peripheral bone scan images can be used to obtain images of cancellous bone features. For example, one or more convolutional neural networks can be used to process peripheral bone scan images to extract cancellous bone features and obtain images of cancellous bone features.

[0516] Perialar bone scan images can also be used to obtain cortical bone feature images. For example, one or more convolutional neural networks can be used to process perialar bone scan images to obtain jawbone feature images and cancellous bone images, and then cortical bone feature images can be obtained based on both. Or, for example, one or more convolutional neural networks can be used to process perialar bone scan images to extract cortical bone features and obtain cortical bone feature images.

[0517] The apparatus 200 for determining calcaneal features can also be configured based on the method for determining calcaneal features in any of the technical solutions provided below. Specifically, depending on the relationship between the steps, related steps can be implemented in the same or different modules.

[0518] Electrical or communication connections can be established between the first module 21, the second module 22, the third module 23, the fourth module 24, and the fifth module 25 to enable data transmission.

[0519] One embodiment of this application may also provide a display device.

[0520] The display device includes at least one of the following:

[0521] The sixth module presents a jawbone feature image in a graphical user interface. The jawbone feature image is determined by a neural network based on two inputs: a root bone scan image representing the same jawbone location and a jawbone segmentation image. The jawbone feature image is used to determine a cortical bone feature image based on its difference from a cancellous bone feature image.

[0522] The seventh module presents a cancellous bone feature image in a graphical user interface. The cancellous bone feature image is determined by a neural network based on both a root bone scan image representing the same jawbone location and a jawbone segmentation image as input. The cancellous bone feature image is used to determine a cortical bone feature image based on its difference from the jawbone feature image.

[0523] The eighth module presents a cortical bone feature image in a graphical user interface. The cortical bone feature image is determined by comparing the differences between a jawbone feature image and a cancellous bone feature image. The jawbone feature image and the cancellous bone feature image are determined by a neural network based on both a root bone scan image representing the same jawbone location and a jawbone segmentation image as input.

[0524] The display device can also be configured based on the method for displaying pedicle features in any of the technical solutions provided below. Specifically, depending on the relationship between the steps, related steps can be implemented in the same or different modules.

[0525] Electrical or communication connections can be established between the sixth, seventh, and eighth modules to enable data transmission.

[0526] The device, or its modules or units, may be implemented by a computer chip or physical entity, or by a product with corresponding functions. While the device is described in terms of multiple modules, in some embodiments, the functions of the modules may be implemented in one or more software or hardware components.

[0527] One embodiment of this application provides a storage medium.

[0528] Storage media can be installed in a computer and store applications. Storage media can be any available medium that the computer can access data from, or it can be a storage device such as a server or data center that integrates one or more available media. Available media can be magnetic media such as floppy disks, hard disks, and magnetic tapes; optical media such as DVDs (Digital Video Discs); or semiconductor media such as SSDs (Solid State Disks).

[0529] In one embodiment, when the application is executed, it implements the step of a method for determining calcaneal bone features. In a specific embodiment, the method for determining calcaneal bone features can be implemented according to any technical solution of this application.

[0530] In one embodiment, when the application is executed, it implements the step of a method for displaying calcaneal features. In a specific embodiment, the method for displaying calcaneal features can be implemented according to any technical solution of this application.

[0531] electronic devices

[0532] One embodiment of this application provides an electronic device 100, as shown in FIG26.

[0533] Electronic device 100 can be a computer, mobile phone, tablet computer, etc., and this application does not limit the specific type of electronic device 100.

[0534] The electronic device 100 includes at least one processor 11, at least one memory 12, and a communication bus 13.

[0535] At least one processor 11 and at least one memory 12 communicate with each other via a communication bus 13.

[0536] Memory 12 is used to store application programs.

[0537] In one embodiment, the processor 11 is configured to implement a method for determining calcaneal features when executing an application stored in the memory 12. In a specific embodiment, the method for determining calcaneal features can be implemented according to any technical solution of this application.

[0538] In one embodiment, the processor 11 is configured to implement a method for displaying calcaneal features when executing an application stored in the memory 12. In a specific embodiment, the method for displaying calcaneal features can be implemented according to any technical solution of this application.

[0539] In one embodiment, the communication bus 13 may include any number of buses and bridge circuits. In some embodiments, in addition to connecting the processor and memory, the communication bus may also be used to connect peripheral devices or other peripheral circuits.

[0540] In one embodiment, the electronic device 100 may include a user interface 14 and at least one network interface 15. The user interface 14 may include a display, keyboard, mouse, trackball, click wheel, buttons, a touchpad, or a touch screen, etc.

[0541] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0542] The detailed descriptions listed above are merely specific descriptions of feasible implementation methods of this application and are not intended to limit the scope of protection of this application. All equivalent implementation methods or modifications made without departing from the spirit of the art of this application should be included within the scope of protection of this application.

Claims

1. A method for determining calcaneal features, characterized in that, include: Obtain radicular bone scan images; Based on the radicular bone scan image, the jawbone is segmented to determine the segmented jawbone image; Based on the segmented images of the jawbone, the cortical bone feature information is determined; The root bone scan image is input into the localization model to obtain the first localization information of N teeth. The first localization information is used to determine the position information of the teeth in the root bone scan image. The center information of different teeth in the root bone scan image is identified to obtain the second positioning information of M teeth; N and M are positive integers greater than 1; based on the first localization information of N teeth and the second localization information of M teeth, the tooth segmentation result is obtained; Based on the cortical bone feature information and the tooth segmentation results, the root bone feature information is determined.

2. A method for determining the characteristics of the pedicle, characterized in that, include: Obtain radicular bone scan images; The root bone scan image is input into the localization model to obtain the first localization information of N teeth. The first localization information is used to determine the position information of the teeth in the root bone scan image. The center information of different teeth in the root bone scan image is identified to obtain the second positioning information of M teeth; N and M are positive integers greater than 1; Based on the first localization information of N teeth and the second localization information of M teeth, the region of interest for tooth segmentation is determined; The image corresponding to the region of interest in the segmented teeth is input into the segmentation model to obtain the tooth segmentation result.

3. A method for determining the characteristics of the calcaneus, characterized in that, include: Obtain radicular bone scan images; Based on the radicular bone scan image, the jawbone is segmented to determine the segmented jawbone image; Using a neural network, the jawbone feature image and the cancellous bone feature image are determined by taking both the root bone scan image and the jawbone segmentation image representing the same jawbone location as input. Compare jawbone feature images with cancellous bone feature images, and determine cortical bone feature images based on the differences; Determine and output the bone cortex feature information based on the bone cortex feature image.

4. The method according to claim 1, characterized in that, include: Using a neural network, the jawbone feature image and the cancellous bone feature image are determined by taking both the root bone scan image and the jawbone segmentation image representing the same jawbone location as input. Compare jawbone feature images with cancellous bone feature images, and determine cortical bone feature images based on the differences; Determine the cortical bone feature information based on the cortical bone feature image.

5. The method according to any one of claims 1-4, characterized in that, include: Based on the root bone scan images, the boundary information of the teeth is determined. Based on the morphology and boundary information of the teeth, the boundary distance information is determined. The boundary distance information is used to characterize the distance between the pixels within the tooth range in the root bone scan image and the boundary point of the tooth, and / or the distance between the pixels within the tooth range in the root bone scan image and the center of the tooth. Based on the boundary distance information, the image corresponding to the region of interest segmented from the first tooth is determined in the root bone scan image.

6. The method according to any one of claims 1-5, characterized in that, The first tooth includes the difference between N teeth and M teeth, where N is less than M.

7. The method according to any one of claims 1-6, characterized in that, include: Based on the initial localization information, determine N regions of interest for tooth segmentation; The images corresponding to the regions of interest (ROIs) of N segmented teeth, and the images corresponding to the ROIs of differentially segmented teeth, are input into the segmentation model to obtain the tooth segmentation results, or... The images corresponding to the regions of interest of N teeth are input into the segmentation model to obtain the segmentation results of N teeth. The image corresponding to the region of interest in the segmented teeth is input into the segmentation model to obtain the segmentation result of the teeth with differences. Based on the segmentation results of N teeth and the segmentation results of the differential teeth, the final tooth segmentation result is obtained.

8. The method according to claim 7, characterized in that, include: Based on the initial localization information, determine N regions of interest for tooth segmentation; The images corresponding to the regions of interest of N teeth are input into the segmentation model to obtain the segmentation results of N teeth; Based on the second positioning information, a three-dimensional reconstruction is performed to obtain the center reconstruction results of M teeth; Compare the segmentation results of N teeth with the center reconstruction results of M teeth to identify the teeth with discrepancies.

9. The method according to any one of claims 1-8, characterized in that, The first positioning information includes information about the bounding box surrounding the N teeth; the method includes: Based on the bounding boxes of N teeth, the root bone scan image is cropped to obtain images corresponding to the regions of interest segmented from the N teeth; The images corresponding to the regions of interest of N teeth are input into the segmentation model to obtain the segmentation results of N teeth.

10. The method according to any one of claims 1-9, characterized in that, The second positioning information includes a central mask image, which is used to characterize the center position of the M teeth.

11. The method according to claim 10, characterized in that, The method includes: Based on the root bone scan images, determine the boundary information of the teeth; Based on the morphology and boundary information of the teeth, the boundary distance information is determined. The boundary distance information is used to characterize the distance between the pixels within the tooth range in the root bone scan image and the boundary point of the tooth, and / or the distance between the pixels within the tooth range in the root bone scan image and the center of the tooth. Based on the boundary distance information of the teeth, a central mask image of the teeth is determined. Pixels with boundary distance information less than a preset threshold have a first gray value, and pixels with boundary distance information greater than the preset threshold have a second gray value.

12. The method according to any one of claims 1-11, characterized in that, include: The cortical bone feature image is determined by pixel-by-pixel subtraction between the jawbone feature image and the cancellous bone feature image.

13. The method according to any one of claims 1-12, characterized in that, include: Using a neural network, feature extraction is performed on the root bone scan image by referring to the segmented image of the jawbone, and the jawbone feature image and the cancellous bone feature image are determined.

14. The method according to any one of claims 1-13, characterized in that, Includes at least one of the following: An oral scan model is obtained, and the cortical bone feature information is transformed by coordinate transformation to display the cortical bone feature information at the oral scan model. Based on the bone trabecular feature image, determine the bone trabecular feature information, obtain the oral scanning model, and perform coordinate transformation on the bone trabecular feature information to display the bone trabecular feature information at the oral scanning model; Determine and output the size information of the bone cortex based on the feature image of the bone cortex; Determine and output bone density information based on the characteristic images of cancellous bone; A three-dimensional model of the jawbone is established based on several jawbone feature images corresponding to several jawbone locations. A three-dimensional model of cancellous bone is established based on several cancellous bone feature images corresponding to several jawbone locations. A three-dimensional model of the bone cortex is established based on several cortical bone feature images corresponding to several jawbone locations.

15. The method according to any one of claims 1-14, characterized in that, include: Obtain a root bone scan image and a jaw segmentation image representing the same jawbone position in a first direction, wherein the first direction is perpendicular to the display plane of the root bone scan image and the jaw segmentation image; Based on the segmented images of the jawbone, the regions representing the jawbone areas in the radicular bone scan images are determined; The pedicle scan image is cropped.

16. The method according to claim 15, characterized in that, include: Determine the regions representing the jawbone area in the segmented jawbone image; Cropping segmented jawbone images; Based on the regions representing the jawbone area in the segmented jawbone image, determine the regions representing the jawbone area in the root bone scan image.

17. The method according to any one of claims 1-16, characterized in that, The calcaneal scan images include cone-beam computed tomography (CBCT) images.

18. The method according to any one of claims 1-17, characterized in that, include: Using a neural network, the maxilla and mandible are segmented based on the root bone scan image to determine the segmented images of the maxilla and mandible.

19. A method for displaying calcaneal features, characterized in that, Performing the method as described in any one of claims 1-18 includes at least one of the following: A jawbone feature image is presented in a graphical user interface. The jawbone feature image is determined by a neural network based on both a root bone scan image representing the same jawbone location and a jawbone segmentation image as input. The jawbone feature image is used to determine a cortical bone feature image based on its difference from a cancellous bone feature image. A cancellous bone feature image is presented in a graphical user interface. The cancellous bone feature image is determined by a neural network based on both a root bone scan image representing the same jawbone location and a jawbone segmentation image as input. The cancellous bone feature image is used to determine a cortical bone feature image based on its difference from the jawbone feature image. The graphical user interface presents a cortical bone feature image, which is determined by comparing the differences between a jawbone feature image and a cancellous bone feature image. The jawbone feature image and the cancellous bone feature image are determined by a neural network based on both a root bone scan image and a jawbone segmentation image representing the same jawbone location as input. Present the tooth segmentation results in a graphical user interface; The root bone scan image is presented in a graphical user interface, and the tooth segmentation result is superimposed on the root bone scan image. A 3D model of the teeth after segmentation is presented in a graphical user interface. The 3D model is generated by 3D reconstruction based on the tooth segmentation results.

20. The method according to claim 19, characterized in that, Includes at least one of the following: The jawbone feature images, cancellous bone feature images, and cortical bone feature images are three-dimensional models; Jawbone feature images, cancellous bone feature images, and cortical bone feature images are two-dimensional images.

21. The method according to any one of claims 19-20, characterized in that, Includes at least one of the following: The oral scan model and the corresponding cortical bone feature image are presented in the graphical user interface. The cortical bone feature image is used to characterize the cortical bone region. The oral scan model and the corresponding cancellous bone feature image are presented in the graphical user interface. The cancellous bone feature image is used to characterize the cancellous bone region. The graphical user interface presents an oral scan model and a corresponding jawbone feature image, which is used to characterize the jawbone region.

22. The method according to any one of claims 19-21, characterized in that, Includes at least one of the following: A selection box surrounding the upper jaw area is displayed in the graphical user interface; A selection box surrounding the jawline is displayed in the graphical user interface; The selection box can be a two-dimensional selection box or a three-dimensional selection box; The selection box is determined based on the region representing the jawbone in the segmented jawbone image, which is segmented based on the root bone scan image using a neural network. The graphical user interface displays a bounding box surrounding each tooth. The bounding box can be a three-dimensional bounding box or a two-dimensional bounding box. The bounding box is determined by a neural network based on the location of different teeth according to the root bone scan image.

23. The method according to any one of claims 1-22, characterized in that, A first device is connected, which is used to implement a cone-beam ring headgear to obtain radicular bone scan images, which are used to obtain jawbone feature images and / or cancellous bone feature images.

24. An electronic device comprising a processor, a memory, and a communication bus, characterized in that, The processor and the memory communicate with each other via the communication bus; The memory is used to store application programs; The processor is configured to implement the steps of the method according to any one of claims 1-23 when executing an application stored in the memory.

25. A storage medium having an application program stored thereon, characterized in that, When the application is executed, it implements the steps of the method according to any one of claims 1-23.

Images