WO2016197370A1 - 牙齿和牙槽骨的分割与重构方法及装置 - Google Patents

牙齿和牙槽骨的分割与重构方法及装置 Download PDF

Info

Publication number
WO2016197370A1
WO2016197370A1 PCT/CN2015/081262 CN2015081262W WO2016197370A1 WO 2016197370 A1 WO2016197370 A1 WO 2016197370A1 CN 2015081262 W CN2015081262 W CN 2015081262W WO 2016197370 A1 WO2016197370 A1 WO 2016197370A1
Authority
WO
WIPO (PCT)
Prior art keywords
tooth
alveolar bone
segmented
teeth
dimensional
Prior art date
Application number
PCT/CN2015/081262
Other languages
English (en)
French (fr)
Inventor
夏泽洋
甘阳洲
熊璟
Original Assignee
深圳先进技术研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳先进技术研究院 filed Critical 深圳先进技术研究院
Priority to PCT/CN2015/081262 priority Critical patent/WO2016197370A1/zh
Priority to CN201580000447.1A priority patent/CN107106117B/zh
Publication of WO2016197370A1 publication Critical patent/WO2016197370A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
    • A61B6/51—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for dentistry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion

Definitions

  • the invention relates to the technical field of orthodontics, in particular to a method and a device for segmenting and reconstructing teeth and alveolar bone.
  • Malocclusion is a high incidence. In addition to affecting the patient's aesthetics, it will cause psychological damage to the patient, and it will also affect the maxillofacial development and oral function to cause physical damage to the patient. Orthodontics is a discipline that specializes in the etiology, diagnosis, treatment, and prevention of malocclusion.
  • the dental jaw physical model is an important tool for recording patient records and expressing the shape of the patient's teeth, and is the most important way to reflect the true three-dimensional anatomy of the jaw.
  • the measurement, analysis and operation of the dental jaw physical model are the main basis for the diagnosis of malocclusion, treatment planning and orthodontic instrument design.
  • the dental physics model can only provide three-dimensional information on the surface of the patient's crown, but lacks the three-dimensional information of the root and alveolar bone necessary for orthodontic treatment. Therefore, it is difficult to achieve objective and accurate treatment based on the dental physics model.
  • the dental jaw physical model also has great limitations in terms of model operation difficulty, storage life, storage space, and transmission convenience.
  • Computed Tomography (CT) scanning has been widely used in clinical orthodontic treatment.
  • Reconstructing the digital three-dimensional model of the patient's complete jaw using oral CT images and applying it to orthodontic treatment is an effective way to solve the many shortcomings of the orthodontic physical model in orthodontic treatment.
  • Embodiments of the present invention provide a method for segmenting and reconstructing teeth and alveolar bone, which simultaneously divides the contours of individual teeth and the contour of alveolar bone, and reconstructs a digitized three-dimensional model and alveolar bone of each individual tooth.
  • Digital three-dimensional model comprises: segmenting a bone tissue region in an oral three-dimensional CT image;
  • a digitized three-dimensional model of each individual tooth is reconstructed according to the two-dimensional contour of the individual teeth, and a digitized three-dimensional model of the alveolar bone is reconstructed according to the two-dimensional contour of the alveolar bone.
  • the skeletal tissue region is segmented in the oral three-dimensional CT image, including:
  • the bone tissue region in the three-dimensional CT image including a tooth tissue of the lower jaw and an alveolar bone connecting tissue region, an upper tooth
  • the teeth of the jaw communicate with the alveolar bone and other areas of the jaw tissue.
  • the teeth and alveolar bone tissue regions are segmented from the bone tissue region, including:
  • the volume of each connected region is analyzed in the skeletal tissue region after the opening operation, and the teeth of the lower jaw and the alveolar bone communicating tissue region and the upper jaw teeth and teeth are extracted from the skeletal tissue region according to the volume.
  • the trough connects the tissue area.
  • the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone are segmented from the region of the tooth and alveolar bone tissue, including:
  • Dividing the two-dimensional contours of the individual teeth of the teeth to be segmented from the teeth of the lower or upper jaw and the alveolar bone tissue, respectively, and separating the teeth and the alveolar bone tissue region to be segmented The area outside the area occupied by the two-dimensional contour of each individual tooth is determined as the alveolar bone area, and the two-dimensional contour of the alveolar bone is extracted from the alveolar bone area.
  • the two-dimensional contours of the individual teeth of the dental jaw to be segmented are respectively segmented from the teeth of the lower or upper jaw and the alveolar bone tissue region, including:
  • the two-dimensional contours of the individual teeth of the teeth to be segmented are respectively segmented from the teeth of the lower or upper jaw and the area of the alveolar bone by the following steps:
  • a slice of the crown portion is selected as an initial slice in which all the teeth of the tooth to be segmented are contained.
  • one pixel is selected as a seed point inside the region of each tooth of the tooth to be segmented, and the seed point is used to identify the tooth to be segmented from the tooth and the alveolar bone region in the initial slice.
  • the segments are sliced one by one to segment the individual teeth of the respective segments to be segmented.
  • the two-dimensional contour, in the current slice in which the segmentation operation is performed, the two-dimensional contour of each individual tooth segmented in the previous slice is used as the initial contour of the corresponding individual tooth in the current slice, and the segmentation of the tooth contour is continued to obtain the current slice.
  • Segmentation result when the current slice divides the two-dimensional contour of each individual tooth to be segmented into an empty set, the segmentation operation of the tooth segment to be segmented is ended, otherwise, the segmentation to be segmented is continued in the latter slice The two-dimensional contour of the individual teeth of the jaw.
  • the region of the tooth to be segmented is identified using the seed point, including:
  • the region of the tooth and the alveolar bone tissue region that intersects the seed point is determined as the region of the tooth to be segmented.
  • the two-dimensional contour of each individual tooth to be segmented is segmented, including :
  • the two-dimensional contour of each individual tooth segmented by the previous slice is used as the initial contour of the corresponding individual tooth in the current slice, and the segmentation of the tooth contour is continued to obtain the segmentation result of the current slice, including:
  • the two-dimensional contour of each individual tooth segmented in the previous slice is used as the initial contour of the corresponding individual tooth in the current slice, and the segmentation result of the current slice is obtained by iterative segmentation using the level set algorithm.
  • the two-dimensional contour of each individual tooth segmented in the previous slice is used as the initial contour of the corresponding individual tooth in the current slice, and iteratively segmented by using the level set algorithm.
  • the current two-dimensional profile of each individual tooth including:
  • the two-dimensional contour of each of the individual teeth segmented in the previous slice is used as the initial contour of the corresponding individual teeth in the current slice, the separation line between the adjacent teeth is acquired, and the local region of interest of each tooth of the current slice is obtained, wherein Each separation line divides the local region of interest of the tooth to be segmented corresponding to the separation line into two regions, one region containing the teeth to be segmented and the other region containing the adjacent teeth of the teeth to be segmented;
  • a two-dimensional contour of the teeth to be segmented is iteratively segmented using a level set algorithm.
  • obtaining a separation line separating adjacent teeth comprises:
  • the Ladong transform is used to obtain the separation line between adjacent teeth.
  • the segmentation operation of each sliced tooth in the current segmentation direction is ended, and the current slice and each segment after the current slice along the current segmentation direction are determined. There is no tooth area in the slice, and the current slice and each slice after the current slice along the current segmentation direction are determined.
  • the tooth and the alveolar bone region of the tooth to be segmented are both alveolar bone regions, and the two-dimensional alveolar bone is extracted. profile.
  • the embodiment of the invention further provides a device for segmenting and reconstructing teeth and alveolar bone, which simultaneously divides the contours of the individual teeth and the contour of the alveolar bone, and reconstructs the digitized three-dimensional model and the alveolar of each individual tooth.
  • a digital three-dimensional model of bone comprises: a bone tissue segmentation module, configured to segment a bone tissue region in an oral three-dimensional CT image;
  • a tooth and alveolar bone tissue segmentation module for segmenting a tooth and alveolar bone tissue region from the bone tissue region
  • a contour segmentation module for segmenting a two-dimensional contour of each individual tooth and a two-dimensional contour of the alveolar bone from the region of the tooth and alveolar bone tissue;
  • a model reconstruction module configured to reconstruct a digitized three-dimensional model of each individual tooth according to the two-dimensional contour of the independent teeth, and reconstruct a digitized three-dimensional model of the alveolar bone according to the two-dimensional contour of the alveolar bone.
  • the bone tissue is segmented in the three-dimensional CT image of the oral cavity, and the tooth and the alveolar bone tissue are segmented from the bone tissue, and finally the two-dimensional shape of each individual tooth is segmented from the tooth and the alveolar bone tissue.
  • the two-dimensional contour of the contour and the alveolar bone enables the simultaneous division of the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone, thereby realizing the digital three-dimensional reconstruction of the individual teeth according to the two-dimensional contour of each individual tooth.
  • the model reconstructs a digital three-dimensional model of the alveolar bone based on the two-dimensional contour of the alveolar bone to facilitate digitized orthodontic treatment.
  • FIG. 1 is a flow chart of a method for segmenting and reconstructing teeth and alveolar bone according to an embodiment of the present invention
  • FIG. 2 is a flow chart of segmentation and extraction of teeth and alveolar bone tissue according to an embodiment of the present invention
  • FIG. 3 is a flow chart showing the segmentation of individual tooth contours by sliced oral three-dimensional CT images of a single jaw according to an embodiment of the present invention
  • Fig. 5 is a structural view showing a device for dividing and reconstructing teeth and alveolar bone according to an embodiment of the present invention.
  • FIG. 1 is a flow chart of a method for segmenting and reconstructing teeth and alveolar bone according to an embodiment of the present invention. As shown in FIG. 1 , the method for segmenting and reconstructing teeth and alveolar bone in the embodiment of the present invention may include:
  • Step 101 segmenting a bone tissue region in an oral three-dimensional CT image
  • Step 102 Segmenting a tooth and an alveolar bone tissue region from the bone tissue region;
  • Step 103 segmenting the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone from the tooth and alveolar bone tissue region;
  • Step 104 Reconstruct a digitized three-dimensional model of each individual tooth according to the two-dimensional contour of the individual teeth, and reconstruct a digitized three-dimensional model of the alveolar bone according to the two-dimensional contour of the alveolar bone.
  • the bone tissue region is segmented in the oral three-dimensional CT image, and then the tooth and alveolar bone tissue region is segmented from the bone tissue region, and then the tooth and the alveolar bone are removed.
  • the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone are segmented in the tissue region, so that the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone can be simultaneously divided, thereby realizing the individual teeth according to the individual teeth.
  • Two-dimensional contour reconstruction of the digital three-dimensional model of each individual tooth reconstruction of the digital three-dimensional model of the alveolar bone according to the two-dimensional contour of the alveolar bone, in order to facilitate the digital orthodontic treatment assistance, enabling digital orthodontic treatment In the assist, the physician can perform interactive operation on the individual teeth when planning the orthodontic treatment plan.
  • the root portion of the tooth is fixed and buried in the alveolar bone, it can be used in the tooth arrangement stage of the treatment plan. To ensure that the crown and root of the tooth are arranged neatly, the root can maintain a good positional relationship with the alveolar bone.
  • a surface mesh reconstruction method such as Marching Cube can be used to reconstruct a three-dimensional surface patch model of each individual tooth and alveolar bone, or a surface fitting method such as NURBS can be used to reconstruct each individual tooth and tooth.
  • a three-dimensional surface model of the trough can be used to reconstruct a three-dimensional surface patch model of each individual tooth and alveolar bone.
  • the present application proposes to segment the bone tissue region in the oral three-dimensional CT image, and then perform two-dimensional modeling of each individual tooth based on the bone tissue region. Segmentation of the contour and the two-dimensional contour of the alveolar bone.
  • the segmentation of the bone tissue region in the three-dimensional CT image of the oral cavity comprises: according to the difference in the gray value of the pixel in the three-dimensional CT image,
  • the skeletal tissue region is segmented in the three-dimensional CT image, the skeletal tissue region including the teeth of the lower jaw and the alveolar bone communicating tissue region, the teeth of the upper jaw and the alveolar bone communicating tissue region, and other jaw bones Organization area.
  • the inventors have found that oral three-dimensional CT image pixels are composed of several types of target pixels such as bone tissue, soft tissue and air. Among them, the pixels of the bone tissue have higher gray scale than the soft tissue and air pixels, and the pixel gray value difference Larger.
  • the segmentation of the bone tissue region can be directly segmented in the three-dimensional space of the CT image according to the difference in the gray value of the pixel, or can be segmented slice by slice on the two-dimensional CT slice.
  • the segmentation method may employ, for example, a segmentation method such as threshold segmentation, watershed, region growth, and level set.
  • the segmented skeletal tissue region is composed of a plurality of parts such as a tooth region of the lower jaw and an alveolar bone communicating tissue region, a tooth of the upper jaw, an alveolar bone communicating tissue region, and other jaw tissue regions.
  • Segmenting the tooth and alveolar bone tissue region in the skeletal tissue region including: first, opening the bone tissue region in a three-dimensional space; and then, for each connected region in the bone tissue region after the opening operation
  • the volume is analyzed, and the teeth and the alveolar bone communicating tissue region of the lower jaw and the tooth and alveolar bone communicating tissue region of the upper jaw are extracted from the open bone tissue region according to the volume.
  • the teeth and alveolar bone tissue regions are segmented by the following steps:
  • Step 201 the three-dimensional CT image
  • Step 202 segmenting a bone tissue region in the three-dimensional CT image
  • Step 203 Perform an opening operation on the bone tissue region
  • Step 204 After the skeletal tissue region is opened, the largest volume region in each connected region is the tooth and alveolar bone communicating tissue region of the lower jaw, followed by the upper jaw tooth and the alveolar bone connecting tissue region. Therefore, after the opening operation is performed in the bone tissue region, the size analysis of the volume of the three-dimensional connected region is performed;
  • Step 205 extract the largest volume area according to the size of the volume
  • Step 206 extracting a region connecting the teeth of the lower jaw to the alveolar bone
  • Step 207 Extract a second largest area according to the size of the volume
  • Step 208 Extracting the area of the upper jaw and the alveolar bone to connect the tissue.
  • each tooth After segmenting the teeth of the lower jaw and the alveolar bone tissue region and the teeth of the upper jaw and the alveolar bone tissue region, the teeth and the alveolar bone tissue regions of each tooth can be simultaneously separated in sequence or in parallel.
  • the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone, in particular, the tooth can be obtained simultaneously by dividing the two-dimensional contour of each individual tooth from the tooth to be segmented and the alveolar bone tissue region.
  • the two-dimensional contour of the trough bone, or a two-dimensional contour of the individual teeth can be obtained simultaneously by dividing the two-dimensional contour of the alveolar bone from the tooth to be segmented and the alveolar bone tissue region.
  • the present invention removes the separated individual teeth from the teeth of the teeth to be segmented and the alveolar bone tissue region by dividing the two-dimensional contour of each individual tooth from the teeth of the tooth to be segmented and the alveolar bone tissue region.
  • the area outside the area occupied by the two-dimensional contour is determined as the alveolar bone region, and the two-dimensional contour of the alveolar bone is extracted from the alveolar bone region, thereby realizing two-dimensional independent teeth in the tooth to be segmented teeth and the alveolar bone tissue region. Simultaneous segmentation of the contour and the two-dimensional contour of the alveolar bone.
  • the two-dimensional contour of each of the independent teeth to be segmented is segmented from the teeth of the lower or upper jaw and the alveolar bone tissue region, including:
  • the two-dimensional contour of the individual teeth of the tooth to be segmented is segmented in the tooth and alveolar bone tissue region of the segmented jaw (for example, the lower or upper jaw):
  • Step 301 In the three-dimensional CT image slice of the tooth and the alveolar bone tissue region from which the tooth to be segmented (the upper or lower jaw) has been extracted, select a slice of the crown portion as the initial slice, and the segmentation will be This initial slice begins.
  • the initial section should be selected from the section of the crown close to the neck of the tooth.
  • the complete crown containing all the teeth of the dental dentition to be segmented, and the tooth region of the initial segment in which the tooth is to be segmented is not connected to the alveolar bone region, and if the tooth region is connected to the alveolar bone region , the curve of the boundary between the crown and the alveolar bone can be drawn, and the connected teeth are separated from the alveolar bone.
  • the crown is not connected to the alveolar bone tissue; but for the lower dentition, the crown of the molar may be connected to the alveolar bone tissue, and the crown can be manually drawn.
  • the boundary of the alveolar bone separates the connected crown from the alveolar bone.
  • Step 302 In the initial section, the tooth area is no longer connected to the alveolar bone area.
  • a pixel is selected as a seed point in the region of each tooth of the tooth to be segmented in the initial slice, and the tooth region of the tooth to be segmented is identified by the seed point.
  • a seed point may be manually selected in each tooth region in the initial slice, and the seed point may be a pixel or a pixel region; in the initial slice, the tooth and the alveolar bone tissue are The region in the region that intersects the seed point is determined as the region of the tooth to be segmented.
  • Step 303 in the tooth region extracted from the initial slice, adjacent teeth may stick together, thereby causing failure of independent tooth segmentation.
  • a separation line of adjacent teeth is obtained, thereby separating adjacent teeth.
  • Open to obtain a two-dimensional contour of each individual tooth in the initial slice.
  • the separation line separates adjacent teeth into independent individuals, thereby obtaining a two-dimensional contour of each individual tooth in the initial slice, and other methods can be used to obtain the relationship between adjacent teeth.
  • Offline In exchange for the separation line between adjacent teeth, the separation line separates adjacent teeth into independent individuals, thereby obtaining a two-dimensional contour of each individual tooth in the initial slice, and other methods can be used to obtain the relationship between adjacent teeth.
  • Step 304 Divide each slice other than the initial slice in the three-dimensional CT image slice of the tooth and the alveolar bone tissue region that has been segmented from the tooth to be segmented, and segment each slice to be segmented in each slice.
  • the two-dimensional contour of the independent tooth In the current slice in which the segmentation operation is performed, the two-dimensional contour of each individual tooth segmented in the previous slice is the initial contour of the corresponding individual tooth in the current slice, and the segmentation of the tooth contour is continued.
  • the segmentation result of the current slice ends the segmentation operation of the segmental tooth to be segmented when the current slice divides the two-dimensional contour of each individual tooth to be segmented into an empty set, otherwise, the segmentation continues in the latter slice.
  • the two-dimensional contour of each individual tooth to be segmented The two-dimensional contour of each individual tooth to be segmented.
  • step 304 the three-dimensional CT image slice of the tooth to be segmented and the alveolar bone tissue region is divided into upper and lower parts by the initial slice, and the upper and lower parts are The segmentation of the slice starts with the initial slice, and divides the two-dimensional contour of the independent tooth in each slice in the upper and lower portions in a slice direction toward the upward segmentation direction and the downward segmentation direction, respectively.
  • the two-dimensional contour of each individual tooth segmented in the previous slice eg, the current slice is the second slice, the previous slice is the initial slice
  • the current slice corresponding to each tooth is used as the current slice corresponding to each tooth.
  • the initial contour is iteratively segmented using a level set algorithm to obtain a two-dimensional contour of each tooth. If the contour of each tooth of the dentition is an empty set in the segmentation result of the current slice performing the segmentation operation, the segmentation of the contour of the subsequent slice on the segmentation direction (the upward or downward segmentation direction) is ended, and the current slice is determined.
  • the tooth region is not present in each slice after the current slice along the current segmentation direction, and the current slice and each slice after the current slice along the current segmentation direction are determined, and the tooth and the alveolar bone region of the tooth to be segmented are both alveolar
  • the bone region extracts the two-dimensional contour of the alveolar bone; otherwise, the segmentation result of the current slice is transmitted to the latter slice, and the segmentation process is repeated for the latter slice.
  • the segmentation of the upper and lower image slices obtained by the initial slice division may be performed in parallel or sequentially.
  • step 304 for the non-initial slice, in order to realize the fully automatic division of the two-dimensional contour of each individual tooth in each slice, and at the same time improve the accuracy of the tooth segmentation, in the current slice of the segmentation operation, all the previous slices
  • the two-dimensional contour of each individual tooth segmented in the slice is used as the initial contour corresponding to each tooth, and the segmentation result of the two-dimensional contour of each independent tooth of the current slice is obtained by using the level set algorithm to iteratively segment, including: the independent segments separated in the previous slice.
  • the level set algorithm is used to iteratively segment the two-dimensional contour of the tooth to be segmented. Specifically, as shown in FIG. 4, in the current slice in which the segmentation operation is performed, the two-dimensional contour of each individual tooth in the current slice is obtained by the following steps:
  • Step 401 The current slice of the tooth and the alveolar bone tissue region of the tooth to be segmented has been extracted
  • Step 402 Acquire a two-dimensional contour of each individual tooth segmented in the previous slice
  • Step 403 The two-dimensional contour of each of the independent teeth segmented in the previous slice is the initial contour of the corresponding individual teeth in the current slice, obtain the separation line between the adjacent teeth, and acquire the local interest of each tooth of the current slice. a region, wherein, for each separation line, the local region of interest of the tooth to be segmented corresponding to the separation line is divided into two regions, one region containing the teeth to be segmented and the other region containing adjacent teeth of the teeth to be segmented ;
  • Step 404 In the region including the tooth in the local region of interest of the tooth to be segmented, the level set algorithm is used to segment the two-dimensional contour of the tooth to be segmented by using the level set algorithm.
  • the local region of interest for each tooth is used.
  • the division in the region containing the tooth may be performed in parallel or sequentially in series.
  • Step 405 Obtain a two-dimensional contour of each individual tooth in the current slice.
  • a device for dividing and reconstructing a tooth and an alveolar bone is further provided.
  • the device for dividing and reconstructing the tooth and the alveolar bone comprises:
  • the skeletal tissue segmentation module 501 is configured to segment the skeletal tissue region in the oral three-dimensional computed tomography CT image; specifically, the skeletal tissue segmentation module 501 is responsible for completing the segmentation of the skeletal tissue in the segmentation and reconstruction device of the tooth and the alveolar bone
  • the part of the area function may be software, hardware or a combination of the two, and may be, for example, a component such as a processing chip that performs the image dividing function.
  • a tooth and alveolar bone segmentation module 502 for segmenting the tooth and alveolar bone tissue region from the bone tissue region; specifically, the tooth and alveolar bone tissue segmentation module 502 is a segmentation of the tooth and alveolar bone
  • the part of the reconstruction device responsible for completing the function of dividing the teeth and the alveolar bone tissue region may be software, hardware or a combination of the two, and may be, for example, a processing chip or the like that performs the image segmentation function.
  • a contour segmentation module 503 configured to segment a two-dimensional contour of each individual tooth and a two-dimensional contour of the alveolar bone from the tooth and alveolar bone tissue region; specifically, the contour segmentation module 503 is a tooth and an alveolar bone
  • the part of the segmentation and reconstruction device responsible for performing the function of dividing the teeth and the alveolar bone contour may be software, hardware or a combination of the two, and may be, for example, a processing chip or the like that performs the image segmentation function.
  • the model reconstruction module 504 is configured to reconstruct a digitized three-dimensional model of each individual tooth according to the two-dimensional contour of the independent teeth, and reconstruct a digitized three-dimensional model of the alveolar bone according to the two-dimensional contour of the alveolar bone.
  • the model is heavy
  • the structure module 504 is a part of the division and reconstruction device of the tooth and the alveolar bone and is responsible for completing the model reconstruction function, and may be software, hardware or a combination of the two, for example, a processing chip or the like that performs the reconstruction function of the model. Device.
  • the bone tissue region is segmented in the oral three-dimensional CT image, and the tooth and alveolar bone tissue region is segmented from the bone tissue region, and finally the teeth and the alveolar bone tissue region are separated.
  • the two-dimensional contour of the individual teeth and the two-dimensional contour of the alveolar bone make it possible to simultaneously segment the two-dimensional contour of each individual tooth and the two-dimensional contour of the alveolar bone, thereby realizing the reconstruction according to the two-dimensional contour of each individual tooth.
  • a digital three-dimensional model of the independent tooth reconstructing the digital three-dimensional model of the alveolar bone according to the two-dimensional contour of the alveolar bone, in order to facilitate the digitized orthodontic treatment, so that in the digital computer orthodontic treatment, the physician performs positive
  • the orthodontic treatment plan can be used to interactively operate the individual teeth.
  • the crown of the teeth can be ensured during the tooth arrangement stage of the treatment plan.
  • the roots are arranged neatly, and the roots can maintain a good positional relationship with the alveolar bone.
  • embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Multimedia (AREA)
  • Biophysics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Dentistry (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Optics & Photonics (AREA)
  • Veterinary Medicine (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

一种牙齿和牙槽骨的分割与重构方法及装置，其中，该方法包括：在口腔三维CT图像中分割出骨骼组织区域(101)；从骨骼组织区域中分割出牙齿与牙槽骨组织区域(102)；从牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓(103)；根据各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据牙槽骨的二维轮廓重构牙槽骨的数字化三维模型(104)。该方案可以实现同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，进而实现根据各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据牙槽骨的二维轮廓重构牙槽骨的数字化三维模型，以利于实现数字化的正畸治疗辅助。

Description

牙齿和牙槽骨的分割与重构方法及装置 技术领域
本发明涉及口腔正畸技术领域，尤其涉及牙齿和牙槽骨的分割与重构方法及装置。
背景技术
错颌畸形具有较高的发病率。其除了影响患者美观给患者带来心理伤害外，还会影响颌面发育和口腔功能给患者造成身体伤害。口腔正畸学是专门研究错颌畸形的病因、诊断、治疗和预防的学科。
在传统的正畸治疗中，牙颌物理模型是记录患者病案、表达患者牙齿形态的重要工具，也是反映牙颌真实三维解剖结构的最主要途径。对牙颌物理模型的测量、分析和操作是错颌畸形诊断、治疗方案规划和矫治器械设计的主要依据。牙颌物理模型只能提供患者牙冠表面的三维信息，而缺乏正畸治疗所必须的牙根及牙槽骨的三维信息。因此，基于牙颌物理模型的治疗方式难以实现客观、准确的治疗。同时，牙颌物理模型在模型的操作难度、存储寿命、存储空间、传输便利性等方面也存在很大局限性。近年来，口腔电子计算机断层扫描(Computed Tomography，CT)扫描已逐步在临床正畸治疗中得到广泛应用。利用口腔CT图像重构患者完整牙颌的数字化三维模型并应用于正畸治疗，是解决牙颌物理模型在正畸治疗应用中诸多不足的有效途径。
要实现上述口腔CT图像在正畸治疗中的应用，首先需要从口腔CT图像中分割出各独立牙颌组织，并重构出各独立牙颌组织的数字化三维模型。目前，已有多种从口腔CT图像中实现独立牙齿的分割与重构的系统和方法。然而，这些系统和方法都未涉及牙槽骨的分割问题，从而不能重构出牙槽骨的数字化三维模型。而在正畸治疗应用中，尤其是在治疗方案规划阶段，独立牙齿的数字化三维模型和牙槽骨的数字化三维模型都是必须的。
发明内容
本发明实施例提供一种牙齿和牙槽骨的分割与重构方法，用以同时分割出各独立牙齿的轮廓和牙槽骨的轮廓，并重构各独立牙齿的数字化三维模型和牙槽骨的数字化三维模型。该方法包括：在口腔三维CT图像中分割出骨骼组织区域；
从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；
从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；
根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。
在一个实施例中，在口腔三维CT图像中分割出骨骼组织区域，包括：
根据所述三维CT图像中像素的灰度值差异，在所述三维CT图像中分割出所述骨骼组织区域，所述骨骼组织区域包括下牙颌的牙齿与牙槽骨连通组织区域、上牙颌的牙齿与牙槽骨连通组织区域以及其他的颌骨组织区域。
在一个实施例中，从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域，包括：
在三维空间中，对所述骨骼组织区域进行开运算；
在开运算后的骨骼组织区域中对各连通区域的体积进行分析，根据体积大小在所述骨骼组织区域中提取出下牙颌的牙齿与牙槽骨连通组织区域和上牙颌的牙齿与牙槽骨连通组织区域。
在一个实施例中，从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，包括：
从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分别分割出待分割牙颌的各独立牙齿二维轮廓，将待分割牙颌的牙齿与牙槽骨组织区域中除去分割出的各独立牙齿二维轮廓所占区域外的区域确定为牙槽骨区域，从牙槽骨区域提取牙槽骨的二维轮廓。
在一个实施例中，从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分别分割出待分割牙颌的各独立牙齿的二维轮廓，包括：
通过以下步骤从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分别分割出待分割牙颌的各独立牙齿的二维轮廓：
在已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片中，选择一张牙冠部位的切片为初始切片，在所述初始切片中，含有待分割牙颌的所有牙齿的完整牙冠：
在所述初始切片中，在待分割牙颌的各牙齿的区域内部分别选择一个像素作为种子点，利用所述种子点从所述初始切片中的牙齿与牙槽骨区域中识别出待分割牙颌的牙齿区域；
在所述识别出的牙齿区域中，获取分离相邻牙齿的分离线，得到所述初始切片中各独立牙齿的二维轮廓；
对已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片中除了所述初始切片之外的其他切片，循环逐个切片分割出各自切片中待分割牙颌的各独立牙齿的二维轮廓，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应的各独立牙齿的初始轮廓，继续进行牙齿轮廓的分割得到当前切片的分割结果，在当前切片分割出待分割牙颌的各独立牙齿的二维轮廓为空集时，结束待分割牙颌牙齿逐切片式的分割操作，否则，在后一切片中继续分割待分割牙颌各独立牙齿的二维轮廓。
在一个实施例中，在所述初始切片中，利用所述种子点识别出待分割牙颌的牙齿区域，包括：
将牙齿与牙槽骨组织区域中与所述种子点相交的区域确定为待分割牙颌的牙齿区域。
在一个实施例中，从已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像的非所述初始切片中，分割出待分割牙颌的各独立牙齿的二维轮廓，包括：
以所述初始切片为界将待分割牙颌的所有切片分为上、下两个部分，以所述初始切片为起始位置，分别朝着向上或向下的分割方向分割各部分中各切片内独立牙齿的二维轮廓；
在进行分割操作的当前切片中，以前一切片分割出的各独立牙齿的二维轮廓作为当前切片中对应各独立牙齿的初始轮廓，继续进行牙齿轮廓的分割得到当前切片的分割结果，包括：
以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应各独立牙齿的初始轮廓，采用水平集算法进行迭代分割得到当前切片的分割结果。
在一个实施例中，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应的各独立牙齿的初始轮廓，采用水平集算法进行迭代分割得到当前切片各独立牙齿的二维轮廓，包括：
以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应的各独立牙齿的初始轮廓，获取相邻牙齿之间的分离线并获取当前切片各牙齿的局部感兴趣区域，其中，每条分离线，将与该分离线对应的待分割牙齿的局部感兴趣区域分为了两个区域，一个区域包含待分割牙齿，另一个区域包含待分割牙齿的相邻牙齿；
在所述各独立牙齿的局部感兴趣区域中包含待分割牙齿的区域内，采用水平集算法迭代分割出待分割牙齿的二维轮廓。
在一个实施例中，获取分离相邻牙齿的分离线，包括：
利用拉东变换获取相邻牙齿之间的分离线。
在一个实施例中，在当前切片分割出的各独立牙齿的二维轮廓为空集时，结束当前分割方向上各切片牙齿的分割操作，确定当前切片和沿当前分割方向在当前切片之后的各切片中不存在牙齿区域，确定当前切片和沿当前分割方向在当前切片之后的各切片中，待分割牙颌的牙齿与牙槽骨区域都为牙槽骨区域，提取出牙槽骨的二维轮廓。
本发明实施例还提供一种牙齿和牙槽骨的分割与重构装置，用以同时分割出各独立牙齿的轮廓和牙槽骨的轮廓，并重构各独立牙齿的数字化三维模型和牙槽骨的数字化三维模型。该装置包括：骨骼组织分割模块，用于在口腔三维CT图像中分割出骨骼组织区域；
牙齿与牙槽骨组织分割模块，用于从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；
轮廓分割模块，用于从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；
模型重构模块，用于根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。
本发明实施例中，通过在口腔三维CT图像中分割出骨骼组织，进而从骨骼组织中分割出牙齿与牙槽骨组织，最后再从牙齿与牙槽骨组织中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，使得可以实现同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，进而实现根据各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据牙槽骨的二维轮廓重构牙槽骨的数字化三维模型，以利于实现数字化的正畸治疗辅助。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。在附图中：
图1为本发明实施例中牙齿和牙槽骨的分割与重构方法的流程图；
图2为本发明实施例中牙齿与牙槽骨组织的分割提取流程图；
图3为本发明实施例中对单颌的口腔三维CT图像逐切片进行各独立牙齿轮廓分割的流程图；
图4为本发明实施例中非初始切片进行各独立牙齿轮廓分割的流程图；
图5为本发明实施例中牙齿和牙槽骨的分割与重构装置的结构图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚明白，下面结合附图对本发明实施例做进一步详细说明。在此，本发明的示意性实施例及其说明用于解释本发明，但并不作为对本发明的限定。
图1为本发明实施例中牙齿和牙槽骨的分割与重构方法的流程图。如图1所示，本发明实施例中牙齿和牙槽骨的分割与重构方法可以包括：
步骤101：在口腔三维CT图像中分割出骨骼组织区域；
步骤102：从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；
步骤103：从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；
步骤104：根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。
由图1可以得知，在本发明实施例中，通过在口腔三维CT图像中分割出骨骼组织区域，然后从骨骼组织区域中分割出牙齿与牙槽骨组织区域，再从牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，使得可以实现同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，进而实现根据各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据牙槽骨的二维轮廓重构牙槽骨的数字化三维模型，以利于实现数字化的正畸治疗辅助，使得在进行数字化的计算机正畸治疗辅助中，医师进行正畸治疗方案规划时可以对各独立牙齿进行交互式操作，同时，由于牙齿的牙根部位是固定和埋藏在牙槽骨中的，在治疗方案规划的排牙阶段时，可以保证牙齿的牙冠与牙根排列整齐外，还可以使牙根与牙槽骨保持良好的位置关系。
具体实施时，可以采用Marching Cube等表面网格重构方法，重构得到各独立牙齿以及牙槽骨的三维表面面片模型，或采用NURBS等曲面拟合方法，重构得到各独立牙齿以及牙槽骨的三维曲面模型。
为了实现可以同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，本申请提出先在口腔三维CT图像中分割出骨骼组织区域，再基于骨骼组织区域进行各独立牙齿的二维轮廓和牙槽骨的二维轮廓的分割，具体的，在本实施例中，在口腔三维CT图像中分割出骨骼组织区域，包括：根据所述三维CT图像中像素的灰度值差异，在所述三维CT图像中分割出所述骨骼组织区域，所述骨骼组织区域包括下牙颌的牙齿与牙槽骨连通组织区域、上牙颌的牙齿与牙槽骨连通组织区域以及其他的颌骨组织区域。发明人发现口腔三维CT图像像素由骨骼组织、软组织和空气等几类目标像素构成，其中，骨骼组织的像素与软组织和空气的像素相比，具有较高的灰度，且像素灰度值差异较大。因此，提出骨骼组织区域的分割可根据像素的灰度值差异直接在CT图像三维空间中进行分割，也可在二维CT切片上逐切片地进行分割。分割方法例如可采用阈值分割、分水岭、区域生长、水平集等分割方法。分割出的骨骼组织区域由下牙颌的牙齿与牙槽骨连通组织区域、上牙颌的牙齿与牙槽骨连通组织区域及其他的颌骨组织区域等多个部分构成。
为了实现可以同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，还需要从骨骼组织区域中分割出牙齿与牙槽骨组织区域，具体的，在本实施例中，从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域，包括：首先，在三维空间中，对所述骨骼组织区域进行开运算；然后，在开运算后的骨骼组织区域中对各连通区域的体积进行分析，根据体积大小在所述开运算后的骨骼组织区域中提取出下牙颌的牙齿与牙槽骨连通组织区域和上牙颌的牙齿与牙槽骨连通组织区域。例如，如图2所示，通过以下步骤分割出牙齿与牙槽骨组织区域：
步骤201：所述三维CT图像；
步骤202：在所述三维CT图像中，分割出骨骼组织区域；
步骤203：对所述骨骼组织区域进行开运算；
步骤204：骨骼组织区域进行开运算后，可以发现各连通区域中，体积最大的区域是下牙颌的牙齿与牙槽骨连通组织区域，其次是上牙颌的牙齿与牙槽骨连通组织区域，因此，在骨骼组织区域进行开运算后，进行三维连通区域体积的大小分析；
步骤205：根据体积的大小，提取出体积最大区域；
步骤206：提取出下牙颌的牙齿与牙槽骨连通组织区域；
步骤207：根据体积的大小，提取出体积第二大区域；
步骤208：提取出上牙颌的牙齿与牙槽骨连通组织区域。
分割出下牙颌的牙齿与牙槽骨组织区域和上牙颌的牙齿与牙槽骨组织区域后，则可以依次或并行地从各牙颌的牙齿与牙槽骨组织区域中同时分割出其中的各独立牙齿的二维轮廓和牙槽骨的二维轮廓，具体的，只要从待分割牙颌的牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓，即可同时得到牙槽骨的二维轮廓，或者，只要从待分割牙颌的牙齿与牙槽骨组织区域中分割出牙槽骨的二维轮廓，即可同时得到各独立牙齿的二维轮廓。因此，本发明通过从待分割牙颌的牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓，将待分割牙颌的牙齿与牙槽骨组织区域中除去分割出的各独立牙齿二维轮廓所占区域外的区域确定为牙槽骨区域，从牙槽骨区域提取牙槽骨的二维轮廓，从而实现待分割牙颌的牙齿与牙槽骨组织区域中各独立牙齿二维轮廓和牙槽骨二维轮廓的同时分割。
具体实施时，如图3所示，从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分割出待分割牙颌的各独立牙齿的二维轮廓，包括：通过以下步骤从所述待分割牙颌(例如，下牙颌或上牙颌)的牙齿与牙槽骨组织区域中分割出待分割牙颌各独立牙齿的二维轮廓：
步骤301：在已提取出待分割牙颌(上牙颌或下牙颌)的牙齿与牙槽骨组织区域的三维CT图像切片中，选择一张牙冠部位的切片为初始切片，分割将从这张初始切片开始。初始切片应尽量从靠近牙颈部位的牙冠切片中选择。在所述初始切片中，含有待分割牙颌牙列所有牙齿的完整牙冠，同时初始切片中待分割牙颌的牙齿区域与牙槽骨区域不相连，若存在牙齿区域与牙槽骨区域相连，则可以绘制牙冠与牙槽骨间边界的曲线，将相连的牙齿与牙槽骨分割开。例如，对于上牙颌牙列，牙冠不会与牙槽骨组织相连；但对于下牙颌牙列，磨牙的牙冠可能与牙槽骨组织相连，此时就可以手动地绘制牙冠与牙槽骨的边界，将相连的牙冠与牙槽骨分割开。
步骤302：在所述初始切片中，牙齿区域与牙槽骨区域不再相连。在所述初始切片中的待分割牙颌各牙齿的区域内分别选择一个像素作为种子点，利用种子点识别出待分割牙颌的牙齿区域。具体的，例如，可以手动在所述初始切片中的各牙齿区域内分别选择一个种子点，该种子点可以是一个像素也可以是像素区域；所述初始切片中，将牙齿与牙槽骨组织区域中与所述种子点相交的区域确定为待分割牙颌的牙齿区域。
步骤303：在从所述初始切片提取出的牙齿区域中，相邻的牙齿可能相互粘连在一起，从而造成独立牙齿分割失败，为此，获取相邻牙齿的分离线，从而将相邻牙齿分离开，得到所述初始切片中各独立牙齿的二维轮廓。具体的，可以利用Radon(拉东)变 换求取相邻牙齿间的分离线，该分离线将相邻的牙齿分离为独立的个体，从而得到初始切片中各独立牙齿的二维轮廓，也可以采用其他方法求取相邻牙齿间的分离线。
步骤304：对已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片中除了所述初始切片之外的其他切片，循环逐个切片分割出各自切片中待分割牙颌的各独立牙齿的二维轮廓，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓为当前切片中对应的各独立牙齿的初始轮廓，继续进行牙齿轮廓的分割得到当前切片的分割结果，在当前切片分割出待分割牙颌的各独立牙齿的二维轮廓为空集时，结束待分割牙颌牙齿逐切片式的分割操作，否则，在后一切片中继续分割待分割牙颌各独立牙齿的二维轮廓。
具体的，在步骤304中，以所述初始切片为界将待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片分为了上、下两个部分，对所述上、下两部分切片的分割，以所述初始切片为起点，分别朝着向上的分割方向和向下的分割方向逐切片地完成上、下两部分中各切片内独立牙齿二维轮廓的分割。同时，在进行分割操作的当前切片中，以前一切片(如当前切片是第二个切片，则前一切片为初始切片)中分割出的各独立牙齿的二维轮廓作为当前切片对应各牙齿的初始轮廓，利用水平集算法迭代分割得到各牙齿的二维轮廓。若所述进行分割操作的当前切片的分割结果中，牙列各牙齿的轮廓都为空集，则结束该分割方向(向上或向下的分割方向)上后续切片牙齿轮廓的分割，确定当前切片和沿当前分割方向在当前切片之后的各切片中不存在牙齿区域，确定当前切片和沿当前分割方向在当前切片之后的各切片中，待分割牙颌的牙齿与牙槽骨区域都为牙槽骨区域，提取出牙槽骨的二维轮廓；否则，将当前切片的分割结果传递到后一切片，对后一切片重复上述分割过程。
具体实施时，由所述初始切片划分得到的上、下两部分图像切片的分割可以并行进行，也可以先后依次进行。
具体实施时，在上述步骤304中，对于非初始切片，各切片中为了实现全自动地分割出各独立牙齿的二维轮廓，同时提高牙齿分割精度，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓作为对应各牙齿的初始轮廓，采用水平集算法迭代分割得到当前切片各独立牙齿二维轮廓的分割结果，包括：以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应各牙齿的初始轮廓，获取相邻牙齿之间的分离线以及各牙齿的局部感兴趣区域；对于所述每条分离线，其将与该分离线对应的待分割牙齿的局部感兴趣区域分为两个区域，一个区域包含待分割牙齿，另一个区域包 含待分割牙齿的相邻牙齿。在所述局部感兴趣区域内的两区域中只有包含待分割牙齿的区域才是待分割牙齿轮廓的可行域。在所述局部感兴趣区域内包含待分割牙齿的区域中，采用水平集算法迭代分割出待分割牙齿的二维轮廓。具体的，如图4所示，在进行分割操作的当前切片中，通过以下步骤得到当前切片中各独立牙齿的二维轮廓：
步骤401：已提取出待分割牙颌的牙齿与牙槽骨组织区域的当前切片；
步骤402：获取前一切片中分割出的各独立牙齿的二维轮廓；
步骤403：以前一切片中分割出的各独立牙齿的二维轮廓为当前切片中对应的各独立牙齿的初始轮廓，获取相邻牙齿之间的分离线，并获取当前切片各牙齿的局部感兴趣区域，其中，对于每条分离线，其将与该分离线对应的待分割牙齿的局部感兴趣区域分为了两个区域，一个区域包含待分割牙齿，另一个区域包含待分割牙齿的相邻牙齿；
步骤404：在待分割牙齿局部感兴趣区域内包含该牙齿的区域中，采用水平集算法通过水平集迭代分割出待分割牙齿的二维轮廓，具体实施时，针对各牙齿在其局部感兴趣区域内包含该牙齿的区域中的分割可并行进行，也可依次串行进行。
步骤405：得到当前切片中各独立牙齿的二维轮廓。
在本发明实施例中，还提供了一种牙齿和牙槽骨的分割与重构装置，如图5所示，该牙齿和牙槽骨的分割与重构装置，包括：
骨骼组织分割模块501，用于在口腔三维电子计算机断层扫描CT图像中分割出骨骼组织区域；具体的，骨骼组织分割模块501是牙齿和牙槽骨的分割与重构装置中负责完成分割骨骼组织区域功能的部分，可以是软件、硬件或二者的结合，例如可以是完成该图像分割功能的处理芯片等元器件。
牙齿与牙槽骨组织分割模块502，用于从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；具体的，牙齿与牙槽骨组织分割模块502是牙齿和牙槽骨的分割与重构装置中负责完成牙齿与牙槽骨组织区域分割功能的部分，可以是软件、硬件或二者的结合，例如可以是完成该图像分割功能的处理芯片等元器件。
轮廓分割模块503，用于从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；具体的，轮廓分割模块503是牙齿和牙槽骨的分割与重构装置中负责完成牙齿与牙槽骨轮廓分割功能的部分，可以是软件、硬件或二者的结合，例如可以是完成该图像分割功能的处理芯片等元器件。
模型重构模块504，用于根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。具体的，模型重 构模块504是牙齿和牙槽骨的分割与重构装置中负责完成模型重构功能的部分，可以是软件、硬件或二者的结合，例如可以是完成该模型重构功能的处理芯片等元器件。
在本发明实施例中，通过在口腔三维CT图像中分割出骨骼组织区域，进而从骨骼组织区域中分割出牙齿与牙槽骨组织区域，最后再从牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，使得可以实现同时分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，进而实现根据各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据牙槽骨的二维轮廓重构牙槽骨的数字化三维模型，以利于实现数字化的正畸治疗辅助，使得在进行数字化的计算机正畸治疗辅助中，医师进行正畸治疗方案规划时可以对各独立牙齿进行交互式操作，同时，由于牙齿的牙根部位是固定和埋藏在牙槽骨中的，在治疗方案规划的排牙阶段时，可以保证牙齿的牙冠与牙根排列整齐外，还可以使牙根与牙槽骨保持良好的位置关系。
本领域内的技术人员应明白，本发明的实施例可提供为方法、系统、或计算机程序产品。因此，本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本发明可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或 其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
以上所述的具体实施例，对本发明的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本发明的具体实施例而已，并不用于限定本发明的保护范围，凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (11)

  1. 一种牙齿和牙槽骨的分割与重构方法，其特征在于，包括：
    在口腔三维电子计算机断层扫描CT图像中分割出骨骼组织区域；
    从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；
    从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；
    根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。
  2. 如权利要求1所述的方法，其特征在于，在口腔三维CT图像中分割出骨骼组织区域，包括：
    根据所述三维CT图像中像素的灰度值差异，在所述三维CT图像中分割出所述骨骼组织区域，所述骨骼组织区域包括下牙颌的牙齿与牙槽骨连通组织区域、上牙颌的牙齿与牙槽骨连通组织区域以及其他的颌骨组织区域。
  3. 如权利要求2所述的方法，其特征在于，从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域，包括：
    在三维空间中，对所述骨骼组织区域进行开运算；
    在开运算后的骨骼组织区域中对各连通区域的体积进行分析，根据体积大小在所述骨骼组织区域中提取出下牙颌的牙齿与牙槽骨连通组织区域和上牙颌的牙齿与牙槽骨连通组织区域。
  4. 如权利要求1至3中任一项所述的方法，其特征在于，从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓，包括：
    从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分别分割出待分割牙颌的各独立牙齿二维轮廓，将待分割牙颌的牙齿与牙槽骨组织区域中除去分割出的各独立牙齿二维轮廓所占区域外的区域确定为牙槽骨区域，从牙槽骨区域提取牙槽骨的二维轮廓。
  5. 如权利要求4所述的方法，其特征在于，从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分割出待分割牙颌的各独立牙齿的二维轮廓，包括：
    通过以下步骤从下牙颌或上牙颌的牙齿与牙槽骨组织区域中分别分割待分割牙颌各独立牙齿的二维轮廓：
    在已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片中，选择一张牙冠部位的切片为初始切片，在所述初始切片中，含有待分割牙颌牙列所有牙齿的完整牙冠；
    在所述初始切片中，在待分割牙颌的各牙齿的区域内部分别选择一个像素作为种子点，利用所述种子点从所述初始切片中的牙齿与牙槽骨区域中识别出待分割牙颌的牙齿区域；
    在所述识别出的牙齿区域中，获取分离相邻牙齿的分离线，得到所述初始切片中待分割牙颌的各独立牙齿的二维轮廓；
    对已分割出待分割牙颌的牙齿与牙槽骨组织区域的三维CT图像切片中除了所述初始切片之外的其他切片，循环逐个切片分割出各自切片中待分割牙颌的各独立牙齿的二维轮廓，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应的各独立牙齿的初始轮廓，继续进行牙齿轮廓的分割得到当前切片的分割结果，在当前切片分割出待分割牙颌的各独立牙齿的二维轮廓为空集时，结束待分割牙颌牙齿逐切片式的分割操作，否则，在后一切片中继续分割待分割牙颌各独立牙齿的二维轮廓。
  6. 如权利要求5所述的方法，其特征在于，在所述初始切片中，利用所述种子点识别出待分割牙颌的牙齿区域，包括：
    将牙齿与牙槽骨组织区域中与所述种子点相交的区域确定为待分割牙颌的牙齿区域。
  7. 如权利要求5所述的方法，其特征在于，从已分割出待分割牙颌牙齿与牙槽骨组织区域的三维CT图像的非所述初始切片中，分割出待分割牙颌的各独立牙齿的二维轮廓，包括：
    以所述初始切片为界将待分割牙颌的所有切片分为上、下两个部分，以所述初始切片为起始位置，分别朝着向上或向下的分割方向分割各部分中的各切片内独立牙齿的二维轮廓；
    在进行分割操作的当前切片中，以前一切片分割出的各独立牙齿的二维轮廓作为当前切片中对应各独立牙齿的初始轮廓，继续进行牙齿轮廓的分割得到当前切片的分割结果，包括：
    以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应各独立牙齿的初始轮廓，采用水平集算法迭代分割得到当前切片的分割结果。
  8. 如权利要求7所述的方法，其特征在于，在进行分割操作的当前切片中，以前一切片中分割出的各独立牙齿的二维轮廓作为当前切片中对应的各独立牙齿的初始轮廓，采用水平集算法迭代分割得到当前切片各独立牙齿的二维轮廓，包括：
    以前一切片中分割出的各独立牙齿的二维轮廓为当前切片中对应的各独立牙齿的初始轮廓，获取相邻牙齿之间的分离线并获取当前切片各牙齿的局部感兴趣区域，其中，对于每条分离线，其将与该分离线对应的待分割牙齿的局部感兴趣区域分为了两个区域，一个区域包含待分割牙齿，另一个区域包含待分割牙齿的相邻牙齿；
    在待分割牙齿的局部感兴趣区域中包含待分割牙齿的区域内，采用水平集算法迭代分割出待分割牙齿的二维轮廓。
  9. 如权利要求5所述的方法，其特征在于，获取分离相邻牙齿的分离线，包括：
    利用拉东变换获取分离相邻牙齿的分离线。
  10. 如权利要求7所述的方法，其特征在于，在当前切片分割出的各独立牙齿的二维轮廓为空集时，结束当前分割方向上各切片牙齿的分割操作，确定当前切片和沿当前分割方向在当前切片之后的各切片中不存在牙齿区域，确定当前切片和沿当前分割方向在当前切片之后的各切片中，待分割牙颌的牙齿与牙槽骨区域都为牙槽骨区域，提取出牙槽骨的二维轮廓。
  11. 一种牙齿和牙槽骨的分割与重构装置，其特征在于，包括：
    骨骼组织分割模块，用于在口腔三维电子计算机断层扫描CT图像中分割出骨骼组织区域；
    牙齿与牙槽骨组织分割模块，用于从所述骨骼组织区域中分割出牙齿与牙槽骨组织区域；
    轮廓分割模块，用于从所述牙齿与牙槽骨组织区域中分割出各独立牙齿的二维轮廓和牙槽骨的二维轮廓；
    模型重构模块，用于根据所述各独立牙齿的二维轮廓重构各独立牙齿的数字化三维模型，根据所述牙槽骨的二维轮廓重构牙槽骨的数字化三维模型。
PCT/CN2015/081262 2015-06-11 2015-06-11 牙齿和牙槽骨的分割与重构方法及装置 WO2016197370A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2015/081262 WO2016197370A1 (zh) 2015-06-11 2015-06-11 牙齿和牙槽骨的分割与重构方法及装置
CN201580000447.1A CN107106117B (zh) 2015-06-11 2015-06-11 牙齿和牙槽骨的分割与重构方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/081262 WO2016197370A1 (zh) 2015-06-11 2015-06-11 牙齿和牙槽骨的分割与重构方法及装置

Publications (1)

Publication Number Publication Date
WO2016197370A1 true WO2016197370A1 (zh) 2016-12-15

Family

ID=57502879

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/081262 WO2016197370A1 (zh) 2015-06-11 2015-06-11 牙齿和牙槽骨的分割与重构方法及装置

Country Status (2)

Country Link
CN (1) CN107106117B (zh)
WO (1) WO2016197370A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109754870A (zh) * 2019-03-12 2019-05-14 北京羽医甘蓝信息技术有限公司 基于深度学习的全景片牙槽骨吸收分级的方法和装置
CN111275710A (zh) * 2019-12-06 2020-06-12 上海正雅齿科科技股份有限公司 一种牙颌模型分割方法及装置
CN111388125A (zh) * 2020-03-05 2020-07-10 深圳先进技术研究院 一种口腔正畸治疗前后牙齿移动量计算方法和装置
CN115177387A (zh) * 2021-04-01 2022-10-14 北京大学口腔医学院 一种口腔修复的设计方法及系统
CN117437250A (zh) * 2023-12-21 2024-01-23 天津医科大学口腔医院 一种基于深度学习的三维牙颌图像分割方法及系统

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108665533A (zh) * 2018-05-09 2018-10-16 西安增材制造国家研究院有限公司 一种通过牙齿ct图像和三维扫描数据重建牙列的方法
CN109978841B (zh) * 2019-03-12 2021-08-03 北京羽医甘蓝信息技术有限公司 基于深度学习的全景片阻生牙识别的方法和装置
CN110619646B (zh) * 2019-07-23 2023-08-29 同济大学 一种基于全景图的单牙提取方法
KR102453932B1 (ko) * 2020-12-28 2022-10-14 주식회사 크리에이티브마인드 3차원 치아이미지에서 치아영역을 인식하는 방법 및 그 장치
CN113240681B (zh) * 2021-05-20 2022-07-08 推想医疗科技股份有限公司 图像处理的方法及装置
CN113409333B (zh) * 2021-06-16 2022-07-22 青岛海信医疗设备股份有限公司 一种三维图像的切割方法及电子设备
CN113506302B (zh) * 2021-07-27 2023-12-12 四川九洲电器集团有限责任公司 一种交互式对象更新方法、装置及处理系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1516074A (zh) * 2002-12-28 2004-07-28 三星电子株式会社 从牙齿图像提取牙齿区域的方法及个人识别方法和装置
CN101393653A (zh) * 2008-10-16 2009-03-25 浙江大学 一种通过牙颌石膏模型ct数据和牙颌的全景透视图重建全牙的三维模型方法
CN101950430A (zh) * 2010-09-23 2011-01-19 北京大学 基于曲面断层片的三维牙齿重建方法
KR101099732B1 (ko) * 2010-07-30 2011-12-28 경희대학교 산학협력단 컴퓨터 단층촬영영상 및 자기공명영상에서 3차원 개별 치아 모델링을 위한 치아 분리 시스템
US20130308846A1 (en) * 2011-07-21 2013-11-21 Carestream Health, Inc. Method for teeth segmentation and alignment detection in cbct volume
CN104240241A (zh) * 2014-09-05 2014-12-24 哈尔滨医科大学 一种Micro-CT图像分析量化评估磨牙牙槽骨的方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101706971A (zh) * 2009-11-27 2010-05-12 西安科技大学 一种牙齿模型中牙冠的自动分割方法
CN102254317A (zh) * 2011-03-25 2011-11-23 苏州迪凯尔医疗科技有限公司 牙种植导航中牙弓曲面的自动提取方法
WO2013060563A1 (en) * 2011-10-25 2013-05-02 Universität Bern Device for treating gingiva/ mucosa at teeth or implants
US8605973B2 (en) * 2012-03-17 2013-12-10 Sony Corporation Graph cuts-based interactive segmentation of teeth in 3-D CT volumetric data

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1516074A (zh) * 2002-12-28 2004-07-28 三星电子株式会社 从牙齿图像提取牙齿区域的方法及个人识别方法和装置
CN101393653A (zh) * 2008-10-16 2009-03-25 浙江大学 一种通过牙颌石膏模型ct数据和牙颌的全景透视图重建全牙的三维模型方法
KR101099732B1 (ko) * 2010-07-30 2011-12-28 경희대학교 산학협력단 컴퓨터 단층촬영영상 및 자기공명영상에서 3차원 개별 치아 모델링을 위한 치아 분리 시스템
CN101950430A (zh) * 2010-09-23 2011-01-19 北京大学 基于曲面断层片的三维牙齿重建方法
US20130308846A1 (en) * 2011-07-21 2013-11-21 Carestream Health, Inc. Method for teeth segmentation and alignment detection in cbct volume
CN104240241A (zh) * 2014-09-05 2014-12-24 哈尔滨医科大学 一种Micro-CT图像分析量化评估磨牙牙槽骨的方法及装置

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109754870A (zh) * 2019-03-12 2019-05-14 北京羽医甘蓝信息技术有限公司 基于深度学习的全景片牙槽骨吸收分级的方法和装置
CN111275710A (zh) * 2019-12-06 2020-06-12 上海正雅齿科科技股份有限公司 一种牙颌模型分割方法及装置
CN111275710B (zh) * 2019-12-06 2024-02-27 正雅齿科科技(上海)有限公司 一种牙颌模型分割方法及装置
CN111388125A (zh) * 2020-03-05 2020-07-10 深圳先进技术研究院 一种口腔正畸治疗前后牙齿移动量计算方法和装置
CN111388125B (zh) * 2020-03-05 2022-03-08 深圳先进技术研究院 一种口腔正畸治疗前后牙齿移动量计算方法和装置
CN115177387A (zh) * 2021-04-01 2022-10-14 北京大学口腔医学院 一种口腔修复的设计方法及系统
CN115177387B (zh) * 2021-04-01 2024-02-20 北京大学口腔医学院 一种口腔修复的设计方法及系统
CN117437250A (zh) * 2023-12-21 2024-01-23 天津医科大学口腔医院 一种基于深度学习的三维牙颌图像分割方法及系统
CN117437250B (zh) * 2023-12-21 2024-04-02 天津医科大学口腔医院 一种基于深度学习的三维牙颌图像分割方法及系统

Also Published As

Publication number Publication date
CN107106117A (zh) 2017-08-29
CN107106117B (zh) 2020-07-03

Similar Documents

Publication Publication Date Title
WO2016197370A1 (zh) 牙齿和牙槽骨的分割与重构方法及装置
US11651494B2 (en) Apparatuses and methods for three-dimensional dental segmentation using dental image data
CN112638312B (zh) 训练深度神经网络的系统和确定牙齿的最终位置的系统
AU2017281290B2 (en) Method for estimating at least one of shape, position and orientation of a dental restoration
US11645746B2 (en) Dental image segmentation and registration with machine learning
US11464467B2 (en) Automated tooth localization, enumeration, and diagnostic system and method
US11734825B2 (en) Segmentation device and method of generating learning model
JP2022000211A (ja) 歯列メッシュ矯正具除去のための方法および装置
KR101744080B1 (ko) 치과 시술 시뮬레이션을 위한 치아모델 생성 방법
CA3162711A1 (en) Method, system and computer readable storage media for creating three-dimensional dental restorations from two dimensional sketches
CN111784754B (zh) 基于计算机视觉的牙齿正畸方法、装置、设备及存储介质
US20220084267A1 (en) Systems and Methods for Generating Quick-Glance Interactive Diagnostic Reports
CN114757960B (zh) 一种基于cbct图像的牙齿分割与重建方法及存储介质
US20220361992A1 (en) System and Method for Predicting a Crown and Implant Feature for Dental Implant Planning
CN106214175B (zh) 一种估算三维牙轴的方法和装置
Deleat-Besson et al. Automatic segmentation of dental root canal and merging with crown shape
Gao et al. Automatic tooth region separation for dental CT images
Ben-Hamadou et al. 3DTeethSeg'22: 3D Teeth Scan Segmentation and Labeling Challenge
Grzegorzek et al. A multi-stage approach for 3D teeth segmentation from dentition surfaces
US20220358740A1 (en) System and Method for Alignment of Volumetric and Surface Scan Images
Deleat-Besson et al. Merging and annotating teeth and roots from automated segmentation of multimodal images
KR20220145758A (ko) 컴퓨팅 장치를 이용한 하악골 비대칭 평가 방법 및 이를 이용한 수술 시뮬레이션 방법
Yuan et al. Bio-information based segmentation of 3D dental models
US20230252748A1 (en) System and Method for a Patch-Loaded Multi-Planar Reconstruction (MPR)
US20230051400A1 (en) System and Method for Fusion of Volumetric and Surface Scan Images

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15894640

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 16.02.2018)

122 Ep: pct application non-entry in european phase

Ref document number: 15894640

Country of ref document: EP

Kind code of ref document: A1