WO2018035837A1 - 一种估算三维牙轴的方法和装置 - Google Patents
一种估算三维牙轴的方法和装置 Download PDFInfo
- Publication number
- WO2018035837A1 WO2018035837A1 PCT/CN2016/096851 CN2016096851W WO2018035837A1 WO 2018035837 A1 WO2018035837 A1 WO 2018035837A1 CN 2016096851 W CN2016096851 W CN 2016096851W WO 2018035837 A1 WO2018035837 A1 WO 2018035837A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- dimensional
- tooth
- image
- projection
- dental
- Prior art date
- Legal status (The legal status 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 status listed.)
- Ceased
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61C—DENTISTRY; APPARATUS OR METHODS FOR ORAL OR DENTAL HYGIENE
- A61C19/00—Dental auxiliary appliances
- A61C19/04—Measuring instruments specially adapted for dentistry
Definitions
- the present application relates to image processing, and more particularly to a method and apparatus for estimating a three-dimensional dental shaft.
- the present application provides a method and apparatus for estimating a three-dimensional dental shaft.
- the present application provides a method for estimating a three-dimensional dental shaft, comprising:
- a three-dimensional dental CT image is converted into a plurality of two-dimensional projection images, and three-dimensional dental axes are extracted from the two-dimensional projection images.
- the three-dimensional dental CT image is converted into a plurality of two-dimensional projection images, and the three-dimensional dental axes are extracted according to the two-dimensional projection image, which specifically includes:
- the above method, the dividing the two-dimensional projection image into a tooth contour specifically includes: [0014] preprocessing the two-dimensional projection image;
- the two-dimensional contour of the tooth is extracted by automatic or manual interaction.
- the above method, the extracting the two-dimensional axis according to the tooth contour specifically includes:
- a two-dimensional axis is extracted on the two-dimensional projected image, the extracted two-dimensional axis including a straight line along the long axis direction of the tooth contour and passing through the centroid point of the tooth contour.
- the three-dimensional dental CT image is projected to obtain a two-dimensional projection image, and specifically includes:
- the tooth alveolar bone region determined by the two-dimensional tooth axis of the first two-dimensional projection image on the left and right parallel axes of the tooth contour is a projection area to pass the two-dimensional tooth axis of the first two-dimensional projection image
- the normal of the plane is the projection direction, and the projection area is projected to obtain a plurality of two-dimensional projection images.
- the present application provides an apparatus for estimating a three-dimensional dental axis, the apparatus for converting a three-dimensional oral CT image into a plurality of two-dimensional projection images, and extracting according to the two-dimensional projection image Three-dimensional dental shaft.
- the device includes:
- a projection module configured to perform multiple projections on the three-dimensional dental CT image to obtain a plurality of two-dimensional projection images
- a segmentation module configured to respectively segment the tooth contours in the two-dimensional projection image
- an extraction module configured to extract a two-dimensional axis according to the tooth contour
- a synthesis module configured to synthesize a three-dimensional dental shaft according to the plurality of the two-dimensional axes.
- the segmentation module is further configured to preprocess the two-dimensional projection image, and extract the extracted two-dimensional contour of the tooth by using an automatic or manual interaction method on the pre-processed two-dimensional projection image.
- the extraction module is further configured to extract a two-dimensional axis on the two-dimensional projection image, the extracted two-dimensional axis includes a longitudinal axis along the contour of the tooth, and passes through a centroid of the contour of the tooth The line of points.
- the projection module is further configured to remove a tooth alveolar bone region from the three-dimensional oral CT image Extracting from the image, obtaining a dental arch curve from the tooth alveolar bone region, projecting the dental alveolar bone region in a projection direction of the dental arch curve to obtain a first two-dimensional projection image; a method in which a two-dimensional tooth axis of the first two-dimensional projection image defines a tooth alveolar bone region defined by left and right parallel axes of the tooth contour as a projection region, and a plane of the two-dimensional tooth axis of the first two-dimensional projection image The line is a projection direction, and the projection area is projected to obtain a plurality of two-dimensional projection images.
- the present application since the three-dimensional oral CT image is converted into multiple two-dimensional projection images, and the three-dimensional dental axis is extracted according to the two-dimensional projection image, the present application comprehensively considers the crown and root information, and only needs to The contouring of the teeth on multiple two-dimensional projection images eliminates the need for complete tooth segmentation in the patient's oral CT image, and the efficiency of the dental axis estimation is higher. At the same time, since the difference between the root and alveolar bone in the two-dimensional projection image is more obvious than the difference between the original oral CT images, the contour segmentation of the teeth is easier, and the estimated three-dimensional dental axes are more accurate.
- FIG. 1 is a flow chart of a method of the present application in an embodiment
- FIG. 2 is a flow chart of obtaining a two-dimensional tooth contour in one embodiment of the method of the present application
- FIG. 3 is a flow chart of obtaining a two-dimensional projected image in one embodiment of the method of the present application.
- FIG. 4 is a flow chart of obtaining a two-dimensional projected image in another embodiment of the method of the present application.
- FIG. 5 is a schematic diagram of functional modules of an apparatus of the present application in an embodiment.
- the method of estimating a three-dimensional dental shaft of the present application comprises three stages. The first stage is the extraction of the first two-dimensional axis.
- the first projection of the input original oral CT image is performed, and then the two-dimensional contour of the tooth is segmented in the projected image, and the two-dimensional axis of the tooth is extracted.
- the segmentation result is recorded as the tooth two-dimensional contour 1
- the extraction result is recorded as the tooth two-dimensional axis 1.
- the second stage is the extraction of the second two-dimensional axis.
- the input original oral CT image is subjected to the second projection (the second projection direction is different from the first projection direction), and then the two-dimensional contour of the tooth is segmented on the projection image, and the two-dimensional axis of the tooth is extracted.
- the segmentation result is recorded as the tooth 2D contour 2, and the extraction result is recorded as the tooth 2D axis 2.
- the third stage is the synthesis of three-dimensional dental axes.
- the three-dimensional axis that is, the estimated three-dimensional dental axis of the tooth, is synthesized based on the two two-dimensional axes extracted, that is, the two-dimensional axis 1 of the tooth and the two-dimensional axis 2 of the tooth.
- the technique does not require the complete segmentation of the teeth on hundreds of oral CT images, only the contour segmentation of each tooth in two projection images; and the difference between the root and the alveolar bone in the two projections is smaller than the original
- the difference between the two in the oral CT image is more pronounced, the contouring of the teeth is easier, and the estimated three-dimensional dental axes are more accurate.
- Embodiment 1 is a diagrammatic representation of Embodiment 1:
- the method for estimating a three-dimensional dental shaft of the present application includes:
- the three-dimensional dental CT image is converted into a plurality of two-dimensional projection images, and the three-dimensional dental axes are extracted from the two-dimensional projection images.
- another method for estimating a three-dimensional dental shaft of the present application includes the following steps:
- Step 102 Perform multiple projections on the three-dimensional oral CT image to obtain a plurality of two-dimensional projection images.
- Step 104 Segment the tooth contours in the two-dimensional projection image.
- Step 106 Extract a two-dimensional axis from the contour of the tooth.
- Step 108 synthesize a three-dimensional dental shaft according to a plurality of two-dimensional axes.
- Step 104 specifically, the method may include:
- Step 1042 Perform preprocessing on the two-dimensional projection image
- Step 1044 Extracting the two-dimensional contour of the tooth by using an automatic or manual interaction method on the pre-processed two-dimensional projection image.
- An automatic or manual interaction method on the pre-processed two-dimensional projection image.
- Image preprocessing is first performed on the two-dimensional projection image.
- adaptive pre-processing algorithm Otsu threshold algorithm, high and low hat filtering, butterfly filtering, bilateral filtering, adaptive scaling, etc. can be used to perform image preprocessing to improve tooth contour in 2D projection image. The clarity of the border.
- nth (n ⁇ 2) The projected image contains only one tooth, which is itself a separate tooth area, so there is no need to isolate the adjacent teeth described below.
- the gray value between the upper and lower jaws and the boundary between adjacent teeth is lower, so it is possible to use gray between the upper and lower jaws and between adjacent teeth.
- Straight lines or curves formed by lower pixels divide the upper and lower jaws and adjacent teeth to isolate the isolated teeth.
- the two-dimensional projection image will be divided into small pieces by these separation lines, and there is only one tooth in each small area, that is, the isolation of the independent teeth is realized on the two-dimensional projection image.
- step 106 may specifically include:
- the first two-dimensional projection image is obtained by projecting the three-dimensional dental CT image, which may specifically include:
- Step 1022 extracting a tooth alveolar bone region in a three-dimensional oral CT image
- Step 1024 extracting an arch curve according to the alveolar bone region of the tooth
- Step 1026 Projecting the tooth alveolar bone region with the normal direction of the arch curve as a projection direction
- the first projection image is obtained.
- the composition flow of the first projection image is as shown in FIG. 3.
- the projection area of the two-dimensional projection image 1 is defined in the dental alveolar bone region in the oral CT image.
- the process of extracting the alveolar bone region in the oral CT image is as follows: In the oral CT image, since the gray scale difference between the bone tissue (tooth, alveolar bone, other jaw tissue) and other tissues is large, a threshold can be used. Segmentation and other methods extract bone tissue from oral CT images. For the extracted bone tissue, the alveolar bone tissue in the bone tissue and other bone tissues such as the neck bone are broken by a ⁇ algorithm. In this case, the bone tissue is divided into several parts, and the largest three-dimensional connected area is the tooth alveolar bone area.
- the dental arch curve is obtained according to the extracted tooth alveolar bone region, and the normal line of the dental arch curve is used as the projection direction.
- the arch curve can be extracted
- the maximum grayscale projection of the tooth alveolar bone region is obtained by refinement, manual mapping combined with spline curve or other polynomial curve fitting.
- the first projected image is obtained by projecting the extracted alveolar bone region along the normal line of the arch curve.
- step 102 the obtaining of the nth (n ⁇ 2) projection images by the projection of the three-dimensional dental CT image may specifically include:
- Step 1028 The synthesis flow of the nth (n ⁇ 2) projection images is as shown in FIG. 4.
- the area of the alveolar bone defined by the two-dimensional axis of the first projected image on the left and right parallel axes of the contour of the tooth is the projection area.
- the projection area is projected by the normal line of the two-dimensional tooth axis passing through the first projected image, and the nth (n ⁇ 2) projection images are obtained.
- Embodiment 2 is a diagrammatic representation of Embodiment 1
- the apparatus for estimating a three-dimensional dental axis of the present application is for converting a three-dimensional dental CT image into a plurality of two-dimensional projection images, and extracting three-dimensional dental axes according to the two-dimensional projection images.
- another embodiment of the apparatus for estimating a three-dimensional dental shaft of the present application may include a projection module, a segmentation module, an extraction module, and a synthesis module.
- the projection module is configured to perform multiple projections on the three-dimensional oral CT image to obtain a plurality of two-dimensional projection images;
- the segmentation module is configured to respectively segment the two-dimensional projection image out of the tooth contour;
- the extraction module is configured to extract the two-dimensional axis according to the contour of the tooth a synthesis module for synthesizing a three-dimensional dental axis from a plurality of two-dimensional axes.
- the segmentation module is further configured to preprocess the two-dimensional projection image, and extract the two-dimensional contour of the tooth by using an automatic or manual interaction manner on the pre-processed two-dimensional projection image.
- FIG. 1 One embodiment of tooth contour segmentation in a synthetic two-dimensional projection image is shown in FIG. Image preprocessing is first performed on the 2D projection image.
- adaptive pre-processing algorithm Otsu threshold algorithm, high and low hat filtering, butterfly filtering, bilateral filtering, adaptive scaling, etc. can be used to perform image preprocessing to improve tooth contour in two-dimensional projection image. The clarity of the border.
- the nth (n ⁇ 2) projection images contain only one tooth, which is itself a separate tooth region, so there is no need to isolate the adjacent teeth described below.
- the gray value between the upper and lower jaws and the boundary between adjacent teeth is lower, so it is possible to use gray between the upper and lower jaws and between adjacent teeth.
- a straight line or curve composed of lower pixels divides the upper and lower jaws and adjacent teeth to achieve independent teeth The tooth is isolated.
- the two-dimensional projection image will be divided into small pieces by these separation lines, and there is only one tooth in each small area, that is, the isolation of the independent teeth is realized on the two-dimensional projection image.
- the extraction module of the present application is further configured to extract a two-dimensional axis on the two-dimensional projection image, the extracted two-dimensional axis including a straight line along the longitudinal direction of the tooth contour and passing through the centroid point of the tooth contour.
- the projection module is further configured to acquire the first and nth (n ⁇ 2) projection images.
- the first projection image is obtained by: extracting a tooth alveolar bone region in a three-dimensional oral CT image; extracting an arch curve according to the alveolar bone region; and performing the dental arch curve on the alveolar bone region of the tooth
- the normal direction is projected in the projection direction to obtain the first projected image.
- the composition flow of the first projected image is as shown in FIG. 3.
- the present application limits the projection area of the two-dimensional projection image 1 to the alveolar bone region in the oral CT image.
- the process of extracting the alveolar bone region in the oral CT image is as follows: In the oral CT image, since the gray scale difference between the bone tissue (tooth, alveolar bone, other jaw tissue) and other tissues is large, a threshold can be used. Segmentation and other methods extract bone tissue from oral CT images. For the extracted bone tissue, the alveolar bone tissue and other bone tissues such as the neck bone in the bone tissue are broken by a sputum algorithm. In this case, the bone tissue is divided into several parts, and the largest three-dimensional connected area is the tooth alveolar bone area.
- the dental arch curve is obtained according to the extracted tooth alveolar bone region, and the normal line of the dental arch curve is used as the projection direction.
- the arch curve can be obtained by refining, manual drawing combined with spline curve or other polynomial curve fitting on the maximum grayscale projection of the extracted alveolar bone region.
- the first projection image is obtained by projecting the extracted alveolar bone region along the normal line of the dental arch curve.
- Obtaining the three-dimensional oral CT image to obtain the nth (n ⁇ 2) projection images may specifically include: the dental teeth determined by the two-dimensional tooth axis of the first projection image on the left and right parallel axes of the tooth contours
- the trough area is the projection area. Projecting the projection area with a normal to the two-dimensional tooth axis of the first projected image as a projection direction, That is, the nth (n ⁇ 2) projection images are obtained.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Epidemiology (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
一种估算三维牙轴的方法，包括：将三维口腔CT图像进行多次投影获得多张二维投影图像（102）；在所述二维投影图像中分别分割出牙齿轮廓（104）；根据所述牙齿轮廓提取二维轴线（106）；根据多个所述二维轴线合成三维牙轴（108）。还公开了估算三维牙轴方法相应的装置。
Description
发明名称：一种估算三维牙轴的方法和装置 技术领域
[0001] 本申请涉及图像处理， 尤其涉及一种估算三维牙轴的方法和装置。
[0002] 背景技术
[0003] 在口腔正畸与牙齿种植中， 口腔医师需要借助牙轴进行治疗规划， 牙轴的不准 确和缺失可能导致方案规划的不合理和失败。 现有很多技术仅利用牙冠进行牙 轴估算， 这样估算的牙轴因不能反映牙根的方位， 结果不准确； 且仅考虑牙冠 信息不能确定仅有部分牙冠萌出或无牙冠萌出的阻生牙和埋伏牙的牙轴。 随着 计算机技术和口腔 CT扫描技术的发展， 口腔 CT图像在口腔正畸与口腔修复中受 到广泛应用。 有些技术虽然利用口腔 CT图像中完整的牙冠与牙根信息进行牙轴 估算， 但都需要首先从口腔 CT图像中分割出完整的牙齿。 由于口腔 CT图像中牙 根与牙槽骨灰度接近， 完整牙齿的分割非常困难， 这些技术容易造成牙轴估算 失败。 同吋， 同一患者的口腔 CT图像一般有几百张， 分割耗吋长， 效率低。
[0004] 发明内容
[0005] 本申请提供一种估算三维牙轴的方法和装置。
[0006] 根据本申请的第一方面， 本申请提供一种估算三维牙轴的方法， 其特征在于， 包括：
[0007] 将三维口腔 CT图像转换为多张二维投影图像， 根据所述二维投影图像提取三 维牙轴。
[0008] 上述方法， 所述将三维口腔 CT图像转换为多张二维投影图像， 根据所述二维 投影图像提取三维牙轴， 具体包括：
[0009] 对三维口腔 CT图像进行多次投影获得多张二维投影图像；
[0010] 在所述二维投影图像中分别分割出牙齿轮廓；
[0011] 根据所述牙齿轮廓提取二维轴线；
[0012] 根据多个所述二维轴线合成三维牙轴。
[0013] 上述方法， 所述将所述二维投影图像分割出牙齿轮廓， 具体包括：
[0014] 对所述二维投影图像进行预处理；
[0015] 采用自动或手动交互方式分割提取出牙齿二维轮廓。
[0016] 上述方法， 所述根据所述牙齿轮廓提取二维轴线， 具体包括：
[0017] 在二维投影图像上提取二维轴线， 提取的二维轴线包括沿所述牙齿轮廓的长轴 方向， 并通过所述牙齿轮廓的质心点的直线。
[0018] 上述方法， 对三维口腔 CT图像进行投影获得二维投影图像， 具体包括：
[0019] 将牙齿牙槽骨区域从所述三维口腔 CT图像中提取出来；
[0020] 从所述牙齿牙槽骨区域获取牙弓曲线；
[0021] 将所述牙齿牙槽骨区域以所述牙弓曲线的法线方向为投影方向进行投影， 得到 第一张二维投影图像；
[0022] 以所述第一张二维投影图像的二维牙齿轴线在其牙齿轮廓的左右平行轴线所确 定的牙齿牙槽骨区域为投影区域， 以过所述第一张二维投影图像的二维牙齿轴 线的平面的法线为投影方向， 对所述投影区域进行投影， 得到多张二维投影图 像。
[0023] 根据本申请的第二方面， 本申请提供一种估算三维牙轴的装置， 所述装置， 用 于将三维口腔 CT图像转换为多张二维投影图像， 并根据所述二维投影图像提取 三维牙轴。
[0024] 上述装置中， 所述装置包括：
[0025] 投影模块， 用于对三维口腔 CT图像进行多次投影获得多张二维投影图像； [0026] 分割模块， 用于在所述二维投影图像中分别分割出牙齿轮廓；
[0027] 提取模块， 用于根据所述牙齿轮廓提取二维轴线；
[0028] 合成模块， 用于根据多个所述二维轴线合成三维牙轴。
[0029] 上述装置中， 所述分割模块， 还用于对所述二维投影图像进行预处理， 对预处 理后的二维投影图像， 采用自动或手动交互方式分割提取出牙齿二维轮廓。
[0030] 上述装置中， 所述提取模块， 还用于在二维投影图像上提取二维轴线， 提取的 二维轴线包括沿所述牙齿轮廓的长轴方向， 并通过所述牙齿轮廓的质心点的直 线。
[0031] 上述装置中， 所述投影模块， 还用于将牙齿牙槽骨区域从所述三维口腔 CT图
像中提取出来， 从所述牙齿牙槽骨区域获取牙弓曲线， 将所述牙齿牙槽骨区域 以所述牙弓曲线的法线方向为投影方向进行投影， 得到第一张二维投影图像； 以所述第一张二维投影图像的二维牙齿轴线在其牙齿轮廓的左右平行轴线所确 定的牙齿牙槽骨区域为投影区域， 以过所述第一张二维投影图像的二维牙齿轴 线的平面的法线为投影方向， 对所述投影区域进行投影， 得到多张二维投影图 像。
[0032] 由于采用了以上技术方案， 使本申请具备的有益效果在于：
[0033] 在本申请的具体实施方式中， 由于将三维口腔 CT图像转换为多张二维投影图 像， 根据二维投影图像提取三维牙轴， 本申请综合考虑了牙冠和牙根信息， 只 需要对每颗牙齿在多张二维投影图像上进行轮廓分割， 无需在患者口腔 CT图像 中进行完整牙齿的分割， 牙轴估算效率更高。 同吋， 由于二维投影图像上牙根 与牙槽骨的差异比原始的口腔 CT图像中两者的差异更明显一些， 牙齿轮廓分割 更加容易， 且估算的三维牙轴也更准确。
[0034] 附图说明
[0035] 图 1为本申请的方法在一种实施方式中的流程图；
[0036] 图 2为本申请的方法在一种实施方式中获得二维牙齿轮廓的流程图；
[0037] 图 3为本申请的方法在一种实施方式中获得二维投影图像的流程图；
[0038] 图 4为本申请的方法在另一种实施方式中获得二维投影图像的流程图；
[0039] 图 5为本申请的装置在一种实施方式中的功能模块示意图。
[0040] 具体实施方式
[0041] 下面通过具体实施方式结合附图对本申请作进一步详细说明。
[0042] 本申请的估算三维牙轴的方法， 首先对口腔 CT图像进行 n(n≥2)次投影， 然后在 n张二维投影图像里分别分割出牙齿轮廓、 提取牙齿轮廓的二维轴线， 最后根据 n根二维轴线合成一根三维轴线， 即所求牙齿的三维牙轴。 随着 n的值增大， 即 投影次数增多， 估算的牙轴精确度增高。 以下以 n=2为例进行说明。 本申请的估 算三维牙轴的方法， 包括三个阶段。 第一阶段即第 1根二维轴线的提取。 对输入 的原始口腔 CT图像进行第 1次投影， 然后在投影图像中分割牙齿二维轮廓、 提取 牙齿二维轴线。 分割结果记为牙齿二维轮廓 1， 提取结果记为牙齿二维轴线 1。
第二阶段即第 2根二维轴线的提取。 对输入的原始口腔 CT图像进行第 2次投影 （ 第 2次投影方向与第 1次投影方向不同） ， 然后在投影图像上分割牙齿二维轮廓 、 提取牙齿二维轴线。 分割结果记为牙齿二维轮廓 2， 提取结果记为牙齿二维轴 线 2。 第三阶段即三维牙轴的合成。 根据提取得到的两根二维轴线即牙齿二维轴 线 1和牙齿二维轴线 2来合成三维轴线， 即所估计的牙齿的三维牙轴。 本技术不 需要在几百张的口腔 CT图像上进行完整牙齿的分割， 只需要对每颗牙齿在 2张投 影图像中进行轮廓分割； 且两张投影图上牙根与牙槽骨的差异比原始的口腔 CT 图像中两者的差异更明显， 牙齿轮廓分割更加容易， 所估算的三维牙轴更为准 确。
[0043] 实施例一：
[0044] 本申请的估算三维牙轴的方法， 其一种实施方式， 包括：
[0045] 将三维口腔 CT图像转换为多张二维投影图像， 根据二维投影图像提取三维牙 轴。
[0046] 如图 1所示， 本申请的估算三维牙轴的方法， 其另一种实施方式， 包括以下步 骤：
[0047] 步骤 102: 对三维口腔 CT图像进行多次投影获得多张二维投影图像。
[0048] 步骤 104: 在二维投影图像中分别分割出牙齿轮廓。
[0049] 步骤 106: 根据牙齿轮廓提取二维轴线。
[0050] 步骤 108: 根据多个二维轴线合成三维牙轴。
[0051] 其中步骤 104， 具体可以包括：
[0052] 步骤 1042: 对二维投影图像进行预处理；
[0053] 步骤 1044: 对预处理后的二维投影图像， 采用自动或手动交互方式分割提取出 牙齿二维轮廓。 在合成的二维投影图像中牙齿轮廓分割的一种实施方式如图 2所 示。 首先对二维投影图像进行图像预处理。 这里可以采用自适应阈值算法、 Otsu 阈值算法、 高低帽滤波、 butterfly滤波、 双边滤波、 自适应比例变换等方法中的 一种或几种结合来进行图像预处理， 提高二维投影图像中牙齿轮廓边界的清晰 度。
[0054] 由于第 1张投影图像中包含了所有的牙齿， 需要实现相邻牙齿的孤立。 而第 n(n
≥2)张投影图像上仅包含一颗牙齿， 其本身即为独立牙齿区域， 所以不需要进行 下述相邻牙齿的孤立。 在预处理之后的第 1张投影图像上， 上下牙颌之间的缝隙 处及相邻牙齿之间的边界处灰度值较低， 因此可以用上下牙颌之间及相邻牙齿 之间灰度较低像素构成的直线或曲线将上下牙颌及相邻牙齿分幵， 实现独立牙 齿的孤立。 此吋， 二维投影图像将被这些分离线分割成小块， 每一小块区域中 只有一颗牙齿， 即在二维投影图像上实现了独立牙齿的孤立。
[0055] 在独立牙齿区域中， 牙根周围常常会粘连一些牙槽骨组织， 因此可以采用阈值 分割、 形态学运算、 形状分析等算法去除牙根周围的部分牙槽骨组织， 并以此 吋独立牙齿区域中面积最大的联通区域的边界作为初始牙齿轮廓。 然后， 可采 用 Moore边界追踪算法、 gmph-cut算法、 活动轮廓、 水平集等算法优化初始牙齿 轮廓， 即得二维牙齿轮廓。
[0056] 其中步骤 106， 具体可以包括：
[0057] 在二维投影图像上提取二维轴线， 提取的二维轴线包括沿牙齿轮廓的长轴方向
， 并通过牙齿轮廓的质心点的直线。
[0058] 步骤 102中， 对三维口腔 CT图像进行投影获得第 1张二维投影图像， 具体可以 包括：
[0059] 步骤 1022: 在三维口腔 CT图像中提取牙齿牙槽骨区域；
[0060] 步骤 1024: 根据牙齿牙槽骨区域提取牙弓曲线；
[0061] 步骤 1026: 对牙齿牙槽骨区域以所述牙弓曲线的法线方向为投影方向进行投影
， 得第 1张投影图像。
[0062] 第 1张投影图像的合成流程如图 3所示。 本申请中将二维投影图像 1的投影区域 限定在口腔 CT图像中牙齿牙槽骨区域。 在口腔 CT图像中提取牙齿牙槽骨区域的 过程如下： 在口腔 CT图像中， 由于骨骼组织 （牙齿、 牙槽骨、 其它颌骨组织） 与其他组织之间灰度差异较大， 可采用阈值分割等方法从口腔 CT图像中提取骨 骼组织。 对提取出的骨骼组织， 采用幵运算算法等方法将骨骼组织中的牙齿牙 槽骨组织与颈骨等其他骨骼组织断幵。 此吋， 骨骼组织被分为几部分， 其中体 积最大的三维连通区域即为牙齿牙槽骨区域。 然后， 根据提取得到的牙齿牙槽 骨区域获取牙弓曲线， 以牙弓曲线的法线为投影方向。 牙弓曲线可以在提取得
到的牙齿牙槽骨区域的最大灰度投影图上通过细化、 手动描点法结合样条曲线 或其他多项式曲线拟合等方法得到。 对提取得到的牙齿牙槽骨区域沿着牙弓曲 线的法线进行投影即得第 1张投影图像。
[0063] 步骤 102中， 对三维口腔 CT图像进行投影获得第 n(n≥2)张投影图像的获取具体 可以包括：
[0064] 步骤 1028: 第 n(n≥2)张投影图像的合成流程如图 4所示。 以所述第 1张投影图像 的二维牙齿轴线在其牙齿轮廓的左右平行轴线所确定的牙齿牙槽骨区域为投影 区域。 以过所述第 1张投影图像的二维牙齿轴线的法线为投影方向， 对所述投影 区域进行投影， 即得第 n(n≥2)张投影图像。
[0065] 实施例二：
[0066] 本申请的估算三维牙轴的装置， 用于将三维口腔 CT图像转换为多张二维投影 图像， 并根据二维投影图像提取三维牙轴。
[0067] 如图 5所示， 本申请的估算三维牙轴的装置， 其另一种实施方式， 可以包括投 影模块、 分割模块、 提取模块和合成模块。 其中， 投影模块， 用于对三维口腔 C T图像进行多次投影获得多张二维投影图像； 分割模块， 用于分别将二维投影图 像分割出牙齿轮廓； 提取模块， 用于根据牙齿轮廓提取二维轴线； 合成模块， 用于根据多个二维轴线合成三维牙轴。
[0068] 在一种实施方式中， 分割模块还用于对二维投影图像进行预处理， 对预处理后 的二维投影图像， 采用自动或手动交互方式分割提取出牙齿二维轮廓。
[0069] 在合成的二维投影图像中牙齿轮廓分割的一种实施方式如图 2所示。 首先对二 维投影图像进行图像预处理。 这里可以采用自适应阈值算法、 Otsu阈值算法、 高 低帽滤波、 butterfly滤波、 双边滤波、 自适应比例变换等方法中的一种或几种结 合来进行图像预处理， 提高二维投影图像中牙齿轮廓边界的清晰度。
[0070] 由于第 1张投影图像中包含了所有的牙齿， 需要实现相邻牙齿的孤立。 而第 n(n ≥2)张投影图像上仅包含一颗牙齿， 其本身即为独立牙齿区域， 所以不需要进行 下述相邻牙齿的孤立。 在预处理之后的第 1张投影图像上， 上下牙颌之间的缝隙 处及相邻牙齿之间的边界处灰度值较低， 因此可以用上下牙颌之间及相邻牙齿 之间灰度较低像素构成的直线或曲线将上下牙颌及相邻牙齿分幵， 实现独立牙
齿的孤立。 此吋， 二维投影图像将被这些分离线分割成小块， 每一小块区域中 只有一颗牙齿， 即在二维投影图像上实现了独立牙齿的孤立。
[0071] 在独立牙齿区域中， 牙根周围常常会粘连一些牙槽骨组织， 因此可以采用阈值 分割、 形态学运算、 形状分析等算法去除牙根周围的部分牙槽骨组织， 并以此 吋独立牙齿区域中面积最大的联通区域的边界作为初始牙齿轮廓。 然后， 可采 用 Moore边界追踪算法、 gmph-cut算法、 活动轮廓、 水平集等算法优化初始牙齿 轮廓， 即得二维牙齿轮廓。
[0072] 本申请的提取模块， 还用于在二维投影图像上提取二维轴线， 提取的二维轴线 包括沿所述牙齿轮廓的长轴方向， 并通过牙齿轮廓的质心点的直线。
[0073] 在一种实施方式中， 投影模块还用于获取第 1张和第 n(n≥2)张投影图像。
[0074] 第 1张投影图像的获取方式为： 在三维口腔 CT图像中提取牙齿牙槽骨区域； 根 据牙齿牙槽骨区域提取牙弓曲线； 对牙齿牙槽骨区域以所述牙弓曲线的法线方 向为投影方向进行投影， 得第 1张投影图像。
[0075] 第 1张投影图像的合成流程如图 3所示。 本申请将二维投影图像 1的投影区域限 定在口腔 CT图像中牙齿牙槽骨区域。 在口腔 CT图像中提取牙齿牙槽骨区域的过 程如下： 在口腔 CT图像中， 由于骨骼组织 （牙齿、 牙槽骨、 其它颌骨组织） 与 其他组织之间灰度差异较大， 可采用阈值分割等方法从口腔 CT图像中提取骨骼 组织。 对提取出的骨骼组织， 采用幵运算算法等方法将骨骼组织中的牙齿牙槽 骨组织与颈骨等其他骨骼组织断幵。 此吋， 骨骼组织被分为几部分， 其中体积 最大的三维连通区域即为牙齿牙槽骨区域。 然后， 根据提取得到的牙齿牙槽骨 区域获取牙弓曲线， 以牙弓曲线的法线为投影方向。 牙弓曲线可以在提取得到 的牙齿牙槽骨区域的最大灰度投影图上通过细化、 手动描点法结合样条曲线或 其他多项式曲线拟合等方法得到。 对提取得到的牙齿牙槽骨区域沿着牙弓曲线 的法线进行投影即得第 1张投影图像。
[0076] 第 n(n≥2)张投影图像的合成流程如图 4所示。 对三维口腔 CT图像进行投影获得 第 n(n≥2)张投影图像的获取具体可以包括： 以所述第 1张投影图像的二维牙齿轴 线在其牙齿轮廓的左右平行轴线所确定的牙齿牙槽骨区域为投影区域。 以过所 述第 1张投影图像的二维牙齿轴线的法线为投影方向对所述投影区域进行投影，
即得第 n(n≥2)张投影图像。
以上内容是结合具体的实施方式对本申请所作的进一步详细说明， 不能认定本 申请的具体实施只局限于这些说明。 对于本申请所属技术领域的普通技术人员 来说， 在不脱离本申请构思的前提下， 还可以做出若干简单推演或替换。
技术问题
问题的解决方案
发明的有益效果
Claims
[权利要求 1] 一种估算三维牙轴的方法， 其特征在于， 包括：
将三维口腔 CT图像转换为多张二维投影图像， 根据所述二维投影图 像提取三维牙轴。
[权利要求 2] 如权利要求 1所述的估算三维牙轴的方法， 其特征在于， 所述将三维 口腔 CT图像转换为多张二维投影图像， 根据所述二维投影图像提取 三维牙轴， 具体包括：
对三维口腔 CT图像进行多次投影获得多张二维投影图像； 在所述二维投影图像中分别分割出牙齿轮廓；
根据所述牙齿轮廓提取二维轴线；
根据多个所述二维轴线合成三维牙轴。
[权利要求 3] 如权利要求 2所述的估算三维牙轴的方法， 其特征在于， 所述根据所 述二维投影图像分割出牙齿轮廓， 具体包括：
对所述二维投影图像进行预处理；
对预处理后的二维投影图像， 采用自动或手动交互方式分割提取出牙 齿二维轮廓。
[权利要求 4] 如权利要求 2所述的估算三维牙轴的方法， 其特征在于， 所述根据所 述牙齿轮廓提取二维轴线， 具体包括：
在二维投影图像上提取二维轴线， 提取的二维轴线包括沿所述牙齿轮 廓的长轴方向， 并通过所述牙齿轮廓的质心点的直线。
[权利要求 5] 如权利要求 2所述的估算三维牙轴的方法， 其特征在于， 对三维口腔
CT图像进行投影获得二维投影图像， 具体包括： 将牙齿牙槽骨区域从所述三维口腔 CT图像中提取出来；
从所述牙齿牙槽骨区域获取牙弓曲线；
将所述牙齿牙槽骨区域以所述牙弓曲线的法线方向为投影方向进行投 影， 得到第一张二维投影图像；
以所述第一张二维投影图像的二维牙齿轴线在其牙齿轮廓的左右平行 轴线所确定的牙齿牙槽骨区域为投影区域， 以过所述第一张二维投影
图像的二维牙齿轴线的平面的法线为投影方向， 对所述投影区域进行 投影， 得到多张二维投影图像。
一种估算三维牙轴的装置， 其特征在于，
所述装置， 用于将三维口腔 CT图像转换为多张二维投影图像， 并根 据所述二维投影图像提取三维牙轴。
如权利要求 6所述的估算三维牙轴的装置， 其特征在于， 包括： 投影模块， 用于对三维口腔 CT图像进行多次投影获得多张二维投影 图像；
分割模块， 用于在所述二维投影图像中分别分割出牙齿轮廓； 提取模块， 用于根据所述牙齿轮廓提取二维轴线；
合成模块， 用于根据多个所述二维轴线合成三维牙轴。
如权利要求 6所述的估算三维牙轴的装置， 其特征在于，
所述分割模块， 还用于对所述二维投影图像进行预处理， 对预处理后 的二维投影图像， 采用自动或手动交互方式分割提取出牙齿二维轮廓 如权利要求 6所述的估算三维牙轴的装置， 其特征在于，
所述提取模块， 还用于在二维投影图像上提取二维轴线， 提取的二维 轴线包括沿所述牙齿轮廓的长轴方向， 并通过所述牙齿轮廓的质心点 的直线。
如权利要求 6所述的估算三维牙轴的装置， 其特征在于，
所述投影模块， 还用于将牙齿牙槽骨区域从所述三维口腔 CT图像中 提取出来， 从所述牙齿牙槽骨区域获取牙弓曲线， 将所述牙齿牙槽骨 区域以所述牙弓曲线的法线方向为投影方向进行投影， 得到第一张二 维投影图像； 以所述第一张二维投影图像的二维牙齿轴线在其牙齿轮 廓的左右平行轴线所确定的牙齿牙槽骨区域为投影区域， 以过所述第 一张二维投影图像的二维牙齿轴线的平面的法线为投影方向， 对所述 投影区域进行投影， 得到多张二维投影图像。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/096851 WO2018035837A1 (zh) | 2016-08-26 | 2016-08-26 | 一种估算三维牙轴的方法和装置 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2016/096851 WO2018035837A1 (zh) | 2016-08-26 | 2016-08-26 | 一种估算三维牙轴的方法和装置 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018035837A1 true WO2018035837A1 (zh) | 2018-03-01 |
Family
ID=61245347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/096851 Ceased WO2018035837A1 (zh) | 2016-08-26 | 2016-08-26 | 一种估算三维牙轴的方法和装置 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2018035837A1 (zh) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108470365A (zh) * | 2018-03-07 | 2018-08-31 | 天津工业大学 | 一种基于上下牙颌的牙弓线绘制方法 |
| CN114757956A (zh) * | 2022-03-01 | 2022-07-15 | 四川锋准机器人科技有限公司 | 一种种植体自动提取以及种植体中轴线自动确定方法 |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101099732B1 (ko) * | 2010-07-30 | 2011-12-28 | 경희대학교 산학협력단 | 컴퓨터 단층촬영영상 및 자기공명영상에서 3차원 개별 치아 모델링을 위한 치아 분리 시스템 |
-
2016
- 2016-08-26 WO PCT/CN2016/096851 patent/WO2018035837A1/zh not_active Ceased
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101099732B1 (ko) * | 2010-07-30 | 2011-12-28 | 경희대학교 산학협력단 | 컴퓨터 단층촬영영상 및 자기공명영상에서 3차원 개별 치아 모델링을 위한 치아 분리 시스템 |
Non-Patent Citations (1)
| Title |
|---|
| LIN BIN: "Computation and Application of tooth -Root orientation from 3D CT Images", CHINA MASTER'S THESIS, vol. 2013, no. 07, 15 July 2013 (2013-07-15), pages 5-6, 11-20, 27 - 34, 58-59 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108470365A (zh) * | 2018-03-07 | 2018-08-31 | 天津工业大学 | 一种基于上下牙颌的牙弓线绘制方法 |
| CN114757956A (zh) * | 2022-03-01 | 2022-07-15 | 四川锋准机器人科技有限公司 | 一种种植体自动提取以及种植体中轴线自动确定方法 |
| CN114757956B (zh) * | 2022-03-01 | 2024-05-10 | 四川锋准机器人科技有限公司 | 一种种植体自动提取以及种植体中轴线自动确定方法 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106214175B (zh) | 一种估算三维牙轴的方法和装置 | |
| US11232573B2 (en) | Artificially intelligent systems to manage virtual dental models using dental images | |
| CN109310488B (zh) | 用于估计牙齿修复体的形状、位置以及取向中的至少一个的方法 | |
| CN109363786B (zh) | 一种牙齿正畸矫治数据获取方法及装置 | |
| JP2022516488A5 (zh) | ||
| US20200350059A1 (en) | Method and system of teeth alignment based on simulating of crown and root movement | |
| CN108932716B (zh) | 用于牙齿图像的图像分割方法 | |
| WO2016197370A1 (zh) | 牙齿和牙槽骨的分割与重构方法及装置 | |
| CN105488849B (zh) | 一种基于混合水平集的三维牙齿建模方法 | |
| Fornai et al. | Variability of Australopithecus second maxillary molars from Sterkfontein Member 4 | |
| US12307693B2 (en) | Automated method for aligning three dimensional dental library model to three dimensional oral scan data and computer readable medium having program for performing the method | |
| CN109377534B (zh) | 自动采样厚度检测的非线性口腔ct全景图像合成方法 | |
| US12469237B2 (en) | System and method for alignment of volumetric and surface scan images | |
| EP4555481B1 (en) | METHOD AND SYSTEM FOR ESTIMATING TOOTH PLACEMENT | |
| WO2018035837A1 (zh) | 一种估算三维牙轴的方法和装置 | |
| WO2024114343A1 (zh) | 牙龈线提取方法、牙科器械的制作方法及电子设备 | |
| CN116309302A (zh) | 一种头颅侧位片关键点提取方法 | |
| CN115517792A (zh) | 基于人工智能的牙列匹配及颌骨分割方法、系统及介质 | |
| CN117670956B (zh) | 牙齿边界的自动提取方法、图像矫正方法及牙齿治疗系统 | |
| Wu et al. | ChatIOS: Improving automatic 3-dimensional tooth segmentation via GPT-4V and multimodal pre-training | |
| CN113313722A (zh) | 一种牙根图像交互标注方法 | |
| CN110960333A (zh) | 基于ai算法的错颌畸形矫治模型匹配方法 | |
| US12579639B2 (en) | System and method for fusion of volumetric and surface scan images | |
| CN121942047A (zh) | 正畸与牙科修复联合处理 | |
| CN116350370B (zh) | 义齿设计方法、装置、设备及计算机可读存储介质 |
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: 16913859 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 1205A DATED 03/07/2018) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16913859 Country of ref document: EP Kind code of ref document: A1 |