CN113393470A - Full-automatic tooth segmentation method - Google Patents

Full-automatic tooth segmentation method Download PDF

Info

Publication number
CN113393470A
CN113393470A CN202110517354.3A CN202110517354A CN113393470A CN 113393470 A CN113393470 A CN 113393470A CN 202110517354 A CN202110517354 A CN 202110517354A CN 113393470 A CN113393470 A CN 113393470A
Authority
CN
China
Prior art keywords
segmentation
model
level set
result
tooth
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.)
Pending
Application number
CN202110517354.3A
Other languages
Chinese (zh)
Inventor
李纯明
李茜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
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 University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202110517354.3A priority Critical patent/CN113393470A/en
Publication of CN113393470A publication Critical patent/CN113393470A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30036—Dental; Teeth

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)
  • Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)

Abstract

The invention discloses a full-automatic tooth segmentation method, which comprises the following steps: s1: preprocessing the collected tooth image to obtain a preprocessing result; s2: performing rough segmentation on the preprocessing result by adopting a deep learning model to obtain a rough segmentation result; s3: carrying out mutual exclusion level set model segmentation on the rough segmentation result to obtain a mutual exclusion level set model segmentation result; s4: and performing three-dimensional reconstruction on the segmentation result of the mutually exclusive level set model. The tooth full-automatic segmentation method provided by the invention can solve the problems that the existing tooth segmentation is difficult and full automation in a complete sense cannot be realized.

Description

Full-automatic tooth segmentation method
Technical Field
The invention relates to the technical field of dental medical treatment, in particular to a full-automatic tooth segmentation method.
Background
In the modern society, the first time "health preservation" becomes a popular word in the society, and the people frequently board the hot search charts of various search engines, which is enough to explain that the attention of the contemporary people to the health is increasing. The least significant of these is the oral problems. The medical imaging technology becomes an indispensable technical means for comprehensively, accurately and accurately acquiring the data of the patient and further providing powerful guarantee for the diagnosis and treatment of oral diseases. The oral cavity CT equipment is a revolutionary tool of dentistry, and the CBCT is applied to the field of oral medicine from the late stage of the nineties of the last century, so that the three-dimensional structure of a jaw face can be truly reflected, the defect of two-dimensional imaging is overcome, the relationship between teeth and bones can be evaluated more three-dimensionally, and a more reasonable scheme can be formulated by an orthodontist. In order to reduce the burden of doctors, alleviate the market relation of insufficient supply and demand between doctors and patients, and to alleviate points to some extent, the research on tooth segmentation is becoming the focus of research of various scholars, certainly, is also a research focus. The emerging CBCT imaging technology dedicated to tooth imaging is receiving much attention, and research hot spots are beginning to shift to the research of CBCT tooth images. The CBCT also provides important data support for assisting oral and maxillofacial diagnosis and treatment.
Because of the imaging technical defects of medical images and the particularity of tooth structures, the tooth segmentation treatment can be realized only by means of a certain amount of interactive operation, so that the tooth segmentation difficulty is huge, and the existing level set algorithm needs a large amount of manual participation and cannot realize full automation in a complete sense.
Disclosure of Invention
The invention aims to provide a full-automatic tooth segmentation method, which aims to solve the problems that the existing tooth segmentation is difficult and full automation in a complete sense cannot be realized.
The technical scheme for solving the technical problems is as follows:
the invention provides a full-automatic tooth segmentation method, which comprises the following steps:
s1: preprocessing the collected tooth image to obtain a preprocessing result;
s2: carrying out rough segmentation on the preprocessing model by adopting a deep learning model to obtain a rough segmentation result;
s3: carrying out mutual exclusion level set model segmentation on the rough segmentation result to obtain a mutual exclusion level set model segmentation result;
s4: and performing three-dimensional reconstruction on the segmentation result of the mutually exclusive level set model.
Alternatively, the step S1 includes the following substeps:
s11: collecting tooth image data which is stored in a layered mode in a DICOM format;
s12: and carrying out gray scale normalization processing on the tooth image data by adopting a gray scale linear change model.
Optionally, in step S12, the gray scale linear variation model includes:
Figure BDA0003062802480000021
wherein D' (x, y) refers to the gray value of the output image at the pixel point (x, y) after gray conversion, and D (x, y) represents the gray value of the input image at the pixel point (x, y).
Optionally, in step S2, the rough segmentation includes:
and selecting an initial layer to be segmented, and applying a deep learning model to perform initial segmentation.
Optionally, the deep learning model is a U-Net model.
Alternatively, the step S3 includes the following substeps:
s31: taking the rough segmentation result as an initialization result of the mutual exclusion level set model;
s32: and according to the initialization result, performing interlayer sequence two-dimensional iterative segmentation on the tooth image data to obtain a segmentation result of the mutually exclusive level set model.
Optionally, the performing interlayer sequence two-dimensional iterative segmentation on the dental image data comprises:
and performing layer-by-layer iterative segmentation in the directions of the dental crown and the dental root by utilizing the interlayer information from the initial layer of the tooth image data.
Optionally, the interlayer information includes:
the dental image data is divided into a target region including teeth and a background region including gums using a clustering algorithm.
Optionally, in step S3, the mutually exclusive level set model includes:
Φ1＝max(Φ1,-Φ2)
Φ2＝max(Φ2,-Φ1)
wherein phi1，Φ2Respectively, a first class level set and a second class level set, and max () represents the maximum value taken for them.
Optionally, the mutually exclusive level set model is a DRLSE model, and the DRLSE model includes:
Figure BDA0003062802480000031
wherein the content of the first and second substances,
Figure BDA0003062802480000032
representing the function of the DRLSE model,
Figure BDA0003062802480000033
indicating a level set function, λ and α are constant values,
Figure BDA0003062802480000037
and
Figure BDA0003062802480000034
representing horizontal and edge terms, respectively, deltaε()，Hε() Representing the unit impact function and the step function which are commonly used, g is a newly introduced boundary detection factor in the model and is expressed as
Figure BDA0003062802480000035
I is the image to be processed and,
Figure BDA0003062802480000036
is a horizontally centered gradient mode wherein
Figure BDA0003062802480000038
Is a gradient operator, is a convolution sign, GσExpressed as σAs a Gaussian function of standard deviation, Ω represents
The invention has the following beneficial effects:
the tooth full-automatic segmentation method provided by the invention can be used for well initializing the level set by combining the deep learning model U-Net, and the introduced mutually exclusive level set model can avoid the error of identifying adjacent teeth as a tooth for segmentation. The higher the automation degree of the segmentation model is, the more time-consuming manual operation can be reduced, and the working efficiency and diagnosis and treatment accuracy of doctors are improved. The model can realize full-automatic full-oral-cavity tooth segmentation, greatly reduces the workload of manual operation compared with the prior semi-automatic segmentation model, and is more favorable for landing application.
Drawings
FIG. 1 is a flowchart illustrating a method for fully automatically segmenting teeth according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating the substeps of step S1 in FIG. 1;
FIG. 3 is a flowchart illustrating the substeps of step S3 in FIG. 1;
fig. 4 is a schematic diagram illustrating a rough segmentation result in the fully automatic tooth segmentation method according to the embodiment of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Examples
The technical scheme for solving the technical problems is as follows:
the invention provides a full-automatic tooth segmentation method, which is shown in a figure 1 and comprises the following steps:
s1: preprocessing the collected tooth image to obtain a preprocessing result;
s2: performing rough segmentation on the preprocessing result by using a deep learning model to obtain a rough segmentation result (shown in reference to fig. 4);
s3: carrying out mutual exclusion level set model segmentation on the rough segmentation result to obtain a mutual exclusion level set model segmentation result;
s4: and performing three-dimensional reconstruction on the segmentation result of the mutually exclusive level set model.
The invention has the following beneficial effects:
the tooth full-automatic segmentation method provided by the invention can be used for well initializing the level set by combining the deep learning model U-Net, and the introduced mutually exclusive level set model can avoid the error of identifying adjacent teeth as a tooth for segmentation. The higher the automation degree of the segmentation model is, the more time-consuming manual operation can be reduced, and the working efficiency and diagnosis and treatment accuracy of doctors are improved. The model can realize full-automatic full-oral-cavity tooth segmentation, greatly reduces the workload of manual operation compared with the prior semi-automatic segmentation model, and is more favorable for landing application.
Alternatively, referring to fig. 2, the step S1 includes the following sub-steps:
s11: collecting tooth image data which is stored in a layered mode in a DICOM format;
DICOM (digital Imaging and Communications in medicine), which is an international standard for medical images and related information (ISO 12052), is one of the most widely deployed medical information standards among tens of thousands of medical Imaging devices in use.
S12: and carrying out gray scale normalization processing on the tooth image data by adopting a gray scale linear change model.
Since the gray scale distribution range of the original tooth image is about 5000-6000, when the gray scale normalization is performed on the original tooth image, the gray scale range of pixels needs to be reduced to [0,255], and in step S12, the gray scale linear variation model includes:
Figure BDA0003062802480000051
wherein D (x, y) represents the gray level of the input image at the pixel (x, y), D' (x, y) refers to the gray level of the output image at the pixel (x, y) after gray level transformation, and the maximum gray level is generally 255. The general normalization transformation only reduces the gray scale range, and does not change the contrast of the gray scale distribution of the pixel.
Optionally, in step S2, the rough segmentation includes:
and selecting an initial layer to be segmented, and applying a deep learning model to perform initial segmentation.
U-Net is a network model which has great success in the field of medical image segmentation, is a structure similar to an encoder-decoder, and is an improvement of FCN (fuzzy C-means) due to the fact that the model presents symmetry (U-shaped); because the down-sampled feature graph and the up-sampled feature graph with the same dimension are spliced and connected through skip connection, the features of high and low levels are better fused; the method is applicable to application scenes with small sample size and high operation speed, so that the deep learning model is adopted as the U-Net model in the invention. Of course, one skilled in the art can also use conventional algorithms, such as models like MICO, to give the level set algorithm a good initialization, and also achieve fully automatic segmentation of teeth.
Alternatively, referring to fig. 3, the step S3 includes the following sub-steps:
s31: taking the rough segmentation result as an initialization result of the mutual exclusion level set model;
here, since the teeth include the upper teeth and the lower teeth, when the rough division result is used as the initialization result of the exclusive level set model, the initialization process should be performed on the initial layers of the upper teeth and the lower teeth, respectively, and thus two initial layer processed results are obtained.
S32: and according to the initialization result, performing interlayer sequence two-dimensional iterative segmentation on the tooth image data to obtain a segmentation result of the mutually exclusive level set model.
Optionally, the performing interlayer sequence two-dimensional iterative segmentation on the dental image data comprises:
and performing layer-by-layer iterative segmentation in the directions of the dental crown and the dental root by utilizing the interlayer information from the initial layer of the tooth image data.
Optionally, the interlayer information includes:
the dental image data is divided into a target region including teeth and a background region including gums using a clustering algorithm. The tooth image data is divided into a target area and a background area by adopting a K-Means clustering algorithm.
Optionally, in step S3, the mutually exclusive level set model includes:
Φ1＝max(Φ1,-Φ2)
Φ2＝max(Φ2,-Φ1)
wherein phi1，Φ2Respectively, a first class level set and a second class level set, and max () represents the maximum value taken for them.
Specifically, the following are shown:
Figure BDA0003062802480000061
Figure BDA0003062802480000062
wherein the content of the first and second substances,
Figure BDA0003062802480000063
and
Figure BDA0003062802480000064
is a description of the time evolution, div () represents the divergence,
Figure BDA0003062802480000065
a first type of level set curve is represented,
Figure BDA0003062802480000066
a second type of level set curve is represented,
Figure BDA0003062802480000067
representing the corresponding gradients of the two level sets, dp(s) represents a function defined by a parameter s and
Figure BDA0003062802480000068
p'(s) is the derivation of the function p,
Figure BDA0003062802480000069
s represents [0,1 ]]P is a distance regularization term correlation function, p2(. cndot.) is a function of the level set gradient mode,
Figure BDA00030628024800000610
is a horizontally centered gradient mode wherein
Figure BDA0003062802480000071
Is a gradient operator, GσIs a gaussian function with standard deviation sigma, I is the original image, is the convolution sign, g is the edge detection factor,
Figure BDA0003062802480000072
optionally, the mutually exclusive level set model is a DRLSE model, and the DRLSE model includes:
Figure BDA0003062802480000073
wherein the content of the first and second substances,
Figure BDA0003062802480000074
representing the function of the DRLSE model,
Figure BDA0003062802480000075
indicating a level set function, λ and α are constant values,
Figure BDA0003062802480000076
and
Figure BDA0003062802480000077
representing horizontal and edge terms, respectively, deltaε()，Hε() Indicating the unit impact in generalFunction and step function, g is a newly introduced boundary detection factor in the model, and is expressed as
Figure BDA0003062802480000078
I is the image to be processed and,
Figure BDA0003062802480000079
is a horizontally centered gradient mode wherein
Figure BDA00030628024800000710
Is a gradient operator, is a convolution sign, GσDenotes a gaussian function with σ as the standard deviation, and Ω denotes an image region.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (10)

1. A full-automatic tooth segmentation method is characterized by comprising the following steps:
s1: preprocessing the collected tooth image to obtain a preprocessing result;
s2: performing rough segmentation on the preprocessing result by adopting a deep learning model to obtain a rough segmentation result;
s3: carrying out mutual exclusion level set model segmentation on the rough segmentation result to obtain a mutual exclusion level set model segmentation result;
s4: and performing three-dimensional reconstruction on the segmentation result of the mutually exclusive level set model.
2. The method for fully automatically segmenting teeth according to claim 1, wherein the step S1 includes the following sub-steps:
s11: collecting tooth image data which is stored in a layered mode in a DICOM format;
s12: and carrying out gray scale normalization processing on the tooth image data by adopting a gray scale linear change model.
3. The method for fully automatically segmenting teeth according to claim 2, wherein in the step S12, the gray scale linear variation model comprises:
Figure FDA0003062802470000011
wherein D' (x, y) refers to the gray value of the output image at the pixel point (x, y) after gray conversion, and D (x, y) represents the gray value of the input image at the pixel point (x, y).
4. The method for fully automatically segmenting teeth according to claim 2, wherein in the step S2, the rough segmentation comprises:
and selecting an initial layer to be segmented, and applying a deep learning model to perform initial segmentation.
5. A tooth full-automatic segmentation method according to any one of claims 1 to 4, characterized in that the deep learning model is a U-Net model.
6. The method for fully automatically segmenting teeth according to claim 2, wherein the step S3 includes the following sub-steps:
s31: taking the rough segmentation result as an initialization result of the mutual exclusion level set model;
s32: and according to the initialization result, performing interlayer sequence two-dimensional iterative segmentation on the tooth image data to obtain a segmentation result of the mutually exclusive level set model.
7. The method of claim 6, wherein the performing the interlayer sequence two-dimensional iterative segmentation on the dental image data comprises:
and performing layer-by-layer iterative segmentation in the directions of the dental crown and the dental root by utilizing the interlayer information from the initial layer of the tooth image data.
8. The method of claim 7, wherein the interlayer information comprises:
the dental image data is divided into a target region including teeth and a background region including gums using a clustering algorithm.
9. The method according to claim 1, wherein in step S3, the mutually exclusive level set model comprises:
Φ1＝max(Φ1,-Φ2)
Φ2＝max(Φ2,-Φ1)
wherein phi1，Φ2Respectively, a first class level set and a second class level set, and max () represents the maximum value taken for them.
10. The method of claim 1, wherein the mutually exclusive level set model is a DRLSE model, and the DRLSE model comprises:
Figure FDA0003062802470000021
wherein the content of the first and second substances,
Figure FDA0003062802470000022
representing the function of the DRLSE model,
Figure FDA0003062802470000023
indicating a level set function, λ and α are constant values,
Figure FDA0003062802470000024
and
Figure FDA0003062802470000025
respectively represent waterArea term and edge term of the sum, δε()，Hε() Representing the unit impact function and the step function which are commonly used, g is a newly introduced boundary detection factor in the model and is expressed as
Figure FDA0003062802470000026
I is the image to be processed and,
Figure FDA0003062802470000027
is a horizontally centered gradient mode wherein
Figure FDA0003062802470000028
Is a gradient operator, is a convolution sign, GσDenotes a gaussian function with σ as the standard deviation, and Ω denotes an image region.
CN202110517354.3A 2021-05-12 2021-05-12 Full-automatic tooth segmentation method Pending CN113393470A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110517354.3A CN113393470A (en) 2021-05-12 2021-05-12 Full-automatic tooth segmentation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110517354.3A CN113393470A (en) 2021-05-12 2021-05-12 Full-automatic tooth segmentation method

Publications (1)

Publication Number Publication Date
CN113393470A true CN113393470A (en) 2021-09-14

Family

ID=77616982

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110517354.3A Pending CN113393470A (en) 2021-05-12 2021-05-12 Full-automatic tooth segmentation method

Country Status (1)

Country Link
CN (1) CN113393470A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113920253A (en) * 2021-11-22 2022-01-11 广州大学 Tooth model fast cutting method based on three-dimensional oral cavity model
CN114241173A (en) * 2021-12-09 2022-03-25 电子科技大学 Tooth CBCT image three-dimensional segmentation method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109816661A (en) * 2019-03-22 2019-05-28 电子科技大学 A kind of tooth CT image partition method based on deep learning
CN109886965A (en) * 2019-04-09 2019-06-14 山东师范大学 The layer of retina dividing method and system that a kind of level set and deep learning combine
US20190392267A1 (en) * 2018-06-20 2019-12-26 International Business Machines Corporation Framework for integrating deformable modeling with 3d deep neural network segmentation
CN110889850A (en) * 2019-12-13 2020-03-17 电子科技大学 CBCT tooth image segmentation method based on central point detection
CN112102495A (en) * 2020-09-15 2020-12-18 北京朗视仪器有限公司 Dental arch curved surface generation method based on CBCT image

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190392267A1 (en) * 2018-06-20 2019-12-26 International Business Machines Corporation Framework for integrating deformable modeling with 3d deep neural network segmentation
CN109816661A (en) * 2019-03-22 2019-05-28 电子科技大学 A kind of tooth CT image partition method based on deep learning
CN109886965A (en) * 2019-04-09 2019-06-14 山东师范大学 The layer of retina dividing method and system that a kind of level set and deep learning combine
CN110889850A (en) * 2019-12-13 2020-03-17 电子科技大学 CBCT tooth image segmentation method based on central point detection
CN112102495A (en) * 2020-09-15 2020-12-18 北京朗视仪器有限公司 Dental arch curved surface generation method based on CBCT image

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
刘祎: "基于CBCT图像的牙齿水平集分割", 《中国优秀博硕士学位论文全文数据库(硕士)医药卫生科技辑》 *
孟闫凯: "对比MRI平扫、增强图像的影像组学标签对直肠癌生存期的预测价值", 《中华放射学杂志》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113920253A (en) * 2021-11-22 2022-01-11 广州大学 Tooth model fast cutting method based on three-dimensional oral cavity model
CN113920253B (en) * 2021-11-22 2022-05-17 广州大学 Tooth model fast cutting method based on three-dimensional oral cavity model
CN114241173A (en) * 2021-12-09 2022-03-25 电子科技大学 Tooth CBCT image three-dimensional segmentation method and system
CN114241173B (en) * 2021-12-09 2023-03-21 电子科技大学 Tooth CBCT image three-dimensional segmentation method and system

Similar Documents

Publication Publication Date Title
US8970581B2 (en) System and method for interactive contouring for 3D medical images
CN113223010B (en) Method and system for multi-tissue full-automatic segmentation of oral cavity image
Tian et al. Efficient computer-aided design of dental inlay restoration: a deep adversarial framework
CN112785609B (en) CBCT tooth segmentation method based on deep learning
CN111415419A (en) Method and system for making tooth restoration model based on multi-source image
CN110889850B (en) CBCT tooth image segmentation method based on central point detection
CN113393470A (en) Full-automatic tooth segmentation method
CN114757960B (en) Tooth segmentation and reconstruction method based on CBCT image and storage medium
CN114332123A (en) Automatic caries grading method and system based on panoramic film
CN113344950A (en) CBCT image tooth segmentation method combining deep learning with point cloud semantics
CN111325754B (en) Automatic lumbar vertebra positioning method based on CT sequence image
Cristian et al. A cone beam computed tomography annotation tool for automatic detection of the inferior alveolar nerve canal
CN114549540A (en) Method for automatically fusing oral scanning tooth data and CBCT (Cone Beam computed tomography) data and application thereof
CN114565601A (en) Improved liver CT image segmentation algorithm based on DeepLabV3+
Du et al. Mandibular canal segmentation from CBCT image using 3D convolutional neural network with scSE attention
Ben-Hamadou et al. 3DTeethSeg'22: 3D Teeth Scan Segmentation and Labeling Challenge
Tian et al. Efficient tooth gingival margin line reconstruction via adversarial learning
Chen et al. Detection of various dental conditions on dental panoramic radiography using Faster R-CNN
Qian et al. An automatic tooth reconstruction method based on multimodal data
CN116958169A (en) Tooth segmentation method for three-dimensional dental model
CN111437057A (en) Three-dimensional tooth shape restoration method and system based on two-dimensional tooth beautifying characteristic line
CN116823729A (en) Alveolar bone absorption judging method based on SegFormer and oral cavity curved surface broken sheet
Yu et al. A benchmark dual-modality dental imaging dataset and a novel cognitively inspired pipeline for high-resolution dental point cloud synthesis
CN114677516B (en) Automatic oral mandibular tube segmentation method based on deep neural network
CN115409811A (en) Tooth model reconstruction method, device, equipment and medium based on curvature enhancement

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210914