CN114549559A - Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) - Google Patents
Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) Download PDFInfo
- Publication number
- CN114549559A CN114549559A CN202210196096.8A CN202210196096A CN114549559A CN 114549559 A CN114549559 A CN 114549559A CN 202210196096 A CN202210196096 A CN 202210196096A CN 114549559 A CN114549559 A CN 114549559A
- Authority
- CN
- China
- Prior art keywords
- tooth
- result
- segmentation
- post
- connected domain
- 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
Links
- 238000007408 cone-beam computed tomography Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012805 post-processing Methods 0.000 title claims abstract description 26
- 238000013473 artificial intelligence Methods 0.000 title description 30
- 230000011218 segmentation Effects 0.000 claims abstract description 40
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 230000010339 dilation Effects 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 3
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 208000014151 Stomatognathic disease Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
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
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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)
- Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)
Abstract
The invention relates to a post-processing method for segmenting tooth results based on CBCT (Cone Beam computed tomography) data AI, which comprises the following steps: s1 loading CBCT data; s2 performing AI segmentation on the CBCT data loaded in the step S1; s3 performing connected domain analysis on the result of the AI segmentation in step S2; s4 finding a connected domain of other tags intersecting each tooth based on step S3; s5 in step S4, { R }j}iWhether the connected domain in (b) can merge with tooth i; s6 generates a result. The post-processing method for the AI segmentation tooth result based on the CBCT data can optimize the AI segmentation tooth result of the CBCT data, does not need a user to manually check and correct the problems of label confusion and cross in the AI segmentation result, can automatically post-process the segmentation result in a short time, combines the labels of the confusion and cross, eliminates the condition of most label confusion, greatly improves the efficiency, improves the AI segmentation result, and provides more informationFor the orthodontist, improve the precision of stealthy orthodontic treatment.
Description
Technical Field
The invention relates to the field of oral medical image processing, in particular to a post-processing method and a post-processing system for segmenting tooth results based on CBCT (Cone Beam computed tomography) data AI.
Background
Cone Beam CT (Cone Beam Computer tomogry, hereinafter CBCT) is currently widely used in the diagnosis and treatment of dental diseases. Along with the improvement of the living standard of people and the updating of the oral concept, more and more people select orthodontics, teeth need to be segmented according to influence data acquired by CBCT in orthodontic treatment, the existing automatic segmentation of teeth through AI can cause the conditions of partial confusion, crossing and the like of labels of all teeth, the later manual modification of the labels is needed, the efficiency is low, and the accuracy of the invisible orthodontic treatment can be influenced.
Disclosure of Invention
The invention aims to solve the technical problem of designing a processing method capable of optimizing AI segmentation tooth results of CBCT data and improving segmentation effect, and solves the existing technical problem.
In order to solve the technical problem, the post-processing method for segmenting the tooth result based on the CBCT data AI comprises the following steps:
s1: loading CBCT data;
s2: AI segmentation is performed on the CBCT data loaded in step S1;
s3: performing connected domain analysis on the result of the AI segmentation in step S2;
s4: on the basis of the step S3, searching the connected domain of other labels intersected with each tooth;
s5: in step S4, { R } is judgedj}iWhether the connected component in (a) can be merged with tooth i;
s6: and generating a result.
Further, the CBCT data format in step S1 is several hundreds of files in the format of a. dcm suffix.
Further, in step S2, the dicom data of CBCT is subjected to dental segmentation by using a conventional AI segmentation network framework, such as V-Net network, U-Net3D network, or nn-UNet network framework.
Further, the method for analyzing the connected component in step S3 includes the following steps:
s31: computing bounding box positions p for all connected regions of each toothlabelAnd size vlabel;
S32: calculating the volume of the connected domain in each bounding box and the belonged label, and marking the connected domain from large to small, for example, marking the maximum connected domain of the tooth 1 as '0101';
s33: storing the information obtained in the steps S31 and S32 as a dictionary structure D as follows:
D={{″0101″:{p1,v1}},...};
subsequent indexing is facilitated, after which traversal operations with respect to the connected component are restricted to being performed in the corresponding bounding box, rather than the entire image.
Further, the method for each tooth to find the connected domain of the other label intersected with the tooth in step S4 includes the following steps:
s41: go through all teeth, take tooth i as an example, choose the largest connected domain CiAs a reference;
s42: setting the shape and size of the convolution kernel (e.g., spherical convolution kernel with radius 3), for CiMorphological dilation is performed and the result of the dilation is denoted Di;
S43: calculating DiThe bounding box of (1), traversed in the bounding box region, finds all other connected domains that intersect the inflation result, denoted as { R }j}iAnd j is the tooth to which the other connected domain belongs.
Further, { R ] in step S5j}iThe method for judging whether the connected component can be combined with the tooth i comprises the following steps:
s51: set the combined volume upper bound U of tooth iiAnd a connected domain RjLower bound of volume Lj,UiAnd LjObtained by calculating statistical information of tooth data, and recording the mean value as meaniVariance is recorded as stdiAnd then:
Ui=meani+2*stdi;
Lj=meanj-stdj;
s52: for it { Rj}iTraversal with R for the middle connected domainjFor example, the sum of the volume and the volume of tooth i is calculated, and if the value is less than UiIndicating that the combined volume is in the normal range; then, judging RjWhether the volume is less than LjIf the value is less than the predetermined value, the connected domain is not belonging to the toothAnd then merging, which can avoid merging the whole tooth into other teeth.
S53: and after merging, modifying the position and size information of the influenced bounding boxes.
The invention also provides a post-processing system for segmenting tooth results based on CBCT data AI, comprising:
one or more processors; and
one or more memories having stored therein a computer-executable program that, when executed by the processor, performs the aforementioned post-processing method of segmenting dental results based on CBCT data AI.
The invention has the beneficial effects that:
the post-processing method of the AI segmentation tooth result based on the CBCT data can optimize the AI segmentation tooth result of the CBCT data, does not need a user to manually check and correct the problems of label confusion and cross in the AI segmentation result, can automatically perform post-processing on the segmentation result in a short time, combines the labels which are confused and crossed, eliminates the condition of most label confusion, greatly improves the efficiency, improves the AI segmentation result, provides more information for orthodontists, and improves the accuracy of invisible orthodontics treatment.
Drawings
The following further explains embodiments of the present invention with reference to the drawings.
FIG. 1 is a block flow diagram of a post-processing method for segmenting dental results based on CBCT data AI in accordance with the present invention;
FIG. 2 is a schematic diagram of the AI segmentation result of CBCT data;
FIG. 3 is a schematic diagram of the CBCT data AI segmentation result after post-processing.
Detailed Description
Referring to fig. 1, the post-processing method for segmenting tooth results based on CBCT data AI of the present embodiment includes the following steps:
s1: loading CBCT data; in this embodiment, the CBCT data format is hundreds of files with a suffix of a.dcm format.
S2: AI segmentation is performed on the CBCT data loaded in step S1; in the embodiment, the dicom data of the CBCT is subjected to dental segmentation by using a conventional AI segmentation network framework, such as a V-Net network, a U-Net3D network, an nn-UNet network framework, etc., and the segmentation result is shown in FIG. 2.
S3: performing connected domain analysis on the result of the AI segmentation in step S2; the method for analyzing the connected domain in the embodiment comprises the following steps:
s31: computing bounding box positions p for all connected regions of each toothlabelAnd size vlabel;
S32: calculating the volume of the connected domain in each bounding box and the belonged label, and marking the connected domain from large to small, for example, marking the maximum connected domain of the tooth 1 as '0101';
s33: storing the information obtained in the steps S31 and S32 as a dictionary structure D as follows:
D={{″0101″:{p1,v1}},...};
subsequent indexing is facilitated, after which traversal operations with respect to the connected component are restricted to being performed in the corresponding bounding box, rather than the entire image.
S4: on the basis of the step S3, searching the connected domain of other labels intersected with each tooth; the method for searching the connected domain of other labels intersected with each tooth in the embodiment comprises the following steps:
s41: go through all teeth, take tooth i as an example, choose the largest connected domain CiAs a reference;
s42: set the shape and size of the convolution kernel (e.g., spherical convolution kernel with a radius of 3 mm), for CiMorphological dilation is performed and the result of the dilation is denoted Di;
S43: calculating DiIs traversed in the bounding box region, finds all other connected domains that intersect the inflation result, denoted as { R }j}iAnd j is the tooth to which the other connected domain belongs.
S5: in step S4, { R } is judgedj}iWhether the connected component in (a) can be merged with tooth i; in this embodiment { Rj}iThe method for judging whether the connected domain can be combined with the tooth i comprises the following steps:
s51: set the combined volume upper bound U of tooth iiAnd a connected domain RjLower bound of volume Lj,UiAnd LjObtained by calculating statistical information of tooth data (the amount of data counted in this example is 100 cases), and the mean value thereof is referred to as meaniVariance is recorded as stdiThen:
Ui=meani+2*stdi;
Lj=meanj-stdj;
s52: for it { Rj}iTraversal with R for the middle connected domainjFor example, the sum of the volume and the volume of tooth i is calculated, and if the value is less than UiIndicating that the combined volume is in the normal range; then, judging RjWhether the volume is less than LjIf the value is less than the preset value, the connected domain is not in the normal range of the tooth, and then the connected domain is merged, so that the condition that the whole tooth is merged into other teeth can be avoided.
S53: and after merging, modifying the position and size information of the influenced bounding boxes.
S6: the results are generated as shown in fig. 3.
The present embodiment further provides a post-processing system for segmenting tooth results based on CBCT data AI, comprising:
one or more processors; and
one or more memories having stored therein a computer-executable program that, when executed by the processor, performs the aforementioned post-processing method of segmenting dental results based on CBCT data AI.
The post-processing method based on the CBCT data AI segmentation tooth result can optimize the AI segmentation tooth result of the CBCT data, does not need a user to manually check and correct the problems of label confusion and cross in the AI segmentation result, can automatically perform post-processing on the segmentation result in a short time, combines the labels with the confusion and cross, eliminates the condition of most labels in disorder, greatly improves the efficiency, improves the AI segmentation result, provides more information for an orthodontist, and improves the accuracy of invisible orthodontic treatment.
In the previous description, numerous specific details were set forth in order to provide a thorough understanding of the present invention. The foregoing description is only a preferred embodiment of the invention, which can be embodied in many different forms than described herein, and therefore the invention is not limited to the specific embodiments disclosed above. And that those skilled in the art may, using the methods and techniques disclosed above, make numerous possible variations and modifications to the disclosed embodiments, or modify equivalents thereof, without departing from the scope of the claimed embodiments. Any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present invention are within the scope of the technical solution of the present invention.
Claims (7)
1. A post-processing method for segmenting tooth results based on CBCT data AI is characterized in that: the method comprises the following steps:
s1: loading CBCT data;
s2: AI segmentation is performed on the CBCT data loaded in step S1;
s3: performing connected domain analysis on the result of the AI segmentation in step S2;
s4: on the basis of the step S3, searching the connected domain of other labels intersected with each tooth;
s5: in step S4, the { R } is judgedj}iWhether the connected component in (a) can be merged with tooth i;
s6: and generating a result.
2. Post-processing method of tooth result segmentation based on CBCT data AI according to claim 1, characterized in that: in step S1, the CBCT data format is hundreds of files in the format of a. dcm suffix.
3. Post-processing method of tooth result segmentation based on CBCT data AI according to claim 1, characterized in that: in step S2, the dicom data of CBCT is subjected to dental segmentation using a conventional AI segmentation web framework.
4. Post-processing method of tooth result segmentation based on CBCT data AI according to claim 1, characterized in that: the method for analyzing the connected component in step S3 includes the following steps:
s31: computing bounding box positions p for all connected regions of each toothlabelAnd size vlabel;
S32: calculating the volume of the connected domain in each bounding box and the belonged label, and marking the connected domain from large to small, for example, marking the maximum connected domain of the tooth 1 as '0101';
s33: storing the information obtained in the steps S31 and S32 as a dictionary structure D as follows:
D={{″0101″:{p1,v1}},...};
subsequent indexing is facilitated, after which traversal operations with respect to the connected component are restricted to occur in the corresponding bounding box.
5. Post-processing method of tooth result segmentation based on CBCT data AI according to claim 1, characterized in that: the method for each tooth to find the connected domain of other tags intersected with the tooth in step S4 includes the following steps:
s41: go through all teeth, take tooth i as an example, choose the largest connected domain CiAs a reference;
s42: set the shape and size of the convolution kernel, pair CiMorphological dilation is performed and the result of the dilation is denoted Di;
S43: calculating DiIs traversed in the bounding box region, finds all other connected domains that intersect the inflation result, denoted as { R }j}iAnd j is the tooth to which the other connected domain belongs.
6. Post-processing method of tooth result segmentation based on CBCT data AI according to claim 1, characterized in that: step S5Rj}iThe method for judging whether the connected component can be combined with the tooth i comprises the following steps:
s51: set the combined volume upper bound U of tooth iiAnd a connected domain RjLower bound of volume Lj,UiAnd LjObtained by calculating statistical information of tooth data, and recording the mean value as meaniVariance is recorded as stdiAnd then:
Ui=meani+2*stdi;
Lj=meanj-stdj;
s52: for it { Rj}iTraversal with R for the middle connected domainjFor example, the sum of the volume and the volume of tooth i is calculated, and if the value is less than UiIndicating that the combined volume is in the normal range; then, judging RjWhether the volume is less than LjIf the value is less than the preset value, the connected domain is not in the normal range of the tooth, and then merging is carried out.
S53: and after merging, modifying the position and size information of the influenced bounding boxes.
7. A post-processing system for segmenting dental results based on CBCT data AI, comprising: the method comprises the following steps:
one or more processors; and
one or more memories having stored therein a computer-executable program that, when executed by the processor, performs the post-processing method of any one of claims 1-6 for segmenting dental results based on CBCT data AI.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210196096.8A CN114549559A (en) | 2022-03-01 | 2022-03-01 | Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210196096.8A CN114549559A (en) | 2022-03-01 | 2022-03-01 | Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114549559A true CN114549559A (en) | 2022-05-27 |
Family
ID=81660833
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210196096.8A Pending CN114549559A (en) | 2022-03-01 | 2022-03-01 | Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114549559A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117095018A (en) * | 2023-10-16 | 2023-11-21 | 北京朗视仪器股份有限公司 | Multi-class tooth segmentation method and device based on CBCT image |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942780A (en) * | 2014-03-27 | 2014-07-23 | 北京工业大学 | Fuzzy-connectedness-algorithm-based segmentation method of thalamus and substructures of thalamus |
CN110599505A (en) * | 2019-09-17 | 2019-12-20 | 上海微创医疗器械(集团)有限公司 | Organ image segmentation method and device, electronic equipment and storage medium |
CN112150477A (en) * | 2019-11-15 | 2020-12-29 | 复旦大学 | Full-automatic segmentation method and device for cerebral image artery |
CN112785609A (en) * | 2021-02-07 | 2021-05-11 | 重庆邮电大学 | CBCT tooth segmentation method based on deep learning |
CN112884792A (en) * | 2021-02-02 | 2021-06-01 | 青岛海信医疗设备股份有限公司 | Lung image segmentation method and device, electronic equipment and storage medium |
CN113516784A (en) * | 2021-07-27 | 2021-10-19 | 四川九洲电器集团有限责任公司 | Tooth segmentation modeling method and device |
-
2022
- 2022-03-01 CN CN202210196096.8A patent/CN114549559A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942780A (en) * | 2014-03-27 | 2014-07-23 | 北京工业大学 | Fuzzy-connectedness-algorithm-based segmentation method of thalamus and substructures of thalamus |
CN110599505A (en) * | 2019-09-17 | 2019-12-20 | 上海微创医疗器械(集团)有限公司 | Organ image segmentation method and device, electronic equipment and storage medium |
CN112150477A (en) * | 2019-11-15 | 2020-12-29 | 复旦大学 | Full-automatic segmentation method and device for cerebral image artery |
CN112884792A (en) * | 2021-02-02 | 2021-06-01 | 青岛海信医疗设备股份有限公司 | Lung image segmentation method and device, electronic equipment and storage medium |
CN112785609A (en) * | 2021-02-07 | 2021-05-11 | 重庆邮电大学 | CBCT tooth segmentation method based on deep learning |
CN113516784A (en) * | 2021-07-27 | 2021-10-19 | 四川九洲电器集团有限责任公司 | Tooth segmentation modeling method and device |
Non-Patent Citations (1)
Title |
---|
陈燕琳: "口腔CBCT图像牙齿自动分割方法研究", 中国优秀硕士学位论文全文数据库-医药卫生科技辑, 15 January 2021 (2021-01-15), pages 074 - 6 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117095018A (en) * | 2023-10-16 | 2023-11-21 | 北京朗视仪器股份有限公司 | Multi-class tooth segmentation method and device based on CBCT image |
CN117095018B (en) * | 2023-10-16 | 2023-12-22 | 北京朗视仪器股份有限公司 | Multi-class tooth segmentation method and device based on CBCT image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jang et al. | A fully automated method for 3D individual tooth identification and segmentation in dental CBCT | |
CN100517344C (en) | Method and apparatus for medical intervention procedure planning | |
Yi et al. | Automatic catheter and tube detection in pediatric x-ray images using a scale-recurrent network and synthetic data | |
CN112052896B (en) | Image processing method and device, and classification model training method and device | |
CN111145160B (en) | Method, device, server and medium for determining coronary artery branches where calcified regions are located | |
US20130188846A1 (en) | Method, system and computer readable medium for automatic segmentation of a medical image | |
EP3971830A1 (en) | Pneumonia sign segmentation method and apparatus, medium and electronic device | |
CN112150472A (en) | Three-dimensional jaw bone image segmentation method and device based on CBCT (cone beam computed tomography) and terminal equipment | |
CN115205469A (en) | Tooth and alveolar bone reconstruction method, equipment and medium based on CBCT | |
CN111062963A (en) | Blood vessel extraction method, system, device and storage medium | |
JP2020171687A (en) | Systems and methods for processing 3d anatomical volumes based on localization of 2d slices thereof | |
CN112150473A (en) | Three-dimensional jaw bone image segmentation modeling method and device based on CT and terminal equipment | |
CN111166362A (en) | Medical image display method and device, storage medium and electronic equipment | |
Kondo et al. | Computer-based extraction of the inferior alveolar nerve canal in 3-D space | |
CN114549559A (en) | Post-processing method and system for segmenting tooth result based on CBCT (Cone Beam computed tomography) data AI (Artificial Intelligence) | |
Yuce et al. | Detection of pulpal calcifications on bite-wing radiographs using deep learning | |
CN111986217A (en) | Image processing method, device and equipment | |
Cunha et al. | A method for segmentation of dental implants and crestal bone | |
US12026879B2 (en) | Method for detecting the presence of pneumonia area in medical images of patients, detecting system, and electronic device employing method | |
US20220351813A1 (en) | Method and apparatus for training automatic tooth charting systems | |
CN114037665A (en) | Mandibular neural tube segmentation method, mandibular neural tube segmentation device, electronic apparatus, and storage medium | |
CN113724264A (en) | Image segmentation method, system and readable storage medium | |
Wu et al. | Clinical tooth segmentation based on local enhancement | |
CN115578554B (en) | Vascular focus identification method, device, electronic equipment and readable storage medium | |
Zou et al. | A Semi-automatic Segmentation for Tooth on Cone Beam CT Volume Following the Anatomic Guidance. |
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 |