CN112164075A - Segmentation method for maxillary sinus membrane morphological change - Google Patents
Segmentation method for maxillary sinus membrane morphological change Download PDFInfo
- Publication number
- CN112164075A CN112164075A CN202011012803.0A CN202011012803A CN112164075A CN 112164075 A CN112164075 A CN 112164075A CN 202011012803 A CN202011012803 A CN 202011012803A CN 112164075 A CN112164075 A CN 112164075A
- Authority
- CN
- China
- Prior art keywords
- maxillary sinus
- vertex
- model
- bone cavity
- sector
- 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.)
- Granted
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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention provides a segmentation method aiming at the form change of a maxillary sinus membrane, which comprises the steps of segmenting and extracting a maxillary sinus air cavity model and a maxillary sinus bone cavity initial model from an obtained maxillary sinus three-dimensional sectional image, scanning the inner surface of the maxillary sinus bone cavity initial model by using a vertex screening method to obtain an inner surface vertex, screening and deleting abnormal vertices by using an average vertex distance filtering method, and constructing a vertex curved surface by using the screened vertices. Attaching the vertex curved surface to the inner surface of the maxillary sinus bone cavity initial model to generate a maxillary sinus bone cavity reconstruction model; and subtracting the maxillary sinus air cavity model from the maxillary sinus bone cavity reconstruction model to obtain the segmentation result of the cyst or the mucosa. The invention can effectively separate cyst or mucosa with the thickness less than 2 mm.
Description
Technical Field
The invention relates to the field of biomedical images, in particular to a segmentation method aiming at morphological changes of a maxillary sinus membrane.
Background
Recently, studies on the association of abnormal morphological changes in the morphology and location of the cyst and mucosal structures of the maxillary sinus with dental lesions using CBCT (cone beam computed tomography) images have been receiving attention. The maxillary sinuses are a pair of sinuses of the maxilla located below the eyes, which are surrounded by four adjacent bones: alveolar process of maxilla, zygomatic bone, infraorbital wall and lateral wall of nasal cavity. Geometrically, the maxillary sinus is a pyramidal structure with its apex at the lateral boundary of the zygomatic bone and base at the intranasal boundary, and its sagittal slice shows how its cross section changes from triangular to quadrangular. The maxillary sinus cavity has a large number of openings and incomplete thin bone boundaries, which bring great difficulty to the cyst segmentation in the sinus cavity. Thus, accurate cutting and quantification of mucosal thickening provides a basis for exploring pathological findings in greater detail in the future. The most common morphological changes of the maxillary sinus are cysts and thickened mucosa, and in order to quantify the thickening of the mucosa and the thickness of the cysts, a mucosal segmentation of the sinus cavity of the maxillary sinus should be obtained first. For the segmentation of cysts (more than 5mm), the region growing method can be suitable for most cyst segmentation and can be accurately cut by using a segmentation tool; however, for the mucosa hyperplasia of the maxillary sinus with the mucosa thickness less than 2mm, the general algorithm for region growing cannot accurately segment, and the maxillary sinus mucosa hyperplasia usually grows insufficiently or overgrows.
Another approach is to use an active contour multi-object segmentation tool, such as the robust statistical segmenter in 3d selier, whose iterative evolution is driven by local robust statistical data, updating the median and quartile spacing in each iteration. It is more resistant to noise and intensity non-uniformities, and is an effective interactive medical image segmentation method using a fast growth, where the user manually draws labels on the target segmented region of the slice image and then grows automatically up to the cyst boundary. However, the abnormal structure of the maxillary sinus may cause the loss of the bone boundary of the bone cavity of the maxillary sinus, the maxillary sinus with cyst or mucosa thickening symptoms may have structural deformation of various degrees, and meanwhile, the thin bone of the wall of the posterior sinus cannot be extracted by the thresholding segmentation, which may cause a large hole and a wide mouth and a large number of openings existing at one side of the maxillary sinus close to the nasal sinus, which may generally cause growth leakage at the fracture boundary of the opening and the thin bone. In addition, for cysts with small thickness (< 3mm), the difficulty of growth and segmentation is further complicated, the shape of the mucosa is more irregular, the thickness is smaller, and the segmentation difficulty is very large, which brings great difficulty to the mucosa or cyst segmentation algorithm.
Chinese patent CN109886969A published in 2019, 6, month and 14 provides a three-dimensional automatic segmentation method of an endoscopic optical coherence tomography image of a respiratory tract, which comprises two steps of image preprocessing and respiratory tract tissue segmentation; the method comprises the following steps: (1) image preprocessing, specifically: (1.1) denoising; (1.2) removing the plastic protective sleeve; (1.3) carrying out binarization and region filtering; (1.4) detecting an upper boundary; (2) the respiratory tract tissue segmentation method comprises the following specific steps: (2.1) detecting the boundary of the respiratory tract lumen surface in the image by using a three-dimensional map search algorithm; (2.2) using a three-dimensional map search method to divide mucosa and submucosal tissues in different search regions, respectively. The method can rapidly and smoothly obtain mucosa segmentation result, but the thickness of the segmented mucosa is not thin enough.
Disclosure of Invention
The invention provides a segmentation method aiming at the morphological change of a maxillary sinus membrane, aiming at overcoming the defect that the thickness of mucosa or cyst of the maxillary sinus segmented by the prior art is not thin enough.
The technical scheme of the invention is as follows:
the invention provides a segmentation method aiming at the morphological change of a maxillary sinus membrane, which comprises the following steps:
s1: obtaining a three-dimensional tomography image of the maxillary sinus by using a CT scanning technology, and recording the three-dimensional tomography image as R;
s2: extracting a maxillary sinus air cavity model marked as A by segmentation from the maxillary sinus three-dimensional sectional image R;
s3: preliminarily segmenting and extracting a maxillary sinus bone cavity initial model from the maxillary sinus three-dimensional sectional image R, and marking as B;
s4: taking a point protruding from the inner surface of the maxillary sinus bone cavity initial model B as a vertex, and constructing a vertex curved surface by using the vertex;
s5: and converting the vertex curved surface into a label voxel and fitting the label voxel onto the inner surface of the initial model B of the maxillary sinus bone cavity to form a maxillary sinus bone cavity reconstruction model, recording the model as C, subtracting the maxillary sinus air cavity model A from the maxillary sinus bone cavity reconstruction model C, and obtaining a division result of mucosa or cyst.
Preferably, the specific operation method for extracting and segmenting the maxillary sinus air cavity model a in S2 is as follows: segmentation was performed using a Robust statistical Segmenter (Robust Statistics Segmenter) in the multi-object segmentation tool (3D Slicer).
Preferably, the specific steps of S3 are:
s3.1: scanning all voxels in the three-dimensional tomography image R of the maxillary sinus, dividing the voxels into three fields according to the intensity values of the voxels; setting intensity value thresholds p and q, wherein voxels are divided into the air field when the voxel intensity value is less than p, voxels are divided into the soft tissue field when the voxel intensity value is between p and q, and voxels are divided into the bone field when the voxel intensity value is greater than q;
s3.2: scanning the voxels which are divided into the soft tissue field again to ensure that the voxels of which the voxel intensity values are close to the intensity value threshold q are accurately divided;
s3.3: voxels belonging to the bone domain are labeled and an initial model B of the maxillary sinus bone cavity is generated.
Preferably, the specific method for dividing the voxels in S3.1 is as follows: the division is carried out by a fuzzy C-mean method.
Preferably, the specific steps of S4 are:
s4.1: scanning the inner surface of the initial maxillary sinus bone cavity model B;
s4.2: defining the vertex set of the initial model B of the bone cavity of the maxillary sinus as follows: all the vertexes of the inner surface of the B,respectively the coordinates and normal vectors of the vertex in the 3D Cartesian space;
s4.3: scanning and selecting a vertex;
for each slice along the path, representing the slice center O as the origin, with coordinates (0, 0, 0); equally dividing the slice into M sectors, wherein the angle of each sector is thetathAnd defining a direction vector for each sector as
At each angle thetathIn the sector part, calculating the distance from each vertex to the circle center O, and selecting the vertex with the minimum distance; let the normal vector of the scanned vertex beWhen in useAnd the fan-shaped direction vectorAngle between theta < thetathThen, the vertex is selected preliminarily; if the vertex is finally selected, the following two conditions are required:
normal vector of vertexNormal vector at scanned vertexThe upper projection angle phi is smaller than the threshold value phith,
S4.4: screening the vertexes obtained in the S4.3 by using an average vertex distance filtering method and deleting abnormal vertexes;
s4.5: and after the vertex screening is finished, constructing a vertex curved surface by using the vertex.
Preferably, the specific steps of S4.4 are as follows:
the distance from a vertex to the center O in a sector of a slice is d (i), and the average of the distances from all vertices to the center O in a sector is dsect-mean(k) Then, thenWherein, i ═ 1.., Nk,NkIs the number of vertices in the sector;
let d be the sector of the slicesect-mean(k) Has an average value of dslice-meanThen, thenWherein k is 1.., M is the number of the sector parts; if | dsect-mean(k)-dslice-mean|<daThe vertices in the corresponding sector will be preserved, daIs a global threshold;
d of each sectorsect-mean(k) D of the preceding three sectorssect-mean(k) For further comparison, if the difference is denoted as z (m), then z (m) ═ dsect-mean(k)-dsect-mean(k + m) |, wherein m ═ 1, 2, 3; if any one of Z (m) > dlThe vertices in the corresponding sector are considered as outlier fixed points and are deleted, dlIs a local threshold.
Preferably, before S4.1 is implemented, any 4 misaligned points need to be marked on the nasal wall surface in the initial maxillary sinus cavity model B: m1, M2, M3 and M4, and marking any point Apex outside the nasal wall surface in the initial model B of the maxillary sinus bone cavity, calculating the geometric constraint relation of the initial model B of the maxillary sinus bone cavity by using the 5 marked points, and generating a slice path.
Preferably, in S4.1, the scanning operation is implemented using a vertex screening algorithm.
Preferably, the vertex surfaces in S4.5 are constructed by poisson reconstruction (poissons reconstruction).
Preferably, the specific operation method for converting the vertex curved surface into the label voxel in S5 is as follows: the translation is implemented using Model to label mapping (Model to label map) in the multi-object segmentation tool (3D Slicer).
Compared with the prior art, the technical scheme of the invention has the beneficial effects that: the method screens peaks which meet the requirements of the inner surface of the initial model of the maxillary sinus bone cavity, constructs a peak curved surface through the peaks, further generates a maxillary sinus bone cavity reconstruction model, and subtracts the maxillary sinus air cavity model from the maxillary sinus bone cavity reconstruction model to obtain the segmentation result of cyst or mucosa; the thickness of the cyst or the mucosa divided by the invention is far smaller than that of the cyst or the mucosa divided by the prior art.
Drawings
Fig. 1 is a flowchart of a segmentation method for morphological changes of maxillary sinus membrane according to embodiment 1.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
The present embodiment provides a segmentation method for morphological changes of maxillary sinus membrane, as shown in fig. 1, the method comprises the following steps:
s1: obtaining a three-dimensional tomography image of the maxillary sinus by using a CT scanning technology, and recording the three-dimensional tomography image as R;
s2: extracting a maxillary sinus air cavity model marked as A by segmentation from the maxillary sinus three-dimensional sectional image R;
s3: preliminarily segmenting and extracting a maxillary sinus bone cavity initial model from the maxillary sinus three-dimensional sectional image R, and marking as B;
s4: taking a point protruding from the inner surface of the maxillary sinus bone cavity initial model B as a vertex, and constructing a vertex curved surface by using the vertex;
s5: and converting the vertex curved surface into a label voxel and fitting the label voxel onto the inner surface of the initial model B of the maxillary sinus bone cavity to form a maxillary sinus bone cavity reconstruction model, recording the model as C, subtracting the maxillary sinus air cavity model A from the maxillary sinus bone cavity reconstruction model C, and obtaining a division result of mucosa or cyst.
The specific operation method for extracting and segmenting the maxillary sinus air cavity model A in the S2 comprises the following steps: segmentation is performed using a Robust statistical Segmenter (Robust Statistics Segmenter) in a multi-object segmentation tool.
The specific steps of S3 are as follows:
s3.1: : scanning all voxels in the three-dimensional tomography image R of the maxillary sinus, dividing the voxels into three fields according to the intensity values of the voxels; setting intensity value thresholds p and q, wherein voxels are divided into the air field when the voxel intensity value is less than p, voxels are divided into the soft tissue field when the voxel intensity value is between p and q, and voxels are divided into the bone field when the voxel intensity value is greater than q;
s3.2: scanning the voxels which are divided into the soft tissue field again to ensure that the voxels of which the voxel intensity values are close to the intensity value threshold q are accurately divided;
s3.3: voxels belonging to the bone domain are labeled and an initial model B of the maxillary sinus bone cavity is generated.
The specific method for dividing the voxels in the S3.1 comprises the following steps: and adopting a fuzzy C-mean algorithm for division.
The specific steps of S4 are as follows:
s4.1: scanning the inner surface of the initial maxillary sinus bone cavity model B;
s4.2: defining the vertex set of the initial model B of the bone cavity of the maxillary sinus as follows: all the vertexes of the inner surface of the B,respectively the coordinates and normal vectors of the vertex in the 3D Cartesian space;
s4.3: scanning and selecting a vertex;
for each slice along the path, representing the slice center O as the origin, with coordinates (0, 0, 0); equally dividing the slice into M sectors, wherein the angle of each sector is thetathAnd defining a direction vector for each sector asIn the present embodiment, the slice is equally divided into 90 slices, and each sector has an angle of 4 °, i.e., M is 90, θth=4°;
At each angle thetathIn the sector part, calculating the distance from each vertex to the circle center O, and selecting the vertex with the minimum distance; let the normal vector of the scanned vertex beWhen in useAnd the fan-shaped direction vectorAngle between theta < thetathThen, the vertex is selected preliminarily; if the vertex is finally selected, the following two conditions are required:
normal vector of vertexNormal vector at scanned vertexThe upper projection angle phi is smaller than the threshold value phith,
S4.4: screening the vertexes obtained in the S4.3 by using an average vertex distance filtering method and deleting abnormal vertexes;
s4.5: and after the vertex screening is finished, reconstructing the curved surface by using the vertex.
The S.4.4 comprises the following specific steps:
the distance from a vertex to the center O in a sector of a slice is d (i), and the average of the distances from all vertices to the center O in a sector is dsect-mean(k) Then, thenWherein, i ═ 1.., Nk,NkIs the number of vertices in the sector;
let d be the sector of the slicesect-mean(k) Has an average value of dslice-meanThen, thenWherein k is 1.., M is the number of the sector parts; in the embodiment, the number of the fan-shaped parts is 90, namely M is 90; if | dsect-mean(k)-dslice-mean|<daThe vertices in the corresponding sector will be preserved, daIs a global threshold; the global threshold in this embodiment is 5, da=5;
D of each sectorsect-mean(k) D of the preceding three sectorssect-mean(k) For further comparison, if the difference is denoted as z (m), then z (m) ═ dsect-mean(k)-dsect-mean(k + m) |, wherein m ═ 1, 2, 3; if Z (m) > dlThe vertices in the corresponding sector are considered as outlier fixed points and are deleted, dlIs a local threshold; in this embodiment the local threshold is 3, dl=3;
Before the implementation of the S4.1, any 4 misaligned points need to be marked on the nasal wall surface in the maxillary sinus bone cavity initial model B: m1, M2, M3 and M4, and marking any point Apex outside the nasal wall surface in the initial model B of the maxillary sinus bone cavity, calculating the geometric constraint relation of the initial model B of the maxillary sinus bone cavity by using the 5 marked points, and generating a slice path.
In S4.1, a vertex screening algorithm is used to implement the scanning operation.
And adopting Poisson reconstruction (PoissonsourceReconstruction) when the vertex curved surface in the S4.5 is constructed.
The vertices selected in S4.5 were mirrored on the mirror plane of the nasal wall constructed from M1-M4 when reconstructed.
The specific operation method of converting the reconstructed surface into the label voxel in S5 is to use a Model to label map (Model to label map) in a multi-object segmentation tool (3D Slicer) to implement the conversion.
The technical scheme of the embodiment has the beneficial effects that: in the embodiment, the vertexes of the internal surface of the initial model of the maxillary sinus bone cavity, which meet the requirements, are screened out, the vertex curved surface is constructed through the vertexes, the maxillary sinus bone cavity reconstruction model is further generated, and the maxillary sinus bone cavity reconstruction model is used for subtracting the maxillary sinus air cavity model to obtain the segmentation result of the cyst or mucosa; the cyst or mucosa with the thickness less than 2mm can be effectively separated.
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A segmentation method aiming at the morphological change of a maxillary sinus membrane is characterized by comprising the following steps:
s1: obtaining a three-dimensional tomography image of the maxillary sinus by using a CT scanning technology, and recording the three-dimensional tomography image as R;
s2: extracting a maxillary sinus air cavity model marked as A by segmentation from the maxillary sinus three-dimensional sectional image R;
s3: preliminarily segmenting and extracting a maxillary sinus bone cavity initial model from the maxillary sinus three-dimensional sectional image R, and marking as B;
s4: taking a point protruding from the inner surface of the maxillary sinus bone cavity initial model B as a vertex, and constructing a vertex curved surface by using the vertex;
s5: and converting the vertex curved surface into a label voxel and fitting the label voxel onto the inner surface of the initial model B of the maxillary sinus bone cavity to form a maxillary sinus bone cavity reconstruction model, recording the model as C, subtracting the maxillary sinus air cavity model A from the maxillary sinus bone cavity reconstruction model C, and obtaining a division result of mucosa or cyst.
2. The method for segmenting maxillary sinus membrane morphological change according to claim 1, wherein the S2 is implemented by extracting a maxillary sinus air cavity model a by: segmentation is performed using a robust statistical segmenter in a multi-object segmentation tool.
3. The method for segmenting the morphological change of the maxillary sinus membrane according to the claim 2, wherein the specific steps of the step S3 are as follows:
s3.1: scanning all voxels in the three-dimensional tomography image R of the maxillary sinus, dividing the voxels into three fields according to the intensity values of the voxels; setting intensity value thresholds p and q, wherein voxels are divided into the air field when the voxel intensity value is less than p, voxels are divided into the soft tissue field when the voxel intensity value is between p and q, and voxels are divided into the bone field when the voxel intensity value is greater than q;
s3.2: scanning the voxels which are divided into the soft tissue field again to ensure that the voxels of which the voxel intensity values are close to the intensity value threshold q are accurately divided;
s3.3: voxels belonging to the bone domain are labeled and an initial model B of the maxillary sinus bone cavity is generated.
4. A segmentation method aiming at the morphological change of the maxillary sinus membrane according to claim 3, wherein the specific method for dividing the voxels in S3.1 is as follows: and adopting a fuzzy C-mean algorithm for division.
5. The method for segmenting maxillary sinus membrane morphological change according to claim 4, wherein the specific steps of S4 are as follows:
s4.1: scanning the inner surface of the initial maxillary sinus bone cavity model B;
s4.2: defining the vertex set of the initial model B of the bone cavity of the maxillary sinus as follows: all the vertexes of the inner surface of the B,respectively the coordinates and normal vectors of the vertex in the 3D Cartesian space;
s4.3: scanning and selecting a vertex;
for each slice along the path, representing the slice center O as the origin, with coordinates (0, 0, 0); equally dividing the slice into M sectors, wherein the angle of each sector is thetathAnd defining a direction vector for each sector as
At each angle thetathIn the sector part, calculating the distance from each vertex to the circle center O, and selecting the vertex with the minimum distance; let the normal vector of the scanned vertex beWhen in useAnd the fan-shaped direction vectorAngle between theta < thetathThen, the vertex is selected preliminarily; if the vertex is finally selected, the following two conditions are required:
normal vector of vertexNormal vector at scanned vertexThe upper projection angle phi is smaller than the threshold value phith,
S4.4: screening the vertexes obtained in the S4.3 by using an average vertex distance filtering method and deleting abnormal vertexes;
s4.5: and after the vertex screening is finished, constructing a vertex curved surface by using the vertex.
6. The method for segmenting the morphological change of the maxillary sinus membrane according to the claim 6, wherein the S4.4 comprises the following specific steps:
the distance from a vertex to the center O in a sector of a slice is d (i), and the average of the distances from all vertices to the center O in a sector is dsect-mean(k) Then, thenWherein, i ═ 1.., Nk,NkIs the number of vertices in the sector;
let d be the sector of the slicesect-mean(k) Has an average value of dslice-meanThen, thenWherein k is 1.., M is the number of the sector parts; if | dsect-mean(k)-dslice-mean|<daThe vertices in the corresponding sector will be preserved, daIs a global threshold;
d of each sectorsect-mean(k) D of the preceding three sectorssect-mean(k) For further comparison, if the difference is denoted as z (m), then z (m) ═ dsect-mean(k)-dsect-mean(k + m) |, wherein m ═ 1, 2, 3; if any one of Z (m) > dlThe vertices in the corresponding sector are considered as outlier fixed points and are deleted, dlIs a local threshold.
7. The method for segmenting maxillary sinus membrane morphological change according to claim 6, wherein before S4.1 is implemented, any 4 misaligned points are marked on the nasal wall surface in the maxillary sinus bone cavity initial model B: m1, M2, M3 and M4, and marking any point Apex outside the nasal wall surface in the initial model B of the maxillary sinus bone cavity, calculating the geometric constraint relation of the initial model B of the maxillary sinus bone cavity by using the 5 marked points, and generating a slice path.
8. The method for segmenting the morphological change of the maxillary sinus membrane according to the claim 7, wherein the scanning operation is implemented by using a vertex screening algorithm in the S4.1.
9. The method for segmenting the morphological change of the maxillary sinus membrane according to the claim 8, wherein the vertex curved surface in the S4.5 is constructed by Poisson reconstruction.
10. A segmentation method for morphological changes of maxillary sinus membrane according to claim 9, wherein the specific operation method of converting vertex curved surface into label voxel in S5 is as follows: the transformation is achieved using model-to-label mapping in a multi-object segmentation tool.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011012803.0A CN112164075B (en) | 2020-09-23 | 2020-09-23 | Segmentation method for maxillary sinus membrane morphology change |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011012803.0A CN112164075B (en) | 2020-09-23 | 2020-09-23 | Segmentation method for maxillary sinus membrane morphology change |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112164075A true CN112164075A (en) | 2021-01-01 |
CN112164075B CN112164075B (en) | 2023-08-01 |
Family
ID=73863611
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011012803.0A Active CN112164075B (en) | 2020-09-23 | 2020-09-23 | Segmentation method for maxillary sinus membrane morphology change |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112164075B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102615323B1 (en) * | 2023-05-10 | 2023-12-19 | 서울대학교산학협력단 | System for automatic segmentation for maxillary sinus and lesion, method for operation thereof and method for artificial intelligent training of automatic segmentation model |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105913424A (en) * | 2016-04-08 | 2016-08-31 | 北京大学口腔医院 | Tooth-based age estimation method and device |
CN106875432A (en) * | 2017-03-09 | 2017-06-20 | 南京医科大学附属口腔医院 | Remporomandibular joint moves method for reconstructing and system |
CN107730542A (en) * | 2017-08-29 | 2018-02-23 | 北京大学 | Cone beam computed tomography image corresponds to and method for registering |
CN108629839A (en) * | 2018-05-09 | 2018-10-09 | 西安增材制造国家研究院有限公司 | The method for obtaining full dental cast using the oral cavity CT images under dental articulation state |
CN108665533A (en) * | 2018-05-09 | 2018-10-16 | 西安增材制造国家研究院有限公司 | A method of denture is rebuild by tooth CT images and 3 d scan data |
CN110189352A (en) * | 2019-05-21 | 2019-08-30 | 重庆布瑞斯科技有限公司 | A kind of root of the tooth extracting method based on oral cavity CBCT image |
CN110895816A (en) * | 2019-10-14 | 2020-03-20 | 广州医科大学附属口腔医院(广州医科大学羊城医院) | Method for measuring alveolar bone grinding amount before mandibular bone planting plan operation |
-
2020
- 2020-09-23 CN CN202011012803.0A patent/CN112164075B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105913424A (en) * | 2016-04-08 | 2016-08-31 | 北京大学口腔医院 | Tooth-based age estimation method and device |
CN106875432A (en) * | 2017-03-09 | 2017-06-20 | 南京医科大学附属口腔医院 | Remporomandibular joint moves method for reconstructing and system |
CN107730542A (en) * | 2017-08-29 | 2018-02-23 | 北京大学 | Cone beam computed tomography image corresponds to and method for registering |
CN108629839A (en) * | 2018-05-09 | 2018-10-09 | 西安增材制造国家研究院有限公司 | The method for obtaining full dental cast using the oral cavity CT images under dental articulation state |
CN108665533A (en) * | 2018-05-09 | 2018-10-16 | 西安增材制造国家研究院有限公司 | A method of denture is rebuild by tooth CT images and 3 d scan data |
CN110189352A (en) * | 2019-05-21 | 2019-08-30 | 重庆布瑞斯科技有限公司 | A kind of root of the tooth extracting method based on oral cavity CBCT image |
CN110895816A (en) * | 2019-10-14 | 2020-03-20 | 广州医科大学附属口腔医院(广州医科大学羊城医院) | Method for measuring alveolar bone grinding amount before mandibular bone planting plan operation |
Non-Patent Citations (1)
Title |
---|
谭珂, 郭光友, 潘新华, 王大君: "适用于虚拟手术的鼻腔模型三维重建", 军医进修学院学报, no. 04, pages 1 - 4 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102615323B1 (en) * | 2023-05-10 | 2023-12-19 | 서울대학교산학협력단 | System for automatic segmentation for maxillary sinus and lesion, method for operation thereof and method for artificial intelligent training of automatic segmentation model |
Also Published As
Publication number | Publication date |
---|---|
CN112164075B (en) | 2023-08-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108171703B (en) | Method for automatically extracting trachea tree from chest CT image | |
US8200355B2 (en) | Method for manufacturing artificial implants | |
Xia et al. | Individual tooth segmentation from CT images scanned with contacts of maxillary and mandible teeth | |
US20130022251A1 (en) | Method and system for tooth segmentation in dental images | |
US20130022252A1 (en) | Panoramic image generation from cbct dental images | |
CN112136157B (en) | Method, system and computer program for segmenting dental pulp area from image | |
CN106909947B (en) | Mean Shift algorithm-based CT image metal artifact elimination method and system | |
CN106846346B (en) | Method for rapidly extracting pelvis outline of sequence CT image based on key frame mark | |
CN114830174A (en) | Image analysis system for identifying lung features | |
CN109493943B (en) | Three-dimensional visual scalp craniotomy positioning method combined with optical surgical navigation | |
CN106780491B (en) | Initial contour generation method adopted in segmentation of CT pelvic image by GVF method | |
CN112164075B (en) | Segmentation method for maxillary sinus membrane morphology change | |
US20220358740A1 (en) | System and Method for Alignment of Volumetric and Surface Scan Images | |
KR101625955B1 (en) | Method of classifying artery and vein of organ | |
Sarada et al. | MLO Mammogram Pectoral Masking with Ensemble of MSER and Slope Edge Detection and Extensive Pre-Processing | |
Tschirren et al. | Airway segmentation framework for clinical environments | |
Zhang et al. | Automatic rib segmentation in chest CT volume data | |
Roy et al. | MDL-IWS: multi-view deep learning with iterative watershed for pulmonary fissure segmentation | |
CN111145353B (en) | Method for generating 3D point cloud through image segmentation and grid feature point extraction algorithm | |
Păvăloiu et al. | 3D dental reconstruction from CBCT data | |
CN106934785B (en) | Medical image segmentation method for liver model in robot virtual training system | |
CN109767396B (en) | Oral cavity CBCT image denoising method based on image dynamic segmentation | |
CN113706687A (en) | Nose environment modeling method and device for path planning | |
Pazokifard et al. | 3-d segmentation of human sternum in lung mdct images | |
CN111489434A (en) | Medical image three-dimensional reconstruction method based on three-dimensional graph cut |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |