CN104504737B - A kind of method that three-dimensional tracheae tree is obtained from lung CT image - Google Patents
A kind of method that three-dimensional tracheae tree is obtained from lung CT image Download PDFInfo
- Publication number
- CN104504737B CN104504737B CN201510009239.XA CN201510009239A CN104504737B CN 104504737 B CN104504737 B CN 104504737B CN 201510009239 A CN201510009239 A CN 201510009239A CN 104504737 B CN104504737 B CN 104504737B
- Authority
- CN
- China
- Prior art keywords
- image
- dimensional
- tracheae
- bronchus
- segmenta
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 210000004072 lung Anatomy 0.000 title claims abstract description 31
- 210000000621 bronchi Anatomy 0.000 claims abstract description 64
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 20
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 230000003044 adaptive effect Effects 0.000 claims abstract description 7
- 238000005457 optimization Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 13
- 239000000284 extract Substances 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 241000270295 Serpentes Species 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 3
- 241000272525 Anas platyrhynchos Species 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 241001632422 Radiola linoides Species 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 230000000877 morphologic effect Effects 0.000 description 3
- 241001632427 Radiola Species 0.000 description 2
- 210000003484 anatomy Anatomy 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000002685 pulmonary effect Effects 0.000 description 2
- 208000006545 Chronic Obstructive Pulmonary Disease Diseases 0.000 description 1
- 206010062717 Increased upper airway secretion Diseases 0.000 description 1
- 208000029523 Interstitial Lung disease Diseases 0.000 description 1
- 235000004338 Syringa vulgaris Nutrition 0.000 description 1
- 244000297179 Syringa vulgaris Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 210000001124 body fluid Anatomy 0.000 description 1
- 239000010839 body fluid Substances 0.000 description 1
- 210000003123 bronchiole Anatomy 0.000 description 1
- 238000013276 bronchoscopy Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004195 computer-aided diagnosis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 208000026435 phlegm Diseases 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 210000003437 trachea Anatomy 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/005—Tree description, e.g. octree, quadtree
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- 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]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Software Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- High Energy & Nuclear Physics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Computer Graphics (AREA)
- Multimedia (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Endoscopes (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention provides a kind of method that three-dimensional tracheae tree is obtained from lung CT image, comprises the following steps：Main bronchus is extracted using adaptive three-dimensional spatial area growth method；Other segmenta bronchopulmonalias in addition to main bronchus are extracted using the method for optimization image characteristics extraction；Main bronchus and the segmenta bronchopulmonalia are carried out by " suture " using fuzzy connectedness algorithm, obtain three-dimensional tracheae tree.In the processing scheme of the present invention, main bronchus and segmenta bronchopulmonalia are first extracted respectively, efficiently solves the problem of main bronchus in conventional art easily blocks, segmenta bronchopulmonalia is easily revealed.Then, the main bronchus extracted, segmenta bronchopulmonalia are combined into one, obtain complete three-dimensional tracheae tree.
Description
Technical field
The present invention relates to computer image processing technology field, more particularly to one kind obtains three-dimensional tracheae from lung CT image
The method of tree.
Background technology
It is the basis that lung qi pipe relevant diseases parameter diagnoses automatically to obtain accurate intratracheal tree, and lung qi Guan Shu's is accurate
Extraction is significant for the computer-aided diagnosis system of PUD D.Using preoperative lung's three-dimensional CT image, with reference to
Medical Image Processing, computer graphics techniques and modern electronic technology reconstruct three-dimensional tracheae tree, realize in art in real time
Guiding, occurs unexpected probability when can greatly reduce patient's row bronchoscopy.
In order to rebuild the bronchus of more stages, many researchers propose the Anatomical Structure information and phase using bronchial tree
The method for closing information.This kind of method can be roughly divided into five classes：1) method of knowledge based or rule；2) template matching method；3)
Morphological method；4) shape analysis method；5) mixed method.The method of knowledge based or rule is attempted when bronchial tree is rebuild
Introduce anatomy relationship priori, tracheae geometry, local image characteristics, fuzzy logic and the connection of lung's bronchus and blood vessel
The knowledge such as property.The principle of template matching method be according to bronchus anatomical structure priori predefine one group of mask of different sizes or
Template, the extraction of aided two-dimensional or three-dimensional image space mesobronchus architectural feature.For example, Kaftan then attempts a tree path
Stay in place form be used for bronchial reconstruction.Morphological method is commonly used for refinement, and original reconstruction (such as passes through 3D region
Growth algorithm) three-dimensional bronchial tree, its design philosophy is an attempt to utilize various morphological operators, the two of connection or mixed fracture
Dimension/three-dimensional bronchial image-region.For example, Aykac etc. proposes is connected to adjacent image layers using expansion form operator
Same area, to improve the bronchial reconstruction accuracy rate of single layer image.In the literature, Fetita etc. is proposed based on connection cost
The Mathematical Morphology Method of function, detect bronchiolar region.Local image characteristics method is to be usually expressed as pipe according to bronchus
The characteristics of road structure, using the method for the characteristic value for solving Hessian matrixes, increased by analyzing the second dervative of airway boundary
By force with rebuilding three-dimensional tubular trachea tree.
Those skilled in the art are aware that, leakage and obstruction are the two big main bugbears that current bronchial tree is rebuild.Cause
Leakage is that CT images have partial volume effect with the main reason for obstruction, causes the bronchial tube wall of lung and air in air flue
Lumen contrast reduce.Leakage merges the tracheae tree for causing to rebuild with its periphery lung tissue (for example, pulmonary parenchyma)；And block
The tracheae tree for then causing to rebuild is broken and rebuild the discontinuous of tracheae.In addition, breathing fortune when picture noise, image artifacts and imaging
Motion artifacts or image caused by moving obscure, and all can bring huge challenge to the reconstruction of bronchial tree.And work as and run into some lungs
When portion's disease, such as chronic obstructive pulmonary disease or interstitial lung disease, the difficulty of reconstruction can be more obvious.And reveal and block
It is the conflict of three-dimensional tracheae tree extraction, this contradiction is impossible to be completed under single algorithm frame.
Therefore, a kind of side that three-dimensional tracheae tree is obtained from lung CT image of leakage and the influence of obstruction can be reduced by needing badly
Method.
The content of the invention
It can reduce the one of leakage and the influence of obstruction during tracheal reconstruction it is an object of the invention to provide a kind of
The method that kind obtains three-dimensional tracheae tree from lung CT image.This method comprises the following steps：Using adaptive three dimensions area
Domain growth method extracts main bronchus；Other branch gas in addition to main bronchus are extracted using the method for optimization image characteristics extraction
Pipeline section；Main bronchus and the segmenta bronchopulmonalia are carried out by " suture " using fuzzy connectedness algorithm, obtain three-dimensional tracheae tree.
As a kind of preferred scheme, before described the step of extracting main bronchus, first CT images are smoothly located
Reason, by the CT values of certain tissue points in binding analysis CT images and the second order arrangement architecture of the local brightness variation around it, and
Analyze whether the tissue points belong to tubular structure, filter out and belong to bronchial tissue points, collect and belong to branch gas described in whole
The tissue points of pipe obtain one-level tracheae pretreatment image.
As a kind of preferred scheme, before described the step of extracting main bronchus, the one-level tracheae is pre-processed
Image carry out closed operation, first the structural element of the one-level tracheae pretreatment image is expanded, then again to expansion after
Image is corroded with structural element, for filling duck eye, makes the edge smoothing of object, obtains two level tracheae pretreatment image.
As a kind of preferred scheme, in the step of extraction main bronchus, following steps are specifically included：Choose starting
Seed point；Selection adaptive local adjacent thresholds method obtains threshold value as the criterion that region increases, above or equal to the threshold
The tissue points of value are integrated into seed region as seed point；After when periphery, all seed points are all integrated into seed region, led
Bronchi image.
As a kind of preferred scheme, the step of other segmenta bronchopulmonalias extracted in addition to main bronchus described in, specifically
Comprise the following steps：Some characteristics of image are extracted to be used to build the energy term in cost function；Using the method for Multiple Kernel Learning, make
It is embedded into the compound kernel function of the feature in three-dimensional bronchial seed point extraction algorithm, obtains forming the segmenta bronchopulmonalia
Seed point；The seed point for forming the segmenta bronchopulmonalia is carried out being continuously available each independent segmenta bronchopulmonalia according to continuity.
As a kind of preferred scheme, it is special that some characteristics of image include the 3D pipelines based on multiple dimensioned Hessian matrixes
Levy coefficient, local phase, SIFT feature, low-definition version Haar-like features.
As a kind of preferred scheme, in the seed point of the composition segmenta bronchopulmonalia is carried out into continuous step, adopt
With the method based on Snake Spline Models.
It is first in described the step of main bronchus and the segmenta bronchopulmonalia are carried out into " suture " as a kind of preferred scheme
First main bronchus and segmenta bronchopulmonalia independent described in adjacent one end are sutured, are then iterated calculating until all
Segmenta bronchopulmonalia " suture " to together.
As a kind of preferred scheme, in described the step of main bronchus and the segmenta bronchopulmonalia are carried out into " suture ", profit
With central line pick-up algorithm, the center line of the main bronchus, general space distance, local image characteristics, bronchus solution are extracted
The one of which in structure or multinomial degree of sentencing are cutd open, constructs Fuzzy connected degree function, from the end of the main bronchus, is utilized
Three-dimensional fuzzy connection algorithm, main bronchus and the segmenta bronchopulmonalia are connected.
As a kind of preferred scheme, in described the step of main bronchus and the segmenta bronchopulmonalia are carried out into " suture ", profit
With central line pick-up algorithm, the center line of each segmenta bronchopulmonalia is extracted, general space distance, local image characteristics, bronchus are dissected
One of which or multinomial degree of sentencing in structure, Fuzzy connected degree function is constructed, using three-dimensional fuzzy connection algorithm, iteration connection is each
The segmenta bronchopulmonalia, until reconstructing whole bronchial tree.
Implement the present invention, a kind of method that three-dimensional tracheae tree is obtained from lung CT image can be obtained, reduced in tracheae tree
Due to influence caused by leakage and blocking in process of reconstruction.
Brief description of the drawings
Fig. 1 is a kind of Technology Roadmap of method that three-dimensional tracheae tree is obtained from lung CT image provided by the invention；
Fig. 2 is a kind of the human body gas that the method for three-dimensional tracheae tree obtains to be obtained from lung CT image using provided by the invention
Pipe tree construction simulation drawing；
Fig. 3 is a kind of the main branch gas that the method for three-dimensional tracheae tree obtains to be obtained from lung CT image using provided by the invention
Pipe extracts result simulation drawing；
Fig. 4 is a kind of the bronchus that the method for three-dimensional tracheae tree obtains to be obtained from lung CT image using provided by the invention
Section extraction process simulation drawing；
Fig. 5 is a kind of the bronchus that the method for three-dimensional tracheae tree obtains to be obtained from lung CT image using provided by the invention
The extraction result simulation drawing of section；
Fig. 6 is to carry out " suture " using a kind of method that three-dimensional tracheae tree is obtained from lung CT image provided by the invention
Result schematic diagram.
Embodiment
With reference to figure 1, Fig. 2, the present invention provides a kind of method that three-dimensional tracheae tree is obtained from lung CT image, it is main including with
Lower step：As shown in arrow 1, main bronchus is extracted using adaptive three-dimensional spatial area growth method, will for ease of description
The step is referred to as step S101；As shown in arrow 2 and arrow 3, extracted using the method for optimization image characteristics extraction except master
Other extrabronchial segmenta bronchopulmonalias, the step is referred to as step S103；As shown in arrow 4, using fuzzy connectedness algorithm
Main bronchus and segmenta bronchopulmonalia are subjected to " suture ", three-dimensional tracheae tree as shown in Figure 2 is obtained, the step is referred to as step
S105。
When carrying out step S101, preferably first the CT images of three-dimensional are smoothed, schemed by binding analysis CT
The CT values of certain tissue points and the second order arrangement architecture of the local brightness variation around it as in, and analyze whether the tissue points belong to
In tubular structure, filter out and belong to bronchial tissue points, collect whole bronchial tissue points that belong to and obtain one-level tracheae
Pretreatment image.Specific method is to eliminate influence of the noise to extraction tracheae by analyzing Hessian matrixes.
Relation table of the various possible structures with Hessian matrix exgenvalues under the three-dimensional situation of table 1
Relation of the upper table for various possible structures under three-dimensional situation with Hessian matrix exgenvalues, λ k represent k-th of amplitude
Minimum characteristic value, 3 eigenvalue λs 1 of Hessian matrixes, λ 2, λ 3 (| λ 1 |≤| λ 2 |≤| λ 3 |) in, the spy of amplitude maximum
Characteristic vector corresponding to value indicative represents the direction of certain tissue points maximum curvature, and characteristic vector corresponding to the minimum characteristic value of amplitude
Represent the minimum direction of certain tissue points curvature.In CT images, tracheae is always dark, so where lung CT image tracheae
The characteristic value of said three-dimensional body vegetarian refreshments should be that λ 1 is smaller, almost 0, λ 2 and λ 3 are positive number.Each voxel is calculated to CT images
The Hessian matrixes of point, and its characteristic value is calculated, determine whether tracheae voxel.
After implementing above-mentioned steps, enhanced air pipe structure more clearly is demonstrated by out.But due to possible
Gray scale by pipe caused by phlegm or other human body fluids is inconsistent, and tracheae missing inspection caused by tracheae branch, or
The reasons such as other local noises so that the main bronchus extracted still there may be discontinuous phenomenon.So to this one
Level tracheae pretreatment image carries out closed operation, i.e., first the structural element of one-level tracheae pretreatment image is expanded, will be with thing
All background dots of body contact are merged into the object, border is expanded to outside, final result is that image integrally expands one
Circle, is then corroded to the image after expansion with structural element again, is eliminated boundary point, border is internally shunk, entirely close
The process of computing makes the airway boundary of CT images smooth, obtains two level tracheae pretreatment image.
Step S101 is carried out on the basis of two level tracheae pretreatment image is obtained, specifically includes following steps：
Starting seed point is chosen, method for optimizing is manual, can more accurately more rapidly find accurate main bronchus 100
Voxel is as starting seed point；
Selection adaptive local adjacent thresholds method obtains threshold value as the criterion that region increases, above or equal to the threshold value
Tissue points be integrated into seed region as seed point；
Until after all seed points in periphery are all integrated into seed region, the image of main bronchus 100 as shown in Figure 3 is obtained.
Following steps are specifically included in step S103：
Extract some characteristics of image such as 3D pipeline features coefficient, local phase, SIFT based on multiple dimensioned Hessian matrixes
Feature, low-definition version Haar-like features, in the case where cost function energy term form determines, using Multiple Kernel Learning
Method, be embedded into using the compound kernel function of the feature in three-dimensional bronchial seed point extraction algorithm, obtain different characteristic
Suitable weight, the energy term in cost function is built to drive shape distortion；
Obtain forming the seed point of the segmenta bronchopulmonalia 200；
The seed point for forming the segmenta bronchopulmonalia 200 is carried out continuously according to continuity using Snake Spline Models method
Each independent segmenta bronchopulmonalia 200 is obtained, as shown in Figure 4, Figure 5.
Bronchus was both avoided using image characteristics extraction algorithm " fracture " occurs during growth structure --- growth
Stop, turn avoid generation " leakage " --- grow into outside lung qi pipe and organize, such as the region such as alveolar.Can according to clinical experience
Know, trickle segmenta bronchopulmonalia 200, be to have significant multi-scale image feature in partial structurtes, but do not ensure that each other
Connection, it is therefore desirable to which implementation steps S105 carries out " suture ".
Following steps are specifically included in step S105：
The independent segmenta bronchopulmonalia 200 of main bronchus 100 and adjacent one end is carried out " suture " first, specific method is base
" fuzzy " connection is carried out to adjacent segmenta bronchopulmonalia 200 in the centerline direction of local main bronchus 100.By main bronchus 100
After being sutured with the independent segmenta bronchopulmonalia 200 of adjacent one end, the end of the segmenta bronchopulmonalia 200 sutured is recycled to remove mould
Paste and couple adjacent segmenta bronchopulmonalia 200, then arrived together until the segmenta bronchopulmonalia 200 " suture " of whole using iterative calculation,
Obtain measurements of the chest, waist and hips tracheae tree as shown in Figure 6.
Central line pick-up algorithm is employed in above-mentioned stitching step, extracts the center line of main bronchus 100, using same
Method extract the center line, general space distance, local image characteristics, bronchus anatomical structure etc. of each segmenta bronchopulmonalia 200 and sentence
Degree, Fuzzy connected degree function is constructed, from the end of the main bronchus 100, using three-dimensional fuzzy connection algorithm, iteration connects
Each segmenta bronchopulmonalia 200 is connect, until reconstructing whole bronchial tree.
The largest benefit of " suture " is to solve that bronchus tube wall is thinning, liquid is full, bronchus collapses and pathology
Local bronchiole characteristics of image caused by the clinical pulmonary lesions such as change disappears without the problem of method connection.Although most end
3D tracheae results in portion have certain geometric unsharpness, but in clinical tolerance interval, have no effect on the operation of bronchoscope
Evaluation and the whole structure of surgical simulation training.
In the processing scheme of the present invention, main bronchus and segmenta bronchopulmonalia 200 are first extracted respectively, efficiently solves tradition
The problem of main bronchus easily blocks in technology, segmenta bronchopulmonalia 200 is easily revealed.Then, by the main bronchus extracted, branch gas
Pipeline section 200 is combined into one, and obtains complete three-dimensional tracheae tree.
For the person of ordinary skill of the art, without departing from the inventive concept of the premise, if can also make
Dry modification and improvement, these belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be with appended right
It is required that it is defined.
Claims (8)
- A kind of 1. method that three-dimensional tracheae tree is obtained from lung CT image, it is characterised in that comprise the following steps：Main bronchus is extracted using adaptive three-dimensional spatial area growth method；Other segmenta bronchopulmonalias in addition to main bronchus are extracted using the method for optimization image characteristics extraction；Main bronchus and the segmenta bronchopulmonalia are carried out by " suture " using fuzzy connectedness algorithm, obtain three-dimensional tracheae tree；Before described the step of extracting main bronchus, first CT images are smoothed, pass through binding analysis CT images In the CT values of certain tissue points and the second order arrangement architecture of the local brightness variation around it, and analyze whether the tissue points belong to Tubular structure, filter out and belong to bronchial tissue points, collect and belong to bronchial tissue points described in whole and obtain one-level tracheae Pretreatment image；In the step of other segmenta bronchopulmonalias extracted in addition to main bronchus described, following steps are specifically included：Some characteristics of image are extracted to be used to build the energy term in cost function；Using the method for Multiple Kernel Learning, it is embedded into three-dimensional bronchial seed point extraction algorithm using the compound kernel function of the feature In, obtain forming the seed point of the segmenta bronchopulmonalia；The seed point for forming the segmenta bronchopulmonalia is carried out being continuously available each independent segmenta bronchopulmonalia according to continuity.
- 2. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that in the extraction Before the step of going out main bronchus, the one-level tracheae pretreatment image is subjected to closed operation, first the one-level tracheae located in advance The structural element of reason image is expanded, and then the image after expansion is corroded with structural element again, for filling duck eye, Make the edge smoothing of object, obtain two level tracheae pretreatment image.
- 3. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that in the extraction In the step of main bronchus, following steps are specifically included：Choose starting seed point；Selection adaptive local adjacent thresholds method obtains threshold value as the criterion that region increases, above or equal to the threshold value Tissue points are integrated into seed region as seed point；After all seed points are all integrated into seed region when periphery, main bronchus image is obtained.
- 4. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that some figures As feature includes the 3D pipeline features coefficient based on multiple dimensioned Hessian matrixes, local phase, SIFT feature, low resolution version This Haar-like features.
- 5. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that by described group Seed point into the segmenta bronchopulmonalia is carried out in continuous step, employs the method based on Snake Spline Models.
- 6. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that described by master In the step of bronchus and the segmenta bronchopulmonalia carry out " suture ", first by main bronchus and the independent branch gas of adjacent one end Pipeline section is sutured, and is then iterated to calculate and is arrived together until the segmenta bronchopulmonalia " suture " of whole.
- 7. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that described by master In the step of bronchus and the segmenta bronchopulmonalia carry out " suture ", using central line pick-up algorithm, the main bronchus is extracted Center line, general space distance, local image characteristics, the one of which in bronchus anatomical structure or multinomial degree of sentencing, construct mould Connectivity function is pasted, from the end of the main bronchus, using three-dimensional fuzzy connection algorithm, by main bronchus and the branch Tracheae section connects.
- 8. the method for three-dimensional tracheae tree is obtained from lung CT image as claimed in claim 1, it is characterised in that described by master In the step of bronchus and the segmenta bronchopulmonalia carry out " suture ", using central line pick-up algorithm, extract in each segmenta bronchopulmonalia Heart line, general space distance, local image characteristics, the one of which in bronchus anatomical structure or multinomial degree of sentencing, construction are fuzzy Connectivity function, using three-dimensional fuzzy connection algorithm, iteration connects each segmenta bronchopulmonalia, until reconstructing whole bronchus Tree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510009239.XA CN104504737B (en) | 2015-01-08 | 2015-01-08 | A kind of method that three-dimensional tracheae tree is obtained from lung CT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510009239.XA CN104504737B (en) | 2015-01-08 | 2015-01-08 | A kind of method that three-dimensional tracheae tree is obtained from lung CT image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104504737A CN104504737A (en) | 2015-04-08 |
| CN104504737B true CN104504737B (en) | 2018-01-12 |
Family
ID=52946131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510009239.XA Expired - Fee Related CN104504737B (en) | 2015-01-08 | 2015-01-08 | A kind of method that three-dimensional tracheae tree is obtained from lung CT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104504737B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10181191B2 (en) | 2014-12-02 | 2019-01-15 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for identifying spine or bone regions in computed tomography image sequence |
| CN104809730B (en) * | 2015-05-05 | 2017-10-03 | 上海联影医疗科技有限公司 | The method and apparatus that tracheae is extracted from chest CT image |
| CN106485704B (en) * | 2016-09-30 | 2021-02-19 | 上海联影医疗科技股份有限公司 | Method for extracting center line of blood vessel |
| CN106875405B (en) * | 2017-01-19 | 2019-05-21 | 浙江大学 | CT image pulmonary parenchyma template tracheae removing method based on breadth first search |
| EP3633612A4 (en) * | 2017-06-30 | 2020-06-03 | Shanghai United Imaging Healthcare Co., Ltd. | Method and system for segmenting image |
| CN107481251A (en) * | 2017-07-17 | 2017-12-15 | 东北大学 | A kind of method that terminal bronchi tree is extracted from lung CT image |
| CN107507171A (en) * | 2017-08-08 | 2017-12-22 | 东北大学 | A kind of lung CT image air flue three-dimensional framework tree extraction and labeling method |
| CN108171703B (en) * | 2018-01-18 | 2020-09-15 | 东北大学 | Method for automatically extracting trachea tree from chest CT image |
| CN108564564A (en) * | 2018-03-09 | 2018-09-21 | 华南理工大学 | Based on the medical image cutting method for improving fuzzy connectedness and more seed points |
| CN108742838B (en) * | 2018-03-29 | 2020-06-16 | 四川大学华西医院 | Preparation method of lung segment model with quantified inter-segment marker points |
| US11071591B2 (en) * | 2018-07-26 | 2021-07-27 | Covidien Lp | Modeling a collapsed lung using CT data |
| CN111161344B (en) * | 2019-12-31 | 2021-08-10 | 广州永士达医疗科技有限责任公司 | Airway elasticity measuring method, system, equipment and medium based on OCT equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | Hepatic portal vein tree modeling method and system thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036484B (en) * | 2013-03-06 | 2017-08-08 | 东芝医疗系统株式会社 | Image segmenting device, image partition method and medical image equipment |
-
2015
- 2015-01-08 CN CN201510009239.XA patent/CN104504737B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393644A (en) * | 2008-08-15 | 2009-03-25 | 华中科技大学 | Hepatic portal vein tree modeling method and system thereof |
Non-Patent Citations (6)
| Title |
|---|
| Leakage suppression in human airway tree segmentation using shape optimization based on fuzzy connectivity method;Rizi.FY et al;《INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY》;20130213;第23卷(第1期);第71-84页 * |
| Segmentation Of Brain Tumors In Computed Tomography Images Using SVM Classifier;B.Shanmugapriya et al;《2014 International Conference on Electronics and Communication System》;20140213;第1-3页 * |
| 一种基于区域生长的目标提取算法;李旭辉 等;《微电子学与计算机》;20081231;第25卷(第12期);第183-186页 * |
| 一种新的三维气管树提取方法;程远雄 等;《中国医学物理学杂志》;20110730;第20卷(第04期);第2759-2763页 * |
| 医学图像分割与三维可视化技术研究;王树秀;《中国优秀硕士学位论文全文数据库 信息科技辑》;20100415(第04期);第I138-643页 * |
| 基于支持向量机的医学图像相关技术研究;李小娟;《中国优秀硕士学位论文全文数据库 信息科技辑》;20131215(第S1期);第I138-562页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104504737A (en) | 2015-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104504737B (en) | A kind of method that three-dimensional tracheae tree is obtained from lung CT image | |
| Hua et al. | Segmentation of pathological and diseased lung tissue in CT images using a graph-search algorithm | |
| JP5701138B2 (en) | Medical image processing apparatus and method, and program | |
| JP5258694B2 (en) | Medical image processing apparatus and method, and program | |
| US9990743B2 (en) | Suppression of vascular structures in images | |
| Jin et al. | Object recognition in medical images via anatomy-guided deep learning | |
| Dakua | Performance divergence with data discrepancy: a review | |
| Yu et al. | Break: Bronchi reconstruction by geodesic transformation and skeleton embedding | |
| CN106846330A (en) | Human liver's feature modeling and vascular pattern space normalizing method | |
| Qi et al. | Automatic pulmonary fissure detection and lobe segmentation in CT chest images | |
| Hong et al. | Automatic lung nodule matching on sequential CT images | |
| Cheng et al. | Segmentation of the airway tree from chest CT using tiny atrous convolutional network | |
| Xiao et al. | Deep learning-based lung image registration: A review | |
| Zhang et al. | Pathological airway segmentation with cascaded neural networks for bronchoscopic navigation | |
| CN111080556A (en) | Method, system, equipment and medium for strengthening trachea wall of CT image | |
| JP2023548041A (en) | Method and system for segmentation and identification of at least one tubular structure in medical images | |
| Bauer et al. | Airway tree reconstruction based on tube detection | |
| Huang et al. | BronchusNet: Region and Structure Prior Embedded Representation Learning for Bronchus Segmentation and Classification | |
| Shao et al. | A segmentation method of airway from chest CT image based on VGG-Unet neural network | |
| WO2022183851A1 (en) | Lung lobe segmentation method based on digital human technology | |
| Zhao et al. | Automatic segmentation method for bone and blood vessel in murine hindlimb | |
| Tan et al. | A novel method for automated segmentation of airway tree | |
| Feuerstein et al. | Adaptive model based pulmonary artery segmentation in 3D chest CT | |
| Tan et al. | Automatic extraction of 3d airway tree from multislice computed tomography images | |
| Xu et al. | A pilot study to utilize a deep convolutional network to segment lungs with complex opacities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201218 Address after: 518057 c702, c704, 7th floor, Tsinghua Ziguang science and Technology Park, No. 13, Langshan Road, North District, high tech Zone, Nanshan District, Shenzhen City, Guangdong Province Patentee after: SHENZHEN BELTER HEALTH MEASUREMENT AND ANALYSIS TECHNOLOGY Co.,Ltd. Address before: 518060 No. 3688 Nanhai Road, Shenzhen, Guangdong, Nanshan District Patentee before: SHENZHEN University Patentee before: But fruit |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180112 |