CN102982531B - Bronchial dividing method and system - Google Patents
Bronchial dividing method and system Download PDFInfo
- Publication number
- CN102982531B CN102982531B CN201210423958.2A CN201210423958A CN102982531B CN 102982531 B CN102982531 B CN 102982531B CN 201210423958 A CN201210423958 A CN 201210423958A CN 102982531 B CN102982531 B CN 102982531B
- Authority
- CN
- China
- Prior art keywords
- bronchus
- lung
- repeatedly
- overall area
- gray
- 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.)
- Active
Links
Abstract
The present invention is applicable to Medical Image Segmentation Techniques field, provide a kind of bronchial dividing method and system, it is as follows that described method comprises step: the first segmentation step, split extract lung and bronchus overall area by 3D region growth algorithm from original lung images; Bronchus strengthens step, carries out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area; Second segmentation step, from repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area, segmentation extracts bronchus.Whereby, the present invention strengthens bronchus by the information by blood vessel, to avoid the impact of gray-value variation, makes bronchial segmentation result more stable and accurate.
Description
Technical field
The present invention relates to Medical Image Segmentation Techniques field, particularly relate to a kind of bronchial dividing method and system.
Background technology
Lung is the respiratory apparatus that body weight for humans is wanted, and comprises left lung and right lung, lays respectively at the both sides of chest, and each lung is connected with tracheae by bronchus.Right lung is split by two lungs and is divided into three blades, and left lung is split by a lung and is divided into two blades.Bronchial tree originates in tracheae, by main bronchus, and lobar bronchi, segmental bronchi, and the tree structure that above branch is step by step formed.Left and right lung can be divided into ten sections according to segmental bronchi.The image technologies such as CT (Computed Tomography, computed tomography), MRI (Magnetic ResonanceImaging, magnetic resonance imaging) can show the bronchus that the above diameter of segmental bronchi is millimeter magnitude.For Diagnosis and Treat asthma, bronchiectasis, lung tumors or cancerous issue excision provide iconography basis.
But CT, although MRI two-dimensional ct image provides the information of focus, doctor by virtue of experience can only judge lesion information, adds risk and the difficulty of operation.Doctor rule of thumb judges tumour and bronchial position relationship, excises the upper and lower lobe of the lung or the upper, middle and lower lobe of the lung.The process developing into medical image of computer graphic image technology provides technical support.From medical image, extract the basis that target object is Iamge Segmentation and three-dimensional reconstruction exactly, be the prerequisite of computer-aided diagnosis, has important using value to raising medical diagnosis rate.The method of employing Iamge Segmentation carries out lung and bronchus extracts, and according to segmental bronchi, lung is divided into ten sections, and doctor passes through the relativeness of the data visual pattern ground observation tumour after rebuilding or cancerous issue and each lung section, and then optionally excises lung section.The risk of operation can be reduced, improve the success ratio of operation.
Chinese patent application CN201110034523.4 discloses a kind of medical image diagnosis assisting technology, realizes the aided diagnosis technique of the 3 d medical images of subject chest region.Extracted by lobe of the lung region and bronchus, infer the sub-section in the local of the lobe of the lung according to bronchial branched structure region, carry out lung evaluation of estimate according to the sub-section in the local of the lobe of the lung.The program adopts region extension method to extract the set of pixels zoarium of bronchiolar region, but because region extension method splits according to bronchial gray feature, easily be subject to the impact of gray-value variation, and be difficult to distinguish in the region that bronchus is similar to lung gray scale.Due to intensive intersections of Tissue distribution such as chest endoluminal vascular, organ, bronchuses, and bronchus tip is very thin, and now chest area is difficult to distinguish with pulmonary air region, therefore adopts and easily leaks into lung areas based on area extension method.
In summary, obviously there is inconvenience and defect in actual use, so be necessary to be improved in the bronchi segmentation technology of existing medical image.
Summary of the invention
For above-mentioned defect, the object of the present invention is to provide a kind of bronchial dividing method and system.It strengthens bronchus by the information by blood vessel, to avoid the impact of gray-value variation, makes bronchial segmentation result more stable and accurate.
To achieve these goals, the invention provides a kind of bronchial dividing method, comprise step as follows:
First segmentation step, is split by 3D region growth algorithm and extracts lung and bronchus overall area from original lung images;
Bronchus strengthens step, carries out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area;
Second segmentation step, from repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area, segmentation extracts bronchus.
According to bronchial dividing method of the present invention, described first segmentation step comprises further:
Lung in described original lung images or bronchus choose the first Seed Points, and first threshold scope is set, calculate with described first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described first threshold, as described lung and bronchus overall area carry out segmentation extraction;
Described bronchus strengthens step and comprises further:
The gray-scale value that gray-scale value in described lung and bronchus overall area is greater than the pixel of 0 is set to 0, obtains the described lung after greyscale transformation and bronchus overall area;
Described lung after greyscale transformation and bronchus overall area are carried out to the repeatedly expansive working of some neighborhoods, and the repeatedly etching operation of some neighborhoods is carried out to the data after repeatedly expansive working;
The gray-scale value of each pixel in described lung after the gray-scale value of each pixel in repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, chooses larger gray-scale value as gray count end value;
The gray-scale value of the described lung after described gray count end value and greyscale transformation and bronchus overall area is carried out second time compare operation, if be greater than predetermined threshold value, then alternatively bronchus.
According to bronchial dividing method of the present invention, described bronchus strengthens in step, adopt different number of operations to carry out described expansive working, described etching operation, described first time compare operation and described second time compare operation to the bronchus of different radii, and the thicker then described number of operations of described bronchus is more.
According to bronchial dividing method of the present invention, described some neighborhoods are four neighborhoods or eight neighborhood.
According to bronchial dividing method of the present invention, described second segmentation step comprises further:
The second Seed Points is chosen in described candidate's bronchus, and Second Threshold scope is set, calculate with described second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described Second Threshold, as final bronchus carry out segmentation extraction.
The present invention also provides a kind of bronchial segmenting system, includes:
First segmentation module, extracts lung and bronchus overall area for being split from original lung images by 3D region growth algorithm;
Bronchus strengthens module, for carrying out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area;
Second segmentation module, extracts bronchus for splitting from repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area.
According to bronchial segmenting system of the present invention, described first segmentation module also chooses the first Seed Points in the lung in described original lung images or bronchus, and first threshold scope is set, calculate with described first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described first threshold, as described lung and bronchus overall area carry out segmentation extraction;
Described bronchus strengthens module and comprises further:
Greyscale transformation submodule, is set to 0 for gray-scale value gray-scale value in described lung and bronchus overall area being greater than the pixel of 0, obtains the described lung after greyscale transformation and bronchus overall area;
Expansion/etching operation submodule, for carrying out the repeatedly expansive working of some neighborhoods to the described lung after greyscale transformation and the bronchiolar region in bronchus overall area, and carries out the repeatedly etching operation of some neighborhoods to the data after repeatedly expansive working;
First comparison sub-module, for the gray-scale value of each pixel in the described lung after the gray-scale value of each pixel in repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, choose larger gray-scale value as gray count end value;
Second comparison sub-module, for the gray-scale value of the described lung after described gray count end value and greyscale transformation and bronchus overall area is carried out second time compare operation, if be greater than predetermined threshold value, then alternatively bronchus.
According to bronchial segmenting system of the present invention, described bronchus strengthens module and is used for adopting different number of operations to carry out described expansive working, described etching operation, first time compare operation and second time compare operation to the bronchus of different radii, and the thicker then described number of operations of described bronchus is more.
According to bronchial segmenting system of the present invention, described some neighborhoods are four neighborhoods or eight neighborhood.
According to bronchial segmenting system of the present invention, described second segmentation module also for choosing the second Seed Points in described candidate's bronchus, and Second Threshold scope is set, calculate with described second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described Second Threshold, as final bronchus carry out segmentation extraction.
The present invention considers because lung tissue and bronchus all belong to density regions, is easily subject to the interference of lung tissue when therefore the difficult point of bronchi segmentation is segmentation bronchus.Therefore, the present invention is according to the relevance of bronchus and blood vessel, the i.e. characteristic that always grows together of bronchus and blood vessel, bronchus is strengthened by the information by blood vessel, can avoid the impact of gray-value variation like this, bronchial segmentation result is more stable and accurate, and can split segmental bronchi and above branch, meet the requirement to bronchi segmentation in clinical medicine, there is higher clinical value.
Accompanying drawing explanation
Fig. 1 is the structural representation of the bronchial segmenting system of the present invention;
Fig. 2 is the structural representation of the preferred bronchial segmenting system of the present invention;
Fig. 3 is the process flow diagram of the bronchial dividing method of the present invention;
Fig. 4 is the process flow diagram of the bronchial dividing method of the preferred embodiment of the present invention; And
Fig. 5 is bronchi segmentation result figure of the present invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Fig. 1 is the structural representation of the bronchial segmenting system of the present invention, and described system 100 includes the first segmentation module 10, bronchus strengthens module 20 and the second segmentation module 30, wherein:
Described first segmentation module 10, extract lung and bronchus overall area for being split from original lung images by 3D region growth algorithm, described lung and bronchus overall area comprise lung tissue and bronchus.Because the gray-scale value of lung tissue and intrabronchial air is starkly lower than surrounding tissue, so 3D region growth algorithm can be adopted to carry out lung tissue and bronchial extraction.3D region growth algorithm to need in lung tissue or bronchus automatically or interactive mode chooses first Seed Points, suitable first threshold scope is set, as [-2000,-250], calculate and the first Seed Points being connected within the scope of spatial neighborhood and all pixels of tonal range within the scope of first threshold, be described lung and bronchus overall area.Described original lung images can be lung CT image or lung MRI image.
Described bronchus strengthens module 20, for carrying out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area.Because lung tissue and bronchus all belong to density regions, when the difficult point of bronchi segmentation is segmentation bronchus, be easily subject to the interference of lung tissue.The blood of pulmonary alveoli and capillary has gas exchanges, carbon dioxide in blood is transported to external by bronchus, bronchus enters blood from the oxygen of external conveying, bronchus and blood vessel grow together, and the dilation and erosion computing of the Two-dimensional morphology that can adopt is to realize bronchial enhancing.Because bronchial gray-scale value is usually less than-750, the gray-scale value of blood vessel higher than 0, so be easy to bronchus to be separated with blood vessel.For the gray level image of lung and bronchus overall area, strengthen bronchus by the maximal value that repeatedly expansive working calculates in contiguous range; , and removed the impact of lung tissue by the repeatedly etching operation minimum value calculated in contiguous range, like this can by all bronchi segmentation of accompanying with blood vessel out.
Described second segmentation module 30, extracts bronchus for splitting from repeatedly expansive working and the lung repeatedly after etching operation and bronchus overall area.
The present invention is according to the relevance of bronchus and blood vessel, bronchus is strengthened by the information by blood vessel, the impact of gray-value variation can be avoided like this, segmentation result is stablized, and can split segmental bronchi and above branch, meet clinically to the requirement of bronchi segmentation, there is higher clinical value.
Fig. 2 is the structural representation of the preferred bronchial segmenting system of the present invention, and described system 100 includes the first segmentation module 10, bronchus strengthens module 20 and the second segmentation module 30, wherein:
Described first segmentation module 10, also in the lung in original lung images or bronchus, choose the first Seed Points, and first threshold scope is set, as [-2000,-250], calculate with the first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of first threshold, as lung and bronchus overall area carry out segmentation extraction.Because the gray-scale value of lung tissue and intrabronchial air is starkly lower than surrounding tissue, so 3D region growth algorithm can be adopted to carry out lung tissue and bronchial extraction.
Described bronchus strengthens module 20 and comprises further:
Greyscale transformation submodule 21, is set to 0 for gray-scale value gray-scale value in lung and bronchus overall area being greater than the pixel of 0, obtains the lung after greyscale transformation and bronchus overall area.Such as in lung CT image, lung and bronchial tonal range are all less than 0, so gray-scale value gray-scale value in lung CT image being greater than the pixel of 0 is set to 0.The interference of high-density region can be eliminated like this.
Expansion/etching operation submodule 22, for carrying out the repeatedly expansive working of some neighborhoods to the bronchiolar region in the lung after greyscale transformation and bronchus overall area, and the data after repeatedly expansive working are carried out to the repeatedly etching operation of some neighborhoods, described some neighborhoods are preferably four neighborhoods or eight neighborhood, and preferred computing formula is as follows:
First comparison sub-module 23, for the gray-scale value of each pixel in the lung after the gray-scale value of each pixel in repeatedly expansive working and the lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, choose larger gray-scale value as gray count end value.The preferred computing formula of f is as follows:
Formula two represents that the image I after to greyscale transformation (i.e. lung and bronchus overall area) carries out the dilation operation of b time and the erosion operation of b time, gets the value larger with image I.In order to accelerate computing velocity, dilation and erosion operation can be carried out in the partition data calculating lung and bronchus overall area.
Second comparison sub-module 24, for the gray-scale value of the lung after described gray count end value and greyscale transformation and bronchus overall area is carried out second time compare operation, if be greater than predetermined threshold value T, then alternatively bronchus.Often carry out once first time compare operation and will carry out second compare operation, the number of times of visible second time compare operation is identical with the number of times of first time compare operation, and preferred computing formula is as follows:
Preferably, described bronchus strengthens module 20 and carries out described expansive working, described etching operation, described first time compare operation and described second time compare operation for adopting different number of operations to the bronchus of different radii, and the thicker then number of operations of bronchus is more.Such as, for tiny bronchus, the number of times of expansion is less, as 2 times; Expansion number of times is then increased for thicker bronchus.In order to extract the bronchus of different radii, can carry out above-mentioned expansive working, etching operation by different b values, first time compare operation and second time compare operation, obtain candidate's bronchus.In the present embodiment, M is 10 to the maximum, and preferred computing formula is as follows:
In one embodiment, the image I (i.e. lung and bronchus overall area) after greyscale transformation is carried out to repeatedly expansive working and the repeatedly etching operation of 4 neighborhoods, described 4 neighborhoods refer to four neighbors up and down of a pixel.
Suppose that the gray-scale value of the partial pixel point in image I is as follows:
-800 -780 -960 -920
0 0 0 0
-650 -920 -790 -800 -1020
0 0 0
-650 -920 -790 -800 -1020 -950
0 0
In above numerical value, there is the gray-scale value of the numeric representation blood vessel of underscore, there is no the bronchial gray-scale value of the numeric representation of underscore, after an expansive working, obtain result as follows:
-650 -780 -780 0 0 0 0 0
-650 -650 -790 -790 0 0 0 0
-650 -650 -790 -790 -800 0 0 0
Mainly comprise the lobe of the lung in lung CT data, pulmonary vascular and bronchus, gray level image expands and refers to get the maximum numerical value of gray-scale value in 4 neighborhoods here.
Such as
Maximum number is-650, so the value after expanding is-650.After above-mentioned expansive working, can see and to improve at the bronchus gray-scale value of blood vessels adjacent.
Described second segmentation module 30, also for choosing the second Seed Points in described candidate's bronchus, and Second Threshold scope is set, calculate with the second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of Second Threshold, as final bronchus carry out segmentation extraction.Consider in described candidate's bronchus and contain more impurity, 3D region growth algorithm can be adopted in conjunction with candidate's bronchus, result is revised.Second Seed Points is chosen in candidate's bronchus, because bronchial extracted region out, the threshold range that one very desirable can be selected, Second Threshold scope grown in conjunction with the condition of the bronchial result of candidate as region growing, the result of region growing is bronchial segmentation result.
Fig. 3 is the process flow diagram of the bronchial dividing method of the present invention, and it realizes by bronchial segmenting system 100 as shown in Figure 1 or 2, comprises step as follows:
Step S301, the first segmentation step: split from original lung images by 3D region growth algorithm and extract lung and bronchus overall area, described lung and bronchus overall area comprise lung tissue and bronchus.Because the gray-scale value of lung tissue and intrabronchial air is starkly lower than surrounding tissue, so 3D region growth algorithm can be adopted to carry out lung tissue and bronchial extraction.3D region growth algorithm to need in lung tissue or bronchus automatically or interactive mode chooses first Seed Points, suitable first threshold scope is set, as [-2000,-250], calculate and the first Seed Points being connected within the scope of spatial neighborhood and all pixels of tonal range within the scope of first threshold, be described lung and bronchus overall area.Described original lung images can be lung CT image or lung MRI image.
Step S302, bronchus strengthens step: carry out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area.Because lung tissue and bronchus all belong to density regions, when the difficult point of bronchi segmentation is segmentation bronchus, be easily subject to the interference of lung tissue.The blood of pulmonary alveoli and capillary has gas exchanges, carbon dioxide in blood is transported to external by bronchus, bronchus enters blood from the oxygen of external conveying, bronchus and blood vessel grow together, and the dilation and erosion computing of the Two-dimensional morphology that can adopt is to realize bronchial enhancing.Because bronchial gray-scale value is usually less than-750, the gray-scale value of blood vessel higher than 0, so be easy to bronchus to be separated with blood vessel.For the gray level image of lung bronchiolar region, bronchus is strengthened by the maximal value that repeatedly expansive working calculates in contiguous range, and the impact of lung tissue is removed by the repeatedly etching operation minimum value calculated in contiguous range, like this can by all bronchi segmentation of accompanying with blood vessel out.
Step S303, the second segmentation step: segmentation extracts bronchus from repeatedly expansive working and the lung repeatedly after etching operation and bronchus overall area.
The present invention adopts simple and quick method to realize bronchi segmentation, segmentation result can reach segmental bronchi and more than, to meet in medical science bronchial requirement, there is higher using value.Be more preferably, the present invention can be generalized to the extraction adopting similar scheme to realize tubular structure from medical image.
Fig. 4 is the process flow diagram of the preferred bronchial dividing method of the present invention, and it realizes by bronchial segmenting system 100 as shown in Figure 2, comprises step as follows:
Step S401, is split by 3D region growth algorithm and extracts lung and bronchus overall area from original lung images.Specifically, lung in original lung images or bronchus choose the first Seed Points, and first threshold scope is set, calculate with the first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of first threshold, as lung and bronchus overall area carry out segmentation extraction.
Step S402, is set to 0 by the gray-scale value that gray-scale value in lung and bronchus overall area is greater than the pixel of 0, obtains the lung after greyscale transformation and bronchus overall area.Such as in lung CT image, lung and bronchial tonal range are all less than 0, so gray-scale value gray-scale value in lung CT image being greater than the pixel of 0 is set to 0.The interference of high-density region can be eliminated like this.
Step S403, to the repeatedly expansive working described lung after greyscale transformation and bronchus overall area being carried out to some neighborhoods, and carries out the repeatedly etching operation of some neighborhoods to the data after repeatedly expansive working.Some neighborhoods are four neighborhoods or eight neighborhood, and preferred computing formula is as follows:
Step S404, the gray-scale value of each pixel in lung after the gray-scale value of each pixel in repeatedly expansive working and the lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, choose larger gray-scale value as gray count end value, preferred computing formula is as follows:
Formula two represents that the image I after to greyscale transformation (i.e. lung and bronchus overall area) carries out the dilation operation of b time and the erosion operation of b time, gets the value larger with image I.In order to accelerate computing velocity, dilation and erosion operation can be carried out in the partition data calculating lung and bronchus overall area.
Step S405, carries out second time compare operation by the gray-scale value of the lung after gray count end value and greyscale transformation and bronchus overall area, if be greater than predetermined threshold value T, then and alternatively bronchus.Often carry out once first time compare operation and will carry out second compare operation, the number of times of visible second time compare operation is identical with the number of times of first time compare operation, and preferred computing formula is as follows:
Step S406, extracts bronchus by the segmentation from repeatedly expansive working and the lung repeatedly after etching operation and bronchus overall area of 3D region growth algorithm.Specifically, the second Seed Points is chosen in candidate's bronchus, and Second Threshold scope is set, calculate and the second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of Second Threshold, carry out segmentation as final bronchus to extract, the bronchus that segmentation extracts as shown in Figure 5.Consider in described candidate's bronchus and contain more impurity, 3D region growth algorithm can be adopted in conjunction with candidate's bronchus, result is revised.Second Seed Points is chosen in candidate's bronchus, because bronchial extracted region out, the threshold range that one very desirable can be selected, Second Threshold scope grown in conjunction with the condition of the bronchial result of candidate as region growing, the result of region growing is bronchial segmentation result.
Preferably, described bronchus strengthens module 20 and carries out described expansive working, described etching operation, described first time compare operation and described second time compare operation for adopting different number of operations to the bronchus of different radii, and the thicker then number of operations of bronchus is more.Such as, for tiny bronchus, the number of times of expansion is less, as 2 times; Expansion number of times is then increased for thicker bronchus.In order to extract the bronchus of different radii, can carry out above-mentioned expansive working, etching operation by different b values, first time compare operation and second time compare operation, obtain candidate's bronchus.In the present embodiment, M is 10 to the maximum, and preferred computing formula is as follows:
In sum, the present invention considers because lung tissue and bronchus all belong to density regions, is easily subject to the interference of lung tissue when therefore the difficult point of bronchi segmentation is segmentation bronchus.Therefore, the present invention is according to the relevance of bronchus and blood vessel, the i.e. characteristic that always grows together of bronchus and blood vessel, bronchus is strengthened by the information by blood vessel, can avoid the impact of gray-value variation like this, bronchial segmentation result is more stable and accurate, and can split segmental bronchi and above branch, meet the requirement to bronchi segmentation in clinical medicine, there is higher clinical value.
Certainly; the present invention also can have other various embodiments; when not deviating from the present invention's spirit and essence thereof; those of ordinary skill in the art are when making various corresponding change and distortion according to the present invention, but these change accordingly and are out of shape the protection domain that all should belong to the claim appended by the present invention.
Claims (8)
1. a bronchial dividing method, is characterized in that, comprises step as follows:
First segmentation step, is split by 3D region growth algorithm and extracts lung and bronchus overall area from original lung images;
Bronchus strengthens step, carries out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area;
Second segmentation step, from repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area, segmentation extracts bronchus;
Described first segmentation step comprises further:
Lung in described original lung images or bronchus choose the first Seed Points, and first threshold scope is set, calculate with described first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described first threshold, as described lung and bronchus overall area carry out segmentation extraction;
Described bronchus strengthens step and comprises further:
The gray-scale value that gray-scale value in described lung and bronchus overall area is greater than the pixel of 0 is set to 0, obtains the described lung after greyscale transformation and bronchus overall area;
Described lung after greyscale transformation and bronchus overall area are carried out to the repeatedly expansive working of some neighborhoods, and the repeatedly etching operation of some neighborhoods is carried out to the data after repeatedly expansive working;
The gray-scale value of each pixel in described lung after the gray-scale value of each pixel in repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, chooses larger gray-scale value as gray count end value;
The gray-scale value of the described lung after described gray count end value and greyscale transformation and bronchus overall area is carried out second time compare operation, if be greater than predetermined threshold value, then alternatively bronchus.
2. bronchial dividing method according to claim 1, it is characterized in that, described bronchus strengthens in step, adopt different number of operations to carry out described expansive working, described etching operation, described first time compare operation and described second time compare operation to the bronchus of different radii, and the thicker then described number of operations of described bronchus is more.
3. bronchial dividing method according to claim 1, is characterized in that, described some neighborhoods are four neighborhoods or eight neighborhood.
4. bronchial dividing method according to claim 1, is characterized in that, described second segmentation step comprises further:
The second Seed Points is chosen in described candidate's bronchus, and Second Threshold scope is set, calculate with described second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described Second Threshold, as final bronchus carry out segmentation extraction.
5. a bronchial segmenting system, is characterized in that, includes:
First segmentation module, extracts lung and bronchus overall area for being split from original lung images by 3D region growth algorithm;
Bronchus strengthens module, for carrying out repeatedly expansive working and repeatedly etching operation to described lung and bronchus overall area;
Second segmentation module, extracts bronchus for splitting from repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area;
Described first segmentation module also chooses the first Seed Points in the lung in described original lung images or bronchus, and first threshold scope is set, calculate with described first Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described first threshold, as described lung and bronchus overall area carry out segmentation extraction;
Described bronchus strengthens module and comprises further:
Greyscale transformation submodule, is set to 0 for gray-scale value gray-scale value in described lung and bronchus overall area being greater than the pixel of 0, obtains the described lung after greyscale transformation and bronchus overall area;
Expansion/etching operation submodule, for carrying out the repeatedly expansive working of some neighborhoods to the described lung after greyscale transformation and the bronchiolar region in bronchus overall area, and carries out the repeatedly etching operation of some neighborhoods to the data after repeatedly expansive working;
First comparison sub-module, for the gray-scale value of each pixel in the described lung after the gray-scale value of each pixel in repeatedly expansive working and the described lung repeatedly after etching operation and bronchus overall area and greyscale transformation and bronchus overall area is carried out first time compare operation, choose larger gray-scale value as gray count end value;
Second comparison sub-module, for the gray-scale value of the described lung after described gray count end value and greyscale transformation and bronchus overall area is carried out second time compare operation, if be greater than predetermined threshold value, then alternatively bronchus.
6. bronchial segmenting system according to claim 5, it is characterized in that, described bronchus strengthens module and is used for adopting different number of operations to carry out described expansive working, described etching operation, first time compare operation and second time compare operation to the bronchus of different radii, and the thicker then described number of operations of described bronchus is more.
7. bronchial segmenting system according to claim 5, is characterized in that, described some neighborhoods are four neighborhoods or eight neighborhood.
8. bronchial segmenting system according to claim 5, it is characterized in that, described second segmentation module also for choosing the second Seed Points in described candidate's bronchus, and Second Threshold scope is set, calculate with described second Seed Points be connected in contiguous range and all pixels of tonal range within the scope of described Second Threshold, as final bronchus carry out segmentation extraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210423958.2A CN102982531B (en) | 2012-10-30 | 2012-10-30 | Bronchial dividing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210423958.2A CN102982531B (en) | 2012-10-30 | 2012-10-30 | Bronchial dividing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102982531A CN102982531A (en) | 2013-03-20 |
| CN102982531B true CN102982531B (en) | 2015-09-16 |
Family
ID=47856473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210423958.2A Active CN102982531B (en) | 2012-10-30 | 2012-10-30 | Bronchial dividing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102982531B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| WO2016086744A1 (en) | 2014-12-02 | 2016-06-09 | Shanghai United Imaging Healthcare Co., Ltd. | A method and system for image processing |
| CN106097305B (en) * | 2016-05-31 | 2019-03-01 | 上海理工大学 | The intratracheal tree dividing method that two-pass region growing combining form is rebuild |
| CN108470331B (en) * | 2017-02-23 | 2021-12-21 | 富士通株式会社 | Image processing apparatus, image processing method, and program |
| US20190069953A1 (en) * | 2017-09-06 | 2019-03-07 | Biosense Webster (Israel) Ltd. | Method and apparatus for enhancement of bronchial airways representations using vascular morphology |
| CN109345517B (en) * | 2018-09-19 | 2022-11-25 | 上海联影医疗科技股份有限公司 | Image segmentation method and device, computer equipment and storage medium |
| CN109360219A (en) * | 2018-10-23 | 2019-02-19 | 东北大学 | A kind of augmented reality auxiliary operation method and system |
| CN116348908A (en) * | 2020-10-22 | 2023-06-27 | 柯惠有限合伙公司 | Video laryngoscope system and method for quantitative assessment of trachea |
| CN114419077B (en) * | 2022-03-31 | 2022-07-05 | 深圳市一图智能科技有限公司 | Automatic lung bronchus segmentation method and system based on CT image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1912927A (en) * | 2006-08-25 | 2007-02-14 | 西安理工大学 | Semi-automatic partition method of lung CT image focus |
| US7206462B1 (en) * | 2000-03-17 | 2007-04-17 | The General Hospital Corporation | Method and system for the detection, comparison and volumetric quantification of pulmonary nodules on medical computed tomography scans |
-
2012
- 2012-10-30 CN CN201210423958.2A patent/CN102982531B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7206462B1 (en) * | 2000-03-17 | 2007-04-17 | The General Hospital Corporation | Method and system for the detection, comparison and volumetric quantification of pulmonary nodules on medical computed tomography scans |
| CN1912927A (en) * | 2006-08-25 | 2007-02-14 | 西安理工大学 | Semi-automatic partition method of lung CT image focus |
Non-Patent Citations (3)
| Title |
|---|
| "基于CT图像的肺气管树3D分割方法的研究";李翠芳等;《中国医学物理杂志》;20110930;第28卷(第5期);第2868页至2871页左栏最后1行 * |
| "基于数学形态学的新方法在脑组织分割中的应用";郑小林等;《仪器仪表学报》;20100228;第31卷(第2期);第465页左栏第4段至第466页第4段，第3节 * |
| "基于数学膨胀和腐蚀的医学磁共振图像的边缘检测";刘亚洁;《生物医学工程与临床》;20120131;第16卷(第1期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102982531A (en) | 2013-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102982531B (en) | Bronchial dividing method and system | |
| US11783498B2 (en) | Feature-based registration method | |
| JP2007144177A (en) | Method and apparatus for semi-automatic segmentation technique for low-contrast tubular shaped objects | |
| CN106651895B (en) | Method and device for segmenting three-dimensional image | |
| AU759501B2 (en) | Virtual endoscopy with improved image segmentation and lesion detection | |
| EP2157905B1 (en) | A method for tracking 3d anatomical and pathological changes in tubular-shaped anatomical structures | |
| CN111798451B (en) | 3D guide wire tracking method and device based on blood vessel 3D/2D matching | |
| CN107067398B (en) | Completion method and device for missing blood vessels in three-dimensional medical model | |
| CN106355582B (en) | A method of the liver medical image segmentation based on shape prior | |
| US20120249546A1 (en) | Methods and systems for visualization and analysis of sublobar regions of the lung | |
| US20130113802A1 (en) | Methods and systems for visualization of 3d parametric data during 2d imaging | |
| US9619938B2 (en) | Virtual endoscopic image display apparatus, method and program | |
| CN102693540B (en) | A kind of method of liver subsection and system thereof | |
| US10524823B2 (en) | Surgery assistance apparatus, method and program | |
| CN106600609A (en) | Spine partition method and system in medical image | |
| CN110432986B (en) | System and method for constructing virtual radial ultrasound images from CT data | |
| EP1761897A1 (en) | Displaying a tracheobronchial tree | |
| CN105139377A (en) | Rapid robustness auto-partitioning method for abdomen computed tomography (CT) sequence image of liver | |
| Chen et al. | Liver segmentation from CT images based on region growing method | |
| US9558589B2 (en) | Medical image display apparatus, method, and program | |
| KR101540402B1 (en) | Method for generating insertion trajectory of surgical needle | |
| CN117100393A (en) | Method, system and device for video-assisted surgical target positioning | |
| Miao et al. | The distance transform method in virtual endscope | |
| Du et al. | The three-dimensional reconstruction and visualization of direct intrahepatic portacaval shunt |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |