CN107909581A - Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images - Google Patents
Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images Download PDFInfo
- Publication number
- CN107909581A CN107909581A CN201711070729.6A CN201711070729A CN107909581A CN 107909581 A CN107909581 A CN 107909581A CN 201711070729 A CN201711070729 A CN 201711070729A CN 107909581 A CN107909581 A CN 107909581A
- Authority
- CN
- China
- Prior art keywords
- lung
- segmentation
- lobe
- section
- images
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present invention relates to a kind of lobe of the lung section dividing method of CT images, device, system, storage medium and equipment, this method to include：Detecting step, in CT images, detection output lung outlines, it includes intrapulmonary region, lung exterior domain；Screening step, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and as candidate region；Segmentation step, in the 3D aspects of candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；Constitution step, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；Integration step, splits segmentation result by vascular tree and lung and combines, and carries out lobe of the lung section segmentation, obtains the final segmentation result of candidate region.Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images provided by the invention based on deep learning, effectively reduce error, improve diagnosis and accuracy rate, and do not limited be subject to individual lung's morphological differences.
Description
Technical field
The present invention relates to image segmentation field, more particularly to a kind of lobe of the lung section dividing method of CT images, device, system,
Storage medium and equipment.
Background technology
At present, lung cancer is the highest cancer of the death rate in all cancers.Lung neoplasm is the Findings form of lung cancer, in CT
The shade of increase in density is shown as in iconography, it is right by the detection and segmentation to Lung neoplasm in CT images middle lobe and lung section
The early screening of lung cancer and assessment are of great significance.In existing detection, 3D CT (Emmenlauer, M., etc.:“free,
fast and reliable stitching of large 3d datasets.”J Microscopy,233,no.1,
Pp.42-60,2009.) image data amount is big, and there are larger individual difference, is brought to the lobe of the lung and the cutting techniques of lung section
Certain difficulty.
It is different from the computer vision technique application of broad sense, in order to reduce answering for depth convolutional neural networks (DCNN) calculating
Miscellaneous degree, efficiently utilize limited training data, and the prior art is by the way of medical ground information is combined with deep learning
Split, for example, 1) being now with successively progressive partitioning scheme for the lobe of the lung and the best technology of lung section segmentation effect
(K.Hayashi,etc.:“Radiographic and CT appearances of the major fissures,”
Radiographics, vol.21, no.4, pp.861-874,2001.), first lung outlines are detected with positioning, is recycled deep
The method of degree study is split (S.Hu, etc. to the lobe of the lung:“Automatic lung segmentation for
accurate quantitation of volumetric X-ray CT images.”IEEE Trans.Med.Imag.,
Vol.20, no.6, pp.490-498, Jun.2001.), the point coordinates according to lung section is characterized to split lung section (E.M.van
Rikxoort,etc.:“Supervised enhancement filters:Application to fissure
detection in chest CT scans,”IEEE Trans.Med.Imag.,vol.27,no.1,pp.1–10,
Jan.2008.).This mode have ignored the lung mechanics difference between different patients, and testing result has uncertainty
(I.C.Sluimer,etc.:“Towards automated segmentation of the pathological lung in
CT,”IEEE Trans.Med.Imag.,vol.24,no.8,pp.1025–1038,Aug.2005.).2) by tracheae and arteries and veins
A washing deformation is established in the analysis of pipe, and it is preferable that this mode splits segmentation effect for visible lung, but for partly not
Visible point of column split effect poor (E.van Rikxoort, etc.:“Automatic segmentation of
pulmonary lobes robust against incomplete fissures.”IEEE Transactions on
Medical Imaging,vol.29,no.6,pp.1286–1296,2010.).Above method carries out in mass data
Experiment, segmentation effect are bad.
Chinese patent literature CN103035009A discloses a kind of reconstruction based on CT images Lung neoplasm edge and segmentation side
Method, the patent carry out spatial alternation first with the mapping mode to Gradient Features with rarefaction representation ability to image；Then
CT Evaluation of Image Segmentation systems are established, mainly marginal information is amplified, rebuild Lung neoplasm edge.
Although above-mentioned patent improves the accuracy rate of Lung neoplasm edge detection and segmentation in CT images, starting to figure
Before the edge of picture is detected, it is necessary to compared with the threshold value initially set, the ability only in the case where meeting condition
Start to detect, otherwise, it is necessary to be handled again image, this method use scope is limited be subject to individual difference, it is necessary to not
It is disconnected to be adjusted.
As the state-of-the-art technology LDCT (low-dose spiral CT) of screening lung cancer, LDCT dose of radiations are relatively low, more conventional CT agent
Amount reduces by 75%~90%, does not influence lung's imaging, the same sensitiveness possessed with CT, from American National lung in 2011 but
The Notes of Key Datas of cancer Screening tests (National Lung Screening Trial, NLST) LDCT can improve the early stage of lung cancer
Diagnosis and after effectively reducing lung cancer case fatality rate, every clinical research on screening lung cancer successively provides soul-stirring
As a result, applications of the LDCT in screening lung cancer is effectively promoted.Therefore, how fully to activate and utilize intimate silent storage
Chest LDCT big datas, one of the problem of being those skilled in the art's urgent need to resolve.
In conclusion how to improve the accuracy and precision of the detection and segmentation to Lung neoplasm in CT images, Yi Jiru
What makes full use of and activates the big data of CT images, is one of those skilled in the art's urgent problem to be solved.
The content of the invention
To solve the above-mentioned problems, the present invention propose a kind of CT images lobe of the lung section dividing method based on deep learning,
Device, system, storage medium and equipment.
To achieve these goals, first aspect present invention provides a kind of lobe of the lung section dividing method of CT images, the party
Method comprises the following steps：
Detecting step, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, lung exterior domain；
Screening step, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；
Segmentation step, in the 3D aspects of candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Constitution step, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Integration step, splits segmentation result by vascular tree and lung using lobe of the lung section partitioning algorithm and is combined, and carries out the lobe of the lung
Section segmentation, obtains the final segmentation result of candidate region.
Further, the FCN networks using multilayer convolutional layer as agent structure are used to be detected and export in detecting step.
Specifically, in segmentation step, blood vessel is carried out respectively to lung section and the lobe of the lung using 3D U-net and along two branch lines
Segmentation splits segmentation with lung.
Further, in constitution step, tracheae distribution situation is inferred according to three-dimensional vascular distribution, obtains bronchial tree.
Specifically, in integration step, lobe of the lung section partitioning algorithm is by inputting lobe of the lung section disaggregated model, by vascular tree and lung
Segmentation result integration is split, and exports final segmentation result.
Second aspect of the present invention provides a kind of lobe of the lung section segmenting device based on CT images, which includes：
Detection module, in CT images, detection output lung outlines, lung outlines to include intrapulmonary region, lung outskirt
Domain；
Screening module, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and by intrapulmonary area
Domain is as candidate region；
Split module, for being split in the 3D aspects of candidate region, while to lung section and lobe of the lung progress blood vessel segmentation with lung
Segmentation；
Constructing module, for according to vessel segmentation, by constructing vascular tree, obtaining the three-dimensional blood vessel point of lung
Cloth；
Module is integrated, is combined for vascular tree and lung to be split segmentation result using lobe of the lung section partitioning algorithm, and carry out
Lobe of the lung section is split, and obtains the final segmentation result of candidate region.
Further, used in detection module the FCN networks using multilayer convolutional layer as agent structure be detected with it is defeated
Go out.
Specifically, in module is split, blood vessel is carried out respectively to lung section and the lobe of the lung using 3D U-net and along two branch lines
Segmentation splits segmentation with lung.
Further, in constructing module, tracheae distribution situation is inferred according to three-dimensional vascular distribution, obtains bronchial tree.
Specifically, in module is integrated, lobe of the lung section partitioning algorithm is by inputting lobe of the lung section disaggregated model, by vascular tree and lung
Segmentation result integration is split, and exports final segmentation result.
Third aspect present invention provides a kind of system for the lobe of the lung section segmentation realized based on CT images, including second aspect
Any one of the lobe of the lung section segmenting device based on CT images.
Fourth aspect present invention provides a kind of non-volatile memory medium, which, which has, stores it
In instruction, when executed so that processor performs the lobe of the lung section dividing method based on CT images, and instruction includes：
Detection instruction, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, lung exterior domain；
Screening instruction, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；
Split order, in the 3D aspects of candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Construction instruction, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Instruction is integrated, vascular tree and lung are split by segmentation result using lobe of the lung section partitioning algorithm and are combined, and carries out the lobe of the lung
Section segmentation, obtains the final segmentation result of candidate region.
Fifth aspect present invention provides a kind of equipment, including memory and processor, memory storage have computer can
Execute instruction, processor are configured as execute instruction to implement the process that the lobe of the lung section based on CT images is split, and process includes：
Detecting step, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, lung exterior domain；
Screening step, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；
Segmentation step, in the 3D aspects of candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Constitution step, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Integration step, splits segmentation result by vascular tree and lung using lobe of the lung section partitioning algorithm and is combined, and carries out the lobe of the lung
Section segmentation, obtains the final segmentation result of candidate region.
A kind of CT images lobe of the lung section dividing method, device, system, storage medium and equipment provided by the invention, are adopted first
By the use of multilayer convolutional layer as the FCN networks of agent structure, the lung outlines in CT images are accurately detected and exported；Secondly screen
Candidate region simultaneously selects 3D U-net expansion to split segmentation to lung section and lobe of the lung progress blood vessel segmentation and lung, according to vessel segmentation
Vascular tree is constructed, and tracheae distribution situation is inferred by three-dimensional vascular distribution；Further in conjunction with vascular tree and lung split segmentation result into
Row lobe of the lung section is split, and finally obtains the segmentation result of candidate region, this method can efficiently control cutting procedure accuracy and
Speed, the segmentation result accuracy finally obtained is high, greatly reduces easily existing error in conventional segmentation mode, improves and examine
Disconnected rate, and do not limited be subject to individual lung's morphological differences, the present invention is proposed the big number of LDCT examinations first in international and national
Combined according to artificial intelligence technology, for solve the problems, such as lung section split automatically, mark with Lung neoplasm precise positioning.
Brief description of the drawings
Fig. 1 is a kind of lobe of the lung section dividing method flow chart based on CT images provided by the invention；
Fig. 2 is a kind of lobe of the lung section segmenting device module map based on CT images provided by the invention；
Fig. 3 is that a kind of lobe of the lung section based on CT images provided by the invention splits flow diagram.
Embodiment
Embodiments of the present invention are illustrated by particular specific embodiment below, those skilled in the art can be by this specification
Revealed content understands other advantages and effect of the present invention easily.Although description of the invention will combine preferred embodiment
Introduce together, but this feature for not representing the invention is only limitted to the embodiment.On the contrary, invented with reference to embodiment
The purpose of introduction is to be possible to the other selections or transformation extended to cover the claim of the present invention.In order to provide pair
The depth of the present invention understands, and many concrete details will be included in being described below.The present invention can not also use these details real
Apply.In addition, in order to avoid the emphasis of the chaotic or fuzzy present invention, some details will be omitted in the de-scription.
Deep learning is a new field in machine learning research, its motivation, which is to establish, simulates human brain is divided
The neutral net of study is analysed, the mechanism that it imitates human brain explains data, such as (" deep learning is so for image, sound and text
Fire, what it can do actually" be selected from《The deep learning world》, quote date 2016-05-03.).The present invention provides one kind
CT images lobe of the lung section dividing method, device, system, storage medium and equipment based on deep learning.
An embodiment of the present invention provides a kind of CT images lobe of the lung section dividing method, is walked as shown in Figure 1, this method includes detection
Rapid 1, screening step 2, segmentation step 3, constitution step 4, integration step 5.Specifically, in detecting step 1, CT images are inputted
Data, and the lung outlines in CT images are detected and exported, the lung outlines of output include intrapulmonary region, lung outskirt
Domain；In screening step 2, lung outlines are inputted, select the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；In segmentation step 3, blood vessel segmentation and lung are carried out to the lung section of candidate region and the lobe of the lung in 3D aspects and is split point
Cut, obtain blood vessel segmentation and lung splits segmentation result, specific form is a three-dimensional lattice, has corresponded to the three-dimensional position in CT images
Put, whether each element value is represented in dot matrix splits algorithm by blood vessel/lung and be determined as that blood vessel/lung is split；In constitution step 4, root
According to vessel segmentation, by constructing vascular tree, the three-dimensional vascular distribution of lung is obtained；In integration step 5, using the lobe of the lung
Vascular tree and lung are split segmentation result and are combined by section partitioning algorithm, and carry out lobe of the lung section segmentation, obtain the final of candidate region
Segmentation result.The present invention efficiently controls the accuracy and speed of segmentation using splitting layer by layer, improves the efficiency and not of segmentation
It is limited to the limitation of individual lung's morphological differences.
In recent years, based on deep learning image Segmentation Technology (Hariharan, B., Arbel_aez, P.,
Girshick,R.,Malik,J.:Hypercolumns for object segmentation and_ne-grained
localization.In:Proc.CVPR.pp.447-456 (2015)) yield unusually brilliant results in major international memory, general deep
Spend in convolutional neural networks (DCNN), several full articulamentums are usually added after convolutional network so that the feature of convolutional network
Vector (Io_e, S., Szegedy, the C. of regular length can be mapped to:Batch normalization:Accelerating
deep network training by reducing internal covariate shift.CoRR abs/
1502.03167 it is used (2015)), and finally.And full convolutional network such as FCN (Long, J., Shelhamer, E.,
Darrell,T.:Fully convolutional networks for semantic segmentation.In:
Proc.CVPR.pp.3431-3440 (2015)) or U-net (Ronneberger, O., Fischer, P., Brox, T.:U-
net:Convolutional networks for biomedical image segmentation.In:MICCAI.LNCS,
Vol.9351, pp.234-241.Springer (2015)), then image can be realized pixel scale classification (Milletari,
F.,Ahmadi,S.,etc.:Hough-cnn:Deep learning for segmentation of deep brain
Regions in MRI and ultrasound.CoRR abs/1601.07014 (2016)) inputs the image of arbitrary size
In FCN networks, by convolutional layer feature extraction and return the up-sampling of convolutional layer, final classification results can revert to artwork
Size, forms end-to-end (Tran, D., Bourdev, L.D., Fergus, R., Torresani, L., Paluri, a M.:
Deep end2end voxel2voxel prediction.CoRR abs/1511.06681 (2015)) network structure
(Seyedhosseini,M.,Sajjadi,M.,Tasdizen,T.:Image segmentation with cascaded
hierarchical models and logistic disjunctive normal networks.In:
Proc.ICCV.pp.2168-2175 (2013)), and image to be detected can be not only divided into prospect background by this classification
Two classes, and the classification of object can be provided automatically.U-net is also one of more typical image segmentation network, and in medicine
The image particularly lobe of the lung, the segmentation of lung section have very big advantage：U-net first can change into 3-D (Fedorov, A.,
Beichel,R.,et al.:3D slicer as an image computing platform for the
Quantitative imaging network.J.Magn Reson Imaging 30 (9), 1323-1341 (2012)) aspect
On image segmentation (Kleesiek, J., Urban, G., Hubert, A., Schwarz, D., Maier-Hein, K.,
Bendszus,M.,Biller,A.:Deep mri brain extraction:A 3d convolutional neural
Network for skull stripping.NeuroImage (2016)), at the same U-net presented in medical domain it is huge
Big achievement, such as the detection of kidney portion lesion segmentation, retinal vessel segmentation etc..Split using the image in deep learning field and calculated
Method, can greatly reduce error, improve diagnosis efficiency and accuracy.
Specifically, it will be understood by those skilled in the art that conventional segmentation mode is by Hu Shiying et al. (Hu
Shiying,Hoffman E A,Reinhardt J M.Automatic Lung segmentation for accurate
quantitation of volumetric X-ray CT images[J].IEEE Trans on Medical Imaging,
2001,20(6):490-498)) start one classical lung segmentation flow, i.e. gray level threshold segmentation, Connected degree analysis, a left side
Right lung separation, morphology closure intrapulmonary high density structures.The flow employs a kind of iteration threshold optimization method, then looks for most
Big two connected components, when adhesion of lung, are isolated by the method for continuous morphological erosion, and this kind of method is referred to as classical
Lung dividing method.And the present invention employs the full convolutional neural networks different from classical dividing method in detecting step 1
(Fully Convolutional Networks, FCN), with traditional convolutional neural networks (Convolutional Neural
Network, CNN) image segmentation method compare, FCN has obvious advantage：(1) the input figure of arbitrary size can be received
Picture, and without requiring all training images and test image that there is same size；(2) more efficiently because avoid by
In brought using block of pixels repetition storage and calculate convolution the problem of.Present invention employs ripe FCN networks to be divided
Cut, wherein agent structure of more convolutional layers as FCN networks, and traditional full linking layer of neutral net is replaced with into convolutional layer,
Guarantee whole network is full convolutional network structure.
In order to ensure that the resolution ratio for exporting result is consistent with input picture, by last several layers of turnup lamination by network
Output characteristic figure is extended, and is expanded to consistent with input picture.And it is of the invention by the following aspects, in lung CT shadow
As upper progress re -training or finely tune to lift the accuracy of segmentation：(1) cut using lung CT images XY directions (as shown in Figure 3)
Face marks out intrapulmonary region and is marked as training as training image；(2) amount of training data：Adopted at random from 10K magnitudes CT
Sample produces 100K magnitude CT tangent plane pictures；(3) accuracy is marked：Need to ensure training mark error within 3mm；(4) need
Specific aim ensures the data volume in specific CT regions (apex pulmonis, membrane, outside lung)；(5) needing to ensure specific CT types distribution, (patient is not
Same age bracket, illness, CT dosage etc.) so that lung outlines are efficiently exported, since human body distributed architecture similitude is higher, CT
Lung outlines in image are relatively clear, so this process of detecting step 1 can be very quick.
Then in screening step 2, the mode of machine segmentation is selected to filter out intrapulmonary region, machine segmentation is to be based on machine
The automatic lung segmentation of study, and using intrapulmonary region as candidate region, for the emphasis examination subsequently further split, effectively
Reduce noise, accelerate detection speed.Wherein machine learning is a multi-field cross discipline, be related to probability theory, statistics,
The multi-door subjects such as Approximation Theory, convextiry analysis, algorithm complex theory.Specialize in the study that the mankind were simulated or realized to computer how
Behavior, to obtain new knowledge or skills, reorganizes the existing structure of knowledge and is allowed to constantly improve the performance of itself.
Further, in segmentation step 3, using 3D U-net and along two branch lines to lung section and the lobe of the lung respectively into promoting circulation of blood
Pipe is split splits segmentation with lung.Two of which branch line is obtained by screening candidate region (intrapulmonary region), that is, candidate regions
The edge in domain.It will be understood by those skilled in the art that encoder gradually decreases the Spatial Dimension of pond layer, decoder is progressively repaired
The details and Spatial Dimension of object.Usually there is quick connection between encoder and decoder, therefore decoder can be helped more preferable
Repair the details of target in ground.U-Net is most common structure in this method.3D U-net are full convolution deep neural networks
One kind, is one kind realization for traditional FCN networks, top-down network and bottom-up network can be passed through great-jump-forward transmission
(Skip connection) connection is to ensure that final output layer can obtain that high-order is semantic and low order local grain letter at the same time at the same time
Breath.Blood vessel segmentation and lung split segmentation to be split using 3D U-net, both are using identical algorithm frame and network
Structure, difference lies in the mark semantic differential inputted when training, the former needs to mark out blood vessel, and the latter needs to mark out lung
Split.Lung section and the lobe of the lung are split by 3D U-net, obtain the distribution of lung section medium vessels and 5 independent space (i.e. lungs
Leaf) and membrane.
Specifically, in constitution step 4, on the basis of vessel segmentation, vascular tree is constructed, obtains lung intersegmental part three
Vascular distribution is tieed up, since 3D U-net are a 3D parted patterns (as shown in Figure 3), it is inputted as the segmentation knot of a 3d space
In fruit, i.e. 3d space each point whether be blood vessel confidence level.When vascular tree is constructed, first according to a fixed threshold
Segmentation result is carried out binaryzation by value, and is smoothed, and then obtains angiosomes most by flood fill algorithms
Big connected component, as the vascular tree constructed, the i.e. final vessel segmentation of output, the confidence level based on 3d space point
To carry out binary conversion treatment, where concrete decision is blood vessel.According to the anatomical configurations of intrapulmonary, in intrapulmonary Pulmonary Vascular and tracheae
Trend has strong consistency, often has near lung's bronchuses at different levels the blood vessel adjoint.But in lung CT imaging, Pulmonary Vascular
Visually more significantly, therefore algorithm is to obtain pulmonary branches tracheorrhaphy by detecting vascular tree, and according to the position approximation of blood vessel
Cloth, so as to extract the characteristic informations such as bronchial tree distribution and shape, the information integration for subsequent step.Wherein, flood
Fill algorithms are that points that several connections are extracted from a region distinguish from other adjacent areas and (or dye different face respectively
Color) classic algorithm because its thinking similar to flood from a regional diffusion to it is all can reach regions and gain the name.
Further, in integration step 5, lobe of the lung section partitioning algorithm is the disaggregated model based on neutral net.Algorithm is defeated
Enter to be characterized as the relative position of intrapulmonary specified point (as shown in figure 3, calculating phase of the region in XYZ directions based on lung segmentation result
Comparative example), the blood vessel and lung of the areas adjacent split result (being represented using 64 × 64 × 64 cubes), and closest lung is split
With vector (position+distance) information of blood vessel.Algorithm trains lobe of the lung section disaggregated model by learning above-mentioned input, so that by blood
Guan Shu and lung split segmentation result and are combined.Algorithm output is the position lobe of the lung and the classification results of lung section.The present invention can be accurate
The accuracy of ground control cutting procedure and speed, for example, it is desired to ensure amount of training data and mark accuracy, amount of training data needs
Ensure 100K magnitudes intrapulmonary point (lobe of the lung belonging to mark and lung section), and ensure to come from 10K magnitude CT data, and ensure CT classes
Type is distributed (patient's different age group, illness, CT dosage etc.)；Need to ensure mark precision more than 95%.Can be by using
The network structure (reducing convolutional layer depth and width) suitably simplified, and shared adjacent domain feature operate calculated to reduce together
Amount, so as to achieve the purpose that acceleration.
Wherein, it is distributed on CT types, first should be according to patient age before the lobe of the lung section segmentation of CT images is carried out
Section, illness, CT dosage (such as routine CT, LDCT) are classified；Secondly split for same CT types, can effectively be subtracted
Few error come by different CT type belt, while data statistics can be targetedly carried out, it is that diagnosis relevant disease (such as shifts
Knurl, celiothelioma, Lung neoplasm etc.) valuable reference frame is provided.
As shown in figure 3, the preferred embodiment of the present invention inputs CT images data first, using FCN networks in CT images
Lung outlines are detected and export, and the lung outlines of output include intrapulmonary region, lung exterior domain；Secondly machine segmentation is selected
Mode filters out intrapulmonary region as candidate region；Again, in 3D aspects, lung section and lung using 3D U-net to candidate region
Leaf carries out blood vessel segmentation and splits segmentation with lung respectively, respectively obtains blood in 5 independent spaces (i.e. the lobe of the lung) and membrane and lung section
Pipe is distributed, and the result being distributed according to lung section medium vessels constructs vascular tree；Finally combine blood vessel segmentation and split segmentation result simultaneously with lung
Lobe of the lung section partitioning algorithm carries out lobe of the lung section segmentation, and lobe of the lung section partitioning algorithm is the disaggregated model based on neutral net, and algorithm passes through
Learn above-mentioned input and train lobe of the lung section disaggregated model, be combined so that vascular tree and lung are split segmentation result.Algorithm exports
For the position lobe of the lung and the classification results of lung section, precisely and rapidly obtain having multiple cut zone (region 1 in such as Fig. 3,3,
4 etc.) result.
Therefore, the lobe of the lung section dividing method of a kind of CT images proposed by the present invention, using the form split layer by layer, and introduces
FCN and 3D U-net are split respectively, and the segmentation result accuracy finally obtained is high, greatly reduces in conventional segmentation mode
Easily existing error, improves diagnosis.
An embodiment of the present invention provides a kind of lobe of the lung section segmenting device of CT images based on deep learning, as shown in Fig. 2,
The device includes detection module 10, screening module 20, segmentation module 30, constructing module 40, integration module 50.
Detection module 10, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, outside lung
Region；
Screening module 20, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and by intrapulmonary
Region is as candidate region；
Split module 30, for carrying out blood vessel segmentation and lung in the 3D aspects of candidate region, while to the lobe of the lung and lung section
Split segmentation；
Constructing module 40, for according to vessel segmentation, by constructing vascular tree, obtaining the three-dimensional blood vessel point of lung
Cloth；
Module 50 is integrated, is combined, goes forward side by side for vascular tree and lung to be split segmentation result using lobe of the lung section partitioning algorithm
Row lobe of the lung section is split, and obtains the final segmentation result of candidate region.
Further, in detection module 10, the FCN networks for employing maturation are split, wherein more convolutional layer conducts
The agent structure of FCN networks, and traditional full linking layer of neutral net is replaced with into convolutional layer, guarantee whole network is full convolution
Network structure, efficiently and fast exports the lung outlines in CT images.
Then in screening step 2, the mode of machine segmentation is selected to filter out intrapulmonary region, and using intrapulmonary region as time
Favored area, for the emphasis examination subsequently further split.
Further, in module 30 is split, using 3D U-net and along two branch lines to the lung section and the lobe of the lung
The blood vessel segmentation is carried out respectively splits segmentation with the lung.That is in the 3D aspects of candidate region, 3D U-net are passed through
Lung section and the lobe of the lung are split, obtain the distribution of lung section medium vessels and 5 independent spaces (i.e. the lobe of the lung) and membrane.
Further, in constructing module 40, on the basis of vessel segmentation, vascular tree is constructed, obtains lung section
Interior three-dimensional vascular distribution.According to the anatomical configurations of intrapulmonary, there is strong consistency in the trend of intrapulmonary Pulmonary Vascular and tracheae, it is past
Nearby there is the blood vessel adjoint toward lung bronchuses at different levels.But in lung CT imaging, Pulmonary Vascular is visually more notable, because
This algorithm is to obtain the distribution of pulmonary branches tracheae by detecting vascular tree, and according to the position approximation of blood vessel.
Further, in module 50 is integrated, lobe of the lung section partitioning algorithm is the disaggregated model based on neutral net.Algorithm is defeated
Enter to be characterized as the relative position of intrapulmonary specified point (as shown in figure 3, calculating phase of the region in XYZ directions based on lung segmentation result
Comparative example), the blood vessel and lung of the areas adjacent split result (being represented using 64 × 64 × 64 cubes), and closest lung is split
With vector (position+distance) information of blood vessel.Algorithm trains lobe of the lung section disaggregated model by learning above-mentioned input, so that by blood
Guan Shu and lung split segmentation result and are combined.Algorithm output is the position lobe of the lung and the classification results of lung section.
Therefore, a kind of lobe of the lung section segmenting device based on CT images of the embodiment of the present invention, using the form split layer by layer,
And introduce FCN and 3D U-net and split respectively, the segmentation result accuracy finally obtained is high, greatly reduces conventional segmentation
Easily existing error, improves diagnosis in mode.
The embodiment of the present invention additionally provides a kind of system for the lobe of the lung section segmentation realized based on CT images, including foregoing implementation
The lobe of the lung section segmenting device based on CT images in mode.
The embodiment of the present invention provides a kind of non-volatile memory medium again, which, which has, stores it
In instruction, when the instruction is performed so that processor perform the lobe of the lung section dividing method based on CT images, the instruction bag
Include：
Detection instruction, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, lung exterior domain；
Screening instruction, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；
Split order, in the 3D aspects of candidate region, while carries out blood vessel segmentation to the lobe of the lung and lung section and splits segmentation with lung；
Construction instruction, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Instruction is integrated, vascular tree and lung are split by segmentation result using lobe of the lung section partitioning algorithm and are combined, and carries out the lobe of the lung
Section segmentation, obtains the final segmentation result of candidate region.
The embodiment of the present invention also provides a kind of equipment, including memory and processor, memory storage have computer can
Execute instruction, processor are configured as performing the instruction to implement the process that the lobe of the lung section based on CT images is split, the process bag
Include：
Detecting step, in CT images, detection output lung outlines, lung outlines include intrapulmonary region, lung exterior domain；
Screening step, in lung outlines, selects the mode of machine segmentation to filter out intrapulmonary region, and intrapulmonary region is made
For candidate region；
Segmentation step, in the 3D aspects of candidate region, while carries out blood vessel segmentation to the lobe of the lung and lung section and splits segmentation with lung；
Constitution step, according to vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Integration step, splits segmentation result by vascular tree and lung using lobe of the lung section partitioning algorithm and is combined, and carries out the lobe of the lung
Section segmentation, obtains the final segmentation result of candidate region.
In conclusion a kind of lobe of the lung section dividing method of CT images based on deep learning provided by the invention, device, being
System, storage medium and equipment, first using FCN network of the multilayer convolutional layer as agent structure, accurately detect and export CT
Lung outlines in image；Secondly screening candidate region and select 3D U-net expansion lung section and the lobe of the lung are carried out blood vessel segmentation and
Lung splits segmentation, and vascular tree is constructed according to vessel segmentation, and has vascular distribution to infer tracheae distribution situation；Further in conjunction with blood
Guan Shu and lung split segmentation result and carry out lobe of the lung section segmentation, finally obtain the segmentation result of candidate region, and this method can be controlled efficiently
The accuracy and speed of cutting procedure processed, the segmentation result accuracy finally obtained is high, greatly reduces in conventional segmentation mode
Easily existing error, improves diagnosis, and is not limited be subject to individual lung's morphological differences, and the present invention is first in international and national
Secondary proposition is combined LDCT examination big datas with artificial intelligence technology, is split automatically for solving lung section, is marked and Lung neoplasm essence
Quasi- orientation problem.
The above-described embodiments merely illustrate the principles and effects of the present invention, not for the limitation present invention.It is any ripe
Know the personage of this technology all can carry out modifications and changes under the spirit and scope without prejudice to the present invention to above-described embodiment.Cause
This, those of ordinary skill in the art is complete without departing from disclosed spirit and institute under technological thought such as
Into all equivalent modifications or change, should by the present invention claim be covered.
Claims (13)
1. the lobe of the lung section dividing method of a kind of CT images, it is characterised in that this method comprises the following steps：
Detecting step, in CT images, detection output lung outlines, the lung outlines include intrapulmonary region, lung exterior domain；
Screening step, in the lung outlines, selects the mode of machine segmentation to filter out the intrapulmonary region, and by the lung
Inner region is as candidate region；
Segmentation step, in the 3D aspects of the candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Constitution step, according to the vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Integration step, splits segmentation result by the vascular tree and the lung using lobe of the lung section partitioning algorithm and is combined, and carries out
Lobe of the lung section is split, and obtains the final segmentation result of the candidate region.
2. the lobe of the lung section dividing method of CT images according to claim 1, it is characterised in that adopted in the detecting step
It is detected and exports for the FCN networks of agent structure to multilayer convolutional layer.
3. the lobe of the lung section dividing method of CT images according to claim 1, it is characterised in that in the segmentation step,
The blood vessel segmentation and the lung is carried out respectively to the lung section and the lobe of the lung using 3D U-net and along two branch lines to split point
Cut.
4. the lobe of the lung section dividing method of CT images according to claim 1, it is characterised in that in the constitution step,
Tracheae distribution situation is inferred according to the three-dimensional vascular distribution, obtains bronchial tree.
5. the lobe of the lung section dividing method of CT images according to claim 1, it is characterised in that in the integration step,
The lobe of the lung section partitioning algorithm splits segmentation result by inputting lobe of the lung section disaggregated model, by the vascular tree and the lung to be integrated,
And export final segmentation result.
6. the lobe of the lung section segmenting device of a kind of CT images, it is characterised in that the device includes：
Detection module, in CT images, detection output lung outlines, the lung outlines to include intrapulmonary region, lung outskirt
Domain；
Screening module, in the lung outlines, selects the mode of machine segmentation to filter out the intrapulmonary region, and by institute
Intrapulmonary region is stated as candidate region；
Split module, for being split in the 3D aspects of the candidate region, while to lung section and lobe of the lung progress blood vessel segmentation with lung
Segmentation；
Constructing module, for according to the vessel segmentation, by constructing vascular tree, obtaining the three-dimensional blood vessel point of lung
Cloth；
Module is integrated, is combined for the vascular tree and the lung to be split segmentation result using lobe of the lung section partitioning algorithm, and
Lobe of the lung section segmentation is carried out, obtains the final segmentation result of the candidate region.
7. the lobe of the lung section segmenting device of CT images according to claim 6, it is characterised in that adopted in the detection module
It is detected and exports for the FCN networks of agent structure to multilayer convolutional layer.
8. the lobe of the lung section segmenting device of CT images according to claim 6, it is characterised in that in the segmentation module,
The blood vessel segmentation and the lung is carried out respectively to the lung section and the lobe of the lung using 3D U-net and along two branch lines to split point
Cut.
9. the lobe of the lung section segmenting device of CT images according to claim 6, it is characterised in that in the constructing module,
Tracheae distribution situation is inferred according to the three-dimensional vascular distribution, obtains bronchial tree.
10. the lobe of the lung section segmenting device of CT images according to claim 6, it is characterised in that in the integration module,
The lobe of the lung section partitioning algorithm splits segmentation result by inputting lobe of the lung section disaggregated model, by the vascular tree and the lung to be integrated,
And export final segmentation result.
11. a kind of system for the lobe of the lung section segmentation for realizing CT images, including claim 6-10 any one of them CT images
Lobe of the lung section segmenting device.
12. a kind of non-volatile memory medium, which, which has, stores instruction therein, when described instruction quilt
During execution so that processor performs the lobe of the lung section dividing method of CT images, and described instruction includes：
Detection instruction, in CT images, detection output lung outlines, the lung outlines include intrapulmonary region, lung exterior domain；
Screening instruction, in the lung outlines, selects the mode of machine segmentation to filter out the intrapulmonary region, and by the lung
Inner region is as candidate region；
Split order, in the 3D aspects of the candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Construction instruction, according to the vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Instruction is integrated, the vascular tree and the lung are split by segmentation result using lobe of the lung section partitioning algorithm and are combined, and is carried out
Lobe of the lung section is split, and obtains the final segmentation result of the candidate region.
13. a kind of equipment, including memory and processor, the memory storage has computer executable instructions, the processing
Device is configured as performing described instruction to implement the process of the lobe of the lung section of CT images segmentation, and the process includes：
Detecting step, in CT images, detection output lung outlines, the lung outlines include intrapulmonary region, lung exterior domain；
Screening step, in the lung outlines, selects the mode of machine segmentation to filter out the intrapulmonary region, and by the lung
Inner region is as candidate region；
Segmentation step, in the 3D aspects of the candidate region, while carries out blood vessel segmentation to lung section and the lobe of the lung and splits segmentation with lung；
Constitution step, according to the vessel segmentation, by constructing vascular tree, obtains the three-dimensional vascular distribution of lung；
Integration step, splits segmentation result by the vascular tree and the lung using lobe of the lung section partitioning algorithm and is combined, and carries out
Lobe of the lung section is split, and obtains the final segmentation result of the candidate region.
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811505220.4A CN109615636B (en) | 2017-11-03 | 2017-11-03 | Blood vessel tree construction method and device in lung lobe segment segmentation of CT (computed tomography) image |
| CN201811506463.XA CN109636811B (en) | 2017-11-03 | 2017-11-03 | Integration method and device for lung lobe segment segmentation of CT (computed tomography) image |
| CN202010540968.9A CN111709953B (en) | 2017-11-03 | 2017-11-03 | Output method and device in lung lobe segment segmentation of CT (computed tomography) image |
| CN201711070729.6A CN107909581B (en) | 2017-11-03 | 2017-11-03 | Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images |
| CN201811505228.0A CN109584252B (en) | 2017-11-03 | 2017-11-03 | Lung lobe segment segmentation method and device of CT image based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711070729.6A CN107909581B (en) | 2017-11-03 | 2017-11-03 | Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images |
Related Child Applications (4)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811506463.XA Division CN109636811B (en) | 2017-11-03 | 2017-11-03 | Integration method and device for lung lobe segment segmentation of CT (computed tomography) image |
| CN202010540968.9A Division CN111709953B (en) | 2017-11-03 | 2017-11-03 | Output method and device in lung lobe segment segmentation of CT (computed tomography) image |
| CN201811505220.4A Division CN109615636B (en) | 2017-11-03 | 2017-11-03 | Blood vessel tree construction method and device in lung lobe segment segmentation of CT (computed tomography) image |
| CN201811505228.0A Division CN109584252B (en) | 2017-11-03 | 2017-11-03 | Lung lobe segment segmentation method and device of CT image based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107909581A true CN107909581A (en) | 2018-04-13 |
| CN107909581B CN107909581B (en) | 2019-01-29 |
Family
ID=61843485
Family Applications (5)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010540968.9A Active CN111709953B (en) | 2017-11-03 | 2017-11-03 | Output method and device in lung lobe segment segmentation of CT (computed tomography) image |
| CN201711070729.6A Active CN107909581B (en) | 2017-11-03 | 2017-11-03 | Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images |
| CN201811506463.XA Active CN109636811B (en) | 2017-11-03 | 2017-11-03 | Integration method and device for lung lobe segment segmentation of CT (computed tomography) image |
| CN201811505228.0A Active CN109584252B (en) | 2017-11-03 | 2017-11-03 | Lung lobe segment segmentation method and device of CT image based on deep learning |
| CN201811505220.4A Active CN109615636B (en) | 2017-11-03 | 2017-11-03 | Blood vessel tree construction method and device in lung lobe segment segmentation of CT (computed tomography) image |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010540968.9A Active CN111709953B (en) | 2017-11-03 | 2017-11-03 | Output method and device in lung lobe segment segmentation of CT (computed tomography) image |
Family Applications After (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811506463.XA Active CN109636811B (en) | 2017-11-03 | 2017-11-03 | Integration method and device for lung lobe segment segmentation of CT (computed tomography) image |
| CN201811505228.0A Active CN109584252B (en) | 2017-11-03 | 2017-11-03 | Lung lobe segment segmentation method and device of CT image based on deep learning |
| CN201811505220.4A Active CN109615636B (en) | 2017-11-03 | 2017-11-03 | Blood vessel tree construction method and device in lung lobe segment segmentation of CT (computed tomography) image |
Country Status (1)
| Country | Link |
|---|---|
| CN (5) | CN111709953B (en) |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765369A (en) * | 2018-04-20 | 2018-11-06 | 平安科技（深圳）有限公司 | Detection method, device, computer equipment and the storage medium of Lung neoplasm |
| CN108765422A (en) * | 2018-06-13 | 2018-11-06 | 云南大学 | A kind of retinal images blood vessel automatic division method |
| CN108764241A (en) * | 2018-04-20 | 2018-11-06 | 平安科技（深圳）有限公司 | Divide method, apparatus, computer equipment and the storage medium of near end of thighbone |
| CN109003299A (en) * | 2018-07-05 | 2018-12-14 | 北京推想科技有限公司 | A method of the calculating cerebral hemorrhage amount based on deep learning |
| CN109146854A (en) * | 2018-08-01 | 2019-01-04 | 东北大学 | A kind of analysis method of Lung neoplasm and pulmonary vascular association relationship |
| CN109255782A (en) * | 2018-09-03 | 2019-01-22 | 图兮深维医疗科技（苏州）有限公司 | A kind of processing method, device, equipment and the storage medium of Lung neoplasm image |
| CN109308695A (en) * | 2018-09-13 | 2019-02-05 | 镇江纳兰随思信息科技有限公司 | Based on the cancer cell identification method for improving U-net convolutional neural networks model |
| CN109345538A (en) * | 2018-08-30 | 2019-02-15 | 华南理工大学 | A kind of Segmentation Method of Retinal Blood Vessels based on convolutional neural networks |
| CN109410170A (en) * | 2018-09-14 | 2019-03-01 | 沈阳东软医疗系统有限公司 | Image processing method, device and equipment |
| CN109492519A (en) * | 2018-09-12 | 2019-03-19 | 浙江浙大列车智能化工程技术研究中心有限公司 | The generation method of training dataset mark for the identification of deep learning underground railway track |
| CN109685787A (en) * | 2018-12-21 | 2019-04-26 | 杭州依图医疗技术有限公司 | Output method, device in the lobe of the lung section segmentation of CT images |
| CN109727251A (en) * | 2018-12-29 | 2019-05-07 | 上海联影智能医疗科技有限公司 | The system that lung conditions are divided a kind of quantitatively, method and apparatus |
| CN109870730A (en) * | 2018-12-28 | 2019-06-11 | 中国科学院重庆绿色智能技术研究院 | A kind of method and system for X-ray machine image analytic degree test body regular inspection |
| CN109886967A (en) * | 2019-01-16 | 2019-06-14 | 成都蓝景信息技术有限公司 | Lung anatomy position location algorithms based on depth learning technology |
| CN109949300A (en) * | 2018-06-03 | 2019-06-28 | 北京昆仑医云科技有限公司 | Method, system and the computer-readable medium of anatomical tree structure analysis |
| CN109961446A (en) * | 2019-03-27 | 2019-07-02 | 深圳视见医疗科技有限公司 | CT/MR three-dimensional image segmentation processing method, device, equipment and medium |
| CN110084816A (en) * | 2019-03-21 | 2019-08-02 | 深圳大学 | Method for segmenting objects, device, computer readable storage medium and computer equipment |
| CN110197712A (en) * | 2019-06-05 | 2019-09-03 | 桂林电子科技大学 | A kind of medical image stocking system and storage method |
| WO2020001086A1 (en) * | 2018-06-28 | 2020-01-02 | 深圳先进技术研究院 | Deep learning-based cardiac ct image segmentation method and apparatus, device, and medium |
| CN110887707A (en) * | 2019-08-28 | 2020-03-17 | 江苏大学 | Grain collection device and grain impurity-containing crushing state real-time monitoring system and method based on U-Net network |
| CN111062955A (en) * | 2020-03-18 | 2020-04-24 | 天津精诊医疗科技有限公司 | Lung CT image data segmentation method and system |
| CN111260669A (en) * | 2020-02-17 | 2020-06-09 | 北京推想科技有限公司 | Lung lobe segmentation method and device based on CT image |
| CN111260671A (en) * | 2020-05-07 | 2020-06-09 | 北京精诊医疗科技有限公司 | Lung leaf segmentation method and system for CT image |
| CN111275722A (en) * | 2020-02-18 | 2020-06-12 | 广州柏视医疗科技有限公司 | Lung segment and liver segment segmentation method and system |
| CN111784700A (en) * | 2019-04-04 | 2020-10-16 | 阿里巴巴集团控股有限公司 | Lung lobe segmentation, model training, model construction and segmentation method, system and equipment |
| WO2020211293A1 (en) * | 2019-04-18 | 2020-10-22 | 北京市商汤科技开发有限公司 | Image segmentation method and apparatus, electronic device and storage medium |
| CN112017136A (en) * | 2020-08-06 | 2020-12-01 | 杭州深睿博联科技有限公司 | Lung CT image parameter reconstruction method, system, terminal and storage medium based on deep learning |
| CN113870167A (en) * | 2020-06-11 | 2021-12-31 | 通用电气精准医疗有限责任公司 | Method, system and storage medium for segmenting lung images |
| US11436720B2 (en) | 2018-12-28 | 2022-09-06 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for generating image metric |
| CN117808975A (en) * | 2024-02-27 | 2024-04-02 | 天津市肿瘤医院（天津医科大学肿瘤医院） | Deep learning-based three-dimensional reconstruction method for lung image surgery planning |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110101401B (en) * | 2019-04-18 | 2023-04-07 | 浙江大学山东工业技术研究院 | Liver contrast agent digital subtraction angiography method |
| US11424021B2 (en) | 2019-05-10 | 2022-08-23 | National Taiwan University | Medical image analyzing system and method thereof |
| TWI745940B (en) * | 2019-05-10 | 2021-11-11 | 國立臺灣大學 | Medical image analyzing system and method thereof |
| CN110136119A (en) * | 2019-05-16 | 2019-08-16 | 杭州健培科技有限公司 | A kind of lung based on deep learning splits the method and system of segmentation and integrity assessment |
| CN112330686A (en) * | 2019-08-05 | 2021-02-05 | 罗雄彪 | Method for segmenting and calibrating lung bronchus |
| CN111242874B (en) * | 2020-02-11 | 2023-08-29 | 北京百度网讯科技有限公司 | Image restoration method, device, electronic equipment and storage medium |
| CN111311583B (en) * | 2020-02-24 | 2021-03-12 | 广州柏视医疗科技有限公司 | Method for naming pulmonary trachea and blood vessel by sections |
| CN111539918B (en) * | 2020-04-15 | 2023-05-02 | 复旦大学附属肿瘤医院 | Ground glass pulmonary nodule risk layered prediction system based on deep learning |
| CN112164074B (en) * | 2020-09-22 | 2021-08-10 | 江南大学 | 3D CT bed fast segmentation method based on deep learning |
| CN112950554B (en) * | 2021-02-05 | 2021-12-21 | 慧影医疗科技（北京）有限公司 | Lung lobe segmentation optimization method and system based on lung segmentation |
| CN112489047B (en) * | 2021-02-05 | 2021-06-01 | 四川大学 | Deep learning-based pelvic bone and arterial vessel multi-level segmentation method thereof |
| CN113160186B (en) * | 2021-04-27 | 2022-10-25 | 青岛海信医疗设备股份有限公司 | Lung lobe segmentation method and related device |
| CN113222006B (en) * | 2021-05-08 | 2021-10-08 | 推想医疗科技股份有限公司 | Method, device, equipment and storage medium for grading segmental bronchus |
| CN113516669B (en) * | 2021-06-23 | 2023-04-25 | 湖北英库科技有限公司 | CT image-based trachea extraction method, device, equipment and storage medium |
| CN114092470B (en) * | 2021-12-08 | 2022-08-09 | 之江实验室 | Deep learning-based automatic detection method and device for pulmonary fissure |
| CN115147359B (en) * | 2022-06-06 | 2023-04-07 | 北京医准智能科技有限公司 | Lung lobe segmentation network model training method and device, electronic equipment and storage medium |
| CN117078698B (en) * | 2023-08-22 | 2024-03-05 | 山东第一医科大学第二附属医院 | Peripheral blood vessel image auxiliary segmentation method and system based on deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254097A (en) * | 2011-07-08 | 2011-11-23 | 普建涛 | Method for identifying fissure on lung CT (computed tomography) image |
| US20140105472A1 (en) * | 2012-10-11 | 2014-04-17 | Vida Diagnostics, Inc. | Visualization and characterization of pulmonary lobar fissures |
| CN106529555A (en) * | 2016-11-04 | 2017-03-22 | 四川大学 | DR (Digital Radiography) slice lung contour extraction method based on fully convolutional network |
| CN107230204A (en) * | 2017-05-24 | 2017-10-03 | 东北大学 | A kind of method and device that the lobe of the lung is extracted from chest CT image |
Family Cites Families (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2003216295A1 (en) * | 2002-02-15 | 2003-09-09 | The Regents Of The University Of Michigan | Lung nodule detection and classification |
| WO2008125910A2 (en) * | 2006-11-10 | 2008-10-23 | Superdimension, Ltd. | Adaptive navigation technique for navigating a catheter through a body channel or cavity |
| JP2008142481A (en) * | 2006-12-13 | 2008-06-26 | Med Solution Kk | Apparatus and program for carrying out segmentation of lungs to units of pulmonary segment automatically |
| CN101763644B (en) * | 2010-03-10 | 2011-11-30 | 华中科技大学 | Pulmonary nodule three-dimensional segmentation and feature extraction method and system thereof |
| JP6273291B2 (en) * | 2012-12-03 | 2018-01-31 | コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． | Image processing apparatus and method |
| CN103136788B (en) * | 2013-03-04 | 2016-01-20 | 重庆大学 | The visual method for reconstructing of a kind of three-dimensional blood vessel bifurcation |
| KR101472558B1 (en) * | 2013-10-04 | 2014-12-16 | 원광대학교산학협력단 | The system and method for automatic segmentation of lung, bronchus, pulmonary vessels images from thorax ct images |
| CN103886599B (en) * | 2014-03-26 | 2017-10-13 | 北京工业大学 | A kind of blood vessel ROI dividing methods based on ivus image |
| US20150305612A1 (en) * | 2014-04-23 | 2015-10-29 | Mark Hunter | Apparatuses and methods for registering a real-time image feed from an imaging device to a steerable catheter |
| CN104574413B (en) * | 2015-01-22 | 2017-04-12 | 深圳大学 | Blood vessel bifurcation extracting method and system of lung CT picture |
| CN104751178B (en) * | 2015-03-31 | 2018-04-03 | 上海理工大学 | Lung neoplasm detection means and method based on shape template matching combining classification device |
| US10078895B2 (en) * | 2015-12-30 | 2018-09-18 | Case Western Reserve University | Prediction of recurrence of non-small cell lung cancer with tumor infiltrating lymphocyte (TIL) graphs |
| 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 |
| CN106780460B (en) * | 2016-12-13 | 2019-11-08 | 杭州健培科技有限公司 | A kind of Lung neoplasm automatic checkout system for chest CT images |
| CN106875379A (en) * | 2017-01-10 | 2017-06-20 | 陕西渭南神州德信医学成像技术有限公司 | Lung splits integrity degree appraisal procedure, device and system |
| CN107016665B (en) * | 2017-02-16 | 2021-05-04 | 浙江大学 | CT pulmonary nodule detection method based on deep convolutional neural network |
| CN107203989A (en) * | 2017-04-01 | 2017-09-26 | 南京邮电大学 | End-to-end chest CT image dividing method based on full convolutional neural networks |
| CN107103187B (en) * | 2017-04-10 | 2020-12-29 | 四川省肿瘤医院 | Lung nodule detection grading and management method and system based on deep learning |
-
2017
- 2017-11-03 CN CN202010540968.9A patent/CN111709953B/en active Active
- 2017-11-03 CN CN201711070729.6A patent/CN107909581B/en active Active
- 2017-11-03 CN CN201811506463.XA patent/CN109636811B/en active Active
- 2017-11-03 CN CN201811505228.0A patent/CN109584252B/en active Active
- 2017-11-03 CN CN201811505220.4A patent/CN109615636B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254097A (en) * | 2011-07-08 | 2011-11-23 | 普建涛 | Method for identifying fissure on lung CT (computed tomography) image |
| US20140105472A1 (en) * | 2012-10-11 | 2014-04-17 | Vida Diagnostics, Inc. | Visualization and characterization of pulmonary lobar fissures |
| CN106529555A (en) * | 2016-11-04 | 2017-03-22 | 四川大学 | DR (Digital Radiography) slice lung contour extraction method based on fully convolutional network |
| CN107230204A (en) * | 2017-05-24 | 2017-10-03 | 东北大学 | A kind of method and device that the lobe of the lung is extracted from chest CT image |
Non-Patent Citations (3)
| Title |
|---|
| BIANCA LASSEN等: "Lung and Lung Lobe Segmentation Methods by Fraunhofer MEVIS", 《HTTP://WWW.MEVIS.FRAUNHOFER.DE》 * |
| ZHONGLIU XIE等: "3D Region Proposal U-Net with Dense and Residual Learning for Lung Nodule Detection", 《HTTP://LUNA16.GRAND-CHALLEGE.ORG/.../20171011_032308_REPORT_XIE.PDF》 * |
| 彭双等: "结合线性形状特征与模糊连接度的三维CT图像肺裂检测", 《计算机应用与软件》 * |
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764241A (en) * | 2018-04-20 | 2018-11-06 | 平安科技（深圳）有限公司 | Divide method, apparatus, computer equipment and the storage medium of near end of thighbone |
| WO2019200740A1 (en) * | 2018-04-20 | 2019-10-24 | 平安科技（深圳）有限公司 | Pulmonary nodule detection method and apparatus, computer device, and storage medium |
| WO2019200747A1 (en) * | 2018-04-20 | 2019-10-24 | 平安科技（深圳）有限公司 | Method and device for segmenting proximal femur, computer apparatus, and storage medium |
| CN108765369A (en) * | 2018-04-20 | 2018-11-06 | 平安科技（深圳）有限公司 | Detection method, device, computer equipment and the storage medium of Lung neoplasm |
| CN109949300A (en) * | 2018-06-03 | 2019-06-28 | 北京昆仑医云科技有限公司 | Method, system and the computer-readable medium of anatomical tree structure analysis |
| CN108765422A (en) * | 2018-06-13 | 2018-11-06 | 云南大学 | A kind of retinal images blood vessel automatic division method |
| WO2020001086A1 (en) * | 2018-06-28 | 2020-01-02 | 深圳先进技术研究院 | Deep learning-based cardiac ct image segmentation method and apparatus, device, and medium |
| CN109003299A (en) * | 2018-07-05 | 2018-12-14 | 北京推想科技有限公司 | A method of the calculating cerebral hemorrhage amount based on deep learning |
| CN109146854A (en) * | 2018-08-01 | 2019-01-04 | 东北大学 | A kind of analysis method of Lung neoplasm and pulmonary vascular association relationship |
| CN109345538A (en) * | 2018-08-30 | 2019-02-15 | 华南理工大学 | A kind of Segmentation Method of Retinal Blood Vessels based on convolutional neural networks |
| CN109255782A (en) * | 2018-09-03 | 2019-01-22 | 图兮深维医疗科技（苏州）有限公司 | A kind of processing method, device, equipment and the storage medium of Lung neoplasm image |
| CN109492519A (en) * | 2018-09-12 | 2019-03-19 | 浙江浙大列车智能化工程技术研究中心有限公司 | The generation method of training dataset mark for the identification of deep learning underground railway track |
| CN109308695A (en) * | 2018-09-13 | 2019-02-05 | 镇江纳兰随思信息科技有限公司 | Based on the cancer cell identification method for improving U-net convolutional neural networks model |
| CN109410170A (en) * | 2018-09-14 | 2019-03-01 | 沈阳东软医疗系统有限公司 | Image processing method, device and equipment |
| CN109685787A (en) * | 2018-12-21 | 2019-04-26 | 杭州依图医疗技术有限公司 | Output method, device in the lobe of the lung section segmentation of CT images |
| US11436720B2 (en) | 2018-12-28 | 2022-09-06 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for generating image metric |
| CN109870730A (en) * | 2018-12-28 | 2019-06-11 | 中国科学院重庆绿色智能技术研究院 | A kind of method and system for X-ray machine image analytic degree test body regular inspection |
| CN109870730B (en) * | 2018-12-28 | 2020-11-20 | 中国科学院重庆绿色智能技术研究院 | Method and system for regular inspection of X-ray machine image resolution test body |
| CN109727251A (en) * | 2018-12-29 | 2019-05-07 | 上海联影智能医疗科技有限公司 | The system that lung conditions are divided a kind of quantitatively, method and apparatus |
| CN109886967A (en) * | 2019-01-16 | 2019-06-14 | 成都蓝景信息技术有限公司 | Lung anatomy position location algorithms based on depth learning technology |
| CN110084816A (en) * | 2019-03-21 | 2019-08-02 | 深圳大学 | Method for segmenting objects, device, computer readable storage medium and computer equipment |
| CN110084816B (en) * | 2019-03-21 | 2021-04-06 | 深圳大学 | Object segmentation method, device, computer-readable storage medium and computer equipment |
| CN109961446A (en) * | 2019-03-27 | 2019-07-02 | 深圳视见医疗科技有限公司 | CT/MR three-dimensional image segmentation processing method, device, equipment and medium |
| CN109961446B (en) * | 2019-03-27 | 2021-06-01 | 深圳视见医疗科技有限公司 | CT/MR three-dimensional image segmentation processing method, device, equipment and medium |
| CN111784700A (en) * | 2019-04-04 | 2020-10-16 | 阿里巴巴集团控股有限公司 | Lung lobe segmentation, model training, model construction and segmentation method, system and equipment |
| CN111784700B (en) * | 2019-04-04 | 2022-07-22 | 阿里巴巴集团控股有限公司 | Lung lobe segmentation, model training, model construction and segmentation method, system and equipment |
| JP2022515722A (en) * | 2019-04-18 | 2022-02-22 | ベイジン センスタイム テクノロジー デベロップメント カンパニー， リミテッド | Image segmentation methods and devices, electronic devices and storage media |
| WO2020211293A1 (en) * | 2019-04-18 | 2020-10-22 | 北京市商汤科技开发有限公司 | Image segmentation method and apparatus, electronic device and storage medium |
| CN110197712B (en) * | 2019-06-05 | 2023-09-15 | 桂林电子科技大学 | Medical image storage system and storage method |
| CN110197712A (en) * | 2019-06-05 | 2019-09-03 | 桂林电子科技大学 | A kind of medical image stocking system and storage method |
| CN110887707B (en) * | 2019-08-28 | 2022-03-22 | 江苏大学 | Real-time monitoring system and method for impurity-containing crushing state of grains based on U-Net network |
| CN110887707A (en) * | 2019-08-28 | 2020-03-17 | 江苏大学 | Grain collection device and grain impurity-containing crushing state real-time monitoring system and method based on U-Net network |
| CN111260669A (en) * | 2020-02-17 | 2020-06-09 | 北京推想科技有限公司 | Lung lobe segmentation method and device based on CT image |
| CN111275722A (en) * | 2020-02-18 | 2020-06-12 | 广州柏视医疗科技有限公司 | Lung segment and liver segment segmentation method and system |
| CN111062955A (en) * | 2020-03-18 | 2020-04-24 | 天津精诊医疗科技有限公司 | Lung CT image data segmentation method and system |
| CN111260671A (en) * | 2020-05-07 | 2020-06-09 | 北京精诊医疗科技有限公司 | Lung leaf segmentation method and system for CT image |
| CN113870167A (en) * | 2020-06-11 | 2021-12-31 | 通用电气精准医疗有限责任公司 | Method, system and storage medium for segmenting lung images |
| CN112017136A (en) * | 2020-08-06 | 2020-12-01 | 杭州深睿博联科技有限公司 | Lung CT image parameter reconstruction method, system, terminal and storage medium based on deep learning |
| CN117808975A (en) * | 2024-02-27 | 2024-04-02 | 天津市肿瘤医院（天津医科大学肿瘤医院） | Deep learning-based three-dimensional reconstruction method for lung image surgery planning |
| CN117808975B (en) * | 2024-02-27 | 2024-05-03 | 天津市肿瘤医院（天津医科大学肿瘤医院） | Deep learning-based three-dimensional reconstruction method for lung image surgery planning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111709953A (en) | 2020-09-25 |
| CN107909581B (en) | 2019-01-29 |
| CN109636811B (en) | 2020-06-12 |
| CN109584252B (en) | 2020-08-14 |
| CN109636811A (en) | 2019-04-16 |
| CN109615636A (en) | 2019-04-12 |
| CN109584252A (en) | 2019-04-05 |
| CN111709953B (en) | 2023-04-07 |
| CN109615636B (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107909581B (en) | Lobe of the lung section dividing method, device, system, storage medium and the equipment of CT images | |
| Su et al. | Lung nodule detection based on faster R-CNN framework | |
| CN109685787A (en) | Output method, device in the lobe of the lung section segmentation of CT images | |
| CN109035255B (en) | Method for segmenting aorta with interlayer in CT image based on convolutional neural network | |
| Li et al. | Automatic cardiothoracic ratio calculation with deep learning | |
| Zheng et al. | Automatic liver tumor segmentation on dynamic contrast enhanced MRI using 4D information: deep learning model based on 3D convolution and convolutional LSTM | |
| Tian et al. | Multi-path convolutional neural network in fundus segmentation of blood vessels | |
| CN108257135A (en) | The assistant diagnosis system of medical image features is understood based on deep learning method | |
| Wen et al. | A novel statistical cerebrovascular segmentation algorithm with particle swarm optimization | |
| CN109345527A (en) | A kind of tumor of bladder detection method based on MaskRcnn | |
| Mu et al. | Segmentation of kidney tumor by multi-resolution VB-nets | |
| CN110910405A (en) | Brain tumor segmentation method and system based on multi-scale cavity convolutional neural network | |
| Ye et al. | Medical image diagnosis of prostate tumor based on PSP-Net+ VGG16 deep learning network | |
| Zhang et al. | Cerebrovascular segmentation from TOF-MRA using model-and data-driven method via sparse labels | |
| Wang et al. | CLCU-Net: Cross-level connected U-shaped network with selective feature aggregation attention module for brain tumor segmentation | |
| CN111179237A (en) | Image segmentation method and device for liver and liver tumor | |
| CN107330953A (en) | A kind of Dynamic MRI method for reconstructing based on non-convex low-rank | |
| Tan et al. | An approach for pulmonary vascular extraction from chest CT images | |
| CN113420826A (en) | Liver focus image processing system and image processing method | |
| CN111784701A (en) | Ultrasonic image segmentation method and system combining boundary feature enhancement and multi-scale information | |
| Qian et al. | Multi-scale context UNet-like network with redesigned skip connections for medical image segmentation | |
| CN113935976A (en) | Method and system for automatically segmenting blood vessels in internal organs by enhancing CT (computed tomography) image | |
| Affane et al. | Literature review of deep learning models for liver vessels reconstruction | |
| Affane et al. | Robust deep 3-D architectures based on vascular patterns for liver vessel segmentation | |
| Xiaojie et al. | Segmentation of the aortic dissection from CT images based on spatial continuity prior model |
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 |