CN112561917A - Image segmentation method and system, electronic device and readable storage medium - Google Patents
Image segmentation method and system, electronic device and readable storage medium Download PDFInfo
- Publication number
- CN112561917A CN112561917A CN202011529902.6A CN202011529902A CN112561917A CN 112561917 A CN112561917 A CN 112561917A CN 202011529902 A CN202011529902 A CN 202011529902A CN 112561917 A CN112561917 A CN 112561917A
- Authority
- CN
- China
- Prior art keywords
- image
- data
- blood vessel
- organ
- identified
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 94
- 238000003709 image segmentation Methods 0.000 title description 27
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 167
- 210000000056 organ Anatomy 0.000 claims abstract description 131
- 210000004072 lung Anatomy 0.000 claims abstract description 122
- 230000011218 segmentation Effects 0.000 claims abstract description 108
- 210000003462 vein Anatomy 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 10
- 210000003437 trachea Anatomy 0.000 abstract description 5
- 210000004185 liver Anatomy 0.000 description 23
- 238000004422 calculation algorithm Methods 0.000 description 20
- 210000005161 hepatic lobe Anatomy 0.000 description 20
- 238000002591 computed tomography Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 12
- 210000002989 hepatic vein Anatomy 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 9
- 230000002685 pulmonary effect Effects 0.000 description 9
- 238000013135 deep learning Methods 0.000 description 8
- 238000003062 neural network model Methods 0.000 description 8
- 210000003492 pulmonary vein Anatomy 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000002271 resection Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000002440 hepatic effect Effects 0.000 description 4
- 210000003041 ligament Anatomy 0.000 description 4
- 210000004224 pleura Anatomy 0.000 description 4
- 238000002604 ultrasonography Methods 0.000 description 4
- 210000000621 bronchi Anatomy 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000012879 PET imaging Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical group [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 238000013170 computed tomography imaging Methods 0.000 description 2
- 238000002059 diagnostic imaging Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000000232 gallbladder Anatomy 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004199 lung function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 230000035790 physiological processes and functions Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012285 ultrasound imaging Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 208000007502 anemia Diseases 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002980 postoperative effect Effects 0.000 description 1
- 210000001147 pulmonary artery Anatomy 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
-
- 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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a segmentation method, a segmentation system, electronic equipment and a readable storage medium of an image, wherein the segmentation method comprises the following steps: acquiring an image to be identified containing a tubular structure organ; acquiring blood vessel data in the image to be identified; obtaining the interface of two adjacent organ sections of the organ according to the blood vessel data; and segmenting the image to be identified based on the boundary surface. The method starts from medical physiological characteristics, trachea and vein blood vessels accompany, veins run between organs containing tubular structures such as lung segments, end point data of the blood vessel segments running between two adjacent organ segments are obtained through blood vessel position data, a space curved surface is fitted according to the end point data and serves as a dividing plane of the two adjacent organ segments, and the intersegmental vein running is utilized to determine a segmentation plane, so that the method is more reliable, efficient and more consistent with medical theoretical standards.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an image segmentation method, an image segmentation system, electronic equipment and a readable storage medium.
Background
The lung focus resection is developed from anatomical lung lobe resection to lung segment resection, so that normal lung tissues can be kept to the maximum extent in a more precise mode, the postoperative survival rate of a patient with poor lung function can be improved, the lung segments are obtained by utilizing CT (computed tomography) examination image segmentation, the lung focus resection method has important significance for lung function analysis, automatic assessment of lung lesion and operation planning, the lung segments are positioned by utilizing medical image CT examination before an operation, and a more precise auxiliary strategy can be provided for lung resection operations by utilizing image navigation in the operation.
In the prior art, manual labeling is mostly adopted in combination with a machine learning algorithm to train historical data to obtain a segmentation model so as to segment lung lobes and lung segments, manual lung segment positioning by using a CT image takes a long time, and parameters obtained by training are difficult to evaluate due to differences of lung structures of different patients, so that a more reliable and efficient segmentation method needs to be provided.
Disclosure of Invention
The invention aims to overcome the defects of long time consumption, low universality and low effectiveness of a lung segmentation method in the prior art, and provides an image segmentation method, an image segmentation system, electronic equipment and a readable storage medium.
The invention solves the technical problems through the following technical scheme:
a method of segmenting an image, the method comprising:
acquiring an image to be identified of an organ containing a tubular structure;
acquiring blood vessel data in the image to be identified;
obtaining the interface of two adjacent organ sections of the organ according to the blood vessel data;
and segmenting the image to be identified based on the boundary surface.
Preferably, said deriving an interface between two adjacent organ segments of said organ from said vessel data comprises:
obtaining a plurality of vessel segments and end point data of each vessel segment according to the vessel data;
and obtaining the interface according to the endpoint data.
Preferably, the obtaining the interface according to the endpoint data includes:
classifying the plurality of vessel segments;
acquiring target endpoint data of vessel segments of the same category;
and fitting according to the target endpoint data to obtain the interface.
Preferably, the classifying the plurality of vessel segments comprises:
and dividing the vessel segments running between two adjacent organ segments into the same category.
Preferably, the segmenting the image to be recognized based on the boundary surface comprises:
randomly selecting a target interface;
and inputting each voxel point in the image to be recognized into the curved function of the target interface so as to determine the organ segment of each voxel point.
Preferably, the organ containing a tubular structure comprises a lung, and after the image to be identified of the organ containing a tubular structure is acquired, the segmentation method further comprises:
identifying lung lobe fissure characteristic data in the image to be identified;
segmenting the image to be identified based on the lung lobe fissure characteristic data to obtain a plurality of lung lobe images;
and in the acquisition of the blood vessel data of the image to be identified, each lung lobe image is respectively acquired.
Preferably, the acquiring the blood vessel data in the image to be identified includes:
dividing veins from the image to be identified, and acquiring vein position data to generate the blood vessel data;
and/or, the acquiring of the blood vessel data in the image to be identified comprises:
and acquiring bifurcation point data of the blood vessel in the image to be identified to generate the blood vessel data.
A segmentation system for an image, the segmentation system comprising:
the image acquisition module is used for acquiring an image to be identified of an organ containing a tubular structure;
the blood vessel data acquisition module is used for acquiring blood vessel data in the image to be identified;
the interface generation module is used for obtaining interfaces of two adjacent organ sections of the organ according to the blood vessel data;
and the segmentation module is used for segmenting the image to be identified based on the boundary surface.
Preferably, the interface generation module comprises an endpoint data acquisition unit;
the end point data acquisition unit is used for acquiring a plurality of blood vessel segments and end point data of each blood vessel segment according to the blood vessel data;
and the interface generation module is used for obtaining the interface according to the endpoint data.
Preferably, the interface generation module further comprises a classification unit;
the classification unit is used for classifying the plurality of blood vessel segments;
the interface generation module is used for acquiring target endpoint data of vessel segments of the same category and obtaining the interface according to the target endpoint data fitting.
Preferably, the classification unit is configured to classify vessel segments running between two adjacent organ segments into the same category.
Preferably, the segmentation module further comprises a selection unit and an organ segment determination unit;
the selection unit is used for randomly selecting a target interface;
the organ segment determining unit is used for inputting each voxel point in the image to be identified into the curved function of the target interface so as to determine the organ segment of each voxel point.
Preferably, the organ containing the tubular structure comprises a lung, and the segmentation system further comprises a lobar fissure data identification module and a lung lobe segmentation module;
the leaf interfissure data identification module is used for identifying lung leaf interfissure characteristic data in the image to be identified;
the lung lobe segmentation module is used for segmenting the image to be identified based on the lung lobe cleft characteristic data to obtain a plurality of lung lobe images;
the blood vessel data acquisition module is used for respectively acquiring each lung lobe image.
Preferably, the blood vessel data acquiring module is configured to segment a vein from the image to be identified, and acquire vein position data to generate the blood vessel data;
and/or the presence of a gas in the gas,
the blood vessel data acquisition module is used for acquiring bifurcation point data of the blood vessel in the image to be identified to generate the blood vessel data.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for segmenting an image as described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the steps of the method for segmenting an image as set forth above.
The positive progress effects of the invention are as follows: the method starts from medical physiological characteristics, trachea and vein blood vessels accompany, veins run between organs containing tubular structures such as lung segments, end point data of the blood vessel segments running between two adjacent organ segments are obtained through blood vessel position data, a space curved surface is fitted according to the end point data and serves as a dividing plane of the two adjacent organ segments, and the intersegmental vein running is utilized to determine a segmentation plane, so that the method is more reliable, efficient and more consistent with medical theoretical standards.
Drawings
Fig. 1 is a flowchart of an image segmentation method according to embodiment 1 of the present invention.
Fig. 2 is a flowchart of step 30 in the image segmentation method according to embodiment 1 of the present invention.
Fig. 3 is a flowchart of step 32 in the image segmentation method according to embodiment 1 of the present invention.
Fig. 4 is a flowchart of step 40 in the image segmentation method according to embodiment 1 of the present invention.
Fig. 5 is a flowchart of a segmentation method for a lung image according to embodiment 2 of the present invention.
Fig. 6 is a schematic structural diagram of a VB-Net neural network in the image segmentation method according to embodiment 2 of the present invention.
Fig. 7 is a schematic diagram of a lung lobe segmentation result in the image segmentation method according to embodiment 2 of the present invention.
Fig. 8a is a schematic diagram of segmenting the left pulmonary blood vessel in the image segmentation method according to embodiment 2 of the present invention.
Fig. 8b is a schematic diagram of right pulmonary vessel segmentation in the image segmentation method according to embodiment 2 of the present invention.
Fig. 9 is a schematic diagram of a segmentation result of a lung segment in the image segmentation method according to embodiment 2 of the present invention.
Fig. 10 is a schematic diagram illustrating a cut-out relationship between a left lung apex posterior segment plane and a left lung superior lobe in the image segmentation method according to embodiment 2 of the present invention.
Fig. 11 is a flowchart of a liver image segmentation method according to embodiment 2 of the present invention.
Fig. 12 is a block diagram of an image segmentation system according to embodiment 3 of the present invention.
Fig. 13 is a block diagram of an image segmentation system according to embodiment 4 of the present invention.
Fig. 14 is a schematic structural diagram of an electronic device according to embodiment 5 of the present invention.
Detailed Description
The invention is further illustrated by the following examples, which are not intended to limit the scope of the invention.
Example 1
Fig. 1 is a flowchart of an image segmentation method provided in this embodiment. The embodiment relates to a specific implementation process for segmenting an image to be identified by acquiring the image to be identified containing a tubular structure organ, identifying the image to be identified to obtain blood vessel data, obtaining an interface of two adjacent organ segments of the organ based on the blood vessel data, and further segmenting the image to be identified through the interface. As shown in fig. 1, the segmentation method includes:
specifically, the organ having a tubular structure includes organs such as a lung and a liver, and the tubular structure means that a blood vessel runs in the organ, and the organ can be segmented based on the running of the blood vessel. The acquisition of the image to be recognized is based on a medical imaging technology commonly used in clinic, and specifically may include: by ultrasound imaging (B-ultrasound, M-ultrasound), by X-ray and CT imaging, MRI imaging, PET imaging, etc.
specifically, the image to be identified may be identified based on a general image identification algorithm (including but not limited to a neural network algorithm, an artificial intelligence algorithm, etc.), a blood vessel is identified from the image to be identified, and blood vessel position data is obtained to generate blood vessel data, preferably, the blood vessel data may be position data of a vein blood vessel; alternatively, the blood vessel data is further generated by acquiring bifurcation point data of each blood vessel in the image to be recognized. It should be noted that, when identifying a blood vessel based on a general image identification algorithm, an image to be identified may be directly input to the image identification algorithm for identification, or data preprocessing such as image resolution adjustment and image normalization may be performed on the image first, and then the preprocessed image is input to the image identification algorithm for identification. The blood vessel data referred to above may be specifically position data of a blood vessel after constructing a three-dimensional coordinate system of the organ.
specifically, the organ segment is used to represent an independent unit after segmenting the organ, taking the organ lung as an example, the lung segment is generally referred to as a bronchopulmonary segment in medicine, and is referred to as an independent blood oxygen exchange unit, the interface may be an interphalangeal segmentation plane of the lung or an intrapolatal segmentation plane, the interphalangeal segmentation plane is a part of an interphalangeal cleft and is used to divide different lung lobes, the intrapolatal segmentation plane refers to a segmentation plane further divided by taking the lung lobe as a unit, similarly, taking the organ liver as an example, the interface of the liver may be an interphalangeal segmentation plane of the liver or an intrapolatal segmentation plane, the interphalangeal segmentation plane is used to divide different liver lobes, and the intrapolatal segmentation plane refers to a segmentation plane further divided by taking the liver lobe as a unit. In this embodiment, after the blood vessel position data is obtained, the blood vessel position data running between two adjacent organ segments is obtained, and a spatial curved surface is fitted according to the blood vessel position data to serve as an interface between the two adjacent organ segments.
And step 40, segmenting the image to be recognized based on the interface.
Specifically, after the interface is determined, the organ may be divided into different segments according to the interface, and reference may be made to fig. 8, where fig. 8 illustrates an exemplary lung as an example, and shows a schematic diagram after segmentation of the lung.
The image segmentation method is used for segmenting the image to be identified containing the tubular structure organ, obtaining the interface of two adjacent organ segments of the organ through the blood vessel data in the image to be identified, and realizing the segmentation of the image to be identified based on the interface. The segmentation method has the advantages that the segmentation plane is determined by the blood vessel data, so that the method is more reliable, efficient and more consistent with the medical theoretical standard, the differences of organ structures of different patients can be effectively overcome, and the method is more universal.
Fig. 2 shows a specific implementation manner of step 30 in the image segmentation method of the present embodiment, which specifically includes:
specifically, the end point data refers to end point position data of each segment of blood vessel, the segments are named according to the relative position and the segment included angle in the organ where each segment is located, the start point and the end point of each segment are taken as the end points of the segment, the end point data constitute the blood vessel data, specifically, the blood vessel continuously branches during running in the organ, when the image to be identified is identified, the branching point data of each blood vessel in the image to be identified is obtained, taking the organ lung as an example, the running blood vessel comprises a leaf-level blood vessel, a segment-level blood vessel and a sub-segment-level blood vessel, for the leaf-level blood vessel, the start point is the start point of each blood vessel, the end point is the branching point of the leaf-level blood vessel, for the segment-level blood vessel, the start point is the end point of the leaf-level blood vessel, the end point is the branching point of the segment-level blood vessel, for the sub-level blood vessel, the starting point is the termination point of the segment-level blood vessels, and the termination point is the termination point of each blood vessel.
And 32, obtaining an interface according to the end point data.
Specifically, after the end point data is obtained, the spatial curved surface is fitted according to the position data of the end point to be used as an interface of the two adjacent organ segments.
In the segmentation method for the image, in order to meet the medical theoretical standard, the segmentation plane of the organ is determined by segmenting the blood vessel running in the organ, specifically, the blood vessel is obtained by segmenting by using the relative information of the anatomical space, and the interface is further determined based on the endpoint data of the blood vessel segmentation, so that the segmentation method is higher in speed and efficiency.
Referring to fig. 3, fig. 3 shows a specific implementation manner of step 32 in the image segmentation method of this embodiment, which specifically includes:
specifically, the blood vessel segments running between two adjacent organ segments are divided into the same category, and the end points of the blood vessel segments of the same category, that is, the inter-segment blood vessel segments between the adjacent organ segments, are further obtained as fitting points of the curved surfaces of the subsequent organ segments.
322, acquiring target endpoint data of the vessel segments of the same category;
the target endpoint data refers to endpoint position data of the blood vessel segments of the same category, and when a boundary surface is obtained, the blood vessel segments of each category need to be processed respectively.
And 323, fitting according to the target endpoint data to obtain an interface.
Specifically, after a three-dimensional coordinate system of the organ is constructed, position data of a target end point is obtained, and a fitting curved surface is performed according to the position data of the target end point to serve as an interface. The least square method can be used for fitting, but is not limited to this method, and other fitting algorithms can also be used for calculation, a surface function is obtained through fitting, and the space surface embodied by the surface function is used as an interface.
In the image segmentation method, the anatomical space relative information is utilized to determine the segmentation plane of the organ by segmenting the blood vessel running between two adjacent organ segments, so that the segmentation plane better accords with the medical theoretical standard.
Fig. 4 shows a specific implementation manner of step 40 in the image segmentation method of the present embodiment, which specifically includes:
specifically, the target interface refers to an arbitrarily selected interface.
And 42, inputting each voxel point in the image to be identified into a curved function of the target interface so as to determine the organ segment of each voxel point.
Specifically, after the three-dimensional coordinate system of the organ is constructed, the curved function confirms each voxel point in the image to be identified containing the tubular structure organ one by one based on the three-dimensional coordinate system, substitutes the coordinate point of each voxel point into the curved function expression of the target interface, calculates the relative relationship between each voxel point and the segmented curved surface, and determines which side of the two sides of the target interface each voxel point is on, wherein the two sides are two sides in any direction in the space divided by the planes, and can be the upper side, the lower side, the left side, the right side, the front side or the rear side. For example, starting from the reference of a three-dimensional coordinate system, 2 interfaces are obtained in a segmentation process for a certain organ, an interface one and an interface two are sequentially obtained from top to bottom, the organ is divided into a segment one, a segment two and a segment three, coordinate points of a voxel point are respectively brought into curved surface functions corresponding to the interface one and the interface two, and then the voxel point is located on the lower side of the interface one and on the upper side of the interface two, so that the organ segment position segment two of the voxel point can be obtained. And marking organ segments to which the residual voxel points belong according to the curved surface function of each interface in the segmented result for all the voxel points and all the interfaces until all the segments are completed.
In this embodiment, the organs containing tubular structures may be lungs, livers, etc., taking the lungs as an example, the intrapulmonary bronchus gradually thins from the pulmonary portal to the under pleura, the pulmonary blood vessels accompany the trachea in physiological features, and the thin layer CT (layer thickness of 1.25mm) shows a high brightness, and the thick layer CT shows an obvious blood vessel lacking zone, the neural network algorithm is used to independently segment blood vessel data on the chest enhancement data, particularly, obvious intersegment blood vessel branches run between each lung segment, end point data of the blood vessel segment running between two adjacent organ segments is obtained through the blood vessel position data, a space curved surface is fitted according to the end point data as a dividing plane of the two adjacent organ segments, the present embodiment determines a segmenting plane by utilizing intersegment venous running, is more reliable and efficient and better meets the medical theoretical standard, and the segmenting method of the present application can effectively overcome the difference of organ structures of different patients, and the method has more universality.
Example 2
This embodiment is a further improvement on embodiment 1, and referring to fig. 5, a segmentation method of a lung image is shown, which after step 10 further comprises:
specifically, the lung lobes are independent subunits of physiological functions of the lung, each lung lobe is independently surrounded by a pleura, the lung lobes are highlighted on a thin-layer CT (layer thickness is 1.25mm), and are obvious poor blood vessel zones on a thick-layer CT, and the interlobal fissure characteristics are identified by using a neural network model.
specifically, different lung lobes are segmented by utilizing a neural network model according to the characteristics of the lobe fissures, when the lung segments are segmented, the lung lobes are segmented by utilizing a deep learning method, and any convolutional neural network can be used for segmenting the lung lobes and the lung blood vessels in the application.
In this embodiment, taking the lung as an example, in order to reduce the segmentation time and improve the efficiency of lung segment segmentation, the lung lobes are segmented based on a deep learning method, and then the lung lobes are segmented respectively.
Taking the lung segmentation as an example, the scheme of the embodiment is further explained by taking a specific example:
(1) obtaining an image to be identified of an organ comprising a tubular structure
Acquiring chest enhanced CT data, and preprocessing an image, wherein the method specifically comprises the following steps: resampling and adjusting the image resolution, and carrying out image normalization, wherein the lung window is adopted: [ -600, 750] adjust the image as a mean and standard deviation, with the formula:
where μ is the mean, σ is the standard deviation, I is the image resolution, and I' is the adjusted image resolution.
(2) Obtaining blood vessel data in the image to be identified
(a) And (5) carrying out lung lobe segmentation by using a neural network method. It should be noted that any convolutional neural network can be used for segmentation of the lung lobes and the pulmonary veins in the present application.
A VB-Net neural network can be used, the network structure is shown in fig. 6, and the lung lobe segmentation results are shown in fig. 7 (upper lobe, middle lobe, lower lobe). The segmentation process is completed by two networks with different resolutions, the low-resolution network adopts [5mm, 5mm, 5mm ] voxel resolution, the low-resolution network is mainly used for coarse lung lobe positioning and can effectively remove background interference information, then the original image after coarse lung lobe positioning and filtering is input into the high-resolution network, the high-resolution network adopts [1mm, 1mm, 1mm ] voxel resolution and can segment fine lung lobe boundaries, wherein, as the 3D image dimension is large, the training adopts the cutting training, the size of the cutting can be selected according to physical display memory, and the selectable range is as follows: 64-256, preferably 96 × 96, selecting Dice Loss (collective similarity measure function) by an optimization function, and selecting Adam (a first-order optimization algorithm capable of replacing the traditional random gradient descent process) by a training optimizer.
(b) And (4) carrying out pulmonary vein segmentation by using a neural network method. The pulmonary vein segmentation is similar to the pulmonary lobe segmentation, a segmentation process from a low-resolution network to a high-resolution network is also adopted, the peripheral diameter of the pulmonary vein is generally smaller than 1mm, so that the high-resolution network adopts [0.5mm, 0.5mm and 0.5mm ], and the selection of a training process, an optimization function and an optimizer is the same as that in the pulmonary lobe segmentation.
(c) Extracting the center line of the pulmonary vein, segmenting the blood vessel by using a bifurcation point (namely a point with neighborhood > 2) on the center line, naming the segments according to the lung lobe, the relative position and the segment included angle of each segment (see table 1) as shown in fig. 8a and 8b, taking the starting point and the ending point of each segment as the end points of the segment, and forming the blood vessel data by the end point data.
TABLE 1 pulmonary segment naming Table
(3) Obtaining the interface of two adjacent organ segments of the organ according to the blood vessel data
According to the following description of the left and right pulmonary vein running of tables 2 and 3, the blood vessel segments running between two adjacent organ segments are divided into the same category, the blood vessel segments of the same category, that is, the end points on the veins between the segments between the adjacent lung segments are obtained as fitting points of the curved surfaces of the lung segments, and least square fitting is used (including but not limited to the method, and other fitting algorithms can be used for calculation), so as to avoid overlarge curvature change and ensure the smoothness of the curved surfaces, a cubic surface is used for fitting, and any cubic surface can be represented as a power basis function:
z＝a1x3+a2y3+a3x2y+a4xy2+a5x2+a6y2+a7xy+a8x+a9y+a10
wherein, a1…a10For the function coefficients, x, y, and z are three-dimensional coordinate values, respectively.
TABLE 2 left pulmonary vein Branch Walking Meter
TABLE 3 Right pulmonary vein Branch Walking List
(4) Segmenting the image to be recognized based on the boundary surface
The surface fitting equation is established based on the anatomical coordinate system, and the segmentation results of all lung segments are shown in fig. 9, wherein the Right side (Right, R direction) of the human body is taken as the x direction, the back to front (a direction) is taken as the y direction, and the foot to head (Superior, S direction) is taken as the z direction. And substituting the coordinate points of the lung lobes into the function expression of the fitting segmented plane, and calculating the relative relation between each voxel point of the lung lobes and the segmented curved surface.
Calculating a1x3+a2y3+a3x2y+a4xy2+a5x2+a6y2+a7xy+a8x+a9y+a10Z, if the result is calculated>If the voxel point is 0, the voxel point is positioned on the upper side of the segmentation surface and is marked as a segment; if the result of the calculation is<0, then the voxel point is located below the segmentation plane, labeled as another segment. The remaining adjacent lung segments are marked according to the segmented result until all lung segment segmentations are completed. Referring to fig. 10, the cut-out relationship of the left lung apex posterior segment plane to the left lung superior lobe is shown.
In this example, the organ with a tubular structure is exemplified by a lung, an intra-pulmonary bronchus is gradually thinned from the lung portal to the under-pleura, pulmonary blood vessels accompany the trachea in physiological features, the organ is highlighted on a thin layer CT (layer thickness of 1.25mm), and is expressed as an obvious anemic vessel zone on a thick layer CT, blood vessel data can be independently segmented on chest enhancement data by using a neural network algorithm, particularly, obvious intersegment blood vessel branches are run between each lung segment, end point data of a blood vessel segment running between two adjacent lung segments is obtained through blood vessel position data, a space curved surface is fitted according to the end point data to be used as a dividing plane of the two adjacent lung segments, a segmentation plane is determined by intersegment vein running in the embodiment, the segmentation method is more reliable and efficient and meets medical theoretical standards, and the segmentation method can effectively overcome differences of lung structures of different patients, and the method has more universality.
In this embodiment, the organ containing the tubular structure may further include a liver, and referring to fig. 11, a segmentation method of a liver image is shown, and for segmentation of the liver image, after step 10, the segmentation method further includes:
specifically, when the hepatic segment is segmented, hepatic vein feature data is obtained by using the neural network model.
specifically, after obtaining hepatic vein feature data, in practical application, the neural network model is used to obtain ligamentous fissure (hepatic circular ligament fissure, venous ligament fissure), gallbladder fossa and the like of the liver, and different liver lobes are segmented based on the data, so that when a liver segment is segmented, a deep learning method is firstly used to segment the liver lobe, and any convolutional neural network can be used for segmenting the liver lobe and the hepatic vein in the application.
In this embodiment, taking the liver as an example, to reduce the segmentation time and improve the efficiency of liver segmentation, liver lobe segmentation is performed based on a deep learning method, and then each liver lobe is segmented.
Example 3
A segmentation system for an image, as shown in fig. 12, the segmentation system comprising:
the image acquisition module 1 is used for acquiring an image to be identified of an organ containing a tubular structure;
specifically, the organ having a tubular structure includes organs such as a lung and a liver, and the tubular structure means that a blood vessel runs in the organ, and the organ can be segmented based on the running of the blood vessel. The acquisition of the image to be recognized is based on a medical imaging technology commonly used in clinic, and specifically may include: by ultrasound imaging (B-ultrasound, M-ultrasound), by X-ray and CT imaging, MRI imaging, PET imaging, etc.
the blood vessel data acquisition module 2 is used for acquiring blood vessel data in the image to be identified;
specifically, the image to be recognized may be recognized based on a general image recognition algorithm (including but not limited to a neural network algorithm, an artificial intelligence algorithm, etc.), a blood vessel is recognized from the image to be recognized, and blood vessel position data is obtained to generate blood vessel data; alternatively, the blood vessel data is further generated by acquiring bifurcation point data of each blood vessel in the image to be recognized. It should be noted that, when identifying a blood vessel based on a general image identification algorithm, an image to be identified may be directly input to the image identification algorithm for identification, or data preprocessing such as image resolution adjustment and image normalization may be performed on the image first, and then the preprocessed image is input to the image identification algorithm for identification. The blood vessel data referred to above may be specifically position data of a blood vessel after constructing a three-dimensional coordinate system of the organ.
The interface generating module 3 is used for obtaining the interfaces of two adjacent organ segments of the organ according to the blood vessel data;
specifically, the organ segment is used to represent an independent unit after segmenting the organ, taking the organ lung as an example, the lung segment is generally referred to as a bronchopulmonary segment in medicine, and is referred to as an independent blood oxygen exchange unit, the interface may be an interphalangeal segmentation plane of the lung or an intrapolatal segmentation plane, the interphalangeal segmentation plane is a part of an interphalangeal cleft and is used to divide different lung lobes, the intrapolatal segmentation plane refers to a segmentation plane further divided by taking the lung lobe as a unit, similarly, taking the organ liver as an example, the interface of the liver may be an interphalangeal segmentation plane of the liver or an intrapolatal segmentation plane, the interphalangeal segmentation plane is used to divide different liver lobes, and the intrapolatal segmentation plane refers to a segmentation plane further divided by taking the liver lobe as a unit. In this embodiment, after the blood vessel position data is obtained, the blood vessel position data running between two adjacent organ segments is obtained, and a spatial curved surface is fitted according to the blood vessel position data to serve as an interface between the two adjacent organ segments.
And the segmenting module 4 is used for segmenting the image to be identified based on the boundary surface.
Specifically, after the interface is determined, the organ may be divided into different segments according to the interface.
The image segmentation system is used for segmenting an image to be identified containing a tubular structure organ, obtaining an interface of two adjacent organ segments of the organ through blood vessel data in the image to be identified, and realizing segmentation of the image to be identified based on the interface. The segmentation method has the advantages that the segmentation plane is determined by the blood vessel data, so that the method is more reliable, efficient and more consistent with the medical theoretical standard, the differences of organ structures of different patients can be effectively overcome, and the method is more universal.
The interface generation module 3 includes an endpoint data acquisition unit 301;
the end point data obtaining unit 301 is configured to obtain a plurality of blood vessel segments and end point data of each blood vessel segment according to the blood vessel data;
specifically, the end point data refers to end point position data of each segment of blood vessel, the segments are named according to the relative position and the segment included angle in the organ where each segment is located, the start point and the end point of each segment are taken as the end points of the segment, the end point data constitute the blood vessel data, specifically, the blood vessel continuously branches during running in the organ, when the image to be identified is identified, the branching point data of each blood vessel in the image to be identified is obtained, taking the organ lung as an example, the running blood vessel comprises a leaf-level blood vessel, a segment-level blood vessel and a sub-segment-level blood vessel, for the leaf-level blood vessel, the start point is the start point of each blood vessel, the end point is the branching point of the leaf-level blood vessel, for the segment-level blood vessel, the start point is the end point of the leaf-level blood vessel, the end point is the branching point of the segment-level blood vessel, for the sub-level blood vessel, the starting point is the termination point of the segment-level blood vessels, and the termination point is the termination point of each blood vessel.
And the interface generation module 3 is used for obtaining the interface according to the endpoint data.
Specifically, after the end point data is obtained, the spatial curved surface is fitted according to the position data of the end point to be used as an interface of the two adjacent organ segments.
In the segmentation method for the image, in order to meet the medical theoretical standard, the segmentation plane of the organ is determined by segmenting the blood vessel running in the organ, specifically, the blood vessel is obtained by segmenting by using the relative information of the anatomical space, and the interface is further determined based on the endpoint data of the blood vessel segmentation, so that the segmentation method is higher in speed and efficiency.
Furthermore, the interface generation module 3 further includes a classification unit 302;
the classification unit 302 is configured to classify the plurality of vessel segments;
specifically, the blood vessel segments running between two adjacent organ segments are divided into the same category, and the end points of the blood vessel segments of the same category, that is, the inter-segment blood vessel segments between the adjacent organ segments, are further obtained as fitting points of the curved surfaces of the subsequent organ segments.
The interface generation module 3 is used for acquiring target endpoint data of vessel segments of the same category and obtaining the interface according to the target endpoint data fitting.
The target endpoint data refers to endpoint position data of the blood vessel segments of the same category, when the interface is obtained, the blood vessel segments of each category need to be processed respectively, after a three-dimensional coordinate system of an organ is constructed, the position data of the target endpoint is obtained, and a fitting curved surface is carried out according to the position data of the target endpoint to be used as the interface. The least square method can be used for fitting, but is not limited to this method, and other fitting algorithms can also be used for calculation, a surface function is obtained through fitting, and the space surface embodied by the surface function is used as an interface.
The segmentation system of the image determines the segmentation plane of the organ by segmenting the blood vessel running between two adjacent organ segments by using the anatomical space relative information, and the segmentation plane of the organ accords with the medical theoretical standard better.
In this embodiment, the segmentation module 4 further includes a selection unit 401 and an organ segment determination unit 402;
the selection unit 401 is configured to select a target interface arbitrarily;
specifically, the target interface refers to an arbitrarily selected interface.
The organ segment determining unit 402 is configured to input each voxel point in the image to be identified into a surface function of the target interface to determine an organ segment to which each voxel point belongs.
Specifically, after the three-dimensional coordinate system of the organ is constructed, the curved function confirms each voxel point in the image to be identified containing the tubular structure organ one by one based on the three-dimensional coordinate system, substitutes the coordinate point of each voxel point into the curved function expression of the target interface, calculates the relative relationship between each voxel point and the segmented curved surface, and determines which side of the two sides of the target interface each voxel point is on, wherein the two sides are two sides in any direction in the space divided by the planes, and can be the upper side, the lower side, the left side, the right side, the front side or the rear side. For example, starting from the reference of a three-dimensional coordinate system, 2 interfaces are obtained in a segmentation process for a certain organ, an interface one and an interface two are sequentially obtained from top to bottom, the organ is divided into a segment one, a segment two and a segment three, coordinate points of a voxel point are respectively brought into curved surface functions corresponding to the interface one and the interface two, and then the voxel point is located on the lower side of the interface one and on the upper side of the interface two, so that the organ segment position segment two of the voxel point can be obtained. And marking organ segments to which the residual voxel points belong according to the curved surface function of each interface in the segmented result for all the voxel points and all the interfaces until all the segments are completed.
In this embodiment, the organs containing tubular structures may be lungs, livers, etc., taking the lungs as an example, the intrapulmonary bronchus gradually thins from the pulmonary portal to the under pleura, the pulmonary blood vessels accompany the trachea in physiological features, and the thin layer CT (layer thickness of 1.25mm) shows a high brightness, and the thick layer CT shows an obvious blood vessel lacking zone, the neural network algorithm is used to independently segment blood vessel data on the chest enhancement data, particularly, obvious intersegment blood vessel branches run between each lung segment, end point data of the blood vessel segment running between two adjacent organ segments is obtained through the blood vessel position data, a space curved surface is fitted according to the end point data as a dividing plane of the two adjacent organ segments, the present embodiment determines a segmenting plane by utilizing intersegment venous running, is more reliable and efficient and better meets the medical theoretical standard, and the segmenting method of the present application can effectively overcome the difference of organ structures of different patients, and the method has more universality.
Example 4
This embodiment is a further improvement on embodiment 3, in which the organ containing the tubular structure includes a lung, and as shown in fig. 13, the segmentation system further includes a leaf cleft data identification module 5 and a lung leaf segmentation module 6;
the leaf fissure data identification module 5 is used for identifying lung leaf fissure characteristic data in the image to be identified;
specifically, the lung lobes are independent subunits of physiological functions of the lung, each lung lobe is independently surrounded by a pleura, the lung lobes are highlighted on a thin layer CT (layer thickness is 1.25mm), and are obvious leaf septation of a poor blood vessel zone on a thick layer CT, and the leaf septation characteristics are obtained by using a neural network model.
The lung lobe segmentation module 6 is configured to segment the image to be identified based on the characteristic data of the lung lobe cleft to obtain a plurality of lung lobe images;
specifically, different lung lobes are segmented according to the lobe fissuring characteristics by using the neural network model, so that when a lung segment is segmented, the lung lobe is segmented by using a deep learning method, and any convolutional neural network can be used for segmenting the lung lobe and a lung blood vessel in the application.
The blood vessel data acquisition module 2 is configured to acquire each lung lobe image respectively.
In this embodiment, taking the lung as an example, in order to reduce the segmentation time and improve the efficiency of lung segment segmentation, the lung lobes are segmented based on a deep learning method, and then the lung lobes are segmented respectively.
In this embodiment, the organ containing the tubular structure may further include a liver, and referring to fig. 11, the segmentation system further includes a hepatic vein data identification module 7 and a hepatic lobe segmentation module 8;
the hepatic vein data identification module 7 is used for identifying the image to be identified to obtain hepatic vein feature data;
specifically, when the hepatic segment is segmented, hepatic vein feature data is obtained by using the neural network model.
The liver lobe segmentation module 8 is configured to segment the image to be identified based on the hepatic vein feature data to obtain a plurality of liver lobe images;
specifically, after obtaining hepatic vein feature data, in practical application, the neural network model is used to obtain ligamentous fissure (hepatic circular ligament fissure, venous ligament fissure), gallbladder fossa and the like of the liver, and different liver lobes are segmented based on the data, so that when a liver segment is segmented, a deep learning method is firstly used to segment the liver lobe, and any convolutional neural network can be used for segmenting the liver lobe and the hepatic vein in the application.
The blood vessel data acquisition module 2 is used for respectively identifying each liver lobe image.
In this embodiment, taking the liver as an example, to reduce the segmentation time and improve the efficiency of liver segmentation, liver lobe segmentation is performed based on a deep learning method, and then each liver lobe is segmented.
Example 5
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of segmenting an image as described in embodiment 1 or 2 when executing the computer program.
Fig. 14 is a schematic structural diagram of an electronic device provided in this embodiment. FIG. 14 illustrates a block diagram of an exemplary electronic device 90 suitable for use in implementing embodiments of the present invention. The electronic device 90 shown in fig. 14 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention. As shown in fig. 12, the electronic device 90 may be embodied in the form of a general purpose computing device, which may be, for example, a server device. The components of the electronic device 90 may include, but are not limited to: at least one processor 91, at least one memory 92, and a bus 93 that connects the various system components (including the memory 92 and the processor 91).
The bus 93 includes a data bus, an address bus, and a control bus.
The processor 91 executes various functional applications and data processing by running a computer program stored in the memory 92.
The electronic device 90 may also communicate with one or more external devices 94 (e.g., keyboard, pointing device, etc.). Such communication may be through an input/output (I/O) interface 95. Also, the electronic device 90 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet) via a network adapter 96. The network adapter 96 communicates with the other modules of the electronic device 90 via the bus 93. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.
It should be noted that although in the above detailed description several units/modules or sub-units/modules of the electronic device are mentioned, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the units/modules described above may be embodied in one unit/module, according to embodiments of the application. Conversely, the features and functions of one unit/module described above may be further divided into embodiments by a plurality of units/modules.
Example 6
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the segmentation method of the image according to embodiment 1 or 2.
More specific examples, among others, that the readable storage medium may employ may include, but are not limited to: a portable disk, a hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible embodiment, the invention can also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps of implementing the method of segmentation of images described in embodiment 1 or 2, when said program product is run on said terminal device.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that this is by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (10)
1. A method of segmenting an image, the method comprising:
acquiring an image to be identified containing a tubular structure organ;
acquiring blood vessel data in the image to be identified;
obtaining the interface of two adjacent organ sections of the organ according to the blood vessel data;
and segmenting the image to be identified based on the boundary surface.
2. The method of segmenting an image according to claim 1, wherein said deriving an interface of two adjacent organ segments of said organ from said vessel data comprises:
obtaining a plurality of vessel segments and end point data of each vessel segment according to the vessel data;
and obtaining the interface according to the endpoint data.
3. The method of segmenting an image according to claim 2, wherein said deriving the interface from the endpoint data comprises:
classifying the plurality of vessel segments;
acquiring target endpoint data of vessel segments of the same category;
and fitting according to the target endpoint data to obtain the interface.
4. A method of segmenting an image as claimed in claim 3, wherein said classifying said plurality of vessel segments comprises:
and dividing the vessel segments running between two adjacent organ segments into the same category.
5. The method of segmenting an image according to claim 1, wherein the segmenting the image to be recognized based on the boundary surface comprises:
randomly selecting a target interface;
and inputting each voxel point in the image to be recognized into the curved function of the target interface so as to determine the organ segment of each voxel point.
6. The method for segmenting an image according to claim 1, wherein the organ containing a tubular structure comprises a lung, and wherein after the acquiring of the image to be identified of the organ containing a tubular structure, the method further comprises:
identifying lung lobe fissure characteristic data in the image to be identified;
segmenting the image to be identified based on the lung lobe fissure characteristic data to obtain a plurality of lung lobe images;
and in the acquisition of the blood vessel data of the image to be identified, each lung lobe image is respectively acquired.
7. The method for segmenting an image according to claim 1, wherein the acquiring of the blood vessel data in the image to be identified comprises:
dividing veins from the image to be identified, and acquiring vein position data to generate the blood vessel data;
and/or, the acquiring of the blood vessel data in the image to be identified comprises:
and acquiring bifurcation point data of the blood vessel in the image to be identified to generate the blood vessel data.
8. A segmentation system for an image, the segmentation system comprising:
the image acquisition module is used for acquiring an image to be identified of an organ containing a tubular structure;
the blood vessel data acquisition module is used for acquiring blood vessel data in the image to be identified;
the interface generation module is used for obtaining interfaces of two adjacent organ sections of the organ according to the blood vessel data;
and the segmentation module is used for segmenting the image to be identified based on the boundary surface.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of segmenting an image according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of segmenting an image according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011529902.6A CN112561917A (en) | 2020-12-22 | 2020-12-22 | Image segmentation method and system, electronic device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011529902.6A CN112561917A (en) | 2020-12-22 | 2020-12-22 | Image segmentation method and system, electronic device and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112561917A true CN112561917A (en) | 2021-03-26 |
Family
ID=75032135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011529902.6A Pending CN112561917A (en) | 2020-12-22 | 2020-12-22 | Image segmentation method and system, electronic device and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112561917A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113223013A (en) * | 2021-05-08 | 2021-08-06 | 推想医疗科技股份有限公司 | Method, device, equipment and storage medium for pulmonary vessel segmentation positioning |
| CN114445391A (en) * | 2022-01-30 | 2022-05-06 | 推想医疗科技股份有限公司 | Blood vessel segmentation method and device, electronic device and computer readable storage medium |
| WO2023032480A1 (en) * | 2021-08-31 | 2023-03-09 | 富士フイルム株式会社 | Medical image processing device, liver segment division method and program |
| WO2024114478A1 (en) * | 2022-12-02 | 2024-06-06 | 珠海赛纳数字医疗技术有限公司 | Pulmonary recognition processing method and apparatus, and server |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN107392910A (en) * | 2017-07-06 | 2017-11-24 | 沈阳东软医疗系统有限公司 | A kind of lobe of the lung dividing method and device based on CT images |
| CN108961273A (en) * | 2018-07-03 | 2018-12-07 | 东北大学 | A kind of method and system for dividing pulmonary artery and pulmonary vein from CT images |
| US20180365838A1 (en) * | 2015-12-22 | 2018-12-20 | Koninklijke Philips N.V. | Heart model guided coronary artery segmentation |
| US20190066301A1 (en) * | 2017-08-30 | 2019-02-28 | Siemens Healthcare Gmbh | Method for segmentation of an organ structure of an examination object in medical image data |
| 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 |
| 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 |
-
2020
- 2020-12-22 CN CN202011529902.6A patent/CN112561917A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180365838A1 (en) * | 2015-12-22 | 2018-12-20 | Koninklijke Philips N.V. | Heart model guided coronary artery segmentation |
| 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 |
| CN107392910A (en) * | 2017-07-06 | 2017-11-24 | 沈阳东软医疗系统有限公司 | A kind of lobe of the lung dividing method and device based on CT images |
| US20190066301A1 (en) * | 2017-08-30 | 2019-02-28 | Siemens Healthcare Gmbh | Method for segmentation of an organ structure of an examination object in medical image data |
| CN108961273A (en) * | 2018-07-03 | 2018-12-07 | 东北大学 | A kind of method and system for dividing pulmonary artery and pulmonary vein from CT images |
| 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 |
| 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 |
Non-Patent Citations (3)
| Title |
|---|
| ROSS JAMES C: "Pulmonary lobe segmentation based on ridge surface sampling and shape model fitting.", MEDICAL PHYSICS, 13 December 2013 (2013-12-13) * |
| 严凯: "基于深度学习的肺分叶分段", 中国优秀硕士学位论文全文数据库, 15 July 2020 (2020-07-15), pages 1 - 4 * |
| 边子健;覃文军;刘积仁;赵大哲;: "肺部CT图像中的解剖结构分割方法综述", 中国图象图形学报, no. 10, 16 October 2018 (2018-10-16) * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113223013A (en) * | 2021-05-08 | 2021-08-06 | 推想医疗科技股份有限公司 | Method, device, equipment and storage medium for pulmonary vessel segmentation positioning |
| CN113223013B (en) * | 2021-05-08 | 2022-02-22 | 推想医疗科技股份有限公司 | Method, device, equipment and storage medium for pulmonary vessel segmentation positioning |
| WO2023032480A1 (en) * | 2021-08-31 | 2023-03-09 | 富士フイルム株式会社 | Medical image processing device, liver segment division method and program |
| CN114445391A (en) * | 2022-01-30 | 2022-05-06 | 推想医疗科技股份有限公司 | Blood vessel segmentation method and device, electronic device and computer readable storage medium |
| CN114445391B (en) * | 2022-01-30 | 2022-10-28 | 推想医疗科技股份有限公司 | Blood vessel segmentation method and device, electronic device and computer readable storage medium |
| WO2024114478A1 (en) * | 2022-12-02 | 2024-06-06 | 珠海赛纳数字医疗技术有限公司 | Pulmonary recognition processing method and apparatus, and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107622492B (en) | Lung fissure segmentation method and system | |
| CN112561917A (en) | Image segmentation method and system, electronic device and readable storage medium | |
| Torosdagli et al. | Deep geodesic learning for segmentation and anatomical landmarking | |
| WO2021238438A1 (en) | Tumor image processing method and apparatus, electronic device, and storage medium | |
| WO2022037696A1 (en) | Bone segmentation method and system based on deep learning | |
| US8150132B2 (en) | Image analysis apparatus, image analysis method, and computer-readable recording medium storing image analysis program | |
| Damas et al. | Forensic identification by computer-aided craniofacial superimposition: a survey | |
| CN110738701B (en) | Tumor three-dimensional positioning system | |
| US8150113B2 (en) | Method for lung lesion location identification | |
| JP2019506208A (en) | System and method for segmentation of endoscopic medical images | |
| CN107545584A (en) | The method, apparatus and its system of area-of-interest are positioned in medical image | |
| CN106327479B (en) | The device and method that congenital heart disease art medium vessels recognize under angiography intermediary | |
| JP7346553B2 (en) | Determining the growth rate of objects in a 3D dataset using deep learning | |
| WO2022213654A1 (en) | Ultrasonic image segmentation method and apparatus, terminal device, and storage medium | |
| JP2011024799A (en) | Medical image processing device, method, and program | |
| Wu et al. | Coarse-to-fine lung nodule segmentation in CT images with image enhancement and dual-branch network | |
| CN116797612B (en) | Ultrasonic image segmentation method and device based on weak supervision depth activity contour model | |
| CN111724389B (en) | Method, device, storage medium and computer equipment for segmenting CT image of hip joint | |
| WO2020114332A1 (en) | Segmentation-network-based ct lung tumor segmentation method, apparatus and device, and medium | |
| CN116309647B (en) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device | |
| Radlak et al. | Automatic detection of bones based on the confidence map for rheumatoid arthritis analysis | |
| Nysjö et al. | BoneSplit-a 3D texture painting tool for interactive bone separation in CT images | |
| CN116310114A (en) | Coronary artery point cloud data processing optimization method of multitask label model | |
| Xu et al. | Automatic segmentation of orbital wall from CT images via a thin wall region supervision-based multi-scale feature search network | |
| CN115761228A (en) | Coronary artery calcified plaque segmentation method, device, equipment and storage medium |
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 |