CN112907577A - Quantitative analysis method, device, computer equipment and storage medium for three-dimensional structural characteristics of hepatic vein blood vessels - Google Patents
Quantitative analysis method, device, computer equipment and storage medium for three-dimensional structural characteristics of hepatic vein blood vessels Download PDFInfo
- Publication number
- CN112907577A CN112907577A CN202110321994.7A CN202110321994A CN112907577A CN 112907577 A CN112907577 A CN 112907577A CN 202110321994 A CN202110321994 A CN 202110321994A CN 112907577 A CN112907577 A CN 112907577A
- Authority
- CN
- China
- Prior art keywords
- vein
- blood vessel
- vessel
- hepatic
- segmentation
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30056—Liver; Hepatic
-
- 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)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Computational Biology (AREA)
- Image Processing (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention provides a quantitative analysis method for three-dimensional structural characteristics of hepatic vein blood vessels, which comprises the following steps: acquiring a three-dimensional image of a liver to be analyzed, and acquiring a segmentation region of a vein; calculating a central line; performing blood vessel segmentation treatment, namely dividing vein blood vessels into left, middle and right hepatic veins and left and right hepatic portal veins; and quantitatively analyzing the three-dimensional structural characteristics of the vein according to the central line of each section and the vein model reconstructed by using the three-dimensional surface reconstruction algorithm, wherein the parameter indexes comprise one or more of vessel morphology, branch and grid characterization indexes. Related apparatus, computer devices and storage media are also provided. The invention can realize full-automatic analysis without manual intervention, and can meet the requirement of batch analysis of blood vessel parameters, thereby being capable of analyzing the three-dimensional structural characteristic quantitative index of the hepatic vein blood vessel rapidly, accurately and in batches, providing more comprehensive and detailed quantitative analysis of the three-dimensional structural characteristic of the blood vessel, and providing more comprehensive reference for doctors to evaluate the blood vessel.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to the technical field of quantitative analysis of three-dimensional structural features of blood vessels, and specifically relates to a quantitative analysis method, device, computer equipment and storage medium for the three-dimensional structural features of hepatic veins.
Background
Many diseases can cause the change of the blood vessel morphology to a certain extent, so that the quantitative analysis of the blood vessel can be used for early prediction and clinical diagnosis of certain diseases. Because the two-dimensional blood vessel quantitative analysis is often influenced by the structure overlapping, the three-dimensional blood vessel quantitative analysis result is more accurate.
Some software or patents for evaluating blood vessel parameters are already on the market, and the contents of the software or patents have certain defects or imperfections. For example, the parameter evaluation of the blood vessel adopts an interactive evaluation mode, that is, a series of points are placed at the position of the blood vessel in a man-machine interaction mode, and parameters such as the length of the blood vessel, the section diameter of the blood vessel, the angle of the bifurcation of the blood vessel and the like are calculated according to the positions of the points, but the method cannot meet the requirement of batch processing. Of course, there are some fully automatic algorithms for evaluating vascular structural features, but the evaluation indexes of the algorithms for evaluating the vascular structural features are single, and a comprehensive quantitative analysis cannot be provided for the health condition of the blood vessels.
For example, chinese patent application No. cn202010153299.x (title of the invention, "blood vessel parameter measurement method, apparatus, computer device, and storage medium") discloses a blood vessel parameter measurement method, which has the following specific technical solution:
1. performing blood vessel segmentation based on deep learning on medical image data (CT and MRI) to be detected;
firstly, preprocessing marked medical image data, including at least one of noise adding, over-under balance sampling processing, data whitening and the like, to realize the expansion of training samples and increase the generalization capability of the trained model so as to obtain more accurate segmentation results.
2. Segmenting a seed point region of medical image data to be detected in a deep learning mode, and extracting seed points of the seed point region;
and training the seed point region model by using the blood vessel image sample data and the seed point region marking data (used for marking the seed point region in the whole artery) of the blood vessel image sample to obtain the seed point region segmentation model.
3. Generating a center line of the blood vessel according to the seed points;
and performing region growing on the seed points extracted from the seed point region segmented by the seed point region segmentation model in the segmented blood vessel data, and connecting the centroids of all the grown regions to be used as the central line of the blood vessel. The method comprises the following specific steps:
(1) and taking the whole artery in the segmented image as a current unmarked area, taking the seed point as a queue starting point, establishing a first central line queue, and taking the seed point as a current point of the first central line queue.
(2) And based on the area where the current point is located, performing area growth in the current unmarked area to obtain a growth area, and taking the growth area as the marked area.
(3) And when the growing region does not reach the vessel end of the whole artery and the growing region is communicated, adding the centroid of the growing region to the first center line queue, taking the centroid as the current point, taking the region of the whole artery except the marked region as the current unmarked region, returning to the region based on the current point, and performing region growing in the current unmarked region to continue execution.
(4) And when the growing area does not reach the tail end of the blood vessel of the whole artery and the growing area is a plurality of branch areas which are not communicated, respectively taking the mass centers of all the branch areas, adding one of the mass centers to the first central line queue, and taking the mass center as the current point of the first central line queue. And respectively establishing corresponding second center line queues by taking the residual centroids as queue starting points, and respectively taking the residual centroids as current points of the corresponding second center line queues. And taking the region of the whole artery except the marked region as a current unmarked region, and performing region growing in the current unmarked region based on the region where the current point is located to obtain a growing region and subsequent steps until the growing region reaches the vessel end of the whole artery.
(5) And when the growing region reaches the vessel end of the whole artery and no branch region exists, generating the central line of the whole artery according to the first central line queue and the second central line queue.
4. And performing interactive measurement on blood vessel parameters on Web according to the central line and the segmented blood vessel data, and generating a corresponding report.
The method comprises the following steps of receiving a blood vessel parameter measurement instruction based on a measurement page, and generating a measurement result corresponding to the blood vessel parameter measurement instruction according to a segmentation image and a center line, wherein the method specifically comprises the following steps:
(1) receiving a blood vessel parameter measurement instruction based on the measurement page;
(2) and determining the selected measurement type and the selected measurement point on the central line in response to the blood vessel parameter measurement instruction. The measurement type is a type of a blood vessel parameter to be measured. Such as: equivalent diameter on the cross section of the blood vessel, length distance between two points on the center line of the blood vessel, included angle formed between three points on the center line of the blood vessel and the like. (ii) a
(3) And according to the measuring points and the segmentation images, carrying out measurement processing corresponding to the measurement type on the whole artery to obtain a measurement result corresponding to the measurement type. Based on the measurement results, an adjustable blood vessel parameter report list can be generated, and the blood vessel parameter report list includes the measurement points and the measurement types contained in the blood vessel parameter measurement instruction, the measurement region image, and the measurement results. According to the sequence of the measurement results in the blood vessel parameter report list, a blood vessel parameter graphic report can be generated.
However, the above-described technical solutions have the following problems:
1. the measurement of the quantitative index of the three-dimensional structural feature of the blood vessel is incomplete, that is, the measured parameters are not comprehensive enough, and the health condition of the blood vessel cannot be evaluated in an all-round way. The above invention only measures the maximum/small value of the diameter on the cross section of the blood vessel, the equivalent diameter of the cross section of the blood vessel, the length of the blood vessel, and the angle of the blood vessel at the bifurcation.
2. The adopted interactive measurement scheme requires manual intervention in the measurement process, and the requirement of batch measurement cannot be met.
Therefore, it is required to provide a quantitative analysis method for three-dimensional structural characteristics of hepatic vein vessels, which can rapidly, accurately and massively analyze quantitative indexes of three-dimensional structural characteristics of hepatic vein vessels.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a method, a device, a computer device and a storage medium for quantitatively analyzing the three-dimensional structural characteristics of the hepatic vein blood vessel, which can quickly, accurately and quantitatively analyze the quantitative indexes of the three-dimensional structural characteristics of the hepatic vein blood vessel in batches.
In order to achieve the above object, in a first aspect of the present invention, there is provided a method for quantitatively analyzing three-dimensional structural features of hepatic vein blood vessels, comprising the steps of:
(1) acquiring a three-dimensional image of a liver to be analyzed, and acquiring a segmentation region of a vein of the three-dimensional image of the liver to be analyzed;
(2) performing centerline calculation on the vein vessel in the segmented region to obtain a centerline of the vein vessel;
(3) performing vessel segmentation processing on the vein vessels in the segmentation region, and dividing the vein vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein branch and a right hepatic portal vein branch;
(4) quantitatively analyzing the three-dimensional structural characteristics of the vein according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic portal vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein the parameter indexes of quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel torsion degree, a blood vessel diameter, a blood vessel volume and a volume percentage, the blood vessel branch characterization index comprises one or more of a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood vessel end node number and a blood vessel branch node number, the vessel mesh characterization indicator includes a vessel distribution density.
Preferably, in the step (1), the three-dimensional image of the liver to be analyzed is a liver MRI three-dimensional image.
Preferably, in the step (1), the step of obtaining the segmented region of the vein of the three-dimensional image of the liver to be analyzed specifically includes:
(11) and inputting the liver three-dimensional image to be analyzed into a pre-trained vein vessel segmentation model, and outputting the segmentation region of the vein vessel.
Preferably, in the step (11), the pre-trained vein vessel segmentation model is obtained by the following steps:
(a) acquiring a sample liver three-dimensional image, and acquiring venous vessel labeling data of the sample liver three-dimensional image;
(b) and taking the sample liver three-dimensional image and the vein vessel marking data as a training set, and carrying out deep learning training on the vein vessel segmentation model iteration to obtain the pre-trained vein vessel segmentation model.
Further, in the step (a), the step of obtaining a three-dimensional image of a sample liver specifically includes:
(a1) preprocessing an original sample liver three-dimensional image to obtain a newly-added sample liver three-dimensional image, wherein the preprocessing comprises one or more of histogram equalization, rotation, noise addition, distortion and mirror image;
(a2) and merging the original sample liver three-dimensional image and the newly-added sample liver three-dimensional image to obtain the sample liver three-dimensional image.
Further, in the step (b), the deep learning network adopted by the vein segmentation model is a fully supervised segmentation network based on a combination of U-Net and a mixed domain attention mechanism.
Preferably, in the step (3), the vein vessel segmentation processing is performed on the vein vessel in the segmentation region according to the topological shape of the centerline of the vein vessel and the direction information of the vein vessel in the three-dimensional space in the segmentation region.
In a second aspect of the present invention, there is provided a device for quantitatively analyzing three-dimensional structural features of hepatic vein blood vessels, comprising:
the vein vessel segmentation module is used for acquiring a segmentation area of a vein vessel of a liver three-dimensional image to be analyzed;
a central line calculating module which is connected with the vein vessel segmentation module in a signal mode and is used for calculating the central line of the vein vessel in the segmentation region so as to obtain the central line of the vein vessel;
the blood vessel segmentation processing module is in signal connection with the central line calculation module and is used for performing blood vessel segmentation processing on the vein blood vessels in the segmentation region, and dividing the vein blood vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein and a right hepatic portal vein;
a three-dimensional structural feature quantitative analysis module in signal connection with the blood vessel segmentation processing module and used for carrying out quantitative analysis on the three-dimensional structural feature of the vein blood vessel according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein parameter indexes of the quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel tortuosity, a blood vessel diameter, a blood vessel volume and a volume percentage, and the blood vessel branch characterization index comprises a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood, The number of the end nodes of the blood vessel and the number of the branch nodes of the blood vessel are one or more, and the characterization index of the blood vessel grid comprises the distribution density of the blood vessel.
In a third aspect of the present invention, there is provided a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the method for quantitative analysis of three-dimensional structural features of hepatic vein vessels when executing the computer program.
In a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program is executed by a processor to implement the method for quantitative analysis of three-dimensional structural features of hepatic vein vessels described above.
By adopting the method, the device, the computer equipment and the storage medium for quantitatively analyzing the three-dimensional structural characteristics of the hepatic vein blood vessels, the full-automatic analysis can be realized without manual intervention, and the requirement of batch analysis of blood vessel parameters can be met, so that the quantitative indexes of the three-dimensional structural characteristics of the hepatic vein blood vessels can be analyzed quickly, accurately and in batches.
Drawings
Fig. 1 is a schematic flow chart of a method for quantitatively analyzing three-dimensional structural features of hepatic vein vessels according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a deep learning training method of a vein vessel segmentation model in the embodiment shown in fig. 1.
Fig. 3 is a schematic front view of a segmented region and a center line of a vein of a three-dimensional image of a liver obtained by the embodiment shown in fig. 1.
Fig. 4 is a schematic frame diagram of an embodiment of the device for quantitatively analyzing three-dimensional structural features of hepatic vein vessels according to the present invention.
Detailed Description
In order to clearly understand the technical contents of the present invention, the following examples are given in detail. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1 to 2, in an embodiment of the present invention, a method for quantitatively analyzing three-dimensional structural characteristics of hepatic vein vessels includes the following steps:
(1) acquiring a three-dimensional image of a liver to be analyzed, and acquiring a segmentation region of a vein of the three-dimensional image of the liver to be analyzed;
(2) performing centerline calculation on the vein vessel in the segmented region to obtain a centerline of the vein vessel;
(3) performing vessel segmentation processing on the vein vessels in the segmentation region, and dividing the vein vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein branch and a right hepatic portal vein branch;
(4) quantitatively analyzing the three-dimensional structural characteristics of the vein according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic portal vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein the parameter indexes of quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel torsion degree, a blood vessel diameter, a blood vessel volume and a volume percentage, the blood vessel branch characterization index comprises one or more of a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood vessel end node number and a blood vessel branch node number, the vessel mesh characterization indicator includes a vessel distribution density.
In the step (1), the three-dimensional image of the liver is a three-dimensional image taken of the liver. It is understood that the three-dimensional image of the liver includes three-dimensional images of hepatic vein vessels including the hepatic left vein, the hepatic middle vein, the hepatic right vein, the hepatic portal left branch, and the hepatic portal right branch. The portal vein enters from the portal and then extends into the left and right branches of the portal vein, and is gradually subdivided into the capillaries of the liver. The left, middle and right branch vessels of the hepatic vein are connected to the inferior vena cava, and the hepatic vein collects blood in the liver and flows into the inferior vena cava. The three-dimensional image of the liver to be analyzed is the three-dimensional image of the liver which needs to be subjected to quantitative analysis of blood vessel parameter indexes.
In the step (1), the three-dimensional image of the liver to be analyzed may be any suitable type of three-dimensional image of the liver, such as a CT (Computed Tomography) three-dimensional image or a MRI (Magnetic Resonance Imaging) three-dimensional image, and in the step (1), the three-dimensional image of the liver to be analyzed is an MRI three-dimensional image of the liver.
In the step (1), the step of acquiring the segmented region of the vein blood vessel of the three-dimensional image of the liver to be analyzed is a step of performing vein blood vessel segmentation on the three-dimensional image of the liver to be analyzed to obtain the segmented region of the vein blood vessel, and any suitable method may be specifically adopted, and in an embodiment of the present invention, in the step (1), the step of acquiring the segmented region of the vein blood vessel of the three-dimensional image of the liver to be analyzed specifically includes:
(11) and inputting the liver three-dimensional image to be analyzed into a pre-trained vein vessel segmentation model, and outputting the segmentation region of the vein vessel.
The vein segmentation model is a machine learning model for performing a vein segmentation process. The pre-trained vein vessel segmentation model can be obtained through deep learning training.
In the step (11), the pre-trained vein segmentation model may be obtained by any suitable method, and in a specific embodiment of the present invention, in the step (11), the pre-trained vein segmentation model is obtained by the following steps:
(a) acquiring a sample liver three-dimensional image, and acquiring venous vessel labeling data of the sample liver three-dimensional image;
(b) and taking the sample liver three-dimensional image and the vein vessel marking data as a training set, and carrying out deep learning training on the vein vessel segmentation model iteration to obtain the pre-trained vein vessel segmentation model.
Namely, inputting the sample liver three-dimensional image in the training set into a deep learning network to obtain vein blood vessel prediction data, then performing difference comparison on the vein blood vessel prediction data and vein blood vessel marking data in the training set, and iteratively updating the deep learning network according to the difference until an iteration termination condition is met to obtain a vein blood vessel segmentation model.
In the step (a), the step of obtaining a three-dimensional image of a sample liver may specifically adopt any suitable method, and in an embodiment of the present invention, in the step (a), the step of obtaining a three-dimensional image of a sample liver specifically includes:
(a1) preprocessing an original sample liver three-dimensional image to obtain a newly-added sample liver three-dimensional image, wherein the preprocessing comprises one or more of histogram equalization, rotation, noise addition, distortion and mirror image;
(a2) and merging the original sample liver three-dimensional image and the newly-added sample liver three-dimensional image to obtain the sample liver three-dimensional image.
The original sample liver three-dimensional image is a sample liver three-dimensional image which is not subjected to pretreatment. Because the image characteristics of liver image data acquired by different instruments are different, in order to make the generalization capability of the blood vessel segmentation model stronger, a certain preprocessing method is required to be adopted during model training, wherein the preprocessing method comprises histogram equalization, rotation, noise addition, distortion, mirror image and other operations. The image histogram equalization is a method for enhancing the contrast of an image, and can change the histogram distribution of the image into approximately uniform distribution to realize the effect of enhancing the contrast. The rotation, distortion and mirror image function is to perform data transformation on liver image data, that is, pixel points in the image are mapped to new positions according to a certain rule, so that the characteristics of the image can be changed appropriately. The changed images are put into the training set, so that the number of samples of the training set can be increased, and the generalization capability of the blood vessel segmentation model after the training is finished can be increased. The noise is added to the image, which is equivalent to adding some interference factors, and gaussian noise and some other random noise are added commonly.
In the step (b), the deep learning network adopted by the vein segmentation model may be any suitable deep learning network, please refer to fig. 2, and in a specific embodiment of the present invention, in the step (b), the deep learning network adopted by the vein segmentation model is a fully supervised segmentation network based on a combination of U-Net and a mixed domain attention mechanism.
As shown in fig. 2, the process schematic diagram of obtaining the pre-trained vein vessel segmentation model through deep learning training is shown. As can be seen from fig. 2, the overall training process is such that: inputting an original medical image (namely a sample liver three-dimensional image) into a deep learning network to obtain a prediction result of the deep learning network, comparing the prediction result (namely vein vessel prediction data) with an artificial label (namely vein vessel label data), feeding back the result to the deep learning network, taking the artificial label as a target, continuously updating the deep learning network according to the fed-back comparison information until the prediction result is close to the artificial label, and thus obtaining the pre-trained vein vessel segmentation model in the embodiment.
It should be noted that the vein extraction method described above uses a vein extraction method based on deep learning, and other vein segmentation schemes, such as vein data extracted by other conventional vein segmentation algorithms or vein data manually labeled by professionals, may be used in the actual use process.
In the step (2), the centerline calculation may specifically adopt any suitable method, and in a specific embodiment of the present invention, in the step (2), the centerline calculation adopts a vmtkNetworkExtraction function in an open source tool vmtk (vascular modeling toolkit) to extract an approximate centerline of the blood vessel and extract a blood vessel branch distal end point based on the approximate centerline, and then performs the centerline calculation using a vmtkcentrilines function. It is also possible to use the three-dimensional skeleton extraction method in the medical image segmentation and registration tool ITK, where the skeleton is composed of three-dimensionally discrete points and then the centerline is generated from the skeleton.
The step (2) may further include any other suitable steps, and in a specific embodiment of the present invention, the step (2) further includes:
and optimizing the vein vessel according to the central line of the vein vessel.
In the step (2), the optimization process may adopt any suitable method, and in a specific embodiment of the present invention, in the step (2), the optimization process is performed according to an end point of the centerline of the vein vessel along a length direction of the centerline of the vein vessel.
In step (2), the optimization process may be any suitable optimization process, and in a specific embodiment of the present invention, in step (2), the optimization process includes one or more of eliminating fine spurs on the surface of the vein vessel and connecting disconnected branches of the vein vessel.
In the step (3), the vessel segmentation process may adopt any suitable method, and in a specific embodiment of the present invention, in the step (3), the vessel segmentation process is performed on the vein vessel in the segmented region according to the topological shape of the centerline of the vein vessel and the directional information of the vein vessel in the three-dimensional space in the segmented region. And finally, determining the left hepatic vein, the right hepatic vein, the middle hepatic vein and the right hepatic vein according to the direction of each branch of the central line from the intersection of the hepatic vein and the inferior vena cava. In the same way, from the entrance of the portal vein, the left portal vein and the right portal vein are determined according to the direction of the center line of the portal vein.
It should be noted that, regarding the blood vessel segmentation processing, in addition to the above-mentioned method mentioned in the present invention, vein vessel segmentation data labeled by a professional or data generated by other segmentation methods with reliable results may be used.
In the step (4), the blood vessel distortion degree refers to a ratio of a length of the blood vessel to a linear distance between two end points of the blood vessel, and is used for measuring the distortion degree of the blood vessel; the vessel diameter comprises the diameter maximum value, the diameter minimum value and the equivalent diameter of the vessel cross section; the blood vessel length refers to the length from one bifurcation of the blood vessel to the next bifurcation; the blood vessel volume refers to the volume of blood that can be accommodated by the whole liver blood vessel; the volume percentage is the proportion of the volume which can be contained by the left branch, the middle branch, the right branch of the hepatic vein, the left branch and the right branch of the hepatic portal vein respectively to the volume of the whole blood vessel; the blood vessel branch number refers to the number of blood vessel branches, wherein the blood vessel branches are linear structures between two nodes of a blood vessel (a blood vessel endpoint and a bifurcation point, a bifurcation point and a bifurcation point). The blood vessel bifurcation angle refers to an angle between axial direction vectors of two blood vessels at a blood vessel branch; the number of the blood vessel root nodes refers to the number of the blood vessel root nodes, the number of the blood vessel end nodes refers to the number of the blood vessel end nodes, and the number of the blood vessel branch nodes refers to the number of the blood vessel branch nodes. The root node is equivalent to the entrance of the blood vessel, the end node is the end point of the blood vessel end, and the branch node is the axial intersection point of the bifurcation of the blood vessel; the blood vessel distribution density refers to the absolute number of blood vessels in a certain designated area, and can quantitatively reflect the distribution trend of the blood vessels.
The quantitative analysis method for the three-dimensional structural characteristics of the hepatic vein blood vessels of the invention is specifically described below by taking 500 MRI three-dimensional images of an original sample liver and an MRI three-dimensional image of a liver of a certain liver disease patient as the MRI three-dimensional images of the liver to be analyzed.
1. Obtaining of a pre-trained vein vessel segmentation model
Performing adaptive histogram equalization on each sample data of 500 original sample liver MRI three-dimensional images, and simultaneously alternately performing rotation, noise, distortion and mirror image on each sample data at intervals of 3 slices in sequence, wherein the rotation angle is 15 degrees and 30 degrees, the mirror image is a vertical mirror image and a horizontal mirror image, and Gaussian noise is randomly added in the images. 500 sample liver MRI three-dimensional images are obtained, 350 parts (70%) of the sample liver MRI three-dimensional images and corresponding liver vessel labeling data are used as a training set, and 150 parts (30%) of the sample liver MRI three-dimensional images and corresponding liver vessel labeling data are used as a verification set.
The training set is input into the fully supervised split network based on a combination of U-Net and mixed domain attention mechanisms as shown in fig. 2. The network is generally divided into encoding and decoding processes, and the encoding process adopts convolution and downsampling operations to extract shallow features in an image. In the decoding process, the convolution and up-sampling operation is adopted to extract deep features of the image, the image information of different layers is fused through the jump layer, and a mixed domain attention mechanism is combined in the fusion process, namely certain features of the image are more obvious. And finally, comparing the output result with the manually marked data and updating the parameters of the model. The size of batch is set to be 8 in the whole training process, the learning rate is 0.0001, the size of a convolution kernel is 3 multiplied by 3 in the downsampling process, the size of a window is set to be 2 multiplied by 2 in the pooling process, an Adam optimizer is adopted for training, and a multi-classification cross entropy loss function is adopted as a loss function.
The verification set is input into the pre-trained vein vessel segmentation model obtained in the above way, the performance in the verification set is better, the Dice value (used for evaluating the effect of image segmentation, the value range is between 0 and 1, the closer to the real result, the value is represented) of the whole verification set is between 0.82 and 0.91, and the average Dice value of the whole verification set is 0.846.
2. Quantitative analysis of three-dimensional structural features of blood vessels of liver MRI (magnetic resonance imaging) three-dimensional image to be analyzed
And inputting the MRI three-dimensional image of the liver to be analyzed into the obtained pre-trained vein vessel segmentation model to obtain a segmentation region of the vein vessel, and performing center line calculation on the vein vessel in the segmentation region to obtain a center line of the vein vessel. Referring to fig. 3, fig. 3 shows a segmented region of vein blood vessels (100 is the segmented vein blood vessels) in a three-dimensional image of a liver and a centerline (200) of the vein blood vessels in the segmented region.
The direction information of the liver data can be obtained from the file header of the collected original liver image data, and the obtained direction information is corresponding to the blood vessel center line data. Then starting from the root node of the blood vessel to the first bifurcation node, judging a direction vector formed by the two points, comparing the direction vector with the direction information, wherein the blood vessel corresponding to the central line beam closest to the right in the direction information is the right branch of the hepatic vein, the blood vessel closest to the left side is the left branch of the hepatic vein, and the rest is the hepatic mid-vein. The left branch of the hepatic portal vein and the right branch of the hepatic portal vein can be judged by the same method.
After the vessel segmentation is completed, the three-dimensional structural characteristics of the vessel can be quantitatively analyzed. Such as the length of a vessel may be measured by the length of the centerline of the segment of the vessel; the diameter of the blood vessel is the maximum value, the minimum value and the diameter equivalent value of the diameter of the section of the blood vessel and a normal plane formed by two adjacent points of the central line; calculating the ratio of the volume of the branch of the blood vessel to the total volume of the blood vessel according to the volume of the blood vessel; the number of various nodes (root nodes, end nodes and branch nodes) of the blood vessel is determined according to the number of the entrance of the central line, the end points of the central line and the branch nodes of the central line; the number of branches of a vessel can be measured in terms of the end-point to bifurcation point of the vessel centerline and the number of bifurcation points to bifurcation points; the bifurcation angle of the blood vessel can also be an angle formed by taking a point from the bifurcation position of the center line of the blood vessel and each of two center lines of the bifurcation; the vessel distribution density is the number of vessels calculated in a region designated by the user, and may be measured by the number of centerlines of the region.
Taking the left hepatic vein as an example in liver data, the results are obtained after parameter evaluation: the distance from the root node of the left hepatic vein to the first branch point is 1.6cm, and the distance from the first branch node to the second branch node is 3.58 cm; the two branch angles of the first branch node are 34.785 degrees; the number of branch nodes is 14; the number of end nodes is 16; the average value of the twist from the root node to the end node is 0.25; the pipe diameter equivalent diameter from the root node to the first branch node is 1.06cm, the maximum value is 1.36cm, and the minimum value is 0.76 cm; the volume of the left branch of the whole hepatic vein is32.740cm3Corresponding to a volume percentage of 26.73%; the number of branches of the blood vessel is 24; from the specified region, the distribution density 6 of the blood vessels is calculated.
Referring to fig. 4, in an embodiment of the present invention, the present invention further provides a device for quantitatively analyzing three-dimensional structural features of hepatic vein vessels, including:
the vein vessel segmentation module is used for acquiring a segmentation area of a vein vessel of a liver three-dimensional image to be analyzed;
a central line calculating module which is connected with the vein vessel segmentation module in a signal mode and is used for calculating the central line of the vein vessel in the segmentation region so as to obtain the central line of the vein vessel;
the blood vessel segmentation processing module is in signal connection with the central line calculation module and is used for performing blood vessel segmentation processing on the vein blood vessels in the segmentation region, and dividing the vein blood vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein and a right hepatic portal vein;
a three-dimensional structural feature quantitative analysis module in signal connection with the blood vessel segmentation processing module and used for carrying out quantitative analysis on the three-dimensional structural feature of the vein blood vessel according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein parameter indexes of the quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel tortuosity, a blood vessel diameter, a blood vessel volume and a volume percentage, and the blood vessel branch characterization index comprises a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood, The number of the end nodes of the blood vessel and the number of the branch nodes of the blood vessel are one or more, and the characterization index of the blood vessel grid comprises the distribution density of the blood vessel.
The vein segmentation module is configured to obtain a segmented region of a vein of a three-dimensional liver image to be analyzed, that is, to perform vein segmentation on the three-dimensional liver image to be analyzed to obtain the segmented region of the vein, and specifically, any suitable method may be adopted.
The device for quantitatively analyzing the three-dimensional structural feature of the hepatic vein may further include any other suitable components, please refer to fig. 4, and in a specific embodiment of the present invention, the device for quantitatively analyzing the three-dimensional structural feature of the hepatic vein may further include a vein segmentation model training module, which is in signal connection with the vein segmentation module and is configured to obtain a three-dimensional image of a sample liver and obtain vein labeling data of the three-dimensional image of the sample liver; and taking the sample liver three-dimensional image and the vein vessel marking data as a training set, and carrying out deep learning training on the vein vessel segmentation model iteration to obtain the pre-trained vein vessel segmentation model.
In order to make the pre-trained vein segmentation model more accurate, in a specific embodiment of the present invention, the vein segmentation model training module is further configured to pre-process the original sample liver three-dimensional image to obtain a newly added sample liver three-dimensional image, where the pre-processing includes one or more of histogram equalization, rotation, noise addition, distortion, and mirroring; and merging the original sample liver three-dimensional image and the newly-added sample liver three-dimensional image to obtain the sample liver three-dimensional image.
The deep learning network adopted by the vein segmentation model can be any suitable deep learning network, and in a specific embodiment of the invention, the deep learning network adopted by the vein segmentation model is a fully supervised segmentation network based on the combination of U-Net and a mixed domain attention mechanism.
The device for quantitatively analyzing the three-dimensional structural feature of the hepatic vein may further include any other suitable components, please refer to fig. 4, and in an embodiment of the present invention, the device for quantitatively analyzing the three-dimensional structural feature of the hepatic vein may further include a centerline optimization module, respectively connected to the centerline calculation module and the blood vessel segmentation processing module in a signal manner, and configured to perform optimization processing on the vein according to a centerline of the vein.
The vessel segmentation processing module is configured to perform vessel segmentation processing on the vein vessel in the segmented region by any suitable method, and in a specific embodiment of the present invention, the vessel segmentation processing module is specifically configured to perform the vessel segmentation processing on the vein vessel in the segmented region according to a topological shape of a center line of the vein vessel and direction information of the vein vessel in a three-dimensional space in the segmented region.
Other specific limitations of the device for quantitatively analyzing the three-dimensional structural feature of the hepatic vein blood vessel can be referred to the above limitations of the method for quantitatively analyzing the three-dimensional structural feature of the hepatic vein blood vessel, and will not be described herein again. All or part of the modules in the device for quantitatively analyzing the three-dimensional structural characteristics of the hepatic vein blood vessels can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In an embodiment of the present invention, the present invention further provides a computer device, which includes a memory and a processor, where the memory stores a computer program, and the processor implements the method for quantitatively analyzing the three-dimensional structural feature of the hepatic venous blood vessel when executing the computer program.
In an embodiment of the present invention, the present invention further provides a computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the method for quantitative analysis of three-dimensional structural features of hepatic vein vessels.
Therefore, the invention mainly realizes the rapid and automatic quantitative analysis of the three-dimensional structural characteristics of hepatic veins and hepatic portal vein vessels. The method comprises the steps of realizing accurate extraction of hepatic vein blood vessels from collected hepatic data by a deep learning method; and carrying out central line calculation according to the extracted venous blood vessel data, segmenting the venous blood vessels according to the topological structure of the central line and the spatial information of the venous blood vessels, and carrying out quantitative analysis on each segment of the venous blood vessels by combining a venous blood vessel model reconstructed by using a three-dimensional surface reconstruction algorithm with the central line.
1. The quantitative analysis method for the three-dimensional structural characteristics of the hepatic vein blood vessels is a full-automatic analysis method without manual intervention, and can meet the requirement of analyzing blood vessel parameters in batches.
2. Compared with the prior art, the method provides more comprehensive and detailed analysis of the vessel structure characteristics, and can provide more comprehensive reference for doctors to evaluate the vessels.
Therefore, by adopting the method, the device, the computer equipment and the storage medium for quantitatively analyzing the three-dimensional structural characteristics of the hepatic vein blood vessels, the full-automatic analysis can be realized without manual intervention, and the requirement of analyzing blood vessel parameters in batches can be met, so that the quantitative indexes of the three-dimensional structural characteristics of the hepatic vein blood vessels can be analyzed quickly, accurately and in batches.
It will thus be seen that the objects of the invention have been fully and effectively accomplished. The functional and structural principles of the present invention have been shown and described in the embodiments, and the embodiments may be modified without departing from the principles. Therefore, this invention includes all modifications encompassed within the spirit and scope of the claims.
Claims (10)
1. A quantitative analysis method for three-dimensional structural features of hepatic vein vessels is characterized by comprising the following steps:
(1) acquiring a three-dimensional image of a liver to be analyzed, and acquiring a segmentation region of a vein of the three-dimensional image of the liver to be analyzed;
(2) performing centerline calculation on the vein vessel in the segmented region to obtain a centerline of the vein vessel;
(3) performing vessel segmentation processing on the vein vessels in the segmentation region, and dividing the vein vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein branch and a right hepatic portal vein branch;
(4) quantitatively analyzing the three-dimensional structural characteristics of the vein according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic portal vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein the parameter indexes of quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel torsion degree, a blood vessel diameter, a blood vessel volume and a volume percentage, the blood vessel branch characterization index comprises one or more of a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood vessel end node number and a blood vessel branch node number, the vessel mesh characterization indicator includes a vessel distribution density.
2. The method for quantitative analysis of three-dimensional structural features of hepatic vein vessels according to claim 1, wherein in the step (1), the three-dimensional image of liver to be analyzed is a liver MRI three-dimensional image.
3. The method for quantitatively analyzing the three-dimensional structural features of hepatic vein blood vessels according to claim 1, wherein in the step (1), the step of obtaining the segmented regions of the vein blood vessels of the three-dimensional image of the liver to be analyzed specifically comprises:
(11) and inputting the liver three-dimensional image to be analyzed into a pre-trained vein vessel segmentation model, and outputting the segmentation region of the vein vessel.
4. The method for quantitatively analyzing the three-dimensional structural features of hepatic vein blood vessels as claimed in claim 3, wherein in the step (11), the pre-trained vein blood vessel segmentation model is obtained by the following steps:
(a) acquiring a sample liver three-dimensional image, and acquiring venous vessel labeling data of the sample liver three-dimensional image;
(b) and taking the sample liver three-dimensional image and the vein vessel marking data as a training set, and carrying out deep learning training on the vein vessel segmentation model iteration to obtain the pre-trained vein vessel segmentation model.
5. The method for quantitative analysis of three-dimensional structural features of hepatic vein vessels according to claim 4, wherein in the step (a), the step of obtaining the three-dimensional image of the sample liver specifically comprises:
(a1) preprocessing an original sample liver three-dimensional image to obtain a newly-added sample liver three-dimensional image, wherein the preprocessing comprises one or more of histogram equalization, rotation, noise addition, distortion and mirror image;
(a2) and merging the original sample liver three-dimensional image and the newly-added sample liver three-dimensional image to obtain the sample liver three-dimensional image.
6. The method for quantitatively analyzing the three-dimensional structural features of hepatic vein blood vessels according to claim 4, wherein in the step (b), the deep learning network adopted by the vein blood vessel segmentation model is a fully supervised segmentation network based on a combination of U-Net and a mixed domain attention mechanism.
7. The quantitative analysis method for three-dimensional structural features of hepatic vein vessels according to claim 1, wherein in the step (3), the vessel segmentation process is performed on the vein vessels in the segmented region according to the topological morphology of the central line of the vein vessels and the directional information of the vein vessels in the three-dimensional space in the segmented region.
8. A quantitative analysis device for three-dimensional structural features of hepatic vein blood vessels is characterized by comprising:
the vein vessel segmentation module is used for acquiring a segmentation area of a vein vessel of a liver three-dimensional image to be analyzed;
a central line calculating module which is connected with the vein vessel segmentation module in a signal mode and is used for calculating the central line of the vein vessel in the segmentation region so as to obtain the central line of the vein vessel;
the blood vessel segmentation processing module is in signal connection with the central line calculation module and is used for performing blood vessel segmentation processing on the vein blood vessels in the segmentation region, and dividing the vein blood vessels into a left hepatic vein, a middle hepatic vein, a right hepatic vein, a left hepatic portal vein and a right hepatic portal vein;
a three-dimensional structural feature quantitative analysis module in signal connection with the blood vessel segmentation processing module and used for carrying out quantitative analysis on the three-dimensional structural feature of the vein blood vessel according to the central line of the left hepatic vein, the central line of the middle hepatic vein, the central line of the right hepatic vein, the central line of the left hepatic vein, the central line of the right hepatic vein and a vein model reconstructed by using a three-dimensional surface reconstruction algorithm in the segmentation area, wherein parameter indexes of the quantitative analysis comprise one or more of a blood vessel morphology characterization index, a blood vessel branch characterization index and a blood vessel grid characterization index, the blood vessel morphology characterization index comprises one or more of a blood vessel length, a blood vessel tortuosity, a blood vessel diameter, a blood vessel volume and a volume percentage, and the blood vessel branch characterization index comprises a blood vessel branch number, a blood vessel bifurcation angle, a blood vessel root node number, a blood, The number of the end nodes of the blood vessel and the number of the branch nodes of the blood vessel are one or more, and the characterization index of the blood vessel grid comprises the distribution density of the blood vessel.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program implements a method for quantitative analysis of three-dimensional structural features of hepatic venous vessels according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements a method for quantitative analysis of three-dimensional structural features of hepatic venous blood vessels according to any one of claims 1 to 7.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110321994.7A CN112907577A (en) | 2021-03-25 | 2021-03-25 | Quantitative analysis method, device, computer equipment and storage medium for three-dimensional structural characteristics of hepatic vein blood vessels |
CN202110807559.5A CN113436185B (en) | 2021-03-25 | 2021-07-16 | Liver vein blood vessel three-dimensional structure characteristic quantitative analysis method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110321994.7A CN112907577A (en) | 2021-03-25 | 2021-03-25 | Quantitative analysis method, device, computer equipment and storage medium for three-dimensional structural characteristics of hepatic vein blood vessels |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112907577A true CN112907577A (en) | 2021-06-04 |
Family
ID=76106557
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110321994.7A Withdrawn CN112907577A (en) | 2021-03-25 | 2021-03-25 | Quantitative analysis method, device, computer equipment and storage medium for three-dimensional structural characteristics of hepatic vein blood vessels |
CN202110807559.5A Active CN113436185B (en) | 2021-03-25 | 2021-07-16 | Liver vein blood vessel three-dimensional structure characteristic quantitative analysis method, device, computer equipment and storage medium |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110807559.5A Active CN113436185B (en) | 2021-03-25 | 2021-07-16 | Liver vein blood vessel three-dimensional structure characteristic quantitative analysis method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (2) | CN112907577A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114010216A (en) * | 2021-12-20 | 2022-02-08 | 重庆医科大学附属第一医院 | Airway morphological parameter quantitative acquisition method and device and airway stent design method |
CN115170912A (en) * | 2022-09-08 | 2022-10-11 | 北京鹰瞳科技发展股份有限公司 | Method for training image processing model, method for generating image and related product |
CN115690411A (en) * | 2022-09-22 | 2023-02-03 | 推想医疗科技股份有限公司 | Liver segment segmentation method and device, electronic equipment and storage medium |
CN117011242A (en) * | 2023-07-10 | 2023-11-07 | 珠海市人民医院 | Method and system for predicting hepatic encephalopathy after internal portal bypass operation through jugular vein |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101425186B (en) * | 2008-11-17 | 2012-03-28 | 华中科技大学 | Liver subsection method based on CT image and system thereof |
CN102048550B (en) * | 2009-11-02 | 2013-07-17 | 上海交通大学医学院附属仁济医院 | Method for automatically generating liver 3D (three-dimensional) image and accurately positioning liver vascular domination region |
CN104657624A (en) * | 2015-03-18 | 2015-05-27 | 汪艳 | Full-quantitative analysis method for liver cirrhosis |
CN106446360B (en) * | 2016-09-08 | 2019-07-09 | 天津大学 | Defects with skull scaffold for vascular tissue engineering parameterization design method based on Micro-CT scanning measurement |
CN110176004A (en) * | 2019-04-24 | 2019-08-27 | 艾瑞迈迪科技石家庄有限公司 | A kind of liver subsection method and system |
CN110310726B (en) * | 2019-08-12 | 2024-03-26 | 杨晓鹏 | Liver segment resection auxiliary system based on medical image three-dimensional reconstruction and holographic display |
CN111462047B (en) * | 2020-03-06 | 2024-03-12 | 深圳睿心智能医疗科技有限公司 | Vascular parameter measurement method, vascular parameter measurement device, vascular parameter measurement computer device and vascular parameter measurement storage medium |
CN111640124B (en) * | 2020-05-25 | 2023-06-02 | 浙江同花顺智能科技有限公司 | Blood vessel extraction method, device, equipment and storage medium |
CN113052785A (en) * | 2021-03-25 | 2021-06-29 | 上海志御软件信息有限公司 | Method and device for constructing automatic liver segmentation model based on deep learning, computer equipment and storage medium |
-
2021
- 2021-03-25 CN CN202110321994.7A patent/CN112907577A/en not_active Withdrawn
- 2021-07-16 CN CN202110807559.5A patent/CN113436185B/en active Active
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114010216A (en) * | 2021-12-20 | 2022-02-08 | 重庆医科大学附属第一医院 | Airway morphological parameter quantitative acquisition method and device and airway stent design method |
CN115170912A (en) * | 2022-09-08 | 2022-10-11 | 北京鹰瞳科技发展股份有限公司 | Method for training image processing model, method for generating image and related product |
CN115690411A (en) * | 2022-09-22 | 2023-02-03 | 推想医疗科技股份有限公司 | Liver segment segmentation method and device, electronic equipment and storage medium |
CN117011242A (en) * | 2023-07-10 | 2023-11-07 | 珠海市人民医院 | Method and system for predicting hepatic encephalopathy after internal portal bypass operation through jugular vein |
CN117011242B (en) * | 2023-07-10 | 2024-05-14 | 珠海市人民医院 | Method and system for predicting hepatic encephalopathy after internal portal bypass operation through jugular vein |
Also Published As
Publication number | Publication date |
---|---|
CN113436185A (en) | 2021-09-24 |
CN113436185B (en) | 2022-11-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113436185B (en) | Liver vein blood vessel three-dimensional structure characteristic quantitative analysis method, device, computer equipment and storage medium | |
US10762637B2 (en) | Vascular segmentation using fully convolutional and recurrent neural networks | |
US9959615B2 (en) | System and method for automatic pulmonary embolism detection | |
US9836849B2 (en) | Method for the autonomous image segmentation of flow systems | |
CN104217418B (en) | The segmentation of calcification blood vessel | |
US7773791B2 (en) | Analyzing lesions in a medical digital image | |
CN110689543A (en) | Improved convolutional neural network brain tumor image segmentation method based on attention mechanism | |
CN107563983A (en) | Image processing method and medical imaging devices | |
JP2013537445A (en) | Visualization adapted for direct use by physicians | |
CN111462047A (en) | Blood vessel parameter measuring method, blood vessel parameter measuring device, computer equipment and storage medium | |
CN112308846B (en) | Blood vessel segmentation method and device and electronic equipment | |
CN113034507A (en) | CCTA image-based coronary artery three-dimensional segmentation method | |
CN116309571B (en) | Three-dimensional cerebrovascular segmentation method and device based on semi-supervised learning | |
US20210020304A1 (en) | Systems and methods for generating classifying and quantitative analysis reports of aneurysms from medical image data | |
CN111178420A (en) | Coronary segment labeling method and system on two-dimensional contrast image | |
Damseh et al. | Laplacian flow dynamics on geometric graphs for anatomical modeling of cerebrovascular networks | |
CN113674291A (en) | Full-type aortic dissection real-false lumen image segmentation method and system | |
Khan et al. | Three-dimensional morphometric analysis of the renal vasculature | |
WO2015150320A1 (en) | Segmentation of tubular organ structures | |
CN114782302A (en) | System and method for predicting physiologically relevant parameters based on medical images | |
CN112651960A (en) | Image processing method, device, equipment and storage medium | |
EP3891704A1 (en) | Systems and methods for quantifying vessel features in ultrasound doppler images | |
WO2019190641A1 (en) | System and method for evaluation of dynamic data | |
CN114880960A (en) | Method for evaluating radioactive embolism dose and injection position based on fluid dynamics | |
CN117078711A (en) | Medical image segmentation method, system, electronic device 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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210604 |
|
WW01 | Invention patent application withdrawn after publication |