CN111161240B - Blood vessel classification method, apparatus, computer device, and readable storage medium - Google Patents
Blood vessel classification method, apparatus, computer device, and readable storage medium Download PDFInfo
- Publication number
- CN111161240B CN111161240B CN201911380101.5A CN201911380101A CN111161240B CN 111161240 B CN111161240 B CN 111161240B CN 201911380101 A CN201911380101 A CN 201911380101A CN 111161240 B CN111161240 B CN 111161240B
- Authority
- CN
- China
- Prior art keywords
- image
- blood vessel
- vessel
- angiography
- angiographic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 210000004204 blood vessel Anatomy 0.000 title claims abstract description 327
- 238000000034 method Methods 0.000 title claims abstract description 66
- 230000011218 segmentation Effects 0.000 claims abstract description 191
- 238000002583 angiography Methods 0.000 claims abstract description 161
- 238000013145 classification model Methods 0.000 claims abstract description 115
- 238000012549 training Methods 0.000 claims abstract description 42
- 238000004590 computer program Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 11
- 230000004927 fusion Effects 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 18
- 238000010968 computed tomography angiography Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 11
- 230000000694 effects Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 230000000903 blocking effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000002591 computed tomography Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- XCWPUUGSGHNIDZ-UHFFFAOYSA-N Oxypertine Chemical compound C1=2C=C(OC)C(OC)=CC=2NC(C)=C1CCN(CC1)CCN1C1=CC=CC=C1 XCWPUUGSGHNIDZ-UHFFFAOYSA-N 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 210000002376 aorta thoracic Anatomy 0.000 description 1
- 210000001841 basilar artery Anatomy 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 210000002168 brachiocephalic trunk Anatomy 0.000 description 1
- 210000001168 carotid artery common Anatomy 0.000 description 1
- 210000004004 carotid artery internal Anatomy 0.000 description 1
- 230000002490 cerebral effect Effects 0.000 description 1
- 239000002872 contrast media Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 208000019553 vascular disease Diseases 0.000 description 1
- 210000002385 vertebral artery Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—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
- G06F18/24—Classification techniques
-
- 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
- G06T2207/10—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
- G06T2207/20—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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The present invention relates to a blood vessel classification method, a computer device and a readable storage medium, the method comprising: acquiring an angiographic image to be processed and a vessel segmentation image corresponding to the angiographic image; inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in an angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel. In the method, the computer equipment can accurately classify the blood vessels in the input blood vessel segmentation image through the pre-trained first blood vessel classification model, and the blood vessel segmentation image corresponds to the angiography image, so that the classification result of the blood vessels in the angiography image can be accurately obtained, and the accuracy of the classification result of the blood vessels in the angiography image is improved.
Description
Technical Field
The present invention relates to the field of medical image processing, and in particular, to a blood vessel classification method, apparatus, computer device, and readable storage medium.
Background
Computed tomography angiography (Computed Tomography Angiography, CTA) is a technique for contrast agent injection prior to computed tomography (Computed Tomography, CT) scanning to visualize and enhance blood vessels. The CTA is applied to different parts of a human body to obtain blood vessel enhanced images of different parts, such as a head and neck CTA image, a coronary CTA image and the like. At present, CTA technology has become an important radiation technology for clinical disease diagnosis, and has important diagnostic significance in head and neck vascular disease diagnosis.
When analyzing blood vessels in the obtained CTA image, blood vessel classification can remove blood vessels which are not concerned, so that the working pressure of doctors is reduced, and diseases caused by good blood vessels at different positions are different, so that the blood vessel classification is an important step in blood vessel analysis. In the conventional technology, the classification of blood vessels is mainly performed by locating key points in CTA images of blood vessels already segmented, and tracking the blood vessels through the located key points.
However, the conventional method of classifying blood vessels has a problem of low classification accuracy.
Disclosure of Invention
Based on this, it is necessary to provide a blood vessel classification method, apparatus, computer device and readable storage medium for solving the problem of low classification accuracy in the conventional blood vessel classification method.
In a first aspect, an embodiment of the present invention provides a blood vessel classification method, including:
acquiring an angiographic image to be processed and a blood vessel segmentation image corresponding to the angiographic image;
inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
In one embodiment, the method further comprises:
inputting the angiography image and the blood vessel segmentation image into a preset second blood vessel classification model to obtain a blood vessel classification result in the angiography image; the second blood vessel classification model is obtained by training according to a sample blood vessel angiography image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample blood vessel angiography image.
In one embodiment, the method further comprises:
determining position information of pixel points in the angiographic image aiming at the angiographic image;
Inputting the angiography image, the blood vessel segmentation image and the position information into a preset third blood vessel classification model to obtain a blood vessel classification result in the angiography image; the third blood vessel classification model is obtained by training according to a training sample with blood vessel classification labels and blood vessel position information.
In one embodiment, the inputting the angiographic image, the vessel segmentation image and the position information into a preset third vessel classification model to obtain a classification result of vessels in the angiographic image includes:
intercepting the angiographic image, the blood vessel segmentation image and the position information respectively to obtain image blocks of the angiographic image, the image blocks of the blood vessel segmentation image and the block information of the position information; wherein the intercepting order of the angiographic image, the vessel segmentation image and the position information is consistent;
and inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model to obtain a classification result of vessels in the angiography image.
In one embodiment, the inputting the image block of the angiographic image, the image block of the vessel segmented image, and the segmented information of the position information into the third vessel classification model, to obtain a classification result of a vessel in the angiographic image, includes:
inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model to obtain a vessel classification result in the image blocks of the angiography image;
and according to the intercepting sequence, performing splicing processing on the classification results of the blood vessels in the image blocks of the angiography image to obtain the classification results of the blood vessels in the angiography image.
In one embodiment, the inputting the image block of the angiographic image, the image block of the vessel segmented image, and the segmented information of the position information into the third vessel classification model, to obtain a classification result of the vessel in the image block of the angiographic image, includes:
inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model through different channels, and extracting the characteristics of the image blocks of the angiography image, the characteristics of the image blocks of the vessel segmentation image and the characteristics of the segmentation information;
Feature fusion is carried out on the features of the image blocks of the angiography image, the features of the image blocks of the vessel segmentation image and the features of the segmentation information of the position information, so that fused features are obtained;
and classifying blood vessels in the image block of the angiography image according to the fused characteristics to obtain a classification result of the blood vessels in the image block of the angiography image.
In one embodiment, the determining, for the angiographic image, the location information of the pixel point in the angiographic image includes:
determining the relative position of each pixel point in the angiography image and a preset origin of coordinates, wherein the origin of coordinates is positioned at the preset position of the angiography image;
and determining the position information according to the relative position of each pixel point.
In one embodiment, the vessel classification model is a three-dimensional model.
In a second aspect, embodiments of the present invention provide a blood vessel classification device, the device comprising:
the first acquisition module is used for acquiring an angiography image to be processed and a blood vessel segmentation image corresponding to the angiography image;
the second acquisition module is used for inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
acquiring an angiographic image to be processed and a blood vessel segmentation image corresponding to the angiographic image;
inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
In a fourth aspect, embodiments of the present invention provide a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
acquiring an angiographic image to be processed and a blood vessel segmentation image corresponding to the angiographic image;
inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
In the blood vessel classification method, the blood vessel classification device, the computer equipment and the readable storage medium provided by the embodiment, the computer equipment acquires an angiography image to be processed and a blood vessel segmentation image corresponding to the angiography image, and inputs the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel. In the method, the computer equipment can accurately classify the blood vessels in the input blood vessel segmentation image through the pre-trained first blood vessel classification model, and the blood vessel segmentation image corresponds to the angiography image, so that the classification result of the blood vessels in the angiography image can be accurately obtained, and the accuracy of the classification result of the blood vessels in the angiography image is improved.
Drawings
FIG. 1 is a schematic diagram of an internal structure of a computer device according to one embodiment;
FIG. 2 is a flow chart of a method of classifying blood vessels according to one embodiment;
FIG. 3 is a flow chart of a blood vessel classification method according to another embodiment;
FIG. 4 is a flow chart of a blood vessel classification method according to another embodiment;
FIG. 5 is a flow chart of a blood vessel classification method according to another embodiment;
fig. 6 is a schematic structural diagram of a blood vessel classifying device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The blood vessel classification method provided by the embodiment of the application can be applied to the computer equipment shown in the figure 1. The computer device comprises a processor, a memory, and a computer program stored in the memory, wherein the processor is connected through a system bus, and when executing the computer program, the processor can execute the steps of the method embodiments described below. Optionally, the computer device may further comprise a network interface, a display screen and an input means. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, which stores an operating system and a computer program, an internal memory. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. Optionally, the computer device may be a server, may be a personal computer, may also be a personal digital assistant, may also be other terminal devices, such as a tablet computer, a mobile phone, etc., and may also be a cloud or remote server.
In the conventional technology, the blood vessel is classified mainly by locating key points in CTA images of the segmented blood vessel and tracking the blood vessel through the located key points, but if the located key points are not accurate enough, the accuracy of classifying the blood vessel is affected. The blood vessel classification method, the computer equipment and the readable storage medium provided by the embodiment of the application aim to solve the technical problems.
The following describes the technical scheme of the present invention and how the technical scheme of the present invention solves the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.
Fig. 2 is a flow chart of a blood vessel classification method according to an embodiment. The embodiment relates to a specific implementation process of inputting a blood vessel segmentation image corresponding to an angiography image to be processed into a preset first blood vessel classification model by computer equipment to obtain a blood vessel classification result in the angiography image. As shown in fig. 2, the method may include:
s201, acquiring a blood vessel segmentation image corresponding to an angiographic image to be processed.
The angiographic image to be processed is a blood vessel enhancement image obtained by applying a computed tomography angiographic (Computed Tomography Angiography, CTA) technology to a certain part of a human body. The blood vessel segmentation image corresponding to the angiography image is an image obtained by segmenting blood vessels in the angiography image through a preset segmentation method, for example, the blood vessels in the angiography image can be segmented through a blood vessel segmentation model based on U-Net or a blood vessel segmentation model based on high-resolution network HRNet to obtain the blood vessel segmentation image corresponding to the angiography image. It will be appreciated that the application of CTA techniques to different parts of the human body will result in different angiographic images. For example, the angiographic image to be processed may be an angiographic image of the head and neck of a human body. Optionally, the computer device may acquire the angiographic image to be processed from a PACS (Picture Archiving and Communication Systems, image archiving and communication system) server, or may acquire the angiographic image to be processed from the medical imaging device in real time, and then acquire the segmented image of the blood vessel corresponding to the angiographic image by using a preset segmentation method. Optionally, after the computer device obtains an angiographic image to be processed and a vessel segmentation image corresponding to the angiographic image, the angiographic image and the vessel segmentation image may be preprocessed, where the preprocessing includes converting a data format of the angiographic image and the vessel segmentation image, converting the angiographic image and the vessel segmentation image from a digital imaging and communications (Digital Imaging and Communications in Medicine, DICOM) format into a NIFTY format, then respectively setting window levels of window frames for the converted angiographic image and the vessel segmentation image HU values to clip the image, and then performing resampling processing and normalization processing on the clipped image. Alternatively, when the image is cut out, the window frame may be set to 600 and the window level may be set to 300. Alternatively, the cropped image may be resampled to a sampling interval of 0.45×0.45×0.7.
S202, inputting a blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in an angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
Specifically, the computer equipment inputs the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in the angiography image. Taking an angiographic image to be processed as an angiographic image of a head and a neck as an example, the head and neck comprises a plurality of blood vessels, each blood vessel has a medical name, but a clinician does not pay attention to all blood vessels in the blood vessels of the head and neck, and the blood vessels mainly paid attention to include blood vessels such as aortic arch, brachiocephalic trunk, common carotid artery, internal carotid artery, vertebral artery, basilar artery and cerebral arterial loop, and the like.
The first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of the current blood vessel; that is, the training process of the first vessel classification model includes: obtaining a sample angiography image and a sample vessel segmentation image corresponding to the sample angiography image, labeling the names of vessels in the sample vessel segmentation image to obtain a sample vessel segmentation image with vessel classification labels, inputting the sample vessel segmentation image into a preset initial first vessel segmentation model to obtain a sample classification result of the vessels in the sample angiography image, comparing the obtained sample classification result with the vessel classification labels in the sample vessel segmentation image to obtain the value of a loss function of the initial first vessel segmentation model, and training the initial first vessel segmentation model according to the value of the loss function of the initial first vessel segmentation model to obtain a first vessel classification model. Optionally, the first vessel classification model is a three-dimensional model. Alternatively, the first vessel classification model may be a U-Net network model or a V-Net network model. Alternatively, the loss function of the initial first vessel classification model may be a Dice loss function, a Cross Entropy (CE) loss function, or another type of loss function. Optionally, when adjusting the network parameters of the initial first vessel classification model, an Adam adaptive optimizer may be used, and the update amplitude of the network parameters may be automatically adjusted according to the training state. Optionally, a In the training of the initial first blood vessel classification model, considering the size of the GPU video memory of the computer device, the problem that the memory overflows may exist when the whole blood vessel segmentation image is processed, so that the computer device may sequentially intercept blood vessel segmentation image blocks with the same size (for example, intercept 128 3 The method comprises the steps of inputting each blood vessel segmentation image block into an initial first blood vessel classification model to obtain a blood vessel classification result in each blood vessel segmentation image block, and then splicing the blood vessel classification results in each blood vessel segmentation image block according to a preset interception sequence to obtain a blood vessel classification result in the whole blood vessel segmentation image, so as to obtain a blood vessel classification result in an angiography image.
In this embodiment, the computer device can accurately classify the blood vessel in the input blood vessel segmentation image through the pre-trained first blood vessel classification model, and since the blood vessel segmentation image corresponds to the angiography image, the classification result of the blood vessel in the angiography image can be accurately obtained, and the accuracy of the classification result of the blood vessel in the angiography image is improved.
On the basis of the above embodiment, as an optional implementation manner, the above method further includes: inputting the angiography image and the blood vessel segmentation image into a preset second blood vessel classification model to obtain a blood vessel classification result in the angiography image; the second blood vessel classification model is obtained by training according to a sample angiography image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample angiography image.
Specifically, the computer equipment inputs the angiographic image and the blood vessel segmentation image into a preset second blood vessel classification model to obtain a blood vessel classification result in the angiographic image, wherein the second blood vessel classification model is obtained by training according to the angiographic image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample angiographic image. The classification result of the blood vessel in the angiographic image obtained in the present embodiment may refer to the description in S202, and will not be described herein. Can be used forIt is understood that the training process of the second vessel classification model includes: obtaining a sample angiography image and a sample vessel segmentation image corresponding to the sample angiography image, labeling names of vessels in the sample angiography image and the sample vessel segmentation image to obtain a sample angiography image and a sample vessel segmentation image with vessel classification labels, inputting the sample angiography image and the sample vessel segmentation image into a preset initial second vessel segmentation model to obtain a sample classification result of vessels in the sample angiography image, comparing the obtained sample classification result with the vessel classification labels in the sample angiography image and the sample vessel segmentation image to obtain a value of a loss function of the initial second vessel segmentation model, and training the initial second vessel segmentation model according to the value of the loss function of the initial second vessel segmentation model to obtain a second vessel classification model. Optionally, the second vessel classification model is a three-dimensional model. Alternatively, the second vessel classification model may be a U-Net network model or a V-Net network model. Alternatively, the loss function of the initial second vessel classification model may be a Dice loss function, a Cross Entropy (CE) loss function, or other type of loss function. Optionally, when adjusting the network parameters of the initial second vessel classification model, an Adam adaptive optimizer may be used, and the update amplitude of the network parameters may be automatically adjusted according to the training state. Optionally, during training the initial second vessel classification model, considering the size of the GPU video memory of the computer device, the problem that memory overflow may exist in processing the whole angiographic image and the vessel segmentation image may be considered, so that the computer device may sequentially cut angiographic image blocks and vessel segmentation image blocks of the same size according to a preset cutting order for the angiographic image and the vessel segmentation image (for example, cut 128 3 The image blocks with the sizes) input the angiography image blocks and the blood vessel segmentation image blocks into an initial second blood vessel classification model to obtain the classification result of the blood vessels in each angiography image block, and then splicing the classification result of the blood vessels in each angiography image block according to a preset interception sequence to obtain the classification result of the blood vessels in the whole angiography image.
In this embodiment, when segmenting the blood vessel in the angiography image, the edge information loss in the angiography image causes the obtained segmentation image of the blood vessel to be inaccurate, and the angiography image and the corresponding segmentation image of the blood vessel are input into the preset second blood vessel classification model, so that the blood vessel in the angiography image can be accurately classified by combining the angiography image and the segmentation image of the blood vessel, the name of the blood vessel in the angiography image can be accurately obtained, the problem that the classification result of the blood vessel in the angiography image is inaccurate due to the inaccurate segmentation image of the blood vessel is avoided, and the accuracy of the classification result of the blood vessel in the obtained angiography image is improved.
Fig. 3 is a flow chart of a blood vessel classifying method according to another embodiment. The embodiment relates to a specific implementation process of obtaining a blood vessel classification result in an angiography image by computer equipment according to the angiography image, a corresponding blood vessel segmentation image and position information of pixel points in the angiography image. As shown in fig. 3, on the basis of the above embodiment, as an alternative implementation manner, the above method further includes:
S301, determining position information of a pixel point in an angiographic image with respect to the angiographic image.
Specifically, the computer device determines, for the angiographic image, positional information of pixel points in the angiographic image. Optionally, the computer device may determine a relative position between each pixel point in the angiographic image and a preset origin of coordinates, and determine, according to the relative position of each pixel point in the angiographic image, position information of the pixel points in the angiographic image, where the preset origin of coordinates is located at a preset position of the angiographic image, optionally, taking the angiographic image of the head and neck as an example, the computer device may determine the preset origin of coordinates by the following procedure: the angiographic image is set as a bone window, and the largest circumscribed rectangle wrapping the head and neck is selected, so that the right upper part (generally the top of the cranium) of the determined largest circumscribed rectangle can be determined as a preset origin of coordinates.
S302, inputting an angiography image, a blood vessel segmentation image and position information into a preset third blood vessel classification model to obtain a blood vessel classification result in the angiography image; the third blood vessel classification model is obtained by training according to a training sample with blood vessel classification labels and blood vessel position information.
Specifically, the computer equipment inputs the angiography image, a blood vessel segmentation image corresponding to the angiography image and position information of pixel points in the angiography image into a preset third blood vessel classification model to obtain a blood vessel classification result in the angiography image. The third blood vessel classification model is obtained by training according to training samples with blood vessel classification labels and blood vessel position information, namely the third blood vessel classification model is obtained by training a preset initial third blood vessel classification model according to sample angiography images with blood vessel classification labels, sample blood vessel segmentation images with blood vessel classification labels and sample position information of pixel points in the sample angiography images. The classification result of the blood vessel in the angiographic image obtained in the present embodiment may refer to the description in S202, and will not be described herein. It will be appreciated that the training process for the third vessel classification model includes: acquiring a sample angiography image and a sample vessel segmentation image corresponding to the sample angiography image, determining sample position information of pixel points in the sample angiography image, labeling names of vessels in the sample angiography image and the sample vessel segmentation image to obtain the sample angiography image and the sample vessel segmentation image with vessel classification labels, inputting the sample position information of the pixel points in the sample angiography image, the sample vessel segmentation image and the sample vessel segmentation image into a preset initial third vessel segmentation model to obtain a sample classification result of the vessels in the sample angiography image, comparing the obtained sample classification result with the vessel classification labels in the sample angiography image and the sample vessel segmentation image to obtain a value of a loss function of the initial third vessel segmentation model, and training the initial third vessel segmentation model according to the value of the loss function of the initial third vessel segmentation model to obtain a third vessel classification model. Optionally, a third The vessel classification model is a three-dimensional model. Optionally, the third vessel classification model may be a U-Net network model or a V-Net network model. It should be noted that, in the convolutional neural network, features from deep layers learn higher-level semantic information, while shallow-layer features retain abundant spatial structural details, so that it is very important to fuse information of different layers in order to accurately divide a target into different parts. When the third vessel classification model is a U-Net network model, the U-Net network can fuse multiple levels of features, but it is obviously disadvantageous to give the same weight to different channels, so an attention feature aggregation (Attention Feature Aggregation, AFA) module can be added to the U-Net network, and the AFA module can extract information-rich features and suppress indiscriminate features. Alternatively, the loss function of the initial third vessel classification model may be a Dice loss function, a Cross Entropy (CE) loss function, or another type of loss function. Optionally, when adjusting the network parameters of the initial third vessel classification model, an Adam adaptive optimizer may be used, and the update amplitude of the network parameters may be automatically adjusted according to the training state. Optionally, when training the initial third vessel classification model, considering the size of the GPU video memory of the computer device, the problem of memory overflow may occur in processing the whole angiographic image, the vessel segmentation image, and the sample position information of the pixel points in the angiographic image, so that the computer device may sequentially intercept the angiographic image blocks, the vessel segmentation image blocks, and the sample block information of the sample position information of the same size according to a preset interception order (e.g. intercept 128 3 The image blocks with the sizes), inputting the angiography image blocks, the blood vessel segmentation image blocks and the sample segmentation information into an initial third blood vessel classification model to obtain the classification result of the blood vessels in each angiography image block, and then splicing the classification result of the blood vessels in each angiography image block according to a preset interception sequence to obtain the classification result of the blood vessels in the whole angiography image.
In this embodiment, the computer device determines, for the angiographic image, the position information of the pixel points in the angiographic image, inputs the angiographic image, the vessel segmentation image, and the position information of the pixel points in the angiographic image into a preset third vessel classification model, where the third vessel classification model can accurately classify the vessels in the angiographic image according to the position information of the pixel points in the angiographic image, in combination with the angiographic image and the vessel segmentation image, thereby improving the accuracy of the classification result of the vessels in the angiographic image.
In some scenes, considering the limitation of GPU video memory, the neural network model cannot process the input whole image, the image needs to be intercepted, and the intercepted image blocks are input into the neural network model for processing. Fig. 4 is a flow chart of a blood vessel classifying method according to another embodiment. The embodiment relates to a specific implementation process of inputting position information of pixels in an angiography image, a vessel segmentation image and the angiography image into a third vessel classification model by computer equipment to obtain a vessel classification result in the angiography image. As shown in fig. 4, based on the above embodiment, as an alternative implementation manner, the step S302 includes:
S401, respectively intercepting an angiography image, a vessel segmentation image and position information to obtain image blocks of the angiography image, image blocks of the vessel segmentation image and block information of the position information; wherein, the interception sequence of the angiography image, the blood vessel segmentation image and the position information is consistent.
Specifically, the computer device performs a clipping process on the angiographic image, the vessel segmentation image and the position information of the pixel points in the angiographic image respectively to obtain an image block of the angiographic image, an image block of the vessel segmentation image and a block information of the position information of the pixel points in the angiographic image. The cutting order of the position information of the pixel points in the angiography image, the blood vessel segmentation image and the angiography image is consistent. Alternatively, the computer device may perform the clipping processing on the angiographic image, the vessel segmentation image, and the positional information of the pixel points in the angiographic image in order of clipping from left to right and from top to bottom.
S402, inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the block information of the position information into a third vessel classification model to obtain a vessel classification result in the angiography image.
Specifically, the computer equipment inputs the image block of the angiography image, the image block of the vessel segmentation image and the block information of the position information of the pixel points in the angiography image into a third vessel classification model to obtain the classification result of the vessels in the angiography image. Optionally, the computer device may input the image block of the angiographic image, the image block of the vessel segmented image, and the segmented information of the position information of the pixel point in the angiographic image into the third vessel classification model to obtain a classification result of the vessel in the image block of the angiographic image, and then, according to a sequence of interception performed on the position information of the vessel segmented image and the pixel point in the angiographic image, perform a stitching process on the classification result of the vessel in the image block of the angiographic image to obtain a classification result of the vessel in the angiographic image.
In this embodiment, the computer device performs the clipping processing on the angiographic image, the vessel segmentation image, and the position information of each pixel in the angiographic image according to the consistent clipping sequence, so as to obtain the image block of the angiographic image, the image block of the vessel segmentation image, and the block information of the position information of each pixel in the angiographic image, and inputs the image block of the angiographic image, the image block of the vessel segmentation image, and the block information of the position information of each pixel in the angiographic image into the third vessel classification model, so that the third vessel classification model can quickly obtain the classification result of the vessel in the angiographic image block, and further can quickly obtain the classification result of the vessel in the angiographic image according to the classification result of the vessel in the angiographic image block, thereby improving the efficiency of obtaining the classification result of the vessel in the angiographic image.
Fig. 5 is a flow chart of a blood vessel classifying method according to another embodiment. The embodiment relates to a specific implementation process of inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the block information of the position information of the pixel points in the angiography image into a third vessel classification model by computer equipment to obtain the vessel classification result in the image blocks of the angiography image. As shown in fig. 4, based on the above embodiment, as an alternative implementation manner, the classification result of the blood vessel in the image block from which the angiographic image is obtained includes:
s501, inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the blocking information of the position information into a third vessel classification model through different channels, and extracting the characteristics of the image blocks of the angiography image, the characteristics of the image blocks of the vessel segmentation image and the characteristics of the blocking information.
Specifically, the computer device inputs the image blocks of the angiography image, the image blocks of the vessel segmentation image and the blocking information of the position information of the pixel points in the angiography image into a third vessel classification model through different channels, and extracts the characteristics of the image blocks of the angiography image, the characteristics of the image blocks of the vessel segmentation image and the characteristics of the blocking information. Optionally, the third vessel classification model may extract the features of the image blocks of the feature vessel segmentation image and the features of the segmentation information of the image blocks of the angiographic image at the same time, or may sequentially extract the features of the image blocks of the feature vessel segmentation image and the features of the segmentation information of the image blocks of the angiographic image.
S502, carrying out feature fusion on the features of the image blocks of the angiography image, the features of the image blocks of the vessel segmentation image and the features of the segmentation information to obtain fused features.
Specifically, after the features of the image block of the angiographic image, the features of the image block of the vessel segmentation image and the features of the block information are extracted by the third vessel classification model, the features of the image block of the angiographic image, the features of the image block of the vessel segmentation image and the features of the block information are subjected to feature fusion through convolution operation, so that the fused features are obtained.
And S503, classifying blood vessels in the image block of the angiography image according to the fused characteristics to obtain a classification result of the blood vessels in the image block of the angiography image.
Specifically, the computer device classifies blood vessels in the image block of the angiography image according to the fused features to obtain a classification result of the blood vessels in the image block of the angiography image. Alternatively, the classification result of the blood vessel in the image block of the angiographic image may refer to the description of the classification result of the blood vessel in the angiographic image in S202, which is not described herein.
In this embodiment, the computer device inputs the image block of the angiographic image, the image block of the vessel segmented image, and the segmented information of the position information of the pixel point in the angiographic image into the third vessel classification model through different channels, so that the features of the image block of the angiographic image, the features of the image block of the vessel segmented image, and the features of the segmented information can be extracted quickly, the extracted features can be fused quickly, the fused features can be obtained, and further, the vessels in the image block of the angiographic image can be classified quickly according to the fused features, and the efficiency of obtaining the classification result of the vessels in the image block of the angiographic image is improved.
It should be understood that, although the steps in the flowcharts of fig. 2-5 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2-5 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily occur sequentially, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or steps.
Fig. 6 is a schematic structural diagram of a blood vessel classifying device according to an embodiment. As shown in fig. 6, the apparatus may include: a first acquisition module 10 and a second acquisition module 11.
Specifically, the first acquiring module 10 is configured to acquire an angiographic image to be processed and a vessel segmentation image corresponding to the angiographic image;
a second obtaining module 11, configured to input the blood vessel segmentation image into a preset first blood vessel classification model, and obtain a classification result of a blood vessel in the angiographic image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
Optionally, the vessel classification model is a three-dimensional model.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above apparatus further includes: and a third acquisition module.
Specifically, a third obtaining module is configured to input an angiographic image and a blood vessel segmentation image into a preset second blood vessel classification model to obtain a classification result of blood vessels in the angiographic image; the second blood vessel classification model is obtained by training according to a sample angiography image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample angiography image.
Optionally, the vessel classification model is a three-dimensional model.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above apparatus further includes: a determining module and a fourth obtaining module.
Specifically, a determining module is used for determining the position information of the pixel points in the angiographic image aiming at the angiographic image;
A fourth acquisition module, configured to input an angiographic image, a vessel segmentation image and position information into a preset third vessel classification model, so as to obtain a classification result of a vessel in the angiographic image; the third blood vessel classification model is obtained by training according to a training sample with blood vessel classification labels and blood vessel position information.
Optionally, the vessel classification model is a three-dimensional model.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the fourth obtaining module includes: an intercepting unit and an acquiring unit.
Specifically, the intercepting unit is used for intercepting the angiographic image, the blood vessel segmentation image and the position information respectively to obtain image blocks of the angiographic image, image blocks of the blood vessel segmentation image and block information of the position information; the angiography image, the blood vessel segmentation image and the position information are intercepted in the same sequence;
and the acquisition unit is used for inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the block information of the position information into the third vessel classification model to obtain the classification result of the vessels in the angiography image.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the obtaining unit is specifically configured to input, into a third blood vessel classification model, image blocks of the angiographic image, image blocks of the vessel segmentation image, and block information of the position information, to obtain a classification result of a vessel in the image blocks of the angiographic image; and according to the intercepting sequence, splicing the classification results of the blood vessels in the image blocks of the angiography image to obtain the classification results of the blood vessels in the angiography image.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the obtaining unit is specifically configured to input, through different channels, block information of an image block of the angiographic image, an image block of the vessel segmentation image, and position information into the third vessel classification model, and extract features of the image block of the angiographic image, features of the image block of the vessel segmentation image, and features of the block information; feature fusion is carried out on the features of the image blocks of the angiography image, the features of the image blocks of the vessel segmentation image and the features of the block information of the position information, so as to obtain fused features; and classifying the blood vessels in the image block of the angiography image according to the fused characteristics to obtain a classification result of the blood vessels in the image block of the angiography image.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above determination module includes a first determination unit and a second determination unit.
Specifically, a first determining unit is configured to determine a relative position between each pixel point in the angiographic image and a preset origin of coordinates, where the origin of coordinates is located at a preset position of the angiographic image;
and the second determining unit is used for determining the position information according to the relative position of each pixel point.
The blood vessel classifying device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
For specific limitations of the blood vessel classification device, reference may be made to the limitations of the blood vessel classification method hereinabove, and no further description is given here. The various modules in the blood vessel classification device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring an angiographic image to be processed and a vessel segmentation image corresponding to the angiographic image;
inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in an angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
The computer device provided in the foregoing embodiments has similar implementation principles and technical effects to those of the foregoing method embodiments, and will not be described herein in detail.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring an angiographic image to be processed and a vessel segmentation image corresponding to the angiographic image;
inputting the blood vessel segmentation image into a preset first blood vessel classification model to obtain a blood vessel classification result in an angiography image; the first blood vessel classification model is obtained by training according to a sample blood vessel segmentation image with a blood vessel classification label, and the blood vessel classification label is used for indicating the name of a current blood vessel.
The computer readable storage medium provided in the above embodiment has similar principle and technical effects to those of the above method embodiment, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (10)
1. A method of vessel classification, the method comprising:
acquiring an angiographic image to be processed and a blood vessel segmentation image corresponding to the angiographic image;
inputting the angiography image and the blood vessel segmentation image into a preset second blood vessel classification model to obtain a blood vessel classification result in the angiography image; the second blood vessel classification model is obtained by training according to a sample blood vessel angiography image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample blood vessel angiography image; the blood vessel classification label is used for indicating the name of the current blood vessel; or,
Determining position information of pixel points in the angiographic image aiming at the angiographic image;
inputting the angiography image, the blood vessel segmentation image and the position information into a preset third blood vessel classification model to obtain a blood vessel classification result in the angiography image; the third blood vessel classification model is obtained by training according to a training sample with blood vessel classification labels and blood vessel position information;
inputting the angiographic image, the vessel segmentation image and the position information into a preset third vessel classification model to obtain a vessel classification result in the angiographic image, wherein the method comprises the following steps:
intercepting the angiographic image, the blood vessel segmentation image and the position information respectively to obtain image blocks of the angiographic image, the image blocks of the blood vessel segmentation image and the block information of the position information; wherein the intercepting order of the angiographic image, the vessel segmentation image and the position information is consistent;
and inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model to obtain a classification result of vessels in the angiography image.
2. The method of claim 1, wherein the third vessel classification model is constructed based on a U-Net network and attention feature aggregation module.
3. The method according to claim 1, wherein inputting the image block of the angiographic image, the image block of the vessel segmentation image, and the segmentation information of the position information into the third vessel classification model, to obtain a classification result of vessels in the angiographic image, comprises:
inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model to obtain a vessel classification result in the image blocks of the angiography image;
and according to the intercepting sequence, performing splicing processing on the classification results of the blood vessels in the image blocks of the angiography image to obtain the classification results of the blood vessels in the angiography image.
4. A method according to claim 3, wherein inputting the image block of the angiographic image, the image block of the vessel segmented image, and the segmented information of the position information into the third vessel classification model, to obtain a classification result of vessels in the image block of the angiographic image, comprises:
Inputting the image blocks of the angiography image, the image blocks of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model through different channels, and extracting the characteristics of the image blocks of the angiography image, the characteristics of the image blocks of the vessel segmentation image and the characteristics of the segmentation information;
feature fusion is carried out on the features of the image blocks of the angiography image, the features of the image blocks of the vessel segmentation image and the features of the segmentation information, so that fused features are obtained;
and classifying blood vessels in the image block of the angiography image according to the fused characteristics to obtain a classification result of the blood vessels in the image block of the angiography image.
5. The method of claim 1, wherein the determining location information of pixels in the angiographic image for the angiographic image comprises:
determining the relative position of each pixel point in the angiography image and a preset origin of coordinates, wherein the origin of coordinates is positioned at the preset position of the angiography image;
and determining the position information according to the relative position of each pixel point.
6. The method of any one of claims 1-5, wherein the vessel classification model is a three-dimensional model.
7. A blood vessel classification device, the device comprising:
the first acquisition module is used for acquiring an angiography image to be processed and a blood vessel segmentation image corresponding to the angiography image;
a third acquisition module, configured to input the angiographic image and the vessel segmentation image into a preset second vessel classification model, so as to obtain a classification result of a vessel in the angiographic image; the second blood vessel classification model is obtained by training according to a sample blood vessel angiography image with a blood vessel classification label and a sample blood vessel segmentation image corresponding to the sample blood vessel angiography image; the blood vessel classification label is used for indicating the name of the current blood vessel; or,
a determining module, configured to determine, for the angiographic image, location information of a pixel point in the angiographic image;
a fourth obtaining module, configured to input the angiographic image, the vessel segmentation image and the position information into a preset third vessel classification model, so as to obtain a classification result of a vessel in the angiographic image; the third blood vessel classification model is obtained by training according to a training sample with blood vessel classification labels and blood vessel position information;
The fourth acquisition module includes:
the intercepting unit is used for intercepting the angiographic image, the blood vessel segmentation image and the position information respectively to obtain image blocks of the angiographic image, image blocks of the blood vessel segmentation image and block information of the position information; wherein the intercepting order of the angiographic image, the vessel segmentation image and the position information is consistent;
and the acquisition unit is used for inputting the image block of the angiography image, the image block of the vessel segmentation image and the segmentation information of the position information into the third vessel classification model to obtain a classification result of vessels in the angiography image.
8. The apparatus according to claim 7, wherein the acquiring unit is specifically configured to input the image block of the angiographic image, the image block of the vessel segmentation image, and the segmentation information of the position information into the third vessel classification model, so as to obtain a classification result of vessels in the image block of the angiographic image; and according to the intercepting sequence, performing splicing processing on the classification results of the blood vessels in the image blocks of the angiography image to obtain the classification results of the blood vessels in the angiography image.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380101.5A CN111161240B (en) | 2019-12-27 | 2019-12-27 | Blood vessel classification method, apparatus, computer device, and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380101.5A CN111161240B (en) | 2019-12-27 | 2019-12-27 | Blood vessel classification method, apparatus, computer device, and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111161240A CN111161240A (en) | 2020-05-15 |
| CN111161240B true CN111161240B (en) | 2024-03-05 |
Family
ID=70558692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911380101.5A Active CN111161240B (en) | 2019-12-27 | 2019-12-27 | Blood vessel classification method, apparatus, computer device, and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111161240B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111882527B (en) * | 2020-07-14 | 2021-12-21 | 上海商汤智能科技有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN115620062A (en) * | 2022-10-24 | 2023-01-17 | 推想医疗科技股份有限公司 | Liver blood vessel classification method, device, equipment and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545269A (en) * | 2016-06-23 | 2018-01-05 | 西门子保健有限责任公司 | The method and system of vascular diseases detection is carried out using recurrent neural network |
| CN107993220A (en) * | 2017-11-14 | 2018-05-04 | 北京理工大学 | The extracting method and device of x-ray imaging image medium vessels structure |
| CN109410188A (en) * | 2017-10-13 | 2019-03-01 | 北京昆仑医云科技有限公司 | System and method for being split to medical image |
| CN109671076A (en) * | 2018-12-20 | 2019-04-23 | 上海联影智能医疗科技有限公司 | Blood vessel segmentation method, apparatus, electronic equipment and storage medium |
| CN109741344A (en) * | 2018-12-28 | 2019-05-10 | 上海联影智能医疗科技有限公司 | Blood vessel segmentation method, apparatus, medical imaging equipment and storage medium |
| CN109816666A (en) * | 2019-01-04 | 2019-05-28 | 三峡大学 | Symmetrical full convolutional neural networks model building method, eye fundus image blood vessel segmentation method, apparatus, computer equipment and storage medium |
| CN110298844A (en) * | 2019-06-17 | 2019-10-01 | 艾瑞迈迪科技石家庄有限公司 | X-ray contrastographic picture blood vessel segmentation and recognition methods and device |
| CN110348541A (en) * | 2019-05-10 | 2019-10-18 | 腾讯医疗健康(深圳)有限公司 | Optical fundus blood vessel image classification method, device, equipment and storage medium |
| CN110443813A (en) * | 2019-07-29 | 2019-11-12 | 腾讯医疗健康(深圳)有限公司 | Blood vessel, the dividing method of eye fundus image, device, equipment and readable storage medium storing program for executing |
| CN110517279A (en) * | 2019-09-20 | 2019-11-29 | 北京深睿博联科技有限责任公司 | Neck vessel centerline extracting method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190014982A1 (en) * | 2017-07-12 | 2019-01-17 | iHealthScreen Inc. | Automated blood vessel feature detection and quantification for retinal image grading and disease screening |
-
2019
- 2019-12-27 CN CN201911380101.5A patent/CN111161240B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545269A (en) * | 2016-06-23 | 2018-01-05 | 西门子保健有限责任公司 | The method and system of vascular diseases detection is carried out using recurrent neural network |
| CN109410188A (en) * | 2017-10-13 | 2019-03-01 | 北京昆仑医云科技有限公司 | System and method for being split to medical image |
| CN107993220A (en) * | 2017-11-14 | 2018-05-04 | 北京理工大学 | The extracting method and device of x-ray imaging image medium vessels structure |
| CN109671076A (en) * | 2018-12-20 | 2019-04-23 | 上海联影智能医疗科技有限公司 | Blood vessel segmentation method, apparatus, electronic equipment and storage medium |
| CN109741344A (en) * | 2018-12-28 | 2019-05-10 | 上海联影智能医疗科技有限公司 | Blood vessel segmentation method, apparatus, medical imaging equipment and storage medium |
| CN109816666A (en) * | 2019-01-04 | 2019-05-28 | 三峡大学 | Symmetrical full convolutional neural networks model building method, eye fundus image blood vessel segmentation method, apparatus, computer equipment and storage medium |
| CN110348541A (en) * | 2019-05-10 | 2019-10-18 | 腾讯医疗健康(深圳)有限公司 | Optical fundus blood vessel image classification method, device, equipment and storage medium |
| CN110298844A (en) * | 2019-06-17 | 2019-10-01 | 艾瑞迈迪科技石家庄有限公司 | X-ray contrastographic picture blood vessel segmentation and recognition methods and device |
| CN110443813A (en) * | 2019-07-29 | 2019-11-12 | 腾讯医疗健康(深圳)有限公司 | Blood vessel, the dividing method of eye fundus image, device, equipment and readable storage medium storing program for executing |
| CN110517279A (en) * | 2019-09-20 | 2019-11-29 | 北京深睿博联科技有限责任公司 | Neck vessel centerline extracting method and device |
Non-Patent Citations (2)
| Title |
|---|
| EARLIER GLAUCOMA DETECTION USING BLOOD VESSEL SEGMENTATION AND CLASSIFICATION;E.Deepika 等;《Proceedings of the Second International Conference on Inventive Systems and Control》;484-490 * |
| 糖尿病视网膜病变眼底图像分类方法;梁平 等;《深圳大学学报理工版》;第34卷(第3期);290-299 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111161240A (en) | 2020-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11880972B2 (en) | Tissue nodule detection and tissue nodule detection model training method, apparatus, device, and system | |
| CN111161270B (en) | Vascular segmentation method for medical image, computer device and readable storage medium | |
| EP4198889A1 (en) | Image processing method apparatus based on artificial intelligence, and computer device and storage medium | |
| EP3992851A1 (en) | Image classification method, apparatus and device, storage medium, and medical electronic device | |
| CN111369542B (en) | Vessel marking method, image processing system, and storage medium | |
| WO2022089221A1 (en) | Medical image segmentation method and apparatus, and device, system and computer storage medium | |
| CN109712217B (en) | Medical image visualization method and system | |
| US8781187B2 (en) | Methods, apparatuses, and computer program products for identifying a region of interest within a mammogram image | |
| US20210033599A1 (en) | Information processing apparatus, control method, and program | |
| CN110363774B (en) | Image segmentation method and device, computer equipment and storage medium | |
| US11967181B2 (en) | Method and device for retinal image recognition, electronic equipment, and storage medium | |
| CN111383259B (en) | Image analysis method, computer device, and storage medium | |
| CN110738643B (en) | Analysis method for cerebral hemorrhage, computer device and storage medium | |
| CN111161240B (en) | Blood vessel classification method, apparatus, computer device, and readable storage medium | |
| WO2021259391A2 (en) | Image processing method and apparatus, and electronic device and storage medium | |
| CN110084868B (en) | Image correction method, apparatus, computer device, and readable storage medium | |
| CN114332132A (en) | Image segmentation method and device and computer equipment | |
| CN111243052A (en) | Image reconstruction method and device, computer equipment and storage medium | |
| CN110598696A (en) | Medical image scanning positioning method, medical image scanning method and computer equipment | |
| CN111160442B (en) | Image classification method, computer device, and storage medium | |
| CN111738975B (en) | Image identification method and image identification device | |
| CN115860067B (en) | Method, device, computer equipment and storage medium for generating countermeasure network training | |
| CN111899273A (en) | Image segmentation method, computer device and storage medium | |
| CN116524158A (en) | Interventional navigation method, device, equipment and medium based on image registration | |
| CN113160199B (en) | Image recognition method and device, computer 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |